genetic polymorphism for glutathione-s- transferase mu in

Occupational and Environmental Medicine 1994;51:812-816

Genetic polymorphism for glutathione-S-transferase mu in asbestos cement workers

K Jakobsson, A Rannug, A-K Alexandrie, L Rylander, M Albin, L Hagmar

AbstractObjective-To investigate whether alack of glutathione-S-transferase mu(GSTMl) activity was related to anincreased risk for adverse outcome afterasbestos exposure.Methods-A study was made of 78 maleformer asbestos cement workers, withretrospective cohort data on exposure,radiographical findings, and lung func-tion. Venous blood samples wereobtained for the analysis ofGSTM1 poly-morphism by the polymerase chain reac-tion technique. Chest x ray films wereclassified according to the InternationalLabour Organisation (ILO) 1980 classifi-cation. Vital capacity (VC) and forcedexpiratory volume during 1 s (FEVy) weredetermined. Individual estimates ofasbestos exposure were calculated, andexpressed as duration of exposure, aver-age exposure intensity, and cumulativedose. Data on smoking were obtainedfrom interviews.Results-The lung function in the studygroup was reduced, compared with refer-ence equations. 23% of the workers hadsmall opacities > 1/0, 29%/o circumscribedpleural thickenings, 14% diffuse thicken-ings, and 12% obliterated costophrenicangles. 54% of the workers were GSTM1deficient. They were comparable with theother workers in age, follow up time(median 30 years), and duration of expo-sure (median 18 years), but had a slightlyhigher cumulated dose (median 18 v 10fibre-years) than the others. Neither inradiographical changes nor lung functionvariables were there any differencesbetween the different GSTM1 groups.The findings were similar when smokinghabits and estimated asbestos exposurewere taken into account.Conclusions-We could not show thatlack of GSTM1 activity was related to anincreased risk for radiographical or lungfiction changes in a group of asbestoscement workers, followed up for a longperiod after the end of exposure.

(Occup Environ Med 1994;5l:812-816)

Keywords: glutathione-S-transferase mu, asbestoscement, lung disease

Although asbestos related disorders areamong the most extensively studied occupa-tional diseases, little is still known about fac-

tors that determine individual susceptibility tothese conditions. As exposure levels decrease,individual characteristics may well assume agreater importance in determining not onlythe actual amount of fibre delivered to thelung tissue and retained there, but also thenature of the response. Peribronchiolar fibrosisis the characteristic early lesion after asbestosexposure, found in both animal and humanstudies.' Based on experimental evidence, ithas been suggested that this is a dose-relatedresponse, whereas individual susceptibilitydetermines the degree of macrophagic alveoli-tis, resulting in subsequent fibrosis.23

Experimental data indicate that active oxy-gen species may be causally involved in thedevelopment of asbestosis.' 4 Data from stud-ies of cell free systems show that active oxygenspecies may be generated by asbestos fibresthemselves. Also, alveolar macrophages thatare activated during the phagocytosis of fibresrelease active oxygen species. At high concen-trations active oxygen species are directlycytotoxic to cells of the respiratory tract; alsothey may cause lipid peroxidation of mem-brane components. Hence, individual differ-ences in activity of enzymes that arescavengers for active oxygen species may bedeterminants of the risk for adverse pul-monary effects.The aim of our study was to investigate

whether the activity of one of these scav-engers, the enzyme glutathione-S-transferasemu (GSTM1), which is polymorphically dis-tributed in humans, was related to individualdifferences in adverse outcome of asbestosexposure. The cytosolic glutathione-S-trans-ferases (GSTs) catalyse glutathione conjuga-tion of electrophilic compounds that can beproduced from exogenous xenobiotics or ariseendogenously during oxidative stress. RatGSTs 3-3 and 4-4, which are homologous tohuman GSTM1 (one of the mu class isoen-zymes of glutathione transferase), are inducedwhen incubated with xanthine and xanthineoxidase in a superoxide producing system.5These enzymes also have glutathione peroxi-dase activity, particularly with DNAhydroperoxides.6 Furthermore, other electro-philic substrates of biological importance likebenzo(a)pyrene-7,8-diol-9, 10-epoxide and 4-hydroxyalkenals are also conjugated with glu-tathione by the rat 3-3 and especially the 4-4isoenzymes.6 Allelic variants of GSTM1 wereassessed in a group of asbestos cement workerswith well characterized individual asbestosexposure, who had been followed up for along period after the end of exposure.

Department ofOccupational andEnvironmentalMedicine, UniversityHospital, S-221 85Lund, SwedenK JakobssonL RylanderM AlbinL HagmarDepartment ofToxicology, NationalInstitute ofOccupational Health,S-171 84 Solna,SwedenA RannugA-K AlexandrieCorrespondence to:Dr Kristina Jakobsson,Department of Occupationaland EnvironmentalMedicine, UniversityHospital, S-221 85 Lund,Sweden.Accepted 18 July 1994

812

on April 1, 2022 by guest. P

rotected by copyright.http://oem

.bmj.com

/O

ccup Environ M

ed: first published as 10.1136/oem.51.12.812 on 1 D

ecember 1994. D

ownloaded from

http://oem.bmj.com/


Table 1 Descriptive data for asbestos cement workers with genotype GSTMI(-) and GSTMJ(+), and non-participants

ParticipantsGSTM1(-) GSTM1(+) Non-participants

Subjects (n) 42 36 17Age (in 1993) 72 (60, 81) 74 (64, 79) 70 (61, 82)Start of employment* 1954 (1938, 1959) 1954 (1945, 1961) 1954 (1945, 1962)End of employment* 1976 (1972, 1976) 1976 (1970, 1977) 1977 (1975, 1977)Employment time (y)* 19 (13, 28) 17 (12, 23) 19 (14, 26)Cumulated dose (f-y)* 18 (8-8, 35) 10 (6-1, 21) 24 (10, 28)Average intensity (f/ml)* 1-0 (0-6, 1-3) 0-6 (0 3, 1-2) 0-9 (0-7, 1-3)Age:

At chest x ray* 62 (51, 73) 66 (55, 70) 61 (50, 70)At lung function test* 69 (57, 79) 68 (60, 75) 62 (46, 72)

Years since start of employment:At chest x ray* 30 (24, 46) 30 (23, 39) 27 (21, 38)At lung function test* 37 (30, 53) 36 (27, 39) 26 (19, 37)

Smoking habits at lung function test:Non-smoker (n(%)) 10 (24) 4 (11) 3 (18)Ex-smoker (n (%)) 21 (50) 17 (47) 8 (47)Smoker (n (%)) 11 (26) 15 (42) 6 (35)Tobacco consumption among

smokers and ex-smokers (pack-y)* 14 (8, 30) 18 (11, 34) 21 (7, 27)

*Median (25, 75 percentiles).

Material and methodsEXPOSED WORKERSAll present and former workers at an asbestoscement plant that had been operating since1907 were invited to a health examination atthe Department of Occupational andEnvironmental Medicine, Lund when theplant closed down in 1977. In 1984 and 1991all employees with at least 10 years of employ-ment were invited to other examinations. Theexaminations included a chest x ray film,spirometry, a work history, and a medical his-tory. We calculated individual cumulativeasbestos doses expressed in fibre-years (f-y).7The average intensity of exposure (f/ml) wasalso estimated, simply by dividing cumulativedose by employment time.On 15 March 1993 there were 95 male

workers with more than 10 years of employ-ment at the plant still alive, who had partic-ipated in at least one of the above mentionedhealth examinations and had had chest x rayfilms classified according to the InternationalLabour Organisation (ILO) 1980 classifica-tion.8 These men formed the study base.Blood samples for determination of geneticpolymorphisms for GSTM1 were obtainedfrom 78 (92%) of them. Table 1 shows demo-graphic data and exposure data for partici-pants and non-participants.

CHEST X RAY FILM FINDINGSThe chest x ray films from these workers hadbeen classified according to the ILO 1980classification by a panel of five readers.9 Themedian reading from the last film classifiedwas used in our study. The profusion of smallopacities regardless of size and shape was con-sidered, as nearly all opacities noted were ofirregular shape. Circumscribed and diffusepleural thickening were noted, regardless ofwidth or extent. Costophrenic angle oblitera-tion was defined as in the ILO standard film.

LUNG FUNCTIONThe results from the last available spirometrywere compared with the reference values mostwidely used in Sweden.'0 As these referenceequations do not take smoking habits into

consideration, we also used reference equa-tions for non-smokers, ex-smokers, andsmokers, obtained from active Swedish con-struction workers (Engholm G, et al, personalcommunication). As our study consists of alarge proportion of old workers, and theexpected values for ages >,70 are based onlinear extrapolation only, we used the spiro-metry results at age .70 in additionalanalyses.

GENOTYPE DETERMINATIONVenous blood samples were collected in tubescontaining sodium citrate and were stored at-70'C. The samples were subjected to lysisand proteinase digestion before extractionwith phenol and chloroform." DNA was iso-lated after ethanol precipation.The analysis of the GSTM1 polymorphism

was performed essentially as described byBrockmoller and collaborators.'2 The poly-merase chain reactions (PCRs) were carriedout in a total volume of 25 ,ul in the presence of10 mM Tris-hydrochloric acid, pH 8.3; 50mM KCI; 0-2 mM of each deoxynucleotidetriphosphate; 1 pM of each primer; 0-8 unitsTaq polymerase; 90 ng genomic DNA as tem-plate; and 1 mM MgCl2. Amplifications were,after an initial denaturation at 940C for 2 min,performed during 35 cycles with denaturationat 940C for 24 s, annealing at 530C for 46 sand extension at 730C for 60 s. Detection ofthe fragments amplified by the PCR wasmade by electrophoresis on a 1-5% agarosegel. All PCRs were carried out in a PCRinstrument (Thermal Cycler 9600) fromPerkin Elmer Cetus.

Deficiency in GSTM1 activity, GSTM1(-),has been ascribed to a homozygous deletion ofthe gene. The PCR assay used in this workdetects the presence or absence of the intactgene, but does not differentiate betweenheterozygous and homozygous carriers. Twoseparate PCRs were performed. The firstprimer set generates an amplification productexceeding exon 4 to 5 if one or both of theGSTM1 alleles are intact. To confirm theresults all samples were further tested with asecond primer set. This amplification yields a

813



.bmj.com

/O

ccup Environ M


ecember 1994. D

ownloaded from

http://oem.bmj.com/

84akobsson, Rannug, Alexandrie, Rylander, Albin, Hagmar

fragment overlapping the first fragment butwhich also includes intron 3. Both fragmentsare absent in GSTM1 ( -) people.

STATISTICSDifferences between subgroups were exam-ined by Fisher's exact test or the Mann-Whitney U test. For radiographic outcomevariables logistic regression was used to esti-mate age (continuous variable), smoking(smoking status at the time of the examina-tion), and exposure adjusted odds ratios(ORs). Duration of employment, averageintensity of employment, and cumulative dose(all variables dichotomised at the median val-ues for the combined group) were examinedin separate models, one at a time. For thelung function variables (expressed as a per-centage of the predicted values, thus adjustedfor age and height) we performed linearregressions, adjusted similarly for exposure.The different statistical methods were per-formed as generally described.13 P valuesgiven are two-tailed. The term significantrefers to P < 0-05 or to a 95% confidenceinterval (95% CI) not including 1 00.

ResultsThe prevalence of genotype GSTM1 (-) was54%. As expected, there were no significantdifferences between the GSTM1 ( -) andGSTM1 ( +) groups in age distribution(table 1). In the group of men aged > 70, thedistribution of GSTM1 genotype was similarto that in younger men (data not shown). Theproportion of lifelong non-smokers wasslightly higher in the GSTM1 ( -) group,24%, compared with 1 % in the GSTM1 (+)group, but the difference in distribution ofsmoking habits (as reported at the time of thelung function test) was not significant (Fisher,P = 0-21). Further, the cumulative tobaccoconsumption among smokers and ex-smokersdid not differ significantly between theGSTM1 groups.The time lapse between start of employ-

Table 2 Radiographic changes and lungfunction among asbestos cement workers withgenotype GSTMI (-) and GSTMJ(+), and among non-participants

Participants

GSTMJ(-) GSTMI(+) Non-partcipantsSmall opacities (n (%)):

<0/1 30 (71) 24 (67) 14 (82)1/0 5 (12) 5 (14) 2 (12)1/1-1/2 7 (17) 7 (19) 1 (6))2/1 0 (0) 0 (0) 0 (0)

Circumscribed pleural thickening (n (%)):Absent 28 (67) 27 (75) 14 (82)Present 14 (33) 9 (25) 3 (18)

Diffuse pleural thickening (n (%)):Absent 37 (88) 30 (83) 13 (76)Present 5 (12) 6 (17) 4 (24)

Costophrenic angle obliteration (n (%)):Absent 38 (90) 31 (86) 16 (94)Present 4 (10) 5 (14) 1 (6)

VC (% predicted)* 79 (70, 89) 78 (69, 83) 79 (72, 86)VC (% predicted, corrected 92 (78, 98) 86 (79, 96) 92 (80, 96)

for smoking habits)*FEVy (% predicted)* 95 (81, 102) 86 (78, 101) 92 (80, 96)FEVy (% predicted, corrected 92 (84, 99) 88 (73, 98) 86 (78, 96)

for smoking habits)*

*Median (25, 75 percentiles)

ment and time of examination did not differbetween the groups. The calendar years forthe start and end of employment, and thusalso the duration of employment, were alsocomparable (table 1). The average intensity ofexposure had been somewhat higher in theGSTM1 ( -) group, median 1-0 f/ml, com-pared with 0-6 f/ml in the GSTM1(+) group(Mann-Whitney, P = 0 05), and thus also thecumulative dose, median 18 v 10 f-y (Mann-Whitney, P = 0 02). In the GSTM1(-)group the proportion of men with an esti-mated cumulative dose >20 f-y was 45%,compared with 21% in the GSTM1(+)group (Fisher, P = 0 24).The distribution of radiographic abnormal-

ities was remarkably similar between the dif-ferent GSTM1 groups (table 2). Neither forparenchymal abnormalities nor for pleuralabnormalities (circumscribed and diffusepleural thickening, obliteration of costo-phrenic angles) were there any differences.The differences in exposure levels betweenthe GSTM1 groups were accounted for in thelogistic regression model, but still showed nosignificant differences between the GSTM1groups for any of the radiographic changes.The odds ratio (OR) for small opacities ofprofusion >1/10 in men with GSTM1(-)was 0-8 (95% CI 0-3-2-1), with cumulativedose as the exposure variable.The two different reference equations used

gave somewhat different estimates of vitalcapacity (VC) (expressed as a percentage ofthe predicted value). There were no consis-tent differences between the GSTM1 groupsregardless of which reference equation wasused, neither when we analysed VC obtainedfrom the last examination (table 2), nor whenwe used lung function data obtained at age< 70 (data not shown).The forced expiratory volume in one second

(FEV,) was slightly higher in the GSTM1 (-)group than in the GSTM1 (+) group, but thedifferences were not significant (table 3),regardless of whether we used reference equa-tions not corrected for smoking (Mann-Whitney, P = 0 21) or corrected for smoking(Mann-Whitney, P = 0-41). The results foranalyses with age limits were similar. The dif-ferences in exposure levels were accounted forin the multiple linear regression model, butshowed no significant differences between theGSTM1 groups for either VC or FEVI.

Non-participants had a slightly, but not sig-nificantly, higher cumulative dose than partic-ipants (table 1). Radiographic and lungfunction changes were no more severe than inthe participants (table 2).

DiscussionThe question whether the biological responseto environmental exposure is modulated byhost polymorphism is currently under exten-sive investigation. The enzymes of interest inthe context of asbestos exposure include theGSTs, which may be involved in the inactiva-tion of active oxygen species. Preliminaryresults have been reported that suggested an

814



.bmj.com

/O

ccup Environ M


ecember 1994. D

ownloaded from

http://oem.bmj.com/


over-representation of GSTM1 (-) genotypein a sample of people with asbestos relateddisease, compared with controls.'4 Studies inpopulations defined by their asbestos expo-sure are to our knowledge not previouslyreported.Our population was defined on the basis of

exposure. The population studied-that is,those with available data on outcome-com-prised 78% of all living workers with 10years of former employment at the plant.Further, the non-participants did not consti-tute a group with more severe asbestos relateddisease than the participants. Thus we canassume that the participants are fairly repre-sentative of the surviving exposed population.The outcome was studied after the end ofexposure, and for all workers > 20 years afterthe start of employment. Thus, we alsoassume that sufficient time had elapsed for thedevelopment of asbestos related disease in ourstudy group.The prevalence of genotype GSTM (-)

was 54%, which is somewhat higher than anearlier study on 70 chimney sweeps, whereGSTM1(-) people comprised 49%.*516 In aprevious estimate of the proportion ofGSTM1(-) people in Sweden, 54% in agroup of 248 examined had a phenotype lack-ing activity towards trans-stilbene oxide.'7We could not show any radiographical and

lung function differences between men withthe genotypes GSTM1(-) and GSTM1(+).The investigated subjects were survivors from apopulation exposed to asbestos cement. Inthis population we have found increased risksof respiratory cancer (including mesothe-liomas), non-malignant respiratory mortality,7and a somewhat shorter life span comparedwith a reference cohort of industrial workers. 18The population investigated might thus theo-retically display an under-representation ofpeople with GSTM1(-), were this genotypestrongly associated with premature deaths(related or not with asbestos exposure). Theoverall prevalence of GSTM1(-) was notlower than expected. Further, we did not findany significantly different distribution ofgenotypes in older than younger workers, andno overall correlation between age and geno-type. Thus, a selective survival bias is not alikely explanation of the present lack of anassociation between genotype and adversepulmonary and pleural outcome.Of the human cytosolic GSTs the GSTA,

GSTM and GSTP forms have all been identi-fied in lung tissue.'9 The GSTM1 enzymeprobably contributes to only a minor part ofthe total GST activity in the lung but it maystill be crucial for the detoxification of poten-tially toxic substrates in certain cell popula-tions in the lung. Increased levels of aromaticDNA adducts found in lung tissue ofGSTMI ( -) people20 and the increased risk oflung cancer of GSTMI(-) smokers2'-23 sup-port the theory that GSTMI has an importantprotective role in the lung.20 The susceptibilityto lung cancer in smokers of the GSTM1 (-)genotype has been most firmly established insmokers with a low tobacco consumption24

and also in lung cancer patients diagnosed at ayoung age.25The relative importance of GST activity in

parenchymal and bronchial lung tissue is notyet clarified. The results from lung cancerstudies are conflicting; some studies correlateGSTM1(-) with an increased risk primarilyfor adenocarcinomas,22 25 whereas other stud-ies report an association with squamous cellcarcinomas only.2627 It has been found thatparenchymal lung preparations have higherconcentrations of reduced glutathione thanbronchial preparations.28 Thus, there is reasonto assume that individual differences in GSTactivity may influence the risk of developmentof a mainly parenchymal disease likeasbestosis.

Despite the fact that the GSTMl(-) menhad a somewhat higher asbestos exposurethan the GSTM1 (+) men, no difference inoutcome was found. From the results it canbe calculated2930 that there is a probability of0l10 for a .50% increase in the relative riskof small opacities 1/0 in the GSTM1(-)group. Thus, although our study group was ofa moderate size only, the power was reason-able. The corresponding probability estimatesfor circumscribed pleural thickenings andobliterated costophrenic angles were 0 50 and0 13. In our study, however, the exposurelevels for most workers were low or moderate,and only a small fraction of the men had radi-ographic changes of ILO grade >.1/1. Truedifferences between GSTM1 groups for moresevere parenchymal disease may well exist,but this cannot be evaluated within our studygroup.

We thank Anita Nilsson, RN, for obtaining the blood samples.Goran Engholm, Gosta von Schmalensee, Kurt Frostrom, andthe Construction Industry Foundation for Industrial Safetyand Health for kind permission to use spirometry referenceequations obtained from active construction workers. We alsothank Dr Margareta Warholm, National Institute forOccupational Health for helpful discussions. The study wassupported by the Swedish Work Environment Fund and theFaculty of Medicine, Lund University.

1 Rom WN, Travis WD, Brody AR. Cellular and molecularbasis of the asbestos-related diseases. Am Rev Respir Dis1991;143:408-22.

2 Begin R, Masse S, Sebastien P, et al. Asbestos exposureand retention as determinants of airway disease andasbestos alveolitis. Am Rev Respir Dis 1986;134:1176-81.

3 Begin R, Masse S, Sebastien P. Alveolar dust clearancecapacity as determinant of individual susceptibility toasbestosis: new experimental observations. Ann OccupHyg 1989;33:279-82.

4 Mossman BT, Bignon J, Corn M, Seaton A, Gee JBL.Asbestos: scientific developments and implications forpublic policy. Science 1990;247:294-301.

5 Murata T, Hatayama S-i, Satoh K, Tsuchida S, Sato K.Activation of rat glutathione transferases in class mu byactive oxygen species. Biochim Biophys Res Commun1990;171:845-51.

6 Tan KH, Meyer DJ, Coles B, Ketterer B. Thyminehydroperoxide, a substrate for rat Se-dependent glu-tathione peroxidase and glutathione transferase isoen-zymes. FEBS Lett 1986;207:231 -3.

7 Albin M, Jakobsson K, Attewell R, Johansson L, WelinderH. Mortality and cancer morbidity in cohorts of asbestoscement workers and referents. Br J Ind Med 1990;47:602-10.

8 International Labour Office. Guidelines for the use of ILOinternational classification of radiographs ofpneumoconiosis.Geneva: ILO, 1980. (Occupational safety and healthseries No 22 revised.)

9 Jakobsson K, Stromberg U, Albin M, Welinder H,Hagmar, L. Radiological changes in asbestos cementworkers. Occup Environ Med 1995; (in press).

815



.bmj.com

/O

ccup Environ M


ecember 1994. D

ownloaded from

http://oem.bmj.com/

Jakobsson, Rannug, Alexandrie, Rylander, Albin, Hagmar

10 Berglund E, Birath G, Bjure J, et al. Spirometric studies innormal subjects. Forced expirograms in subjectsbetween 7 and 70 years of age. Acta Med Scand 1963;173:185-98.

11 Gustafson S, Proper JA, Bowie EJW, Sommer SS.Parameters affecting the yield of DNA from humanblood. Anal Biochem 1987;165:294-9.

12 Brockmoller J, Gross D, Kerb R, Drakoulis N, Roots I.Correlation between trans-stilbene oxide glutathioneconjugation activity and the deletion mutation in theglutathione S-transferase class mu gene detected bypolymerase chain reaction. Biochem Pharmacol 1992;43:647-50.

13 Checkoway H, Pearce NE, Crawford-Brown DJ. Researchmethods in occupational epidemiology. Oxford: OxfordUniversity Press, 1989.

14 Smith CM, Christiani DC, Kelsey KT. Glutathione-S-transferase mu phenotype and genotype in workerswith asbestos-related lung disease [abstract]. In:Proceedings from 9th International Symposium onEpidemiology in Occupational Health. Sept 23-25.Cincinnati: 1992:250.

15 Carstensen U, Alexandrie A-K, Hogstedt B, Rannug A,Bratt I, Hagmar L. B- and T-lymphocyte micronuclei inchimney sweeps with respect to genetic polymorphismfor CYPlA1 and GST1 (class mu). Mutat Res1993;289: 187-95.

16 Ichiba M, Hagmar L, Rannug A, et al. Aromatic DNAadducts, micronuclei and genetic polymorphism forCYPlA1 and GST1 in chimney sweeps. Carcinogenesis1994;15: 1347-52.

17 Seidegird J, Pero RW. The hereditary transmission of highglutathione transferase activity towards trans-stilbeneoxide in human mononuclear leucocytes. Hum Genet1985;69:66-8.

18 Albin M, Horstmann V, Jakobsson K. Survival in cohorts ofasbestos cement and referent workers. In Albin M:Asbestos exposure and risk among asbestos cement workers.Studies of exposure and fibre concentrations, retention pat-terns and histopathological changes in lung tissue, and cancerrisk, mortality and survival. Lund, Sweden: LundUniversity, 1992. (Thesis.)

19 Singhal S, Saxena M, Ahmad H, Awasthi S, Haque A,Awasthi Y. Glutathione S-transferase of human lung:characterization and evaluation of the protective role of

the alpha-class isozymes against lipid peroxidation. ArchBiochem Biophys 1992;299:232-41.

20 Shields PG, Bowman ED, Harrington AM, Doan VT,Weston A. Polycyclic aromatic hydrocarbon-DNAadducts in human lung and cancer suspectibility genes.Cancer Res 1993;53:3486-92.

21 Seidegard J, Pero RW, Miller DG, Beattie EJ. A glu-tathione transferase in human leukocytes as a marker forthe susceptibility for lung cancer. Carcinogenesis 1986;7:751-3.

22 Seidegard J, Pero RW, Markowitz MM, Roush DG, MillerDG, Beattie EJ. Isoenzyme(s) of glutathione transferase(class mu) as a marker for the susceptibility to lung can-cer: a follow up study. Carcinogenesis 1990;11:33-6.

23 Nazar-Stewart V, Motulsky AG, Eaton DL, White E,Hornung SK, Leng ZT, et al. The glutathione S-trans-ferase-mu polymorphism as a marker for susceptibilityfor lung carcinoma. Cancer Res 1993;53:2313-8.

24 Nakachi K, Imai K, Hayashi S-i, Kawajiri K.Polymorphisms of the CYPlAl and glutathione S-trans-ferase genes associated with susceptibility to lung cancerin relation to cigarette dose in a Japanese population.Cancer Res 1993;53:2994-9.

25 Alexandrie A-K, Ingelman Sundberg M, Seidegard J,Tornling G, Rannug A. Genetic susceptibility to lungcancer with special emphasis on CYPIA1 and GSTM1: astudy on host factors in relation to age at onset, gender,and histological cancer type. Carcinogenesis 1994.

26 Zhong S, Howie AF, Ketterer B, et al. Glutathione-S-transferase mu locus: use of genotyping and phenotypingassays to assess association with lung cancer susceptibility.Carcinogenesis 1991;12:1533-7.

27 Hirvonen A, Husgafvel-Pursiainen K, Anttila S, Vainio H.The GSTM1 null genotype as a potential risk modifierfor squamous cell carcinoma of the lung. Carcinogenesis1993;14:1479-81.

28 Petruzxelli S. De Flora S, Bagnasco M, et al. Carcinogenmetabolism studies in human bronchial and lungparenchymal tissues. Am Rev Respir Dis 1989;140:417-22.

29 Smith AH, Bates MN. Confidence limits analysis shouldreplace power calculations in the interpretation of epi-demiologic studies. Epidemiology 1992;3:449-52.

30 Stromberg U. Post-study probability of effect at least equalto a given value [letter]. Epidemiology 1994;5:477.

Vancouver styleAll manuscripts submitted to Occup EnvironMed should conform to the uniformrequirements for manuscripts submitted tobiomedical journals (known as theVancouver style.)

Occup Environ Med, together with manyother international biomedical journals, hasagreed to accept articles prepared in accor-dance with the Vancouver style. The style(described in full in the BMJ, 24 February1979, p 532) is intended to standardiserequirements for authors.

References should be numbered consec-utively in the order in which they are firstmentioned in the text by Arabic numeralsabove the line on each occasion the refer-ence is cited (Manson' confirmed otherreports2-5 . . .). In future references topapers submitted to Occup Environ Med

should include: the names of all authors ifthere are seven or less or, if there are more,the first six followed by et al; the title ofjournal articles or book chapters; the titlesof journals abbreviated according to thestyle of Index Medicus; and the first and finalpage numbers of the article or chapter.Titles not in Index Medicus should be givenin full.

Examples of common forms of refer-ences are:

1 International Steering Committee of Medical Editors,Uniform requirements for manuscripts submitted tobiomedical journals. BrMedJ 1979;1:532-5.

2 Soter NA, Wasserman SI, Austen KF. Cold urticaria:release into the circulation of histamine and eosino-phil chemotactic factor of anaphylaxis during coldchallenge. N EnglJ Med 1976;294:687-90.

3 Weinstein L, Swartz MN. Pathogenic properties ofinvading micro-organisms. In: Sodeman WA Jr,Sodeman WA, eds. Pathologic physiology, mechanismsof disease. Philadelphia: W B Saunders, 1974:457-72.

816



.bmj.com

/O

ccup Environ M


ecember 1994. D

ownloaded from

http://oem.bmj.com/

genetic polymorphism for glutathione-s- transferase mu in

Documents