Human glutathione S-transferase | My Assignment Tutor

Carcinogenesis vol.19 no.2 pp.275–280, 1998Human glutathione S-transferase P1 polymorphisms: relationshipto lung tissue enzyme activity and population frequencydistributionMary A.Watson1, Richard K.Stewart2, GraemeB.J.Smith2, Thomas E.Massey2 and Douglas A.Bell1,31Laboratory of Computational Biology and Risk Assessment, NationalInstitute of Environmental Health Sciences, PO Box 12233, ResearchTriangle Park, NC 27709, USA and 2Department of Pharmacology andToxicology, Queen’s University, Kingston, Ontario, K7L 3N6 Canada3To whom correspondence should be addressedThe association between glutathione S-transferase (GST)activity as measured by 1-chloro-2,4-dinitrobenzene(CDNB) conjugation and genotype at exon 5 and exon 6of the human GSTP1 gene was investigated in normal lungtissue obtained from 34 surgical patients. These sampleswere genotyped for previously identified polymorphisms inexon 5 (Ile105Val) and exon 6 (Ala114Val) by PCR-RFLPand direct sequencing. GST enzyme activity was significantly lower among individuals with the 105 Val allele.Homozygous Ile/Ile samples (n J 18) had a mean cytosolicCDNB conjugating activity of 74.9 K 3.8 nmol/mg permin; heterozygotes (n J 13) had a mean specific activityof 62.1 6 4.2 nmol/mg per min and homozygous Val/Val(n J 3) had a mean specific activity of 52.5 K 4.5 nmol/mg per min. The CDNB conjugating activity measured forthe Ile/Ile genotype group was significantly different fromthat observed in the Ile/Val group (P J 0.03), and fromIle/Val and Val/Val genotypes combined (P J 0.009). MeanGST activity values were consistently lower in individualswith genotypes containing the 105 valine allele, regardlessof smoking exposure. Genotypes at codon 114 were alsoassessed but the mean GST activity was not significantlylower in individuals with the 114 valine allele. A newhaplotype, present in two samples who were homozygous105Ile and had a 114Val, was identified and proposedas GSTP1*D. Frequencies of the exon 5 and exon 6polymorphisms were determined in samples obtained fromEuropean-Americans, African-Americans and Taiwanese.The differences observed were highly significant suggestingthe possibility of GSTP1 genotype-associated, ethnic differences in cancer susceptibility and chemotherapeuticresponse.IntroductionGlutathione transferases (GSTs*) consist of a super-family ofdimeric phase II metabolic enzymes that catalyze the conjugation of reduced glutathione with electrophilic groups of a widevariety of compounds. In general, the reactions catalyzed byGSTs are considered detoxifying, and serve to protect cellularmacromolecules from damage caused by cytotoxic and carcino-*Abbreviations: Ile, isoleucine; Val, valine; GST, glutathione S-transferase;CDNB, 1-chloro-2,4-dinitrobenzene; PCR, polymerase chain reaction; RFLP,restriction fragment length polymorphism; SSCP, single-strand conformationpolymorphism; GSTP1-1a, 105 isoleucine form of the GSTP1 protein, GSTP1-1b, 105 valine form of the GSTP1 protein; BSA, bovine serum albumin.© Oxford University Press 275genic agents (1,2). In some cases, however, glutathione conjugation can result in chemical intermediates that may bemore reactive than the parent compounds (3). Biochemicalcharacteristics separate the cytosolic GST isozymes into fourmajor classes: alpha, mu, theta and pi (4). Different classesof GSTs have distinct substrate specificities, although somesubstrates, such as 1-chloro-2,4-dinitrobenzene (CDNB), areconjugated by several classes (2). GST composition variesamong tissues and the particular combination of GSTsexpressed in a tissue may influence detoxifying capability (2).There is evidence to suggest that GSTs are involved in themetabolism of chemotherapeutic agents, and that they areover-expressed in tumors that are refractory to treatment (5).GST isoforms have been shown to be polymorphic (GSTM1and GSTT1) resulting in reduced enzyme activity (6–8), andmay confer an increased risk for developing specific types ofcancer (9–13).GSTP1-1 is widely expressed in normal human epithelialtissue (14) and is particularly abundant in the lung, esophagusand placenta (15). As much as 10-fold inter-individual variationin GST activity has been reported in normal and tumor tissues(16,17). GSTP1-1 is commonly over-expressed in tumors, andelevated levels have been found in stomach, colon, bladder,oral, breast, skin and lung tumors compared with normalmatched tissues (15–23). In some cancer model systems,GSTP1-1 expression is considered a pre-neoplastic tumormarker (24). Increased levels of GSTP1-1 in tumors mayaccount for part of the inherent drug resistance observed inmany tumors, although the mechanism remains unknown (5).This evidence suggests that genetic polymorphisms in GSTP1could be an important factor in cancer etiology and therapy.Polymorphisms in exon 5 (Ile105Val) and exon 6 (Ala114-Val) of GSTP1 were first reported by Board et al. (25) andthese changes appear to be within the active site of the GSTP1-1 protein (26,27). In vitro cDNA expression studies suggestthat the Ile105Val substitution reduces enzyme activity (26).However, no studies have compared enzyme activities in tissuesamples with the presence of these polymorphisms. In thisstudy, we investigated whether polymorphisms within exons5 and 6 of the GSTP1 gene could account for the interindividual variation seen in GST activity in normal lung tissue.We initially screened for the exon 5 polymorphism by usingsingle-strand conformation polymorphism (SSCP) analysisand then developed a polymerase chain reaction-restrictionfragment length polymorphism (PCR-RFLP) method for rapiddetection. We report the association of GSTP1 polymorphismswith lower CDNB conjugating activity in human lung tissueand provide data on the frequency of GSTP1 polymorphismamong samples collected from individuals with European,African and Asian ancestry.Materials and methodsSubjectsSections of peripheral lung tissue (13–108 g), devoid of macroscopicallyvisible tumors, were obtained from surgical patients at Kingston GeneralM.A.Watson et al.Table I. Primer sequencesExon 5 sequencing PiF2537 CCA ACC CCA GGG CTC TAT Gand SSCP PiR205 GGG GTG AGG GCA CAA GAExon 5 RFLP PiF2306 GTA GTT TGC CCA AGG TCA AG(Ile105Val) PiR2721 AGC CAC CTG AGG GGT AAGExon 6 RFLP PiF3402 GGG AGC AAG CAG AGG AGA AT(Ala114Val) PiR3800 CAG GTT GTA GTC AGC GAA GGA GHospital, Kingston, Ontario, Canada, during clinically indicated lobectomy.Samples were obtained following informed consent and in accord with Queen’sUniversity Research Ethics Board guidelines. In order to minimize any lossof GSTP1-1 activity, immediately after removal, each tissue section wasplaced in 0.9% NaCl and kept on ice. Time between surgical resection andtissue processing was ~20 min. Tissues were homogenized and cytosolsprepared by differential centrifugation as described previously (28). Cytosolicfractions were then snap frozen in liquid N2 and stored at –70°C until use.Following thawing, protein concentrations were measured by the method ofLowry et al. (29), and glutathione S-transferase (GST)-catalyzed conjugationof 1-chloro-2,4-dinitrobenzene (CDNB) (Sigma Corporation, St Louis, MO)was measured as described by Habig et al. (30). DNA was isolated fromfrozen sections of peripheral lung by proteinase digestion followed byphenol:chloroform extraction and ethanol precipitation (31). Patients werecharacterized with respect to diagnosis leading to surgery, drug treatment priorto surgery, smoking status, possible occupational exposure to carcinogens,sex, age and race. Patients were considered former smokers if smokingcessation was .2 months prior to surgery. This period coincides with thetime others have found necessary for loss of smoking-related effects on GSTand cytochrome P450-mediated biotransformation activities in human lungtumors (32,33).Blood samples from African-American and European-American subjectswere obtained from a community-based sample of 424 healthy, unrelatedvolunteers from Durham and Chapel Hill, North Carolina, who were recruitedby newspaper advertisement. Taiwanese DNA samples were obtained from116 unrelated full-term maternity patients with uncomplicated pregnancies atthe Chang Gung Memorial Hospital, Taiwan and were kindly provided by DrL.L.Hsieh. These samples were extracted from placental tissue as describedin Hsieh and Hsieh (34). The population samples have been utilized in severalstudies from this laboratory and are used only for estimating approximategene frequencies (9,13,35,36). Comparison of gene frequency data (GSTM1,NAT2, CYP2E1) derived from these samples with other published andunpublished data on European-American, African-American and Asian populations suggest that in general, these population samples display gene frequenciesthat approximate regional, national and international trends.All samples were obtained with informed consent and under NIH approvedhuman subject protocols. DNA was extracted from whole blood with conventional phenol:chloroform extraction methodology following lysis and nucleipelleting using the Applied Biosystem Inc. protocol on the ABI 340A DNAExtractor (ABI, Foster City, CA).Single-strand conformation polymorphismUsing published sequence data, primers were designed using the Oligoprogram (National Biosciences, Inc., Plymouth, MN) to amplify exon 5 ofthe GSTP1 gene. Two consecutive PCR reactions were performed. Briefly,100 ng of genomic DNA from frozen sections of peripheral lung was addedto a mix containing 25 pmol of primers PiF2537 and PiR205 (Table I), 200µM deoxynucleoside triphosphates, 1 U of Taq polymerase (Promega Corp.,Madison, WI), 1.6 mM MgCl2, and a PCR buffer containing 16.6 mM (NH4)2SO4, 50 mM β-mercaptoethanol, 6.8 µM EDTA, 67 mM Tris (pH 8.8) and80 µg/ml bovine serum albumin (BSA), in a final volume of 30 µl. A ‘hotstart’ was used to prevent non-specific priming in the first cycle of PCR.Following an initial denaturation step at 94°C for 3 min, five cycles of PCRwere carried out (cycle 1: 94°C for 15 s, 64°C for 30 s, 72°C for 60 s) inwhich the annealing temperature decreased by 1°C each cycle. This wasfollowed by 25 cycles of amplification at 94°C for 15 s, 59°C for 30 s and72°C for 1 min in a Perkin–Elmer 9600. Aliquots of 1 µl of the initial PCRreaction were added to a subsequent PCR containing 25 pmol of each SSCPprimer, 2 µCi [α-33P]dATP (Amersham Corp., Arlington Heights, IL), 2 µMdNTPs, 2.5 mM MgCl2, 1 U Taq Polymerase and (NH4)2SO4 PCR buffer ina final volume of 20 µl and amplified for an additional 15 cycles at 94°C for15 s, 59°C for 30 s and 72°C for 1 min. Products were diluted and added toa formamide stop solution, heat denatured, placed on ice and loaded on to a6% acrylamide gel containing 10% glycerol. Samples were electrophoresedfor 5.5 h at 35 W and 4°C. The gel was dried and autoradiographed overnightat room temperature.276Fig. 1. SSCP analysis of exon 5 revealed three distinct fragment patterns.Direct sequencing of these samples determined samples 2 and 3 wereheterozygous and sample 7 was homozygous for a G→A transition atnucleotide 313 of GSTP1.SequencingPhenotyped samples screened by SSCP were directly sequenced in order toelucidate the basis for observed conformational changes detected. Sampleswere amplified prior to sequencing using primers PiF2537 and PiR205 forexon 5 (Table I). Cycling parameters were the same used in the SSCP protocolprior to the radioactive amplification, and both sense and anti-sense strandswere sequenced. The PCR fragment from exon 6 was sequenced using theforward primer for the PCR-RFLP (PiF3402). The PCR product was purifiedwith a Microcon 100 microconcentrator (Amicon Corp., Beverly, MA) and 5µl was added to a sequencing mix and cycle sequenced using ABI PrismAmpliTaq DNA Polymerase FS (Perkin-Elmer Corp., Foster City, CA).Samples were again purified using Centri-sep columns (Princeton Separations,Adelphia, NJ) and analyzed on an ABI 373 Stretch Sequencer (ABI, FosterCity, CA).PCR-RFLP analysisPCR was performed using the primers PiF2306/PiR2721 for exon 5 (Ile105Val),and PiF3402/PiR3800 for exon 6 (Ala114Val). The same cycling parameterswere used as in the non-radioactive amplification of the SSCP protocol.Following PCR, the entire sample was digested for 2 h at 37°C with 5 unitsof Alw26I for exon 5, or AciI for exon 6 (New England Biolabs, Beverly,MA). Approximately 15 µl of the digest was electrophoresed on a 3% 3:1NuSieve:agarose gel (FMC Bioproducts, Rockland, ME) containing ethidiumbromide. While digestion with AciI identifies the polymorphism in exon 6(Ala114Val), digestions with this enzyme were inconsistent. Optimized conditions could not be obtained and results reported for Ala114Val (Table IV)were confirmed by direct sequencing.StatisticsDifferences between means of two groups were tested using Student’s t-test.For comparisons between group data that did not display normal distributions,non-parametric analysis was performed by Mann–Whitney rank sum test (SASstatistical package, SAS, Cary, NC). Analysis of variance (two-way) wascarried out to test the relationship between smoking history (current versusformer), GSTP1 genotype and CDNB conjugating activity (SAS). Genotypeand allele frequency differences were tested by chi-square analysis (SAS).Results and discussionSSCP screening of GSTP1 exon 5 was undertaken at the outsetof the study because the precise nature of the polymorphismwas unclear from the published literature and the genomedatabase. SSCP analysis revealed three distinct fragment patterns for exon 5 of GSTP1 (Figure 1). From direct sequencingof the samples that exhibited mobility shifts, it was determinedthat samples 2 and 3 were heterozygous and sample 7 washomozygous for a G→A transition at nucleotide 313 ofGSTP1. This change represents an amino acid substitutionfrom isoleucine to valine at codon 105; as previously reportedby Board et al. (25) and by Zimniak et al. (26). However, inthe latter study the change was stated to be at codon 104. Thispolymorphism is located at an Alw26I restriction site andunambiguous identification of variants can be achieved byPCR-RFLP, as seen in Figure 2. The results obtained with thepresent method are equivalent to a recently published method(37); however, the approach used here provides a control cutsite for the Alw26I restriction enzyme.Table II shows mean GST activity after grouping samplesby Ile105Val and Ala114Val genotype. GST enzyme activitiesin lung tissue were significantly lower among individualsHuman glutathione S-transferase P1 polymorphismsFig. 2. PCR-RFLP patterns of the Exon 5 amplified fragment resulted in aband of 113 bp in all samples, which represents a control cut forconfirmation of proper digestion. In the wild-type sequence (Ile/Ile), bandsof 329 bp and 113 bp were generated, whereas in the homozygous mutant,(Val/Val), bands at 216 bp, 113 bp and 107 bp were produced. In theheterozygous Ile/Val, all four bands were present.Table II. CDNB conjugating activity in lung tissue cytosols in relationshipto GSTP1 genotypesExon 5 Specific activity Exon 6 genotype Specificgenotype mean (SEM) (Ala114Val) activity mean(Ile105Val) (SEM)Ile/Ile 74.9 (3.8)a Ala/Ala 70.5 (2.8)b(n 5 18) (n 5 27)Ile/Val 62.1 (4.2) Ala/Val 59.2 (10.3)(n 5 13) (n 5 6)Val/Val 52.5 (4.5) Val/Val 55.2(n 5 3) (n 5 1)aIle105Val tests, Ile/Ile versus Ile/Val, P 5 0.014; Ile/Ile versus Val/Val,P 5 0.05; Ile/Ile versus pooled Ile/Val and Val/Val, P 5 0.005 (all Mann–Whitney rank sum test).bAla114Val tests, Ala/Ala versus Ala/Val, P 5 0.36; Ala/Ala versus Ala/Val1 Val/Val, P 5 0.22 (all Mann–Whitney rank sum test).with the 105 Val allele. The mean activities for the Ile/Valheterozygote group (62.1 6 4.2 nmol/mg per min, P 5 0.03)and Ile/Val and Val/Val genotypes combined (60.3 1 3.6 nmol/mg per min, P 5 0.009) were significantly lower than for Ile/Ile genotypes (74.9 6 3.8 nmol/mg per min).Zimniak et al. (26) used Escherichia coli expression tocompare Ile 105 and Val 105 alleles and found reduced (~30%)CDNB conjugating activity for the 105Val allele, and a recentpaper by Ali-Osman and colleagues reproduced this finding(27). The reduced CDNB activities we observe in lung tissuesamples with genotypes containing 105Val were consistentwith these findings. Of note, heterozygote genotypes had lowermean CDNB conjugating activity than Ile/Ile homozygotes.This suggests the possibility that GSTP1a-1b heterodimers,which would be present in the cells of those with heterozygotegenotypes, may have reduced activity.Although CDNB is not a class specific GST substrate, alarge proportion of human lung CDNB activity has beenattributed to GSTP1-1 (38). Contribution to CDNB activity inthe lung from other GST enzymes such as GSTM3 or GSTA1/A2 would tend to dilute the GSTP1 genotype effect that wehave observed. Thus the GSTP1 105Val genotype might havegreater impact on other GSTP1-1 substrates. Interestingly,Zimniak et al. (26) found that the GSTP1-1 enzyme containing105Val had higher rather than lower activity for ethacrynic277Table III. Smoking: GSTP1 genotype and CDNB activityExon 5 GST Specific activity mean 6 SEM of smoking statusgenotypeNon-smoker (n) Former (n) Current (n)All genotypes 77.9 6 8.8 (2) 70.8 6 5.5 (9) 58.9 6 3.3 (23)Ile/Ile 85.1 (1) 61.5 6 4.2 (5)a 79.7 6 4.0 (12)Ile/Val or Val/Val 70.7 (1) 55.5 6 8.9 (4) 61.1 6 4.2 (11)aTo test the difference between current smokers and former smokers,analysis of variance was carried out with genotype as a covariate. Thedifferences observed between current smokers and former smokers were notsignificant (P 5 0.11). The presence of a 105Val allele was positivelyassociated with lower CDNB activity (ANOVA, F 5 4.48, P 5 0.0086).acid and bromo-sulfophthalein conjugation; suggesting thatthe differences in activity reflect differences in binding of theelectrophilic substrate to the hydrophobic binding site ofGSTP1-1. It would be of interest to know how the polymorphism affects activity with regard to other GSTP1-1 substratespresent in cigarette smoke, such as benzo[a]pyrene.Due to limited availability of tissues, this study utilizedhealthy tissue from patients undergoing surgery (as opposedto healthy tissue from healthy patients). While disease statecan sometimes influence expression of metabolizing enzymesin tumors, this has not been observed for GSTP1-1 in normalhuman lung tissue (41). The present study focuses on theeffect of GSTP1 genotype on GSTP1-1 activity levels inhealthy lung tissue, and while the relationship between thesevariables could be influenced by disease state, this seemsunlikely, given that these results are consistent with thebiochemical data (26,27).As shown in Table II, there was a non-significant trendtoward lower mean GST activity among individuals with the114 valine allele. Ali-Osman et al. (27) identified three variantcDNAs with the following haplotypes: (i) 105 Ile/114Ala(GSTP1*A); (ii) 105Val/114Ala (GSTP1*B), and (iii) 105Val/114Val (GSTP1*C) (see Reference 39 for GST nomenclature).Precise haplotype frequencies could not be determined in thepresent study using PCR genotyping methods, but haplotypescan be estimated to occur in the order GSTP1*A .. GSTP1*B. GSTP1*C. In addition, two samples had a 114 valine alleleand were homozygous for 105Ile. Therefore, a fourth rarehaplotype exists for GSTP1 (105Ile/114Val, proposed asGSTP1*D). There were not enough individuals in the studycarrying this haplotype (GSTP1*D) to assess if 114 valinehad an independent association with enzyme activity levels.However, the mean conjugating activity for the two individualscarrying GSTP1*D alleles was 87.9 nmol/mg per min, suggesting that the presence of a valine at position 114 had littleimpact on enzyme activity in whole tissue extracts.Exposure to cigarette smoke can alter expression of somexenobiotic metabolism enzymes (32,33,40,41). Because weobserved considerable variation for CDNB conjugating activitywithin GSTP1 genotypes, we also tested to see if currentsmoking behavior influenced GSTP1-1 activity in lung tissuesamples. There were two non-smokers, nine former smokersand 23 current smokers in the study. As seen in Table III,which displays stratification by smoking and also by smokingand genotype, there is no apparent relationship between smoking history and CDNB activity in these tissue samples. Meanactivity levels for current smokers (68.5) were intermediate tonon-smokers (87.9) and former smokers (62.5). IndividualsM.A.Watson et al.Table IV. GSTP1 genotype and allele frequencies among Taiwanese, African-Americans and European-AmericansExon 5 Ile105Val Exon 6 Ala114Valn Ile/Ile Ile/Val Val/Val Allele frequency n Ala/Ala Ala/Val Val/Val Allele frequency(valine) (valine)Taiwanese 116 78 (67%) 35 (30%) 3 (3%) 0.18a n.d.e n.d.n.d.112106 (95%) 6 (5%) African-American 137 48 (35%) 63 (46%) 26 (19%) 0.42b European-American 287 119 (42%) 147 (51%) 21 (7%) 0.33c,d 114 93 (82%) 20 (18%) 0 (0%) 0.09aAllele frequency Taiwanese versus African-Americans, chi-square test, P , 0.0001.bAllele frequency African-Americans versus European-Americans, chi-square test, P 5 0.01.cAllele frequency Taiwanese versus European-Americans, chi-square test, P , 0.0001.dThe Hardy–Weinberg equilibrium test indicates an excess number of 105Val heterozygotes in this European-American population sample (P , 0.05). Otherpopulations were in equilibrium. HW tests for other genes (GSTM1, NAT2, NAT1, CYP2E1) in these populations were all in equilibrium.en.d., not determined.with the 105 valine allele tended to have lower activities ineach smoking category; however, due to the small size of thisstudy, differences observed in the stratified analysis are notstatistically meaningful. Only GSTP1 genotype appears to bea significant predictor of activity (F 5 4.48, P 5 0.0086). Otherstudies of GSTP1 and smoking have also been inconclusive.Nakajima et al. (41) found no significant differences in GSTP1-1 activity in lung tissue among samples from smokers andnon-smokers, providing further evidence that tobacco may notbe an effective inducer of GSTP1 in lung.In order to determine the distribution of genotypes indifferent populations, Ile105Val and Ala114Val polymorphismswere analyzed in samples obtained from European-Americans,African-Americans and Taiwanese populations. Table IV showsthe frequencies of genotypes and alleles in these populations.The 105 valine allele was most common among AfricanAmericans (0.42) and least common among Taiwanese (0.18)with European-Americans (0.33) intermediate between thesegroups. The allele frequency for European-Americans is notsignificantly different. These frequencies should be consideredpreliminary; in particular, the European-American sampledisplays an excessive number of heterozygotes. While thesamples used to test frequency may not be truly representativeof these racial groups, the differences observed were highlystatistically significant, which suggests that the underlyingallele frequencies in Europeans, Asians and African-Americansare likely to be different. Other data from our laboratorysuggests that Asian populations, in general, may have a low105 valine allele frequency: a Japanese population we havetested had a 0.14 allele frequency for 105 valine (D.A.Belland T.Katoh, unpublished).The Ala114Val polymorphism in exon 6 was less commonthan the Ile105Val in both European-Americans (0.09) andAfrican-Americans (0.05) (Table IV). While the 105Val allelewas more common in African-Americans than in EuropeanAmericans, the opposite was true for the 114Val allele. Thus,the GSTP1*B haplotype must be the common variant allele inAfrican-Americans. The differences in genotype and allelefrequencies that we observed between groups suggests thepossibility of differences in susceptibility to exposure toelectrophilic toxicants, or in effectiveness of drugs that areinactivated by GSTP1-1-catalyzed biotransformation. Forexample GSTP1-1 is over-expressed in certain drug resistanttumor cell lines and GSTP1-1 activity may contribute to thedrug resistant phenotype of these cells (42,43). It will be of278interest if genetic polymorphism modulates GSTP1-1 tumorexpression or drug resistance phenotype.Other GST super-family genes are polymorphic (GSTM1and GSTT1) and variations within these genes have beenassociated with an increased risk for cancer, including bladderand lung (9–13). A recent report by Harries et al. (37) suggeststhat GSTP1 genotype may also influence risk of cancer;individuals who were GSTP1 valine 105 homozygotes hadincreased risk for bladder and testicular cancer, and decreasedrisk for prostate cancer. No information on epidemiologicalrisk factors was available for subjects in the Harries et, but the positive findings in that preliminary studysuggest that more detailed studies might be worthwhile.Levels of expression of the GSTP1-1 enzyme may beimportant for the efficiency of detoxification. Terrier et al. (14)used immunohistochemistry to examine GSTP1-1 expression innormal human tissue and found high levels of GST pi-classexpression in respiratory, urinary and digestive tract epithelia.Thus, tissues that are the most likely to be exposed toexogenous chemicals from the environment also generallyexpress GSTP1-1 at high levels. Hypothetically, these tissuesmay be the most ‘at risk’ of developing cancers among thoseindividuals with genetically determined low activity GSTP1-1 phenotypes. Indeed, a very recent paper suggests thatindividuals with low activity alleles for GSTP1 had higherlevels of smoking related DNA adducts in lung tissue andhigher risk of lung cancer (44). We are conducting follow-upcase-control studies of cancer of the bladder, lung, colon, breastand prostate to determine if risk associated with environmentalexposures is modulated by genetic polymorphism in theGSTP1 gene.In conclusion, the Ile105Val polymorphism in the GSTP1gene significantly alters catalytic activity of this phase IIenzyme in lung tissue samples and this variation occursfrequently in human populations. GSTP1 variability has potential implications for individual susceptibility to electrophiliccarcinogen metabolites, as well as for expression of the drugresistant phenotype.AcknowledgementsWe thank L.L.Hsieh, Chang Gung Medical School for providing DNA samplesfrom Taiwanese maternity patients. We also thank Patrick Herron, NIEHS forPCR optimizations, Lena Heung for photographs and Richard Morris, Analytical Science, Inc. for statistical analysis. This study was supported in part byMedical Research Council of Canada grant MT10382 to T.E.M. The authorswish to express gratitude to Drs Alan Conlan, Ken Reid and Dimitri Petsikas,Human glutathione S-transferase P1 polymorphismsDepartment of Surgery, Queen’s University for providing the human lungtissue and smoking histories. We also thank Sophia Ali and Patty Donnellyfor assistance in tissue preparation.References1.Boyer,T.D. and Kenney,W.C. (1985) Preparation, characterization andproperties of glutathione S-transferases. In Zakim,D. and Vessey,D. (eds)Biochemical Pharmacology and Toxicology. John Wiley and Sons, NewYork, NY, pp. 297–363.2.Mannervik,B. (1985) The isozymes of glutathione S-transferase. Adv.Enzymol., 57, 357–417.3.Dekant,W. and Vamkars,S. (1993) Glutathione-dependent bioactivation ofxenobiotics. Xenobiotica, 23, 873–887.4.Mannervik,B., Alin,P., Guthenberg,C., Jensson,H., Tahir,M.K.,Warholm,M. and Jornvall,H. (1985) Identification of three classes ofcytosolic glutathione transferase common to several mammalian species,correlation between structural data and enzymatic properties. Proc. NatlAcad. Sci. USA, 82, 7202–7206.5.Tsuchida,S. and Sato,K. (1992) Glutathione transferases and cancer. Crit.Rev. Biochem. Mol. Biol., 27, 337–384.6.Seidgard,J., Vorachek,W.R., Pero,R.W. and Pearson,W.R. (1988) Hereditarydifferences in the expression of the human glutathione transferase activeon trans-stilbene oxide are due to a gene deletion. Proc. Natl Acad. Sci.USA, 85, 7293–7297.7.Pemble,S., Schroeder,K.R., Spencer,S.R, Meyer,D.J., Hallier,E., Bolt,H.M,Ketterer,B. and Taylor,J.B. (1994). Human glutathione S-transferase theta(GSTT1): cDNA cloning and the characterization of a geneticpolymorphism. Biochem. J., 300, 271–276.8.Wiencke,J.K., Pemble,S., Ketterer,B. and Kelsey,K.T. (1995) Gene deletionof glutathione transferase theta 1: correlation with induced genetic damageand potential role in endogenous mutagenesis. Cancer Epidem. Biomark.Prevent., 4, 253–260.9.Bell,D.A., Taylor,J.A., Paulson,D.F., Robertson,C.N., Mohler,J.L. andLucier,G.W. (1993) Genetic risk and carcinogen exposure—a commoninherited defect of the carcinogen-metabolism gene glutathioneS-transferase M1 (GSTM1) that increases susceptibility to bladder cancer.J. Natl Cancer. Inst., 85, 1159–1164.10.Seidegard,J., Pero,R.W., Markowitz,M.M., Roush,G., Miller D.G. andBeattie,J. (1990) Isoenzyme(s) of glutathione transferase (class Mu) as amarker for the susceptibility to lung cancer. Carcinogenesis, 11, 33–36.11.Nazar-Stewart,V., Motulsky,A.G., Eaton,D.L., White,E., Hornung,S.K.,Leng,Z.T., Stapleton,P. and Weiss,N.S. (1993) The glutathione S-transferaseµ polymorphism as a marker for the susceptibility to lung cancer. CancerRes., 53, 2313–2318.12.Hirvonen,A., Husgafvel-Pursiainen,K., Anttila,S. and Vainio,H. (1993) TheGSTM1 null genotype as a potential risk modifier for squamous cellcarcinoma of the lung. Carcinogenesis, 14, 1479–1481.13.Chen,H.W., Sandler,D.P., Taylor,J.A, Shore D.L., Liu,E., Bloomfield,C.D.and Bell,D.A. (1996) Increased risk for myelodysplastic syndromes inindividuals with glutathione-transferase theta 1 (GSTT1) gene defect.Lancet, 347, 295–297.14.Terrier,P., Townsend,A.J., Coindre,J.M., Triche,T.J. and Cowan,K.H. (1990)An immunohistochemical study of pi class glutathione S-transferaseexpression in normal human tissue. Am. J. Pathol., 137, 845–853.15.Moscow,J.A., Fairchild,C.R., Madden,M.J. et al. (1989) Expression ofanionic glutathione-S-transferase and P-glycoprotein genes in humantissues and tumors. Cancer Res., 49, 1422–1428.16.Singh,S.V., Xu,B.H., Tkalcevic,G.T., Gupta,V., Roberts,B. and Ruiz,P.(1994) Glutathione-linked detoxification pathway in normal and malignanthuman bladder tissue. Cancer Lett., 77, 15–24.17.Howie,A.F., Forrester,L.M., Glancey,M.J., Schlager,J.J., Powis,G.,Beckett,G.J., Hayes,J.D. and Wolf,C.R. (1990) Glutathione S-transferaseand glutathione peroxidase expression in normal and tumor human tissues.Carcinogenesis, 11, 451–458.18.Singh,S.V., Brunnert,S.R., Roberts,B. and Krishan,A. (1990) Differentialexpression of glutathione S-transferase, glutathione peroxidase andglutathione reductase in normal and malignant human breast tissues.Cancer Lett., 51, 43–48.19.Mulder,T.J., Manni,J.J., Roelofs,H.J., Peters,W.M. and Wiersma,A. (1995)Glutathione S-transferases and glutathione in human head and neck cancer.Carcinogenesis, 16, 619–624.20.Mulder,T.J., Verspaget,H.W., Sier,C.M., Roelofs,H.J., Ganesh,S.,Griffioen,G. and Peters,W.M. (1995) Glutathione S-transferase π incolorectal tumors is predictive for overall survival. Cancer Res., 55,2696–2702.27921.Peters,W.M., Boon,C.W., Roelofs,H.J., Wobbes,T., Nagengast,F.M. andKremers,P.G. (1992) Expression of drug metabolizing enzymes and P-170glycoprotein in colorectal carcinoma and normal mucosa. Gastroenterology,103, 448–455.22.Peters,W.M., Wormskamp,N.M. and Thies,E. (1990) Expression ofglutathione S-transferase in normal gastric mucosa and gastric tumors.Carcinogenesis, 11, 1593–1596.23.Shimizu,K., Toriyama,F., Shang,H. and Yoshida,H. (1995) The expressionof placental-type glutathione S-transferase (GST-π) in human cutaneouscarcinoma in situ, that is, actinic keratosis and Bowen’s disease, comparedwith normal human skin. Carcinogenesis, 10, 2327–2330.24.Sato,K. (1989) Glutathione S-transferases as markers of preneoplasia andneoplasia. Adv. Cancer Res., 52, 205–255.25.Board,P.G., Weber,G.C. and Coggan,M. (1989) Isolation of a cDNA cloneand localization of the human glutathione S-transferase 3 genes tochromosome bands 11q13 and 12q13–14. Ann. Hum. Genet., 53, 205–213.26.Zimniak,P., Nandur,B., Pikula,S., Bandorowics-Pikula,J., Singhal,S.,Srivastava,S., Awashthi,S. and Awasthi,Y. (1994) Naturally occurringhuman glutathione S-transferase GSTP1-1 isoforms with isoleucine andvaline in position 105 differ in enzymatic properties. Eur. J. Biochem.,224, 893–899.27.Ali-Osman,F., Akande,O., Antoun,G., Mao J. and Buolamwin,J. (1997)Molecular cloning, characterization and expression in Escherichia coli offull-length cDNAs of three human glutathione S-transferase Pi genevariants. J. Biol. Chem., 15, 10004–10012.28.Daniels,J.M., Liu,L., Stewart,R.K. and Massey,T.E. (1990) Biotransformation of aflatoxin B1 in rabbit lung and liver microsomes.Carcinogenesis, 11, 823–827.29.Lowry,O.H., Rosebrough,N.J., Farr,A.L. and Randall,R.J. (1951) Proteinmeasurement with the folin phenol reagent. J. Biol. Chem., 193, 265–275.30.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, 7130–7139.31.Devereux,T.R., Belinsky,S.A., Maronpot,R.R., White,C.M., Hegi,M.E.,Patel,A.C., Foley,J.F., Greenwell,A. and Anderson,M.W. (1993)Comparison of pulmonary O-6 methylguanine DNA adduct levels and Kiras activation in lung-tumors from resistant and susceptible mouse strains.Mol. Carcinogenesis, 8, 177–185.32.Bartsch H., Bartsch,H., Petruzzelli,S. et al. (1991) Carcinogen metabolismand DNA adducts in human lung tissues as affected by tobacco smokingor metabolic phenotype: A case-control study on lung cancer patients.Mutat. Res., 250, 103–114.33.McLemore,T.L., Adelberg,S., Liu,M.C., McMahon,N.A., Yu,S.J. andHubbard,W.C. (1990) Expression of CYP1A1 gene in patients with lungcancer: Evidence for cigarette smoke-induced gene expression in normallung tissue and for altered gene regulation in primary pulmonarycarcinomas. J. Natl Cancer Inst., 82, 1333–1339.34.Hsieh,L.L and Hsieh,T.T. (1993) Detection of aflatoxin B1–DNA adductsin human placenta and cord blood. Cancer Res., 53, 1278–1280.35.Bell,D.A., Taylor,J.A., Butler,M.A., Stephens,E., Wiest,J., Brubaker,L.,Kadlubar,F. and Lucier,G.W. (1993) Genotype/phenotype discordance forhuman arylamine N-acetyl transferase (NAT2) reveals a new slowacetylator allele common in African-Americans. Carcinogenesis, 14,1689–1692.36.Stephens,E.A., Taylor,J.A., Kaplan,N., Yang, C.-H., Hsieh,L.L.,Lucier,G.W. and Bell,D.A. (1994) Ethnic variation in the CYP2E1 gene:Polymorphism analysis of 695 African-Americans, European-Americansand Taiwanese. Pharmacogenetics, 4, 185–192.37.Harries,L.W., Stubbins M.J., Forman D., Howard G. and Wolf,C.R. (1997)Identification of genetic polymorphisms at the glutathione S-transferase Pilocus and association with susceptibility to bladder, testicular and prostatecancer. Carcinogenesis, 18, 641–644.38.Singhal,S.S. Saxena,J., Ahmad,H., Awasthi,S., Haque,A.K. andAwasthi,Y.C. (1992) Glutathione S-transferases of human lung:Characterization and evaluation of the protective role of the alphaclass isozymes against lipid peroxidation. Arch. Biochem. Biophys., 299,232–241.39.Mannervik,B., Awasthi,Y.C., Board,P.G. et al. (1992) Nomenclature forhuman glutathione transferases. Biochem. J., 282, 305–308.40.Petruzzelli,S., Camus,A.M., Carrozi,L. et al. (1988) Long lasting effectsof tobacco smoking on pulmonary drug-metabolizing enzymes: a casecontrol study on lung cancer patients. Cancer Res., 48, 4695–4700.41.Nakajima,T., Elovaara,E., Anttilla,S., Hirvonen,A., Camus,A., Hayes,J.,Ketterer,B. and Vainio,H. (1995) Expression and polymorphism ofglutathione S-transferase in human lungs: risk factors in smoking relatedlung cancer. Carcinogenesis, 16, 707–711.M.A.Watson et al.42.Cole,S.P., Downes,H.F., Mirski,S.E. and Clements,D.J. (1990) Alterationsin glutathione and glutathione related enzymes in a multidrug resistantsmall cell lung cancer cell line. Mol. Pharmacol., 37, 192–197.43.Bai,F., Nakanishi,Y., Kawasaki,M., Takayama,K., Yataunami,J., Pei,X.H.,Tsuruta,N., Wakamatsu,K. and Hara,N. (1996) Immunohistochemicalexpression of glutathione S-transferase Pi can predict chemotherapyresponse in patients with non-small cell lung carcinoma. Cancer, 78,416–421.44.Ryberg,D., Skaug,V., Hewer,A. et al. (1997) Genotypes of glutathionetransferase M1 and P1 and their significance for lung DNA adduct levelsand cancer risk. Carcinogenesis, 18, 1285–1289.Received on July 30, 1997; revised on October 22, 1997; accepted onNovember 3, 1997280


Leave a Reply

Your email address will not be published. Required fields are marked *