Introduction

Top Abstract Introduction Materials and Methods Results and Discussion References

4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK)¹ is a tobacco-specific pulmonary carcinogen that is hypothesized to play a significant role as a cause of lung cancer in smokers (Hoffmann and Hecht, 1985; Hecht and Hoffmann, 1988; Hecht, 1998). In support of this hypothesis, numerous analytical studies have shown that NNK is present in substantial quantities in unburned cigarette tobacco as well as cigarette smoke (Hoffmann et al., 1994; Hecht and Tricker, 1999). Furthermore, the uptake of NNK by humans has been conclusively demonstrated by quantification of its metabolites NNAL [4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol], NNAL-Gluc {[4-(methylnitrosamino)-1-(3-pyridyl)but-1-yl]--O-D-glucosiduronic acid}, and NNAL-N-oxide [4-(methylnitrosamino)-1-(3-pyridyl-N-oxide)-1-butanol] (Fig. 1) in urine (Carmella et al., 1993, 1995, 1997; Hecht, 1998; Parsons et al., 1998; Lackmann et al., 1999). The involvement of NNK as a lung carcinogen in smokers also is supported by extensive studies that demonstrate that it is a potent lung carcinogen in rodents (Hoffmann and Hecht, 1985; Hecht and Hoffmann, 1988; Hecht, 1998). In rats, NNK causes lung tumors independent of the route of administration, a property not shown by any other tobacco carcinogen. In mice, NNK induces lung tumors in strains that are either susceptible or resistant to pulmonary tumor induction by other carcinogens. NNK is also an effective lung carcinogen in hamsters.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Major pathways of NNK metabolism.

NNK requires metabolic activation to exert its carcinogenic effects (Hecht, 1998). Metabolic activation of NNK occurs via hydroxylation at the methyl and methylene carbons adjacent to the N-nitroso group (Fig. 1). -Hydroxylation at the methyl group produces intermediate 1 and ultimately diazohydroxide 5 that pyridyloxobutylates DNA. -Hydroxylation at the methylene group leads to intermediate 2 and methanediazohydroxide (6) that methylates DNA. The major NNK metabolite NNAL, which is also a strong lung carcinogen in rats and mice, undergoes similar metabolic transformations leading to diazohydroxides 6 and 7 as illustrated in Fig. 1 (Hecht, 1998). These metabolic pathways have been extensively characterized in rodent lung where they are catalyzed effectively by cytochrome P-450 enzymes (Hecht, 1998). The resulting DNA adducts of NNK are formed in substantial quantities in rodent lung and their persistence is associated with lung tumor induction (Hecht, 1998). They also have been detected in human lung, but at lower levels (Hecht, 1998).

In contrast to the strong above-mentioned supportive evidence, in vitro studies with human lung tissue preparations indicate that the -hydroxylation metabolic activation pathways of NNK are in general poorly catalyzed (Castonguay et al., 1983a; Smith et al., 1992, 1995). The difference in metabolic activation between rodent and human tissues could be due to species differences in metabolism. However, it may be due to other factors. Human lung enzymes may be inhibited by years of exposure to tobacco smoke, by the presence of disease or anesthesia, or may be more sensitive to experimental manipulations than rodent enzymes. Because monkeys are closely related to humans, investigation of NNK metabolism by nonhuman primate lung might provide a better indication of potential species differences between rodents and primates, including humans. Nonhuman primate lung tissue is not readily available because these animals are not common laboratory models and are infrequently euthanized. In this study, we had access to fresh lung tissue from three monkeys, two patas (Erythrocebus patas) and one cynomolgus (Macaca fascicularis). This is the first study to investigate NNK metabolism in cultured nonhuman primate lung tissue. A previous study reported on NNK metabolism by lung microsomes from a patas monkey (Smith et al., 1997). There also have been three in vivo studies of NNK metabolism in monkeys (Castonguay et al., 1985; Hecht et al., 1993; Meger et al., 1999).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results and Discussion References

Animals. Patas and cynomolgus monkeys were maintained in a facility approved by the American Association for the Accreditation of Laboratory Animal Care (BioQual, Inc., Shady Grove, MD). They were fed a dry commercial primate diet containing 24% crude protein, supplemented with 5 ml of a multiple vitamin mixture weekly. Small quantities of fresh fruit were fed twice weekly. Patas 1, no. 665, was a female weighing 6.6 kg, age 16 years. She had been used for breeding and never received any carcinogen treatment. Patas 2, no. 865, was a female weighing 7.4 kg, age 14.5 years. She had received repeated treatments with aflatoxin B₁ during a pregnancy 10 years previously, and with 2-amino-3-methylimidazo[4,5-f]quinoline during a pregnancy 4 years previously. The cynomolgus monkey, no. 681, was a female weighing 4.3 kg, age 13 years. She was born in the wild and received no carcinogen treatment. She suffered from endometriosis. She had received Bupranax until 2 weeks before euthanasia. For comparison, cultures were established from a male F344 rat, age 10 weeks.

Chemicals. [5-³H]NNK, which has tritium at the 5-position of the pyridine ring, was acquired from Chemsyn Science Laboratories (Lenexa, KS), and diluted with unlabeled NNK as required. NNK metabolite standards were synthesized (McKennis et al., 1964; Hecht et al., 1980; Castonguay et al., 1984).

Organ Culture. The method was based on a previous study of NNK metabolism in cultured rat lung (Doerr-O'Rourke et al., 1991). Monkeys were anesthetized with ketamine and then euthanized by exsanguination. The lungs were removed under sterile conditions to dishes kept on wet ice. Lungs from patas 1 and the cynomolgus monkey were removed as large pieces and kept in plastic dishes on wet ice until 0.2-mm² samples were taken at random from the parenchyma for culture. For patas 2, 1-g samples were taken from four different areas of the peripheral lung at necropsy and kept in cold CMRL-1066 medium until sampled for culture. The explants used in all the culture studies reported herein would contain Clara cells, type 1 and 2 alveolar cells, endothelial cells, and macrophages. Explants (30/dish) were allowed to attach to the etched surface of 60-mm petri dishes, 15 min before adding 2 ml of CMRL-1066 medium. Twenty-four hours later, [5-³H]NNK was added such that the final concentration was 10 µM. Media were harvested at time points as indicated in Fig. 2. Each time point had three or four dishes and the values reported are the means. Media controls contained [5-³H]NNK, but no explants. Media were stored at 20°C until analysis.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Formation of NNK metabolites at various times in lung organ cultures from patas monkey 1 (A), patas monkey 2 (B), cynomolgus monkey (C), and F344 rat (D). Values for the monkey lung cultures are means of three or four culture dishes at each time point. Representative reproducibility data are illustrated in Table 1. Values for the rat lung culture are means of two dishes, except 24-h, single values. , diol; , NNAL; , NNK-N-oxide; , NNAL-N-oxide; , keto acid; , hydroxy acid; , keto alcohol. Rates of formation of total -hydroxylated products were compared over the 3- to 12-h time periods. Line slopes and 95% CI were 47.9 (17.9-78.0) for the rat, 266.9 (144-390) for the cynomolgus monkey, 123 (35.0-212) for patas 1, and 71.5 (51.6-91.4) for patas 2.

HPLC Analysis. Metabolites were analyzed by reversed phase HPLC on a 4.6 × 250-mm Rainin Microsorb-MV, 5-µm C-18 column at a flow rate of 1 ml/min. Detection of radioactivity was carried out with a Radiomatic Flo-One beta detector (Meriden, CT). Standard metabolites were detected with a Gilson model 116 UV detector operated at 254 nm. Solvent A was 20 mM sodium phosphate buffer, pH 7.0, and solvent B was methanol. The solvent program was as follows [time (min), %A, %B]: 0, 100, 0; 16, 92, 8; 31, 92, 8; 79, 68, 32; 84, 0, 100; and 99, 100, 0. Metabolites were initially identified by coelution of radioactivity with standards. Confirmation of metabolite identities was accomplished with a second system in which the pH of solvent A was adjusted to 4.5, resulting in a substantial change in retention times of most metabolites (Carmella and Hecht, 1985). Normal phase HPLC also was used for additional confirmation of the identity of 4-hydroxy-1-(3-pyridyl)-1-butanone (keto alcohol). A 4.6 × 250-mm Alltech Econosil 5-µm silica column was eluted with Solvents A (hexane) and B (2:1 isopropanol/ethanol) as follows [time (min), %A, %B]: 0, 100, 0; 5, 100, 0; and 45, 80, 20.

-Glucuronidase Treatment. Fifty microliters of medium from the 24-h time points of patas 1 and the cynomolgus monkey cultures were treated separately with -glucuronidase (Type IXA; Sigma). Each sample was incubated at 37°C for 20 h with 260 U of the enzyme in a final volume of 0.5 ml. An aliquot of each was analyzed by HPLC. To confirm enzyme activity, radiolabeled glucuronides of NNAL collected from the urine of a patas monkey treated with [5-³H]NNK (Hecht et al., 1993) were treated under the same conditions and completely converted to NNAL.

Statistical Analysis. Metabolite levels were compared with Student's t test and linear regression analysis (Instat, Inc., Newton, MA).

	Results and Discussion

Top Abstract Introduction Materials and Methods Results and Discussion References

Metabolites were identified by coelution with standards in at least two HPLC systems. The time course of metabolite formation in monkey and rat lung cultures is illustrated in Fig. 2. Metabolites were not observed in control cultures. In the cynomolgus monkey, keto alcohol formed rapidly and was the most abundant metabolite at 3 and 6 h; at subsequent time points, diol [4-(3-pyridyl)butan-1,4-diol] was found in the highest concentrations, probably due in part to metabolic reduction of keto alcohol. In the patas monkeys, keto alcohol and diol were also formed in substantial quantities. Diol was the most prevalent metabolite at 12 and 24 h in patas 1, and was second in abundance to NNAL in patas 2. Keto alcohol results from methyl hydroxylation of NNK as illustrated in Fig. 1. -HydroxymethylNNK (1) is mutagenic in bacterial and mammalian cells and produces pyridyloxobutyl DNA adducts (Hecht, 1998). Diol results both from methyl hydroxylation of NNAL and carbonyl reduction of keto alcohol. Little is known at present about the DNA-binding properties and mutagenicity of diazohydroxide 7. Levels of the other -hydroxylation products, keto acid [4-oxo-4-(3-pyridyl)butanoic acid] and hydroxy acid [4-hydroxy-4-(3-pyridyl)butanoic acid], were generally lower than those of keto alcohol and diol.

Total -hydroxylation can be assessed by summing levels of keto alcohol, diol, keto acid, and hydroxy acid. These data are summarized in Table 1 and Fig. 3 for the 24-h time point. The results clearly show that -hydroxylation exceeded pyridine-N-oxidation in all monkeys and was the most prevalent metabolic pathway in patas 1 and the cynomolgus monkey. In patas 2, -hydroxylation products were exceeded only by NNAL. -Hydroxylation was the least abundant pathway in the rat lung cultured under identical conditions. The rat data are similar to those observed in a previous study (Doerr-O'Rourke et al., 1991). The rate of formation of -hydroxylation products was significantly greater (P < .05) for the cynomolgus lung compared with the rat lung (Fig. 2).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Comparative levels of -hydroxylation, pyridine-N-oxidation, and NNAL formation in lung explants cultured with NNK (patas monkeys 1 and 2 , cynomolgus monkey , and rat ).

Pyridine-N-oxidation, yielding NNK-N-oxide [4-(methylnitrosamino)-1-(3-pyridyl-N-oxide)-1-butanone] and NNAL-N-oxide, was also a prevalent metabolic pathway in monkey and rat lung (Figs. 2 and 3; Table 1). In all cases, NNK-N-oxide exceeded NNAL-N-oxide. Previous studies in rats and mice have consistently shown that pyridine-N-oxidation is a substantial metabolic pathway in lung, generally exceeding that in liver (Hecht, 1998; Schrader et al., 1998). Because NNK-N-oxide and NNAL-N-oxide are both less tumorigenic than NNK (Castonguay et al., 1983b; Upadhyaya et al., 1999), this pathway is considered to be a detoxification mechanism.

Substantial amounts of carbonyl reduction to NNAL were observed in all cultures (Figs. 2 and 3; Table 1). This was the most prevalent pathway in patas 2, with lower levels being observed in patas 1 and the cynomolgus monkey. Treatment of culture media with -glucuronidase followed by HPLC analysis provided no evidence for the presence of NNAL-Gluc or any other glucuronides. Previous studies in rat lung also have failed to detect NNAL-Gluc (Staretz et al., 1997; Schrader et al., 1998).

The results for patas 1 can be compared with those obtained by Smith et al. (1997) who used lung microsomes from the same animal. They also observed significant amounts of methyl hydroxylation of NNK, with keto alcohol being the major metabolite. The rate of methyl hydroxylation was 6.8-fold greater than that of methylene hydroxylation. These results are consistent with our data. In that study, the rate of formation of NNK-N-oxide was greater than that of NNAL, whereas we found the opposite. This is probably due to the different systems used, e.g., microsomes versus organ culture. NNAL formation from NNK is catalyzed by nonmicrosomal as well as microsomal enzymes, which could explain its higher rate of formation herein (Hecht, 1998). The study of Smith et al. (1997) also suggests that cytochrome P-450s are only partially involved in NNK metabolism by patas lung microsomes; lipoxygenase and cyclooxygenase enzymes may play a role in NNK activation.

NNAL was the predominant metabolite of NNK in cultured human lung, carried out under conditions similar to those used herein (Castonguay et al., 1983a). Levels of hydroxy acid were ~0.1% as great as those of NNAL. An unknown metabolite, constituting ~3 to 5% of the amount of NNAL and eluting halfway between hydroxy acid and NNAL, also was observed. In human lung microsomes, NNAL was also the predominant metabolite of NNK (Smith et al., 1992, 1995). Small amounts of -hydroxylation and pyridine-N-oxidation products have been reported as lung microsomal products of NNK metabolism (Smith et al., 1992, 1995).

In summary, the results of this study clearly demonstrate that cultured monkey lung catalyzes NNK -hydroxylation, pyridine-N-oxidation, and carbonyl reduction in amounts that are comparable to those observed in rat lung. The production of substantial levels of -hydroxylation metabolites in monkey lung is of particular interest. These results are different from those obtained to date with human lung explants and microsomes, where -hydroxylation is a relatively minor pathway. The results indicate that there are probably not major rat-primate differences in pulmonary metabolism of NNK, and suggest that factors other than intrinsic metabolic capacity may be responsible for the discordant results obtained in comparative studies of NNK metabolism by human and rodent lung.