Carcinogenesis vol.13 no. 12 pp.2227-2232, 1992
Metabolism of 1- and 2-naphthylamine in isolated rat hepatocytes
Achim Orzechowski, Dieter Schrenk1 and Karl Walter Bock Institute of Toxicology, University of Tubingen, Wilhelmstrasse 56, W-7400 Tubingen, Germany 'To whom correspondence should be addressed
Introduction Aromatic amines such as 2-naphthylamine (2-NA*) were among the first chemicals recognized to induce cancer in humans and experimental animals (1,2). 2-NA, which has been used industrially and is present in ng amounts in cigarette smoke (3), induces primarily carcinomas of the urinary bladder in humans, dogs and rats (1). In contrast, no increased carcinogenic risk has been observed after treatment of animals with pure 1-naphthylamine (1-NA) (4), although epidemiological studies initially implicated 1-NA as a human carcinogen, probably resulting from concomitant exposure to 2-NA (1). In previous studies a number of metabolites of 2-NA have been isolated and characterized in vivo and in vitro. It was shown that acetyl, glucuronosyl and sulfamyl conjugates at the amino group and C-(ring-) and A'-oxidation products and their conjugates are formed (5,6). A'-oxidation has been recognized as the crucial step in the activation of 2-NA. The hydroxylamine can undergo •Abbreviations: 2-NA, 2-naphthylamine; 1-NA, 1-naphthylamine; MC, 3-methylcholanthrene; PAP, 3'-phosphoadenosine-5'-phosphate; El, electron impact; UGT, UDP-glucuronosyltransferase. © Oxford University Press
Materials and methods Chemicals [3H]2-NA (26 mCi/mmol) was a generous gift from Dr F.F.Kadlubar, National Center for Toxicological Research, Jefferson, AR. 1-NA, 2-NA, 1-nitronaphthalene, 2-nitronaphthalene and 2-hydroxy-l-arninonaphthaJene were obtained from Aldrich (Steinheim, Germany). MC, collagenase type IV, sulfatase type H-5, p-nitrophenylsulfate and 3'-phosphoadenosine-5'-phosphate (PAP) were purchased from Sigma (St Louis, MO), UDP-glucuronic acid from Boehringer (Mannheim, Germany), /S-glucuronidase from Serva (Heidelberg, Germany) and [l4C]UDP-glucuronic acid from New England Nuclear (Dreieich, Germany). The following compounds were synthesized using published procedures: JV-hydroxy-2-NA (13), W-acetyl-2-NA and JV-acetyl-1-NA (14), 2-NA-l-sulfate (15), l-hydroxy-2-NA (16) and 6-hydroxy-2-NA (5). The identity of the synthetic standards was confirmed by mass spectrometry. Biosynthesis of conjugated compounds The yV-glucuronides of 1- and 2-NA and W-hydroxy-2-NA were synthesized using liver microsomes from MC-treated rats in the following incubation mixture (17): Tris-HCl buffer, pH 7.4 (100 raM), magnesium chloride (5 mM), Brij 58 (0.5 mg/mg protein), UDP-glucuronic acid (3 mM) and 1 mg microsomal protein in a total volumtof 500 /d. The substratexoncentration of amines was 0.5 mM. After precipitating protein with methanol, N-glucuronides were isolated using the described HPLC system. The identity of the ghicuronides was confirmed by electron impact (EI)-mass spectroscopy of their silylated derivatives. The sulfate conjugates of 6-hydroxy-2-NA and 2-hydroxy-l-NA were synthesized using the following incubation mixture (18): Tris-HCl buffer, pH 7.4 (50 mM), magnesium chloride (5 mM), PAP (10 mM), /j-nitrophenylsulfate (5 mM), sodium sulfite (0.25 mM), dithiothreitol (100 mM), bovine serum albumin (0.02%) and 100 til liver homogenate (1 mg protein) in a final volume of 800 j j . The reactions were started by addition of 50 /tM of the substrate and terminated after 30 min by addition of two volumes methanol. The su I fairs were isolated by HPLC and their identity was confirmed after incubation of the isolated compounds with arylsulfatase. The liberated aminophenols were identified by their retention times and characteristic fluorescence spectra.
2227
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
The liver probably plays a major role in the metabolic activation of the bladder carcinogen 2-naphthylamine (2-NA) and hi the inactivation of the non-carcinogenic isomer 1-naphthylamine (1-NA). However, metabolic profiles of these compounds (including primary metabolites and directly determined conjugates) hi hepatocytes are not available. Therefore metabolism of 1- and 2-NA was compared hi freshly isolated hepatocytes from 3-methylcholanthrene (MO-treated and untreated rats. At 10 /iM, 2-NA was found to be mainly A'-acetylated (66% of total metabolites after 1 h incubation) and iV-ghicuronidated (19%). Minor pathways led to C-oxidation (7%) and A'-oxidation (3%; 2% present as the W-glucuronide). In hepatocytes from MC-treated rats total metabolism was slightly affected (1.5-fold increase). However, C- and A'-oxidation were markedly increased (63 and 18% respectively), while A'-acetylation and A'-glucuronidation were diminished (5 and 2% respectively). Similar experiments were carried out with 1-NA. Its A'-glucuronide was the predominant metabolite (68%) followed by the A'-acetylated compound (15%) while C-oxidation was low and W-oxidized metabolites could not be detected, even after induction. The results demonstrate that MC treatment markedly shifted 2-NA metabolism from A'-acetylation and A'-glucuronidation to N- and C-oxidation. In the case of 1-NA metabolism extensive A'-glucuronidation together with the lack of A'-oxidation may prevent carcinogenesis.
A'-glucuronidation and the resulting A'-hydroxy-A'-glucuronide is excreted into urine, where it can be hydrolyzed under the slightly acid conditions in the bladder to liberate the hydroxylamine (7). This can be further activated through protonation or acetylation to form the ultimate carcinogen, probably an electrophilic nitrenium cation. In addition, other activation pathways for 2-NA, such as the formation of reactive ortho-iminoquinones catalyzed by prostaglandin H synthase in the urinary bladder have been discussed (8,9). The metabolism of the non-carcinogenic derivative 1-NA has been studied less extensively. The presence of small quantities of A'-hydroxylated metabolites in urine of dogs and in in vitro incubations with bovine bladder epithelial microsomes has been reported (1,4,10). However, with liver microsomes from dogs, rats and humans and with purified rat P450 isozymes no A'-hydroxylation was detectable (11,12). Ring oxidation of 1-NA was consistently found to occur at positions 2 and 4 (4,11). The balance between different activating and inactivating metabolic pathways plays a decisive role in determining the potency and site of action of a given carcinogen. However, no information about the metabolism of 1- and 2-NA on the hepatocellular level is available. To elucidate the mechanisms of toxification and detoxification further, we quantitatively compared the metabolic profiles of 1- and 2-NA in isolated hepatocytes from untreated and 3-methylcholanthrene (MC)treated rats.
A.Orzecbowski, D.Schrenk and K.W.Bock Animals Male Wistar rats (180-250 g) were obtained from Savo (Kisslegg, Germany) and kept on standard diet (Altromin, Lage, Germany) and tap water ad libitum. MC was administered i.p. (40 mg/kg) in corn oil 3 days before death. Untreated animals were used as controls. Hepatocytes were obtained using the sequential EDTA-collagenase method as described by Seglen (19). The viability of the cells was determined using trypan blue. Only preparations exceeding 85% viability were used for metabolism studies. Microsomes were prepared as described previously (20).
10
20
Analysis of metabolites Aliquots (100-200 fd) were injected into a Waters HPLC system (Millipore, Bedford, MA), composed of two pumps (501/510) and a Lamda-Max 481 UV dectector (absorbance wavelength 280 nm). Additionally, either a fluorescence detector (emission 405 nm, excitation 340 nm; Sykam model S3400, Sykam, Gilding, Germany) or a radkjisotope detector (Beckman 171, Beckman Instruments, Fullerton, CA) was linked to trie UV detector. Separation of metabolites was achieved on a 4.6 x 250 mm steel column filled with Spherisorb ODS2 5 nm (Grom, Herrenberg, Germany). A linear gradient from 100% diethylammonium acetate (0.1 % in water, adjusted to pH 7.0 with acetic acid) to 65% acetonitrile over 40 min followed by a linear gradient to 100% acetonitrile over 5 min at a flow rate of 1 ml/min was used. Quantification of metabolites of 2-NA was obtained directly using the radioisotope detector by recording the percentage of total radioactivity eluting with each peak. Total elided radioactivity, including metabolites and unmetabolized substrates, was determined by liquid scintillation spectrometry on a Beckman LS 3801 liquid scintillation counter. To quantify total metabolism, except covalent protein binding, the amount of unmetabolized substrate was deducted from total eluted radioactivity. Because further oxidation of the N-hydroxy metabolite sometimes occurred during the analytic procedure, the rates of JV-hydroxylatk>n were calculated from total N-hydroxy- and nitroso compounds observed as described by Butler el al. (22). For quantification of 1-NA metabolites, standard calibration curves for the peak areas at 280 nm were recorded for A'-acetyl-l-NA and 2-hydroxy-I-NA. Calibration curves for peak areas of the N-glucuronides of 1-NA, 2-NA and N-hydroxy-2-NA were obtained by incubating the amines with [UC]UDPglucuronic acid (see above) and quantifying the glucuronides by liquid scintillation spectrometry after collecting the fractions from the HPLC. The rate of total metabolism of 1-NA was calculated from the decrease of unmetabolized substrate.
30
Retention Tune (min)
Fig. 1. Radiochromatogram of metabolites of [3H]2-NA produced in hepatocytes from (A) untreated controls and (B) MC-treated rats. Compounds identified are: 1, 2-NA-6-sulfate; 2, 2-NA-6-glucuronide; 3, 2-NA-/V-glucuronide + W-hydroxy-2-NA-Ar-glucuronide; 4, 2-NA-l-sulfate; 5, /V-hydroxy-2-NA; 6, /V-acetyl-2-NA; 7, 2-NA; 8, 2-nitrosonaphthalene. Cells were incubated with 50 nM 2-NA.
10
20
15 Rrtrwwn Time (mm)
Covalent protein binding of 2-NA metabolites Covalent protein binding in incubations of hepatocytes with [3H]2-NA was measured by the method of Uehleke et al. (23). The pellet obtained by precipitation with methanol was resuspended in 10% trichloroacetic acid and centrifuged at
Fig. 2. HPLC elution profile showing in detail UV detection of A'-hydroxy-2-NA-A'-glucuronide (3a) and 2-NA-/V-glucuronide (3b). Cells from untreated rats were incubated with 50 /iM 2-NA. Metabolites 1 and 4 are 2-NA-6-sulfate and 2-NA-l-sulfate.
Table I. Metabolite formation [nmol/h/106 cells (% of total metabolites)] from 2-NA in isolated rat hepatocytes. Effects of prctreatment with MC Metabolite
10 uM 2-NA* Controls
W-Acetyl-2-NA 2-NA-N-glucuronide 2-NA-l-sulfate A'-Hydroxy-2-NA-N-glucuronide 2-NA-6-sulfate /V-Hydroxy-2-NA 2-NA-1 -glucuronide Unidentified metabolites Total metabolism
4.4 1.2 0.3 0.15 0.12 0.06 95%.
Naphthylamine metabolism in rat bepatocytes
Instrumentation UV spectra were recorded on a Shimadzu 160A UV spectrophotometer and fluorescence spectra on a Perkin Elmer LS-5B luminescence spectrometer. El-mass spectra were obtained using a Finnigan 4021 mass spectrometer, FAB mass spectra on a Varian MAT 711A mass spectrometer.
Results Separation and quantification of metabolites of 2-NA HPLC analysis of radioactive metabolites of 2-NA formed in hepatocytes from untreated and MC-treated rats is shown in Figure 1. The metabolites identified in order of increasing retention times are 2-NA-6-sulfate, 2-NA-6-glucuronide, A
1
3H
u
2
3
4
-i.
.
*
B
1
2 10
4 30
20
Metabolic profile of 1-NA HPLC chromatograms of metabolites of 1-NA are shown in Figure 3. Identification of metabolites was based on retention times and fluorescence and UV spectra compared to synthetic standards. The number of metabolites produced by hepatocytes
(min) Fig. 3. HPLC elution profile showing UV (A) and fluorescence (B) detection of metabolites of 1-NA produced in hepatocytes from untreated rats. Compounds identified are: 1, l-NA-A'-glucuronide; 2, l-NA-2-sulfate; 3, yV-aceryl-1-NA; 4, 1-NA. The substrate concentration was 10 /iM.
Table II. Metabolite formation [nmol/h/106 cells (% of total metabolites)] from 1-NA in isolated rat hepatocytes. Effects of pretreatment with MC Metabolite
1-NA° Controls
l-NA-A'-glucuronide l-NA-2-sulfate W-Acetyl-l-NA Unidentified metabolites
4.6 0.5 1.2 0.6
± ± ± ±
0.& 0.2 0.3 0.2
Total metabolism
7.1 ± 2.4
1-NA Controls
MC-treated (68) (7) (15) (9)
6.6 0.3 0.15 0.3
± ± ± ±
1.1 0.1 0.06 0.1
7.5
± 2.7
(89) (4) (2) (4)
23.1 1.9 4.6 2.9
± ± ± ±
1.2 0.5 1.0 0.5
MC-treated (70) (7) (14) (9)
33.1 ± 3.2
'1-NA (dissolved in DMSO) was added to 50 ml of hepatocyte suspension to give a final concentration of 10 and ''Values represent means ± SD of at least three independent experiments.
24.8 3.2 1.8 2.8 35.2
± ± ± ± ±
2.1 0.9 0.61 0.7 3.2
(77) (9) (5) (8)
respectively.
2229
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
Af-hydroxy-2-NA-N-glucuronide, 2-NA-7v-glucuronide, 2-NA1-sulfate, N-hydroxy-2-NA and /V'-acetyl-2-NA. Identification of metabolites was based on their retention times and characteristic fluorescence or UV spectra compared to synthetic standards. Separation of free and conjugated compounds was sufficient for direct quantification using radiodetection, except for the N-glucuronides of 2-NA and Ar-hydroxy-2-NA, which could only be distinguished by UV detection (Figure 2). Quantification of these metabolites was achieved by using standard calibration curves. Metabolic profile of 2-NA Table I summarizes the quantitative analysis of the metabolic profile of 2-NA. The metabolites identified accounted for > 88% of total metabolites. In hepatocytes from untreated rats the two major routes of 2-NA metabolism at a substrate concentration of 10 /iM were N-acetylation (66% of total metabolites) and A'-glucuronidation (19%). MC treatment markedly increased C(63% of total metabolites) and N-oxidation (18%). In hepatocytes from untreated rats, mutagenic N-hydroxy-2-NA was present both in free and conjugated form at low levels (1 and 2% of total metabolites respectively). MC treatment led to a 2-fold increase in free N-hydroxy-2-NA, and a > 10-fold increase in N-hydioxy2-NA-Af-glucuronide. Covalent binding to protein was slightly increased from 0.020 ± 0.006 to 0.029 ± 0.007 nmol/mg protein (means ± SD of four experiments). The effect was not significant using Student's f-test. Raising the substrate concentration from 10 to 50 /tM resulted in a 1.9-fold increase in total metabolism. The proportion of N-acetyl-2-NA (31 % of total metabolites) was decreased and an increased proportion of 2-NA-N-glucuronide (35%), ringhydroxylated 2-NA-l-sulfate (19%) and 2-NA-6-sulfate (4%) as well as A/-hydroxy-2-NA-A/-glucuronide (4%) was found. Similar to the results at 10 /iM, MC treatment diminished the formation of N-acetyl-2-NA to 2% of total metabolites, and led to a relative decrease of 2-NA-N-glucuronide (14%). Total metabolism was enhanced by MC treatment by a factor of 2.6, mainly due to a strong increase in the formation of ring-hydroxylated metabolites, and the formation of N-hydroxylated products was stimulated — 5-fold. A concomitant, significant rise in covalent protein binding from 0.079 ± 0.010 nmol/mg in controls to 0.160 ± 0.015 nmol/mg after MC treatment was obtained (means ± SD of four experiments; P 2 0.001). When the incubation time was increased to 2 h, no marked alterations of the metabolic patterns of 2-NA were observed.
100 000 gfor 10 min. The procedure was repeated with 10 and 5% trichloroacetk acid at room temperature and 90°C, with ethanol at 90°C in closed vials, chloroform:methanol (1:1, v/v) and dry ethylether at room temperature. The final pellet was gently heated in a water bath to remove residual ether and was dissolved in 5 ml 1 N NaOH. After neutralization with 1 N HC1, aliquots were removed for liquid scintillation counting and procein determination. Enzyme assays 1- and 2-NA glucuronidation was assayed according to the fluorometric method described by Lilienblum and Bock (17). Protein was determined according to Lowry el at. (24).
A.Orzechowski, D.Schrenk and K.W.Bock
-1
-0.5
0.0
OS
1.0
\S
2.0
25
Table III. Apparent kinetic parameters of 1- and 2-NA glucuronidation Compound (mM) Untreated controls 1-NA 2-NA MC treatment 1-NA 2-NA
*MI
KM2 *MI KM2
V max (nmol/min/mg)
Od/min/mg)
0.45 2.5
90 21
200 8
0.07 0.40 °- 5 3.8
30 220 26 57
600 550 52 15
of untreated and MC-treated rats was much lower than with 2-NA as substrate. Three metabolites could be characterized as 1-NAN-glucuronide, l-NA-2-sulfate and JV-acetyl-1-NA. Compounds eluting at retention times less than 12.5 min were also found in 1-NA-free incubations. Since radiolabeled 1-NA was not available, covalent protein binding could not be assessed. As shown in Table II, at substrate concentrations of 10 and 50 nM, 1-NA-N-glucuronide was the predominant metabolite (68 and 70% of total metabolites respectively). Formation of JV-acetyl-1-NA was lower (15 and 14%) and ring-hydroxylation, finally leading to l-NA-2-sulfate accounted for only 7% of total metabolism. W-Hydroxylated metabolites were not detectable in any incubation with 1-NA. Pretreatment with MC did not lead to a marked alteration of the overall pattern of metabolites except for a decrease in N-acetyl-1-NA formation. Interestingly, in hepatocytes from untreated animals, the total rate of metabolism at 50 /tM was much higher with 1-NA than with 2-NA. 2230
Discussion In isolated rat hepatocytes seven metabolites of 2-NA, also detected in vivo (25), were identified and quantified by direct monitoring of primary and conjugated metabolites. Since primary metabolites of 2-NA such as N-hydroxy-2-NA and 1-hydroxy2-NA are readily oxidized in aqueous solution, recording of the conjugates may allow more exact quantification of metabolic pathways than hydrolyzing the conjugates and analyzing the liberated primary metabolites. With hepatocytes from untreated rats the major routes of metabolism were N-acetylation and N-glucuronidation. These findings may reflect the low constitutive expression of isoforms of the CYP1A family which are primarily responsible for the oxidation of aromatic amines (11,26). The high percentage of N-acetyl-2-NA found in rat hepatocytes is not observed in urine or bile of rats treated with 2-NA (25). Acetylation does not markedly alter the polarity of 2-NA. Hence, further metabolism of this compound may be required in vivo to facilitate its excretion. However, an extended incubation time did not lead to a marked decrease of N-acetyl-2-NA, suggesting that extrahepatic sites are involved in the further metabolism of N-acetyl-2-NA. /V-Acetyl-2-NA probably represents a detoxification product of 2-NA. It was shown in vivo that acetylation shifts the site of ring-oxidation to 6-mono- and 5,6-dihydroxylated metabolites regarded as detoxified metabolites (25) and that no hydroxamic acids were found when animals were treated with N-acetyl-2-NA (5). The /V-glucuronide of /V-hydroxy-2-NA has been suggested as a transport form in a sequence of events leading to bladder cancer. The glucuronide is formed preferentially in the liver by MCinducible UGT isozyme (27) and, after release into the systemic circulation, ultimately reaches the urinary bladder where it liberates the hydroxylamine forming an electrophile which binds to DNA (7). DNA adducts formed in the bladder epithelium are thought to result ultimately in the formation of bladder cancer (2). Our findings demonstrate that /V-hydroxy-2rNA-/V-glucuronide and unconjugated N-hydroxy-2-NA are released from rat hepatocytes. The latter compound may enter the bladder lumen via the systemic circulation and bind directly to DNA as has been proposed for 4-aminobiphenyl (2). Formation of N-hydroxy2-NA-glucuronide was increased in hepatocytes from MC-treated rats. This effect, together with a reduced formation of the detoxified metabolites N-acetyl-2-NA and 2-NA-N-glucuronide
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
Fig. 4. Hanes plots of UGT activity towards 1-NA in microsomes of untreated (A) and MC-treated (B) rats. The S/V value is expressed in mg x min/1.
Kinetic analysis of UDP-glucuronosyltransferase (UGT) activity towards 1- and 2-NA Rat liver microsomes from control and MC-treated rats were used for kinetic analysis of N-glucuronidation. As shown in Figure 4, biphasic enzyme kinetics of UGT activity towards 1-NA can be detected after MC treatment. Similar results were observed with 2-NA (not shown), suggesting the induction of (an) isozyme(s) with higher affinity towards 1- and 2-NA. Kinetic data for both substrates are compared in Table HI. Comparison of the apparent Km values shows that 1-NA is glucuronidated in microsomes from untreated rats with a 5-fold higher affinity than 2-NA. From experiments with liver microsomes of MC-treated rats, two apparent Km values were calculated for each substrate, both constants being ~ 10-fold higher (lower affinity) for 2-NA than for 1-NA. Comparison of the ratio V^JKm (corresponding to the 'intrinsic clearance') demonstrates that 1-NA is glucuronidated far more effectively than 2-NA.
Naphthylamine metabolism in rat bepatocytes
Acknowledgements The authors wish to thank Mrs S.Beck and Mrs S.Pahl for expert technical assistance. This project was supported by a research grant from the Deutsche Forschungsgemeinschaft, Bonn-Bad Godesberg.
References 1. Radomsld J.L. (1979) The primary aromatic amines, their btotogkal properties and structure—activity relationships. Annu. Rev. PharmacoL Toxicol., 19, 129-157. 2. Beland,F.A. and Kadlubar.F.F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. In Cooper.C.S. and Grover.P.L. (eds), Chemical Carcinogenesis and Mutagenesis I. SpringerVerlag, Berlin, pp. 267-325. 3. Patriankos.C. and Hoffmann.D. (1979) Chemical studies on tobacco smoke. LXIV. On the analysis of aromatic amines in cigarette smoke. J. AiuL Toxicol, 3, 150-154. 4. Radomsky.L., Deichmann,W.B., Altmann.N.H. and Radomski.T. (1980) Failure of pure 1-naphthylamine to induce bladder tumors in dogs. Cancer Res., 40, 3537-3539. 5. Boyland,E. and Manson.D. (1963) The biochemistry of aromatic amines: the metabolism of 2-naphthylamine and 2-naphthylhydroxylamine derivatives. Biochem. J., 101, 84-102. 6. Kadlubar.F.F., Unruh.L.E., Flammang.TJ., Sparks.D., Mitchum.R.K. and Mulder.G.J. (1981) Alteration of urinary levels of the carcinogen W-hydroxy2-naphthylamine and its Ar-glucuronide in the rat by control of urinary pH, inhibition of metabolic sulfation, and changes in biliary excretion. Chem.-Biol. Interactions, 33, 129-147. 7.Kadlubar,F.F., MillerJ.A. and Miller.C.E. (1977) Hepatic microsomal Af-glucuronidation and nucleic acid binding of JV-hydroxy arylamines in relation to urinary bladder carcinogenesis. Cancer Res., 37, 805-814. 8. Kadlubar.F.F., Frederick.C.B., Weiss,C.C. and Zenser.T.V. (1982) Prostaglandin endoperoxide synthetase-mediated metabolism of carcinogenic aromatic amines and their binding to DNA and protein. Biochem. Biophys. Res. Commun., 108, 253-258. 9. BoydJ.D. and Eling.E.T. (1987) Prostaglandin H synthase-catalyzed metabolism and DNA binding of 2-naphthylamine. Cancer Res., 47, 4007-4014. 10. PoupkoJ.M., Radomslti.T., Santella.R.M. and RadomskiJ.L. (1983) Organ, species and compound specificity in the metabolic activation of primary aromatic amines. J. NatL Cancer Inst., 70, 1077-1080. 11. Hammons.GJ., Guengerich.F.P., Weis.C.C, Beland.F.A. and Kadlubar.F.F. (1985) Metabolic oxidation of carcinogenic arylamines by rat, dog and human hepatic microsomes and by purified flavin-containing and cytochrome P-450 monooxygenases. Cancer Res., 45, 3578-3585. 12. Brill,E. and RadomskiJ.L. (1971) Comparison of in vitro and in vivo N-oxidation of carcinogenic aromatic amines. Xenobiotica, 1, 347—348. 13. Boyland,E., Manson.D. and Nery.R. (1962) The reaction of phenylhydroxylamine and 2-naphthylhydroxylamine with thiols. /. Chem. Soc., 1962, 606-611. 14. BoothJ., Boyland.E. and Manson.D. (1955) Metabolism of polycyclic compounds. 9. Metabolism of 2-naphthylamine in rat tissue slices. Biochem. J., 60, 6 2 - 7 1 . 15. BoylandJ*., Manson.D. and Sims,P. (1953) The preparation of o-aminophenyl sulfates. J. Chan. Soc., 1953, 3623-3630. 16. Desai.R.D., Hunter.R.F. and Khaladi.R.K. (1938) The unsaturation and tautomeric mobility of heterocyclic compounds. J. Chem. Soc., 1938, 321-329. 17. Lilienblum.W. and Bock.K.W. (1984) N-Glucuronide formation of carcinogenic aromatic amines in rat and human liver microsomes. Biochem. PharmacoL, 33, 2041-2046. 18. FJ Mouelhi.M. and Kaufrman.F.C. (1986) Sublobular distribution of transferases and hydroiases associated with ghicuronide, suliate and ghitathione conjugation in human liver. Hepatology, 6, 450-456. 19. Seglen.P.O. (1976) Preparation of isolated rat liver cells. Methods Cell BioL, 18, 2 9 - 8 3 . 20. Bock.K.W. and White.I.N.H. (1974) UDP-glucuronosyltransferascs in perfused rat liver and in microsomes: influence of phenobarbital and 3-nWhylcholanrhrene. Eur. J. Biochem., 46, 451-459. 21.Schrenk,D. andBockJCW. (1990) Metabolism of benzene in rat nepatocvtes. Influence of inducers on phenol glucuronklalian. Drug. Mewb. Dispos., 18, 720-725. 22. Buuer.M.A., Guengerich.F.P. and Kadlubar.F.F. (1989) Metabolic activation of the carcinogens 4-aminobiphenyl and 4,4'-methylene-bis(2-chloraniline)
2231
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
is likely to increase the burden of genotoxic intermediates reaching the bladder. Similarly, induction of hepatic cytochrome P4501A2 may increase the risk of bladder carcinogenesis in smokers in a synergistic manner. 2-NA is not known to be a hepatocarcinogen. Covalent binding to hepatic macromolecules does not seem to reflect a carcinogenic risk. However, an increased formation of reactive 2-NA metabolites after MC-type induction may lead to an increased binding to liver DNA and may thus initiate hepatocarcinogenesis. In fact, covalent protein binding with 50 /iM 2-NA was significantly increased in hepatocytes from MC-treated rats. Similar experiments using 32P-postlabeling of DNA bases are currently under way to determine effects on DNA binding. Direct N-glucuronidation of 1- and 2-NA probably represents a detoxifying reaction. However, cleavage of the conjugates, e.g. by intestinal glucuronidases, may lead to reabsorption of the parent compound from the gut. Glucuronidation of 1- and 2-NA has been shown to be preferentially catalyzed by MC-inducible phenol UGT (17,28). However, in hepatocytes from MC-treated rats, induction of this enzyme does not appear to compensate for the induction of P4501A isozymes, with the effect that 2-NA-glucuronide was a minor metabolite. The total rate of metabolism at a substrate concentration of 10 /tM was similar for 2-NA and its non-carcinogenic isomer 1-NA. At 50 /iM, however, an almost linear increase in total metabolism was observed with 1-NA but not with 2-NA. A lower than proportional increase may indicate saturation of a ratelimiting step in 2-NA metabolism at this substrate concentration. Comparison of the metabolic profiles with both isomers revealed very different patterns of metabolites. 1-NAN-glucuronide was the predominant metabolite of 1-NA in incubations at both substrate concentrations as well as with cells from MC-treated rats. In agreement with previous findings in liver microsomes no Ak>xidau'on products were detected (10,11). From kinetic studies in rat liver microsomes an apparent Km value of 0.5 mM was calculated from a MC-inducible UGT activity towards 2-NA. This finding suggests that 2-NA is a poor substrate even for the MC-inducible phenol UGT. In contrast, a significantly higher affinity for both constitutive and MC-inducible isoforms of UGT and a much higher 'intrinsic clearance' was calculated with 1-NA as substrate. Induction of a high affinity UGT with a low Km value of ~50/iM appears to compensate the induction of P4501A isozymes resulting in a balance between N-glucuronidation and ring-hydroxylation of 1-NA. Failure of 1-NA to induce tumors in experimental animals has been based solely on the lack of TV-oxidation (10). However, 1-NA was found to act as a mutagen in the Ames test (29) and it has been suggested that 2-hydroxy-l-NA, the major metabolite formed in microsomal incubations, has some potential to form an electrophilic iminoquinone, which might react with DNA (2). UGTs play a major role in the metabolism of aromatic and heteroaromatic amines and their roles in chemical carcinogenesis have been recently reviewed (30). Our results show that 1-NA is extensively A'-glucuronidated and as a consequence can be rapidly excreted. Efficient N-glucuronidation may thus prevent accumulation of 1-NA and its further oxidation to toxic metabolites. Further comparative studies with other carcinogenic arylamines are necessary to extend our understanding of the highly complex pathways of this class of compounds leading to distinct patterns of organ-directed carcinogenicity.
A.Orzechowsld, D.Schrenk and K.W.Bock
Received on July 27, 1992; revised on September 21, 1992; accepted on September 24, 1992
2232
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
by human hepatic microsomes and purified rat hepatic cytochrome P-450 monooxygenases. Cancer Res., 49, 25—31. 23. Uehleke.H., HeUmer.K.H. and Tabarelli-Poplawski.S. (1976) Metabolic activation of halothane and its covalent binding to liver endoplasmatic proteins in vitro. Naunyn Schmiedbergs Arch. Pharmacol., 279, 38-48. 24. Lowry.O.H., Rosebrough.A.L., Farr,A.L. and Randall.RJ. (1951) Protein measurement with the Folin reagent. /. Biol. Chem., 193, 265-275. 25. Gorrod.J.W. and Manson.D. (1986) The metabolism of aromatic amines. Xenobiotica, 16, 933-955. 26. Butler.M.A., Iwasaki.M., Guengerich.F.P. and Kadlubar.F.F. (1989) Human cytochrome P-450pA (P-450IA2), the phenacetin Odecthylase, is primarily responsible for the hepatic 3-demethylation of caffeine and N-oxidation of carcinogenic arylamines. Proc. NatL Acad. Sci. USA, 86, 7696-7700. 27. Bock.K.W., Josting.D., LUienbhim.W. and Pfeil.H. (1979) Purification of rat-liver microsomal UDP-glucuronosyl-transferase. Separation of two forms inducible by 3-methylchc4anthrene or phenobarbhal. Eur. J. Biochem., 98, 19-26. 28. Green.M.D. and Tephly.T.R. (1987) N-Glucuronidation of carcinogenic aromatic amines catalyzed by rat hepatic microsomal preparations and purified undine 5'Kliphosphateghjcuronosyttransferases. Cancer Res., fl, 2028-2031. 29. McCannJ., Choi.E., Yamasaki.E. and Ames.B.N. (1975) Detection of carcinogens as mutagens in the Salmonella Imicrosomc test: assay of 300 chemicals. Proc. Natl. Acad. Sci. USA, 72, 5135-5139. 30. Bock.K.W. (1991) Roles of UDP-glucuronosyltransferases in chemical carcinogenesis. Oil. Rev. Biochem. Mol. Biol., 26, 129-150.
Metabolism of 1- and 2-naphthylamine in isolated rat hepatocytes
Achim Orzechowski, Dieter Schrenk1 and Karl Walter Bock Institute of Toxicology, University of Tubingen, Wilhelmstrasse 56, W-7400 Tubingen, Germany 'To whom correspondence should be addressed
Introduction Aromatic amines such as 2-naphthylamine (2-NA*) were among the first chemicals recognized to induce cancer in humans and experimental animals (1,2). 2-NA, which has been used industrially and is present in ng amounts in cigarette smoke (3), induces primarily carcinomas of the urinary bladder in humans, dogs and rats (1). In contrast, no increased carcinogenic risk has been observed after treatment of animals with pure 1-naphthylamine (1-NA) (4), although epidemiological studies initially implicated 1-NA as a human carcinogen, probably resulting from concomitant exposure to 2-NA (1). In previous studies a number of metabolites of 2-NA have been isolated and characterized in vivo and in vitro. It was shown that acetyl, glucuronosyl and sulfamyl conjugates at the amino group and C-(ring-) and A'-oxidation products and their conjugates are formed (5,6). A'-oxidation has been recognized as the crucial step in the activation of 2-NA. The hydroxylamine can undergo •Abbreviations: 2-NA, 2-naphthylamine; 1-NA, 1-naphthylamine; MC, 3-methylcholanthrene; PAP, 3'-phosphoadenosine-5'-phosphate; El, electron impact; UGT, UDP-glucuronosyltransferase. © Oxford University Press
Materials and methods Chemicals [3H]2-NA (26 mCi/mmol) was a generous gift from Dr F.F.Kadlubar, National Center for Toxicological Research, Jefferson, AR. 1-NA, 2-NA, 1-nitronaphthalene, 2-nitronaphthalene and 2-hydroxy-l-arninonaphthaJene were obtained from Aldrich (Steinheim, Germany). MC, collagenase type IV, sulfatase type H-5, p-nitrophenylsulfate and 3'-phosphoadenosine-5'-phosphate (PAP) were purchased from Sigma (St Louis, MO), UDP-glucuronic acid from Boehringer (Mannheim, Germany), /S-glucuronidase from Serva (Heidelberg, Germany) and [l4C]UDP-glucuronic acid from New England Nuclear (Dreieich, Germany). The following compounds were synthesized using published procedures: JV-hydroxy-2-NA (13), W-acetyl-2-NA and JV-acetyl-1-NA (14), 2-NA-l-sulfate (15), l-hydroxy-2-NA (16) and 6-hydroxy-2-NA (5). The identity of the synthetic standards was confirmed by mass spectrometry. Biosynthesis of conjugated compounds The yV-glucuronides of 1- and 2-NA and W-hydroxy-2-NA were synthesized using liver microsomes from MC-treated rats in the following incubation mixture (17): Tris-HCl buffer, pH 7.4 (100 raM), magnesium chloride (5 mM), Brij 58 (0.5 mg/mg protein), UDP-glucuronic acid (3 mM) and 1 mg microsomal protein in a total volumtof 500 /d. The substratexoncentration of amines was 0.5 mM. After precipitating protein with methanol, N-glucuronides were isolated using the described HPLC system. The identity of the ghicuronides was confirmed by electron impact (EI)-mass spectroscopy of their silylated derivatives. The sulfate conjugates of 6-hydroxy-2-NA and 2-hydroxy-l-NA were synthesized using the following incubation mixture (18): Tris-HCl buffer, pH 7.4 (50 mM), magnesium chloride (5 mM), PAP (10 mM), /j-nitrophenylsulfate (5 mM), sodium sulfite (0.25 mM), dithiothreitol (100 mM), bovine serum albumin (0.02%) and 100 til liver homogenate (1 mg protein) in a final volume of 800 j j . The reactions were started by addition of 50 /tM of the substrate and terminated after 30 min by addition of two volumes methanol. The su I fairs were isolated by HPLC and their identity was confirmed after incubation of the isolated compounds with arylsulfatase. The liberated aminophenols were identified by their retention times and characteristic fluorescence spectra.
2227
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
The liver probably plays a major role in the metabolic activation of the bladder carcinogen 2-naphthylamine (2-NA) and hi the inactivation of the non-carcinogenic isomer 1-naphthylamine (1-NA). However, metabolic profiles of these compounds (including primary metabolites and directly determined conjugates) hi hepatocytes are not available. Therefore metabolism of 1- and 2-NA was compared hi freshly isolated hepatocytes from 3-methylcholanthrene (MO-treated and untreated rats. At 10 /iM, 2-NA was found to be mainly A'-acetylated (66% of total metabolites after 1 h incubation) and iV-ghicuronidated (19%). Minor pathways led to C-oxidation (7%) and A'-oxidation (3%; 2% present as the W-glucuronide). In hepatocytes from MC-treated rats total metabolism was slightly affected (1.5-fold increase). However, C- and A'-oxidation were markedly increased (63 and 18% respectively), while A'-acetylation and A'-glucuronidation were diminished (5 and 2% respectively). Similar experiments were carried out with 1-NA. Its A'-glucuronide was the predominant metabolite (68%) followed by the A'-acetylated compound (15%) while C-oxidation was low and W-oxidized metabolites could not be detected, even after induction. The results demonstrate that MC treatment markedly shifted 2-NA metabolism from A'-acetylation and A'-glucuronidation to N- and C-oxidation. In the case of 1-NA metabolism extensive A'-glucuronidation together with the lack of A'-oxidation may prevent carcinogenesis.
A'-glucuronidation and the resulting A'-hydroxy-A'-glucuronide is excreted into urine, where it can be hydrolyzed under the slightly acid conditions in the bladder to liberate the hydroxylamine (7). This can be further activated through protonation or acetylation to form the ultimate carcinogen, probably an electrophilic nitrenium cation. In addition, other activation pathways for 2-NA, such as the formation of reactive ortho-iminoquinones catalyzed by prostaglandin H synthase in the urinary bladder have been discussed (8,9). The metabolism of the non-carcinogenic derivative 1-NA has been studied less extensively. The presence of small quantities of A'-hydroxylated metabolites in urine of dogs and in in vitro incubations with bovine bladder epithelial microsomes has been reported (1,4,10). However, with liver microsomes from dogs, rats and humans and with purified rat P450 isozymes no A'-hydroxylation was detectable (11,12). Ring oxidation of 1-NA was consistently found to occur at positions 2 and 4 (4,11). The balance between different activating and inactivating metabolic pathways plays a decisive role in determining the potency and site of action of a given carcinogen. However, no information about the metabolism of 1- and 2-NA on the hepatocellular level is available. To elucidate the mechanisms of toxification and detoxification further, we quantitatively compared the metabolic profiles of 1- and 2-NA in isolated hepatocytes from untreated and 3-methylcholanthrene (MC)treated rats.
A.Orzecbowski, D.Schrenk and K.W.Bock Animals Male Wistar rats (180-250 g) were obtained from Savo (Kisslegg, Germany) and kept on standard diet (Altromin, Lage, Germany) and tap water ad libitum. MC was administered i.p. (40 mg/kg) in corn oil 3 days before death. Untreated animals were used as controls. Hepatocytes were obtained using the sequential EDTA-collagenase method as described by Seglen (19). The viability of the cells was determined using trypan blue. Only preparations exceeding 85% viability were used for metabolism studies. Microsomes were prepared as described previously (20).
10
20
Analysis of metabolites Aliquots (100-200 fd) were injected into a Waters HPLC system (Millipore, Bedford, MA), composed of two pumps (501/510) and a Lamda-Max 481 UV dectector (absorbance wavelength 280 nm). Additionally, either a fluorescence detector (emission 405 nm, excitation 340 nm; Sykam model S3400, Sykam, Gilding, Germany) or a radkjisotope detector (Beckman 171, Beckman Instruments, Fullerton, CA) was linked to trie UV detector. Separation of metabolites was achieved on a 4.6 x 250 mm steel column filled with Spherisorb ODS2 5 nm (Grom, Herrenberg, Germany). A linear gradient from 100% diethylammonium acetate (0.1 % in water, adjusted to pH 7.0 with acetic acid) to 65% acetonitrile over 40 min followed by a linear gradient to 100% acetonitrile over 5 min at a flow rate of 1 ml/min was used. Quantification of metabolites of 2-NA was obtained directly using the radioisotope detector by recording the percentage of total radioactivity eluting with each peak. Total elided radioactivity, including metabolites and unmetabolized substrates, was determined by liquid scintillation spectrometry on a Beckman LS 3801 liquid scintillation counter. To quantify total metabolism, except covalent protein binding, the amount of unmetabolized substrate was deducted from total eluted radioactivity. Because further oxidation of the N-hydroxy metabolite sometimes occurred during the analytic procedure, the rates of JV-hydroxylatk>n were calculated from total N-hydroxy- and nitroso compounds observed as described by Butler el al. (22). For quantification of 1-NA metabolites, standard calibration curves for the peak areas at 280 nm were recorded for A'-acetyl-l-NA and 2-hydroxy-I-NA. Calibration curves for peak areas of the N-glucuronides of 1-NA, 2-NA and N-hydroxy-2-NA were obtained by incubating the amines with [UC]UDPglucuronic acid (see above) and quantifying the glucuronides by liquid scintillation spectrometry after collecting the fractions from the HPLC. The rate of total metabolism of 1-NA was calculated from the decrease of unmetabolized substrate.
30
Retention Tune (min)
Fig. 1. Radiochromatogram of metabolites of [3H]2-NA produced in hepatocytes from (A) untreated controls and (B) MC-treated rats. Compounds identified are: 1, 2-NA-6-sulfate; 2, 2-NA-6-glucuronide; 3, 2-NA-/V-glucuronide + W-hydroxy-2-NA-Ar-glucuronide; 4, 2-NA-l-sulfate; 5, /V-hydroxy-2-NA; 6, /V-acetyl-2-NA; 7, 2-NA; 8, 2-nitrosonaphthalene. Cells were incubated with 50 nM 2-NA.
10
20
15 Rrtrwwn Time (mm)
Covalent protein binding of 2-NA metabolites Covalent protein binding in incubations of hepatocytes with [3H]2-NA was measured by the method of Uehleke et al. (23). The pellet obtained by precipitation with methanol was resuspended in 10% trichloroacetic acid and centrifuged at
Fig. 2. HPLC elution profile showing in detail UV detection of A'-hydroxy-2-NA-A'-glucuronide (3a) and 2-NA-/V-glucuronide (3b). Cells from untreated rats were incubated with 50 /iM 2-NA. Metabolites 1 and 4 are 2-NA-6-sulfate and 2-NA-l-sulfate.
Table I. Metabolite formation [nmol/h/106 cells (% of total metabolites)] from 2-NA in isolated rat hepatocytes. Effects of prctreatment with MC Metabolite
10 uM 2-NA* Controls
W-Acetyl-2-NA 2-NA-N-glucuronide 2-NA-l-sulfate A'-Hydroxy-2-NA-N-glucuronide 2-NA-6-sulfate /V-Hydroxy-2-NA 2-NA-1 -glucuronide Unidentified metabolites Total metabolism
4.4 1.2 0.3 0.15 0.12 0.06 95%.
Naphthylamine metabolism in rat bepatocytes
Instrumentation UV spectra were recorded on a Shimadzu 160A UV spectrophotometer and fluorescence spectra on a Perkin Elmer LS-5B luminescence spectrometer. El-mass spectra were obtained using a Finnigan 4021 mass spectrometer, FAB mass spectra on a Varian MAT 711A mass spectrometer.
Results Separation and quantification of metabolites of 2-NA HPLC analysis of radioactive metabolites of 2-NA formed in hepatocytes from untreated and MC-treated rats is shown in Figure 1. The metabolites identified in order of increasing retention times are 2-NA-6-sulfate, 2-NA-6-glucuronide, A
1
3H
u
2
3
4
-i.
.
*
B
1
2 10
4 30
20
Metabolic profile of 1-NA HPLC chromatograms of metabolites of 1-NA are shown in Figure 3. Identification of metabolites was based on retention times and fluorescence and UV spectra compared to synthetic standards. The number of metabolites produced by hepatocytes
(min) Fig. 3. HPLC elution profile showing UV (A) and fluorescence (B) detection of metabolites of 1-NA produced in hepatocytes from untreated rats. Compounds identified are: 1, l-NA-A'-glucuronide; 2, l-NA-2-sulfate; 3, yV-aceryl-1-NA; 4, 1-NA. The substrate concentration was 10 /iM.
Table II. Metabolite formation [nmol/h/106 cells (% of total metabolites)] from 1-NA in isolated rat hepatocytes. Effects of pretreatment with MC Metabolite
1-NA° Controls
l-NA-A'-glucuronide l-NA-2-sulfate W-Acetyl-l-NA Unidentified metabolites
4.6 0.5 1.2 0.6
± ± ± ±
0.& 0.2 0.3 0.2
Total metabolism
7.1 ± 2.4
1-NA Controls
MC-treated (68) (7) (15) (9)
6.6 0.3 0.15 0.3
± ± ± ±
1.1 0.1 0.06 0.1
7.5
± 2.7
(89) (4) (2) (4)
23.1 1.9 4.6 2.9
± ± ± ±
1.2 0.5 1.0 0.5
MC-treated (70) (7) (14) (9)
33.1 ± 3.2
'1-NA (dissolved in DMSO) was added to 50 ml of hepatocyte suspension to give a final concentration of 10 and ''Values represent means ± SD of at least three independent experiments.
24.8 3.2 1.8 2.8 35.2
± ± ± ± ±
2.1 0.9 0.61 0.7 3.2
(77) (9) (5) (8)
respectively.
2229
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
Af-hydroxy-2-NA-N-glucuronide, 2-NA-7v-glucuronide, 2-NA1-sulfate, N-hydroxy-2-NA and /V'-acetyl-2-NA. Identification of metabolites was based on their retention times and characteristic fluorescence or UV spectra compared to synthetic standards. Separation of free and conjugated compounds was sufficient for direct quantification using radiodetection, except for the N-glucuronides of 2-NA and Ar-hydroxy-2-NA, which could only be distinguished by UV detection (Figure 2). Quantification of these metabolites was achieved by using standard calibration curves. Metabolic profile of 2-NA Table I summarizes the quantitative analysis of the metabolic profile of 2-NA. The metabolites identified accounted for > 88% of total metabolites. In hepatocytes from untreated rats the two major routes of 2-NA metabolism at a substrate concentration of 10 /iM were N-acetylation (66% of total metabolites) and A'-glucuronidation (19%). MC treatment markedly increased C(63% of total metabolites) and N-oxidation (18%). In hepatocytes from untreated rats, mutagenic N-hydroxy-2-NA was present both in free and conjugated form at low levels (1 and 2% of total metabolites respectively). MC treatment led to a 2-fold increase in free N-hydroxy-2-NA, and a > 10-fold increase in N-hydioxy2-NA-Af-glucuronide. Covalent binding to protein was slightly increased from 0.020 ± 0.006 to 0.029 ± 0.007 nmol/mg protein (means ± SD of four experiments). The effect was not significant using Student's f-test. Raising the substrate concentration from 10 to 50 /tM resulted in a 1.9-fold increase in total metabolism. The proportion of N-acetyl-2-NA (31 % of total metabolites) was decreased and an increased proportion of 2-NA-N-glucuronide (35%), ringhydroxylated 2-NA-l-sulfate (19%) and 2-NA-6-sulfate (4%) as well as A/-hydroxy-2-NA-A/-glucuronide (4%) was found. Similar to the results at 10 /iM, MC treatment diminished the formation of N-acetyl-2-NA to 2% of total metabolites, and led to a relative decrease of 2-NA-N-glucuronide (14%). Total metabolism was enhanced by MC treatment by a factor of 2.6, mainly due to a strong increase in the formation of ring-hydroxylated metabolites, and the formation of N-hydroxylated products was stimulated — 5-fold. A concomitant, significant rise in covalent protein binding from 0.079 ± 0.010 nmol/mg in controls to 0.160 ± 0.015 nmol/mg after MC treatment was obtained (means ± SD of four experiments; P 2 0.001). When the incubation time was increased to 2 h, no marked alterations of the metabolic patterns of 2-NA were observed.
100 000 gfor 10 min. The procedure was repeated with 10 and 5% trichloroacetk acid at room temperature and 90°C, with ethanol at 90°C in closed vials, chloroform:methanol (1:1, v/v) and dry ethylether at room temperature. The final pellet was gently heated in a water bath to remove residual ether and was dissolved in 5 ml 1 N NaOH. After neutralization with 1 N HC1, aliquots were removed for liquid scintillation counting and procein determination. Enzyme assays 1- and 2-NA glucuronidation was assayed according to the fluorometric method described by Lilienblum and Bock (17). Protein was determined according to Lowry el at. (24).
A.Orzechowski, D.Schrenk and K.W.Bock
-1
-0.5
0.0
OS
1.0
\S
2.0
25
Table III. Apparent kinetic parameters of 1- and 2-NA glucuronidation Compound (mM) Untreated controls 1-NA 2-NA MC treatment 1-NA 2-NA
*MI
KM2 *MI KM2
V max (nmol/min/mg)
Od/min/mg)
0.45 2.5
90 21
200 8
0.07 0.40 °- 5 3.8
30 220 26 57
600 550 52 15
of untreated and MC-treated rats was much lower than with 2-NA as substrate. Three metabolites could be characterized as 1-NAN-glucuronide, l-NA-2-sulfate and JV-acetyl-1-NA. Compounds eluting at retention times less than 12.5 min were also found in 1-NA-free incubations. Since radiolabeled 1-NA was not available, covalent protein binding could not be assessed. As shown in Table II, at substrate concentrations of 10 and 50 nM, 1-NA-N-glucuronide was the predominant metabolite (68 and 70% of total metabolites respectively). Formation of JV-acetyl-1-NA was lower (15 and 14%) and ring-hydroxylation, finally leading to l-NA-2-sulfate accounted for only 7% of total metabolism. W-Hydroxylated metabolites were not detectable in any incubation with 1-NA. Pretreatment with MC did not lead to a marked alteration of the overall pattern of metabolites except for a decrease in N-acetyl-1-NA formation. Interestingly, in hepatocytes from untreated animals, the total rate of metabolism at 50 /tM was much higher with 1-NA than with 2-NA. 2230
Discussion In isolated rat hepatocytes seven metabolites of 2-NA, also detected in vivo (25), were identified and quantified by direct monitoring of primary and conjugated metabolites. Since primary metabolites of 2-NA such as N-hydroxy-2-NA and 1-hydroxy2-NA are readily oxidized in aqueous solution, recording of the conjugates may allow more exact quantification of metabolic pathways than hydrolyzing the conjugates and analyzing the liberated primary metabolites. With hepatocytes from untreated rats the major routes of metabolism were N-acetylation and N-glucuronidation. These findings may reflect the low constitutive expression of isoforms of the CYP1A family which are primarily responsible for the oxidation of aromatic amines (11,26). The high percentage of N-acetyl-2-NA found in rat hepatocytes is not observed in urine or bile of rats treated with 2-NA (25). Acetylation does not markedly alter the polarity of 2-NA. Hence, further metabolism of this compound may be required in vivo to facilitate its excretion. However, an extended incubation time did not lead to a marked decrease of N-acetyl-2-NA, suggesting that extrahepatic sites are involved in the further metabolism of N-acetyl-2-NA. /V-Acetyl-2-NA probably represents a detoxification product of 2-NA. It was shown in vivo that acetylation shifts the site of ring-oxidation to 6-mono- and 5,6-dihydroxylated metabolites regarded as detoxified metabolites (25) and that no hydroxamic acids were found when animals were treated with N-acetyl-2-NA (5). The /V-glucuronide of /V-hydroxy-2-NA has been suggested as a transport form in a sequence of events leading to bladder cancer. The glucuronide is formed preferentially in the liver by MCinducible UGT isozyme (27) and, after release into the systemic circulation, ultimately reaches the urinary bladder where it liberates the hydroxylamine forming an electrophile which binds to DNA (7). DNA adducts formed in the bladder epithelium are thought to result ultimately in the formation of bladder cancer (2). Our findings demonstrate that /V-hydroxy-2rNA-/V-glucuronide and unconjugated N-hydroxy-2-NA are released from rat hepatocytes. The latter compound may enter the bladder lumen via the systemic circulation and bind directly to DNA as has been proposed for 4-aminobiphenyl (2). Formation of N-hydroxy2-NA-glucuronide was increased in hepatocytes from MC-treated rats. This effect, together with a reduced formation of the detoxified metabolites N-acetyl-2-NA and 2-NA-N-glucuronide
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
Fig. 4. Hanes plots of UGT activity towards 1-NA in microsomes of untreated (A) and MC-treated (B) rats. The S/V value is expressed in mg x min/1.
Kinetic analysis of UDP-glucuronosyltransferase (UGT) activity towards 1- and 2-NA Rat liver microsomes from control and MC-treated rats were used for kinetic analysis of N-glucuronidation. As shown in Figure 4, biphasic enzyme kinetics of UGT activity towards 1-NA can be detected after MC treatment. Similar results were observed with 2-NA (not shown), suggesting the induction of (an) isozyme(s) with higher affinity towards 1- and 2-NA. Kinetic data for both substrates are compared in Table HI. Comparison of the apparent Km values shows that 1-NA is glucuronidated in microsomes from untreated rats with a 5-fold higher affinity than 2-NA. From experiments with liver microsomes of MC-treated rats, two apparent Km values were calculated for each substrate, both constants being ~ 10-fold higher (lower affinity) for 2-NA than for 1-NA. Comparison of the ratio V^JKm (corresponding to the 'intrinsic clearance') demonstrates that 1-NA is glucuronidated far more effectively than 2-NA.
Naphthylamine metabolism in rat bepatocytes
Acknowledgements The authors wish to thank Mrs S.Beck and Mrs S.Pahl for expert technical assistance. This project was supported by a research grant from the Deutsche Forschungsgemeinschaft, Bonn-Bad Godesberg.
References 1. Radomsld J.L. (1979) The primary aromatic amines, their btotogkal properties and structure—activity relationships. Annu. Rev. PharmacoL Toxicol., 19, 129-157. 2. Beland,F.A. and Kadlubar.F.F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. In Cooper.C.S. and Grover.P.L. (eds), Chemical Carcinogenesis and Mutagenesis I. SpringerVerlag, Berlin, pp. 267-325. 3. Patriankos.C. and Hoffmann.D. (1979) Chemical studies on tobacco smoke. LXIV. On the analysis of aromatic amines in cigarette smoke. J. AiuL Toxicol, 3, 150-154. 4. Radomsky.L., Deichmann,W.B., Altmann.N.H. and Radomski.T. (1980) Failure of pure 1-naphthylamine to induce bladder tumors in dogs. Cancer Res., 40, 3537-3539. 5. Boyland,E. and Manson.D. (1963) The biochemistry of aromatic amines: the metabolism of 2-naphthylamine and 2-naphthylhydroxylamine derivatives. Biochem. J., 101, 84-102. 6. Kadlubar.F.F., Unruh.L.E., Flammang.TJ., Sparks.D., Mitchum.R.K. and Mulder.G.J. (1981) Alteration of urinary levels of the carcinogen W-hydroxy2-naphthylamine and its Ar-glucuronide in the rat by control of urinary pH, inhibition of metabolic sulfation, and changes in biliary excretion. Chem.-Biol. Interactions, 33, 129-147. 7.Kadlubar,F.F., MillerJ.A. and Miller.C.E. (1977) Hepatic microsomal Af-glucuronidation and nucleic acid binding of JV-hydroxy arylamines in relation to urinary bladder carcinogenesis. Cancer Res., 37, 805-814. 8. Kadlubar.F.F., Frederick.C.B., Weiss,C.C. and Zenser.T.V. (1982) Prostaglandin endoperoxide synthetase-mediated metabolism of carcinogenic aromatic amines and their binding to DNA and protein. Biochem. Biophys. Res. Commun., 108, 253-258. 9. BoydJ.D. and Eling.E.T. (1987) Prostaglandin H synthase-catalyzed metabolism and DNA binding of 2-naphthylamine. Cancer Res., 47, 4007-4014. 10. PoupkoJ.M., Radomslti.T., Santella.R.M. and RadomskiJ.L. (1983) Organ, species and compound specificity in the metabolic activation of primary aromatic amines. J. NatL Cancer Inst., 70, 1077-1080. 11. Hammons.GJ., Guengerich.F.P., Weis.C.C, Beland.F.A. and Kadlubar.F.F. (1985) Metabolic oxidation of carcinogenic arylamines by rat, dog and human hepatic microsomes and by purified flavin-containing and cytochrome P-450 monooxygenases. Cancer Res., 45, 3578-3585. 12. Brill,E. and RadomskiJ.L. (1971) Comparison of in vitro and in vivo N-oxidation of carcinogenic aromatic amines. Xenobiotica, 1, 347—348. 13. Boyland,E., Manson.D. and Nery.R. (1962) The reaction of phenylhydroxylamine and 2-naphthylhydroxylamine with thiols. /. Chem. Soc., 1962, 606-611. 14. BoothJ., Boyland.E. and Manson.D. (1955) Metabolism of polycyclic compounds. 9. Metabolism of 2-naphthylamine in rat tissue slices. Biochem. J., 60, 6 2 - 7 1 . 15. BoylandJ*., Manson.D. and Sims,P. (1953) The preparation of o-aminophenyl sulfates. J. Chan. Soc., 1953, 3623-3630. 16. Desai.R.D., Hunter.R.F. and Khaladi.R.K. (1938) The unsaturation and tautomeric mobility of heterocyclic compounds. J. Chem. Soc., 1938, 321-329. 17. Lilienblum.W. and Bock.K.W. (1984) N-Glucuronide formation of carcinogenic aromatic amines in rat and human liver microsomes. Biochem. PharmacoL, 33, 2041-2046. 18. FJ Mouelhi.M. and Kaufrman.F.C. (1986) Sublobular distribution of transferases and hydroiases associated with ghicuronide, suliate and ghitathione conjugation in human liver. Hepatology, 6, 450-456. 19. Seglen.P.O. (1976) Preparation of isolated rat liver cells. Methods Cell BioL, 18, 2 9 - 8 3 . 20. Bock.K.W. and White.I.N.H. (1974) UDP-glucuronosyltransferascs in perfused rat liver and in microsomes: influence of phenobarbital and 3-nWhylcholanrhrene. Eur. J. Biochem., 46, 451-459. 21.Schrenk,D. andBockJCW. (1990) Metabolism of benzene in rat nepatocvtes. Influence of inducers on phenol glucuronklalian. Drug. Mewb. Dispos., 18, 720-725. 22. Buuer.M.A., Guengerich.F.P. and Kadlubar.F.F. (1989) Metabolic activation of the carcinogens 4-aminobiphenyl and 4,4'-methylene-bis(2-chloraniline)
2231
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
is likely to increase the burden of genotoxic intermediates reaching the bladder. Similarly, induction of hepatic cytochrome P4501A2 may increase the risk of bladder carcinogenesis in smokers in a synergistic manner. 2-NA is not known to be a hepatocarcinogen. Covalent binding to hepatic macromolecules does not seem to reflect a carcinogenic risk. However, an increased formation of reactive 2-NA metabolites after MC-type induction may lead to an increased binding to liver DNA and may thus initiate hepatocarcinogenesis. In fact, covalent protein binding with 50 /iM 2-NA was significantly increased in hepatocytes from MC-treated rats. Similar experiments using 32P-postlabeling of DNA bases are currently under way to determine effects on DNA binding. Direct N-glucuronidation of 1- and 2-NA probably represents a detoxifying reaction. However, cleavage of the conjugates, e.g. by intestinal glucuronidases, may lead to reabsorption of the parent compound from the gut. Glucuronidation of 1- and 2-NA has been shown to be preferentially catalyzed by MC-inducible phenol UGT (17,28). However, in hepatocytes from MC-treated rats, induction of this enzyme does not appear to compensate for the induction of P4501A isozymes, with the effect that 2-NA-glucuronide was a minor metabolite. The total rate of metabolism at a substrate concentration of 10 /tM was similar for 2-NA and its non-carcinogenic isomer 1-NA. At 50 /iM, however, an almost linear increase in total metabolism was observed with 1-NA but not with 2-NA. A lower than proportional increase may indicate saturation of a ratelimiting step in 2-NA metabolism at this substrate concentration. Comparison of the metabolic profiles with both isomers revealed very different patterns of metabolites. 1-NAN-glucuronide was the predominant metabolite of 1-NA in incubations at both substrate concentrations as well as with cells from MC-treated rats. In agreement with previous findings in liver microsomes no Ak>xidau'on products were detected (10,11). From kinetic studies in rat liver microsomes an apparent Km value of 0.5 mM was calculated from a MC-inducible UGT activity towards 2-NA. This finding suggests that 2-NA is a poor substrate even for the MC-inducible phenol UGT. In contrast, a significantly higher affinity for both constitutive and MC-inducible isoforms of UGT and a much higher 'intrinsic clearance' was calculated with 1-NA as substrate. Induction of a high affinity UGT with a low Km value of ~50/iM appears to compensate the induction of P4501A isozymes resulting in a balance between N-glucuronidation and ring-hydroxylation of 1-NA. Failure of 1-NA to induce tumors in experimental animals has been based solely on the lack of TV-oxidation (10). However, 1-NA was found to act as a mutagen in the Ames test (29) and it has been suggested that 2-hydroxy-l-NA, the major metabolite formed in microsomal incubations, has some potential to form an electrophilic iminoquinone, which might react with DNA (2). UGTs play a major role in the metabolism of aromatic and heteroaromatic amines and their roles in chemical carcinogenesis have been recently reviewed (30). Our results show that 1-NA is extensively A'-glucuronidated and as a consequence can be rapidly excreted. Efficient N-glucuronidation may thus prevent accumulation of 1-NA and its further oxidation to toxic metabolites. Further comparative studies with other carcinogenic arylamines are necessary to extend our understanding of the highly complex pathways of this class of compounds leading to distinct patterns of organ-directed carcinogenicity.
A.Orzechowsld, D.Schrenk and K.W.Bock
Received on July 27, 1992; revised on September 21, 1992; accepted on September 24, 1992
2232
Downloaded from http://carcin.oxfordjournals.org/ at North Carolina State University on May 2, 2015
by human hepatic microsomes and purified rat hepatic cytochrome P-450 monooxygenases. Cancer Res., 49, 25—31. 23. Uehleke.H., HeUmer.K.H. and Tabarelli-Poplawski.S. (1976) Metabolic activation of halothane and its covalent binding to liver endoplasmatic proteins in vitro. Naunyn Schmiedbergs Arch. Pharmacol., 279, 38-48. 24. Lowry.O.H., Rosebrough.A.L., Farr,A.L. and Randall.RJ. (1951) Protein measurement with the Folin reagent. /. Biol. Chem., 193, 265-275. 25. Gorrod.J.W. and Manson.D. (1986) The metabolism of aromatic amines. Xenobiotica, 16, 933-955. 26. Butler.M.A., Iwasaki.M., Guengerich.F.P. and Kadlubar.F.F. (1989) Human cytochrome P-450pA (P-450IA2), the phenacetin Odecthylase, is primarily responsible for the hepatic 3-demethylation of caffeine and N-oxidation of carcinogenic arylamines. Proc. NatL Acad. Sci. USA, 86, 7696-7700. 27. Bock.K.W., Josting.D., LUienbhim.W. and Pfeil.H. (1979) Purification of rat-liver microsomal UDP-glucuronosyl-transferase. Separation of two forms inducible by 3-methylchc4anthrene or phenobarbhal. Eur. J. Biochem., 98, 19-26. 28. Green.M.D. and Tephly.T.R. (1987) N-Glucuronidation of carcinogenic aromatic amines catalyzed by rat hepatic microsomal preparations and purified undine 5'Kliphosphateghjcuronosyttransferases. Cancer Res., fl, 2028-2031. 29. McCannJ., Choi.E., Yamasaki.E. and Ames.B.N. (1975) Detection of carcinogens as mutagens in the Salmonella Imicrosomc test: assay of 300 chemicals. Proc. Natl. Acad. Sci. USA, 72, 5135-5139. 30. Bock.K.W. (1991) Roles of UDP-glucuronosyltransferases in chemical carcinogenesis. Oil. Rev. Biochem. Mol. Biol., 26, 129-150.
