Toxicology Letters, 64/65 (1992) 477-463 0 1992 Elsevier Science Publishers B.V., All rights reserved 0376-4274/Z/$5.00
477
Expression of pulmonary cytochrome P450lAl and carcinogen DNA adduct formation in high risk subjects for tobacco-related lung cancer H. Bartscha, M. Castegnaroa, M. RojasaTb,A.-M. Camusa, K. Alexa.ndrotib and M. Langa fir Research on Cancer, Lyon, and bVisitingScientist des Sciences de la Vie, Paris (France)
%ternationalAgency Dkpartement
from CNRS,
Kqy words: Lung cancer; CYPlA-related enzymes; Carcinogen DNA adducts; Metabolic phenotypes; Cigarette smoking
SUMMARY Cigarette smoking is the strongest risk factor for lung cancer (LC), but genetically determined variations in pulmonary metabolism of tobacco-derived carcinogens may affect individual risk. Results from a case-control study on LC patients demonstrated the pronounced effect of tobacco smoke on pulmonary xenobiotic metabolism and prooxidant state, and suggested the existence of a metabolic phenotype at higher risk for tobacco-associated LC: LC patients who were recent smokers had significantly induced BP-3-hydroxylase (AHH) and ethoxycoumarin Odeethylase (ECDE) activities in lung parenchyma, when compared with smoking non-cancer patients. In recent smokers, lung AHH activity was positively correlated with the level of tobacco smokederived DNA adducts as determined by 32P-postlabelling. Pulmonary AHH activity also showed a good correlation with the intensity of immunohistochemical staining for cyt. P4501A by a monoclonal Ab in lung tissue sections: smoking and peripheral type of lung cancers were positively related to high levels of this cyt. P450 species, probably reflecting high rates of induction. These results suggest that high pulmonary CYPlAl expression (controlling in part carcinogen DNA-adduct formation) in tobacco smokers, appears to be associated with LC risk. High risk subjects may thus be identifiable through genotyping assays for CYPlAl polymorphism.
INTRODUCTION
Cigarette smoking is the strongest risk factor for lung cancer, but genetically determined variations in activity of pulmonary enzymes that metabo-
Correspondence to: H. Bartsch, International Agency for Research on Cancer, 150 cools Albert-Thomas, 69372 Lyon Cedex 08, France.
478
lize tobacco-derived carcinogens may affect individual risk [ll. To investigate whether expression of (polymorphic) CYPIA-related enzymes can serve as risk markers for carcinogen-DNA damage and for lung cancer in smokers, lung tissue specimens taken during surgery from patients with either lung cancer or non-neoplastic lung disease were analysed. Phase I and phase II drug-metabolizing enzyme activities were determined in lung parenchyma and/or bronchial tissue preparations. Stereoselective metabolism of benzo[ulpyrene (B[alP)-7,Sdiol by human lung parenchymal microsomes was also investigated. The data were analysed for (i) differences in metabolic profiles between bronchial and parenchymal lung tissues; (ii) the effect of recent exposure to tobacco smoke on enzyme inducibility and B[alP metabolism; (iii) differences in enzyme inducibility between patients with lung cancer and those with other lung disease; (iv) correlation of pulmonary BP-Shydroxylase or aryl hydrocarbon hydrolase (AHH) with carcinogen DNA adduct levels or with metabolism in lung parenchyma, and (v) correlation in lung tissue sections of immunohistochemical staining intensity for CYPlA with enzyme activities, smoking history and tumour types [21. The results demonstrated a pronounced effect of tobacco smoke on pulmonary drug and carcinogen metabolism and suggested the existence of a metabolic phenotype at higher risk for tobacco-associated lung cancer. STUDY SUBJECTS
AND METHODS
Subjects were patients undergoing surgery for lung cancer Qobectomy or pulmonectomy) or other thoracic diseases (a) at the University Hospital, Pisa, Italy, and (b) at the University Central Hospital, Helsinki, Finland. Data regarding pulmonary and extrapulmonary diseases, occupational exposures (asbestos) and smoking habits were collected for later data analysis. Patients were divided into smokers, ex-smokers (those who had refrained from smoking for at least six months) and non-smokers. Not all parameters were investigated for every subject, and results from sub-groups are described below. BIOCHEMICAL
ASSAYS AND IMMUNOHISTOCHEMISTRY
S-12 Fraction and microsomes from lung tissue were prepared and enzyme activities were determined as previously described, [31. AHH activity was determined by the fluorometric method with a detection limit of 0.05 pmol 3-hydroxy-BP x [mini-’ x [mg protein]-i. Microsomal conversion of (-)-BP-7,8-diol into tetrols was quantified using HPLC with fluorescence detection 141.Identification of bulky (aromatic) DNA adducts was carried out after enrichment of adducted nucleotides by dephosphorylation of nor-
479
ma1 nucleotides with nuclease Pi using 32P-postlabelling, chromatography clean-up, transfer chromatography and separation, as described elsewhere [5,61. The quantitation of normal nucleotides and adducts was performed using liquid scintillation counting. The calculation of adduct levels was based on the assumption that the adducts. were completely resistant to 3’-dephosphorylation by nuclease Pi. For immunohistochemistry and enzyme assays of the same sample, specimens were kept at -70°C until analysed. The remaining surgical specimen was fixed with 10% formalin and the histological samples were processed conventionally for light microscopy. For immunohistochemistry (in collaboration with S. Anttila, K. Husgafvel-Pursiainen, Institute of Occupational Health, Helsinki, Finland; H. Vainio, IARC, Lyon, France and E. Hietanen, University of Turku, Turku, Finland) frozen sections were stained with a monoclonal antibody (Mab 1-7-l) raised against a 3-methylcholanthrene-inducible rat cytochrome P450 isozyme (generously provided by Drs H. Gelboin and S.S. Park, NCI, Bethesda, MD, USA), using an avidin-biotin-peroxidase method [71. The immunostaining of peripheral lung tissue was classified into three grades.
RESULTS AND DISCUSSION
Comparison ofphase Iandphase IIdrug-metabolizing enzymes in human bronchial and lung parenchymal tissue: Pulmonary aryl hydrocarbon hydroxylase (AHH), ethoxycoumarin 0-deethylase (ECDE), epoxide hydrolase (EH), glutathione S-transferase (GST) and UDP-glucuronosyltransferase (UGT) activities were measured in S-12 fractions of bronchial tree and peripheral lung parenchyma from 21 patients (University Hospital of Pisa), all smokers or ex-smokers [81.Enzyme activities in lung parenchyma and bronchial preparations of the same patient investigated in a sub-group of 10 were positively correlated with each other (e.g. AHH: rS0.77, p (0.009, GST: r=0.83, p
477
Expression of pulmonary cytochrome P450lAl and carcinogen DNA adduct formation in high risk subjects for tobacco-related lung cancer H. Bartscha, M. Castegnaroa, M. RojasaTb,A.-M. Camusa, K. Alexa.ndrotib and M. Langa fir Research on Cancer, Lyon, and bVisitingScientist des Sciences de la Vie, Paris (France)
%ternationalAgency Dkpartement
from CNRS,
Kqy words: Lung cancer; CYPlA-related enzymes; Carcinogen DNA adducts; Metabolic phenotypes; Cigarette smoking
SUMMARY Cigarette smoking is the strongest risk factor for lung cancer (LC), but genetically determined variations in pulmonary metabolism of tobacco-derived carcinogens may affect individual risk. Results from a case-control study on LC patients demonstrated the pronounced effect of tobacco smoke on pulmonary xenobiotic metabolism and prooxidant state, and suggested the existence of a metabolic phenotype at higher risk for tobacco-associated LC: LC patients who were recent smokers had significantly induced BP-3-hydroxylase (AHH) and ethoxycoumarin Odeethylase (ECDE) activities in lung parenchyma, when compared with smoking non-cancer patients. In recent smokers, lung AHH activity was positively correlated with the level of tobacco smokederived DNA adducts as determined by 32P-postlabelling. Pulmonary AHH activity also showed a good correlation with the intensity of immunohistochemical staining for cyt. P4501A by a monoclonal Ab in lung tissue sections: smoking and peripheral type of lung cancers were positively related to high levels of this cyt. P450 species, probably reflecting high rates of induction. These results suggest that high pulmonary CYPlAl expression (controlling in part carcinogen DNA-adduct formation) in tobacco smokers, appears to be associated with LC risk. High risk subjects may thus be identifiable through genotyping assays for CYPlAl polymorphism.
INTRODUCTION
Cigarette smoking is the strongest risk factor for lung cancer, but genetically determined variations in activity of pulmonary enzymes that metabo-
Correspondence to: H. Bartsch, International Agency for Research on Cancer, 150 cools Albert-Thomas, 69372 Lyon Cedex 08, France.
478
lize tobacco-derived carcinogens may affect individual risk [ll. To investigate whether expression of (polymorphic) CYPIA-related enzymes can serve as risk markers for carcinogen-DNA damage and for lung cancer in smokers, lung tissue specimens taken during surgery from patients with either lung cancer or non-neoplastic lung disease were analysed. Phase I and phase II drug-metabolizing enzyme activities were determined in lung parenchyma and/or bronchial tissue preparations. Stereoselective metabolism of benzo[ulpyrene (B[alP)-7,Sdiol by human lung parenchymal microsomes was also investigated. The data were analysed for (i) differences in metabolic profiles between bronchial and parenchymal lung tissues; (ii) the effect of recent exposure to tobacco smoke on enzyme inducibility and B[alP metabolism; (iii) differences in enzyme inducibility between patients with lung cancer and those with other lung disease; (iv) correlation of pulmonary BP-Shydroxylase or aryl hydrocarbon hydrolase (AHH) with carcinogen DNA adduct levels or with metabolism in lung parenchyma, and (v) correlation in lung tissue sections of immunohistochemical staining intensity for CYPlA with enzyme activities, smoking history and tumour types [21. The results demonstrated a pronounced effect of tobacco smoke on pulmonary drug and carcinogen metabolism and suggested the existence of a metabolic phenotype at higher risk for tobacco-associated lung cancer. STUDY SUBJECTS
AND METHODS
Subjects were patients undergoing surgery for lung cancer Qobectomy or pulmonectomy) or other thoracic diseases (a) at the University Hospital, Pisa, Italy, and (b) at the University Central Hospital, Helsinki, Finland. Data regarding pulmonary and extrapulmonary diseases, occupational exposures (asbestos) and smoking habits were collected for later data analysis. Patients were divided into smokers, ex-smokers (those who had refrained from smoking for at least six months) and non-smokers. Not all parameters were investigated for every subject, and results from sub-groups are described below. BIOCHEMICAL
ASSAYS AND IMMUNOHISTOCHEMISTRY
S-12 Fraction and microsomes from lung tissue were prepared and enzyme activities were determined as previously described, [31. AHH activity was determined by the fluorometric method with a detection limit of 0.05 pmol 3-hydroxy-BP x [mini-’ x [mg protein]-i. Microsomal conversion of (-)-BP-7,8-diol into tetrols was quantified using HPLC with fluorescence detection 141.Identification of bulky (aromatic) DNA adducts was carried out after enrichment of adducted nucleotides by dephosphorylation of nor-
479
ma1 nucleotides with nuclease Pi using 32P-postlabelling, chromatography clean-up, transfer chromatography and separation, as described elsewhere [5,61. The quantitation of normal nucleotides and adducts was performed using liquid scintillation counting. The calculation of adduct levels was based on the assumption that the adducts. were completely resistant to 3’-dephosphorylation by nuclease Pi. For immunohistochemistry and enzyme assays of the same sample, specimens were kept at -70°C until analysed. The remaining surgical specimen was fixed with 10% formalin and the histological samples were processed conventionally for light microscopy. For immunohistochemistry (in collaboration with S. Anttila, K. Husgafvel-Pursiainen, Institute of Occupational Health, Helsinki, Finland; H. Vainio, IARC, Lyon, France and E. Hietanen, University of Turku, Turku, Finland) frozen sections were stained with a monoclonal antibody (Mab 1-7-l) raised against a 3-methylcholanthrene-inducible rat cytochrome P450 isozyme (generously provided by Drs H. Gelboin and S.S. Park, NCI, Bethesda, MD, USA), using an avidin-biotin-peroxidase method [71. The immunostaining of peripheral lung tissue was classified into three grades.
RESULTS AND DISCUSSION
Comparison ofphase Iandphase IIdrug-metabolizing enzymes in human bronchial and lung parenchymal tissue: Pulmonary aryl hydrocarbon hydroxylase (AHH), ethoxycoumarin 0-deethylase (ECDE), epoxide hydrolase (EH), glutathione S-transferase (GST) and UDP-glucuronosyltransferase (UGT) activities were measured in S-12 fractions of bronchial tree and peripheral lung parenchyma from 21 patients (University Hospital of Pisa), all smokers or ex-smokers [81.Enzyme activities in lung parenchyma and bronchial preparations of the same patient investigated in a sub-group of 10 were positively correlated with each other (e.g. AHH: rS0.77, p (0.009, GST: r=0.83, p
