Occupational exposure to mineral fibres. Biomarkers of oxidative damage and antioxidant defence and...

Mutagenesis vol. 23 no. 4 pp. 249–260, 2008 doi:10.1093/mutage/gen004Advance Access Publication 14 February 2008

Occupational exposure to mineral fibres. Biomarkers of oxidative damage andantioxidant defence and associations with DNA damage and repair

Marta Staruchova1, Andrew R. Collins2,Katarina Volkovova1, Csilla Mislanova1,Zuzana Kovacikova1, Jana Tulinska1, Anton Kocan1,Ladislav Staruch3, Ladislava Wsolova1 andMaria Dusinska1,4,*1Research Base of Slovak Medical University, Limbova 12, 833 03 Bratislava,Slovakia, 2Department of Nutrition, University of Oslo, PO Box 1046 Blindern,0316 Oslo, Norway, 3Faculty of Chemical and Food Technology, SlovakUniversity of Technology, Radlinskeho 9, 812 37 Bratislava, Slovakia and4Norwegian Institute of Air Research, Instituttveien 18, NO-2027 Kjeller,Norway

In order to study the effect of mineral wool exposure onoxidative DNA damage and lipid peroxidation, an epide-miological study was conducted in a mineral wool factoryin Slovakia. Altogether 141 subjects were investigated (21–58 years old), 43 controls (20 men and 23 women: 27 non-smokers, 16 smokers) and 98 exposed (75 men and 23women: 61 non-smokers, 37 smokers). We found highermalondialdehyde (MDA) levels in the group of all exposedworkers (P 5 0.025) and in exposed non-smokers (P 50.003) and a significantly suppressed activity of ceruloplas-min oxidase (P 5 0.02, P < 0.02, respectively) and catalase(CAT) (P 5 0.04, P 5 0.01, respectively) in these groups.The activity of glutathione S-transferase (GST) wasaffected by exposure to mineral wool; levels weresignificantly lower in all exposed subjects (P 5 0.04), inthe exposed non-smokers (P 5 0.03) and in exposed men (P< 0.01). Concentrations of vitamin C in plasma and theferric-reducing activity of plasma (FRAP) were notaffected by the mineral wool exposure. There wasa significant negative correlation between the activity ofglutathione peroxidase (GPX) and MDA in the whole group(P < 0.01) and in the exposed group and between CATactivity and MDA in all subjects (P < 0.01). GST activitycorrelated inversely with oxidized pyrimidines in lympho-cyte DNA, in almost all subgroups. We found significantnegative correlations between DNA repair and GPX in allsubjects (P 5 0.03) as well as in control men (P < 0.03) andbetween DNA repair and CAT in all control subjects (P <0.02) and in control men (P < 0.01). Interestingly, we founda positive correlation between DNA repair and MDA in allsubjects (P < 0.01) and in all exposed subjects (P < 0.03).The presented results indicate that mineral wool exposureinduces an increase in oxidative damage to biomoleculesespecially in the group of male non-smokers. However,optimal levels of antioxidants could have a protective effect.Biomarkers such as MDA, antioxidant enzymes andantioxidant vitamins measured in blood may be usefulbiomarkers of oxidative stress and antioxidant protection.We do not recommend FRAP as a marker of antioxidantstatus as interference from other constituents can providefalse or confusing results. Our study supports the idea that

there might also be other mechanisms by which antioxidantenzymes (especially GST) protect cells against oxidativeDNA damage.

Introduction

Free radicals and peroxides are clearly involved in physiolog-ical phenomena such as synthesis of prostaglandins andthromboxanes (1) and in pathogenesis of various diseases,including atherosclerosis, inflammatory diseases and cancer.They are also thought to participate in aging processes. Thebiological effects of these highly reactive compounds arecontrolled in vivo by a wide spectrum of antioxidative defencemechanisms: vitamins E and C, carotenoids, metabolites suchas glutathione and uric acid and antioxidant enzymes (2,3).Among these enzymes, superoxide dismutase (SOD; EC1.15.1.1) catalyzes dismutation of the superoxide anion (O�.)into hydrogen peroxide (H2O2), catalase (CAT; EC 1.11.1.6)breaks down H2O2 and glutathione peroxidase (GPX; EC1.11.1.9) both detoxifies H2O2 and converts lipid hydro-peroxides to non-toxic alcohols. In several clinical studies, oneor several of these antioxidant enzymes were measured inblood as possible biological indicators, especially concerninghyperlipidaemia and atherosclerosis (4), alcoholism (5–8),diabetes (9,10), Down syndrome (11,12), cancer (13,14) andAlzheimer disease (15). However, data concerning the bi-ological variability of these enzymes in blood of apparentlyhealthy subjects are scarce and have often been obtained fromrestricted populations (16–18).

Oxidative stress occurs when reactive oxygen species (ROS)are formed in amounts that exceed the capacity of theantioxidant defence system to remove them. The detrimentaleffects of ROS typically are assessed by the presence ofoxidatively altered biomolecules. Oxidative DNA damageformed during oxidative stress probably plays a critical rolein carcinogenesis (19). During oxidative stress in vivo or whenROS react with DNA in vitro, several types of DNA damageare formed, including strand breaks and small base lesions (20).8-Oxoguanine constitutes one of the most easily formed oxi-dative DNA lesions and can be detected by high-performanceliquid chromatography (HPLC) in both urine and tissues afteroxidative stress (19). Alternatively, oxidative DNA lesions canbe detected by the enzyme-modified single-cell gel electro-phoresis (comet) assay (21).

The adverse effects that arise from exposure to asbestos havestimulated the development of substitute materials, man-mademineral fibres (MMMFs). However, little is known about thehealth effects of these fibres. The potentially harmful effects ofall types of respirable fibres are at present one of the mostimportant fields of interest in industrial hygiene (22).

*To whom correspondence should be addressed. Tel/Fax: þ421 2 59369270; Email: [email protected]

� The Author 2008. Published by Oxford University Press on behalf of the UK Environmental Mutagen Society.

All rights reserved. For permissions, please e-mail: [email protected]. 249

by guest on May 10, 2011

mutage.oxfordjournals.org

Dow

nloaded from

http://mutage.oxfordjournals.org/

Mineral wool, made from natural basic material, is usedmainly for thermal and acoustic insulation and fire protectionof roofs, walls and floors. Mineral wool, formerly regarded bythe International Agency for Research on Cancer as a possiblycarcinogenic material (in the classification group 2B), wasreassessed in 2001 and is now assigned to group 3, i.e. notclassifiable as to human carcinogenicity. There is only weakepidemiological evidence linking mineral wool exposure withhuman cancer (23). There appear to have been few studies ofthe possible genotoxicity of MMMF. In principle, fibres mightinduce carcinogenesis directly, or via inflammation, with itsassociated release of reactive oxygen, damage to DNA and cellproliferation; on the other hand, mineral wool fibres are clearedrapidly from the rodent lung (24). In recent in vitroexperiments in rat alveolar macrophages, mineral wool wasnot cytotoxic (25).

As part of an EC-funded investigation into the possibleconsequences to health of exposure to mineral fibres, we havemonitored various biomarkers related to genotoxicity in 98workers exposed to mineral wool during its manufacture ina factory in Nova Bana (Slovakia) and 43 control employees ofthe same factory working in administrative or other jobs withminimal exposure to mineral wool (26). Strand breaks inlymphocyte DNA were higher in exposed compared to controlnon-smokers, but there was no effect of exposure on specificdamage to bases in DNA (oxidation and alkylation), nor onchromosome aberrations. The frequency of micronuclei washigher in women in the control group than in mineral wool-exposed women. DNA repair 8-oxoguanine DNA glycosylase(OGG) activity was unaffected by exposure, but was negativelycorrelated with micronucleus frequency, implying that unrepaired8-oxoguanine contributes to micronucleus formation. Theconclusion from this study was that, overall, mineral woolexposure has no clearly deleterious effect on genetic stability inhumans. Here we present data on additional biomarkers ofoxidative damage and antioxidant defence in relation to exposureto mineral wool and compare them with previously measuredgenetic stability markers, namely DNA damage and repair.

Materials and methods

Subject selection and samplings

Workers with at least 5 years exposure to mineral wool at an industrial plant inNova Bana, Slovakia, were recruited for this study. The group of 141 healthyadult volunteers (average age 40.8, ranging from 21 to 58) consisted of 43control clerical workers from the factory (20 men and 23 women: 27 non-smokers, 16 smokers) and 98 exposed (75 men and 23 women: 61 non-smokers, 37 smokers) (Table I). The sampling was carried out in September2000. The participants were interviewed by trained personnel. Each participantprovided detailed information on working histories, physical activity, activeand passive smoking and alcohol consumption, exposure to environmentalmaterials, health condition, occupation, education and other socioeconomicvariables. All workers underwent clinical examination, including a functional

spirometry test, radiological and immunological examination. All subjectscontributed a single blood sample in autumn. A urine sample was used formeasurement of cotinine to determine smoking status. Blood was collected byvenipuncture from fasted subjects, with ethylenediaminetetraacetic acid(EDTA) as anticoagulant, and used for isolation of plasma, lymphocytes anderythrocytes for both markers of genetic stability and markers of oxidativedamage and antioxidant protection. Both exposed and referent samples werecollected on the same day to minimize the influence of experimental variation.The study was blinded and all subjects, questionnaires and samples were codedprior to coming to the clinic for examination or to the laboratory for furtheranalyses. The study was conducted in a good laboratory practice laboratoryfollowing standard operating procedures.

All study participants signed an informed consent form. This study wasapproved by the Ethical Committee of the Research base of the Slovak MedicalUniversity (the Institute of Preventive and Clinical Medicine) in Bratislava.

Exposure in workplace and personal dosimetry

Exposure to mineral fibres and polycyclic aromatic hydrocarbon (PAH) in theworkplace was detected using high-volume stationary samplers (GPS-1;Graseby-Andersen, Atlanta, GA, USA) as well as by personal dosimeters.Concentrations of mineral fibres as well as PAH exposure were measured fourtimes a year (including sampling time) in the mineral wool factory. Twosamples of indoor air (one in a production hall and one in an administrativebuilding) were collected in each season.

Personal dosimetry was carried out at the time of sampling in the mineralwool factory. Personal air samples for fibres as well as PAHs were collected bypersonal samplers both at the workplace (8 h) and at home (8 h).

PAH measurements. Thirteen PAH congeners (fluorene, phenanthrene, anthra-cene, fluoranthene, pyrene, benzo[a]anthracene, chrysene, benzo[a]fluoranthene,benzo[k]fluoranthene, benzo[a]pyrene, indeno[1,2,3-c,d]pyrene, dibenzo[a,h]anthracene, benzo[g,h,i]perylene) collected by personal and high-volume samplerswere determined by high-resolution gas chromatography/low-resolution massspectrometry in the selected ion monitoring mode isotope dilution methodaccording to United States Environmental Protection Agency TO-13 method.

Fibre determination. Air particulate samples were taken on membrane filters.Samples were evaluated using a microscope with phase contrast (Nikon, Tokyo,Japan) according to the Reference Method for the Determination of AirborneAsbestos Fibre Concentration at Workplaces by Light Microscopy (MembraneFilter Method), AIA 1979 (London, UK). Fibre identification, size distributionand quantification were confirmed by an electron scanning microscope com-bined with X-ray microanalysis. All fibres were counted for artificial mineralfibres (basalt glass) and then sorted to respirable and non-respirable.

Assay of cotinine in urine

Concentration of cotinine was measured by a radioimmunoassay method usinga commercial kit from Brandeis University (Waltham, MA, USA). The methodis based on competition between labelled and unlabelled cotinine for limitedantibody-binding sites. To the standard diluted with buffer, 3H-cotinine andanti-cotinine-CDI-thyroglobulin were added. After incubation, normal rabbitserum and goat ‘Minnie’ anti-rabbit-c-globulins were added. The reactionmixture was incubated overnight at 4�C. The sediment was diluted in 0.1 MNaOH. After adding scintillation reagent, the activity of the cotinine–antibodycomplex was measured. For every set of samples, a new calibration curve wasmade. The concentration of cotinine is expressed as milligrams of cotinine permillilitre creatinine in urine. Creatinine (two-point rate test) was measured witha VITROS 250 biochemical analyser.

Biochemical screening

Serum biochemical markers were measured including aspartate aminotransa-minase, alanine aminotransferase, c-glutamyltransferase, alkaline phosphatase,amylase, urea, creatinine, albumin, total protein, total bilirubin, cholesterol,triacetylglycerols and glucose. Urine osmolarity and uric acid were measured.Biochemical analyses were performed in a Vitros 250 analyser (Johnson &Johnson, Rochester, NY, USA). The body mass index (BMI) was calculated asbody weight (kg)/height (m2).

Isolation of lymphocytes

After centrifugation of blood for collection of plasma (frozen as aliquots at�80�C), the buffy coat was recovered and mixed with RPMI 1640 mediumwith 10% foetal calf serum (FCS) before layering over an equal volume ofLymphoprep (Nycomed, Oslo, Norway) and centrifuging at 700 � g for 20 minat 20�C. The layer above the Lymphoprep, containing lymphocytes, wasremoved, diluted with medium and centrifuged at 700 � g for 15 min at 20�C.Lymphocytes were used immediately for estimation of DNA damage, while

Table I. Anthropometric data; numbers in each group, numbers of men andwomen, number of smokers and mean age (years) with 95% confidence limitin parenthesis

Total Men Women

Exposed factory workers 98 75 23Smokers 37 31 6Mean age 41.2 (39.3–43.0) 40.7 (38.6–42.9) 42.7 (39.1–46.2)

Factory controls 43 20 23Smokers 16 9 7Mean age 40.4 (37.9–43.0) 39.5 (34.6–44.3) 41.3 (38.7–43.9)

M. Staruchova et al.

250



Dow

nloaded from


surplus cells were suspended in 90% FCS and 10% dimethyl sulphoxide,divided into aliquots and frozen slowly to �80�C. For the DNA repair assay,lymphocytes (5 � 106 in 50-ll aliquots) were snap frozen in liquid nitrogen in45 mM N-2-hydroxyethylpiperazine-N#-2-ethanesulfonic acid, 0.4 M KCl,1 mM EDTA, 0.1 mM dithiothreitol and 10% glycerol, pH 7.8. The remainingblood cells were stored frozen at �80�C for additional analyses.

Measurement of markers of oxidative stress and antioxidant protection

Measurement of antioxidant enzyme activity. After plasma separation, eryth-rocytes were washed three times with isotonic saline (0.9% sodium chloride).Centrifugation for each washing was at 3000 r.p.m. for 10 min at 4�C. Eryth-rocytes were then lysed with hypotonic solution (distilled water) and thehaemolysate was used (at suitable dilutions) for measurement of activities ofantioxidant enzymes. The activity of GPX was determined by the kinetic methodaccording to Paglia and Valentine (27), CAT was measured spectrophotometricallyby a modified method of Cavarocchi et al. (28) and glutathione S-transferase(GST) by a kinetic method according to Habig et al. (29). The activity of SODwas estimated by a commercial test kit (Randox Lab. Ltd, Grumlin, UK).

Measurement of total antioxidant capacity, FRAP. The index of the combinednon-enzymatic antioxidant capacity of plasma [ferric-reducing activity ofplasma (FRAP)] was measured spectrophotometrically according to Benzie andStrain (30). Ferric to ferrous ion reduction at low pH causes formation ofa coloured ferrous–tripyridyltriazine complex. FRAP values were obtained bycomparing the absorbance at 593 nm in test samples with solutions containingferrous ions in known concentration.

Measurement of ceruloplasmin oxidase. Ceruloplasmin (CPL) oxidase activityin plasma was assayed with the use of o-dianisidine dihydrochloride accordingto the method of Schosinsky et al. (31).

Measurement of antioxidant micronutrients. Plasma vitamin C (32), a-tocopherol, c-tocopherol, b-carotene, retinol, xanthophyll and lycopene weredetected by HPLC (33).

Measurement of malondialdehyde. Lipid peroxidation was determined aslevels of malondialdehyde (MDA) by a modified HPLC method in plasma (34).

Measurement of DNA damage and repair. DNA damage was measured usingthe comet assay (single-cell alkaline gel electrophoresis) as previouslydescribed (35,36), and these data were presented previously (26). In additionto frank DNA strand breaks, oxidized bases were measured by conversion tobreaks using endonuclease III (Endo III, which recognizes oxidizedpyrimidines) or formamidopyrimidine DNA glycosylase (FPG, specific foraltered purines including formamidopyrimidines and 8-oxoguanine). Theenzyme 3-methyladenine DNA glycosylase II (AlkA) was used in a similarway to analyse DNA alkylation. Net enzyme-sensitive sites were calculated bysubtracting the comet score after incubation with buffer alone from the scorewith enzyme. Reference standard from frozen human lymphocytes was usedand slides run together with coded samples.

OGG activity was measured using an in vitro assay based on the ability ofa cell-free lymphocyte extract to incise substrate DNA containing 8-oxoguanine. The increase in DNA breaks during 10-min incubation, measuredwith the comet assay, was taken as the measure of repair incision for statisticalanalysis (36,37).

Statistical analysis

To test for significant differences between groups, we used the independentsamples t-test for normally distributed data and the Mann–Whitney U-test fornon-normally distributed data. Differences between three groups were tested byone-way analysis of variance and by Bonferroni’s test if equal variances wereassumed or Tamhane’s test if equal variances were not assumed. For variableswith normal distribution Pearson’s and for non-normally distributed variablesSpearman’s correlation coefficients were calculated. SPSS 13.0 software wasused for statistical analysis. The data are expressed as means � SEM.Differences with P , 0.05 were considered to be statistically significant.

Results

Measurement of exposure to PAHs and fibres in workplace andpersonal dosimetry

Exposure to mineral wool in the factory was monitored forseveral decades by National Public Health Institutions usingoptic microscope, but no published results are available.

The levels of mineral fibres in the workplace were muchhigher in the past than when measured during this study. In thepresent study mineral fibre as well as PAH exposures weremeasured four times a year (including the time of the bloodsampling). The presence of basalt fibres was confirmed in thewool factory, although all measured levels of basalt fibres werevery low (10–1000 times below the Slovak occupationallimits).

Total PAH levels in production and office workplacesdetermined in different seasons were also very low (Table II).Summed PAH concentrations in autumn air samples collectedby stationary samplers were 146 ng/m3 in a production hall and53 ng/m3 in an office room. Results of personal dosimetry ofsubjects working in the mineral wool factory are documentedin Figure 1. Personal monitoring was carried out altogether in29 workers of an industrial plant. Mean values of 161 ng/m3

(the sum of the 13 PAH congeners) and 148 ng/m3 were foundin air samples collected by personal samplers worn by thesmoking and non-smoking male workers, respectively (8-hsampling at work), whereas mean values were 89 and 26 ng/m3, respectively, for 8-h samples collected inside their flats/houses. Mean values of 106 and 82 ng/m3 were found insmoking and non-smoking male clerks, respectively (8-h sam-pling at work), whereas the concentrations were 75 and 53 ng/m3, respectively, in the 8-h samples collected inside their flats/houses. Summed PAH concentrations in air samples collectedby personal samplers carried by factory workers were 216 ng/m3

in the production area and 89 ng/m3 in the office area.

Clinical examination: effect of exposure and smoking onclinical markers

No pathological changes were found on X-rays of workersexposed to mineral wool. Spirometrical examination also didnot show any changes. Forced expiratory vital capacity andforced expiratory volume 1 second markers in exposed workersin the mineral wool factory did not differ from controls.Systolic blood pressure showed interesting variations. Moder-ately elevated (P 5 0.067) systolic blood pressure was foundamong the exposed group in comparison with the controlgroup. Moreover, systolic blood pressure in smokers amongthe factory controls and in all exposed (smokers and non-smokers) was above the normal reference range. Mean valuesof systolic blood pressure in exposed and control smokersreached 140 mm Hg. Smoking exposed workers hadsignificantly higher pulse frequency than non-smoking exposedcolleagues. There was no significant difference in BMIbetween exposed and control groups, nor between anysubgroups. However, the mean BMI in all groups andsubgroups exceeded 25. The lowest BMI were seen in control

Table II. Total PAH levels in production and office workplaces determinedby the stationary sampling in different seasons

Season Summed PAH concentration (ng/m3)

Production Office

Winter 300 163Spring 129 107Summer 118 90Autumn 146a 53a

aPersonal sampling at work and at home involving 22 workers (17 productionand 5 office ones) was carried out in the same time as blood sampling and in thesame month as the stationary sampling.

Biomarkers of exposure to mineral fibres

251



Dow

nloaded from


women (25.76 � 0.96) and the highest in exposed women(27.93 � 1.16).

People who worked with mineral wool had significantlyelevated serum levels of urea in all exposed groups exceptwomen (Table III). Men, both exposed and controls, had higherlevels of urea compared to the corresponding group of women.Other biochemical parameters did not show any changesrelated to exposure to mineral fibres (Table III).

Measurement of lipid peroxidation and antioxidant defencemarkers

Lipid peroxidation measured as level of MDA in plasma showselevation in the group of all exposed workers (P5 0.03) and inexposed non-smokers (P 5 0.003) (Figure 3).

We measured plasma concentrations of vitamin C andtocopherols, carotenoids and other micronutrients; FRAP; CPLoxidase activity; and the activities of antioxidant enzymes,GST, GPX, SOD and CAT in erythrocytes. The results arepresented in Tables IV and V. The activity of GST was affectedby exposure to mineral wool; GST levels were significantlylower in all exposed subjects, in the exposed men and exposednon-smokers. Control men had higher levels of GST thancontrol women. Activities of CPL oxidase and CAT weresignificantly suppressed in all exposed subjects. This effect wasalso seen specifically in non-smokers. Control women hadhigher activities of CPL than men. Activities of SOD, GPX andFRAP were not affected by the mineral wool exposure. Therewere significant differences only between men and women.

Concentrations of vitamins in plasma are shown in Table V.There were no significant differences between exposed andcontrol subjects, except for exposed non-smokers who hadhigher levels of c-tocopherol compared to control non-smokersand control smokers who had higher levels of b-carotenecompared to exposed smokers. Among exposed subjects,women had higher levels of vitamin C but lower levels ofretinol compared to men. Control smokers had higher levels ofc-tocopherol compared to control non-smokers. Women hadhigher levels of b-carotene, both in exposed as well as controlgroups compared to men, as did exposed non-smokerscompared to smokers (Table V).

Association between different markers in exposed and controlgroups and subgroups

In order to assess which markers of oxidative stress,antioxidant defence and genetic stability are the most relevantin relation to exposure to mineral fibres, we sought associations

of different biochemical markers, lipid peroxidation (MDA),markers of antioxidant protection and of genetic stability,in exposed and control groups and different subgroups(Tables VI–XIII). All subjects, groups and subgroups wereanalysed in the same way. We only report those results wheresignificant differences are seen.

Fig. 1. Individual PAH levels measured by personal sampling of non-smokers in September 2000 (Izomat, Nova Bana, Slovakia).

Table III. Levels of biochemical parameters measured in plasma

Exposed N Control N P

Urea (mmol/litre)All 4.81 � 0.14 89 4.09 � 0.14 43 0.001Men 5.05 � 0.16a 70 4.39 � 0.17b 17 0.008Women 3.92 � 0.21a 19 3.86 � 0.19b 22Smokers 5.05 � 0.17 35 4.36 � 4.36 15 0.03Non-smokers 4.65 � 0.20 54 3.92 � 0.15 24 0.005

Cholesterol (mmol/litre)All 5.54 � 0.13 89 5.25 � 0.18 39Men 5.50 � 0.14 70 5.18 � 0.35 17Women 5.68 � 0.30 19 5.30 � 0.18 22Smokers 5.59 � 0.21 35 5.67 � 0.27 15Non-smokers 5.50 � 0.16 54 4.98 � 0.23 24

Triglycerides (mmol/litre)All 2.40 � 0.20 89 2.04 � 0.33 39Men 2.61 � 0.24c 70 2.59 � 0.63 17Women 1.65 � 0.18c 19 1.62 � 0.32 22Smokers 2.94 � 0.37d 35 2.13 � 0.58 15Non-smokers 2.06 � 0.21d 54 1.99 � 0.41 24

Total proteins (g/litre)All 75.47 � 0.56 89 73.96 � 0.77 39Men 75.90 � 0.65 70 73.91 � 1.23 17Women 73.89 � 0.99 19 74.00 � 1.01 22Smokers 75.20 � 0.73 35 74.12 � 0.78 15Non-smokers 75.65 � 0.80 54 73.86 � 1.17 24

Glucose (mmol/litre)All 5.74 � 0.12 89 5.69 � 0.20 39Men 5.76 � 0.14 70 6.06 � 0.43 17Women 3.92 � 0.21 19 3.86 � 0.19 22Smokers 5.94 � 0.24 35 6.13 � 0.46 15Non-smokers 5.60 � 0.13 54 5.41 � 0.14 24

Total bilirubin (mmol/litre)All 13.06 � 0.50 89 14.27 � 1.27 39Men 13.51 � 0.58 70 17.23 � 2.45 17Women 11.36 � 0.93 19 11.99 � 1.09 22Smokers 12.40 � 0.80 35 13.87 � 2.38 15Non-smokers 13.47 � 0.67 54 14.52 � 1.50 24

Mean values are shown, with SEM, for all subjects in each group, for men andwomen and for smokers and non-smokers. P values shown in the table relate toexposed/control comparisons. Differences between men/women or smokers/non-smokers, if significant at P , 0.05, are indicated by a superscript letter:aP 5 0.001, bP 5 0.05, cP 5 0.002, and dP 5 0.04.


252



Dow

nloaded from


Biochemical parameters

We found associations between urea and antioxidant vitamins.There was a significant negative correlation of urea withvitamin C in all subjects (n 5 135, r 5 �0.192, P 5 0.03), allcontrols (n 5 42, r 5 �0.305, P 5 0.05) and control non-smokers (n 5 26, r 5 �0.424, P 5 0.03), with b-carotene incontrol non-smokers (n 5 26, r 5 �0.395, P 5 0.046). Apositive correlation between urea and c-tocopherol was foundonly in the group of control women (n 5 23, r 5 0.569, P 50.005).

The level of glucose positively correlates with cholesteroland with MDA in many groups and in almost all groups andsubgroups with triglycerides (Table VI). The level of glucose isalso associated with BMI as well as correlating with age(Table VI).

Antioxidant capacity

The total antioxidant capacity of plasma (FRAP) surprisinglyshowed a positive correlation with lipid parameters cholesteroland triglycerides in nearly all investigated subgroups andgroups (Table VII); it was also associated in many groups witha marker of lipid peroxidation, MDA (Table IX). There werepositive correlations of FRAP with glucose and BMI in allsubjects, all exposed and controls, all men, exposed men,control women, all smokers and all non-smokers as well asexposed smokers. FRAP correlated with age in all subjects, allmen and in non-smoker controls (Table VII).

We also found, as expected, positive correlations of FRAPwith a-tocopherol in all main groups and many subgroups.Correlations of FRAP with urea were found in all subjects, allexposed, non-smokers and exposed non-smokers. b-Carotenecorrelated negatively with FRAP in all subjects, all exposedsubjects and in non-smokers. No correlations with ascorbicacid or bilirubin were found (Table VIII).

Lipid peroxidation

There was a significant negative correlation between levels ofMDA and the activity of GPX in all subjects and in exposedgroup, in exposed smokers, exposed men and in male non-smokers. MDA correlated inversely also with CAT activity inall subjects, in non-smokers, in exposed non-smokers and innon-smoking women (Table IX).

As already mentioned, we found a positive correlationbetween lipid peroxidation measured as level of MDA andFRAP in all investigated subjects, all exposed, smokers,exposed smokers, all women, exposed women and non-smoking women (Table IX).

Fig. 3. Levels of MDA in plasma.

Fig. 2. Levels of DNA strand breaks in lymphocytes. Data from Dusinskaet al. (26).

Table IV. Levels of antioxidants and antioxidant enzymes measured inlymphocytes and erythrocytes


GST (U/gHb)All 71.2 � 3.3 98 83.7 � 4.9 43 0.038Men 67.2 � 3.8 75 89.9 � 7.7a 20 0.007Women 84.1 � 6.3 23 78.2 � 6.3a 23Smokers 73.7 � 5.4 37 78.3 � 7.3 16Non-smokers 69.8 � 4.2 61 86.6 � 6.5 27 0.032

SOD (U/gHb)All 881.1 � 23.2 98 895.3 � 31.5 43Men 902.2 � 28.4b 75 969.9 � 47.1c 20Women 812.3 � 31.5b 23 830.4 � 38.3c 23Smokers 933.9 � 38.5 37 905.3 � 51.5 16Non-smokers 853.0 � 28.6 61 889.9 � 40.4 27

GPX (U/gHb)All 19.4 � 0.5 98 20.4 � 0.7 43Men 19.0 � 0.5 75 20.9 � 0.9 20Women 20.7 � 1.0 23 20.0 � 1.0 23Smokers 20.3 � 0.7 37 19.8 � 1.0 16Non-smokers 18.9 � 0.6 61 20.8 � 0.9 27

CAT (kU/gHb)All 24.5 � 0.9 98 29.0 � 2.0 43 0.044Men 24.0 � 1.0 75 25.3 � 2.4 20Women 26.1 � 2.3 23 32.3 � 3.1 23Smokers 24.9 � 1.7 37 25.4 � 3.8 16Non-smokers 24.3 � 1.1 61 31.0 � 2.3 27 0.012

CPL (U/litre)All 115.8 � 3.2 98 129.4 � 5.0 43 0.022Men 107.0 � 2.6 75 112.0 � 5.0d 20Women 144.3 � 8.4 23 144.5 � 6.9d 23Smokers 119.8 � 5.2 37 125.5 � 9.6 16Non-smokers 113.8 � 4.1 61 131.5 � 5.8 27 0.016

FRAP (lmol/litre)All 987.8 � 36.3 98 899.3 � 26.3 43Men 1036.8 � 44.8d 75 986.4 � 35.7d 20Women 828.0 � 34.5d 23 823.5 � 31.4d 23Smokers 1037.6 �71.8 37 898.8 � 30.1 16Non-smokers 961.4 � 40.4 61 900.1 � 52.8 27

Mean values are shown, with SEM, for all subjects in each group, for men andwomen and for smokers and non-smokers. P values shown in the table relate toexposed/control comparisons. Differences between men/women or smokers/non-smokers, if significant at P , 0.05, are indicated by a superscript letter:aP 5 0.03, bP 5 0.04, cP 5 0.03 and dP 5 0.001.


253



Dow

nloaded from


DNA damage and biochemical markers and markers ofantioxidant defence

It should be mentioned here that the results of assays for DNAdamage are taken from a previous publication (26) where wereported no difference of FPG or Endo III sites betweenexposed and controls. However, levels of strand breaks ofexposed subjects were higher than controls (Figure 2).

FPG-sensitive sites correlated positively with BMI incontrols (n 5 42, r 5 0.333, P 5 0.03), men (n 5 93, r 50.374, P 5 0.001), exposed men (n 5 73, r 5 0.332, P 50.004), control men (n 5 20, r 5 0.506, P 5 0.03) andsmokers (n 5 51, r 5 0.289, P 5 0.04). The same positivecorrelation was found with Endo III-sensitive sites in theexposed group (n 5 96, r 5 0.223, P 5 0.03) and in exposedsmokers (n 5 37, r 5 0.344, P 5 0.04).

We found an association of DNA damage with age. Thelevel of FPG-sensitive sites correlated with age in all controlsubjects (n5 42, r5 0.389, P5 0.01), in all men (n5 95, r5

0.240, P 5 0.02), in control men (n 5 20, r 5 0.481, P 50.03) and in non-smoker controls (n 5 27, r 5 0.539, P 50.004). The positive association between age and Endo III-sensitive sites was found only in the exposed smokers group(n 5 37, r 5 0.362, P 5 0.03).

We found a positive correlation between FPG sites andcholesterol in exposed men (n5 70, r5 0.414, P5 0.001) buta negative correlation in exposed women (n 5 19, r 5 �0.527,P 5 0.02). A positive association of FPG-sensitive sites withtriglycerides was found in the control group (n 5 38, r 50.443, P 5 0.005), in control women (n 5 21, r 5 0.479, P 50.03) and in non-smoking controls (n 5 24, r 5 0.557, P 50.005). Triglycerides correlated positively also with Endo IIIsites in the group of non-smoking controls (n 5 24, r 5 0.557,P 5 0.005).

Bilirubin correlated positively with FPG sites in the entireexposed group (n 5 89, r 5 0.210, P 5 0.01), in controlsmokers (n 5 14, r 5 0.542, P 5 0.05) and in exposed non-smokers (n 5 54, r 5 0.368, P 5 0.006).

The level of total proteins in plasma correlated inverselywith both FPG as well as Endo III-sensitive sites in all women

Table V. Levels of antioxidant micronutrients in plasma


Vitamin C (lmol/litre)All 32.75 � 0.83 98 35.08 � 1.47 43Men 31.18 � 0.88a 75 32.40 � 1.95 20Women 37.87 � 1.72a 23 37.42 � 2.07 23Smokers 30.57 � 1.33 37 33.50 � 3.14 16Non-smokers 33.91 � 1.04 61 35.88 � 1.53 27

Xanthophyll (mg/litre)All 0.20 � 0.01 98 0.19 � 0.01 43Men 0.20 � 0.01 75 0.20 � 0.02 20Women 0.21 � 0.02 23 0.18 � 0.01 23Smokers 0.19 � 0.01 37 0.18 � 0.01 16Non-smokers 0.21 � 0.01 61 0.19 � 0.02 27

Retinol (mg/litre)All 0.60 � 0.01 98 0.58 � 0.02 43Men 0.62 � 0.01b 75 0.60 � 0.03 20Women 0.53 � 0.02b 23 0.57 � 0.02 23Smokers 0.59 � 0.02 37 0.61 � 0.03 16Non-smokers 0.60 � 0.02 61 0.57 � 0.02 27

c-Tocopherol (mg/litre)All 1.24 � 0.01 98 1.00 � 0.07 43Men 1.30 � 0.10 75 1.13 � 0.12 20Women 1.05 � 0.10 23 0.89 � 0.07 23Smokers 1.23 � 0.08 37 1.19 � 0.11c 16Non-smokers 1.25 � 0.11 61 0.89 � 0.08c 27 0.048

a-Tocopherol (mg/litre)All 8.61 � 0.30 98 8.05 � 0.45 43Men 8.87 � 0.36 75 8.43 � 0.84 20Women 8.23 � 0.53 23 7.73 � 0.45 23Smokers 9.06 � 0.51 37 8.08 � 0.62 16Non-smokers 8.34 � 0.37 61 8.04 � 0.63 27

Lycopene (mg/litre)All 0.22 � 0.01 98 0.23 � 0.01 43Men 0.22 � 0.01 75 0.23 � 0.02 20Women 0.22 � 0.01 23 0.24 � 0.02 23Smokers 0.20 � 0.01 37 0.22 � 0.01 16Non-smokers 0.23 � 0.01 61 0.24 � 0.02 27

b-Carotene (mg/litre)All 0.13 � 0.01 98 0.16 � 0.02 43Men 0.11 � 0.01d 75 0.11 � 0.02e 20Women 0.18 � 0.02d 23 0.19 � 0.03e 23Smokers 0.09 � 0.01a 24 0.14 � 0.02 16 0.017Non-smokers 0.15 � 0.01a 37 0.17 � 0.03 40

Mean values are shown, with SEM, for all subjects in each group, for men andwomen and for smokers and non-smokers. P values shown in the table relate toexposed/control comparisons. Differences between men/women or smokers/non-smokers, if significant at P , 0.05, are indicated by a superscript letter.aP 5 0.001, bP 5 0.002, cP 5 0.03, dP 5 0.003 and eP 5 0.04.

Table VI. Association of glucose with lipid and other parameters measured inplasma or calculated from data

Parameter Group N Correlationcoefficient (r)

P

Cholesterol All 128 0.255 0.004Exposed 89 0.306 0.004Women 41 0.388 0.012Men exposed 70 0.271 0.023Non-smoker 78 0.277 0.014Non-smokers exposed 54 0.277 0.043

Triglycerides All 128 0.381 0.001Exposed 89 0.398 0.001Men 87 0.322 0.002Women 41 0.392 0.011Men exposed 70 0.387 0.001Women control 22 0.614 0.002Non-smokers 78 0.410 0.001Smokers 50 0.336 0.017Non-smokers exposed 54 0.421 0.002Smokers exposed 35 0.359 0.034Smokers control 15 0.530 0.042

MDA All 128 0.208 0.018Exposed 89 0.218 0.040Men 87 0.045 0.045Non-smokers 78 0.238 0.036Non-smokers exposed 54 0.270 0.049

BMI All 126 0.279 0.002Exposed 87 0.304 0.004Men 85 0.322 0.003Men exposed 68 0.400 0.001Women control 22 0.444 0.038Non-smokers 77 0.389 0.001Non-smokers control 24 0.491 0.015Non-smokers exposed 53 0.323 0.015Smokers exposed 34 0.372 0.030

Age All 128 0.303 0.001Exposed 89 0.423 0.001Men 87 0.269 0.012Women 41 0.437 0.004Men exposed 70 0.356 0.002Women exposed 19 0.549 0.015Non-smokers 78 0.398 0.001Non-smokers exposed 54 0.489 0.001Smokers exposed 34 0.372 0.030

We report only significant data (P , 0.05).


254



Dow

nloaded from


(n5 40, r5 �0.318, P5 0.05, respectively, r5 �0.370, P50.02) and in exposed women (n 5 19, r 5 �0.495, P 5 0.03,respectively, r 5 �0.483, P 5 0.036). Urea concentrationscorrelated negatively with Endo III-sensitive sites only in thegroup of control non-smokers (n 5 24, r 5 �0.418, P 50.042).

There were interesting associations between oxidative DNAdamage markers and antioxidant enzymes. GST activitycorrelated inversely with oxidized purines measured as FPG-sensitive sites in all smokers and in male smokers. Theassociation between DNA damage and GST activity was morepronounced when measuring Endo III sites (oxidized pyrimi-dines). We found an inverse correlation in almost all

subgroups: all subjects, exposed, all women, all men, allnon-smokers, exposed men, control non-smokers and womennon-smokers. On the other hand, the level of FPG-sensitivesites correlated positively with antioxidant enzyme GPX in thecontrol group, women, control women, control non-smokersand women non-smokers (Table X).

We also found significant positive correlations ofDNA damage, both FPG- and Endo III-sensitive sites, with

Table VII. Association of FRAP with lipid parameters and other parametersmeasured in plasma or calculated from data

Parameters Group N Correlationcoefficient (r)

P

Cholesterol All 128 0.400 0.001Exposed 89 0.446 0.001Control 39 0.432 0.006Men 87 0.429 0.001Women 41 0.588 0.001Men exposed 70 0.414 0.001Women exposed 19 0.696 0.001Men control 17 0.546 0.023Women control 22 0.467 0.028Non-smokers 78 0.423 0.001Smokers 50 0.486 0.001Non-smokers exposed 54 0.386 0.004Smoker exposed 35 0.511 0.002Smokers control 15 0.567 0.028

Triglycerides All 128 0.444 0.001Exposed 89 0.422 0.001Control 39 0.436 0.006Men 87 0.390 0.001Women 41 0.469 0.002Men exposed 70 0.390 0.001Women exposed 19 0.533 0.019Men control 17 0.687 0.002Women control 22 0.578 0.005Non-smokers 78 0.516 0.001Smokers 50 0.333 0.018Non-smokers exposed 54 0.463 0.001Non-smokers control 24 0.556 0.005Smokers exposed 35 0.376 0.026Smokers control 15 0.743 0.001Exposed 89 0.422 0.001

Glucose All 128 0.270 0.002Exposed 89 0.283 0.007Women 41 0.338 0.031Men exposed 70 0.270 0.024Women control 22 0.444 0.038Non-smokers 78 0.258 0.023Smokers 50 0.322 0.020Smokers exposed 35 0.393 0.049

BMI All 139 0.334 0.001Exposed 89 0.267 0.009Control 43 0.324 0.034Men 85 0.350 0.001Men exposed 73 0.349 0.003Women control 23 0.493 0.017Non-smoker 87 0.302 0.005Smokers 52 0.442 0.001Smokers exposed 36 0.403 0.015

Age All 141 0.170 0.044Men 95 0.225 0.028Non-smokers control 16 0.509 0.044


Table VIII. Association of FRAP with plasma antioxidants measured inplasma

Plasma antioxidant Group N Correlationcoefficient (r)

P

Ascorbic acid – – – –a-Tocopherol All 141 0.425 0.001

Control 43 0.349 0.022Exposed 98 0.478 0.001Men 95 0.462 0.001Women 46 0.385 0.008Men exposed 75 0.517 0.001Women exposed 23 0.424 0.044Non-smokers 88 0.452 0.001Smokers 53 0.358 0.009Non-smokers exposed 61 0.460 0.001Non-smokers control 27 0.428 0.026Smokers expose 37 0.481 0.003

Urea All 128 0.261 0.003Exposed 89 0.218 0.041Non-smokers 78 0.290 0.010Smokers exposed 37 0.347 0.035

Protein All 128 0.263 0.003Exposed 89 0.333 0.001Men 87 0.268 0.012Men exposed 70 0.314 0.008Non-smokers 78 0.289 0.010Smokers exposed 37 0.347 0.035Non-smokers exposed 61 0.398 0.003

Bilirubin – – – –b-Carotene All 141 �0.221 0.008

Exposed 98 �0.231 0.022Non-smokers 88 �0.272 0.010


Table IX. Association of MDA with antioxidant enzymes and FRAPmeasured in plasma or in erythrocytes

Enzyme Group N Correlationcoefficient (r)

P

GPX All 141 �0.227 0.007Exposed 98 �0.226 0.013Exposed smokers 37 �0.332 0.044Men exposed 75 �0.245 0.034Men non-smokers 55 �0.304 0.024

CAT All 141 �0.171 0.043Non-smokers 88 �0.303 0.004Non-smokers exposed 61 �0.252 0.050Women non-smokers 33 �0.422 0.014

FRAP All 141 0.219 0.016Exposed 98 0.251 0.013Smokers 53 0.286 0.038Smokers exposed 37 0.347 0.035Women 46 0.397 0.006Women exposed 23 0.447 0.032Women non-smokers 33 0.427 0.013



255



Dow

nloaded from


a-tocopherol and retinol in several groups. Interestingly, aninverse correlation of FPG as well as Endo III sites with b-carotene was found in several groups. A negative correlation ofEndo III-sensitive sites was found with xanthophyll in smokersand control smokers (Table XI).

DNA repair and biochemical markers and markers ofantioxidant defence

An interesting association between DNA repair capacity(incisions per 10 min at 8-oxoguanine in DNA, measured withthe comet assay as the activity of a lymphocyte extract ona DNA substrate containing 8-oxo-dG) and the level of lipidperoxidation in plasma was found in all investigated subjects,in all exposed subjects and in all male subjects. There werealso positive correlations between DNA repair capacity andserum cholesterol in all subjects and in exposed, as well aswith triglycerides in exposed and exposed smoker groups(Table XII).

On the other hand, we found a significant negativecorrelation between DNA repair capacity and GPX in allsubjects as well as in control men and CAT in all controlsubjects, control men and control smokers. The relationshipsbetween DNA repair rate and antioxidant micronutrientsascorbic acid, c-tocopherol, b-carotene and retinol are alsoshown in Table XIII. There are negative correlations of DNArepair rate with ascorbic acid and b-carotene in several groupsand subgroups and positive correlations with c-tocopherol andretinol in smokers and non-smokers (Table XIII).

Table X. Association of DNA damage (FPG and Endo III sites) withantioxidant enzymes GST and GPX, measured in lymphocytes or inerythrocytes

Antioxidant enzyme Group N Correlationcoefficient (r)

P

GST with Endo III All 141 �0.258 0.002Exposed 98 �0.249 0.014Women 46 �0.422 0.004Men 95 �0.214 0.038Non-smokers 88 �0.309 0.003Men exposed 74 �0.237 0.041Non-smokers exposed 61 �0.280 0.029Non-smokers control 27 �0.390 0.044Women non-smokers 33 �0.426 0.013

GST with FPG Smokers 52 �0.291 0.036Men smokers 40 �0.360 0.023

GPX with FPG Control 42 0.467 0.004Women 45 0.316 0.04Women control 22 0.448 0.04Non-smokers control 27 0.554 0.003Women non-smokers 33 0.388 0.03


Table XI. Association of DNA damage (FPG and Endo III sites) withantioxidant vitamins a-tocopherol, b-carotene and retinol, measured inlymphocytes or in plasma

Antioxidant vitamins Group N Correlationcoefficient(r)

P

a-Tocopherol with FPG All 140 0.261 0.002Men 95 0.296 0.004Smokers 53 0.273 0.048Non-smokers 88 0.359 0.001Men control 20 0.590 0.006Non-smokers control 27 0.690 0.001

a-Tocopherol withEndo III

All 141 0.192 0.023Non-smokers 88 0.275 0.010Non-smokers control 27 0.526 0.005

b-Carotene with FPG All 141 �0.177 0.036Smokers 52 �0.377 0.006Non-smokers exposed 61 �0.276 0.031

b-Carotene with Endo III Smokers 53 �0.348 0.011Smokers exposed 37 �0.460 0.004

Retinol with FPG All 140 0.207 0.014Exposed 98 0.204 0.044Control 43 0.448 0.003Women 45 0.371 0.045Non-smokers 88 0.216 0.043Women exposed 23 0.624 0.001Non-smokers control 27 0.463 0.015Smokers exposed 37 0.333 0.044

Retinol with Endo III All 141 0.195 0.021Non-smokers 88 0.222 0.036Women exposed 23 0.475 0.022Non-smokers control 27 0.498 0.008Smokers exposed 37 0.333 0.044

Xanthophyll withEndo III

Smokers 53 �0.365 0.007Smokers control 16 �0.498 0.049


Table XII. Relationship between DNA repair rate (incisions per 10 min at 8-oxoguanine in DNA, measured with the comet assay) with MDA,triglycerides and cholesterol

Group N Correlationcoefficient (r)

P

MDA All 141 0.224 0.008Exposed 98 0.224 0.026Men 95 0.225 0.028

Cholesterol All 128 0.128 0.04Exposed 89 0.228 0.031

Triglycerides Exposed 89 0.218 0.04Exposed smokers 35 0.380 0.02


Table XIII. Relationship between DNA repair rate (incisions per 10 min at 8-oxoguanine in DNA, measured with the comet assay) with antioxidantenzyme GPX and CAT and with antioxidant vitamins C, c-tocopherol, b-carotene and retinol

Antioxidant Group N Correlationcoefficient (r)

P

GPX All 141 �0.184 0.029Control men 20 �0.492 0.028

CAT Control 43 �0.356 0.019Control men 20 �0.568 0.009Control smokers 16 �0.755 0.001

Ascorbic acid All 141 �0.170 0.004Control 43 �0.373 0.014Smokers 53 �0.281 0.041Smokers control 16 �0.536 0.032

c-Tocopherol Smokers 53 0.273 0.048Non-smokers 88 0.452 0.001

b-Carotene All 141 �0.190 0.024Exposed 98 �0.248 0.014Men 95 �0.232 0.024Non-smokers 88 �0.221 0.038Men exposed 75 �0.239 0.039Non-smokers exposed 61 �0.276 0.031

Retinol Smokers 53 0.358 0.008



256



Dow

nloaded from


Discussion

Reactive oxygen intermediates together with other proin-flammatory mediators are proposed to be involved in themechanism of action of industrial fibrous dusts. The effect ofoccupational exposure to mineral wool on DNA damage andrepair (measured in lymphocytes) was already published byDusinska et al. (26). Though lymphocytes are surrogate cellsand not target cells, we believe that they reflect overall bodyexposure. In this case, blood circulates to all organs includinglung where mineral fibres deposit. Mineral wool exposure didnot increase levels of oxidized bases in lymphocyte DNA asmeasured with the comet assay, nor induce cytogeneticdamage. Nevertheless, results showed that exposure tomineral wool led to an increase in DNA strand breaks inthe lymphocytes of investigated subjects (Figure 2). Whenanalysed according to sex and smoking habit, this effectwas apparent in the group of non-smokers. DNA repairof oxidative damage did not differ between exposed andcontrols. However, DNA repair capacity in lymphocytes ofexposed subjects was higher in men compared to women.DNA repair capacity was higher in control compared toexposed men (26).

In order to investigate the possible mechanism of mineralwool exposure, we focused in the present paper on markers ofoxidative stress and antioxidant defence and compared themwith oxidative DNA damage and repair. We were alsointerested in which markers of oxidative stress, antioxidantdefence and genetic stability are the most relevant and relatedto exposure to mineral fibres, and therefore, we examinedassociations between different biochemical markers, lipidperoxidation (MDA), markers of antioxidant protection andof genetic stability in exposed and control groups and indifferent subgroups.

We analysed typical biochemical and clinical markers whichcould be related to exposure to mineral wool. Sampling of thesame group and all analyses were done at the same time asalready published results of markers of genetic stability (26). Wemeasured plasma concentrations of vitamin C, a-tocopherol, b-carotene, retinol, xanthophyll and lycopene; FRAP; CPLassayed for its oxidase activity; MDA levels in plasma; andthe activities of antioxidant enzymes, GST, GPX, SOD and CATin erythrocytes.

Effect of exposure and smoking on clinical markers

All 141 volunteers in our study were relatively healthy.However, though exposure to mineral wool was very low, wefound a moderate elevation of systolic blood pressure above thenormal reference range among the exposed group, as well as insmokers and a significantly higher mean pulse frequency insmokers. An interesting association of blood glucose levelswith cholesterol, triglycerides and MDA in all subjects, in allexposed and also in many other subgroups (Table VI) showsthe relation of biochemical and clinical markers. In ourprevious study with diabetes II patients, we found anassociation of markers of oxidative DNA damage with bloodglucose levels (38).

The mean BMI for all investigated subjects, both women andmen, was slightly above the recommended range. Severalclinical and biochemical parameters correlated with BMI suchas blood glucose, MDA, FRAP, FPG and Endo III-sensitivesites. Moreover, blood glucose levels and also FRAP as well asDNA damage were associated with age.

MDA as a marker of lipid peroxidation

MDA is a physiologic ketoaldehyde produced by peroxidativedecomposition of unsaturated lipids as a byproduct ofarachidonate metabolism. The excess MDA produced asa result of tissue injury can combine with free amino groupsof proteins, producing MDA-modified protein adducts (39).The clinical relevance of the reaction between MDA andproteins is highlighted in atherosclerosis, which is a majorcause of coronary heart disease and strokes. Plasma MDAconcentrations are increased in diabetes mellitus, and MDA canbe found in the atherosclerotic lesions (39,40). One conse-quence of oxidative stress and lipid peroxidation is theformation of DNA adducts. Since DNA is believed to be thetarget molecule for carcinogens, endogenous DNA adductsderived from oxidative stress, lipid peroxidation and othersources have been proposed to contribute to the aetiology ofhuman cancers (41).

Our study shows that the plasma level of MDA is slightlyelevated in all exposed groups and significantly in all exposedworkers, as well as in exposed non-smokers (Figure 3). Higheroxidative damage in these groups is possibly the consequenceof significantly suppressed activity of CPL oxidase and CAT inthese groups (Table IV). MDA indeed correlated inversely withCAT activity in all subjects, in non-smokers, in exposed non-smokers and in non-smoking women (Table IX). The activityof GST was also apparently affected by exposure to mineralwool, being relatively suppressed in all exposed subjects, inexposed men and in exposed non-smokers. In contrast, GPXand SOD activities measured in erythrocytes did not differbetween exposed and control workers. There was a negativecorrelation between the activity of GPX and MDA whichsuggests that the higher enzyme activity can protect againstlipid peroxidation. Similar results have been obtained inworkers exposed to nickel (42), cadmium (43), chromium(44) and manganese (45). The results of the present studysuggest that increased plasma lipid peroxidation and decreasederythrocyte antioxidant levels could be used as biomarkers ofoxidative stress in exposed workers. In contrast, antioxidantenzymic activities were not considered a suitable marker forchromium exposure (43–45).

The higher MDA level and suppressed GST, CPL and CATactivities found in our study seem together to indicate elevatedoxidative stress in blood cells of exposed subjects and could besuggested as useful markers of oxidative stress and antioxidantprotection.

FRAP as marker of total antioxidant capacity

The FRAP assay was established by Benzie and Strain (30) asan index of total antioxidant capacity of plasma. It is quick andeasy to perform, and the reaction is reproducible and linearlyrelated to the molar concentration of the antioxidants present.There is also no activity-changing interaction betweenantioxidants in this system. Therefore, FRAP was suggestedas a useful marker to measure antioxidant capacity in cells.This marker was used in several population studies includingours (46,47). Benzie and Strain (30) estimated the per centcontribution of plasma antioxidants to total FRAP (Table XIV)where uric acid represents 60% of total FRAP, ascorbic acid15% and proteins 10%.

In our study, we did not find any difference in FRAP valuesbetween exposed and controls but men had significantly higherlevels of FRAP compared to women in both exposed and controlgroups (Table IV). FRAP correlated strongly with a-tocopherol,


257



Dow

nloaded from


urea and total proteins (Table VIII) in many groups andsubgroups. The correlation with proteins was expected in viewof its 10% contribution to FRAP. The strong correlation of FRAPwith a-tocopherol is also not surprising, though it was reportedas contributing to FRAP only 5% (30). Our results suggest thatperhaps the contribution of a-tocopherol to total FRAP might behigher than 5%. An expected correlation between FRAP andvitamin C was not found in any group or subgroup which mightbe perhaps a result of the negative association we foundbetween urea and vitamin C. FRAP correlates positively withurea (Table VIII). We also found that FRAP inversely correlateswith b-carotene. This also might indirectly result from a negativecorrelation between urea and b-carotene, though we found thisonly in the group of non-smoking controls.

The total antioxidant capacity of plasma (FRAP) surpris-ingly correlated positively with the lipid parameters cholesteroland triglycerides in nearly all investigated subgroups andgroups, and it is therefore not surprising that it positivelycorrelates with BMI in almost all groups and subgroups, aswell as with age (Table VII). We would expect totalantioxidant capacity to correlate negatively with MDA, butthe opposite was true in our study (Table IX). This is likely tobe the consequence of the strong positive correlation betweenFRAP and lipid parameters. Our results clearly show thatFRAP as a marker of total antioxidant capacity in humanpopulation studies cannot be recommended due to theinterference with other biochemical, mainly lipid parameters.

Plasma antioxidants

Concentrations of antioxidant micronutrients (except for c-tocopherol in one group) were not affected by the mineral woolexposure (Table V). To assess the contribution of micronu-trients to antioxidant status, it should be taken into consider-ation that together they can form synergistic complexes in cellsand organisms, e.g. vitamin C can regenerate oxidized vitaminE (48). Therefore, assessment of antioxidant status depends onmeasurement of not just one particular micronutrient but ofa complex battery of antioxidant compounds.

DNA damage and repair: an index of antioxidantprotection—role of GST

DNA damage in lymphocytes is a biomarker of exposure toDNA-damaging agents. The link between DNA damage andthe risk of cancer, and possibly other diseases, has yet to beestablished definitively through epidemiological studies,though there is clearly a mechanism for such a link. Webelieve that oxidative damage measured in lymphocytes can

accurately reflect the overall level of exposure to reactiveoxygen in the body. Many events downstream from the initialevent of DNA damage help to determine whether a particulardamage results in carcinogenic change. One of the events isDNA repair. We measured DNA strand breaks and apurinic/apyrimidinic sites using the alkaline comet assay and sitessensitive to the enzymes FPG- and Endo III, i.e. oxidizedpurines and pyrimidines. We also measured the capacity ofa lymphocyte extract for repair of oxidized guanine.

We found an interesting positive correlation of DNA damagewith cholesterol and triglycerides which suggests that lipidparameters can directly contribute to the induction of oxidativeDNA damage. Similar results were found in rats with highcholesterol diet (Stetina et al., in preparation). On the otherhand, DNA damage (strand breaks, FPG- or Endo III-sensitivesites) did not correlate with plasma levels of MDA. It is notsurprising that both FPG- and Endo III-sensitive sitescorrelated with BMI and age since cholesterol and triglyceridesare associated with BMI (49–51) as we also found in this study.

Several negative associations of both FPG- and EndoIII-sensitive sites with b-carotene and xanthophyll suggestthe protective effect of antioxidant vitamins. On the other hand,a positive correlation with a-tocopherol and retinol in manygroups of investigated subjects, both with FPG and Endo IIIsites (Table XI), suggests that more detailed studies are needed,which include measurement of activities of several antioxidantenzymes together with overall antioxidant status.

A negative correlation of FPG- and Endo III-sensitive siteswith GST levels implies that antioxidant enzymes may play animportant role in protection against oxidative DNA damage.Indeed, we found an inverse correlation between activity ofantioxidant enzyme GST and oxidized purines and pyrimidines(Table X). GST has an important role in detoxification ofxenobiotics, drugs and carcinogens and thus protects the cellsagainst redox cycling and oxidative stress (52,53). Xenobiotic-metabolizing enzymes play a major role in regulating the toxic,oxidative damaging, mutagenic and neoplastic effects ofchemical carcinogens. Mounting evidence has indicated thatthe induction of phase 2 detoxification enzymes such as GSTsresults in protection against toxicity and chemical carcinogen-esis, especially, during the initiation phase. The GSTs area family of enzymes that catalyze the nucleophilic addition ofthe thiol of GSH to a variety of electrophiles (54). It seems thatGST may have also some other role in protection of DNAagainst oxidative DNA damage. Recently, it was found that a-class GSTs bind with the dinitrosyl–diglutathionyl–ironcomplex in rat hepatocytes and that a significant part of thebound complex is also associated with the nuclear fraction(55,56). Stella et al. (56) using confocal and electronmicroscopy reported nuclear localization of GSTs in thesecells. Surprisingly, they found that a considerable amount ofGST corresponding to 10% of the cytosolic pool iselectrostatically associated with the outer nuclear membrane,and a similar quantity is compartmentalized inside the nucleus.Mainly a-class GSTs are involved in this double modality ofinteraction. A quantitative analysis of the membrane-boundaGSTs suggests the existence of a multilayer assembly of theseenzymes at the outer nuclear envelope that could representa considerable novelty in cell physiology. The authorsconclude that the interception of potentially noxious com-pounds to prevent DNA damage could be the possiblephysiological role of the perinuclear and intra-nuclear locali-zation of aGSTs.

Table XIV. Relative activity of individual plasma antioxidants and theirestimated contributions to total FRAP value

Plasmaantioxidant

Relative activity(measured range)

Expected fastingplasmaconcentration(lmol/litre)

Estimated %contributionto total FRAP

Ascorbic acid 2.0 (1.9–2.1) 30–100 15a-Tocopherol 2.0 (1.7–2.1) 15–40 5Uric acid 2.0 (2.0–2.4) 150–450 60Bilirubina 4.0 (4.2–4.6) ,20 5Protein 0.1 (0.1–0.15) 800–1100 10Others – – 5

aBased on albumin-corrected readings [Benzie and Strain (30)].


258



Dow

nloaded from


The interesting positive correlation between DNA repaircapacity of lymphocyte extract (on 8-oxoguanine-containingsubstrate DNA) and MDA, cholesterol and triglyceridesreported here suggests that elevated oxidative damage canstimulate DNA repair (Table XII).

On the contrary, negative correlations between the capacityfor repair of oxidized DNA and GPX and CAT (Table XIII)may suggest that these enzymes protect DNA against oxidativeDNA damage and thus indirectly correlate with DNA repair.This could be true with CAT but not with GPX as we founda positive correlation of the enzyme activity with FPG-sensitivesites (Table X) and a negative correlation with repair rate(Table XIII).

The possible relationship between DNA repair rate andantioxidant micronutrients such as vitamin C, c-tocopherol, b-carotene and retinol (Table XIII) is still confusing. Thenegative correlations of DNA repair rate with ascorbic acidand b-carotene might suggest that the higher the level of thesemicronutrients, the lower the level of DNA damage (as wefound with b-carotene) and, consequently, DNA repair activity.In the case of c-tocopherol and retinol, we found a positiveassociation both with DNA damage (FPG- as well as Endo III-sensitive sites) and also with DNA repair capacity for oxidativeDNA damage. More data are needed to provide a sufficientexplanation.

Conclusion

The results presented indicate that mineral wool exposureinduces an increase in oxidative damage of biomoleculesespecially in the group of male non-smokers. However, theoptimal levels of antioxidants could have a protective effect.We assume that indicators such as MDA, antioxidant enzymesand antioxidant vitamins are useful markers to detect the levelof oxidative stress and antioxidant protection in blood cells. Onthe other hand, FRAP does not seem a good marker of totalantioxidant capacity in population studies as the possibleinterference with other parameters can provide false orconfusing results. Our study also shows that additionally toan antioxidant role, there might also be other mechanisms bywhich antioxidant enzymes (especially GST) protect cellsagainst oxidative DNA damage.

Funding

European Union (project FIBRETOX no. QLK4-1999-01629).

Acknowledgements

We thank all participants in the mineral wool factory (Nova Bana, Slovakia), aswell as the management, for their enthusiastic participation. We thank all staffat the Department of Experimental and Applied Genetics, Slovak MedicalUniversity, Bratislava for help with conducting the study, and with processingof samples. Personal sampling, fibre analysis and medical investigation werecarried out with help of the National Institutes of Health (Banska Bystrica,Nitra and Ziar nad Hronom). We thank Mrs Anna Moravkova and AnnaGaziova, Renata Mateova, Kristına Gaval’ova, Zuzana Rostasova, L’ubicaMikloskova, Jarmila Jantoskova and Viera Machalkova for their excellenttechnical help. We also thank Dr Blazıcek for MDA analyses. Conflict ofinterest statement: None declared.

References

1. Southorn, P. A. and Powis, G. (1988) Free radicals in medicine. I. Chemical

nature and biologic reactions. Mayo Clin. Proc., 63, 381–389.

2. Machun, U. and Bendich, A. (1987) Free radical tissue damage: protectiverole of antioxidant nutrients. FASEB J., 1, 441–445.

3. Niki, E. (1987) Antioxidants in relation to lipid peroxidation. Chem. Phys.Lipids, 44, 227–253.

4. Lorand, S., Lafont, H., Bruneau, M. and Thevemn, M. (1987) Etude desactivites de la superoxyde dismutase et de la glutathion peroxydamrythrocytaire chez les malades athromateux. Acta Pharm. Biol. Clin., 4,372–374.

5. Tanner, A. R., Bantock, I., Hinks, L., Lloyd, B., Turner, N. R. andWright, R. (1986) Depressed selenium and vitamin E levels in a alcoholicpopulation. Possible relationship to hepatic injury through increased lipidperoxidation. Dig. Dis. Sci., 31, 1307–1312.

6. Bjorneboe, G. E., Johnsen, J., Bjorneboe, A., Marklund, S. L., Skylv, N.,Hoiseth, A., Bache-Wiig, J. E., Morland, J. and Drevon, C. A. (1988) Someaspects of antioxidant status in blood from alcoholics. Alcohol Clin. Exp.Res., 12, 806–810.

7. Kubota, S., Sate, N., Matsumura, T. and Kamada, T. (1985) Chemilumi-nescence and superoxide dismutase in the plasma in patients with alcoholicand non-alcoholic liver injuries. Alcohol, 2, 469–472.

8. Ledig, M., Doffoel, M., Doffoel, S., Kopp, P., Bockel, R. and Mandel, P.(1988) Blood cell superoxide dismutase and enolase activities as markers ofalcoholic and nonalcoholic liver diseases. Alcohol, 5, 387–391.

9. Kaji, H., Kurasaki, M., Ito, K. et al. (1985) Increased lipoperoxide valueand glutathione peroxidase activity in blood plasma of type 2 (non insulin-dependent) diabetic women. Klin Wochenschr., 63, 765–768.

10. Inagaki, T., Kamei, Y., Yamanouchi, K., Kate, K. and Sawaki, S. (1989)Serum activities of superoxide dismutase in diabetic patients. J. Clin.Biochem. Nutr., 6, 135–146.

11. Anneren, G., Gebre-Medhin, M., Gustavson, K. H. and Plantin, L. O.(1985) Selenium in plasma and erythrocytes in patients with Down’ssyndrome and healthy controls. Variation in relation to age, sex andglutathione peroxidase activity in erythrocytes. Acta Paediatr. Scand., 74,508–514.

12. Neve, J., Sinet, P. M., Molle, U. and Nicole, A. (1983) Selenium, zinc andcopper in Down’s syndrome (trisomy 21): blood levels and relations withglutathione peroxidase and superoxide dismutase. Clin. Chim. Acta, 133,209–214.

13. Bewick, M., Coutie, W. and Tudhope, G. R. (1987) Superoxide dismutase,glutathione peroxidase and catalase in the red cells of patient withmalignant lymphoma. Br. J. Haematol., 65, 347–350.

14. Gonzales, R., Auclair, C., Voisin, E., Gautero, H., Dhermy, D. andBoivin, P. (1984) Superoxide dismutase, catalase and glutathioneperoxidase in red blood cells from patients with malignant diseases.Cancer Res., 44, 4137–4139.

15. Jeandel, C., Nicolas, M. B., Dubois, F., Nabet-Belleville, F., Penin, F. andCuny, G. (1989) Lipid peroxidation and free radical scavengers inAlzheimer’s disease. Gerontology, 35, 275–282.

16. Neve, J., Vertongen, F., Peretz, A. and Carpentier, Y. A. (1989) Usualvalues for selenium and glutathione peroxidase in a Belgian population.Ann. Biol. Clin., 47, 138–143.

17. McMaster, D., Bell, N., Anderson, P. and Love, A. N. G. (1990)Automated measurement of two indicators of human selenium status, andapplicability to population studies. Clin. Chem., 36, 211–216.

18. Henny, J., Houot, O., Steinmetz, J. et al. (1990) Recueil de l’information,controle de qualite, description de la population etudiee. In Siest, G.,Henny, J. and Schiele, F. (eds), Referennces Biologie Clinique. Elsevier,Amsterdam, pp. 43–56.

19. Loft, S. and Poulsen, H. E. (1996) Cancer risk and oxidative DNA damagein man. J. Mol. Med., 74, 297–312.

20. Moller, P. and Wallin, H. (1998) Adduct formation, mutagenesis andnucleotide excision repair of DNA damage produced by reactive oxygenspecies and lipid peroxidation product. Mutat. Res., 410, 271–290.

21. Collins, A. R., Duthie, S. J. and Dobson, V. L. (1993) Direct enzymicdetection of endogenous oxidative base damage in human lymphocyteDNA. Carcinogenesis, 14, 1733–1735.

22. IARC (1977) Asbestos. IARC Monographs on the Evaluation of theCarcinogenic Risks of Chemicals to Man. IARC Lyon, France, Vol. 14,pp. 1–106.

23. IARC (2002) Man-made Vitreous Fibres. IARC Monographs. IARC. Lyon,France, Vol. 81.

24. Hesterberg, T. W. and Hart, G. A. (2001) Synthetic vitreous fibers: a reviewof toxicology research and its impact on hazard classification. Crit. Rev.Toxicol., 31, 1–53.

25. Kudo, Y., Watanabe, M., Okada, M., et al. (2003) Comparative cytotoxicitystudy of rock wool and chrysotile by cell magnetometric evaluation. Inhal.Toxicol., 15, 1275–1295.


259



Dow

nloaded from


26. Dusinska, M., Barancokova, M., Kazimirova, A., Harrington, V.,Volkovova, K., Staruchova, M., Horska, A., Wsolova, L. and Collins, A.(2004) Does occupational exposure to mineral fibres cause DNA orchromosome damage? Mutat. Res., 553, 103–110.

27. Paglia, D. E. and Valentine, W. N. (1967) Studies on the quantitative andqualitative characterization of erythrocyte glutathione peroxidase. J. Lab.Clin. Med., 70, 158–169.

28. Cavarocchi, N. C., England, M. D., O’Brien, J. F., Solis, E., Russo, P.,Schaff, H. V., Orszulak, T. A., Pluth, J. R. and Kaye, M. P. (1986)Superoxide generation during cardiopulmonary bypass: is there a role forvitamin E? J. Surg. Res., 40, 519–527.

29. Habig, W. H., Pabst, M. J. and Jakoby, W. B. (1974) Glutathione S-transferases. The first enzymatic step in mercapturic acid formation. J. Biol.Chem., 249, 7130–7139.

30. Benzie, I. F. F. and Strain, J. J. (1996) The ferric reducing ability of plasma(FRAP) as a measure of ‘‘antioxidant power’’: the FRAP assay. Anal.Biochem., 239, 70–76.

31. Schosinsky, K. H., Lehmann, H. P. and Beeler, M. F. (1974) Measurementof ceruloplasmin from its oxidase activity in serum by use of o-dianisidinedihydrochloride. Clin. Chem., 20, 1556–1563.

32. Cerhata, D., Bauerova, A. and Ginter, E. (1994) Determination of ascorbicacid in blood serum using high-performance liquid chromatography and itscorrelation with spectrophotometric (colorimetric) determination. CeskaSlov. Farm, 43, 166–168.

33. Hess, D., Keller, H. E., Oberlin, B., Bonfanti, R. and Schuep, W. (1991)Simultaneous determination of retinol, tocopherols, carotenes and lycopenein plasma by means of high-performance liquid chromatography onreversed phase. Int. J. Vitam. Nutr. Res., 61, 232–238.

34. Richard, M. J., Guiraud, P., Meo, J. and Favier, A. (1992) High-performance liquid chromatographic separation of malondialdehyde-thiobarbituric acid adduct in biological materials (plasma and human cells)using a commercially available reagent. J. Chromatogr., 577, 9–18.

35. Collins, A. R. (2004) The comet assay for DNA damage and repair. Mol.Biotechnol., 26, 249–261.

36. Dusinska, M., Collins, A., Kazimirova, A., et al. (2004) Genotoxic effectsof asbestos in humans. Mutat. Res., 553, 91–102.

37. Collins, A. R., Dusinska, M., Horvathova, E., Munro, E., Savio, M. andStetina, R. (2001) Inter-individual differences in DNA base excisionrepair activity measured in vitro with the comet assay. Mutagenesis, 16,297–301.

38. Collins, A. R., Raslova, K., Somorovska, M., Petrovska, H.,Ondrusova, A., Vohnout, B., Fabry, R. and Dusinska, M. (1998) DNAdamage in diabetes: correlation with a clinical marker. Free Radic. Biol.Med., 25, 373–377.

39. Stocker, R. and Keaney, J. F., Jr. (2004) Role of oxidative modifications inatherosclerosis. Physiol. Rev., 84, 1381–1478.

40. Slatter, D. A., Bolton, C. H. and Bailey, A. J. (2000) The importance oflipid-derived malondialdehyde in diabetes mellitus [Review]. Diabetologia,43, 550–557.

41. Ames, B. (1989) Endogenous DNA damage as related to cancer and aging.Mutat. Res., 214, 41–46.

42. Kalahasthi, R. B., Hirehal Raghavendra Rao, R. and Bagalur KrishnaMurthy, R. (2006) Plasma lipid peroxidation and erythrocyte antioxidantsstatus in workers exposed to nickel. Biomarkers, 11, 241–249.

43. Babu, K. R., Rajmohan, H. R., Rajan, B. K. and Kumar, K. M. (2006)Plasma lipid peroxidation and erythrocyte antioxidant enzymes status inworkers exposed to cadmium. Toxicol. Ind. Health, 22, 329–235.

44. Huang, Y. L., Chen, C. Y., Sheu, J. Y., Chuang, I. C., Pan, J. H. andLin, T. H. (1999) Lipid peroxidation in workers exposed to hexavalentchromium. J. Toxicol. Environ. Health A, 56, 235–247.

45. Yiin, S. J., Lin, T. H. and Shih, T. S. (1996) Lipid peroxidation inworkers exposed to manganese. Scand. J. Work Environ. Health, 22,381–386.

46. Kazimirova, A., Barancokova, M., Volkovova, K., Staruchova, M.,Krajcovicova-Kudlackova, M., Wsolova, L., Collins, A. R. andDusinska, M. (2004) Does a vegetarian diet influence genomic stability?Eur. J. Nutr., 43, 32–38.

47. Choi, S. W., Benzie, I. F., Collins, A. R., Hannigan, B. M. and Strain, J. J.(2004) Vitamins C and E: acute interactive effects on biomarkers ofantioxidant defence and oxidative stress. Mutat. Res., 551, 109–117.

48. Fryer, M. J. (1992) The antioxidant effects of thylakoid vitamin E(-tocopherol). Plant Cell Environ., 15, 381–392.

49. Hu, D., Hannah, J., Gray, R. S., Jablonski, K. A., Henderson, J. A.,Robbins, D. C., Lee, E. T., Welty, T. K. and Howard, B. V. (2000) Effects ofobesity and body fat distribution on lipids and lipoproteins in nondiabeticAmerican Indians: the strong heart study. Obes. Res., 8, 411–421.

50. Freedman, D. S., Serdula, M. K., Perey, C. A., Ballew, C. and Whitle, L.(1997) Obesity, levels of lipids and glucose, and smoking among Navajoadolescents. J. Nutr., 127, 2120–2127.

51. Despres, J. P., Moorjani, S., Lupien, P. J., Tremblay, A., Nadeau, A. andBouchard, C. (1990) Regional distribution of body fat, plasma lipoproteins,and cardiovascular disease. Arteriosclerosis, 10, 497–511.

52. Mates, M. (2000) Effects of antioxidant enzymes in the molecular controlof reactive oxygen species toxicology. Toxicology, 153, 83–104.

53. Casalino, E., Sblano, C., Landriscina, V., Calzaretti, G. and Landrascina, C.(2004) Rat liver glutathione S-transferase activity stimulation followingacute cadmium or manganese intoxication. Toxicology, 200, 29–38.

54. Hayes, J. D. and Pulford, D. J. (1995) The glutathione S-transferasesupergene family: regulation of GST and the contribution of theisoenzymes to cancer chemoprotection and drug resistance. Crit. Rev.Biochem. Mol. Biol., 30, 445–600.

55. Pedersen, J. Z., De Maria, F., Turella, P., et al. (2007) Glutathionetransferases sequester toxic dinitrosyl-iron complexes in cells. A protectionmechanism against excess nitric oxide. J. Biol. Chem., 282, 6364–6371.

56. Stella, L., Pallottini, V., Moreno, S. et al. (2007) Electrostatic association ofglutathione transferase to the nuclear membrane. Evidence of an enzymedefense barrier at the nuclear envelope. J. Biol. Chem., 282, 6372–6379.

Received on October 20, 2007; revised on December 21, 2007;accepted on December 21, 2007


260



Dow

nloaded from


Occupational exposure to mineral fibres. Biomarkers of oxidative damage and antioxidant defence and...

Documents

Transcript of Occupational exposure to mineral fibres. Biomarkers of oxidative damage and antioxidant defence and...