Vol. 5. 36/-369. May /996 Cancer Epidemiology, Biomarkers & Prevention 36/

32P-Postlabeling Detection of DNA Adducts in Peripheral White Blood

Cells of Greenhouse Floriculturists from Western Liguria, l

Marco Peluso, Franco Merlo, Armelle Munnia,Claudia Bolognesi, Riccardo Puntoni, and Silvio Parodi2

Centro Nazionale per lo Studio dci Tumori di Origine Ambientale IM. P..

F. M.. A. M.. C. B.. R. P.. 5. P.1. Servizio di Epidemiologia Ambientale eBiostatistica IF. M.. R. P.1. Unit#{227}di Valutazione Tossicologica IC. B.l,

Servi,io di Oncologia Sperimentale IS. P.1. Istituto Nazionale per Ia Ricerca

sul Cancro; and t,Jniversit#{224} degli Studi di Genova IS. P.), Genova, Italy

Abstract

Pesticides are widely used in agriculture to enhance cropyields and to control disease vectors. Floriculturists workfrequently in greenhouses and may be exposed to highlevels of pesticides, which may result in adverse healtheffects. To evaluate the relationship between exposure topesticides and DNA adduct formation in peripheralWBCs of Italian floriculturists, the nuclease P1modification of a 32P-postlabeling assay was used toanalyze WBC DNA from floriculturists (n = 26) andmatched controls (n 22). DNA adduct-positive samples

were more frequent in floriculturists (11126; 42%) than inmatched controls (2/22; 9%) (P < 0.01). Slightly higherfrequencies of DNA adduct-positive samples wereobserved in floriculturists �44 years of age (53%) and infemale floriculturists (57%). Floricultural practice wasfound to be associated with a significantly higher DNAadduct-positive rate in WBCs (rate ratio, 5.12; 95%confidence interval, 1.1-23.7) after allowing for the effectsof age and gender. These two latter covariates were notsignificantly associated with DNA adduct-positive rates.The quantitative levels of DNA adducts were significantlyhigher in floriculturists than in matched controlsaccording to the Mann-Whitney nonparametric statistic(P = 0.0052). The median adduct level for positivesamples among floriculturists was 1.5/108 bases. A

specific, well-visible spot, named a adduct, was detectedin 7 out of the 1 1 DNA adduct-positive samples fromfloriculturists but in none of the (22 + 20) referentsamples (P = 0.0004). The presence of pesticide-relatedDNA adducts was confirmed clearly using the butanolextraction procedure. Six of 8 floriculturists and 0 of 10

Received 5/30/95; revised 1/5/96; accepted 1/26/96.The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked adrertise,ne,tt in

accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

I This work was supported partially by Consiglio Nazionale delle Ricerche

targeted project Prevention and Control of Disease Factors. Consiglio Nazionale

delle Richerche project Applicazioni Cliniche della Ricerca Oncologica (sub-pr.

I and 2). a Finalized Project of Ministry of Health, European Community

Environmental Research Programme Contract #91-01 61. and Associazione Itali-

ana per Ia Ricerca sul Cancro (AIRC 1995).

2 To whom requests kr reprints should be addressed, at Istituto Nazionale per Ia

Ricerca sul Cancro. Laboratory of Experimental Oncology. Viale Benedetto. XV.

n.l0. 16132 Genova. Italy.

referents were found positive with this method. Themedian adduct level for positive samples was 6.0/10�bases. Two strong spots close to the origin could beidentified in all six positive floriculturists, using thebutanol extraction procedure. No association betweenDNA adducts and use of specific pesticides was observed.

Introduction

Pesticides are a heterogeneous category of biologically active

compounds. Their application remains the most effectivemethod used to control disease vectors, improve agriculturalproduction, and protect stored crops. Some 120 chemical in-

gredients are the constituents of the most used pesticides ( I)among an estimated and rapidly increasing figure of 680 reg-istered active ingredients (2). Many of these compounds. be-cause of their environmental persistence, will be in our envi-ronment for many years to come. Despite the fact that the main

peculiarity of pesticides is their selectivity of action, com-

pounds that are toxic to pests may produce adverse effects inother organisms. including acute toxicity and delayed healtheffects. such as reproductive effects, degenerative disease, and

cancer in humans (3, 4). The evidence of carcinogenicity aris-ing from human epidemiologic studies is generally related to

exposure to complex chemical mixtures rather than to singleagents. Although a potential association between human expo-

sure to phenoxy herbicides and chlorophenols and non-Hodgkin’s lymphoma and soft-tissue sarcoma has been sug-

gested, the vast majority of the published epidemiologicinvestigations has dealt with exposure to broad categories of

agrochemical compounds such as insecticides, herbicides, andfungicides (5). Active agrochemical ingredients have been stud-ied extensively in animals to assess the human cancer hazard.but only few of these studies have been published in openliterature. The IARC evaluation of the carcinogenicity of 44

pesticides studied adequately in laboratory animals indicates 25active agrochemical ingredients with sufficient evidence and I 9with limited evidence of carcinogenicity in rodents (6-12).

Experimental data are available only for the active agrochemi-

cal compounds, and their extrapolation to much more complexhuman exposure situations is still of limited value.

Pesticide use involves exposure to complex mixtures ofdifferent types of chemicals. e.g. , by-products present in tech-

nical formulations, impurities, solvents, and other compoundsproduced during the storage procedure. Moreover, although“inert” ingredients have no pesticidal activity, they may bebiologically active and sometimes be the most toxic portion ofa pesticide formulation (13). The development of biomonitor-ing techniques has improved the estimation of health risksassociated with occupational/environmental exposure. Bio-monitoring studies on human populations exposed to pesticides

( 14) have measured alterations on cytogenetic parameters. suchas chromosomal aberrations, sister chromatid exchanges, and

micronuclei frequency. The majority of these studies (26 of 35)

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

362 DNA Adducts in WBC of Flonculturists

has reported positive findings, suggesting a causal relationshipbetween exposure to pesticides and genotoxic effects (14). Thenegative results obtained in a limited number of studies could

be attributed generally to the length of employment (14). Flo-riculturists work frequently in confined or semiconfined spaces

(i.e., greenhouses), where temperature and humidity are kept toa desired level that is ideal for specific crops, the use ofpesticide products is elevated, and their dissipation time may belonger than in open fields ( 15). For these reasons, workers arelikely to be exposed to high levels of complex chemical mix-tures (including impurities associated with technical-gradecomponents of commercial preparations). which may result in

undesired health effects.A large part of Italian flower production is confined to the

Provinces of Imperia and Savona, located in the western part of

the region of Liguria, near the French coast. Examination of themost recent agriculture statistics shows a yield of 255,600,000

roses, corresponding to 54% of Italian production, and a farm-land area devoted to floriculture of about 4800 hectares; 45.5%of which is greenhouse cropland ( 16, 1 7). The use of pesticidesin this area is elevated, i.e., 1,780,441 kg/year (16), and new

marketable agrochemicals are frequently introduced to improvethe intensive production of flowers. and perhaps also to prevent

the adverse effects of resistant strains of pests. Our previousstudies of fioriculturists of the Province of Imperia revealed asignificant association between occupational exposure to pes-ticides and cytogenetic alterations in terms of chromosomalaberrations and micronuclei frequencies detected in peripheralblood lymphocytes (18. 19). The present investigation is aimedat studying whether there is any relationship between exposureto pesticides experienced by floriculturists and the occurrenceof 32P-postlabeling DNA adducts measured in peripheral blood

nucleated cells, a biomarker of the biologically effective dose.

Assessment of human exposure to pesticides by measurementof DNA adducts has not been reported in the literature. DNAadduct detection is considered indicative of biologically rele-vant exposure to potential environmental carcinogens (20). The

postlabeling procedure developed by Randerath el a!. (21) hasbeen applied extensively to quantify DNA adducts in differentcellular systems and in human populations exposed to environ-mental carcinogens or potentially at increased risk for cancer.

Materials and Methods

Subjects and Data Collection. Twenty-six floriculturists and22 population referents, all nonsmokers, were selected from thesame geographic area including the provinces of Impena andSavona, located in the west side of the region of Liguria in

northwestern Italy, and enrolled in the study. Nonsmokingsubjects were part of a larger investigation aimed at assessing

the prevalence of selected biomarkers in nucleated blood cellsof 68 floriculturists and 55 age- and gender-matched referents.Nonsmoking floriculturists included 19 males and 7 females,and referents included 13 males and 9 females; their agesranged between 26 and 65 years and 28 and 6 1 years, respec-

lively. A standardized questionnaire was submitted to eachsubject at the time of blood collection, including items regard-

ing the last 5 years’ illnesses, hospitalizations, exposure to

therapeutic and/or diagnostic X-ray and past and present occu-pational exposures to solvents, tar products. dyes, asbestos,ionizing radiation, and use of pesticides. Floriculturists were

asked to check a detailed list of pesticides known to be sold in

the geographic area from which they were selected, to report theuse of other compounds, and to indicate the amount used duringthe previous 12 months (i.e. , kg/year). Moreover, floriculturists

had to declare the type of cultivation specifying the greenhouse

and open field areas, whether they handled pesticides, and thekind of protective devices used during preparation and appli-

cation of pesticides and regular work. Blood samples (5 ml)

were collected in heparinized Vacuette tubes between 9 a.m.

and 10 am., coded, stored properly in a thermic container, andsent immediately to the laboratory, where they were processed.

Blood Sampling and DNA Isolation. Blood samples (4-5 ml)were centrifuged at 800 g for 45 mm and the buffy coat wasseparated. To lyse the containing red cells, the buffy coats were

suspended in 3 volume 0. 17 M ammonium chloride at 4#{176}Cfor10 mm and centrifuged at 800 X g for 10 mm. The pellet

containing WBC was washed with ammonium chloride andstored at -80#{176}C.DNA was isolated from the stored cell pellets

by enzymatic digestion of RNA and proteins followed by

solvent extraction (22).

32P DNA Postlabeling Techniques. WBC DNA sampleswere analyzed blindly using the nuclease P1 procedure (23)

and the butanol extraction technique (24). The nuclease P1method is effective at detecting bulky hydrophobic adducts,such as those derived from polycyclic aromatic hydrocar-

bons and some single-ring aromatics that bind to exocyclic

amino groups in DNA purines (25). Conversely, the extrac-tion with butanol works predominantly for the majority ofthe arylamines bound to the C-8 position of the guanine and

for hydrophobic, relatively low-molecular-weight adductssuch as those formed by some alkylating agents (25). Sam-

ples of 10 i.�g DNA were digested for 3.5 h at 37#{176}Cto3’mononucleotides with 4 .tg of spleen phosphodiesterase(Boeringer Mannheim, Mannheim, Germany) and 0.4 units

of micrococcal nuclease (Sigma Chemical Co., St Louis,

MO) in 17 mM sodium succinate, 8 msi calcium chloride

buffer (pH 6.0; Ref. 26). DNA hydrolisates were evaporatedto dryness and redissolved appropriately. Ten pg of normalnucleotides were dephosphorylated by treatment with 4 jsg

of nuclease P1 (Boeringer Mannheim) for 40 mm at 37#{176}Casdescribed (27). Eight floriculturist and 10 referent DNAsamples were also analyzed by the butanol extraction pro-cedure as described already (27). The hydrolysates enrichedin adducted nucleotides were labeled by incubation with 75

�.tCi of carrier free [y-32P]ATP (3000 Ci/mmol; DuPont,Wilmington, DE) and S units of T4 polynucleotide kinase

(Pharmacia, Uppsala, Sweden) at 37#{176}Cfor 40 mm in 25 pA

of bicine buffer mixtures (pH 9) composed as describedpreviously (27). Excess ATP in the reaction mixtures and

efficiency of the removal of normal nucleotides were

checked by analyzing samples of the 32P-labeled postnucle-

ase P1 digest with one-dimensional PEI3-cellulose TLC(Merck, Darmstadt, Germany; Ref. 28). The labeled digest

DNA samples treated with nuclease P1 or extracted with

butanol were then resolved on PEI-cellulose TLC plates. Theplates were developed first with 1 M sodium phosphate (pH

6.0) onto a paper wick. Hydrophobic adducted nucleotidesremaining at the origin were then transferred onto a PEI-cellulose plate by 3 M lithium formate, and 7 M urea (pH 3.5),

and, after turning the plate by 90#{176},developed in 0.7 M LiC1,

0.45 M Tris-HC1, 7.7 M urea (pH 8.0) followed by 1.7 M

sodium phosphate (pH 5.0) onto a Whatman no. 1 wick. The

3 The abbreviations used are: PEI, polyethyleneimine; RAL, relative adduct

labeling: BaP, benzo(a)pyrene; RR, rate ratio; CI, confidence interval; cpm,counts/mm.

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

RAL =

Cancer Epidemiology, Biomarkers & Prevention 363

adduct spots were detected by screen-intensified autoradiog-raphy at -80#{176}C for 72 h.

For the quantitation of adducts, the radioactive areas of the

spots were cut out and Cerenkov radiation counted. DNA

adduct levels were measured as:

cpm in adducted nucleotides

cpm in total nucleotides

To calculate cpm in total nucleotides, samples of DNA digestswere diluted appropriately and reacted in mixtures as describedabove for postlabeling analysis of DNA adducts (obviouslyomitting the P 1 nuclease step). The labeled normal nucleotideswere then separated by one-dimensional PEI-cellulose TLC and

counted for their radioactivity (29). When using the P1 nucleasemethod, a radioactive spot marked x was sometimes foundclose to the right side of the plates (approximately 6.7 cm from

the X axis and 3.8 cm from the Y axis) in DNA samples from

floriculturists and referents (see Fig. 2, C and D: x spots). The

x spot was considered a priori as an artifact quantified sepa-rately. A spot in the same area of the chromatograms hasalready been observed when high specific activity [y-32P]ATPwas used to label DNA samples, regardless of tissue or species

origin (30), and it has been suggested that this type of spot maynot be due to a molecule bound covalently to DNA (30). DNAfrom liver of mice treated i.p. with 0.06 mg/kg BaP and sac-rificed after 24 h was assayed routinely (P1 nuclease method)

as a positive control along with the experimental samples. TheRAL of BaP-derived DNA adducts was S 1 ± 9 X 10#{176}(av-erage ± SD). The BaP adduct pattern obtained was in keepingwith that observed by Reddy and Randerath (22). The slight

difference in migration of BaP DNA adducts may result from

the slightly different solvent systems used and from the fact thatthe PEI-TLC plates were not identical, commercial in our caseand laboratory prepared in the Reddy and Randerath (22) case.

The 48 human samples analyzed were accompanied by eightstandard BaP experiments, performed as described above, i.e.,approximately one BaP experiment for every six human sam-

ples. In the standard BaP experiments (see Fig. 2F, spots 1-3),we were regularly able to detect three major spots. For the

strongest BaP spots (see Fig. 2F, spot I), we evaluated thevariability of the quantitative migration of the spot, in respectto the X and Y axes of the 12 X 12-cm plate (31), and obtainedthe following results: migration along the X axis, 2. 1 ± 0.7 cm;

and migration along the Y axis, 2. 1 ± 0.2 cm, starting from theorigin of the acceptor chromatogram (at 2 cm from the left and

bottom side).

Statistical Analysis. The association between exposure to pes-ticides and occurrence of DNA adducts in WBCs was tested bycomparing the rates of DNA adduct-positive samples in ex-

posed and referent subjects, allowing for the effects of age andgender, through multiple logistic regression analysis. The esti-mated associations were expressed as RRs and their 95% CIs.

Given the limited sample size, RRs were computed after adding0.5 to each empty cell to avoid sampling zeros, i.e., cells withno counts (32). The level of DNA adducts detected in floricul-

turists and referents were compared by the non-parametricMann-Whitney rank sum test (33). The Fischer exact test was

used for the analysis of the results obtained with the butanol

extraction method.

Results

We have examined 26 nonsmoker floriculturists and 22 non-

smoker referents. Floriculturists worked mainly in greenhouses

with a cultivated area ranging between 2,000 and 13,000 m2.

The mean number of years spent in floricultural activity was 22± 12 (SD) and was highly correlated with age (r 0.83, P

0.0001). All the subjects considered in our study are exposeddaily to a complex mixture of pesticides and to by-products

present in agrochemical formulations, impurities, solvents, and

chemicals produced during the storage procedures. Generally,

the active ingredients represent about 20-50% of the technical

formulation (34). The production of ornamental plants and

flowers in the greenhouses of Western Liguria is continuous,without appreciable oscillations all year long. No significantseasonal difference in pesticide use was observed. The expo-

sure to pesticides occurs during loading, mixing, and applica-

tion of pesticides, as well as during manual activities, by the

continuous contact with foliar surface and with contaminatedcut flowers. The number of active ingredients used during the

12 months before blood collection ranged between 2 and 14.

corresponding to amounts that ranged between 4 and 70 kg/

year. Table 1 shows the most important subclasses of pesticidesused by the floriculturists considered in our study. Among the

most intensively used pesticides we have identified a subgroupof DNA-damaging agents (Table 2).

Nuclease P1 Method. Blood DNA samples were analyzedblindly for the exposure status and considered to be negative forDNA adduction when no spots or only spots in the x position,

considered a technical artifact, were observed (see Fig. I and

“Materials and Methods”). Thirteen out of 26 floriculturists and7 out of 22 referents showed WBC DNA adducts in the xposition (Fig. I). Samples were considered positive for DNA

adduction when spots other than the x spot were observed.

They were stratified according to exposure status (i.e. , referentsversus floriculturists), age <44, and �44 years (i.e. , the median

age of referents), and gender. Table 3 shows the frequencies of

32P-postlabeling DNA adduct-positive samples detected among

referents and floriculturists by age and gender. The DNA ad-duct-positive rate was higher in floriculturists (11/26; 42%)

than in referents (2/22; 9%). The fraction of positive floricul-

turists was increased further by about 20% (see Table 6), ifpositivity in at least one of the two methods is considered. Thehighest 32P-DNA adduct-positive rates were detected in flori-

culturists aged �44 years (9/17; 53%) and in female floricul-turists (4/7; 57%). Floricultural practice was found to be asso-

ciated with a significantly increased DNA adduct-positive ratein the DNA of nucleated peripheral blood samples (RR, 5.12;

95% CI, I . 1-23.7) after allowing for the effects of age and

gender (Table 3). Conversely, the associations between DNAadduction, age (RR, 4.02; 95% CI, 0.9-23.7), and gender (RR,

1.94; 95% CI = 0.4-8.6) failed to reach a significant relevance.No association between DNA adducts and use of specific

pesticides was found, and no relationship was observed be-tween DNA adduct levels and the number or the amount of

agrochemicals used during the I 2 months before blood collec-tion (data not shown). The non-x extra spots were also meas-

ured quantitatively and expressed as RAL values. The resultsobtained (Table 4) indicate that higher levels of DNA adductswere found in floriculturists than in referents. According to the

nonparametric Mann-Whitney rank sum test (33), the one-tailed

probability that the higher quantitative prevalence of DNAadducts in floriculturists is determined only by chance was P =

0.0052. As we have illustrated in “Materials and Methods” for

the case of the BaP standard, our technique is highly reproduc-ible not only in qualitative terms and in terms of RAL values,

but also in terms of quantitative migration in the chromato-

grams. As a consequence, we were able to identify three re-

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

2 4 6 8 10 12

364 DNA Adducts in WBC of Floriculturists

Table / Pesticides used by the floriculturists included i n the study

Category Chemical classes kg/yr (%)

Fungicides Dithiocarbamates (vapam)

Nitrosulphorganics (dazomet)

Phthalimides (captan, folpet)

Benzimidazolics (benomyl. carbendazins)

3078 (55.2)

Insecticides Organophosphates (monocrotophos,

metamidophos)

Chlororganics (endosulfan)

Carbamates (methomyl)

2326 (41.6)

Herbicides Bipyridilics (paraquat)

Organophosphates (glyphosate)

178 (3.2)

Table 2 DNA-damaging agents mainly used by the floriculturists included in

the study

. .Pesticide Trade name

Chemical Abstracts.

registry no.Category

Captan Orthocide 133-06-2 Fungicide

Carbendazim Bavistin 10605-21 -7 Fungicide

Dazomet Basamid 533-74-4 Insecticide

Folpet Ridomil 133-07-3 Fungicide

Glyphosate Roundup 1071-83-6 Herbicide

Methomyl Lannate 16752-77-5 Insecticide

Methyl parathion Bladan 298-0()-0 Insecticide

Monocrotophos Azodrin 1 1 1 3-02-6 Insecticide

Paraquat Gramoxone 1910-42-S Herbicide

Zineb Aspor 12122-67-7 Fungicide

producible positions in the chromatograms that were specificfor the floriculturists (Fig. 1). These extra spots were arbitrarilynamed a, �, and y. By only using the solvent system selected,it was possible to obtain a clear resolution of the a spots fromthe origin in the DNA of WBC floriculturists. Fig. 1 shows alsothe reproducible position for the x spot (considered bona fide,

a technical artifact), and the � and e spots, two “unusual spots”that were detected in two referents. Fig. 2 reports typical

chromatograms showing the a spot (Fig. 14). the � spot (Fig.2B). and the y spot (Fig. 2C) detected in fioriculturists and the

x spot (Fig. 2!)) as well as a DNA adduct-negative sample (Fig.2E) detected among referents. Fig. 2F shows a typical BaP

standard. The unusual spots observed in two referents have the

following coordinates: � spot, X axis = 7.3 cm, and Y axis =

4.6 cm (sample code 415AP); and c spot, X axis = 5.5 cm, andY axis = 1 .2 cm (sample code 46AP). The possible explana-tions for these two spots are that these people may have expe-rienced some other unknown exposure undocumented in thequestionnaire or that the two spots observed are unusual spotsof endogenous origin. In Table S are shown from each positivesubject the membership (floriculturist or referent), type of

spots, RAL values in decreasing order, gender, and age. Thelevel of DNA adducts found in WBCs of positive floriculturistswas in the range of 0.04 fmol/j.tg DNA (median value, 1.5

adducts/108 bases). The lack of the a, 3, and y spots in

nonsmoking referents was confirmed by the analysis of WBCDNA of 20 nonsmoking subjects, comparable by age (range,25-59 years) and gender (1 1 males, 9 females) distributions,who served as referents in a study of urban police officers

exposed to city air pollutants. The prevalence of these spotsonly among floriculturists, and never among the two groups of

referents, was statistically significant (P = 0.0004, binomialdistribution). To assess better the relevance of our data, weinvestigated the repeatability of our measurements by randomly

STATIONARY PHASE[polyamine-ethylene ion-exchange resin (PEI) cellulose F]

Lrt�

C’)

I

ace-A)

<N.

c-a)

r0�O�

C’)

1 cmMOBILE PHASE 2 (D2)

[0.65M LiCI; 0.45M tris; 7.7M urea, pH8.0]

Fig. I. Quantitative migration of spot adducts in 26 nonsmoker floriculturists

and 22 matched controls. Quantitative migration of a, 13, v and x spots in cm

(average ± SD from the X and Y origins, respectively). a, 2.5 ± 0.6 and 2. 1 ±0.4; �3, 4.5 ± 0.4 and 4.2 ± 0.2; ‘y, 5.9 ± 0.1 and 8.1 ± 0.1; 8, 7.3 and 4.6; e,

5.5 and 1.2; x. 6.7 ± 0.3 and 3.8 ± 0.2. One floriculturist(Fl+) showed both an

a and a y spot; for this reason a + (3 + y = 12 instead of I 1. Two steps of the

TLC procedure are also a preliminary “cleanup” with 1 si sodium phosphatebuffer (pH 6.8) and a final “cleanup” with 1 .7 si sodium phosphate buffer (pH

5.0). The median intensity of the a, �3, and ‘y spots is about three times (1.5 X

108 adducts/base) the median intensity ofthe x spots (0.5 X lO”8 adducts/base).The median intensity of the 6 and e spots is I X l0 � adducts/base. DI. first

dimension; D2, second dimension.

selecting 20 of 48 samples ( 10 of 26 floriculturists and I 0 of 22referents). Seven out of I 1 DNA adduct-positive samples and 3out of 15 DNA adduct-negative samples were selected amongfloriculturists and 1 out of the 2 DNA adduct-positive samplesand 9 out of 20 negative samples among referents. A second

independent 32P-DNA postlabeling analysis was performed onthe 6-12-month-stored DNA samples, and the findings ob-

tamed are reported in Table 6. Comparing the first and secondsets of experiments, the average variation of relative intensityof the different spots was about 13%. The correlation coeffi-

cient was comfortably high (r = 0.83) and statistically signif-icant (P < 0.001). Considering that r� = 0.69, this means thatthe variability of the second set of measurements is able toexplain about 69% of the variability of the first set measure-ments. A 100% concordance was obtained in terms of qualita-tive positive and negative results. Table 7 shows the reproduc-

ibility of our chromatographic experiments from the point of

view of the degree of migration along the X and Y axes. The

average migration was very similar in the two sets of experi-ments. Different batches of PEI-cellulose TLC plates wereutilized during the time period between the two experiments.The internal variability (SD within experiments for five a

spots) for the first and the second set of experiments was 9%and 0.1%, respectively. The very small internal variabilityobserved during the second set of experiments is probably

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 365

Table 3 52P-Postlabeling DNA adduct-negative (-) and positive (+) samples and rates (% +), crude (OR”) and logistic regression (OR5) adjusted odds ratio point

estimates and their 95% Cl by exposure status, age, and gender (nuclease P1 method)

DNA adducts

Covariates Referents Floriculturists OR” (95% CI) OR” (95% CI)

No. - + (%+) No. - + (%+)

Exposure

Referents 20 2 (9) 1 1

Floriculturists 22 26 15 11 (42) 7.33 (1.2-41.6) 5.12 (1.1-23.7)

Age

<44 years II II - (-) 9 7 2 (22) 1 1

�44 years II 9 2 (18) 17 8 9 (53) 5.82(0.5-35.1) 4.02 (0.9-18.1)

Sex

Males 13 12 1 (8) 19 12 7 (37) 1 1

Females 9 8 1 (11) 7 3 4 (57) 1.36(0.3-6.1) 1.90(0.4-8.6)

“ Crude and “ logistic regression odds ratio point estimates adjusted for the other covariates based on adding 0.5 to each empty cell to avoid sampling zeros (33).

Table 4 ‘2P-Postlabeling DNA adduct levels detected in 22 referents and 26

floriculturists, all nonsmokers (nuclease P1 method)

Referents FloriculturistsDNA adduct level (RAL X I0�)

No. % No. %

Negative 20 91.0 15 57.7

<10 1 4.5 5 19.3

10-19.99 1 4.5 4 15.4

20-29.99 0 0 I 3.8

�30 0 0 1 3.8

related to the use of a single batch of TLC plates during a short

time interval. The between-experiment variability was slightlylarger (about 18%). It is likely that different batches of PEI-

cellulose sheets could be the major cause responsible for thelarger variations observed between the first and second sets of

experiments.

Butanol Extraction Method. The samples analyzed in two

independent experiments with the nuclease P1 method werealso analyzed using the butanol extraction method (Table 6). Ifwe compare the spots observed using the nuclease P1 method

with the spots observed with the butanol extraction method, themost straightforward comments suggested by the results ob-served (Table 6) are the following: nonsmoker floriculturist no.

2 (positive with the nuclease P1 method, but negative with thebutanol extraction method) presented a f3-type spot; similarly,nonsmoker referent no. I presented an s-type spot; samples 1,

3, 5, and 7 (nonsmoker floriculturists), positive with both meth-ods, always presented an a spot when we used the nuclease P1method.

Looking at the situation from the point of view of the

butanol extraction method, in an area of the chromatogramclose to the a spot observed with the nuclease P1 method, wecould observe at least two different spots (Fig. 3A); no otherevident spots could be observed in different parts of the chro-

matogram. The results observed with the nonsmoker floricul-turists of samples 8 and 9 (Table 6) suggest that a very intense

spot observed in the a zone using the butanol extraction methodcan sometimes correspond to a type of adduct completely

different from the a-type adduct observed with the nuclease P1method. Using the butanol extraction method, we have testedblindly eight nonsmoker floriculturists and 10 nonsmoker ref-

erents. Six of eight floriculturists were positive for DNA ad-ducts, whereas all referents were negative. The proportion 6/8

is different from the proportion 0/10 at a P 0.0015 level ofsignificance (Fisher’s exact test, two-tailed). Two strong spots

close to the origin were present in all six positive floriculturists(Fig. 3A). The median adduct level/l08 bases observed in

positive samples using the butanol extraction procedure was 6.0(Table 6), higher than the median value observed with thenuclease P1 method.

Discussion

Several biomonitoring studies have utilized successfully themeasurements of DNA adduct formation as a biological marker

of the DNA-targeted doses resulting from occupational expo-sure to genotoxic mixture (20). Floriculturists are widely ex-

posed to pesticides, which are complex mixtures of severalsubstances, as active agrochemical compounds, surface agents,stabilizers, dyers, solvents, and impurities ( 1). and which may

sometimes have genotoxic properties and induce DNA adductformation. Two examples of pesticides positive with the 32P-postlabeling technique and capable of forming DNA adductsare the fungicide dichlofluanid (34) and the technical formula-

tion of insecticide methomyl, Lannate 25 (26). Furthermore,some pesticides used by floriculturists, such as vinclozolin,

have a parent aromatic amine in their structure, which may be

activated metabolically to genotoxic carcinogens (35) and

might form DNA adducts. In the present study, we examinedthe formation of DNA adducts in WBC DNA from 26 non-smoker floriculturists and 22 nonsmoker referents. Using the

nuclease P1 method, our findings revealed a higher frequencyof 32P-DNA adduct-positive blood samples (42%) among fib-

riculturists exposed to chemical mixtures used during their

occupational practices than among referent subjects (9%) notexposed to pesticides. Floriculturists were found to have anincreased risk of DNA adduct positivity (RR, 5. 12; 95% CI,1.1-23.7) after adjustment for the effects of age and gender.

These latter two covariates were not associated with signifi-cantly increased risks for DNA adducts. 32P-DNA quantitative

assessment revealed statistically significantly higher levels ofadducts in peripheral nucleated cells of floriculturists as com-pared with referents (P = 0.0052). We also investigatedwhether the total amount of pesticides used by floriculturistsduring the last year before blood collection or the extent ofgreenhouse area could be correlated with DNA adduct levels.No significant correlation was found, nor we could detect anyrelationship with the use of specific pesticides (data not report-ed). These negative findings may be explained by the fact thatfloriculturists, by frequently treating and handling their crops,

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

a

V

D E F

I

366 DNA Adducts in WBC of Floriculturists

A B C

p4

� �

] �0_� ‘� �‘ � [

Fig. 2. Autoradiogranis of TLC naps of nuclease P1-treated ‘P-labeled DNA digests isolated from peripheral WBCs of floriculturists (A. a spot; B, � spot; C. y spot

and x spot); from WBCs of referents (D. x sp)t; E. negative sample; and from liver of mice treated with 0.06 mg/kg BaP as positive control (F. spot nos. 1-3).Autoradiography was pertormed at - 80C for 72 h.

Table 5 Frequency of a. (3. y. b. and � adduct spots and levels of adducts in nonsmoker positive subjects. expressed as number of adducts/105 nucleotides

)nuclease P1 method)

Positive subjects a spot (3 spot y spot b spot E spot RAL X 10 ‘� Sex Age (yr)

I” ** 31 M 55

2” - S - - 29 M 653” *

- - - - 16.8 M 634” *

- - - 15.4 F 485” - - - - * 15.1 M 56

6” *- - - - 14.4 M 61

7” S - - - 13.7 M 54

8’ - -. S - 9.7 F 269” - *

- - - 8.8 M 26

10” * - - - 8.7 F 56

II” - - - S S F 53

12” - S - - 2.5 F 54

13” * - - - 1.7 M 27

,‘ Floriculturists.

I, Referents.S Presence of the type of spot.

are continuously exposed to complex chemical mixtures. The

peculiarity of their exposure makes it practically impossible todetermine biologically meaningful time intervals between thelast important exposure and the collection of blood samples. In

a well-known work of biomonitoring of a situation of heavyenvironmental pollution, Hemminky et al. (36), using a nude-ase P1 enhancement procedure. reported levels of RALs about10 times higher than in our positive floriculturists. The order ofmagnitude of our numbers could suggest a milder exposure.

Obviously, the significance of our adducts in terms of geno-toxicity is unknown; therefore, no extrapolation is possible in

terms of risk assessment. The interindividual variation of DNAadduct level observed in our floriculturists (Table 5) may de-pend on several parameters such as adduct stability. cell turn-over, and DNA repair rates (20). Although we have no ideaabout the persistence of the adducts observed in floriculturists,

DNA adducts, even in long-lived lymphocytes, are generallyremoved after exposure within a period of a few hours up to

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 367

Table 6 Quantitative and qualitative results obtained from two independent analyses of 20 DNA samples of 10 nonsmoker floriculturists and 10 nonsmoker referents

(nuclease P1 method). plus a third analysis performed using the butanol extraction method

The paramrmetric correlation coefficient between the eight couples of positive results obtained in the first and second analysis, respectively, was r 0.83 (constant -0.01;

B = 1. 13; P � 0.01 ). For negative results a complete concordance was observed. When, for positive results, the average of analyses I and 2 were compared with the

corresponding values of analysis 3. no quantitative correlation was observed. At a qualitative level, 6 of 8 nonsmoker floriculturists and 0 of 10 nonsmoker referents were

positive with the butanol extraction method.

Floriculturists Analysis 1” Analysis 2” Analysis 3” Referents Analysis 1” Analysis 2” Analysis 3”

I 31 20 33 1 15.1 10 0

2 29 25 0 2 0 0 0

3 15.4 18 26 3 0 0 0

4 14.4 16.5 ND” 4 0 0 0

5 13.7 15 64 5 0 0 0

6 9.7 6.4 ND” 6 0 0 0

7 8.7 10 56 7 0 0 0

8 0 0 72 8 0 0 0

9 0 0 97 9 0 0 0

10 0 0 0 10 0 0 0

“RAL x l0’�.I, Not done. because not enough DNA was available for the third analysis.

Table 7 Comparative chromatographic migration of the major spot adducts of eight positive cases, in two subsequent independent analyses (nuclease P1 method)

Migrations along the X and Y axes are starting from t

of X”:X’ = 0.96 ± 0.19 (SD); average of Y”/Y’ =

he origin of the acceptor. Migration’ and migration” correspond to the first and second set of chrom

1.04 ± 0.17 (S.D.) The X’ and Y” values of migration’ were used as a reference standard.atograms. Average

Major spot type Migration’ X’ (cm) Migration’ Y’ (cm) Migration” A” (cm) Ratio X”:X’ Migration” Y” (cm) Ratio Y”:Y’

I. cm 2.1 1.8 2.9 1.38 2.2 1.22

2. a 2.5 2.1 2.5 1.00 2.2 1.05

3. a 2.8 2.2 2.5 0.89 2.2 1.00

4. a 3 1.8 2.5 0.83 2.2 1.22

5. a 3.5 3 2.8 0.80 2.2 0.73

6. f3 4.5 4.2 4.2 0.93 3.8 0.90

7. y 5.8 8 5.2 0.90 8.5 1.06

8. 6 7.3 4.6 6.8 0.93 5.4 1.10

several months, except for a small fraction that may persist for

years (37). For instance, an average reduction in adduct levelsof WBC of about nine times was observed in foundry workers

examined after 6 weeks of continuous exposure and immedi-ately after 4 weeks of vacation (38). Analysis of WBC DNA

adduct patterns revealed the presence in the chromatograms ofthree reproducible spots specific for the fioriculturists andnamed a (7/26), � (3/26), and y (2/26) (Table 5). Jahnke et al.

(39) and Gallagher et al. (40) have reported the presence of arelatively weak (i.e., 0.81 adducts/108 nucleotides) very hydro-

phobic spot-adduct similar to our a spot, in WBC of nonsmok-ing subjects, perhaps related to the consumption of charcoal-

broiled food. The eluents used by the above authors for thechromatographic migration along dimension 1 and 2 were sim-ilar but not identical to our eluents.

As already mentioned and discussed in “Materials andMethods,” the nuclease P1 method and the butanol extraction

method tend to detect, at least in part, different types of adducts.For this reason, we deemed it convenient to investigate a

significant part of our samples using both methods. Both meth-ods have confirmed that positive chromatograms are muchmore frequent in nonsmoker floriculturists than in nonsmokerreferents. According to the analysis of the spots observed (see“Results”). probably the a-spots observed with the nuclease P1method can also be detected with the butanol extraction method

(Table 6, floriculturist sample nos. 1 , 3, 5, and 7); other spotsdetected with the nuclease P1 method in different areas of thechromatogram (13’ �r’ & and a have never been observed with

the butanol extraction method; in an area (a zone) of the

chromatogram approximately 2-2.7 cm from the X and Y axes

strong adducts not detected by the nuclease P1 method weredetected in two out of eight samples of nonsmoker floricultur-

ists (Table 6, fioriculturist sample nos. 8 and 9). In these twocases, the butanol extraction procedure showed clearly its ca-pability of enlarging the global spectrum of detectable adducts.

In conclusion, we feel that it was very useful to use bothmethods in our analysis, a larger spectrum of DNA adducts

could be observed, and the statistical difference between flori-culturists and referents became much stronger, considering theglobality of the results obtained with the two complementary

methods. Our investigation indicates that the studied floricul-turists of Western Liguria are exposed to genotoxic agents,which results in increased levels of DNA adducts. The general

versatility and usefulness of the 2P-DNA postlabeling tech-nique as a workplace dosimeter and indicator of potential riskof cancer, is confirmed by our investigation. Our adduct levelswere measured at a single time point, with no information about

the pharmacokinetics of adduct formation and persistence, andwith no information concerning the potential mutagenicity ofthese adducts. Although we will never stress enough the un-

certainty of interpretation of the study findings, at the sametime we would not consider the observed DNA adduct level asnegligible, therefore recommending procedures of crop treat-ment aimed at reducing human exposure. At the level of con-trols by governmental agencies, allowed pesticides in generalhave been selected for their weak or negative carcinogenicity.Therefore, we were rather surprised by our positive findings,clearly significant from a statistical point of view and con-

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

368 DNA Adducts in WBC of Floriculturists

24. Gupta, R. C. Enhanced sensitivity of 32P-postlabeling analysis of aromatic

carcinogen: DNA adducts. Cancer Res., 45: 5656-5662, 1985.

B

Fig. 3. Autoradiograms of butanol

extracted 32P-labeled DNA digests

from floriculturist WBC (A) and from

referent WBC (B). Autoradiography

was performed at -80CC for 72 h.

firmed with two independent subsequent dosages of the samesamples and with two different methods, nuclease P1 and

butanol extraction. Our educated guess is that technical prep-arations can often display rather different genotoxic propertiesin respect to pure chemical agents. As an alternative expla-nation, we could suggest an indirect effect of modulation ofthe level of endogenous adducts, by some of the agents

present in the technical preparations. The long route towardthe characterization of our putative adducts could involve asystematic testing of all major technical products utilized byour floriculturists in vivo in mice. In case of sufficientsimilarity between human and rodent metabolism, we could

perhaps identify the equivalent of the a spot detected withthe nuclease P1 method, or of the typical spots detected withthe butanol extraction procedure, in peripheral WBCs ofmice exposed to a unique technical product. This could be

the first step toward the characterization of the adducts. Thislong work of screening is now considered among our futureplans.

References

I. Worthing. C. R.. and Hance, R. J. (eds.). The Pesticide Manual. A World

Compendium. Surrey. United Kingdom: British Crop Protection Council, Farn-

ham. 1991.

2. Briggs. S. A.. and the Staff of the Rachel Carson Council. Basic Guide to

Pesticides: Their Characteristics and Hazards, Washington, D.C.: Hemisphere

Publishing Corporation (Taylor and Francis Group). 1992.

3. Levine, R. Recognized and possible effects of pesticides in humans. In: W. J.

Hayes and E. R. Laws (eds.). Handbook of Pesticide Toxicology, Vol. 1, GeneralPrinciples. pp. 275-360. San Diego, CA: Academic Press Inc., 1991.

4. Hayes. W. J., and Laws, E. R. (eds.). Handbook of Pesticide Toxicology, Vol.1, General Principles. San Diego. CA: Academic Press Inc.. 1991.

5. Merlo, F.. Bolognesi. C., and Reggiardo. G. Carcinogenic risk of pesticides. J.Exp. Clin. Cancer Res., 13: 5-20. 1994.

6. IARC. Some Organochlorine Pesticides. IARC Monographs on the Evaluation

of Carcinogenic Risk to Humans. Vol. 5. Lyon. France: IARC, 1974.

7. IARC. Some Fumigants. the Herbicides 2,4-D and 2,4,5-T, Chlorinated Diben-

zodioxins and Miscellaneous Industrial Chemicals. IARC Monographs on the

Evaluation ofCarcinogenic Risk to Humans. Vol. 15. Lyon. France: IARC, 1977.

8. IARC. Some Halogenated Hydrocarbons. IARC Monographs on the Evalua-

tion of Carcinogenic Risk to Humans, Vol. 20. Lyon. France: IARC. 1979.

9. IARC. Miscellaneous Pesticides. IARC Monographs on the Evaluation of

Carcinogenic Risk to Humans, Vol. 30. Lyon, France: IARC, 1983.

10. IARC. Some Halogenated Hydrocarbons and Pesticide Exposures. IARC

Monographs on the Evaluation of Carcinogenic Risk to Humans, Vol. 41 . Lyon,

France: IARC, 1986.

I I . IARC. Overall Evaluations of Carcinogenicity: An Updating of IARC Mono-

graphs Volumes 1 to 42. IARC Monographs on the Evaluation of Carcinogenic

Risk to Humans, Suppl. 7. Lyon, France: International Agency for Research on

Cancer, 1987.

12. IARC. Occupational Exposures in Insecticide Application. and Some Pesti-

cides. IARC Monographs on the Evaluation of Carcinogenic Risk to Humans,

Vol. 53. Lyon, France: IARC, 1991.

13. Bernard, C. B., and Philogene, B. J. R. Insecticide synergists: role, impor-

tance, and perspectives. J. Toxicol. Environ. Health, 38: 199-223, 1993.

14. Bolognesi. C., and Merlo, F. Biomonitoring of human populations exposed to

pesticides. In: P. Cheremisinoff (ed), Encyclopedia of Environmental Control

Technology, Vol. 8, Work Area Hazard, pp. 673-737. Houston, Texas: Gulf

Publishing Co.. 1995.

15. Edelman, P. A. Prevention of injury by pesticides. In: W. J. Hayes and E. R.

Laws. (eds.), Handbook of Pesticide Toxicology, Vol. 1, General Principles, pp.

405-452. San Diego, CA: Academic Press, Inc., 1991.

16. Istituto Nazionale di Statistica. Statistiche dell’agricoltura, zootecnia e mezzi

di produzione. Annuario ISTAT, No. 38, 1993.

17. Istituto Nazionale di Statistica. 4) Censimento generale dell’agricoltura

1990-91 (Fourth General Census of Agriculture, 1990-91). ISTAT, 1994.

18. De Ferrari, M., Artuso, M., Bonassi, S., Bonatti, S., Cavalieri, Z., Pescatore,D., Marchini, E., Pisano, V., and Abbondandolo, A. Cytogenetic biomonitoring of

an Italian population exposed to pesticides: chromosome aberration and sister-

chromatid exchange analysis in peripheral blood lymphocytes. Mutat. Res., 260:

105-113, 1991.

19. Bolognesi, C., Parrini, M., Bonassi, S., IaneIlo, G., and Salanitto, A. Cyto-

genetic analysis of a human population occupationally exposed to pesticides.

Mutat. Res., 285: 239-249, 1993.

20. Beach, A. C., and Gupta. R. C. Human biomonitoring and the 32P-postla-

beling assay. Carcinogenesis (Lond.), 13: 1053-1074, 1992.

21. Randerath, K., Reddy, M. V., and Gupta, R. C. 32P-labeling test for DNA

damage. Proc. Natl. Acad. Sci. USA, 78: 6126-6129, 1981.

22. Reddy, M. V., and Randerath, K. A comparison of DNA adduct formation in

white blood cells and internal organs of mice exposed to benzo[alpyrene, diben-

zo[c,g]carbazole, safrole and cigarette smoke condensate. Mutat. Res., 241:

37-48, 1990.

23. Reddy, M. V. 32P-Postlabeling analysis of small aromatic and bulky non-aromatic DNA adducts. in: D. H. Phillips, M. Castegnaro, and H. Bartsch. (eds.),

Postlabeling Methods for Detection of DNA Adducts, pp. 25-34. Lyon, France:

IARC, 1993.

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

Cancer Epidemiology, Biomarkers & Prevention 369

25. Gupta, R. C., and Earley, K. 32P-adduct assay: comparative recoveries of

structurally diverse DNA adducts in the various enhancement procedures. Car-

cinogenesis (Lond.). 9: 1687-1693, 1988.

26. Bolognesi, C., Peluso, M., Degan, P., Rabboni, R., Munnia, A., and Abbon-

dandolo, A. Genotoxic effects of the carbamate insecticide, methomyl. 11. In viva

studies with pure compound and the technical formulation. Laminate 25. Environ.

Mol. Mutagen.. 24: 235-242. 1994.

27. Peluso. M.. Castegnaro. M., Malaveille, C., Friesen, M., Garren, L., Haute-

feuille. A.. Vineis. P., Kadlubar. F., and Bartsch, H. 32Postlabelling analysis of

urinary mutagens from smokers of black tobacco implicates 2-amino- 1-methyl-

6-phenylimidazol4.5-blpyridine (PhIP) as a major DNA-damaging agent. Carci-nogenesis (Lond.), 12: 713-717, 1991.

28. Randerath, E., Mittal, D., and Randerath. K. Tissue distribution of covalent

DNA damage in mice treated dermally with cigarette ‘Star”: preference for lung

and heart DNA. Carcinogenesis (Lond.), 9: 75-80, 1988.

29. Taningher. M., Peluso, M., Parodi, S., Ledda-Columbano, G. M., and Colum-

bano, A. Genotoxic and non-genotoxic activities of 2,4- and 2,6-diaminotoluene,

as evaluated in Fisher-344 rat liver. Toxicology. 99: 1-10. 1995.

30. Phillips, D. H., Hemminki, K., Alhonen, A., Hewer, A., and Grover, P. L.Monitoring occupational exposure to carcinogens: detection by 32P-postlabelling

of aromatic DNA adducts in white blood cells from iron foundry workers. Mutat.

Res.. 204: 531-541, 1988.

3 1 . Lu, L-J. W., Disher. R. M., Reddy. M. V.. and Randerath, K. 32P-Postlabel-ling assay in mice of transplacental DNA damage induced by the environmental

carcinogen safrole, 4-aminobiphenyl, and benzo(a)pyrene. Cancer Res., 46:

3046-3054, 1986.

32. Hollander, M., and Wolfe, D. A. Nonpararnetric statistical methods. In: Hill

Book. Wiley Series in Probability and Mathematical Statistics. New York: John

Wiley & Sons, Inc., 1973.

33. Agresti. A. Categorical Data Analysis. Wiley Series in Probability and

Mathematical Statistics. New York: John Wiley & Sons, Inc.. 1990.

34. Beach, A. C.. and Gupta, R. C. Human biornonitoring and the 32P-postla-

belling assay. Carcinogenesis (Lond.), 13: 1053-1074. 1992.

35. Sabbioni, G., and Neumann, H. G. Biomonitoring of arylamines: hemoglobin

adducts of urea and carbamate pesticides. Carcinogenesis (Lond.), 11: 1 1 1-1 15,

1990.

36. Hemminky, K., Grzybowska, E.. Chorazy. M.. Twardowska-Saucha, K.,

Sroczynski. J. W.. Putman, K. L. Randerath. K., Phillips. D. H., Hewer, A..

Santella, R. M., and Perera, F. P. DNA adducts in human related to occupationally

environmental exposure to aromatic compounds. In: H. Vainio, M. Santa. and A.

J. McMichael (eds.), Complex Mixtures and Cancer Risk, pp. 181-192. Lyon,

France: IARC, 1990.

37. Perera, F. P.. and Whyast. R. M. Biomarkers and molecular epidemiology in

mutation/cancer research. Mutat. Res., 313: 1 17-129, 1994.

38. Perera, F. P.. Hemminki. K., Young. 1. L.. Brenner. D., Kelly. G.. and

Santella. R. M. Detection of polycyclic aromatic hydrocarbon-DNA adducts in

white blood cells of foundry workers. Cancer Res., 48: 2288-2291. 1988.

39. Jahnke. G. D.. Thompson. C. L., Walker, M. P., Gallagher, J. E., Lucier, G.

W.. and DiAugustine, R. P. Multiple DNA adducts in lymphocytes of smokersand nonsmokers determined by 32P-postlabelling analysis. Carcinogenesis

(Lond.), 11: 205-211. 1990.

40. Gallagher, J., Mumford, J., Li, X., Shank. T., Manchester, D., and Lewtas. J.

DNA adduct profiles and levels in placenta. blood and lung in relation to cigarette

smoking and smokey coal emissions. In: D. H. Phillips, M. Castegnaro. and H.Bartsch. (eds.), Postlabelling Methods for Detection of DNA Adducts. pp. 283-

292. Lyon, France: IARC. 1993.

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

1996;5:361-369. Cancer Epidemiol Biomarkers Prev   M Peluso, F Merlo, A Munnia, et al.   Italy.blood cells of greenhouse floriculturists from western Liguria, (32)P-postlabeling detection of DNA adducts in peripheral white

  Updated version

  http://cebp.aacrjournals.org/content/5/5/361

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cebp.aacrjournals.org/content/5/5/361To request permission to re-use all or part of this article, use this link

on June 1, 2018. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from