HES”; 31293015914611 IBBARY 117W "TlIii/illiilliflii/iiii/WIT/7! "1 ‘ Michigan State University “9-- This is to certify that the dissertation entitled ‘ Canine In Vivo and lg_Vitro Metabolism of the Bladder CarEinogen 4,4'-Methyienebis(2-Ch10roani1ine) presented by Melanie Otten Manis has been accepted towards fulfillment of the requirements for Ph.D. Pharmacology/ degree in . Tox1cology /fi\ I 'I .2! 1' "’ , Major professor I ’ C’ t.” DateAQZ/“JJ 71 // y / ¢// 1/ MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771 MSU * LIBRARIES “ RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. CANINE IN_VIVO AND IN_VITRO METABOLISM OF THE BLADDER CARCINOGEN 4,4'-METHYLENEBIS(2-CHLOROANILINE) By Melanie Otten Manis A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pharmacology and Toxicology l984 fi/ I '- ABSTRACT Canine In VlVO and ln_Vitro Metabolism of the Bladder Carcinogen 4,4'-Methylenebis(2-Chloroaniline) By Melanie Otten Manis Metabolism of arylamines enhances excretion and results in both detoxified and activated products. Metabolism of known bladder car- cinogens has been investigated in the liver, kidney medulla, and bladder but not in the renal cortex. 4,4'-Methylenebis(2—chloroaniline) (MBOCA), an arylamine bladder carcinogen in dogs, is a potential human carcinogen whose biotransformation has not been extensively investigated. The ob- jectives of this investigation were 2-fold: (l) to determine the struc- ture of the major canine urinary metabolite of MBOCA and (2) to test the hypothesis that liver and kidney play a role in arylamine metabolism, using MBOCA as a model. Analysis of the major urinary metabolite of MBOCA demonstrated that it was 5-hydroxy-3,3'-dichloro-4,4'-diaminodiphenylmethane-S-sulphate, an ortho-hydroxy sulphate conjugate. Arylsulfatase but not B-glucu- ronidase or citric acid hydrolyzed the metabolite jn_!itrg, indicating that it was a sulphate conjugate. Electron impact mass spectrometry following derivatization and transesterification indicated that the metabolite was ring hydroxylated and fast atom bombardment mass spec- trometry confirmed the molecular weight as consistent with a sulfate h“: 81' 55 CC ll Melanie Otten Manis ester. Proton nuclear magnetic resonance studies indicated that the ring substitution was ortho to an amine. The major urinary metabolite of MBOCA corresponded to the same type of conjugate formed in dogs administered similar to known bladder carcinogens. Tissue slice incubations from liver and kidney produced glucuro- nide, glucoside, and sulphate metabolites of MBOCA as determined by enzymatic, physical, and instrumental methods. Liver slices produced seven metabolites separable by high performance liquid chromatography and kidney cortex slices produced six. An N-glycoside and the ortho- hydroxy sulphate were identified in both organs. A hepatic metabolite was characterized as an o-glucuronide and both liver and kidney produced a metabolite with characteristics of an N-hydroxy-N-glucuronide. 14C-MBOCA bound covalently to both liver and kidney slices in a substrate concentration and time-dependent manner. Binding was not altered in the presence of D(+)galactosamine or p-nitrophenyl sulphate. Binding decreased with a general inhibition of metabolism in the pre- sence of 2,6-dichloro-4-nitrophenol, indicating that oxidation but not conjugation may be important. This investigation provided the first demonstration of production of conjugated arylamine metabolites by the renal cortex. Thus, both liver and kidney may play a role in arylamine metabolism in the dog. In memory of Hans Hinnrich Mehrkens ii ACKNOWLEDGEMENTS My deepest thanks go to my husband Dr. Jack Jay Manis who makes this possible by his patience, love, wit, and stamina. I extend sincere thanks to Dr. H. Emmett Braselton for his guidance and flexibility. His high standards of performance were an excellent model. My thanks also to the members of my guidance committee Drs. Theodore M. Brody, Jay I. Goodman, Jerry B. Hook and Veronica Maher for their assistance during the course of this project. I appreciated the calm guidance of Dr. Michael Slanker in dog handling and the skillful technical assistance of Ms. Laura Everett and Ms. Gay DeShone in doing ig_vitrg_studies. The help of Brian Musselman and Brad Ackerman who assisted with fast atom bombardment mass spectro- metry, Dr. Klaus Halenga and Kermit Johnson who produced the nuclear magnetic resonance spectrum, and Dr. Veronica Maher and Kathleen Debien who assisted with mutagenicity studies was appreciated. iii TABLE OF CONTENTS Page LIST OF TABLES --------------------------------------------------- vii LIST OF FIGURES -------------------------------------------------- viii LIST OF ABBREVIATIONS -------------------------------------------- x INTRODUCTION ----------------------------------------------------- l A. Background --------------------------------------------- l B. Arylamine metabolism ----------------------------------- l0 C. Extrahepatic metabolism -------------------------------- 17 D. 4,4'-Methylenebis(2-chloroaniline) (MBOCA) ------------- 22 E. Purpose ------------------------------------------------ 24 METHODS ---------------------------------------------------------- 27 A. Animals -------- g --------------------------------------- 27 B. Chemicals ---------------------------------------------- 27 C. Isolation and identification of the major metabolite detected in canine urine and in hepatic tissue slices-- 28 l. HPLC characterization ----------------------------- 28 2. Isolation------; ---------------------------------- 29 3. Hydrolysis ---------------------------------------- 3O 4. Mass spectrometry --------------------------------- 31 5. Nuclear magnetic resonance (NMR) ------------------ 3l 6. Binding in vitro to DNA and protein --------------- 32 7. MutageniEity -------------------------------------- 33 D. Assessment of metabolism jg_vitro ---------------------- 34 l. Incubation procedures ----------------------------- 34 2. Metabolite separation and quantitation ------------ 35 3. Macromolecular binding ---------------------------- 36 E. Individual ig_vitro tissue slice experiments ----------- 36 1. Preliminary investigation of metabolism ----------- 36 a. Enzyme concentration dependence of binding--- 36 b. MBOCA concentration dependence of binding---- 37 c. Time dependence of metabolism ---------------- 37 iv TABLE OF CONTENTS (continued) Page 2. Elucidation of metabolites - sulfation ------------ 37 a. Incubation with Na235504 --------------------- 37 b. Incubation with p-nitrophenyl sulphate ------- 38 c. Incubation with 2,6-dichloro-4-nitr0phenol--- 38 3. Elucidation of metabolites - glucuronidation and glucosidation ------------------------------------- 38 a. Incubation with 14C-glucose ------------------ 38 b. Incubation with D(+)galactosamine ------------ 38 4. Elucidation of metabolites - glutathione ---------- 38 a. Incubation with diethylmaleate --------------- 39 b. Incubation with methionine sulfoximine ------- 39 5. Post-incubation analyses -------------------------- 39 a. Citric acid hydrolysis ----------------------- 39 b. B-Glucuronidase hydrolysis ------------------- 40 c. Arylsulfatase hydrolysis --------------------- 40 d. Thermal hydrolysis --------------------------- 4O 6. Mass spectrometry --------------------------------- 40 a. Isolation of peaks from in thro incubations- 40 b. Fast atom bombardment mass spectrometry (FAB- MS) ------------------------------------- 41 c. Electron- -impact mass spectrometry (EI -MS)---- 4l 7. Renal medullary metabolism ------------------------ 42 F. Statistical methods ------------------------------------ 42 RESULTS ---------------------------------------------------------- 44 A. Isolation and characterization of the major metabolite detected in canine urine and hepatic tissue slices ----- 44 1. HPLC characterization ----------------------------- 44 2. Hydrolysis ---------------------------------------- 44 3. Mass spectrometry --------------------------------- 47 4. Nuclear magnetic resonance ------------------------ 47 5. Binding in thro to protein and DNA --------------- 55 6. Mutagenicity -------------------------------------- 55 B. lg_vitro incubations ------- L ---------------------------- 61 1. Preliminary investigation of metabolism ----------- 6l 2. Elucidation of metabolites - sulfation ------------ 73 TABLE OF CONTENTS (continued) Page 3. Elucidation of metabolites - glucuronidation and glucosidation ..................................... 39 4. Elucidation of metabolites - glutathione conju- gation -------------------------------------------- 96 5. Post-incubation analyses .......................... 103 6. Mass spectrometry of peaks isolated from ifl_vitro incubations ................................. """ ...... l08 7. Renal medullary metabolism-----------------------+ l08 DISCUSSION ....................................................... 115 A. Characterization of the major metabolites in canine urine -------------------------------------------------- ll5 B. lg_vitro metabolism of MBOCA by liver and kidney slices llB l. Macromolecular binding in hepatic and renal tissue slice --------------------------------------------- l20 2. Elucidation of metabolites in renal and hepatic tissue slices ------------------------------------- l23 SUMMARY AND CONCLUSIONS ------------------------------------------ l28 BIBLIOGRAPHY ----------------------------------------------------- l32 vi LIST OF TABLES Table Page 1 250 MHz NMR spectral data of the major canine urinary metabolite of 4,4'-methylenebis(2-chloroaniline) ------- 52 2 Mutagenicity of the major urinary metabolite of MBOCA-- 60 vii Figure 10 ll l2 13 LIST OF FIGURES Aromatic amine structures .............................. HPLC characterization of the major urinary metabolite of MBOCA ----------------------------------------------- Arylsulfatase hydrolysis of the major urinary metabo- lite of MBOCA .......................................... FAB-MS of the major urinary metabolite of MBOCA in a saturated KCl-glycerol matrix at 6.5 Kev ............... 250 MHz NMR of the major urinary metabolite of MBOCA--- Protein binding of the major urinary metabolite of MBOCA during arylsulfatase hydrolysis in 0.2 M Tris buffer, pH 7.l, using 2 pg of metabolite --------------- DNA binding of the major urinary metabolite of MBOCA during arylsulfatase hydrolysis in 0.2 M Tris buffer with 2 pg metabolite and 1 mg/ml DNA ------------------- HPLC chromatogram of metabolites from a hepatic tissue Slice incubation ....................................... HPLC chromatogram of metabolites from a renal cortical tissue slice incubation -------------------------------- MBOCA concentration-dependent macromolecular binding in hepatic tissue slice incubations -------------------- MBOCA concentration-dependent macromolecular binding in renal cortical and medullary tissue slice incuba- tions -------------------------------------------------- Time-dependent macromolecular binding of MBOCA in tissue slices .......................................... Appearance of metabolites in hepatic tissue slice incu- bations ------------------------------------------------ viii Page 6 45 48 50 53 56 58 62 64 67 67 7l 74 3C LIST OF FIGURES (continued) Figure l4 l5 l6 l7 l8 l9 20 21 22 23 24 25 26 27 28 29 30 Page Appearance of metabolites in renal cortical tissue slice incubations -------------------------------------- 76 Labeling of hepatic metabolites with Na23SSO4 ---------- 78 Labeling of renal cortical metabolites with Na235504--- 80 Production of metabolites by hepatic tissue slices in the presence of p-nitrophenyl sulphate ----------------- 83 Production of metabolites by renal cortical tissue slices in the presence of p-nitrophenyl sulphate ------- 85 Macromolecular binding in tissue slices in the presence of p-nitrophenyl sulphate------— ----------------------- 87 Production of metabolites by hepatic tissue slices in - the presence of 2,6-dichloro-4-nitrophenol ------------- 90 Production of metabolites by renal cortical tissue slices in the presence of 2,6-dichloro-4-nitrophenol--- 92 Macromolecular binding in tissue slices in the presence of 2,6-dichloro-4-nitrophenol- ------------------------- 94 Production of metabolites by hepatic tissue slices in the presence of D(+)galactosamine ---------------------- 97 Production of metabolites by renal cortical tissue slices in the presence of D(+)galactosamine ------------ 99 Macromoleular binding in tissue slices in the presence of D(+)galactosamine ----------------------------------- lOl Post-incubation hydrolysis of metabolites produced by hepatic tissues slices --------------------------------- lO4 Post-incubation hydrolysis of metabolites produced by renal cortical tissue slices --------------------------- l06 EI-MS of the penta-TMS derivatized carbohydrate moiety from the 9.5 min HPLC peak and a glucose standard ------ 109 Selected ion chromatogram of the penta-TMS derivative of the carbohydroate moiety from the 9.5 min HPLC peak and mannose, galactose and glucose standards ----------- lll Macromolecular binding in renal medullary slices ------- 113 ix ACN 2AAF 2AF 4A8 ACS BZ BSA DNA DCB DCNP DPM EI-MS FAB-MS HPLC ip iv LSS MBOCA 2NA NMR PIC A PNPS RCM RNA SC TCA LIST OF ABBREVIATIONS acetonitrile 2-acetylaminofluorene 2-aminofluorene 4-aminobiphenyl aqueous counting solution benzidine bovine serum albumin deoxyribonucleic acid 3,3'-dichlorobenzidine 2,6-dichloro-4-nitrophenol disintegrations per minute electron-impact mass spectrometry fast atom bombardment mass spectrometry high performance liquid chromatography intraperitoneal intravenous liquid scintillation spectroscopy 4,4'-methylenebis(2-chloroaniline) 2-naphthylamine nuclear magnetic resonance tetrabutylammonium phosphate; reagent for paired ion chromatography, acids p-nitrophenyl sulphate radial compression module ribonucleic acid subcutaneous trichloroacetic acid INTRODUCTION A. Background Bladder cancer is a significant cause of human death in indus- trialized nations. Bladder tumors represent about 2% of the tumors of the body and result in some 3% of the cancer deaths each year (Mal- try, l97l). Urban populations have higher rates of bladder cancer than rural populations (Mommsen, l983) and men have a three-fold greater incidence than women (King, l982; Mommsen, l983; Kern, l984). This sex difference is probably related to occupation and personal habits, e.g. smoking (King, l982). In some areas occupation-related bladder cancer accounts for 15-30% of the total (Lower, 1983). This indicates that bladder cancer may be, in part, a preventable disease. Human bladder cancer is also a problem in less developed countries where Schistosoma haematobium is endemic in the water supply. Larvae penetrate the skin and migrate to the bladder where the organism causes a chronic infection (Boyland, l963). Ova are passed in the urine. The disease, bilharziasis, is associated with an increased risk of bladder cancer (Boyland, l963; King, l982) which occurs equally in men and women. The cause is unknown but may be due to the chronic irritation or to the production of carcinogenic compounds (King, l982). Improvement of sanitation may combat bladder disease in these areas. 2 The urinary bladder is susceptible to toxicity and carcinogenicity due to its role as a holding tank for concentrated metabolic wastes. It is a highly distensible organ, whose transitional cell epithelium is unique in its ability to change thickness and cell shape during bladder distension and contraction. The epithelium extends from the bladder, up the ureters and throughout the renal pelvis. Because of its exposure to urine, the transitional cell epithelium is constantly in contact with concentrated metabolic wastes. It is from this cell layer that bladder cancer arises. Bladder tumors vary in their origin and severity. They range from benign papillomas to transitional cell carcinomas, squamous cell car- cinomas, and adenocarcinomas. The latter two groups comprise less than 3% of the bladder cancers (Vidyarthi, l97l). Bladder cancers are rated in severity based on their invasive properties in the bladder, on lymph node involvement and on metastatic behavior. They were thought to originate mainly as papillomas (Vidyarthi, l97l). Recent surveys have demonstrated that non-papillary tumors are more invasive (Kern, l984) and that nonpapillary lesions may be a more frequent precursor of bladder cancer (Brawn, l984). Early detection remains essential for effective treatment. Epidemiologic studies have elucidated the etiology of bladder cancer. Various aspects of lifestyle including smoking habits, coffee drinking, diet, saccharin consumption, and type of employment have been investigated to determine their relationship to bladder cancer. For some of these including smoking, the results are definitive. 3 Cigarette smoking increases the risk of bladder cancer. Relative risk for cigarette smokers increased to l.89-5.8 in men and 2.0-4.5 in women (Cole g£_§l,, 1971; Simon g£_al,, 1975; Howe gt 21,, 1980). It increased in a dose-dependent manner in both males and females according to smoking frequency per day, duration of smoking in years and total life cigarette consumption (Howe g§_al,, 1980). Other tobacco products have also been surveyed. In contrast to the strong positive correlation with cigarette smoking, investigations into other types of tobacco use have had conflicting results. Cole g§_al, (1971) and Nynder g§_al, (1963) reported a weak or absent correlation between bladder cancer and cigar or pipe smoking. However, Howe g£_al, (1980) reported a significant increase in risk for pipe smoking by males and Mommsen e§_gl, (1983) found increased risk due to cheroot smoking by females. No differences were noted for tobacco chewing or snuff use (Wynder g£_gl,, 1963). Any apparent relationship between coffee drinking and bladder cancer may be noncausal (Simon g£_al,, 1975). An increased risk was reported for men but not women (Bross and Tiding, 1973; Cartwright §§_ a1,, 1981; Hartge g£_gl,, 1983; Mommsen §t_gl,, 1983). Conversely, an increased risk for women has also been reported (Simon 2; 31,, 1975; Howe g§_gl,, 1980; Morrison §__al,, 1982). This contradiction, the low positive relationship between bladder cancer and coffee drinking and the lack of dose response indicated that, if an association exists, it is weak (Simon gt_§l,, 1975). Diet is not often included in epidemiologic studies of bladder cancer but may play a role in its appearance. Mettlin and Graham (1979) studied the diet of patients and controls using 29 edibles in addition 4 to coffee, tea, and milk. They reported that the risk of bladder cancer varied inversely with milk and carrot intake. Coffee drinkers, sub- divided into groups by milk drinking habits, had a decreased risk of bladder cancer if they also drank milk. The differences observed were not significant, however. There was an inverse relationship associated with carrot consumption as well. Milk and carrots are high in vitamin A which has been associated with a protective role in lung cancer. The reduced risk of bladder cancer may have been due to vitamin A consump- tion (Mettlin and Graham, 1979). Epidemiologic studies on saccharin usage had mixed results. Saccha- rin enhanced the risk of bladder cancer for men in a dose-dependent manner in one study (Howe g£_al,, 1980), and enhanced risk only for male nonsmokers in another (Cartwright g§_al,, 1981). In the latter investi- gation there was a nonsignificant increase in risk for female nonsmokers and no increase for smokers of either sex. Mommsen g§_al, (1983) re- ported an enhanced risk for nonsmoking women and Simon g§_al, (1975) found no increased risk for women using cyclamates or saccharin. This weak association has been difficult to characterize. Other lifestyle factors have been studied. No increased risk was found for use of hair dyes (Howe g; 21,, 1980; Hartge §5_al,, 1982), use of estrogens (Mommsen g;_gl,, 1983), alcohol drinking habits (Cartwright g§_§l,, 1981) and tea consumption (Simon g£_al,, 1975). Although most lifestyle factors are not associated with an increased risk of bladder cancer, certain occupations clearly stand out. Occupations with in- creased risk of bladder cancer include truck driving (perhaps due to exposure to diesel fumes) (Howe g£_al,, 1980; Silverman g§_gl,, l983), 5 food processing (Howe g; al,, 1980), and jobs in chemical and rubber industry, cable manufacture, textile works, leather works, coal tar and gas industry, and pigment and paint manufacture (Tola, 1980). Although the etiologic agent is unknown in most cases, exposure to arylamines is suspected in most of these industries. Arylamines provide the starting material for the manufacture of dyes used in leather, paper, textiles, and paints. They have been used as antioxidants and as polymeryzing agents in the rubber and cable industries and are a byproduct of the coal tar and gas industries (Tola, 1980). In a few well studied cases, arylamines in the workplace have been associated with an increased risk of bladder cancer. Arylamines had been suspect bladder carcinogens for over 40 years before their carcinogenicity was demonstrated in an animal model. It was another 15 years before epidemiologic studies demonstrated the role of a particular arylamine, benzidine (BZ) (Figure l), in human bladder cancer. In 1894 Rehn, an industrial surgeon, noted an increased inci- dence of bladder tumors in dye workers (Parkes, 1978). The dye industry in Germany was a scant 30 years old. He and others catalogued the cases in the early twentieth century as they developed parallel to the de- velopment of the dye industry. Then, in 1938, Heuper and Wolfe reported the development of bladder cancer in dogs treated with 2-napthylamine (2NA) (Figure l) (Clayson and Garner, 1978). At that time it was the only animal tested which developed bladder cancer from 2NA (Clayson and Garner, 1978) and it has been used to test other industrial arylamines since then. It was not until the 1950's that Case and coworkers (1954) finished a controlled epidemiologic study and definitely demonstrated the role of 2NA and 82 in human bladder cancer. They demonstrated that 2 - naphthylamine @.@ NH2 2 ' aminoiluorene ”NH2 4 - aminobiphenyl H2NHNH2 benzidine CI H I . H N‘©~C—@>NH 4,4 - methylenebls 2 H 2 (2 - chloroaniline) Figure 1. Structures of Aromatic Amines 7 2NA and 82, but neither l-naphthylamine (lNA) nor aniline, were human bladder carcinogens. Because of the efforts of Case and others the industrial use of demonstrated potent arylamine bladder carcinogens has been curtailed. 2NA and 4-aminobipheny1 (4A8) (Figure l) are no longer used at all. 82, although not used in many countries, remains a potential problem because of its use in the manufacture of imported dyes, which retain a residue (Boeniger 22.21:: 1981). Exposure in the workplace is controlled more tightly than before and is monitored by testing for urinary excretion of the parent compound (Meal g£_gl,, 1981). Bladder cancer remains a concern in paint, pigment, dyestuff, chemical, rubber, and cable manu- facture as well as in textile and leather works and the coal tar and gas industries. The sensitivity of humans to arylamine-induced bladder cancer may, in part, be determined by acetylator phenotype. Acetylation is believed to be a detoxication step for arylamine bladder carcinogens. Liver N- acetyltransferase is a polymorphic enzyme in humans and in some recently developed animal models. These models have not yet been used to deter- mine whether or not a different pattern of tumorigenesis will occur in rapid and slow acetylators but human epidemiologic studies suggest a direct relationship between slow acetylators and sensitivity to aryl- amine-induced bladder cancer. Cartwright g£__l, (1982) reported that although no relationship existed between acetylator phenotype and bladder cancer on a population-wide basis, a significant proportion of dye workers with bladder cancer were slow acetylators (96%). Evans g: 21, (1983), also in the United Kingdom, found that bladder cancer in 8 general but not industrial exposure specifically was associated with slow acetylators. Miller and Cosgriff (1983) (Rochester, NY) reported no difference from control in acetylator status of their bladder cancer population as a whole or when subgrouped according to industrial expo- sure. In these studies, the number of exposed workers was small (less than 25), the control groups were from different populations, and patients were not phenotyped at diagnosis. The differences in bladder cancer according to phenotype may have been due to a greater survival of slow acetylators (Evans, 1983). However, differences in the physiology of bladder cancer patients grouped by geographical region has been noted previously (Brown and Price, 1969) and may play a role in epidemiologic studies of this kind. Clearly more research is needed. Fast and slow acetylator animal models will be useful to study the problem. The tumorigenesis of arylamines has been investigated in a number of animal species. Although arylamines differ in their potency and organ specificity, a general pattern of tumorigenesis is observed in each species. In rats, arylamines with two aromatic rings generally cause Zymbal and mammary gland tumors. Benzidine congeners including 82, 3,3‘dichlorobenzidine (DCB), dimethylbenzidine and dimethoxybenzi- dine cause Zymbal gland tumors (IARC, 1971; 1973; 1982b). BZ, DCB, 4A8, dimethylbenzidine and 4,4'-methylenebis(2-methy1aniline) produce mammary tumors (IARC 1971; 1973; 1982b). Arylamines with one aromatic ring produce bladder cancer in rats. These include o-anisidine, m- and p- cresidine and 4-chloro-o-pheny1enediamine (IARC, 1982a). Bladder cancer is also produced in rats by a few arylamines with two aromatic rings, including 2NA (Hicks g§_al,, 1982) and tetrachlorobenzidine (IARC, 1982a). 9 Other animal species have not been as extensively used. Arylamines cause hepatomas in mice and can result in bladder cancer in rabbits and hamsters. Mice are sensitive to hepatomas from 4A8, 82, 2NA, and DCB but not 1NA (IARC, 1973; l982b). Although bladder cancer is observed in rabbits treated with 4A8, BZ and 2NA, the latter two are but weak car- cinogens (IARC, 1971; 1982b). In hamsters 2NA and DCB produce bladder cancer, DCB and 82 produce hepatomas and dimethoxybenzidine produces forestomach carcinoma (IARC, 1973; 1982b). Dogs have been used as a test animal for bladder carcinogens ever since Heuper and Wolfe demonstrated bladder cancer from 2NA in 1938 (Bonser, 1969). 4A8, 82 and DCB also produced bladder cancer in dogs (IARC, 1973; 1982b). In addition, the carcinogenic potency of these amines has been evaluated partly on the basis of the latency in dogs. Using this method, 4A8 is the most potent bladder carcinogen; where administration of 30 9 produced transitional cell carcinoma after 33 months (IARC, l97l). 2NA is also a potent carcinogen, it produces cancer in 30 months following administration of 100-200 g (IARC, 1973). Both 82 and DCB were weak carcinogens. They required 150-325 g and latent periods of 6-10 years (IARC, l982b). Dogs have been a test animal for arylamine bladder carcinogens because they are sensitive to bladder tumors from arylamines, the tumors are of the same type and cell origin as observed in humans, and because carcinogenic potency can be evaluated using the latent period. In spite of the reasons in favor of using dogs in arylamine bladder carcinogenicity testing their use in routine testing has been dis- couraged due to the long latent period and concomitant costs involved 10 (Bonser, 1969). Compounds that are bladder carcinogens in dogs are also carcinogens in rodents. Tests in rodents require significantly less time and resources. In addition, although spontaneous tumors are rare, they are observed in dogs after age 7, making results difficult to interpret (Bonser, 1969). Thus, the use of dogs in long-term testing might be inappropriate. They might be useful, however, in metabolism studies. Metabolism is often necessary for carcinogenesis and differ- ences therein are reflected in differences in sensitivity and target organ specificity. B. Arylamine Metabolism Arylamines are lipophilic compounds which require metabolism for rapid excretion. Metabolism results primarily in detoxication, e.g. ring hydroxylation and conjugation (Clayson and Garner, 1978; Masson e_t_ 31,, 1983) resulting in metabolites that are more polar, more water soluble, and more rapidly eliminated than the parent compound. Some metabolites are common to several species. The major canine urinary metabolite of 82, 2NA, 4A8, and 1NA, the ortho-hydroxy sulphate (Wiley, 1938; Clayson and Ashton, 1956; Sciarini and Meigs, 1958; Clayson g5 21,, 1959) has been identified in rats (2NA, Booth g;_gl,, 1955; 1NA, Clayson and Ashton, 1956; 82, Clayson §3_al,, 1959) and rabbits (1NA, Clayson and Ashton, 1956). The ortho-hydroxy glucuronide is also formed in mice (82, Sciarini and Meigs, 1961), dogs (2NA, Booth g;_al,, 1955; 1NA, Clayson and Ashton, 1956) and rabbits (1NA, Clayson and Ashton, 1956). Arylamines can be detoxified by direct conjugation to the amine. The N-glucuronides of 82 (dog, rabbit, rat, Clayson g§_gl,, 1959; rat, 11 Lynn g§_al,, 1983), 2NA (rat, rabbit, Boyland g£_al,, 1957) and 1NA (rat, Clayson and Ashton, 1956) and sulphamates of 2NA and lNA(rabbit, Boyland gt 31,, 1957) and 82 (Clayson g;_al,, 1959) are examples of the detoxifying reaction. N-Acetylation, sometimes with added ring metabolism is observed in animal species except dogs. Rats, rabbits and mice produce N-acetyl— and N,N'-diacetylbenzidine (Sciarini and Meigs, 1961; Lynn 33 31., 1980, 1983). N-Acetylation with ring hydroxylation produces metabolites such as 4'-acetamido-4-amino-3-hydroxybenzidine sulphate (rat, mouse, rabbit, Clayson g§.gl,, 1959), 2-acetamido-6-naphthol (rat, rabbit, Booth g§_al,, 1955) and monoacetyl-3-hydroxybenzidine sulphate or glucuronide (mouse, Sciarini and Meigs, 1961). Other metabolites in- clude hydroxylation and conjugation at ring positions farther from the amine (Booth _t__l,, 1955; Clayson and Ashton, 1956). Induction of ring hydroxylation also decreases tumorigenicity (Lotlikar g;_al,, 1967). These metabolites are all considered detoxication products because of the decrease in tumorigencitiy, and due to their stability and rapid elimination. Arylamine N-acetylation may play a role in the target organ speci- ficity. In most species, except the dog, a rapid equilibrium exists bewteen the arylamine and the arylamide, favoring the arylamide (Irving, 1979; Kriek, 1979). As described above, these species, including rats, mice, and hamsters develop tumors of the liver, mammary gland, Zymbal gland and occasionally the bladder. This organ specificity may be due in part to arylamine metabolism in the individual organ. Hepatic car- cinogenesis is believed to be due, in part, to sulfotransferase mediated 12 formation of a labile ester from hydroxamic acids (weisburger g§_al,, 1972; Beland g£_al,, 1982; Poirer g§_al,, 1982) and arylamines may require acetylation prior to initiating liver cancer. Zymbal gland tumors may be initiated by another route. The Zymbal gland is sensitive to arylamine and arylamide carcinogenesis and contains a high deacetyl- ase activity in addition to P450 monooxygenase activity (Irving, 1979; Pohl and Fouts, 1983). Thus, arylamides may be deacetylated during or prior to the activation process. Bladder cancer from arylamines and arylamides in dogs may be due, in part, to their inability to acetylate arylamines (Poirer _;__l,, 1963; Kriek and Restra, 1979; Weber g§_gl,, 1980) and their renal deacetylase activity (Lower and Bryan, 1976). The renal deacetylase activity towards 2-acetylaminof1uorene (2AAF), 4- acetylaminobiphenyl, and 2-acety1aminonaphtha1ene was directly related to the susceptibility of the dog to bladder cancer from these arylamides (Lower and Bryan, 1976). Sensitivity to bladder cancer in humans may also be due, in part, to acetylator status (see Background). These observations, indicate that acetylation and deacetylation can alter the target organ specificity of carciongenesis. The carcinogenicity of specific metabolites has been tested. The orthohydroxy metabolites of arylamines have not been consistently positive or negative in the mouse bladder implantation model. 3-Hydroxy- benzidine, 3-hydroxybenizidine sulphate, 2-amino-l-naphthylsulphate, and 2-amino-1-f1uorenol were negative (Allen g$_al,, 1957; Bonser g£_gl,, 1963; IARC, 1982). However, 1-amino-2-naphthol, 2-amino-1-naphthol, 2- amino-l-naphthyl glucosiduronate, and 3-hydroxy-4-aminobiphenyl sulphate produced bladder tumors (Bonser t 1., 1963). The mouse bladder model has! form main Benz benz (Mor prod hydr impl When lhes effe geni reac 13 has been questioned because pellet implantation alone has enhanced tumor formation. Thus, the carcinogenicity of orthohydroxy metabolites re- mains unclear. Other metabolites have also been tested for carcinogenicity. Benzidine metabolites N,N'-diacetylbenzidine and N-hydroxy-N,N'-diacety1- benzidine injected i.p. produced mammary and Zymbal gland tumors in rats (Morton g£_gl,, 1981), N-hydroxy-Z-NA and N-hydroxy-lNA injected s.c. produced pulmonary and hepatic tumors in mice (IARC, 1973) and injected i.p. produced peritoneal cancer in rats (Belman g§_al,, 1968). N- hydroxy-ZNA also produced bladder cancer in mice by bladder pellet implantation (Bonser $3.21,, 1963) and produced bladder cancer in dogs when instilled into the bladder in dimethylsulfoxide (IARC, 1973). These data clearly demonstrate the carcinogenicity of the hydroxylamine. Arylamines require metabolic activation to exert their carcinogenic effect. N-Oxidation is the first step and a requirement for carcino- genic activity (Irving, 1979; Kriek, 1979). N-Hydroxylation is a minor reaction that takes place in the microsomal subcellular fraction in both the cytochrome P450 monooxygenase system and in the FAD containing monooxygenases (Frederick g§_al,, 1982; Cummings and Prough, 1983). Formation of the hydroxylamine does not increase polarity and it must be further metabolized to be excreted. The relative degree of carcino- genicity of 4A8, 2NA, and 1NA is related to the concentration of their N-oxidation products in canine urine (Radomski and Brill, 1970, 1971). Phenobarbital induced N-hydroxylation of 4A8 and 2NA (Uehleke and Brill, 1968) and reduced the latency of bladder cancer from 4A8 in dogs (Mac- Donald £12.31” 1973). The hydroxylamine is not considered the ultimate carcinogen, however. 14 Hydroxylamines may be activated by conjugation with a sulphate or glucuronide. O-Esterification results in an intermediate which rapidly decomposes to a reactive electrophile, the arylnitrenium ion. This ion may be the ultimate carcinogenic species (Kriek, 1979). It has a posi- tive charge on the nitrogen and interconverts at slightly acidic pH with the carbocation (Kadlubar g;_al,, 1981), a resonance structure with the positive charge delocalized over the aromatic ring system. The carbo- cation reacts with nucleophiles at a position ortho to the amine (Morton g§_§1,, 1980) and the arylnitrenium ion reacts at the nitrogen atom. Since the majority of DNA binding occurs via the nitrogen atom (Martin and Ekers, 1980) and is thought to result in the heritable changes involved in initiation, the arylnitrenium ion is believed important in carcinogenesis. Hartman and Schlegel (1979) concluded that the arylnitrenium ion singlet stability is related to carcinogenicity. Molecular orbital calculations indicated that carcinogenic arylamines formed arylnitrenium ions with singlet and triplet states of similar energy, the singlet being a stable species. Noncarcinogenic arylamines formed singlet states of higher energy and less stability than the corresponding triplet states. The carcinogenic or mutagenic potential of substituted arylamines was successfully predicted. This work indicated that aryl- nitrenium ions with enhanced opportunity to react as a singlet, i.e. longer singlet lifetime, would be more likely to act in initiation processes. In addition, calculation of nitrenium ion singlet stability might be useful as a predictive tool. acid ferr 0XY9‘ este tent: to.i wins- but I mou51 acti' tran: type. tran: ence 9609‘ the 1 Sing CYto. and l Form. Sona' 15 Labile hydroxylamine conjugates can be formed from arylhydroxamic acids. Cytosolic N,O-acetyltransferase activity is capable of trans- ferring the acetyl group of a hydroxamic acid from the nitrogen to the oxygen (King and Glowinski, 1983). This produces an extremely labile ester, the N-acyloxyarylamine. It has not been isolated but its exis- tence has been inferred by C-8 guanine adducts formed on addition of DNA to jg_vitro incubations. They lack the acetyl group (King and Glo- winski, 1983). Microsomal N,O-acetyltransferase activity exists as well but has not been as thoroughly characterized. In the rabbit but not the mouse, liver cytosolic N,O-acetyltransferase and N-acetyltransferase activities are on the same enzyme (Glowinski g;_al,, 1980), (N-acetyl- transferase is the polymorphic enzyme responsible for acetylator pheno- type). 009 tissues have neither N-acetyltransferase nor N,O-acetyl- transferase activity (King and Glowinski, 1983). This species differ- ence may be involved in differential sensitivity to arylamine carcino- genesis. Formation of free radicals following oxidation may play a role in the carcinogenicity of arylamines. Free radicals may be formed on single electron transfer by NADH-cytochrome b5 reductase or NADPH- cytochrome c reductase and during peroxidation by horseradish peroxidase and prostaglandin synthase (Mason and Chignell, 1982; Mason, 1982). Formation of free radicals from arylamines has been studied in micro- somal preparations by measuring the electron spin resonance signal. Free radical formation by N-hydroxy-N-methylaminoazobenzene and its congeners paralleled the carcinogenicity of the compounds (Nakayama g5 al:, 1980). Nakayama §£_gl, (1982) found that 2NA readily formed CY Tn arg duc met for hyd The bla. the of ( live hydr 4A8 mlcr thos 0f t1 frOm QIUCL lamln 16 products representing free radical forms of 2-naphthylhydroxylamine and 2-amino-1-naphthol but 1NA produced little free radical signal. Both cytochrome P450 enzymes and mixed function amine oxidase were involved. These free radicals can react with DNA jn_vj£rg_and may be involved in arylamine carcinogenesis. . Labile ester or reactive free radical arylamine metabolites pro- duced by the liver may not be involved in bladder carcinogenesis. These metabolites are not long lived enough to act at sites distant from their formation, such as the bladder. Other metabolites, detoxified by ring hydroxylation and conjugation are excreted in the urine, unreactive. The N-hydroxy-N-glucuronide conjugate, however, may be involved in bladder carcinogenesis. It is stable at physiological pH and releases the hydroxylamine at a slightly acidic pH such as is found in the urine of dogs and man (Kadlubar g£_al,, 1977, 1978; Poupko g§_al,, 1979). Arylamine N-hydroxy-N-glucuronide conjugates are produced by the liver and excreted in the urine. Kadlubar g§_gl, (1977) produced N- hydroxy-N-glucuronide conjugates from the hydroxylamine of 2NA, 1NA, and 4A8 using uridine-5'-diphosphoglucuronic acid fortified canine hepatic microsomes. Poupko g§_al, (1979) repeated these results and isolated those conjugates of 2NA and 1NA from canine urine after administration of the parent compound. The N-hydroxy-N-glucuronide of 4A8 was isolated from canine urine by Radomski g£_al, (1977). In addition, human and rat glucuronyltransferase-dependent metabolism of the three arylhydroxy- 1amines was observed (Kadlubar §£_gl., 1977). Support for a role of these glucuronide conjugates in the etiology of bladder cancer comes from several studies. The N-hydroxy-N-glucuro- nide conjugates are stable at neutral pH but release the hydroxylamine wit env and dent absc out 11880 any. was c Brill 1977) 17 at slightly acidic pH or with e-glucuronidase (Kadlubar g5_gl,, 1977, 1978). Altering urinary pH of rats to 5.7 and 7.7 altered the amount of 2NA recovered as N-hydroxy-N-glucuronide in the expected manner (Kad- lubar g§_al,, 1981). The lipophilic hydroxylamines may react directly with the bladder epithelium as arylnitrenium ions produced in the acidic environment of the urine or may be absorbed by the bladder epithelium and further activated (Kadlubar g;_§l,, 1977). Oglesby g§_gl, (1981) demonstrated that 2NA, N-hydroxy-ZNA, and N-hydroxy-lNA were readily absorbed by rat bladder epithelium at pH 5 and 7 and distributed through- out the body. The N-hydroxy-N-glucuronide of 2NA was absorbed only at pH 5, under conditions of hydrolysis. Their study indicated that these compounds may be absorbed and recirculated and perhaps metabolized or re-excreted. These studies demonstrated that arylhydroxylamines may be delivered to the bladder from the liver in the form of an N-glucuronide. Metabolism by the bladder epithelium and other extrahepatic metabolism needs to be considered as well. C. Extrahepatic Metabolism Early studies indicated that the bladder played a minor role, if any, in metabolic activation of arylamines. N-Hydroxy-ZNA but not 2NA was carcinogenic on direct instillation into dog bladder (Radomski and Brill, 1970). The same was true for 4A8 in mice (Kadlubar g§_gl,, 1977). These early studies indicated that metabolic activation of arylamines was necessary prior to reaching the bladder. When a section of the bladder was protected from exposure to urine, it did not develop 'tumors. Exposed areas of the bladder in the same animal, treated with 2NA, developed tumors (Kadlubar g£_al,, 1977). This demonstrated that th ge th bo Th $01 1111 la‘ bla 11'; whe 1115 met duc cul C011“ 18 the systemic circulation to the bladder did not contribute to carcino- genesis but that exposure to urine was necessary. Recently, metabolism by the urothelium has been investigated more thoroughly. Poupko §;_al, (1983) compared the metabolism of 4A8 by rat, bovine, and canine liver and bovine and canine bladder P450 enzymes. They demonstrated that N-hydroxylation occurred in both bladder micro- somal preparations as well as in liver microsomes from dogs and rats. The relative rates of N-hydroxylation of 4A8, 2NA, and 1NA were corre- lated with their carcinogenic potency (4AB>2NA>1NA). Canine liver and bladder mucosal microsomes metabolized 4A8 at equal rates when norma- lized per nmol P450 but the liver rate was ten times that of the bladder when normalized to microsomal protein. This study demonstrated that bladder mucosal cells may contribute to N-hydroxylation of arylamines. Bladder organ and cell culture have been used to study arylamine metabolism. 2AAF is converted to ring hydroxylated detoxication pro- ducts as well as to 2AF and N-hydroxy-ZAAF in human and rat bladder culture (Moore g£_al,, l982). Glucuronides of the ring hydroxylated compounds were also formed. Bovine urothelial cells activated 2AF, 2AAF, 4A8, 2NA, and 82 to mutagens in S, typhimurium TA98 or TAlOO and 2AF, 2AAF, and 4A8 to mutagens in V79 cells (Hix g§_al,, 1983; Oglesby g§_gl,, 1983). Thus, recent investigations have demonstrated that bladder cells are capable of oxidation, conjugation, and mutagenic activation of arylamines. Parent compound excreted in the urine or re- leased by hydrolysis in the bladder may be metabolized by the urothe- lium. Arylamine metabolism in the bladder may involve prostaglandin synthase. The transitional cell epithelium contains prostaglandin syr an.‘ 1101 pr: th ar. till til 111' 1116' th- th SP Ci ti du Se 0t ”‘61 mi: 19 synthase activity (Brown g§_al,, 1980). Microsomal metabolism of 2- amino-4-(5-nitro-2-furyl)thiazole produced reactive metabolites which bound to added DNA and tRNA (Mattammal g§_al,, 1981). Aspirin reduced prostaglandin production and lesions from N-[4-(5-nitro-2-furyl)-2- thiazolyl]formamide in rat bladder (Cohen g§_al,, 1981). Thus, if an arylamine reaches the bladder unmetabolized, it may be metabolized there and contribute to carcinogenesis. The kidney may also play a role in arylamine metabolism and activa- tion. The kidney contains two distinct regions, cortex and medulla. There is a decreasing gradient of P450 enzymes from the cortex to the medulla and an increasing gradient of prostaglandin synthase activity in the same direction (Zenser g§.al,, 1978; Armbrecht g§.gl,, 1979; Davis g§_al,, 1981). This results in a difference in metabolic potential in the two regions. The proximity of the renal medulla to the bladder has sparked interest in the potential metabolism of arylamine bladder car- cinogens by prostaglandin synthase. Prostaglandin synthase has a combination of two enzymatic activi- ties, fatty acid cyclooxygenase and prostaglandin hydroperoxidase (Mattammal g;_al,, 1981; Marnett, 1981). Arylamines may be oxidized during the second step, prostaglandin 62 reduction. Thus, in the pre- sence of cyclooxygenase inhibitors such as aspirin and indomethacin, other peroxides e.g. cumene peroxide, can substitute for prostaglandin 62. These tools have been used to investigate prostaglandin synthase- mediated arylamine cooxidation. Prostaglandin synthase was capable of cooxidation of arylamines in microsomal preparations from ram seminal vesicle and rabbit renal 20 medulla. BZ, 2AF, 4A8, and 2-amino-4-(5-nitro-2-furyl)thiazole were cooxidized during arachadonic acid metabolism (Zenser g1_a1,, 1979a,b, 1980; Mattammal g1_a1,, 1981; Morton g; 31., 1983). Metabolism resulted in reactive products as determined by DNA binding (Zenser g1_a1,, 1980; Mattamal _e_t_a1., 1981), RNA binding (Zenser _e_t__a_l_., 1980; Mattammal 321; 31,, 1981; Morton 31_a1,, l983), macromolecular binding (Zenser g1_§1,, 1979b) and mutagenesis (Robertson §1_g1,, 1983). Binding decreased in the presence of indomethacin or aspirin (Zenser g1_a1,, 1979a,b, 1980; Mattammal g1_g1,, 1981). Binding increased in the presence of the sub- strates arachadonic acid (Zenser g1_a1,, 1979b, 1980) and cumene hydro- peroxide (Mattammal g1_g1,, 1981). Metabolism of arylamines in renal medullary slices has also been demonstrated. 82 metabolism, measured by macromolecular binding, in- creased with addition of arachadonic acid and was inhibited by aspirin, indomethacin, and meclofenamate (Rapp g1_a1,, 1980). Mixed function oxidase inhibitors metyrapone and SKF-525A did not inhibit binding, demontrating that P450 enzymes were not involved. Renal cortical arylamine metabolism has been observed, using muta- genicity in the Ames test as an index. Both S9 and microsomal fractions from mouse kidney activated 2AAF (Reddy gg_a1,, 1980). S-9 fractions also activated AF to mutagens (Aune and Dybing, 1979; Robertson and Birnbaum, 1982; Sutter g1_g1,, 1982). Renal mutagenic activation of 2AAF but not 2AF was inducible by 3-methy1cholanthrene and 2.3.7.8- tetrachloro-p-dibenzodioxin (Aune and Dybing, 1979; Reddy g1_g1,, 1980). As in the liver, renal activation decreased with increased age (Robert- son and Birnbaum, 1982; Sutter _e_i;a_1_., l982). be ac SU 8C6 bol fra att SUC 21 Production of arylamine metabolites in the renal cortex has not been extensively studied. The cortex contains mixed function oxidase activities (Rush g1_g1,, 1983) and can form glucuronide (Dutton, 1980), sulphate (Mulder, 1981) and glutathione conjugates (Jakoby and Habig, 1980) with xenobiotics other than arylamines. 2AAF oxidation was ob- served in mouse renal S9. Renal hydroxylation of 2AAF was induced by 2,3,7,8-tetrachloro-p-dibenzodioxin, and N-hydroxylation was decreased (diGiovanni, 1979). Booth g1_a1, (1955) noted that rat kidney slices did not metabolize 2NA. The slices were capable of hydrolyzing and acetylating 2-amino-6-naphthol sulphate and deacetylating several meta- bolites. Zenser g1.a1, (1979) investigated BZ metabolism in rabbit hepatic and renal microsomes. They observed no metabolism with either fraction using NADPH or arachadonic acid as cosubstrates. Thus, attempts to demonstrate metabolism in the renal cortex have had limited success and the production of conjugated metabolites has not been demon- strated. The renal cortex may be well suited for metabolism of bladder carcinogens for several reasons. First, it has a direct route of excre- tion of metabolites in the urine. Second, the cortex receives a high blood flow. The kidney represents 0.5% of the body weight but receives up to 20% of the cardiac output (de Wardener, 1973; Cohen and Kamm, 1976). The cortex represents 70% of the renal weight and receives 94% of the renal blood flow (de Wardener, 1973). Third, the renal cortex contains the enzymes required for oxidation. P450 monooxygenase acti- vity toward AF and AAF was mentioned above. Other substrates include phenetidine, aniline, and 4A8 (Jones g1_a1,, 1980). Finally, the potential for conjugation with glutathione, sulphate, and glucuronide 22 moieties exists, as outlined above and by Aitio and Marniemi, Powell and Roy, and Chasseaud (1980). Thus, the kidney has the potential to play a role in arylamine metabolism. D. 4,4'-Methylenebis(2-chloroaniline) (MBOCA)_ MBOCA (Figure 1), an arylamine, is the main component of an indus- trial curing agent. It is used in the manufacture of polyurethane foams, industrial rubber products, e.g. gaskets, belts, gears, solid tires, and consumer products, e.g. heels and soles of shoes, skate wheels. Although MBOCA is no longer manufactured in the USA, it is imported in amounts of 1-3.5 million pounds a year for the 200-400 plants that use it (TSCA, 1983). Exposure to MBOCA presents a health hazard due to its potential human carcinogenicity. MBOCA has been tested for carcinogenicity in rats, mice, and dogs. In 2 year studies, MBOCA caused liver, lung, and Zymbal gland tumors in rats (Steinhoff and Grundmann, 1971; Russfield g1_a1,, 1975; Stula g;- 31,, 1975; Kommineni £1.31,, 1978). Lung tumors from arylamines are not common in rats (Russfield $1.21,, 1975). MBOCA caused hemangiomas and hemangiosarcomas in mice of both sexes and hepatomas in female mice (Russfield g1_g1,, 1975). Dogs developed only transitional cell car- cinoma of the bladder when MBOCA was administered over 8-9 years (con- trol animals had no tumors) (Stula g1_§1,, 1977). MBOCA is thus clearly a carcinogen in animals. MBOCA has been tested ig_vitro for mutagenicity, genotoxicity, and transformation. It is mutagenic in two assays, the Ames test (pro- caryotic, reverse mutations) (McCann §1_a1,, 1975) and the mouse lym- phoma assay (eukaryotic, forward mutation) (U.S. Dept. Health, 1983). 23 MBOCA is genotoxic to mouse and hamster hepatocytes (McQueen g1_a1,, 1981), and positive in the Balb C 3T3 mouse embryo cell transformation assay (U.S. Dept. Health, 1983). It also inhibited DNA synthesis in cell culture (Aust g1_a1,, 1981). Thus, MBOCA may be considered a potential human health hazard based on its widespread industrial use, its demonstrated carcinogenicity in animal models and its 1g_!11§9_ mutagenicity. Human exposure has occurred during manufacture and use. Workers have had quantifiable urinary concentrations (Linch g;_a1,, 1971) and have experienced acute toxicity due to an industrial accident (Hosein and van Roosmalen, 1978). A two-mile wide area surrounding the site of its previous manufacture in Adrian, Michigan was contaminated and pre- school children residing in the area had detectable urinary MBOCA con- centrations. MBOCA has also been detected in the urine of workers' families (Williams, 1979). Because metabolism is required for rapid excretion and for carcinogenic activation, these observations point to the need for in-depth studies of MBOCA metabolism. Metabolism and disposition have been studied in rats and dogs. MBOCA was extensively metabolized and rapidly excreted in both animals. In rats 69-73% of the administered dose was recovered in feces and 22- 29% in urine in 48 hr (Farmer g1_a1,, 1981; Tobes §1_a1,, 1983). Al- though widely distributed, retained radioactivity was highest in the liver. The parent compound represented just 1-2% of the excreted dose in the urine (Farmer g1_g1,, 1981; Tobes g1_a1,, 1983). In the dog, 4 % of the administered dose was recovered in the urine by 24 hours, 0.54% of that as the parent compound (Manis g1_ 1., 1984). Thirty-two percent da 111 ti me di ur $01 1116' the ani as Sen Aha billl 24 of the dose was excreted into the bile, none of it as the parent com- pound. The liver retained the highest tissue concentration (Manis g: a1,, 1984). Extensive metabolism was demonstrated in both rat and dog by the low concentration of the parent compound in the urine and absence of parent compound in dog bile. Excretion in the feces represents an opportunity for reabsorption, enterohepatic circulation, and continued exposure. Few metabolites of MBOCA have been identified. Farmer g1_a1, (1981) hydrolyzed urine from MBOCA-dosed rats with sulphatase-glucuroni- dase, increasing the yield of the parent compound from 1-2% to 3-6%. Two deconjugated metabolites were characterized by their HPLC retention time. Manis and Braselton (1984) identified the major canine urinary metabolite as the ortho-hydroxy sulphate, 5-hydroxy-3,3'-dichloro-4,4'- diaminodiphenylmethane-S-sulphate. This was similar to the major canine urinary metabolite of 82, 4A8 and 2NA (Wiley, 1938; Bradshaw and Clay- son, 1955; Sciarini and Meigs, 1958). Further investigation of MBOCA metabolism will aid elucidating the mechanism of its carcinogenicity. E. Purpose The objective of this thesis project was to test the hypothesis that the liver and kidney both play a role in the metabolism of aryl- amines and that both can produce reactive metabolites. MBOCA was used as a model arylamine and the dog was used as a test animal due to its sensitivity to arylamine induced bladder carcinogenesis. The initial phase of the investigation concerned elucidation of the structure, binding characteristics and mutagenicity of the major MBOCA metabolite 11' Me 011 an, It me: Are 913i 25 identified in canine urine. Determination of the complete structure indicated that MBOCA may be metabolized in the dog similar to other arylamines and would be a good model for metabolism studies. Binding studies demonstrated the reactivity of the metabolite with DNA and protein 1__vitro. Methods developed in this phase of the investigation were then used to elucidate structure and reactivity of metabolites including the major urinary metabolite, produced in liver and kidney in_ 11139, Metabolism and reactivity 1fl_!11§g_were measured in terms of metabolite production, detected by high performance liquid chromato- graphy (HPLC) and macromolecular binding in tissue slices. The results were used as a measure of the potential to produce metabolites and reactive species 1g_1119, The structures of metabolites were elucidated using chemical, physical, and enzymatic methods both during and after incubation. The studies included an investigation of the contribution of glucuronidation and sulphation to binding in hepatic and renal cortical slices. Induction of bladder cancer by arylamines may involve metabolism at a number of sites in the organism. The potential for metabolism in liver, kidney and bladder have been demonstrated using other arylamines. Metabolism in the liver can produce an array of detoxication products, one of which, the N-hydroxy-N-glucuronide, is stable at physiological pH and labile at the acidic pH of the urine (Kadlubar g__a1,, 1977, 1978). It can release a reactive hydroxylamine in the bladder. The renal medulla is physiologically proximal to the bladder and contains prosta- glandin synthase. It is capable of metabolizing arylamines to reactive products that could be excreted in the urine and react with the bladder epithelium (Zenser 21.31,, 1979a,b, 1980). The renal cortex contains 26 mixed function oxidase activity and the capacity for conjugation. Arylamine oxidation in the renal cortex has been observed 1fl_vitro (diGiovanni, 1979). The specific objectives of this investigation are to determine: 1. The structure, protein and DNA binding, and mutagenicity of the major metabolite of MBOCA detected in liver slices and in canine urine. 2. The metabolism of MBOCA in canine liver and kidney slices in terms of a. rate of metabolism. b. types of conjugates produced. c. degree of macromolecular covalent binding. d. the role of specific types of conjugates in covalent binding. METHODS A. Animals Healthy adult male mongrel dogs weighing 11-19 kg were used for slice incubation studies. They had free access to food and water prior to the experiments and were used within 3 days of receipt. Dogs of less than about 1.5 years did not produce the same metabolic profile 10.19 111§9_incubation. Urine for isolating the major metabolite was obtained in a previous investigation (Manis g1_a1,, 1984) where adult, male, conditioned, beagle-type mongrel dogs of ll-17 kg were used. 8. Chemicals Methane sulfonic acid, gold label deuterium oxide, gold label m- cresol, glucuronic acid and diethyl maleate were obtained from Aldrich Chemical Co. (Milwaukee, WI), aqueous counting solution (ACS) from Amersham Corporation (Arlington Heights, IL), 2,6-dichloro-4-nitrophenol (DCNP) from Alpha (Danvers, MA), tetrabutylammonium phosphate (PIC A) from Waters Associates (Milford, MA), D-glucose [U-14C] 13 mCi/mmol, 3550 2 4 C-toluene (4x10 Na 14 43 Ci/mg, 220 mCi/ml from ICN Pharmaceuticals (Irvine, CA), 5 DPM/ml) from New England Nuclear (Boston, MA), anhy- drous sodium sulphate from Mallinckrodt (Paris, KY), B-glucuronidase free-arylsulfatase, B-glucuronidase, Type I DNA, D(+)galactosamine, 27 28 p-nitrophenyl sulphate, 5,5'-dithiobis(2-nitrobenzoic acid), methionine- DL-sulfoximine and deoxycholic acid, sodium salt from Sigma Chemical Corporation (St. Louis, MO) and Soluene 350 and Demilume from United Technologies Packard (Downers Grove, IL). Solvents were HPLC, UV, spectrometric grade and were purchased from Burdick and Jackson (Muske- gon, MI). Water was purified by passing distilled water through a four- bowl Milli-Q (Millipore Corporation) water purification system con- taining a 0.25u filter. 4,4'-[]4C]Methy1enebis(2-chloroaniline) (MBOCA*) specific activity 58 mCi/mmole and 4,4'-methy1enebis(2-chloro[U-]4C]aniline) (14C-MBOCA) specific activity 10.9 mCi/mmole (Amersham Corporation) were provided by the Michigan Toxic Substances Control Commission through the Office of Radiation, Chemical, and Biological Safety, Michigan State University. They were purified by HPLC on a C reverse-phase column in acetoni- 18 trile:water (47/53, v/v) prior to use. 4,4'-Methylenebis(2-chloroani- line) (MBOCA) was a gift from Daniel E. Williams (Division of Environ- mental Epidemiology. Michigan Department of Public Health) and was purified before use in a similar manner. C. Isolation and Characterization of the Major Metabolite in Canine Urine and Hepatic Tissue Slices 1. HPLC characterization Dog urine was collected in the course of a study on the absorp- tion, disposition, and excretion of 4,4-[14C]methy1enebis(2-chloroani- line) 4.3 mCi/mmol (Manis 21_a1,, 1984). Aliquots of dog urine repre- senting 5% of the urine at each timepoint 1/2-6 hours following intra- ‘venous administration of MBOCA* were pooled, brought to 20% acetonitrile 29 (ACN) and 0.001 M PIC A and filtered for analysis by HPLC. A 7.8 mm x 30 cm C18 uBondapak column (Waters Associates) was used. Fractions were collected every 30 sec for 30 min using a 25 min linear gradient with a 4 min lag of from 20-75% ACN:H20, 0.001 M PIC A, at at 4.5 m1/min. The radioactivity of each fraction was determined by liquid scintillation spectroscopy (LSS). Recovery of radioactivity from the column was 100%. 2. Isolation of the major metabolite Dog urine was extracted 3 times with methylene chloride and stored at 4°C until use. The major metabolite was partially purified and concentrated using C18 reverse phase Sep Paks (Waters Associates) which had been activated with ACN and rinsed with water. Extracted urine, in 10 m1 aliquots was applied to a Sep Pak. The metabolite was eluted with 5 ml 30% ACNzHZO. The ACN was evaporated under Né, two such aliquots combined and this procedure repeated. The second eluant was adjusted to 20% ACN, 0.001 M PIC A and filtered for HPLC. This method concentrated 95 ml of urine to about 25 ml. Two HPLC purification steps were used to isolate the metabo- lite. First, the metabolite was eluted at 16.5 min on the 25 min linear gradient described above with a flow rate of 2.5 ml/min. The ACN was evaporated under N2, the PIC A removed using Sep Paks, and the eluant adjusted to 10% ACN. Second, the metabolite was eluted at 14 min on a 10 min linear gradient of 10-40% ACN, 0.010 M ammonium acetate, pH 6.4, with a flow rate of 2.3 ml/min. The HPLC system was as described above, but used different 7.8 mm x 30 cm C18 pBondapak columns for each isola- tion step so that the second column was not exposed to PIC A. Following the second HPLC purification the ammonium acetate was removed using Sep 30 Paks and the metabolite was stored in water at 4°C until use. It was extremely stable under these conditions. 3. Hydrolysis Hydrolysis of the metabolite was attempted using 3 different methods: citric acid, B-glucuronidase, and arylsulfatase. First, purified metabolite was evaporated to dryness under N2, resuspended in 0.033 M citric acid, pH 3.9, and incubated at 37°C for 1 hour. The reaction mixture was extracted 3 times with dichloromethane, and the extracts pooled, dried over sodium sulfate, evaporated under N2, re- suspended in 40% ACNzHZO (v/v) and filtered for HPLC. The metabolite was isolated and collected at 10 min on a 15 min 10-30% linear gradient of ACN:H20 at 1.8 m1/min using a 3.8 mm x 30 cm uBondapak C18 column. The amount of metabolite present was quantified by LSS, and control and experimental values compared. Hydrolysis with B-glucuronidase was performed in 0.075 M phosphate buffer at pH 6.0 using 312 units of enzyme and about 3 ug purified metabolite in a 1.5 ml assay. This was incubated for 1 hr at 37°C before extraction. The HPLC analysis was performed as described above. Phenophthalein glucuronide was used to assess enzymatic acti- vity. Arylsulfatase hydrolysis was performed using 0.005, 0.05, or 0.5 units of enzyme in 0.2 M Tris buffer, pH 7.1 and 2 pg of the meta- bolite in a 3 m1 reaction mixture. The reaction was terminated by the addition of 0.76 ml 0.005 M PIC A and 3 m1 dichloromethane. The reac- tion mixture was extracted three times and the organic phase prepared for HPLC by drying, evaporating, resuspending and filtering as described 31 for citric acid hydrolysis. The metabolite was collected from a 3.8 mm x 30 cm uBondapak column using a 15 min linear gradient from 4 to 75% ACN (v/v), 0.001 M PIC A at 1.8 ml/min and was quantified by LSS. The enzyme was inactivated for control experiments by placing it in a boiling water bath for 10 min followed by a quick cooling on ice. 4. Mass spectrometry a. E1;M§, EI-MS was done by direct probe on a Finnegan 3200 at 70 eV. Because sulfate conjugates are not volatile and do not give good mass spectra the metabolite was acetylated using a method which simultaneously transesterified the sulfate conjugate (Paulson and Portnoy, 1970). Two hundred ul of 40:1 acetic anhydridezmethane sulfo- nic acid were added to 2 pg of the metabolite and incubated at 100°C for 60 min. The reaction was stopped by placing the sample on ice and adding a chip of ice to the vial. After 5 min, the sample was extracted 3 times with 0.5 ml benzene and the extracts pooled and dried under N2. The sample was resuspended in 20 ul dichloromethane for transfer to the probe capillary. An MBOCA control was derivatized and analyzed in the same manner. b. [AB;M§, FAB-MS was done on a Varian CH5 double focusing instrument at 6.5 Kev using an argon atom beam. A KCl saturated gly- cerol matrix was used. The sample of 10-20 pg was applied to the probe tip and dried in a gentle stream of air. 5. Nuclear magnetic resonance The metabolite was prepared for proton nuclear magnetic reso- nance (NMR) by hydrogen-deuterium exchange on the amine. About 500 pg was added to an activated, 020 rinsed Sep Pak. The Sep Pak was rinsed twice with 5 m1 020, allowing 10 min between and after rinses for 32 exchange. The metabolite was eluted with 50:50 v/v ACN:DZO, evaporated to dryness, resuspended in 0.325 m1 0 0, and passed through an 0.2u 2 filter before use. NMR was carried out on a Bruker WM 250 with fourier transform using peak suppression by selective inversion. 6. Covalent binding in vitro to DNA andprotein a. DNA binding. DNA binding was determined under condi- tions of arylsulfatase hydrolysis using 0.05 or 0.5 units of enzyme and 1 mg/ml DNA in a 2 m1 reaction volume. The reaction was terminated and the DNA isolated as described by Mattammal g1_a1, (1981) using 2% potassium acetate in ethanol to precipitate the DNA. Following extrac- tions with ethanol and ether the DNA was resuspended overnight at 2°C in 1 ml 0.015 M NaCl-0.0015 M sodium citrate pH 7.0. The sample was divided for LSS and DNA determinations. DNA was hydrolyzed by addition of 2 volumes 10% TCA and incubation at 100°C 20 min prior to determina- tion of DNA concentration by the Ceriotti method (Ceriotti, 1952). DNA recoveries were about 50%. b. Protein binding. Protein binding was determined under conditions of arylsulfatase hydrolysis using 0.5 or 0.05 units of enzyme. The reaction was stopped by the addition of carrier BSA, PIC A to 0.001 M and 3 m1 dichloromethane. Following 3 dichloromethane ex- tractions protein was precipitated by adjusting the mixture to 10% TCA and placing at 0°C for 5 min. Protein was pelleted and the precipitate extracted with 10% TCA twice, followed by acetone and ethylacetate 3 times each. This was sufficient to reach background radioactivity in the extracts. The pellet was solubilized overnight at 37°C in 1 m1 1 N NaOH and divided for LSS and protein determination. Binding was 33 quantified by LSS following addition of an equal volume of l M acetic acid and protein was determined by the Lowry method (Lowry 31 31,, 1951). 7. Mut3genicity Salmonella typhimurium strains TA100 and TA1538, containing respectively base pair (McCann 31_31,, 1975b) and frameshift (Ames 33 31,, 1973b) mutation in the histidine operon, where chosen for the assay due to their susceptibility to mutation by arylamines and their wide- spread use with compounds similar to MBOCA (Ames 33_31,, 1973a,b; McCann 33_31,, 1975a; Bos 31_31,, 1982). However, TA100 was not viable under the preincubation conditions of the assay and only TA1538 was used. Bacteria for the assay were grown overnight in nutrient broth and were washed and resuspended with 0.2 M Tris, pH 7.1. Tris buffer was used due to the inhibition of the arylsulfatase by phosphate (Ramm- 1er 31_31,, 1964) and resulted in a 30-40% loss of viability. The bacteria were preincubated for 0 or 1 hr with one unit arylsulfatase and 0.4, 4, 40, 80, or 200 pg of the metabolite of MBOCA in a 2 ml volume. Following the incubation 0.5 ml was plated in triplicate on Vogel Bonner E medium (Vogel and Bonner, 1956) using top agar supplemented with 0.05 M histidine and 0.05 M biotin. An aliquot was diluted 10° and 107 fold with phosphate buffered saline and plated in triplicate on nutrient broth plates for a colony count. All plates were incubated at 37°C. Colonies were scored at 36-48 hr. 34 0. Assessment of Metabolism In Vitro l. Incubation procedure Dogs were killed by lethal injection of pentobarbital (80 mg/kg). Kidneys were quickly excised, divided into 4 pieces by trans- verse cuts, rinsed by agitation in cold saline (0.85% NaCl) and placed in ice-cold phosphate buffered medium (slice buffer) containing 96.7 mM NaCl, 7.4 mM sodium phosphate buffer, 40 mM KCl and 0.74 mM CaCl2 at pH 7.4 (Cross and Taggart, 1950). Liver samples were excised, cut into approximately one inch pieces, rinsed in cold saline and placed in a separate container of ice-cold slice buffer. Renal medullary tissue was carefully dissected away from cortico-medullary tissue and placed on ice in slice buffer. All phases of the experiments were initiated within 3 hours. Thin slices were prepared from the renal cortex (400 mg i_5%) renal medulla (400 mg :_5%), or liver (200 mg :_5%) by hand using a slicing block and razor blade. The slices were thin enough to curl when they were picked up with forceps. They were incubated in 4 ml of slice buffer. 14C-MBOCA, 100 nmol/m1 (9.09 x 10'4 uCi/nmol) was added in 20 pl dimethyl sulfoxide (DMSO). Incubations were performed in a Dubnoff shaker at 90 cycles/min at 37°C under 100% 02 for 60 min with liver slices and 90 min with kidney slices. Changes in the above proto- col including slice weight, time, preincubation additions, and substi- tutions are noted where they occur. Incubations were stopped by addi- tion of an equal volume of ice-cold ACN and were placed on ice for 15 min. 35 2. Metabolite sep3ration and quantitation Incubated samples were prepared for HPLC through a multi-step procedure. Slices were homogenized in the incubation media with a Wheaton teflon-glass homogenizer and 1 ml 0.05 M PIC A and 3 ml dichloro- methane were added. The homogenate was extracted 3 times with 3 m1 dichloromethane. The extracts were dried over anhydrous sodium sul- phate, evaporated under N2, resuspended in 150 p1 1:1 ACNzHZO, and filtered through 0.2p nitrocellulose filters (BAS, from Anspec, Ann Arbor, MI). Glass fiber prefilters (Millipore Corporation) were some- times used. Samples were kept at -20°C until use. Total recovery of radioactivity was 90-100% in the extract and 75-85% in the filtrate. Calculations were based on DPM in the incubation. An HPLC method for separation of the metabolites was developed based on their anionic nature as glucuronide or sulphate conjugates. HPLC was performed using M6000 and M45 pumps, a model 720 system con- troller, U6K injector, and model 440 detector (Waters Associates, Mil- ford, MA). Absorbance at 254 nm was monitored and recorded on a Linear chart recorder (Linear Instruments, Corporation, Irving, CA). Separa- tion of metabolites was achieved using a 5p C18 Novapak radial compres- sion column (Waters Associates) with a 35 min linear gradient of 23- 42.5% ACN, 0.0025 M PIC A at 3 ml/min. A second set of conditions was also used, including a 25 min linear gradient of 23-33% ACN, 0.0025 M PIC A followed by a.20 min linear gradient of from 33-42.5% ACN, 0.0025 M PIC A at 2.3 ml/min. During methods development, metabolite peaks were identified by collecting one minute or peak fractions through 60 min and LSS. For metabolite quantitation during the experiments, radio- active peaks only were collected and quantified by LSS using a Packard 36 84600 refrigerated scintillation counter encoded with a predetermined quench curve. Most samples contained radioactivity at 10 times or more background although none less than 3 times background were used for quantification. 3. Assessment of macromolecular covalent binding Macromolecular binding was determined on the extracted homoge- nate. The homogenate was brought to 10% trichloroacetic acid and macro- molecules were allowed to precipitate on ice for 15 min. Following centrifugation the supernatant was discarded. The precipitate was extracted 6 times with 3:1 methanol:ether and 4 times with ethylacetate. Each extraction was a minimum of 15 min on ice and one of the first 6 was overnight at 4°C. Using this protocol background DPM were reached with 4 extractions of 3:1 methanol:ether and 3 extractions of ethyl- acetate. Following the 1ast extraction the precipitates were allowed to dry in a fume hood. They were moistened with 0.3 ml (200 mg) or 0.6 m1 (400 mg) 5% deoxycholic acid overnight and solubilized with 2-3 ml Soluene 350 at 50°C. Fifteen ml demilume was added and binding was quantified by LSS using the internal standard method with 14C-toluene. Binding was expressed as pmol/mg wet weight of tissue. E. Individual Experiments 1. Preliminary investigation of metabolism a. Enzyme concentration dependence. In a preliminary experiment renal cortical slices of 100, 200, and 400 mg and liver slices of 50, 100 and 200 mg were incubated with 0.05 and 0.1 pmol/ml 14C-MBOCA (0.05 pmollml was 1.8x10'3 pCi/nmol). Hepatic slice incuba- tions were 90 min and renal cortical slice incubations were 120 min. 37 Metabolism was evaluated by HPLC as described in section C.2. and by macromolecular binding as described in section C.3. b. Concentration dependence of MBOCA binding_t0 tissue 311333, The effect of MBOCA concentration on binding was determined using 0, 0.005 (1.8x10'2 pCi/nmol - MBOCA*), 0.01 (9x10'3 pCi/nmol - 14c-MBOCA), 0.025 (3.6x10-3 uCi/nmol), 0.05 (1.8x10'3 uCi/nmol), 0.100, and 0.200 (4.5x10'4 pCi/nmol) pmol/ml with 200 mg liver incubated 60 min and 400 mg renal cortex incubated 90 min. iMacromolecular binding was determined after dichloromethane extraction as described in section C.3. l4 Controls were inactivated with ACN prior to adding C-MBOCA. c. Time gependence of metabolism. Appearance of metabo- lites was studied in liver slices 100 mg :_5%, 0.05 pmo1/m1 14C-MBOCA (1.8x10'3 pCi/nmol), 0-90 min and in 400 mg kidney, 0.1 pmol/ml 14C- MBOCA, 0-120 min. Time dependence of macromolecular binding was deter- mined in 200 mg liver or renal cortical slices over a 0-90 min time 14 period for liver, and 0-120 min for kidney using 0.1 pm01/m1 C-MBOCA. 2. Elucidation of metabolites - sulfation a. Incubation with Na93550A. Renal cortical and hepatic slices were incubated as described in section 0.1 except 100 nmol/ml 35 6 8 cold MBOCA and 2 mM Na2 S04 (8x10 DPM and 7.4x10 DPM) were used. Incubations were prepared for HPLC as described in section C.2. One- minute or peak fractions were collected across the gradient (36 min- kidney; 32 min-liver) and DPM were quantified by LSS using the quench l4 curve for C. As a control, CPM were measured 0-167 KeV and found to be no different than the 0-156 KeV used for 14C. 38 b. Incubation with p-nitrgphenyl su1phate (PNPS). Renal cortical and hepatic slices were preincubated 10 min in 0, 0.1, 0.5, and 2.0 mM PNPS under incubation conditions described in 0.1. prior to the addition of ‘4 C-MBOCA. Metabolism was evaluated as described in sec- tions 0.2-0.3. c. Incubation with 2,6-dich10r0-4-nitrophenol (DCNP). Renal cortical and hepatic slices were preincubated 10 min in slice buffer containing 2 mM NaZSO4 and O, 0.1, 1.0, 10.0, and 100.0 pM DCNP 14C- under incubation conditions described in 0.1. prior to adding MBOCA. Metabolism was evaluated as described in sections D.2.-D.3. 3. Elucidation of metabolism - gjucuronides l4 a. Incubation with C-glucose. Hepatic liver and renal cortical slices were incubated as described in 0.1. except that 100 14 7 nmol/ml cold MBOCA and C-glucose (3.7x10 8 DPM, 1.29 pmol) were used. In another experiment 1.85x10 DPM (6.47 pmol) 14C-glucose were used with hepatic slices. Incubations were prepared for HPLC as described in section 0.2. Fractions were collected across the gradient and DPM were quantified by LSS. b. Incubation with D(+)galactosamine. Renal cortical and hepatic slices were preincubated 10 min in 0, 0.6, 3.0, or 15.0 mM D(+)galactosamine under incubation conditions described in 0.1. prior to adding 14C-MBOCA. Metabolism was evaluated as described in sections D.l.-D.3. 4. Elucidation of metabolism - glutathione Preliminary studies were done to inhibit glutathione conjugate formation by depleting available glutathione or inhibiting its forma- tion. The decrease in nonprotein sulfhydryl groups due to the treatment 39 was determined on a second set of slices preincubated 10 min with the inhibitor. Nonprotein sulfhydryl groups were determined by the method of Ellman (1959). Briefly, tissues were homogenized, then precipiated with 20 vol 6% TCA and centrifuged to pellet the precipitate. For the colorimetric reaction 0.5 m1 of the supernatant was used with 2.0 ml 0.3 M Na2P04 pH 8.2 and 0.5 ml 0.04% 5,5'-dithi0-bis(2-nitr0benzoic acid) in 10% Na citrate. Optical density at 412 nm was determined imme- diately and compared to a standard curve of 0-50 pg/ml reduced gluta- thione. a. Incubation with diethyl maleate. Renal cortical or hepatic slices were preincubated 10 min with 0, 0.2, 1.0, or 5.0 mM 14c-MBOCA. diethyl maleate under incubation conditions before adding Metabolism was evaluated as described in sections C.1.-C.3. b. Incubation with methionine sulfoximine. Renal cortical and hepatic slices were preincubated 10 min with 0, 0.2, 2.0, or 10.0 mM methionine sulfoximine before adding 14C-MBOCA. Metabolism was evalu- ated as described in sections C.1.-C.3. 5. Post-incubation analyses For post-incubation analysis the dichloromethane extracts for each tissue described in section C.2. were pooled and stored at -20°C until use. Extracts were divided, evaporated under N2 and resuspended in the appropriate medium for each test. a. Citric acid hydrolysis. Dried extracts were resuspended in 2 m1 0.075 M phosphate buffer pH 7.0. Samples received 50 or 100 pl 1 M citric acid (final pH 3.60-4.70). All tubes were incubated 60 min at 37°C in a shaking water bath. Samples were neutralized with 4 drops 40 0.1 N NaOH. Tubes were extracted and analyzed as described in section C.2. following the addition of 0.2 ml 0.05 M PIC A. b. B-glucuronidase hydrolysis. Dried extracts were resus- pended in 2 ml 0.075 M phosphate buffer pH 7.0. Samples received 500- 100 units glucuronidase and all tubes were incubated 4 hours at 37°C in a shaking water bath. Following addition of 0.2 ml 0.05 M PIC A the tubes were extracted with dichloromethane, prepared for HPLC, and ana- lyzed as described in section C.2. c. Arylsulfatase hydrolysis. Dried extracts were resus- pended in 2 ml 0.2 M Tris-HCl buffer at pH 7.1. Samples received 4-8 units of arylsulfatase and all tubes were incubated at 37°C, 4 hours in a shaking water bath. Samples and controls were extracted, prepared for HPLC and analyzed as described in section C.2. following addition of 0.2 ml 0.05 M PIC A. d. Thermal hydrolysis. Dried samples were resuspended in 2 m1 0.075 M phosphate buffer pH 7.0. Controls were placed in the re- frigerator and samples were placed at 37°C in a shaking waterbath for 24 hr. After adding 0.2 ml 0.05 M PIC A samples and controls were ex- tracted, prepared for HPLC, and analyzed as described in section C.2. 6. Mass spectrometry a. Isolation of peaks from in vitro incubations. Metabo- lites eluting at 9.5 min (liver) and 15.5 min (kidney) were isolated from 13_11333_incubations for mass spectrometry. Incubations were homogenized and extracted as described in section C.2. The dichloro- methane extracts from 3 incubations were pooled, dried and resuspended in 5 ml H20. This was passed over a C18 Sep Pak (Waters Associates) 41 previously activated with 20 m1 ACN and equilibrated with H20. The Sep Pak was rinsed with 10 ml water and the metabolites eluted with 1:1 ACN:H O. The ACN was evaporated under N 2 2 HPLC. A 5p C18 radial compression module (RCM) column (Waters Asso- and the sample filtered for ciates) and a 40 min hyperbolic (#8) gradient was used to isolate the metabolites. The column eluant was evaporated to dryness and resus- pended in ACN for storage. b. Fast atom bombardment mass spectrometry_(FAB-MS). FAB- MS was done on a Varian CH5 double focusing instrument at 6.5 KeV using a xenon atom beam. A KCl saturated glycerol matrix was used. The samples of 1-5 pg were applied to the probe tip and dried in a gentle stream of air. The FAB-MS was done at the MSU Mass Spectrometry Faci- lity which is supported by NIH grant RR00480-13. c. Electron-impact mass spectrometry (EI-MS). The struc- ture of the peak at 9.5 min was investigated by EI-MS. Earlier obser- vations indicated that it was N-conjugated, possibly with a sugar mole- cule. The conjugate was too unstable to derivatize and run directly but the sugar molecule, released on hydrolysis was suitable for derivatiza- tion and mass spectrometry. Thus, 75 nmol of the isolated conjugate was dissolved in water maintained at room temperature for 48 hours. This resulted in complete disappearance of the peak at 9.5 min and concomi- tant appearance of MBOCA as determined by HPLC and LSS as described in section C.3. For EI-MS 60 nmol were derivatized with 1:1:3 trimethyl- silyl imidazole: N,O-bis(trimethylsily1)trifluoroacetamide (BSTFA): pyridine for 20 min at 100°C. This produced a penta-TMS (trimethyl- silyl) derivative. The reagents were evaporated under N2 and the sample 42 was resuspended in dichloromethane. The derivative was analyzed by gas chromatography-mass spectrometry (GC-MS) on a Finnigan 3200 GC/MS with Riber SADR data system. Chromatography was carried out on 2 separate columns, 1% 0Vl7 and 1% 0V1 both on gas chrom 0 using 30 ml/min helium and a temperature program of l40-260°C at 12°C/min beginning 0.5 min after injection. This method was adequate for detection of 3 nmol of the derivatized sugar. Mass spectra were obtained at 70 eV. 7. Renal medullary metabolism Renal medullary metabolism was evaluated through binding studies done on 400 mg tissue incubated 90 min. These studies included the substrate concentration dependence of binding and binding in the presence of 0.2 or 1 mM arachadonic acid, 0.28 mM indomethacin or 1.0 mM aspirin. Incubations with added compounds were done under yellow lights due to the light sensitivity of arachadonic acid. Arachadonic acid was diluted in ethanol and stored in the dark, under N2, in the freezer. Its purity was tested prior to use. TLC on LHP-K plates was done using 99:1 ethylacetatezacetic acid and visualizing in iodine vapors. Ara- chadonic acid had an Rf of 0.67 under these conditions. F. Statistical Methods Differences from control values were determined at p50.05 by one- way or two-way analysis of variance. Metabolites from renal cortical and hepatic slice incubations with added inhibitors or enhancers of conjugation and associated macromolecular binding studies were analyzed. The least significant differences test was used for individual compari- sons where the analysis of variance indicated that differences existed. 43 Renal medullary metabolism was evaluated by one-way analysis of variance in arachadonic acid studies and by a t-test when indomethacin and aspirin were used. RESULTS A. Isolation and Characterization of the Major Metabolite Detected in Canine Urine and Hepatic Tissue Slices 1. HPLC characterization The major metabolite of MBOCA in canine urine and hepatic tissue slices was initially characterized in urine by HPLC and LSS. Figure 2 illustrates a profile of radioactivity which eluted from the column under gradient conditions and indicated a major peak at 16.5 min. This peak contained an average of 75% of the eluted radioactivity in the urine of the 2 dogs tested. Under these conditions MBOCA eluted at 22.5 min. 2. Hydrolysis Both chemical and enzymatic hydrolysis of the major metabolite were used to elucidate the structure of the isolated compound. Mild acid conditions using citric acid and enzymatic hydrolysis with B- glucuronidase have been used to hydrolyze N-hydroxy-N-glucuronides and O-glucuronides 0f arylamines, respectively (Kadlubar 33 31,, 1978). Neither of these treatments altered retention time, peak height, or the amount of the major urinary metabolite of MBOCA recovered from the reaction mixture (data not shown). Arylsulfatase hydrolysis, however, resulted in a time and enzyme concentration-dependent decrease in the amount of the metabolite 44 45 .cmm: mm: Ammpmwuomm< mgmpmzv asapou xmamncom: mpg sop Eu om x as m.n < .cws\PE m.e am < UH; : —oo.o .Lmumznmpwcwwcoumum Nmmimm .mm_ ewe a a sum: pcm_cmcm gooey. :ws mm a sect umuomppou mcowpumce uwm on do mwmzpmcm mm; mgu mo uopg m>wp -mpcmmmeamg < . «F up .v.: o_ om mp mm mmwcopouwop\.oc N m .e.: N m o w mumpa\.o: cowpmnaucwwea gum: o_.~ NF.F NN._ em.o om.o —N._ eo._ *»u__wnmw> op mp .u.c Fm Pm NF PP mmwcopou mop\.o: m FF m.u.c m o_ o_ o tape—a\.oc cowpmnsucwmea oz om nmz o.om o.o~ o.o_ o._ _.o AungQ\mnv cowamcucmucoo o RP co uao nmwggmu mm: mnamemoumeoezu .>m om um comm :mmwccwm m :6 xgumsogpumam mmmztxzamcmoumeoecu mom xn vm~z_ece mm: m>_um>wemn one .uemccmum amouzpm a wee xmmq QJQI ewe m.m as» 262» zumwoe mpecnxgonemu vmmwum>mgmc mZHinucwa ecu mo mztmu .mm mczmwu 110 mm deemed a): 0" OD? 00m Own -..l—- a — . _ . — L _ . — . _ . — . . . n - . - n - q H: O 5:. n... 3 gone So: 62 006 8“ CNN 0.— pptlt—l L _ . _ . . _ _ _ . _ . 3 s . - . mm! on; 0" 8' 00M Ofln - - _ p — . _ . — . - . — . _ n _ . _ p — . — . — O i d 1‘ 1 33:3 . OO— OOM DON ONN OG— optkt— - . .-_ . _ . — . p . _ a I? t. ._ . _ I. t O max enigma: Ausuew! 111 .eoeoodmee eooea ewe m.o meweeemoe eee\u.~_ um odoomioep mo Eeemogq mgzuegmaewu e use anpm; cwe\PE om mcwmz m~>o mp :6 use cupcemu mm: xcgmcmoyeEoggu .mnemuceum mmouzpm new amouumpmm .mmoccme use xmma Uba: are m.m exp 50;» xumwoe mumccxgonemu m>wum>wemu mzeimucma 6:» oo aeemoumeoecu cow umpompmm .om mgzmwu mm oe=m_e Omuo mane arm mm"? ovum Noun EN 5: n20 05: 112 nv oe an an mm on m. o. n :03 b p P L p L b b n a c j .5... 2.3 goon oz: oo§~ H H N h h H H M - OO— O 401 333.9 cox—a n n H F F H H F wo— ouo.uo.ou cox—u H n H J u n u r u nova— 30.202 cox—N oo— Atgsuetu! eAgtoleJ 113 Figure 30. Macromolecular binding in renal medullary slices. N=4 except (a) where N=5 and (b) where N=3. Standard error bars are shown. fissignificantly different from control, p50.05. 114 .“ U! '0 pmol I mg wet weight 9 u: 0 o 0.2 1.0 0.28 1.0 (mM) control arachadonic indo- aspirin . acid methacin Figure 30 DISCUSSION A. Characterization of the Major Metabolite in Canine Urine The major metabolite of MBOCA in canine urine was an ortho-hydroxy sulphate, the same type of major metabolite excreted by dogs after administration of other arylamine bladder carcinogens. The metabolite represented about 75% of the excreted radioactivity (Figure 2) and was hydrolyzed by arylsulfatase in a time and enzyme concentration-dependent manner (Figure 3) but not by B-glucuronidase or citric acid. The EI spectrum indicated that the metabolite was ring hydroxy- lated. It provided evidence of the addition of 5 substituents during derivatization and transesterification, possible only if one substituent was added to a ring position. The molecular ion data obtained from FAB agreed with the molecular weight of a hydroxylated MBOCA sulfate with K+ counter ion and K+ adduct (Figure 4). Proton NMR data (Figure 5 and Table 1) also supported the proposed structure by providing evidence that the ring substitution was ortho to an amine. These data demon- strated that the major metabolite of MBOCA in dog urine was 5-hydroxy- 3,3'-dichloro-4,4'-diamino-diphenylmethane-S-sulfate. Several other aromatic amine carcinogens are excreted in urine primarily as the ortho-sulfate conjugate. Wiley (1938) found that dogs excreted 2-naphthylamine as the sulfuric acid ester of 2-amino-l-naph- thol. The major canine urinary metabolite of 4-aminobiphenyl was 115 116 identified as 4-amino-3-diphenylyl sulfate (Bradshaw and Clayson, 1955) and was hydrolyzed to 3-hydroxy-4-aminobiphenyl. Benzidine was excreted from dogs primarily as 3-hydroxybenzidine sulfate (Sciarini and Meigs, 1958; Troll and Nelson, 1958). That MBOCA had a similar major urinary metabolite indicated that it was metabolized in a similar manner. Metabolism of arylamines can result in mutagenic activation. However, in the present study the major metabolite of MBOCA was not mutagenic in Salmonella typhimurium TA1538 under conditions of aryl- sulphatase hydrolysis (Table 2). The assay conditions were harsh, as demonstrated by the loss of viability in the Tris buffer and may have determined the outcome of the test. TA1538, a frameshift mutant has been used successfully to determine the mutagenicity of 2AF, 4AB, BZ, 2AAF (Ames gt_gl,, 1973a; Fouarge §t_al,, l982), N-hydroxy-ZAF (Ames et 31,, 1973b), and N-hydroxy-Z-AAF (white gt_a1,, 1983) but was not as sensitive as the base pair mutants TA100 and TA1535 to 2NA and MBOCA (Ames gt_§l,, 1973a; McCann gt_§l,, 1975). TA1535 was selected for this study due to its sensitivity to arylamine mutagens, its low background reversion rate and its superior survival under the incubation conditions. The limited quantity of substrate precluded further evaluation of the mutagenicity of the major metabolite in other tester strains. Bacterial mutagenicity may be a Specialized measure of reactivity. Reversion frequency depends on the tester strain and the activation system. Moreover, quantitative determination of carcinogenesis cannot be made from the test results (International Commission, 1982). The DNA adduct responsible for mutagenicity may be different from the adduct involved in carcinogenicity. The only DNA lesion identified in TA1538 117 treated with 2AAF was a 2AF adduct (Beland §t_al,, 1982). This was the same as the major adduct in rat tissues and the only one identified in female rats and in tissues not sensitive to 2AAF carcinogenicity (Beland gt_al,, 1982). In addition, the reversion rate for 4AB, 2AF and 2NA (2AF>2NA>4AB) did not correlate with carcinogenicity in the urinary bladder (4AB>2AF>2NA) but rather with the amount of mutagen bound per 106 nucleotides (Kadlubar g__al,, 1982). Although bacterial mutageni- city may be used as a screening method for carcinogens, it may be a measure of DNA binding under physiological conditions. The ortho-hydroxy sulphate of MBOCA bound to both DNA and protein jn_vitrg_upon hydrolysis with arylsulfatase (Figures 6 and 7). This indicated that the ortho-hydroxy MBOCA metabolite was a reactive com- pound, detoxified by conjugation with the sulphate. Its role in bladder carcinogenesis may be limited by urinary hydrolysis and Boyland and Williams (1960) demonstrated that human urinary arylsulfatase activity did not hydrolyze 2-amino-l-naphthol sulphate. Arylsulphatase A and B are excreted in the urine (Dodgson and Spencer, 1957) and increased arylsulfatase A activity has been observed in patients with urologic cancer (Kosugi 22.91:, 1983). This has been further investigated be- cause aromatic sulphates are hydrolyzed by the A form (Kosugi gt_§l,, 1983). The increased activity was due to an arylsulfatase A isozyme not detected in normal urine (Kosugi gt_al,, 1983) and may reflect an in- crease in arylsulfatase activity due to the malignancy (Maru gt 21,, 1980). Thus, arylamine sulphates may not be hydrolyzed by the urinary enzyme and increased urinary arylsulfatase activity may be a marker for the urologic carcinoma and not a participant. 118 The major MBOCA metabolite in canine urine, the ortho-hydroxy sulphate, was a detoxication product with a structure comparable to the major canine urinary metabolite of other arylamine bladder carcinogens. This demonstrated that MBOCA could be metabolized in a manner similar to those arylamines. It was of interest to determine the roles that liver and kidney played in the production of this and other conjugates and the roles they played in production of reactive metabolites. B. In Vitro Metabolism of MBOCA by Liver and Kidney Slices In order to study the potential roles of individual organs in MBOCA metabolism, they must be isolated from total body metabolic capacity. This can be done by using purified enzymes, isolated liver cell or renal tubule preparations, tissue slices, or organ perfusion (Jones gt_al,. 1980). Broken cell, microsomal and purified enzyme preparations produce Phase I metabolites unless cosubstrates for Phase II metabolism are added. This does not take advantage of cellular compartmentalization and limits the variety of conjugated metabolites produced at one time. Whole cell and organ preparations offer the advantage of obtaining a complement of Phase II metabolites. Tissue slices were used in this investigation because they provided a rapid, convenient method of obtaining a whole cell preparation capable of producing a spectrum of conjugated metabolites. Unlike organ perfusion, tissue slices may be used to study several sets of experimental conditions simultaneously and without possible carryover in metabolism or binding. Tissue slices have limitations, however. Slice thickness and media hinder oxygen diffusion and can alter metabolism (Cohen and Ramm, 1976). In addition, xeno- biotic delivery is not the same as j__vivo. In spite of these 119 limitations tissue slices represented a convenient method for investi- gating MBOCA metabolism. An HPLC system was developed to separate metabolites in order to investigate conjugate structure and the effects of various compounds on their production. Most arylamine conjugates are glucuronides or sul- phates and are ionized in aqueous media. An ion-pair reagent such as tetrabutylammonium phosphate (PIC A) combines with the ion, forming a non-ionic pair which is retained on a reversed-phase column. Reversed phase chromatography was used because the more polar metabolites eluted prior to the parent compound, MBOCA (Figures 8 and 9). This method separated three MBOCA conjugates sensitive to B-glucuronidase (Figure 26) demonstrating the power of the technique and allowing investigation of these metabolites individually. These three metabolites (10.1, 11.5, 12.2 min) did not change in parallel during experiments to study the production of reactive metabolites and their separation was essential to their structural elucidation. MBOCA metabolism to reactive products was determined in terms of macromolecular or nonextractable binding. Metabolism can produce reactive products that are extremely unstable and bind to the organelle in which they are produced. They may also migrate within the cell and react with cell membranes, proteins, RNA and DNA (Gillette and Pohl, 1977). A correlation between covalent binding and carcinogenesis has been demonstrated for related alkylating agents, azo dyes, and poly- cyclic hydrocarbons (Brookes, 1977) and covalent binding may play a role in mutagenesis and cell necrosis as well (Cohen, 1983). Nonextractable binding has been used to evaluate formation of these reactive species. It includes covalent binding and high affinity noncovalent binding but 120 omits lipid bound metabolites that are extracted with organic solvents (Gillette and Pohl, 1977). Although it may therefore underestimate total covalent binding, it is a simple, rapid, and widely used method. Nonextractable binding is used here to determine covalent binding. 1. Macromolecular binding in hepatic and renal tissue slices 14 Macromolecular binding of C-MBOCA occurred in both liver and kidney, demonstrating that the renal cortex and medulla metabolized MBOCA to reactive products. ‘4 C-MBOCA binding increased with time (Figure 12) and enzyme concentration to 0.1 pmol/m1 (Figures 10 and 11). It was not an artifact of incubation with tissue slices as demonstrated by the low binding in acetonitrile-poisoned controls (Figures 10 and 11). Increasing the substrate concentration to 0.2 pmol/m1 produced no further increase in binding and the 0.1 pmol/m1 conditions (200 mg liver, 60 min; 400 mg kidney, 90 min) were used for further study. Renal medullary metabolism was low. Rapp et_gl, (1980) demonstrated a linear increase of BZ binding to rabbit medullary slices up to 6 pmol/mg at 0.2 mM arachadonic acid. In the present study MBOCA binding to canine medullary slices was not increased at 0.2 mM and reached just 2.2 pmol/mg at 1.0 mM arachadonic acid (Figure 30). These differences may be due to differences in the tissues from rabbit and dog or the difference in incubation conditions. However, MBOCA was not metabolized to tRNA binding species by prostaglandin synthase jn_yitrg_ under conditions where B2 and 2AF were highly bound (Morton gt_al,. 1983). Thus, MBOCA may not be as readily cooxidized as other aryl- amines. 121 Macromolecular binding was investigated further in the kidney cortex and liver. Alteration of incubation conditions by addition of D(+)galactosamine or PNPS had little effect on binding but did alter metabolite production. D(+)galactosamine inhibits glucuronidation two ways. It traps uridine nucleotides as UDP-galactosamine and that com- pound inhibits UDP-glucose dehydrogenase (Schwarz, 1980). Trapping uridine nucleotides could decrease glucosidation as well. D(+)galacto- samine inhibited production of metabolites in liver (Figure 23 - 9.5, 10.1 and 12.2 min) and kidney (Figure 24 - 9.5 and 12.2 min) without altering macromolecular binding (Figure 25). This indicated that these metabolites were not involved in macromolecular binding and were de- toxication products. PNPS also decreased metabolites in liver (Figure 17 - 9.5, 10.1, 12.2) and kidney (Figure 18 - 15.5 min) without altering binding (Figure 19). PNPS contains sulphate group potential comparable to that of the metabolic sulphate donor, 3'-phosphoadenosine-5'-phosphosulphate (PAPS) and can enhance the endogenous pool of PAPS by phenolsulpho- transferase mediated transulfation of 3'-phosphoadenosine-5'-phosphate (Schwartz, 1980). Concentrations of 0.05-4 mM PNPS have been reported to stimulate sulphation of l-naphthol up to 275% in isolated rat liver cells (Schwartz, 1980). In the present experiment PNPS did not enhance production of any metabolite previously labelled with 355 (Figures 15 and 16, vs. 17 and 18). Phenolsulphotransferase is a high affinity, low capacity enzyme (Koster _t_gl,, l982b) and may have been saturated under the incubation conditions. The peak at 9.5 min increased over 300% in the kidney and decreased significantly in liver without increasing or 122 decreasing binding. Thus, it was not a reactive metabolite. These experiments indicated that peaks at 10.1, 12.2, and 15.5 min were also unreactive. Inhibition of sulphation was also used to investigate MBOCA metabolite reactivity and structure. DCNP is a selective inhibitor (Koster gt_gl,, l982a) active towards phenol, harmol, or N-hydroxyphena- cetin sulphation (Mulder, 1981). Glucuronidation is not inhibited, however. Thus, inhibition of phenolsulphotransferase activity towards harmol in the rat is reflected in an increased production of harmol glucuronide (Koster 93.21:, l982b). In the present investigation, DCNP had no effect on binding in either liver or kidney at low concentrations but inhibited binding at 10 and 100 0M (Figure 22) where production of all but one metabolite fell (Figures 20 and 21). The metabolite at 11.5 min increased up to 500% without altering binding, demonstrating that it was unreactive. The depression of metabolite production and concomitant decrease in binding indicated that metabolism was necessary for binding. It is possible that the depression of metabolism was a result of inhi- bition of oxidative phosphorylation and commitant depletion of cellular ATP. DCNP inhibited rat mitochondrial respiration at 2 0M in our laboratory (Karl Ebner, personal communication). The experiments on alteration of incubation conditions demon- strated that metabolism was necessary for binding and indicated that hepatic and renal metabolites eluting at 9.5, 10.1, 11.5, 12.2 and 15.5 min were detoxication products. These metabolites were more sensitive to altered incubation conditions and were produced in greater quantity than those eluting 22-25.5 min except for the MBOCA orthohydroxy sul- phate, peak 25.5 min, which was the major metabolite in liver tissue 123 slices. The structures of other metabolites characterized by HPLC were further elucidated using post-incubation hydrolysis. 2. Elucidation of metabolites in renal and hepatic tissue slices The post-incubation experiments, in concert with previous experiments, aided in elucidation of the structure of MBOCA metabolites from liver and kidney. Post-incubation hydrolyses indicated that peaks from liver and kidney with the same HPLC retention time were identical. This was also true for experiments on alteration of incubation condi- tions, except for the peak at 9.5 min in the presence of PNPS. Metabo- lites from both organs are therefore considered together, in their elution sequence. The peak at 9.5 min was identified as an N-glucoside. The peak was sensitive to citric acid hydrolysis at pH 3.5-4.2 (Figures 26 and 27), a characteristic of N-conjugates. It decomposed to the parent compound on standing at room temperature, indicating that MBOCA was not oxidized prior to conjugation. It was not labeled by 355 (Figures 15 and 16) and was decreased by D(+)galactosamine (Figures 23 and 24) an inhibitor of glucuronidation and glucosidation. Further, the peak was not sensitive to B-glucuronidase or arylsulfatase which hydrolyze 0- conjugates (Figures 26 and 27). These data indicated that the compound was a glucose or glucuronic acid MBOCA N-conjugate. The EI-MS of the liberated sugar following hydrolysis at room temperature (Figure 28) demonstrated a glucose moiety and identified the structure as an N- glucoside. The peak at 10.1 min was sensitive to both citric acid hydro- lysis and B-glucuronidase (Figure 26), a property of N-hydroxy-N-glucu- ronides (Kadlubar gt_al,, 1977, 1978). It was inhibited by PNPS (Figure 124 17), insensitive to arylsulfatase (Figure 26) and not labeled by 355 (Figure 15) indicating that it was not a sulphate. In addition, it was inhibited by D(+)galactosamine (Figure 23). This evidence supports an N-hydroxy-N-glucuronide structure. The 10.1 min MBOCA metabolite may also contain more than one conjugate. Hydrolysis of one moiety would result in disappearance of the peak. The peak at 11.5 min appeared to be an O-glucuronide because it was insensitive to citric acid hydrolysis but sensitive to B-glucu- ronidase (Figure 26). In addition, it was enhanced up to 500% by DCNP (Figure 20), a compound known to enhance the glucuronidation of phenols while inhibiting sulphation (Koster gt 21,, 1982b). However, there was no apparent decrease in the sulphate conjugates eluting 22-25 min in the presence of DCNP. This discrepancy may be due to the relative propor- tions of metabolites. The ortho-hydroxy sulphate represented the major metabolite in liver slice metabolism and a small decrease in sulfation would not be easily discerned. Since the peak at 11.5 was small, the increase was readily observed. Like the peak at 10.1 min, the peak at 12.2 min was decreased by PNPS in the liver (Figure 17), was insensitive to arylsulfatase (Figures 26 and 27) and was not labeled by 355 (Figures 15 and 16) indicating that it was not a sulphate and did not contain a sulfur atom. It was sensitive to both citric acid and B-glucuronidase (Figures 26 and 27), evidence in support of an N-hydroxy-N-glucuronide structure and was decreased in both liver and kidney by D(+)galactosamine (Figures 23 and 24). Unlike the peak at 10.1 min, the peak at 12.2 min was produced in both liver and kidney. It was produced in quantities great enough to be isolated and further elucidation of this structure is warranted. 125 The experiments were not able to define the structure of the peaks at 15.5 and 22 min. Both were produced in renal slice incubations only. The peak at 15.5 min was not labeled by 35S (Figure 16) was decreased by PNPS (Figure 18), and was insensitive to arylsulfatase (Figure 27). It was not affected by e-glucuronidase, heat or citric acid (Figure 27). For these reasons, it is probably not a glucuronide, glucoside. sulphate or sulphamate. It is produced in large enough quantities to make further structural elucidation possible. The peak at 22 min was labeled by 355 (Figure 16) but was unaffected by other incuba- tion or post-incubation treatments. This metabolite was present in very small quantities, making further characterization difficult. The peak at 23.5 may have been a sulphate-sulphamate double conjugate or a combination of an N-glucuronide and sulphate. It was sensitive to citric acid, a characteristic of N-conjugates, and to arylsulfatase, characteristic of o-sulphates (Figures 27). In addition, it was labeled with 355 (Figures 15 and 16). It was produced in quanti- ties too small to evaluate further. The peak at 25.5 min was the ortho-hydroxy sulphate charac- terized earlier after extraction from canine urine. It was labeled by 35S (Figures 15 and 16) and, not surprisingly, was sensitive to aryl- sulfatase in post-incubation hydrolysis (Figures 26 and 27). In the slice incubation studies it represented the major metabolite in liver but was produced in only small quantities in the kidney. Elucidation of MBOCA metabolite structures demonstrated that both liver and kidney produced glucuronide and sulphate conjugates _i_r_1_ vitro. As outlined in the Introduction, ortho-hydroxy glucuronides and 126 sulphates have been identified for BZ, 1NA, 2NA, and 4AB (Wiley, 1983; Clayson and Ashton, 1956; Sciarni and Meigs, 1958; Clayson et_gl,, 1959; Booth gt_al,, 1955; Sciarini and Meigs, 1961), sulphamates for BZ, 1NA, and 2NA (Boyland gt_al,, 1957; Clayson gt_al,, 1959) and N-hydroxy- N-glucuronides for 2NA and 4AB (Kadlubar _t__l,, 1981; Radomski et_al,, 1977). Characterization of MBOCA metabolites from dog liver and kidney indicated that an o-glucuronide and an o-sulphate were formed. A sul- phamate was not identified although the peak at 23.5 min may contain this moiety. One of the two peaks with identical responses to the incubation and post-incubation studies (10.1 and 12.2 min) may have represented an N-hydroxy-N-glucuronide but only one of them was formed in the liver. Further elucidation of their structures may provide. insight into hepatic and renal formation of the conjugate of a proximate bladder carcinogen. The studies indicated that MBOCA metabolism was similar to that of other arylamines. One MBOCA metabolite, identified as an N-glucoside, appeared to be unique in the spectrum of arylamine metabolites. The N-glucoside has not been reported as a metabolite for other arylamines. The forma- tion of 2NA-N-glucoside was observed in rat tissue slice incubations containing 0.2% glucose and in a 0.2% glucose solution without slices but not in slice incubations without added glucose (Booth gt_al,, 1955). The media used in MBOCA metabolism did not contain glucose. In addi- tion, in the presence of PNPS, production of this peak varied in oppo- site directions in liver and kidney, making nonmetabolic production unlikely. Glucosidation can supplement glucuronidation in mammalian systems when the aglycon load is high (Dutton, 1980) as the UDP-gluco- syltransferases are normal cellular constituents in liver and kidney, 127 with steroids and bilirubin as the natural substrates (Dutton, 1980). It is conceivable that the UDP-glucose dehydrogenase, necessary for conversion of UDP-glucose to UDP-glucuronic acid, may have been inhi- bited under the incubation conditions. Further investigation is war- ranted to determine whether the N-glucoside is a product of slice incubations only. MBOCA metabolites formed in the tissue slices included several detoxication products which could be hydrolyzed in the bladder. The N- glucoside (at 9.5 min) was sensitive to mild acid hydrolysis and de- composed to MBOCA. It may also decompose in the urine, releasing MBOCA which could then be metabolized by the urothelium. Thus, both liver and kidney contained a route of metabolism for delivery of the lipophilic parent compound to the bladder. The minor, 35S-labeled metabolite at 23.5 min was also labile in acidic media but the product of that hydro- lysis is unknown. o-Glucuronides may be hydrolyzed by urinary B- glucuronidase. The o-glucuronides of 2-acetamino-6-naphthol, 2-aceta- mido-l-naphthol, l-naphthol, and 2-amino-l-naphthol were hydrolyzed by human urinary B-glucuronidase (Boyland and Williams, 1960). Whether the arylamine aglycons are reactive or undergo further metabolism is not known. SUMMARY AND CONCLUSIONS The purposes of this investigation were 2-fold: (l) to determine the structure and reactivity of the major urinary metabolite of MOBCA in the dog and (2) to test the hypothesis that the liver and kidney play a role in arylamine metabolism, using MBOCA as a model. The dog was used as a model species because it is sensitive to arylamine-induced bladder cancer. The similarity of MBOCA metabolism to known bladder carcinogens was assessed initially by determining the structure and binding character- istics of the major urinary metabolite in dogs. Detailed enzymatic, chemical, and instrumental analyses including arylsulfatase, B-glucuro- nidase, and citric acid hydrolyses, EI-MS, FAB-MS, and NMR determined that the metabolite was an ortho-hydroxy sulphate. It was therefore similar to the major urinary metabolites in dogs of 4AB, BZ, and 2NA (Wiley, 1938; Bradshaw and Clayson, 1955; Sciarini and Meigs, 1958; Troll and Nelson, 1958). Binding studies indicated that the ortho- hydroxy sulphate of MBOCA was inert but upon hydrolysis with arylsulfa- tase it bound to both DNA and protein, jn_vitro. Thus, the oxidized metabolite was reactive and sulphate conjugation was a detoxification mechanism. The major urinary metabolite was also the major metabolite produced by liver slices but was a minor metabolite in slices from the kidney 128 129 cortex. Both organs produced a variety of metabolites 13_11333_in- cluding an N-glucuside, metabolites sensitive to B-glucuronidase, the ortho-hydroxy sulphate, and another metabolite sensitive to arylsulfa- tase. Arylamine metabolism usually results in production of an N- glucuronide, not an N-glycoside. This product has not been reported as a metabolic product of other carcinogenic arylamines 13_1113_or 13_11333 but has been reported as a metabolite 13_1113_of the non-carcinogenic arylamine 1NA (Clayson and Ashton, 1956). Whether or not this differ- ence in metabolism is physiologically important can be debated. Both metabolites are sensitive to mild acid hydrolysis which would release the parent compound in the bladder where it could then be metabolically activated by the urothelium. However, if differences in metabolism are reflected in differences in carcinogenicity, then the production of the MBOCA N-glycoside may be a reflection of its low carcinogenicity. It is therefore important to determine whether or not the MBOCA-N-glycoside is produced by dogs 13_1113_and whether production of an arylamine N-glyco- side can be used as an index of carcinogenicity. A metabolite from liver and kidney slice incubations, identified as peak 12.2 min by its HPLC retention time, had properties of an N-hydroxy- N-glucuronide. Importantly, it was sensitive to citric acid hydrolysis, a property of Nconjugates, and was sensitive to B-glucuronidase, a property of o-glucuronides and N-hydroxy-N-glucuronides (Kadlubar 33_ 31,, 1977, 1978). The N-hydroxy-N—glucuronide has been proposed as the hepatic metabolite which, when excreted in the urine, decomposes to the hydroxylamine, a proximate carcinogen. Production of the N-hydroxy-N- glucuronide by the kidney could make it an important site of metabolism of arylamine bladder carcinogens. 130 Liver and kidney tissue slices produced metabolites unique to each organ. The liver produced a metabolite (at 11.5 min) sensitive to B- glucuronidase and whose production was enhanced in the presence of DCNP, an inhibitor of phenol sulphation known to enhance phenol glucuronida- tion (Koster 33 31,, l982a). Thus, it appeared to be an o-glucuronide. Liver slices produced a second, uncharacterized metabolite at 13.5 min which was not observed in all experiments. The kidney also produced two unique metabolites neither of which was identified. One at 15.5 min, the major metabolite in kidney slice incubations, was insensitive to all hydrolysis methods used. It may be an unusual metabolite and warrants further investigation. The other, at 22 min, was produced in such small quantities that characterization was difficult. The kidney medulla did not produce HPLC identifiable metabolites on incubation with MBOCA and macromolecular binding was used to determine metabolism to reactive products. Prostaglandin synthase mediated aryl- amine cooxidation has been proposed as a route of metabolism in the kidney medulla. Binding was low, even in the presence of the prosta- glandin synthase substrate arachadonic acid, and inhibitors failed to decrease binding. Morton 33_31, (1983) reported low tRNA binding by MBOCA using microsomal prostaglandin synthase to cooxygenate, thus MBOCA may not be a good substrate. Macromolecular binding was also used to determine the production of reactive metabolites in liver and kidney cortex slices. Binding did not change under altered incubation conditions including inhibition of glucuronidation and glucosidation with D(+)galactosamine and addition of PNPS to stimulate sulfation. Binding decreased in concert with the decrease in metabolism in general on incubation with DCNP. Incubation 131 with D(+)galactosamine, PNPS and DCNP did significantly alter 5 MBOCA metabolites from kidney or liver and demonstrated that they were detoxi- cation products. These studies were therefore successful in demonstrating that both liver and kidney can play a role in metabolism of arylamine bladder carcinogens to conjugated and reactive products. Because of its high concentration of oxidizing and conjugating enzymes, the liver has been considered the primary source of arylamine metabolites and the source of the purported proximate carcinogen, the hydroxylamine. The kidney cortex receives a high blood flow and contains the enzymes required for oxidation and conjugation of arylamines, albeit at a lower concentration than the liver. The present investigation demonstrated for the first time that the kidney cortex has the potential to produce reactive and conjugated metabolites. Thus, the kidney cortex must be added to the potential sites of metabolism for arylamine bladder carcinogens. .ii _nh—— BIBLIOGRAPHY BIBLIOGRAPHY Aitio, A. and Marniemi, J. (1980). Extrahepatic glucuronide conjuga- tion. 13; Extrahepatic Metabolism of Drugs and Other Compounds (T.E. Gram, Ed.), Spectrum Publishers, NY, pp. 365-388. Allen, M.J., Boyland, E., Dukes, C.E., Horning, E.S. and Watson, J.G. (1957). Cancer of the urinary bladder induced in mice with meta- bolites of aromatic amines and tryptophan. Brit. J. Cancer 11; 212-228. Ames, B.N., Durston, W.E., Yamasaki, E. and Lee, F.D. (1973a). Car- cinogens are mutagens: A simple test system combining liver homo- genates for activation and bacteria for detection. Proc. Natl. Acad. Sci. USA 19; 2281-2285. ' Ames, B.N., Lee, F.D. and Durston, W.E. (1973b). An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc. Natl. Acad. Sci. USA 19; 782-786. Armbrecht, H.J., Birnbaum, L.S., Zenser, T.V., Mattammal, M.B. and Davis, B.B. (1979). Renal cytochrome P-450's-electrophoretic and electron paramagnetic resonance studies. Arch. Biochem. Biophys. 121; 277-284. Aune, T. and Dybing, E. (1979). Mutagenic activation of 2,4-diamino- anisole and 2-aminofluorene 13_vitro by liver and kidney fractions from aromatic hydrocarbon responsive and nonresponsive mice. Biochem. Pharmacol. 33; 2791-2797. Aust, A.E., Antczak, M.R., Maher, V.M. and McCormick, J.J. (1981). Identifying human cells capable of metabolizing various classes of carcinogens. J. Supramol. Structure Cell. Biochem. 13; 269-279. Beland, F.A., Dooley, K.L. and Jackson, C.D. (l982). Persistance of DNA adducts in rat liver and kidney after multiple doses of the carcinogen N-hydroxy-Z-acetylaminofluorene. Can. Res. 13, 1348- 1354. Belman, 5., Troll, W., Teebor, G. and Mukai, F. (1968). The carcino- genic and mutagenic properties of N-hydroxy-aminonaphthalenes. Can. Res. 33; 535-542. 132 133 Boeniger, M.F., Stein, H.P., Choudhary, G. and Neumeister, C.E. (1981). Residual benzidine in imported and domestic benzidine dyes. Toxicol. Letters 3; 415-420. Bonser, G.M. (1969). How valuable the dog in the routine testing of suspected carcinogens? J. Natl. Canc. Inst., 271-274. Bonser, G.M., Boyland, E., Busby, E.R., Clayson, D.B., Grover, P.L. and Jull, J.W. (1963). A further study of bladder implantation in the mouse as a means of detecting carcinogenic activity: Use of crushed paraffin wax or stearic acid as the vehicle. Brit. J. Cancer 11; 127-136. Booth, J., Boyland, E. and Manson, D. (1955). Metabolism of polycyclic compounds. 9. Metabolism of 2-naphthylamine in rat tissue slices. J. Biochem. 33; 62-71. Bos, R.P., van Doorn, R., Yih-van de Hurk, E., van Gemert, P.J.L. and Henderson, P.Th. (1982). Comparison of the mutagenicities of 4- aminobiphenyl and benzidine in the Salmonella/microsome, Salmonella/ hepatocyte and host-mediated assays. Mutation Res. 33; 317-325. Boyland, E. (ed.) (l963a). Other environmental factors. 13_The Bio- chemistry of Bladder Cancer, Charles C. Thomas, Springfield, 111., p. 13-17. Boyland, E. (ed.) (1963b). Endogenous Carcinogens. 13; The Biochemistry of Bladder Cancer, Charles C. Thomas, Springfield, 111., p. 63-69. Boyland, E., Manson, D. and Orr, S.F.D. (1957). The biochemistry of aromatic amines. 2. The conversion of arylamines into arylsul- phamic acids and arylamine-fl-glucosiduronic acids. J. Biochem. 33, 417-423. Boyland, E. and Williams, K. (1960). The biochemistry of aromatic amines. 7. The enzymic hydrolysis of aminonaphthyl glucosiduronic acids. Biochem. J. 13; 388-396. Bradshaw, L. and Clayson, D.B. (1955). Metabolism of two aromatic amines in the dog. Nature 113; 974-975. Brawn, P.N. (1982). The origin of invasive carcinoma of the bladder. Cancer 33; 515-519. Brookes, P. (1977). Role of covalent binding in carcinogenicity. 13; Biological Reactive Intermediates (D.J. Jollow, J.J. Kocsis, R. Snyder, H. Vainio, eds.), Plenum Press, NY, pp. 470-480. Bross, I.D.J. and Tidings, J. (1973). Another look at coffee drinking and cancer of the urinary bladder. Prevent. Med. 3; 445-451. 134 Brown, R.R. and Price, J.M. (1969). Tryptophan metabolism in patients with bladder cancer: Geographical differences. J. Natl. Canc. Inst. 13; 295-301. Brown, W.W., Zenser, T.V. and Davis, B.B. (l980). Prostaglandin E production by rabbit urinary bladder. Am. J. Physiol. 333; 452- 458. Cartwright, R.A., Adib, R., Glashan, R. and Gray, B.K. (1981). The epidemiblogy of bladder cancer in West Yorkshire. A preliminary report on non-occupational aetiologies. Carcinogenesis 3; 343-347. Cartwright, R.A., Rogers, H.J., Barham-Hall, D., Glashan, R.W., Ahmad, R.A., Higgins, E. and Kahn, M.A. (1982). Role of N-acetyltrans- ferase phenotypes in bladder carcinogenesis: A pharmacogenetic epidemiological approach to bladder cancer. Lancet 11; 842-845. Case, R.A.M., Hosker, M.E., McDonald, D.B. and Pearson, J.T. (1954). Tumors of the urinary bladder in workmen engaged in the manufacture and use of certain dyestuff intermediates in the British chemical industry. Brit. J. Industr. Med. 11; 75-104. Ceriotti, G. (1952). A microchemical determination of deoxyribonucleic acid. J. Biol. Chem. 133; 297-303. Chasseaud, L.F. (l980). Extrahepatic conjugation with glutathione. In: Extrahepatic Metabolism of Drugs and Other Foreign Compounds (T.E. Gram, Ed.), Spectrum Publishers, NY, pp. 427-452. Clayson, D.B. and Ashton, M.J. (1956). The metabolism of l-naphthyl- amine and its bearing on the mode of carcinogenesis of the aromatic amines. Acta-Unio Internationalis Contra Cancrum 13; 539-542. Clayson, D.B. and Garner, R.C. (1978). Carcinogenic aromatic amines and related compounds. 13; Chemical Carcinogens (C.E. Searle, ed.), A.C.S. Monographs 173, Published in the U.S.A.. PP. 366-461. Clayson, D.B., Ward, E. and Ward, L. (1959). The fate of benzidine in various species. Acta-Unio Internationalis Contra Cancrum 13; 581- 586. Cohen, J.J. and Kamm, D.E. (1976). Renal metabolism: Relation to renal function. 13; The Kidney (B.M. Brenner and F.C. Rector, Eds.), W.B. Saunders Co., Philadelphia, pp. 126-214. Cohen, S.M., Zenser, T.V., Murasaki, G., Fukushima, S., Mattammal, M.B., Rapp. N.S. and Davis, B.B. (1981). Aspirin inhibition of N-[4-(5- nitro-Z-furyl)-2-thiazolyly] formamide-induced lesions of the urinary bladder correlated with inhibition of metabolism by bladder prostaglandin endoperoxide synthetase. Canc. Res. 31; 3355-3359. Cole, P., Monson, R.R., Haning, H. and Friedell, G.H. (1971). Smoking and cancer of the lower urinary tract. N. Eng. J. Med. 331; 129- 134. 135 Cross, R.J. and Taggart, J.V. (1950). Renal tubular transport: Accu- mulation of p-aminohippurate by rabbit kidney slices. Am. J. Physiol. 131; 181-190. Cummings, S.W. and Prough, R.A. (1983). Metabolic formation of toxic metabolites. 13; Biological Basis of Detoxication (J. Caldwell and W.B. Jakoby, eds.). Academic Press, N.Y., p. 1-30. Davis, B.B., Mattammal, M.B. and Zenser, T.V. (1981). Renal metabolism of drugs and xenobiotics. Nephron 31; 187-196. de Wardener, H.E. (Ed.) (1973). The renal cirCulation. 13; The Kidney, printed in Great Britain. PP. 103-114. DiGiovanni, J., Berry, D.L., Slaga, T.J., Jones, A.H. and Jachau, M.R. (1979). Effects of pretreatment with 2,3,7,8-tetrachlorodibenzo- p-dioxin on the capacity of hepatic and extrahepatic mouse tissues to convert procarcinogens to mutagens for Salmonella typhimurium auxotrophs. Toxicol. Appl. Pharmacol. 33; 229-239. Dodgson, K.S. and Spencer, B. (1957). Studies on sulphatases. 15. The arylsulphatases of human serum and urine. Biochem. J. 33; 668- 673. Dutton, G.J. (1980a). Extrahepatic glucuronidation. In: Glucuronida- tion of Drugs and Other Compounds (G.J. Dutton, edTTL CRC Press, Inc., Boca Raton, Fla., p. 149-158. Dutton, G.J. (Ed.) (l980b). Relation of other drug metabolizing path- ways to glucuronidation. 13; Glucuronidation of Drugs and Other Compounds, CRC Press, Inc., Boca Raton, Fla., p. 169-182. Ellman, G.L. (1959). Tissue sulfhydryl groups. Arch. Biochem. Bio- phys. 33; 70-77. Evans, D.A.P., Eze, L.C. and Whibley, E.J. (1983). The association of the slow acetylator phenotype with bladder cancer. J. Med. Gen. 20: 330-333. Farmer, P.B. and Rickard, J. (1981). The metabolism and distribution of 4,4'-methy1ene-bis(2-chloroaniline) (MBOCA) in rats. J. Appl. Toxicol. 1; 317-322. Fouarge, M., Mercier, M. and Poncelet, F. (1982). Mutagenicity of three aromatic amines in the presence of fractions from various tissues. Toxicol. Letters 11; 312-320. Frederick, C.B., Mays, J.B., Ziegler, D.M., Guengerich, F.P. and Kadlu- bar, F.F. (l982). Cytochrome P-450- and Flavin-containing mono- oxygenase-catalyzed formation of the carcinogen N-hydroxy-Z-amino- flu3rene and its covalent binding to nuclear DNA. Canc. Res. 33; 267 -2677. 136 Gillette, J.R. and Pohl, L.R. (1977). A prospective on covalent bind- ing and toxicity. J. Toxicol. Environ. Hlth. 3; 849-871. Hartge, P., Hoover, R., Altman, R., Austin, D.F., Cantor, K.P., Child, M.A., Key, C.R., Mason, T.J., Marrett, L.D., Myers, M.H., Narayana, A.S., Silverman, D.T., Sullivan, J.W., Swanson, G.M., Thomas, D.B. and West, D.W. (1982). Use of hair dyes and risk of bladder cancer. Canc. Res. 33; 4784-4787. Hartge, P., Hoover, R., West, D.W. and Lyon, J.L. (1983). Coffee drinking and risk of bladder cancer. J. Natl. Canc. Inst. 13; 1021-1026. Hartman, 6.0. and Schlegel, H.B. (1981). The relationship of the carcinogenic/mutagenic potential of arylamines to their singlet- triplet nitrenium ion energies. Chem.-Biol. Interactions 33; 319-- 330. Hicks, R.M., Wright, R., Wakefield, J. St. J. (1982). The induction of rat bladder cancer by 2-naphthylamine. Brit. J. Canc. 33; 646-661. Hix, C., Oglesby, L., MacNair, P., Sieg, M. and Langenbach, R. (1983). Bovine bladder and liver cell and homogenate-mediated mutagenesis of Salmonella tyghjmurium with aromatic amines. Carcinogenesis 3; 1401:1407. Hosen, H.R. and Van Roosmalen, P.B. (1978). Acute exposure to methy- lene-bis-ortho chloroaniline (MOCA). Am. Ind. Hyg. Assoc. J. 33; 496-497. Howe, G.R., Burch, J.D., Miller, A.B., Cook, G.M., Esteve, J., Morrison, B., Gordon, P., Chambers, L.W., Fodor, G. and Winsor, G.M. (1980). Tobacco use, occupation, coffee, various nutrients, and bladder cancer. J. Natl. Canc. Inst. 31; 701-713. IARC Working Group in Geneva (1971). 4-Aminobiphenyl. In IARC Mono- gr3phs: Carcinogenic Risk of Chemicals to Man, Vol. 1, pp. 74-79. IARC Working Group in Geneva (1971). Benzidine. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 1. PP. 80-86. IARC Working Group in Geneva (1971). 3,3'-Dimethylbenzidine. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 1, pp. 87- IARC Working Group in Lyon (1973). 3,3'-Dimethoxybenzidine. In IARC Monographs: Carcinggenic Risk of Chemicals to Man, Vol. 4. PD. 41- 47. IARC Working Group in Lyon (1973). 3,3'-Dichlorobenzidine. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 4, pp. 49- 55. 137 IARC Working Group in Lyon (1973). 4,4'-Methylenebis(2-methylani- line). In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 4, pp. 73-77. IARC Working Group in Lyon (1973). 4,4'-Methylenedianiline. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 4, pp. 79- 85. IARC Working Group in Lyon (l973). l-Naphthylamine. In IARC Mono- graphs: Carcinogenic Risk of Chemicals to Man, Vol. 4, PP.'§7-96. IARC Working Group in Lyon (l973). 2-Naphthylamine. In IARC Mono- graphs: Carcinogenic Risk of Chemicals to Man, Vol. 4, pp. 97-lll. IARC Working Group in Lyon (l982). 2,2',5,5'-Tetrachlorobenzidine. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 27, pp.*l4lél46. IARC Working Group in Lyon (l982). Benzidine and its sulphate, hydro- chloride and dihydrochloride. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 29, pp. l49-183. IARC Working Group in Lyon (l982). 3,3'-Dichlorobenzidine and its. dihydrochloride. In IARC Monographs: Carcinogenic Risk of Chemicals to Man, Vol. 29, pp. 239-256. International Commission for Protection Against Environmental Mutagens and Carcinogens (l982). Mutagenesis testing as an approach to carcinogenesis. Mutation Res. 22; 73-9l. Irving, C.C. (l979). Species and tissue variations in the metabolic activation of aromatic amines. lg; Carcinogens: Identification and Mechanisms of Action (A.C. Griffin and C.R. Shaw, Eds.), Raven Press, NY, p. le-227. Jakoby, W.B. and Habig, W.H. (1980). Glutathione transferases. Ip_ Enzymatic Basis of Detoxication, Vol. II (W.B. Jakoby, Ed.), Academic Press, NY, p. 63-94. Jones, D.P., Orrenius, S. and Jakobson, S.W. (1980). Cytochrome P450- linked monooxygenase systems in the kidney. 13; Extrahepatic Metabolism of Drugs and Other Foreign Compounds (T.E. Gram, Ed.), Spectrum Publishers, NY, pp. lZ3-l58. Kadlubar, F.F., Beland, F.A., Beranek, D.T., Dooley, K.L., Heflich, R.H. and Evans, F.E. (1982). Arylamine-DNA adduct formation in rela- tion to urinary bladder carcinogenesis and Salmonella typhimurium mutagenesis. 13; Environmental Mutagens and Carcinogens, Pro- ceedings of the Third International Conference, Tokyo, Japan, l98l (T. Sigmura, S. Kondo and H. Takebe, Eds.), pp. 385-396. 138 Kadlubar, F., Flammang, T. and Unruh, L. (l978). The role of N-hydroxy arylamine N-glucuronides in arylamine-induced urinary bladder carcinogenesis: Metabolite profiles in acidic, neutral and alka- line urines of 2-naphthylamine- and 2-nitronaphthalene-treated rats. In Conjugation Reactions in Drug Biotransformation (A. Aito, Ed.), Elsevier/North-Holland Biomedical Press, Amsterdam, pp. 443- 454. Kadlubar, F.F., Miller, J.A. and Miller, E.C. (l977). Hepatic micro- somal N-glucuronidation and nucleic acid binding of N-hydroxy arylamines in relation to urinary bladder carcinogenesis. Canc. Res. p]; 805-814. Kadlubar, F.F., Unruh, L.E., Flammang, T.J., Sparks, 0., Mitchum, R.K. and Mulder, G.J. (l98l). Alteration of urinary levels of the carcinogen, N-hydroxy-Z-naphthylamine, and its N-glucuronide in the rat by control of urinary pH, inhibition of metabolic sulfation, and changes in biliary excretion. Chem.-Biol. Interactions ga; l29-l47. Kern, W.H. (l984). The grade and pathologic stage of bladder cancer. Cancer 58; ll85-ll89. King, C.M. (l982). The origins of urinary bladder cancer. 13; AUA Monographs, Vol. I, Bladder Cancer (W.W. Bonney and G.R. Prout, Jr., Eds.), Williams and Wilkins, Baltimore. PP. 13-26. King, C.M. and Glowinski, I.B. (l983). Acetylation, deacetylation and acyltransfer. Environ. Hlth. Perspect. 42; 43-50. Kommineni, C., Groth, D.H. and Frockt, I.J. (l978). Determination of the tumorigenic potential of methylene-bis-orthochloroaniline. J. Environ. Pathol. Toxicol. g; l49-l7l. Koster, H., Halsema, I., Scholtens, E., Meerman, J.H., Pang, K.S. and Mulder, G.J. (1982a). Selective inhibition of sulphate conjugation in the rat. Biochem. Pharmacol. 81; 19l9-1924. Koster, H., Halsema, 1., Scholtens, E., Pang, K.S. and Mulder, G.J. (l982b). Kinetics of sulphation and glucuronidation of harmol in the perfused rat liver preparation. Biochem. Pharmacol. B1; 3023- 3028. Kosugi, M., Maru, A., Mitsuhashi, K., Koyanagi, T., Ishibashi, T. and Imai, Y. (1983). Clinical significance of a variant form of urinary arylsulphatase A. Japan J. Exp. Med. §§; 73-76. Kriek, E. (l979). Aromatic amines and related compounds as carcino- genic hazards to man. 13; Environmental Carcinogenesis (P. Emmelot and E. Kriek, Eds.), Elsevier/North-Holland Biomedical Press, Amsterdam, pp. l43-l64. 139 Kriek, E. and Westra, J.G. (l979). Metabolic activation of aromatic amines and amides and interactions with nucleic acids. In: Chemical Carcinogens and DNA VII (P.L. Groves, Ed.), CRC—Press Inc., Boca Raton, Fla, pp. 2-28. Liehr, J.G., Beckner, C.F., Ballatore, A.M. and Caprioli, R.M. (l982). Fast atom bombardment mass spectrometry of estrogen glucuronides and sulfates. Steroids 22; 599-605. Linch, A.L., O'Connor G.B., Barnes, J.R., Killian, A.S. and Neeld, W.E. (l97l). Methylene-bis-ortho-chloroaniline (MOCA): Evaluation of hazards and exposure control. Am. Indus. Hyg. Assoc. J. 22; 802- 8l9. Lotlikar, P.D., Enomoto, M., Miller, J.A. and Miller, E.C. (1967). Species variations in the N- and ring-hydroxylation of Z-acetyl- aminofluorene and effects of 3-methylcholanthrene pretreatment. Proc. Soc. Exp. Biol. Med. 122; 341-346. Lower, G.M. and Bryan, G.T. (1976). Enzymic deacetylation of carcino- genic arylacetamides by tissue microsomes of the dog and other species. J. Toxicol. Environ. Hlth. 1; 421-432. Lynn, R.K., Donielson, D.W., Ilias, A.M., Kennish, J.M., Wong, K. and Matthews, H.B. (1980). Metabolism of bisazobiphenyl dyes derived from benzidine, 3,3'-dimethylbenzidine, or 3,3'-dimethoxybenzidine to carcinogenic aromatic amines in the dog and rat. Toxicol. Appl. Pharmacol. 22; 248-258. Lynn, R.K., Garvie-Gould, C., Milam, D.F., Scott, K.F., Eastman, C.L. and Rodgers, R.M. (l983). Metabolism of the human carcinogen, benzidine, in the isolated perfused rat liver. Drug Metab. Disp. ll: l09-ll4. MacDonald, W.E., Anderson, W.A.D., Coplan, M., Woods, F. and Deichmann, W.B. (l973). Effect of phenobarbital on production of bladder cancer (dog) by 4-aminobiphenyl. Toxicol. Appl. Pharmacol. 22; p. 498. Maltry, E. Jr. (Ed.) (1971). Foreword. IQ; Benign and Malignant Tumors of the Urinary Bladder, Medical Examination Publishing Co., Flushing N.Y., p. ll. Manis, M.0. and Braselton, W.E. (l984). Structure elucidation and 12. vitro reactivity of the major metabolite of 4,4'-methylenebis(2- chloroaniline) (MBOCA) in canine urine. Fundamen. Appl. Toxicol., in press. Manis, M.0., Williams, D.E., McCormack, K.M., Schock, R.J., Lepper, L.R., Ng, Y-C. and Braselton, W.E. (l984). Percutaneous absorp- tion, disposition, and excretion of 4,4'-methylenebis(2-chloroani- line) in dogs. Environ. Res. 22; 234-245. 140 Marnett, L.J. (1981). Minireview. Polycyclic aromatic hydrocarbon oxidation during prostaglandin biosynthesis. Life Sci. 22; 531- 546. Maru, A., Ishibashi, T., Imai, Y., Makita, A. and Tsuji, I. (1980). Arylsulphatase A activity of urine in patients with various geni- tourinary tract disorders. Clin. Chim. Acta 122; 155-161. Mason, R.P. (1982). Free radical intermediates in the metabolism of toxic chemicals. 22_Free Radicals in Biology. Vol. 5, Academic Press, p. 161-221. Mason, R.P., Chignell, C.E. (1982). Free Radicals in Pharmacology and Toxicology - Selected Topics. Pharmacol. Rev. 22; 189-21l. Masson, M.A., Ioannides, C., Gorrod, J.W. and Gibson, G.G. (1983). The role of highly purified cytochrome P-450 isozymes in the activation of 4-aminobiphenyl to mutagenic products in the Ames test. Car- cinogenesis 2; 1583-1586. Mattammal, M.B., Zenser, T.V. and Davis, B.B. (1981). Prosta landin hydroperoxidase-mediated 2-amino-4-(5-nitro-2-fury1)-[‘4Cflthiazole metabolism and nucleic acid binding. Canc. Res. 21; 4961-4966. McCann, J., Choi, E., Yamasaki, E. and Ames, B.N. (l975a). Detection of carcinogens as mutagens in the Salmonella/microsome test. Assay of 300 chemicals. Proc. Natl. Acad. Sci. 22(12): 5135-5139. McCann, J., Spingarn, N.E., Kobori, J. and Ames, B.N. (l975b). Detec- tion of carcinogens as mutagens: Bacterial tester strains with R factors plasmids. Proc. Natl. Acad. Sci. 22(3): 979-983. McQueen, C.A., Maslansky, G.J., Crescenzi, 5.8. and Williams, G.M. (1981). The genotoxicity of 4,4'-methylenebis(2-chloroani1ine) in rat, mouse, and hamster hepatocytes. Toxicol. Appl. Pharmacol. 22; 231-235. Meal, P.F., Cocker, J., Wilson, H.K. and Gilmour, J.M. (1981). Search for benzidine and its metabolites in urine of workers weighing benzidine-derived dyes. Brit. J. Industr. Med. 22; 191-193. Meister, A. (1983). Selective modification of glutathione metabolism. Science 222; 472-477. Mettlin, C., Graham, S. (1979). Dietary risk factors in human bladder cancer. Am. J. Epidemiol. 112; 255-263. Miller, H.E. and Cosgriff, J.M. (l983). Acetylator phenotype in human bladder cancer. J. Urol. 122; 65-66. Mommsen, 5., Aagaards, J. and Sell, A. (1983). A case-control study of female bladder cancer. Europ. J. Canc. Clin. Oncol. 12; 725-729. 141 Moore, B.P., Hicks, R.M., Knowles, M.A. and Redgrave, S. (1982). Metabolism and binding of benzo(a)pyrene and 2-acetylaminofluorene by short-term organ cultures of human and rat bladder. Canc. Res. 22; 642-648. Morrison, A.S., Buring, J.E., Verhoek, W.G., Aoki, K., Leck, I., Ohno, Y. and Obata, K. (1982). Coffee drinking and cancer of the lower urinary tract. J. Nat. Canc. Inst. 22; 91-94. Morton, K.C., King, C.M., Vaught, J.B., Wang, C.Y., Lee, M-S. and Mar- nett, L.J. (1983). Prostaglandin H synthase-mediated reaction of carcinogenic arylamines with tRNA and homopolyribonucleotides. Biochem. Biophys. Res. Comm. 12; 96-103. Mulder, G.J. (Ed.) (1981). Sulfation jg_vivo and in isolated cell preparations. 22; Sulfation of drugs and related compounds, CRC Press Inc., Boca Raton, Fla., p. 131-186. Mulder, G.J. and Scholtens, E. (l977). Phenol sulphotransferase and uridine diphosphate glucuronyltransferase from rat liver ifl_V1VO and ig_vitro. Biochem. J. 122; 553-559. Nagata, C., Kodama, M., Kimura, T. and Aida, M. (1980). Metabolically generated free radicals from many types of chemical carcinogens and binding of the radicals with nucleic acid bases. 22; Carcinogene- sis: Fundamental Mechanisms and Environmental Effects (8. Pullman, P.0. Ts'o, and H. Gelboin, Eds.), D. Reidel Publishing Co., Boston, Mass., p. 43-54. Nakayama, T., Kimura, T., Kodama, M. and Nagata, C. (1982). Electron spin resonance study on the metabolism of 2-naphthy1amine and l- naphthylamine in rat liver microsomes. Gann 12; 382-390. Oglesby, L.A., Flammang, T.J., Tullis, D.L. and Kadlubar, F.F. (1981). Rapid absorption, distribution, and excretion of carcinogenic N- hydroxy-arylamines after direct instillation into the rat urinary bladder. Carcinogenesis 2; 15-20. Oglesby, L.A., Hix, C., Snow, L., MacNair, P., Seig, M. and Langenbach, R. (1983). Bovine bladder urothelial cell activation of carcino- gens to metabolites mutagenic to Chinese hamster V79 cells and Salmonella typhimurium. Canc. Res. 22; 5194-5199. Parkes, H.G. (1978). The epidemiology of the aromatic amine cancers. lg; Chemical Carcinogens (C.E. Searle, Ed.), ACS Monographs 173, p. 462-480. Paulson, 0.0. and Portnoy, C.E. (1970). Sulfate ester conjugate: A one-step method for replacing the sulate with an acetyl group. J. Agric. Food Chem. 22; 180-181. 142 Pohl, R.J. and Fouts, J.R. (l983). Cytochrome P-450-dependent xeno- biotic metabolizing activity in Zymbal's gland, a specialized sebaceous gland of rodents. Canc. Res. 22; 3660-3662. Poirier, L.A., Miller, J.A. and Miller, E.C. (1963). The N- and ring- hydroxylation of 2-acetylaminofluorene and the failure to detect the N-acetylation of 2-aminofluorene in the dog. Canc. Res. 22; 790-800. Poirer, M.0., True, B'A, Laishes, B.A. (1982). Formation and removal of (guan-B-yl)-DNA-2-acety1aminofluorene adducts in liver and kidney of male rats given dietary 2-acety1aminofluorene. Canc. Res. 22; 1317-1321. Poupko, J.M., Hearn, W.L. and Radomski, J.L. (l979). N-Glucuronidation of N-hydroxy aromatic amines: A mechanism for their transport and bladder-specific carcinogenicity. Toxicol. Appl. Pharmacol. 22; 479-484. Poupko, J.M., Radomski, T., Santella, R.M. and Radomski, J.L. (l983). Organ, species, and compound specificity in the metabolic activa- tion of primary aromatic amines. J. Natl. Canc. Inst. 22; l077- 080. Powell, G.M. and Roy, A.B. (l980). Sulphate conjugation. In: Extra- hepatic Metabolism of Drugs and Other Foreign Compounds—(T.E. Gram, Ed.), Spectrum Publishers, NY, pp. 389-426. Radomski, J.L. (1979). The primary aromatic amines: Their biological properties and structure activity relationships. Ann. Rev. Phar- macol. Toxicol. 22; 129-157. Radomski, J.L. and Brill, E. (1970). Bladder cancer induction by aromatic amines: Role of N-hydroxy metabolites. Science 121; 992- 993. Radomski, J.L. and Brill, E. (1971). The role of N-oxidation products of aromatic amines in the induction of bladder cancer in the dog. Arch. Toxikol. 22; 159-175. Radomski, J.L., Hearn, W.L., Radomski, T., Moreno, M. and Scott, W.E. (1977). Isolation of the glucuronic acid conjugate of N-hydroxy- 4-aminobiphenyl from dog urine and its mutagenic activity. Canc. Res. 22; 1757-1762. Rammler, D.H., Grado, C. and Fowler, L.R. (1964). Sulfur metabolism of Aerobacter aerogenes. I. A repressible sulfatase. Biochem. 2; 230-237 0 Rapp. N.S., Zenser, T.V., Brown, W.W. and Davis, B.B. (1980). Metabo- lism of benzidine by a prostaglandin-mediated process in renal inner medullary slices. J. Pharmacol. Exp. Ther. 222; 401-406. 143 Reddy, T.V., Weisburger, E.K. and Thorgiersson, 5.5. (1980). Mutagenic activation of N-2-fluorenylacetamide and N-hydroxy-N-Z-fluorenyl- acetamine in subcellular fractions from X/Gf mice. J. Natl. Canc. Inst. 22; l563-1569. Robertson, I.G.C. and Birnbaum, L.S. (1982). Age-related changes in mutagen activation by rat tissues. Chem.-Biol. Interact. 22; 243- 252. Robertson, I.G.C., Sivarajah, K., Eling, T.E. and Zeiger, E. (l983). Activation of some aromatic amines to mutagenic products by prosta- glandin endoperoxide synthase. Canc. Res. 22; 476-480. Rush, G.F., Wilson, D.M. and Hook, J.B. (1983). Selective induction and inhibition of renal mixed function oxidases in the rat and rabbit. Fundam. Appl. Toxicol. 2; 16l-l68. Russfield, A.B., Homburger, F., Boger, E., van Dongen, C.G., Weisburger, E.K. and Weisburger, J.H. (1975). The carcinogenic effect of 4,4'-methylene-bis-(Z-chloroaniline) in mice and rats. Toxicol. Appl. Pharmacol. 21; 47-54. Schwartz, L.R. (1980). Modulation of sulfation and glucuronidation of l-naphthol in isolated rat liver cells. Arch. Toxicol. 22; 137- 145. Sciarini, L.J. and Meigs, J.W. (1958). The biotransformation of benzi- dine (4,4'-diaminobipheny1), an industrial carcinogen, in the dog. I. AMA Arch. Industr. Hlth. 12; 521-530. Sciarini, L.J. and Meigs, J.W. (1961). The biotransformation of benzi- dine. II. Studies in mouse and man. AMA Arch. Environ. Hlth. 2; 423-428. Silverman, D.T., Hoover, R.N., Albert, S. and Graff, K.M. (1983). Occupation and cancer of the urinary tract in Detroit. J. Natl. Simon, 0., Yen, S. and Cole, P. (1975). Coffee drinking and cancer of the lower urinary tract. J. Natl. Canc. Inst. 22; 587-591. Steinhoff, D. and Grundmann, E. (l97l). Zur cancerogenen Wirkung von 3,3'-dichlor-4,4'-diaminodiphenylmethan bei Ratten. Naturwissen- schaffen 22; 578. Stula, E.F., Barnes, J.R., Sherman, H., Reinhardt, C.F. and ZaPP. J.A. Jr. (1977). Urinary bladder tumors in dogs from 4,4'-methy1ene- bis(2-chloroaniline) (MOCA). J. Environ. Pathol. Toxicol. 1; 31- 50. 144 Stula, E.F., Sherman, H., Zapp, J.A. Jr. and Clayton, J.W. Jr. (1975). Experimental neoplasia in rats from oral administration of 3,3- dichlorobenzidine, 4,4'-methylene-bis(2-chloroaniline), and 4,4'- methylene-bis-(Z-methylaniline). Toxicol. Appl. Pharmacol. 21; 159-176. Sutter, M.A., Chernesky, P., Jayaraj, A. and Richardson, A. (1982). Metabolic activation of chemical carcinogens by kidney from rats and mice of various ages. Comp. Biochem. Physiol. 122; 435-448. Tobes, M.C., Brown, L.E., Chin, B. and Marsh, 0.0. (1983). Kinetics of tissue distribution and elimination of 4,4'-methylenebis(2-chloro- aniline) in rats. Toxicol. Letters 11; 69-75. Tola, S. (1980). Occupational cancer of the urinary bladder. J. Toxicol. Environ. Hlth. 2; 333-340. Toxic Substances Control Act (TSCA) (1983). TSCA Chemicals-in-Progress Bulletin 2; 8. Uehleke, H. and Brill, E. (1968). Increased metabolic N-oxidation of 2-naphthy1amine in dogs after phenobarbital pretreatment. Biochem. Pharmacol. 12; 1459-1461. U.S. Dept. of Health and Human Services (1983). NTP Technical Bulletin 2; p. 7. Vidyarthi, S.C. (1971). Pathology of bladder tumors. 12; Benign and Malignant Tumors of the Urinary Bladder (E. Maltry, Jr., Ed.), Medical Examination Publishing Co., Inc., Flushing, N.Y., p. 13-56. Vogel, H.J. and Bonner, D.M. (1956). Acetylornithinase of Escherichia coli: Partial purification and some properties. J. Biol. Chem. 212; 97-106. Weber, W.W., Radtke, H.E. and Tannen, R.H. (l980). Extrahepatic N- acetyltransferase and N-deacetylases. In: Extrahepatic Metabolism of Drugs and Other Foreign Compounds (T.E. Gram, Ed.), Spectrum Publishers, Jamaica, N.Y. Weisburger, J.H., Yamamoto, R.S., Williams, G.M., Grantham, P.H., Mat- sushima, T. and Weisburger, E.K. (1972). On the sulfate ester of N-hydroxy-N-Z-fluorenylacetamide as a key ultimate hepatocarcinogen in the rat. Canc. Res. 22; 491-500. White, G.L., Beranek, D.T. and Heflich, R.H. (1983). Effect of bac- terial concentration on reversions induced in Salmonella typhi- murium TA1538 by N-hydroxy-Z-acetylaminofluorene. Environ. Muta- genes1s 2; 565-575. Wiley, F.H. (1983). The metabolism of B-naphthylamine. J. Biol. Chem. 122: 627-630. 145 Williams, D.E. (1979). Report to the Michigan Toxic Substance Control Commission, October 10. Wynder, E.L., 0nderdonk, J. and Mantel, N. (1963). An epidemiologic investigation of cancer of the bladder. Cancer 12; 1388-1407. Zenser, T.V., Mattammal, M.B., Armbrecht, H.J. and Davis, 8.8. (1980). Benzidine binding to nucleic acids mediated by the peroxidative activity of prostaglandin endoperoxide synthetase. Canc. Res. 22; 2839-2845. Zenser, T.V., Mattammal, M.B., Brown, W.W. and Davis, B.B. (l979a). Cooxygenation by prostaglandin cyclooxygenase from rabbit inner medulla. Kid. Internat. 12; 688-694. Zenser, T.V., Mattammal, M.B. and Davis, 8.8. (1978). Differential distribution of the mixed-function oxidase activities in rabbit kidney. J. Pharmacol. Exp. Ther. 221; 719-725. Zenser, T.V., Mattammal, M.B. and Davis, B.B. (l979b). Cooxidation of benzidine by renal medullary prostaglandin cyclooxygenase. J. Pharmacol. Exp. Ther. 211; 460-464. "111111111111111“