XENOBIOTICA,
1992, VOL. 22,
NO.
12, 1403-1423
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
S. BOEREN?, B. TYRAKOWSKAtS, J. VERVOORTT, E. DE HOFFMAN$, K. TEUNIS", A. VAN VELDHUIZEN" and I . M. C . M. RIETJENSTY
t Department of Biochemistry, Agricultural University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands. 3 Universiti. Catholique de Louvain, Department of Mass Spectrometry, Place Louis Pasteur 1, B-1348, Louvain-la-Neuve, Belgium Department of Organic Chemistry, Agricultural University, Dreijenplein 8, 6703 BC Wageningen, T h e Netherlands Received 2 January 1992; accepted 1I July 1992
1. Rat liver microsomal metabolism of 2-fluoro-, 2-chloro- and 2-bromo-4methylaniline was investigated using h.p.1.c. Metabolites identified include products from side-chain C-hydroxylation (benzyl alcohols and benzaldehydes) and N-hydroxylation (hydroxylamines and nitroso derivatives). Aromatic ring hydroxylation was not a major reaction pathway. 2. A new type of microsomal metabolite was detected which was identified as a secondary amine, i.e. a halogenated N-(4'-aminobenzyl)-4-methylaniline. 3. In addition to these products azoxy, azo and hydrazo derivatives were formed. 4. Benzyl alcohols and halogenated N-(4'-aminobenzyl)-4-methylanilines were the major microsomal metabolites for all three 2-halogenated 4-methylanilines.
5. Quantification of the metabolite patterns demonstrated an influence of the type of halogen substituent on the rate of microsomal metabolism. T h e rate of side-chain Chydroxylation increases in the order 2-fluoro-4-methylaniline < 2-chloro-4-methylaniline < 2-bromo-4-methylaniline. 6. T h e rate of N-hydroxylation increases from 2-bromo-4-methylaniline < 2-fluoro-4That 2-chloro-4-methylaniline is Nmethylaniline < 2-chloro-4-methylaniline. hydroxylated to a larger extent is in accordance with its greater mutagenicity, twice that of 2-bromo-4-methylaniline.
Introduction Methylaniline (toluidine) and its derivatives are used in industrial manufacturing processes of dyes, pigments, agrochemicals and plastics. Exposure to methylaniline derivatives is not limited to production facilities and usage areas only, but may also result from industrial pollution and bacterial or chemical breakdown of the derivatives (Leslie et al. 1988, Knowles and Sen Gupta 1970, Zimmer et al. 1980, Yoshimi et al. 1988). T h e toxicity and especially mutagenicity of various methylanilines has been well documented (Yoshimi et al. 1988, Zimmer et al. 1980, Weisburger et al. 1978). Although it is known that the methylanilines have to be metabolized to their ultimate toxic and mutagenic form, products from N-hydroxylation especially contributing to this toxic effect (Zimmer et al. 1980, Hecht et al. 1979, Weisburger and Weisburger 1973), few studies report on their metabolism.
5 Present address: Institute of Commodity Science, Academy of Economics, Al. Niepodleglosci 10, 60-967 Poznan, Poland. 7 To whom correspondence should be addressed. 0049-8254/92 $3.00 0 1992 Taylor & Francis Ltd.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
1404
S. Boeren et al.
In an in vivo study using rats, Son et al. (1980) showed that the main urine metabolites formed from 2-methylaniline were conjugated derivatives of ring hydroxylated products, and metabolites derived from side-chain C- and N-hydroxylation. Aromatic ring hydroxylation was also reported to occur in in vivo studies with 2-, 3- and 4-methylaniline (Cheever et al. 1980) and 4-trifluoromethylaniline (Wade et al. 1988). From in vitro liver microsomal metabolism data, Daly et al. (1 968) suggested that 4-methylaniline was converted to 4-aminobenzyl alcohol and 4-aminobenzaldehyde as a result of side-chain C-hydroxylation; aromatic ringand N-hydroxylation were not reported. Furthermore, Kiese (1963) provided evidence for the in vivo N-hydroxylation of 2-, 3- and 4-methylaniline in dogs, by demonstrating the formation of the corresponding nitrosotoluenes and/or methaemoglobin formation in the blood of dogs exposed to these methylanilines. For methylanilines halogenated in the aromatic ring, the only metabolism studies have been with 4-chloro-2-methylaniline, the metabolic breakdown product of the insecticide/acaracide chlordimeform (N'-(4-chloro-o-tolyl)-N,N-dimethylformamidine) (Knowles and Sen Gupta 1970), which undergoes side-chain Chydroxylation and further oxidation to give 5-chloro-anthranilic acid; no other metabolic products were identified. In spite of this lack of knowledge on the metabolism of halogenated methylanilines, their mutagenicity is well documented and known to vary with the substituent pattern (Zimmer et al. 1980, Yoshimi et al. 1988). Zimmer et al. (1980) for instance, reported that the mutagenicity towards Salmonella typhimurium TA100 of 2-chloro-4-methylaniline in the presence of S9 mix was twice that of 2-bromo-4methylaniline. Although mutagenicity of halogenated methylanilines is known to result from metabolism, no knowledge exists concerning the effect of a halogen substituent in the aromatic ring on the metabolism of methylanilines. Therefore, we decided to study the biotransformation of a series of three related halogenated methylanilines, namely, 2-fluoro-, 2-chloro- and 2-bromo-4-methylaniline. This paper describes the identification and quantification of the major metabolites formed from 2-fluoro-, 2-chloro- and 2-bromo-4-methylaniline by rat liver microsomes. T h e results demonstrate a qualitative and quantitative influence of the type of halogen substituent on the metabolites obtained.
Experimental Chemicals 2-Fluoro-4-methylaniline and 2-fluoro-4-methyl-nitrobenzene were purchased from Fluorochem (Derbyshire, UK). 4-Methylaniline, 2-chloro-4-methylaniline, 2-bromo-4-methylaniline, 4-nitrobenzyl alcohol and 2-hydroxy-4-methylaniline were purchased from Janssen Chimica (Beerse, Belgium). 3-Fluoro-4-aminobenzaldehyde,3-chloro-4-aminobenzaldehyde and 3-bromo-4-aminobenzaldehyde were synthesized from the corresponding 2-halogenated 4-methylanilines using 2,3-dichloro-5,6dicyano-l,4-benzoquinone(Janssen), according to the method described by La1 et al. (1984) for synthesis of 3,5-dimethyl-4-aminobenzaldehydefrom 2,4,6-trimethylaniline. 3-Fluoro-4-aminobenzyl alcohol, 3-chloro-4-aminobenzyl alcohol and 3-bromo-4-aminobenzyl alcohol were obtained by reduction of the corresponding benzaldehyde with sodium borohydride following the procedure described by Schenker (1961). 4-Aminobenzyl alcohol was synthesized by the reduction of 4-nitrobenzyl alcohol, dissolved in ethyl acetate, under 4atm H, in the presence of O.l%(w/v) palladium/carbon (Vogel 1989). 2-Fluoro-4-methylnitrosobenzene, 2-chloro-4-methylnitrosobenzene and 2-bromo-4methylnitrosobenzene were synthesized by oxidation of the corresponding methylaniline with potassium peroxymonosulphate in a water-acetic acid mixture using the procedure described by Kennedy and Stock (1960) for the synthesis of 1,4-dinitrosobenzene.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1405
2-Fluoro-4-methylhydroxylaminobenzeneand 2-chloro-4-methylhydroxylaminobenzenewere obtained by palladium-catalysed reduction of 2-fluoro-4-methylnitrobenzene and 2-chloro-4methylnitrobenzene carried out as described by Stavenuiter et al. (1985) for the synthesis of N-hydroxy5-phenyl-2-pyridinamine from 2-nitro-5-phenylpyridine using hydrazine hydrate in tetrahydrofuran. 2-Bromo-4-methylhydroxylaminobenzene was synthesized by reduction of 2-bromo-4methylnitrosobenzene in water with 10 times excess of NADH. Synthesis of N-(4'-amino-3'-fluorobenzyl)-2-fluoro-4-methylaniline, N-(4'-amino-3'-chlorobenzyl)2-chloro-4-methylaniline and N-(4'-amino-3'-bromobenzyl)-2-bromo-4-methylaniline was achieved by dropwise adding 1.85 g 2,3-dichloro-5,6-dicyano-l,4-benzoquinone(8.5 mmol), dissolved in 20 ml 1,4dioxane, to a solution of 1.Og 2-fluoro-4-methylaniline or 1.2g 2-chloro-4-methylaniline or 1.5 g 2bromo-4-methylaniline (8.1 mmol) in 5 ml 1,4-dioxane. After stirring at room temperature for 10 min, the mixture was poured into a solution of 3.0 g NaBH, (79 mmol) in lOOml water. T h e solution was stirred for lomin, 200ml 1 M KH,PO, were added and stirring was continued for another 30min. T h e compound was extracted into ethyl acetate, folIowed by washing three times with water, three times with saturated aq. NaCl and filtration over a Whatman 1 ps filter. 2,2'-Difluoro-4,4'-dimethylazoxybenzene and 2,2'-dichloro-4,4'-dimethylazoxybenzenewere synthesized by oxidation of the respective 4-methylanilines with potassium peroxymonosulphate in a 1 : 1 water-acetic acid mixture using a modification of the method described above for the synthesis of 2fluoro- and 2-chloro-4-methylnitrosobenzene. T h e azoxybenzenes were prepared by slowly adding 1.50g potassium peroxymonosulphate (2.4 mmol) dissolved in 10 ml water to a solution of 0.50 g 2-fluoro-4methylaniline or 0 5 7 g 2-chloro-4-methylaniline (4.0 mmol) in acetic acid, stirring at room temperature for 30 min. Crystals formed upon the addition of 50 ml water were collected by filtration and washed with water. 2,2'-Difluoro-4,4'-dimethylazobenzeneand 2,2'-dichloro-4,4-dimethylazobenzenewere synthesized by reduction of 2,2'-difluoro- and 2,2'-dichloro-4,4'-dimethylazoxybenzenerespectively with lithium aluminium hydride, carried out essentially as described by Schenker (1961). A mixture of 2,2'-dibromo-4,4'-dirnethylazoxybenzene and 2,2'-dibromo-4,4'-dimethylazobenzene was obtained using a modification of the procedure described by Holmes and Bayer (1960) for the synthesis of 2,6-dibromonitrosobenzene from 2,6-dibromoaniline. In short, 1.52 g (8 mmol) 2-bromo-4methylaniline in 7 ml acetic acid and 120pl H,SO, were oxidized by addition of 2 ml of 30% hydrogen peroxide and stirring at 4&50"C for 16 h. Upon addition of 5 ml water the compounds were precipitated by centrifugation. T h e pellet was dissolved in ethanol and the mixture of the two compounds was recrystallized upon addition of water and standing at 4°C for 3 days. 2,2'-Dichloro-4,4'-dimethylhydrazobenzene was synthesized from 2,2'-dichloro-4,4'-dimethylazoxybenzene using hydrazine/Raney nickel according to the method described by Furst and Moore (1957). 2,2'-Dibromo-4,4'-dimethylhydrazobenzene was synthesized the same way as 2,2'-dichloro-4,4'dimethylhydrazobenzene. Complete purification of this bromo derivative was not performed because it was hampered by its instability under aerobic conditions resulting in formation of 2,2'-dibromo-4,4'dimethylazobenzene. Generally, compounds were further purified by preparative chromatography on silicagel 60 (Merck, Darmstadt, Germany), using a mixture of hexane plus a suitable modifier (for example dichloromethane or acetonitrile) as the eluent. Structures were confirmed by 'H-n.m.r. mass spectrometry, i.r. and I3Cn.m.r. Details of spectral characteristics of the compounds synthesized by newly described or modified procedures are presented in the results section (table 1). Spectral analysis of synthetic reference compounds 'H-n.m.r. and 13C-n.m.r. spectra were measured on a Bruker AC 200 spectrometer. Non-deuterated impurities in the deuterated solvents were used as internal standard; for 'H-n.m.r. these were CHCl, at 7.24ppm and CH,COCH, at 2.04ppm, for I3C-n.m.r. these were CDCI, at 77.0ppmand CD,COCD, at 29.8 and 2060 ppm. Mass spectra were recorded using a Hewlett Packard 5970B g.1.c.-mass spectrometry system with a 30m capillary DB17 column. Accurate masses were determined on a AEI-902 mass spectrometer equipped with a VG-2AB console, and i.r. spectra were recorded on a Philips PU 9706 spectrometer. Preparation of microsomes Liver microsomes were prepared as described before using untreated male Wistar rats (300-350 g) (Rietjens and Vervoort 1989). Microsomal incubations Microsomal incubations were carried out in 0.1 M (final concentration) potassium phosphate p H 7.6 containing 1 mM 4-methylaniline substrate (added as a 0.1 M stock solution in dimethyl sulphoxide), 2 mM NADPH and 2 p~ microsomal cytochrome P450. Reactions were started by the addition of NADPH. After 10min at 37°C reactions were terminated by freezing the reaction mixture in liquid nitrogen for h.p.1.c. analysis, or by mixing with trichloroacetic acid (2.0 ml incubation mixture +0.6ml 200g/l
1406
S . Boeren et al.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
trichloroacetic acid) for chemical determination of N-hydroxylated products. Blank experiments were performed by adding water instead of NADPH, or dimethyl sulphoxide instead of the 4-methylaniline solution. T o study the involvement of the cytochrome P450 system in the formation of the various metabolites, to some incubations NADH (2mM final concentration) instead of NADPH, or the cytochrome P450 inhibitor octylamine (Cashman and Hanzlik 1981) (7 mM final concentration) were added. Other incubations were performed in the presence of CO, or made anaerobic by three cycles of evacuation and filling with argon and the addition of glucose (1OmM final concentration) plus glucose oxidase (Boehringer, Mannheim, Germany) (0.1 mg/ml final concentration). T o prevent any reaction in these anaerobic samples as a result of aerobic sample preparation for h.p.1.c. injection, the cytochrome P450 inhibitor octylamine was added to a final concentration of 7 mM when terminating the anaerobic reaction.
Preparation of h.p.1.c. samples Microsomal incubation samples, frozen in liquid nitrogen to terminate the reaction, were thawed and centrifuged at 4°C at 13000g for 1Omin. For standard h.p.1.c. analysis Sop1 of the supernatant was injected directly into the h.p.1.c. (see below). For g.1.c.-mass spectrometry of metabolite peaks, 20 ml of incubation mixture were freeze-dried to approximately 1 ml. These concentrated samples were centrifuged for 10min at 13 OOOg and 80Opl of the supernatants thus obtained were injected into the h.p.1.c. using a pre-concentration column as described below. Analysis by h.p.1.c. A Kratos 400 high pressure gradient system, a 100 x 3.0 mm CP-LiChrosorb RP C8 column and a Kratos 757 UVjVIS detector (8 mm light path) were used. T h e detection wavelength was 295 nm. T h e gradient used consisted of eluting with 100%water for 1 rnin, followed b y a linear decrease to 202, (v/v) water in methanol in 26 min and keeping the percentage of water in methanol at 20% (v/v) for 6 min. Standard injections were carried out using an external Sop1 loop. For injections for the g.1.c.-mass spectrometry experiments the Sop1 injection loop was replaced by a 10 x 3.0mm ChromSpher C18 preconcentration column and 800 p1 of the concentrated sample were injected. Analysis by g.1.c.-mass spectrometry Peak fractions collected from the h.p.1.c. run were freeze-dried and the residues were dissolved in 20p1 dichloromethane. Of these samples 1 pI was injected into the g.1.c.-mass spectrometry system. T h e Varian 3700 gas chromatograph of the system was equipped with a 30 m x 0.32 mm chemically bonded RSL-200 column. Helium was used as the carrier gas at a flow of 1 ml/min. During the run a linear temperature gradient from 50 to 200°C was applied in 21 min. The column was directly coupled to a Finnigan-Mat TSQ70 mass spectrometer operating in the electron impact mode at 70eV. Determination of molar extinction coeficients A known amount of compound (10--50mg)was dissolved in 25 ml of a suitable solvent, e.g. 50% (v/v) methanol in water or 20% (v/v) methanol in ethanol. Small amounts of these stock solutions were diluted in the water-methanol mixtures at which the respective metabolites elute from the h.p.1.c. column. The absorbance of these samples at 29.5 nm was determined using a LKB Ultrospec I11 spectrometer. Chemical assay for N-oxidation products N-oxidation products (hydroxylamine plus nitroso derivatives) were determined essentially as described by Herr and Kiese (1959). In short, 1.0ml water and 0.30 ml 10%(w/v) potassium ferricyanide in 6 M HCI were added to 2.0 ml trichloroacetic acid-precipitated microsomal supernatant to oxidize the hydroxylamine to the corresponding nitroso derivative. These samples were extracted with 2 ml CCl, and the tetra phases were washed twice with 2 . 5 H,SO,. ~ T o 1.0ml of the washed tetra phases, 1.0ml of glacial acetic acid and 50pl 20%(w/v) sodium nitrite were added. After lSmin, 1 0 0 ~ 15O%(w/v) ammonium sulphamate were added and the samples were mixed for 1Omin. Finally, 2 5 0 ~ 180%(v/v) acetic acid and 50 pl N-(1 -naphthyl)rthylenediamine dihydrochloride (Aldrich, Steinheim, Germany) were added. After 2 h in the dark the absorbance of the upper water layer at 555 nm was determined. Using synthetic 2-fluoro-, 2-chloro- and 2-bromo-4-methylnitrosobenzene, molar extinction coefficients for this assay at 555nm of respectively 13.5, 14.7 and 16,6mM-'cm-' were obtained. Determination of apparent dissociation constants (Kd,,,) T o determine apparent dissociation constants (Kd,,,) for the binding of the 2-halogenated 4methylanilines to microsomal cytochromes P450, 1.0 ml of a microsomal preparation containing 1.0 p~ cytochrome P450 in 0.1 M potassium phosphate p H 7 6 was titrated with 2-fluoro-4-methylaniline, 2chloro-4-methylaniline or 2-bromo-4-methylaniline using small volumes of a 50 mM methylaniline stock . the solution in dimethyl sulphoxide up to a maximal final methylaniline concentration of 2 . 0 m ~During stepwise addition of the 2-halogenated 4-methylanilines, difference spectra were recorded against reference cuvettes containing a similar microsomal dilution to which identical volumes of dimethyl
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1407
sulphoxide were added. Difference spectra were recorded from 300 to 500 nm using an Aminco DW-2A spectrophotometer. 2-Fluoro-4-methylaniline, 2-chloro-4-methylaniline and 2-bromo-4-methylaniline all gave rise to a so-called type I1 spectral change with a maximum at 430 nm and a minimum at 400410nm. The absorbance differences (AA) between this maximum and minimum in the spectra were determined. The Kdappwas calculated from a linear double reciprocal plot of AA-’ versus the 4methylaniline concentration- I .
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Results Spectral characteristics of newly described reference compounds Table 1 presents the ‘H-n.m.r., 13C-n.m.r., i.r., and mass analysis of the synthetic reference compounds prepared by newly described or modified procedures. The table also presents yields of these syntheses as well as characteristics of the compounds after purification. Benzaldehydes, benzyl alcohols, hydroxylaminobenzenes and nitroso derivatives were prepared by well described procedures. Their structures were also confirmed by spectral and mass analysis (data not shown). Microsomal metabolite patterns of the 2-halogenated-4-methylanilines Figure 1 presents the reversed-phase h.p.1.c. elution pattern of a microsomal incubation with 2-chloro-4-methylaniline (figure 1 A) and the corresponding blanks carried out in the absence of NADPH (figure 1 B) or in the absence of 2-chloro-4methylaniline (figure 1 C). Comparison of the chromatograms in figure 1 reveals that at least 10 metabolites are formed from 2-chloro-4-methylaniline. Additional experiments demonstrated product formation to be linear with the amount of microsomes added up to at least 2 p microsomal ~ cytochrome P450 and linear in time for at least 10 min (data now shown). Figure 2 shows a schematic presentation of the h.p.1.c. elution patterns of rat liver microsomal incubations with 2-fluoro-4methylaniline, 2-chloro-4-methylaniline and 2-bromo-4-methylaniline. I n this presentation, blank peaks were omitted. Identification of the metabolites presented in figures 1 and 2 was performed on the basis of h.p.1.c. retention times of the synthetic reference compounds. I n addition, g.1.c.-mass spectrometry experiments further confirmed the identification of the main metabolites (figure 3). Figure 3 shows the mass spectra obtained for the compounds of the main metabolite peaks in the h.p.1.c. elution patterns. In these g.1.c.-mass spectrometry experiments mass spectra were obtained for N-(4’-amino-3’-fluorobenzyl)-2-fluoro-4-methylaniline (figure 3 A), 3-chloro-4aminobenzyl alcohol (figure 3 B), N-(4-amino-3’-chlorobenzyl)-2-chloro-4methylaniline (figure 3 C), 2,2’-dichloro-4,4’-dimethylhydrazobenzene (figure 3 D), 3-bromo-4-aminobenzyl alcohol (figure 3 E), 3-bromo-4-aminobenzaldehyde (figure 3 F) and N-(4’-amino-3’-bromobenzyl)-2-bromo-4-methylaniline (figure 3 G ) . T h e similarities of the mass spectra of the metabolites with those of the corresponding synthetic reference compounds (inserts in figures 3 A-G) confirmed the structures of the respective metabolites. For other metabolite peaks, mass spectra could not be obtained due to too low concentrations. For these metabolites, identification was accomplished by comparison of the retention times to the retention times of added synthetic reference compounds. Reference compounds for products of aromatic ring hydroxylation were not synthesized, as from the h.p.1.c. elution patterns it already emerged that they did not represent a major reaction pathway. This is concluded from the fact that these
S . Boeren et al.
1408
Table 1. Spectral characteristics of synthetic reference compounds. Information between brackets for the n.m.r. data refer to (1) the splitting pattern of the respective n.m.r. resonances; d =doublet, dd =double doublet, t = triplet, m = multiplet; the respective J-coupling constants are also presented, and to (2) the proton or carbon atoms the signal is ascribed to. Data between brackets for the i.r. data refer to the group the resonance is ascribed to. Values between brackets for the mass spectra present the relative intensity of the respective peaks in the mass spectrum, the highest peak is normalized to 1.00. Compound Spectral data
Appearance
Yield
light yellow needles 20% 'H-n.m.r. (cyclohexane: [d6]-acetnne=Y: 1 viv): G(ppm) =2.23 (CH,), 4.09 (NH,), 4.22 (d: J = 5.3 Hz) (CH,), 4.39 ( N H ) , 6.5-7.1 (m) (6 aromatic H's) '3C-n.m.r. (cyclohexane: [d6]-acetone=9 : 1 v/v): G(pprn)=24.6 (CH,), 52.2 (CH,), 116.9+ 121.2 (C6+C6'), I19.0+ 119.2 (C3 +C3'), 127.8+ 129.2 (C5 +C5'), 1305 134.0 (C4+C4'), 1392 (d: ./=13Hz) (Cl+(Cl'), 156.4 ( d : J=236Hz) (C2+C2') i.r. (CHCI,): v(cm-')=3460+3405 (NH), 2995+2918+2851 (CH), 1630 (NH), 1515, 1440, 1330, 1118, 929, 858 m s . : m / z (rel. int): 248 (019), 125 (0.39), 124 (1.00), 77 (0.11) Accurate mass: calculated: 248.1 125 found: 248.1 126
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
N-(4'-amino-3'-fluorobenzyl)-2-fluoro-4-methylanil~ne
+
N-(4'-amino-3'-chlorobenzyl)-2-chloro-4-methylaniline pale yellow solid 12% 'H-n.m.r. (cyclohexane : [d6]-acetonr= 9 : 1 viv): G(ppm) = 2.22 (CH,), 4.23 (d: J = 5.6 Hz) (CH,), 4.46 (NH,), 4.66 (t: J = 5 , 6 H z ) (NH), 6.5-7.2 (m) (6 aromatic H's) "C-n.m.r. (cyclohexane: [d6]-acetone=9: 1v/v): G(ppm)=23.5 (CH,), 51.3 (CH,), 115.4f119.5 ( C 6 + C 6 ) , 122.9 (C2+C2'), 130.2 (C4+C4'), 131.9+132.0 (C5+C5'), 132.9+133.0 (C3+C3'), 145.8+147.0 (C1 f C 1 ' ) i.r. (CHCI,): v(cm-')=3435+3398 (NH), 3003+2926+2860 (CH), 1614 (NH), 1498, 1320, 1151. 1040. 872 m.s.: m / z (rel. int): 282 (0.09), 280 (0.15), 143 (0.16), 142 (0.57), 141 (0.45), 140 (1.00), 113 (0.06), 104 (0.17), 77 (041) Accurate mass: calculated: 280.0534 found: 280.0531 pale yellow solid 9.3% 'H-n.m.r. (cyclohexane : [d6]-acetone= 9 : 1 vjv): G(ppm) =2.22 (CH,), 4.22 (CH,), 4 . 2 4 7 (NH, NH), 6.5-7.4 (m) (6 aromatic H's) '3C-n.m.r. (cyclohexane: [d6]-acetonr=9.1 v/v): G(ppm)=23.3 (CH,), 51.2 (CH,), 112.6+ 113.3 (C2 +C2'), 115.4+ 119.2 (C6+C6'), 130.6+133.2 (C4+C4'), 130.9+132.6 (C5 +CS), 135.1+136.5 (C3+C3'), 146.8+148.0 ( C l + C l ' ) i.r. (CHCI,): v(cm-')=3435+3400 (NH), 2998+2920+2851 (CH), 1615 (NH), 1500, 1311, 1154, 1033, 868 m.s.: m / z (rel. int): 372 (0.17), 370 (0.33), 368 (0.171, 187 (0.28), 186 (0.84), 185 (0.38), 184 (1.00), 105 (0.27). 104 (0-291, 78 (0.16), 77 (0.15) Accurate mass: calculated: 367-9524 found: 367.9527 RT-(4'-amino-3'-bromobenzyl)-2-bromo-J-methylaniline
+
2,2'-difluoro-4,4'-dimeihylazoxybenzene yellow crystals 20% 'H-n.m.r. (CDCI,): G(ppm) = 2.38 2.40 (each CH,), 7.02 (t: J=7.6 Hz) (H3 H3'+ H5 HS'), 7.78 (t: J = 7 . 8 H z ) (H6'), 8.33 (t: J=8.0Hz) (H6) "C-n.m.r. (CDCI,): G(ppm)=21.1 +21.3 (each CH,), 117.3+118.5 (each d: J = 2 0 H z ) (C3+C3'), 124.4+125.5+ 126.0+1261 (CS+CS'+C6+C6'), 131.1 f136.2 (each d: J = 8 H z ) (Cl +C1'), 143.0+145.0 (each d: J = 8 H z ) (C4+C4'), 154.9f157.1 (each d: J = 250 Hz) (C2+ C2') i.r. (CHCI,): v(cm-')=2923+2855(CH), 1612,1598,1461+1330(N-0), 1270,1122,950,863 m s . : m / z (rel. int): 262 (0.41), 242 (0.76), 137 (024), 123 (0.90), 122 (0.64), 109 (1.00). 96 (0.50), 83 (0.82) Accurate mass: calculated: 262.091 8 found: 262.0913
+
+
+
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1409
2,2’-dichloro-4,4‘-dimethylazoxybenzene light yellow crystals 14% ‘H-n.m.r. (CDCl,): G(ppm)=2.36+2.38 (each CH,), 7.16 (t: J=6.5Hz) (H5+H5’), 7.32 (H3+H3’), 7.62 (d: J = 8.0 Hz) (H6‘), 8.09 (d: J = 8.4) (H6) l3C-n.m.r. (d[6]acetone): G(ppm)=20.9+21.0 (each CH,), 123.9+125.8 (C6+C6’), 126.4+1302 (C2 +C2’), 128.7+129.4 (C5’+C5’), 131.3+131.8 (C3+C3’), 139.5+146.2 (C1 +C1’), 141.4+143.3 (C4+C4’) v(cm-’)=3000+2930 (CH), 1603, 1459f1330 (N-0), 1058, 1047, 873 i.r. (CHCI,): 296 (004), 294 (006), 261 (034), 259 (1.00), 196 (0.16), 153 (0.12), 125 (0.50),104 m.s. m/z (rel. int): (0.32), 89 (0.38), 77 (0.31) Accurate mass: calculated 294.0327 found: 294.0331 2,Z’-dibromo-4,4‘-dimethylazoxybenzene m.s.: m / z (rel. int): 386 (0.04), 384 (006), 382 (003), 305 (l.OO), 304 (0.16), 303 (0.95), 223 (0.12), 196 (036), 171 (0.37), 169 (0.36), 153 (O.ll), 104 (036), 90 (061), 89 (050), 77 (042) Accurate mass: calculated: 381.9316 found: 381.9316
2,Z’-di$uoro-4,4‘-dimethylazobenzene orange crystals 99% ‘H-n.m.r. (CDCI,): G(ppm)= 2.40 (CH,), 701 (t: J = 8.4 Hz) (H3 + H3’f H5 + HS’), 7.68 (t: J = 7 9 H z ) (H6+H6‘) l3C-n.m.r. (CDCI,): G(ppm)=21.3 (CH,), 117.1 (d: J = 2 0 H z ) (C3+C3’), 117.3 (C6+C6‘), 122.8 +129.5 (Cl+Cl’), 124.9+125.0(C5+C5‘), 138,5+143,9(eachd: J = 8 H z ) (C4 +C4‘), 160.0 (d: J=257Hz) (C2+C2’) i.r. (CHCl,): v(cm-’)=2930+2860 (CH), 1614, 1585, 1490, 1270, 1122, 939, 866 246 (048), 137 (0.45), 109 (1.00), 83 (040) m.s. m / z (rel. int): Accurate mass: calculated: 246.0968 found 246.0963 2,2’-dichloro-4,4‘-dimethylazobenzene orange crystals 98% ‘H-n.m.r. (CDCI,): G(ppm)= 2.40 (CH,), 7.13 (d: J =8.O Hz) (H5 HS’), 7.37 (H3 H3’), 7.70 (d: J=7.8 Hz) (H6+H6’) I3C-n.m.r. (CDCl,): S(ppm)=21.1 (CH,), 117.5 (C6+C6‘), 123.0 (C2+C2’), 128.0 (C5+C5’), 1307 (C3+C3’), 135.4 (C4+C4‘), 142.8 (C1 +Cl’) i.r. (CHCI,): v(cm-’)=2924 (CH), 1595, 1480, 1058, 889 m.s.: m / z (rel. int): 282 (0.04), 280 (020), 278 (0.30), 155 (0.16), 153 (0.48), 127 (035), 125 (1.00), 99 (0.20), 89 (0.58) Acccurate mass: calculated: 278.0375 found 278.0372
+
+
2,2’-dibromo-4,4‘-dimethylazobenzene m.s.: m / z (rel. int): 370 (0.43), 368 (0.83), 366 (0.43), 199 (069), 197 (069), 171 (0.92), 169 (091), 90 (1.00), 89 (0.81) Accurate mass calculated: 365.9367 found: 365.9365 2,2’-dichloro-4,4‘-dimethylhydrazobenzene white crystals (99%) ‘H-n.m.r. ([d6]acetone): G(ppm)=220 (CH,), 6.81 (d: J=8.4Hz) (H6+H6’), 6.84 (NH), 6.95 (dd, ,5=8.2Hz, 4J=1.9Hz) (H5+H5’), 712 (d:J=2.0Hz) (H3+H3’) ‘3C-n.m.r. (CDCl,): G(ppm)=202 (CH,), 112.9 (C6+C6’), 117.7 (C2+C2’), 1285 (C5+C5’), 129.7 (C3+C3‘), 129.8 (C4+C4‘), 141.6 (C1 +C1’) i.r. (CHCI,): v(cm-’)=3380 (NH), 2925 (CH), 1612, 1494, 1298, 1042, 875 m.s.: m / z (rel. int): 284 (0.04), 282 (0.19), 280 (0.31), 246 (036), 245 (0.32), 244 (1.00), 243 (0.21), 230 (008), 104 (009) Accurate mass: calculated: 280.0534 found: 280.0527
S . Boeren
1410
et
al.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
I
1
I
in
20
t i m e lminl
(4 Figure 1 . Reversed-phase h.p.1.c.chromatograms of rat liver microsomal incubations with 2-chloro-4methylaniline. ( a ) ,complete incubation mixture (insert: 10 times decreased sensitivity); ( b ) ,blank incubation without NADPH; and (c), blank incubation without 2-chloro-4-methylaniline. Asterisks mark the main unidentified metabolite peaks.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1411
30
2b
10
time lminl
(ti,
I
.
L 10
0
time lminl
(4 Figure 2. Schematic representation of h.p.1.c. elution patterns of rat liver microsomal metabolites from: (a), 2-fluoro-4-methylaniline; (b), 2-chloro-4-methylaniline; and ( c ) , 2-bromo-4-methylaniline. The position of the peaks on the x-axis represent the retention time of the peak maximum in the original chromatogram. Peak heights represent peak areas in arbitrary units (a.u.). The broad peak denoted by the letter S represents the substrate peak. Asterisks mark unidentified metabalite peaks.
1412
S. Boeren
et
al.
124
100
80
-
0 40
120
80
ZLO
200
160
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
60
dc
m U c
: a 40 20.
.
111 L I I,. I I " " I '
50
100
150
200
250
Mass/Charge
151
100
80
d
60
0 m
n a
a
40
20
1 80
121)
I
100
Mass/Charge
(4
120
110
140
160
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
loor
100
I‘
SO
80.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
-
60.
U a 0
m c -0
c
2 a
&O73
r
20280
Mass/Charge
(4
100
80
-
60
al c u U
2
a
40.
20 77
lo4 230
50
100
200
150 MassAharge
(d )
80
300
1413
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
1414
S . Boeren et al.
,
LO
10020
80-
60-
403
61
11
MassAharge
(4
50
100
150
Mass/Charge
(f)
200
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1415
181
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
104
1.’
I 160
150
200 250 MassICharge
300
350
(9) Figure 3.
Mass spectra of rat liver microsomal 2-halogenated 4-methylaniline metabolites
Mass spectra of synthetic reference compounds are shown as inserts. ( a ) , N-(4‘-amino-3’fluorobenzyl)-2-fluoro-4-methylaniline;( b ) , 3-chloro-4-aminobenzyl alcohol; (c), N-(4’-amino3’-chlorobenzyl)-2-chloro-4-methylaniline; ( d ) , 2,2’-dichloro-4,4’-dimethylhydrazobenzene; (e), 3-bromo-4-aminobenzyl alcohol; (f),3-bromo-4-aminobenzaldehyde; and (g), N-(4’-amino-3’bromobenzyl)-2-bromo-4-methylaniline.
metabolites, being more polar, should elute at shorter retention times than the reaction substrates. Furthermore, using the non-halogenated, 2-hydroxy-4methylaniline, 4-aminobenzylalcohol and 4-methylaniline, it was demonstrated that the product from aromatic ring hydroxylation of 4-methylaniline, i.e. 2-hydroxy-4methylaniline (Cheever et al. 1980), elutes after the benzyl alcohol and before the 4methylaniline parent compound (data not shown). From the results presented in figure 2 it follows that for the 2-halogenated-4-methylanilines no unidentified metabolite peak is observed in the region of the h.p.1.c. elution pattern between the peaks of the respective benzyl alcohols and the parent compounds. This also indicates that aromatic ring hydroxylation is not an important microsomal biotransformation pathway for the 2-halogenated 4-methylanilines.
QuantiJication of metabolite patterns T o quantify metabolite patterns, molar extinction coefficients of the various metabolites were determined. This was done using known amounts of the synthetic reference compounds dissolved in methanol/water mixtures that matched the percentage of methanol in water in which the respective compounds elute from the h.p.1.c. column. Results of these determinations are presented in table 2. T h e molar extinction coefficients of the N-hydroxylamines, 2-fluoro-4-methyl-nitrosobenzene, 2-chloro-4-methylnitrosobenzene and 2,2‘-dibromo-4,4’-dimethylhydrazobenzene were not determined. For the nitroso plus N-hydroxylamino derivatives quantification of the amounts formed in the microsomal incubations was carried out using a chemical assay for their detection and quantification. From the results obtained it followed that formation of these products from N-hydroxylation was below the detection limit of
1416
S . Boeren et al.
Table 2. Molar extinction coefficients of 2-halogenated 4-methylaniline metabolites at 295 nm. Molar extinction coefficients were determined using synthetic reference compounds dissolved in waterlmethanol mixtures that equalled the water/methanol mixture in which the respective compounds elute from the h.p.1.c.column. The percentageof methanol in water of these solvents is presented.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Compound
Methanol in water (%)
Molar extinction coefficient at 295nm (mM-lcm-’)
1 .oo
3-Fluoro-4-aminobenzyl alcohol 3-Chloro-4-aminobenzyl alcohol 3-Bromo-4-aminobenzyl alcohol
25 25 25
1.94 210
3-Fluoro-4-aminobenzaldehyde 3-Chloro-4-aminobenzaldehyde 3-Bromo-4-aminobenzaldehyde
40 40 40
7.99 10.0 9.98
N-(4’-amino-3’-fluorobenzyl)-2-fluoro-4-methylaniline N-(4’-amino-3’-chlorobenzyl)-2-chloro-4-methylaniline N-(4‘-amino-3’-bromobenzyl)-2-bromo-4-methylaniline
75 75 75
6.81 7.88 8 70
2,2’-Difluoro-4,4’-dimethylazoxybenzene 2,2’-~ichloro-4,4’-dimethylazoxybenzene
80 80
8.48 7.08
2,2’-Dichloro-4,4‘-dimethylhydrazobenzene
80
101
2,2’-Difluoro-4,4‘-dimethylazobenzene 2,2’-Dichloro-4,4’-dimethylazobenzene
80 80
6.55 5.96
the assay, which had been established to be 0 * 5 0 p ~ ,i.e. below 0.25 nmol/lOmin/nmol P450. This result is in accordance with the h.p.1.c. elution patterns (figures 1 and 2) in which no, or only very small, peaks were observed at the retention times of the synthetic hydroxylamine and nitroso derivatives. Using the peak areas in the h.p.1.c. elution patterns and the molar extinction coefficients presented in table 2, quantification of the various metabolites is possible. T h e results obtained are presented in table 3. From these data it follows that in the microsomal incubations the halogenated N-(4‘-aminobenzyl)-4-methylanilinesand the benzyl alcohols are the main metabolites formed, as together they make up about 80-90% of all metabolites formed from the three 2-halogenated-4-methylanilines. Furthermore, the results in table 3 demonstrate that, based on the formation of the benzyl alcohol and the benzaldehyde, the amount of C-hydroxylation increases going from 2-fluoro-4-methylaniline < 2-chloro-4-methylaniline < 2-bromo-4methylaniline. T h e total rate of conversion is lowest for 2-fluoro-4-methylaniIine, the other two compounds showing similar total conversion rates. Taking the formation of the hydroxylamine plus nitroso, and twice the amount of azoxy plus azo plus hydrazo derivatives to calculate the amount of N-hydroxylation, it follows that 2-chloro-4-methylaniline is N-hydroxylated to the highest extent (table 3). T h e azoxy, azo and hydrazo derivatives are counted twice, because they are believed to be formed from condensation of a nitroso and a hydroxylamino derivative (see Discussion section). Cytochrome P450 dependence of metabolite formation T o investigate the role of the microsomal cytochrone P450 system in the formation of the various metabolites, incubations were carried out in the presence of
4 1 k0.4 0.35 f 0.05
1992, VOL. 22,
NO.
12, 1403-1423
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
S. BOEREN?, B. TYRAKOWSKAtS, J. VERVOORTT, E. DE HOFFMAN$, K. TEUNIS", A. VAN VELDHUIZEN" and I . M. C . M. RIETJENSTY
t Department of Biochemistry, Agricultural University, Dreijenlaan 3, 6703 HA Wageningen, The Netherlands. 3 Universiti. Catholique de Louvain, Department of Mass Spectrometry, Place Louis Pasteur 1, B-1348, Louvain-la-Neuve, Belgium Department of Organic Chemistry, Agricultural University, Dreijenplein 8, 6703 BC Wageningen, T h e Netherlands Received 2 January 1992; accepted 1I July 1992
1. Rat liver microsomal metabolism of 2-fluoro-, 2-chloro- and 2-bromo-4methylaniline was investigated using h.p.1.c. Metabolites identified include products from side-chain C-hydroxylation (benzyl alcohols and benzaldehydes) and N-hydroxylation (hydroxylamines and nitroso derivatives). Aromatic ring hydroxylation was not a major reaction pathway. 2. A new type of microsomal metabolite was detected which was identified as a secondary amine, i.e. a halogenated N-(4'-aminobenzyl)-4-methylaniline. 3. In addition to these products azoxy, azo and hydrazo derivatives were formed. 4. Benzyl alcohols and halogenated N-(4'-aminobenzyl)-4-methylanilines were the major microsomal metabolites for all three 2-halogenated 4-methylanilines.
5. Quantification of the metabolite patterns demonstrated an influence of the type of halogen substituent on the rate of microsomal metabolism. T h e rate of side-chain Chydroxylation increases in the order 2-fluoro-4-methylaniline < 2-chloro-4-methylaniline < 2-bromo-4-methylaniline. 6. T h e rate of N-hydroxylation increases from 2-bromo-4-methylaniline < 2-fluoro-4That 2-chloro-4-methylaniline is Nmethylaniline < 2-chloro-4-methylaniline. hydroxylated to a larger extent is in accordance with its greater mutagenicity, twice that of 2-bromo-4-methylaniline.
Introduction Methylaniline (toluidine) and its derivatives are used in industrial manufacturing processes of dyes, pigments, agrochemicals and plastics. Exposure to methylaniline derivatives is not limited to production facilities and usage areas only, but may also result from industrial pollution and bacterial or chemical breakdown of the derivatives (Leslie et al. 1988, Knowles and Sen Gupta 1970, Zimmer et al. 1980, Yoshimi et al. 1988). T h e toxicity and especially mutagenicity of various methylanilines has been well documented (Yoshimi et al. 1988, Zimmer et al. 1980, Weisburger et al. 1978). Although it is known that the methylanilines have to be metabolized to their ultimate toxic and mutagenic form, products from N-hydroxylation especially contributing to this toxic effect (Zimmer et al. 1980, Hecht et al. 1979, Weisburger and Weisburger 1973), few studies report on their metabolism.
5 Present address: Institute of Commodity Science, Academy of Economics, Al. Niepodleglosci 10, 60-967 Poznan, Poland. 7 To whom correspondence should be addressed. 0049-8254/92 $3.00 0 1992 Taylor & Francis Ltd.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
1404
S. Boeren et al.
In an in vivo study using rats, Son et al. (1980) showed that the main urine metabolites formed from 2-methylaniline were conjugated derivatives of ring hydroxylated products, and metabolites derived from side-chain C- and N-hydroxylation. Aromatic ring hydroxylation was also reported to occur in in vivo studies with 2-, 3- and 4-methylaniline (Cheever et al. 1980) and 4-trifluoromethylaniline (Wade et al. 1988). From in vitro liver microsomal metabolism data, Daly et al. (1 968) suggested that 4-methylaniline was converted to 4-aminobenzyl alcohol and 4-aminobenzaldehyde as a result of side-chain C-hydroxylation; aromatic ringand N-hydroxylation were not reported. Furthermore, Kiese (1963) provided evidence for the in vivo N-hydroxylation of 2-, 3- and 4-methylaniline in dogs, by demonstrating the formation of the corresponding nitrosotoluenes and/or methaemoglobin formation in the blood of dogs exposed to these methylanilines. For methylanilines halogenated in the aromatic ring, the only metabolism studies have been with 4-chloro-2-methylaniline, the metabolic breakdown product of the insecticide/acaracide chlordimeform (N'-(4-chloro-o-tolyl)-N,N-dimethylformamidine) (Knowles and Sen Gupta 1970), which undergoes side-chain Chydroxylation and further oxidation to give 5-chloro-anthranilic acid; no other metabolic products were identified. In spite of this lack of knowledge on the metabolism of halogenated methylanilines, their mutagenicity is well documented and known to vary with the substituent pattern (Zimmer et al. 1980, Yoshimi et al. 1988). Zimmer et al. (1980) for instance, reported that the mutagenicity towards Salmonella typhimurium TA100 of 2-chloro-4-methylaniline in the presence of S9 mix was twice that of 2-bromo-4methylaniline. Although mutagenicity of halogenated methylanilines is known to result from metabolism, no knowledge exists concerning the effect of a halogen substituent in the aromatic ring on the metabolism of methylanilines. Therefore, we decided to study the biotransformation of a series of three related halogenated methylanilines, namely, 2-fluoro-, 2-chloro- and 2-bromo-4-methylaniline. This paper describes the identification and quantification of the major metabolites formed from 2-fluoro-, 2-chloro- and 2-bromo-4-methylaniline by rat liver microsomes. T h e results demonstrate a qualitative and quantitative influence of the type of halogen substituent on the metabolites obtained.
Experimental Chemicals 2-Fluoro-4-methylaniline and 2-fluoro-4-methyl-nitrobenzene were purchased from Fluorochem (Derbyshire, UK). 4-Methylaniline, 2-chloro-4-methylaniline, 2-bromo-4-methylaniline, 4-nitrobenzyl alcohol and 2-hydroxy-4-methylaniline were purchased from Janssen Chimica (Beerse, Belgium). 3-Fluoro-4-aminobenzaldehyde,3-chloro-4-aminobenzaldehyde and 3-bromo-4-aminobenzaldehyde were synthesized from the corresponding 2-halogenated 4-methylanilines using 2,3-dichloro-5,6dicyano-l,4-benzoquinone(Janssen), according to the method described by La1 et al. (1984) for synthesis of 3,5-dimethyl-4-aminobenzaldehydefrom 2,4,6-trimethylaniline. 3-Fluoro-4-aminobenzyl alcohol, 3-chloro-4-aminobenzyl alcohol and 3-bromo-4-aminobenzyl alcohol were obtained by reduction of the corresponding benzaldehyde with sodium borohydride following the procedure described by Schenker (1961). 4-Aminobenzyl alcohol was synthesized by the reduction of 4-nitrobenzyl alcohol, dissolved in ethyl acetate, under 4atm H, in the presence of O.l%(w/v) palladium/carbon (Vogel 1989). 2-Fluoro-4-methylnitrosobenzene, 2-chloro-4-methylnitrosobenzene and 2-bromo-4methylnitrosobenzene were synthesized by oxidation of the corresponding methylaniline with potassium peroxymonosulphate in a water-acetic acid mixture using the procedure described by Kennedy and Stock (1960) for the synthesis of 1,4-dinitrosobenzene.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1405
2-Fluoro-4-methylhydroxylaminobenzeneand 2-chloro-4-methylhydroxylaminobenzenewere obtained by palladium-catalysed reduction of 2-fluoro-4-methylnitrobenzene and 2-chloro-4methylnitrobenzene carried out as described by Stavenuiter et al. (1985) for the synthesis of N-hydroxy5-phenyl-2-pyridinamine from 2-nitro-5-phenylpyridine using hydrazine hydrate in tetrahydrofuran. 2-Bromo-4-methylhydroxylaminobenzene was synthesized by reduction of 2-bromo-4methylnitrosobenzene in water with 10 times excess of NADH. Synthesis of N-(4'-amino-3'-fluorobenzyl)-2-fluoro-4-methylaniline, N-(4'-amino-3'-chlorobenzyl)2-chloro-4-methylaniline and N-(4'-amino-3'-bromobenzyl)-2-bromo-4-methylaniline was achieved by dropwise adding 1.85 g 2,3-dichloro-5,6-dicyano-l,4-benzoquinone(8.5 mmol), dissolved in 20 ml 1,4dioxane, to a solution of 1.Og 2-fluoro-4-methylaniline or 1.2g 2-chloro-4-methylaniline or 1.5 g 2bromo-4-methylaniline (8.1 mmol) in 5 ml 1,4-dioxane. After stirring at room temperature for 10 min, the mixture was poured into a solution of 3.0 g NaBH, (79 mmol) in lOOml water. T h e solution was stirred for lomin, 200ml 1 M KH,PO, were added and stirring was continued for another 30min. T h e compound was extracted into ethyl acetate, folIowed by washing three times with water, three times with saturated aq. NaCl and filtration over a Whatman 1 ps filter. 2,2'-Difluoro-4,4'-dimethylazoxybenzene and 2,2'-dichloro-4,4'-dimethylazoxybenzenewere synthesized by oxidation of the respective 4-methylanilines with potassium peroxymonosulphate in a 1 : 1 water-acetic acid mixture using a modification of the method described above for the synthesis of 2fluoro- and 2-chloro-4-methylnitrosobenzene. T h e azoxybenzenes were prepared by slowly adding 1.50g potassium peroxymonosulphate (2.4 mmol) dissolved in 10 ml water to a solution of 0.50 g 2-fluoro-4methylaniline or 0 5 7 g 2-chloro-4-methylaniline (4.0 mmol) in acetic acid, stirring at room temperature for 30 min. Crystals formed upon the addition of 50 ml water were collected by filtration and washed with water. 2,2'-Difluoro-4,4'-dimethylazobenzeneand 2,2'-dichloro-4,4-dimethylazobenzenewere synthesized by reduction of 2,2'-difluoro- and 2,2'-dichloro-4,4'-dimethylazoxybenzenerespectively with lithium aluminium hydride, carried out essentially as described by Schenker (1961). A mixture of 2,2'-dibromo-4,4'-dirnethylazoxybenzene and 2,2'-dibromo-4,4'-dimethylazobenzene was obtained using a modification of the procedure described by Holmes and Bayer (1960) for the synthesis of 2,6-dibromonitrosobenzene from 2,6-dibromoaniline. In short, 1.52 g (8 mmol) 2-bromo-4methylaniline in 7 ml acetic acid and 120pl H,SO, were oxidized by addition of 2 ml of 30% hydrogen peroxide and stirring at 4&50"C for 16 h. Upon addition of 5 ml water the compounds were precipitated by centrifugation. T h e pellet was dissolved in ethanol and the mixture of the two compounds was recrystallized upon addition of water and standing at 4°C for 3 days. 2,2'-Dichloro-4,4'-dimethylhydrazobenzene was synthesized from 2,2'-dichloro-4,4'-dimethylazoxybenzene using hydrazine/Raney nickel according to the method described by Furst and Moore (1957). 2,2'-Dibromo-4,4'-dimethylhydrazobenzene was synthesized the same way as 2,2'-dichloro-4,4'dimethylhydrazobenzene. Complete purification of this bromo derivative was not performed because it was hampered by its instability under aerobic conditions resulting in formation of 2,2'-dibromo-4,4'dimethylazobenzene. Generally, compounds were further purified by preparative chromatography on silicagel 60 (Merck, Darmstadt, Germany), using a mixture of hexane plus a suitable modifier (for example dichloromethane or acetonitrile) as the eluent. Structures were confirmed by 'H-n.m.r. mass spectrometry, i.r. and I3Cn.m.r. Details of spectral characteristics of the compounds synthesized by newly described or modified procedures are presented in the results section (table 1). Spectral analysis of synthetic reference compounds 'H-n.m.r. and 13C-n.m.r. spectra were measured on a Bruker AC 200 spectrometer. Non-deuterated impurities in the deuterated solvents were used as internal standard; for 'H-n.m.r. these were CHCl, at 7.24ppm and CH,COCH, at 2.04ppm, for I3C-n.m.r. these were CDCI, at 77.0ppmand CD,COCD, at 29.8 and 2060 ppm. Mass spectra were recorded using a Hewlett Packard 5970B g.1.c.-mass spectrometry system with a 30m capillary DB17 column. Accurate masses were determined on a AEI-902 mass spectrometer equipped with a VG-2AB console, and i.r. spectra were recorded on a Philips PU 9706 spectrometer. Preparation of microsomes Liver microsomes were prepared as described before using untreated male Wistar rats (300-350 g) (Rietjens and Vervoort 1989). Microsomal incubations Microsomal incubations were carried out in 0.1 M (final concentration) potassium phosphate p H 7.6 containing 1 mM 4-methylaniline substrate (added as a 0.1 M stock solution in dimethyl sulphoxide), 2 mM NADPH and 2 p~ microsomal cytochrome P450. Reactions were started by the addition of NADPH. After 10min at 37°C reactions were terminated by freezing the reaction mixture in liquid nitrogen for h.p.1.c. analysis, or by mixing with trichloroacetic acid (2.0 ml incubation mixture +0.6ml 200g/l
1406
S . Boeren et al.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
trichloroacetic acid) for chemical determination of N-hydroxylated products. Blank experiments were performed by adding water instead of NADPH, or dimethyl sulphoxide instead of the 4-methylaniline solution. T o study the involvement of the cytochrome P450 system in the formation of the various metabolites, to some incubations NADH (2mM final concentration) instead of NADPH, or the cytochrome P450 inhibitor octylamine (Cashman and Hanzlik 1981) (7 mM final concentration) were added. Other incubations were performed in the presence of CO, or made anaerobic by three cycles of evacuation and filling with argon and the addition of glucose (1OmM final concentration) plus glucose oxidase (Boehringer, Mannheim, Germany) (0.1 mg/ml final concentration). T o prevent any reaction in these anaerobic samples as a result of aerobic sample preparation for h.p.1.c. injection, the cytochrome P450 inhibitor octylamine was added to a final concentration of 7 mM when terminating the anaerobic reaction.
Preparation of h.p.1.c. samples Microsomal incubation samples, frozen in liquid nitrogen to terminate the reaction, were thawed and centrifuged at 4°C at 13000g for 1Omin. For standard h.p.1.c. analysis Sop1 of the supernatant was injected directly into the h.p.1.c. (see below). For g.1.c.-mass spectrometry of metabolite peaks, 20 ml of incubation mixture were freeze-dried to approximately 1 ml. These concentrated samples were centrifuged for 10min at 13 OOOg and 80Opl of the supernatants thus obtained were injected into the h.p.1.c. using a pre-concentration column as described below. Analysis by h.p.1.c. A Kratos 400 high pressure gradient system, a 100 x 3.0 mm CP-LiChrosorb RP C8 column and a Kratos 757 UVjVIS detector (8 mm light path) were used. T h e detection wavelength was 295 nm. T h e gradient used consisted of eluting with 100%water for 1 rnin, followed b y a linear decrease to 202, (v/v) water in methanol in 26 min and keeping the percentage of water in methanol at 20% (v/v) for 6 min. Standard injections were carried out using an external Sop1 loop. For injections for the g.1.c.-mass spectrometry experiments the Sop1 injection loop was replaced by a 10 x 3.0mm ChromSpher C18 preconcentration column and 800 p1 of the concentrated sample were injected. Analysis by g.1.c.-mass spectrometry Peak fractions collected from the h.p.1.c. run were freeze-dried and the residues were dissolved in 20p1 dichloromethane. Of these samples 1 pI was injected into the g.1.c.-mass spectrometry system. T h e Varian 3700 gas chromatograph of the system was equipped with a 30 m x 0.32 mm chemically bonded RSL-200 column. Helium was used as the carrier gas at a flow of 1 ml/min. During the run a linear temperature gradient from 50 to 200°C was applied in 21 min. The column was directly coupled to a Finnigan-Mat TSQ70 mass spectrometer operating in the electron impact mode at 70eV. Determination of molar extinction coeficients A known amount of compound (10--50mg)was dissolved in 25 ml of a suitable solvent, e.g. 50% (v/v) methanol in water or 20% (v/v) methanol in ethanol. Small amounts of these stock solutions were diluted in the water-methanol mixtures at which the respective metabolites elute from the h.p.1.c. column. The absorbance of these samples at 29.5 nm was determined using a LKB Ultrospec I11 spectrometer. Chemical assay for N-oxidation products N-oxidation products (hydroxylamine plus nitroso derivatives) were determined essentially as described by Herr and Kiese (1959). In short, 1.0ml water and 0.30 ml 10%(w/v) potassium ferricyanide in 6 M HCI were added to 2.0 ml trichloroacetic acid-precipitated microsomal supernatant to oxidize the hydroxylamine to the corresponding nitroso derivative. These samples were extracted with 2 ml CCl, and the tetra phases were washed twice with 2 . 5 H,SO,. ~ T o 1.0ml of the washed tetra phases, 1.0ml of glacial acetic acid and 50pl 20%(w/v) sodium nitrite were added. After lSmin, 1 0 0 ~ 15O%(w/v) ammonium sulphamate were added and the samples were mixed for 1Omin. Finally, 2 5 0 ~ 180%(v/v) acetic acid and 50 pl N-(1 -naphthyl)rthylenediamine dihydrochloride (Aldrich, Steinheim, Germany) were added. After 2 h in the dark the absorbance of the upper water layer at 555 nm was determined. Using synthetic 2-fluoro-, 2-chloro- and 2-bromo-4-methylnitrosobenzene, molar extinction coefficients for this assay at 555nm of respectively 13.5, 14.7 and 16,6mM-'cm-' were obtained. Determination of apparent dissociation constants (Kd,,,) T o determine apparent dissociation constants (Kd,,,) for the binding of the 2-halogenated 4methylanilines to microsomal cytochromes P450, 1.0 ml of a microsomal preparation containing 1.0 p~ cytochrome P450 in 0.1 M potassium phosphate p H 7 6 was titrated with 2-fluoro-4-methylaniline, 2chloro-4-methylaniline or 2-bromo-4-methylaniline using small volumes of a 50 mM methylaniline stock . the solution in dimethyl sulphoxide up to a maximal final methylaniline concentration of 2 . 0 m ~During stepwise addition of the 2-halogenated 4-methylanilines, difference spectra were recorded against reference cuvettes containing a similar microsomal dilution to which identical volumes of dimethyl
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1407
sulphoxide were added. Difference spectra were recorded from 300 to 500 nm using an Aminco DW-2A spectrophotometer. 2-Fluoro-4-methylaniline, 2-chloro-4-methylaniline and 2-bromo-4-methylaniline all gave rise to a so-called type I1 spectral change with a maximum at 430 nm and a minimum at 400410nm. The absorbance differences (AA) between this maximum and minimum in the spectra were determined. The Kdappwas calculated from a linear double reciprocal plot of AA-’ versus the 4methylaniline concentration- I .
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Results Spectral characteristics of newly described reference compounds Table 1 presents the ‘H-n.m.r., 13C-n.m.r., i.r., and mass analysis of the synthetic reference compounds prepared by newly described or modified procedures. The table also presents yields of these syntheses as well as characteristics of the compounds after purification. Benzaldehydes, benzyl alcohols, hydroxylaminobenzenes and nitroso derivatives were prepared by well described procedures. Their structures were also confirmed by spectral and mass analysis (data not shown). Microsomal metabolite patterns of the 2-halogenated-4-methylanilines Figure 1 presents the reversed-phase h.p.1.c. elution pattern of a microsomal incubation with 2-chloro-4-methylaniline (figure 1 A) and the corresponding blanks carried out in the absence of NADPH (figure 1 B) or in the absence of 2-chloro-4methylaniline (figure 1 C). Comparison of the chromatograms in figure 1 reveals that at least 10 metabolites are formed from 2-chloro-4-methylaniline. Additional experiments demonstrated product formation to be linear with the amount of microsomes added up to at least 2 p microsomal ~ cytochrome P450 and linear in time for at least 10 min (data now shown). Figure 2 shows a schematic presentation of the h.p.1.c. elution patterns of rat liver microsomal incubations with 2-fluoro-4methylaniline, 2-chloro-4-methylaniline and 2-bromo-4-methylaniline. I n this presentation, blank peaks were omitted. Identification of the metabolites presented in figures 1 and 2 was performed on the basis of h.p.1.c. retention times of the synthetic reference compounds. I n addition, g.1.c.-mass spectrometry experiments further confirmed the identification of the main metabolites (figure 3). Figure 3 shows the mass spectra obtained for the compounds of the main metabolite peaks in the h.p.1.c. elution patterns. In these g.1.c.-mass spectrometry experiments mass spectra were obtained for N-(4’-amino-3’-fluorobenzyl)-2-fluoro-4-methylaniline (figure 3 A), 3-chloro-4aminobenzyl alcohol (figure 3 B), N-(4-amino-3’-chlorobenzyl)-2-chloro-4methylaniline (figure 3 C), 2,2’-dichloro-4,4’-dimethylhydrazobenzene (figure 3 D), 3-bromo-4-aminobenzyl alcohol (figure 3 E), 3-bromo-4-aminobenzaldehyde (figure 3 F) and N-(4’-amino-3’-bromobenzyl)-2-bromo-4-methylaniline (figure 3 G ) . T h e similarities of the mass spectra of the metabolites with those of the corresponding synthetic reference compounds (inserts in figures 3 A-G) confirmed the structures of the respective metabolites. For other metabolite peaks, mass spectra could not be obtained due to too low concentrations. For these metabolites, identification was accomplished by comparison of the retention times to the retention times of added synthetic reference compounds. Reference compounds for products of aromatic ring hydroxylation were not synthesized, as from the h.p.1.c. elution patterns it already emerged that they did not represent a major reaction pathway. This is concluded from the fact that these
S . Boeren et al.
1408
Table 1. Spectral characteristics of synthetic reference compounds. Information between brackets for the n.m.r. data refer to (1) the splitting pattern of the respective n.m.r. resonances; d =doublet, dd =double doublet, t = triplet, m = multiplet; the respective J-coupling constants are also presented, and to (2) the proton or carbon atoms the signal is ascribed to. Data between brackets for the i.r. data refer to the group the resonance is ascribed to. Values between brackets for the mass spectra present the relative intensity of the respective peaks in the mass spectrum, the highest peak is normalized to 1.00. Compound Spectral data
Appearance
Yield
light yellow needles 20% 'H-n.m.r. (cyclohexane: [d6]-acetnne=Y: 1 viv): G(ppm) =2.23 (CH,), 4.09 (NH,), 4.22 (d: J = 5.3 Hz) (CH,), 4.39 ( N H ) , 6.5-7.1 (m) (6 aromatic H's) '3C-n.m.r. (cyclohexane: [d6]-acetone=9 : 1 v/v): G(pprn)=24.6 (CH,), 52.2 (CH,), 116.9+ 121.2 (C6+C6'), I19.0+ 119.2 (C3 +C3'), 127.8+ 129.2 (C5 +C5'), 1305 134.0 (C4+C4'), 1392 (d: ./=13Hz) (Cl+(Cl'), 156.4 ( d : J=236Hz) (C2+C2') i.r. (CHCI,): v(cm-')=3460+3405 (NH), 2995+2918+2851 (CH), 1630 (NH), 1515, 1440, 1330, 1118, 929, 858 m s . : m / z (rel. int): 248 (019), 125 (0.39), 124 (1.00), 77 (0.11) Accurate mass: calculated: 248.1 125 found: 248.1 126
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
N-(4'-amino-3'-fluorobenzyl)-2-fluoro-4-methylanil~ne
+
N-(4'-amino-3'-chlorobenzyl)-2-chloro-4-methylaniline pale yellow solid 12% 'H-n.m.r. (cyclohexane : [d6]-acetonr= 9 : 1 viv): G(ppm) = 2.22 (CH,), 4.23 (d: J = 5.6 Hz) (CH,), 4.46 (NH,), 4.66 (t: J = 5 , 6 H z ) (NH), 6.5-7.2 (m) (6 aromatic H's) "C-n.m.r. (cyclohexane: [d6]-acetone=9: 1v/v): G(ppm)=23.5 (CH,), 51.3 (CH,), 115.4f119.5 ( C 6 + C 6 ) , 122.9 (C2+C2'), 130.2 (C4+C4'), 131.9+132.0 (C5+C5'), 132.9+133.0 (C3+C3'), 145.8+147.0 (C1 f C 1 ' ) i.r. (CHCI,): v(cm-')=3435+3398 (NH), 3003+2926+2860 (CH), 1614 (NH), 1498, 1320, 1151. 1040. 872 m.s.: m / z (rel. int): 282 (0.09), 280 (0.15), 143 (0.16), 142 (0.57), 141 (0.45), 140 (1.00), 113 (0.06), 104 (0.17), 77 (041) Accurate mass: calculated: 280.0534 found: 280.0531 pale yellow solid 9.3% 'H-n.m.r. (cyclohexane : [d6]-acetone= 9 : 1 vjv): G(ppm) =2.22 (CH,), 4.22 (CH,), 4 . 2 4 7 (NH, NH), 6.5-7.4 (m) (6 aromatic H's) '3C-n.m.r. (cyclohexane: [d6]-acetonr=9.1 v/v): G(ppm)=23.3 (CH,), 51.2 (CH,), 112.6+ 113.3 (C2 +C2'), 115.4+ 119.2 (C6+C6'), 130.6+133.2 (C4+C4'), 130.9+132.6 (C5 +CS), 135.1+136.5 (C3+C3'), 146.8+148.0 ( C l + C l ' ) i.r. (CHCI,): v(cm-')=3435+3400 (NH), 2998+2920+2851 (CH), 1615 (NH), 1500, 1311, 1154, 1033, 868 m.s.: m / z (rel. int): 372 (0.17), 370 (0.33), 368 (0.171, 187 (0.28), 186 (0.84), 185 (0.38), 184 (1.00), 105 (0.27). 104 (0-291, 78 (0.16), 77 (0.15) Accurate mass: calculated: 367-9524 found: 367.9527 RT-(4'-amino-3'-bromobenzyl)-2-bromo-J-methylaniline
+
2,2'-difluoro-4,4'-dimeihylazoxybenzene yellow crystals 20% 'H-n.m.r. (CDCI,): G(ppm) = 2.38 2.40 (each CH,), 7.02 (t: J=7.6 Hz) (H3 H3'+ H5 HS'), 7.78 (t: J = 7 . 8 H z ) (H6'), 8.33 (t: J=8.0Hz) (H6) "C-n.m.r. (CDCI,): G(ppm)=21.1 +21.3 (each CH,), 117.3+118.5 (each d: J = 2 0 H z ) (C3+C3'), 124.4+125.5+ 126.0+1261 (CS+CS'+C6+C6'), 131.1 f136.2 (each d: J = 8 H z ) (Cl +C1'), 143.0+145.0 (each d: J = 8 H z ) (C4+C4'), 154.9f157.1 (each d: J = 250 Hz) (C2+ C2') i.r. (CHCI,): v(cm-')=2923+2855(CH), 1612,1598,1461+1330(N-0), 1270,1122,950,863 m s . : m / z (rel. int): 262 (0.41), 242 (0.76), 137 (024), 123 (0.90), 122 (0.64), 109 (1.00). 96 (0.50), 83 (0.82) Accurate mass: calculated: 262.091 8 found: 262.0913
+
+
+
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1409
2,2’-dichloro-4,4‘-dimethylazoxybenzene light yellow crystals 14% ‘H-n.m.r. (CDCl,): G(ppm)=2.36+2.38 (each CH,), 7.16 (t: J=6.5Hz) (H5+H5’), 7.32 (H3+H3’), 7.62 (d: J = 8.0 Hz) (H6‘), 8.09 (d: J = 8.4) (H6) l3C-n.m.r. (d[6]acetone): G(ppm)=20.9+21.0 (each CH,), 123.9+125.8 (C6+C6’), 126.4+1302 (C2 +C2’), 128.7+129.4 (C5’+C5’), 131.3+131.8 (C3+C3’), 139.5+146.2 (C1 +C1’), 141.4+143.3 (C4+C4’) v(cm-’)=3000+2930 (CH), 1603, 1459f1330 (N-0), 1058, 1047, 873 i.r. (CHCI,): 296 (004), 294 (006), 261 (034), 259 (1.00), 196 (0.16), 153 (0.12), 125 (0.50),104 m.s. m/z (rel. int): (0.32), 89 (0.38), 77 (0.31) Accurate mass: calculated 294.0327 found: 294.0331 2,Z’-dibromo-4,4‘-dimethylazoxybenzene m.s.: m / z (rel. int): 386 (0.04), 384 (006), 382 (003), 305 (l.OO), 304 (0.16), 303 (0.95), 223 (0.12), 196 (036), 171 (0.37), 169 (0.36), 153 (O.ll), 104 (036), 90 (061), 89 (050), 77 (042) Accurate mass: calculated: 381.9316 found: 381.9316
2,Z’-di$uoro-4,4‘-dimethylazobenzene orange crystals 99% ‘H-n.m.r. (CDCI,): G(ppm)= 2.40 (CH,), 701 (t: J = 8.4 Hz) (H3 + H3’f H5 + HS’), 7.68 (t: J = 7 9 H z ) (H6+H6‘) l3C-n.m.r. (CDCI,): G(ppm)=21.3 (CH,), 117.1 (d: J = 2 0 H z ) (C3+C3’), 117.3 (C6+C6‘), 122.8 +129.5 (Cl+Cl’), 124.9+125.0(C5+C5‘), 138,5+143,9(eachd: J = 8 H z ) (C4 +C4‘), 160.0 (d: J=257Hz) (C2+C2’) i.r. (CHCl,): v(cm-’)=2930+2860 (CH), 1614, 1585, 1490, 1270, 1122, 939, 866 246 (048), 137 (0.45), 109 (1.00), 83 (040) m.s. m / z (rel. int): Accurate mass: calculated: 246.0968 found 246.0963 2,2’-dichloro-4,4‘-dimethylazobenzene orange crystals 98% ‘H-n.m.r. (CDCI,): G(ppm)= 2.40 (CH,), 7.13 (d: J =8.O Hz) (H5 HS’), 7.37 (H3 H3’), 7.70 (d: J=7.8 Hz) (H6+H6’) I3C-n.m.r. (CDCl,): S(ppm)=21.1 (CH,), 117.5 (C6+C6‘), 123.0 (C2+C2’), 128.0 (C5+C5’), 1307 (C3+C3’), 135.4 (C4+C4‘), 142.8 (C1 +Cl’) i.r. (CHCI,): v(cm-’)=2924 (CH), 1595, 1480, 1058, 889 m.s.: m / z (rel. int): 282 (0.04), 280 (020), 278 (0.30), 155 (0.16), 153 (0.48), 127 (035), 125 (1.00), 99 (0.20), 89 (0.58) Acccurate mass: calculated: 278.0375 found 278.0372
+
+
2,2’-dibromo-4,4‘-dimethylazobenzene m.s.: m / z (rel. int): 370 (0.43), 368 (0.83), 366 (0.43), 199 (069), 197 (069), 171 (0.92), 169 (091), 90 (1.00), 89 (0.81) Accurate mass calculated: 365.9367 found: 365.9365 2,2’-dichloro-4,4‘-dimethylhydrazobenzene white crystals (99%) ‘H-n.m.r. ([d6]acetone): G(ppm)=220 (CH,), 6.81 (d: J=8.4Hz) (H6+H6’), 6.84 (NH), 6.95 (dd, ,5=8.2Hz, 4J=1.9Hz) (H5+H5’), 712 (d:J=2.0Hz) (H3+H3’) ‘3C-n.m.r. (CDCl,): G(ppm)=202 (CH,), 112.9 (C6+C6’), 117.7 (C2+C2’), 1285 (C5+C5’), 129.7 (C3+C3‘), 129.8 (C4+C4‘), 141.6 (C1 +C1’) i.r. (CHCI,): v(cm-’)=3380 (NH), 2925 (CH), 1612, 1494, 1298, 1042, 875 m.s.: m / z (rel. int): 284 (0.04), 282 (0.19), 280 (0.31), 246 (036), 245 (0.32), 244 (1.00), 243 (0.21), 230 (008), 104 (009) Accurate mass: calculated: 280.0534 found: 280.0527
S . Boeren
1410
et
al.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
I
1
I
in
20
t i m e lminl
(4 Figure 1 . Reversed-phase h.p.1.c.chromatograms of rat liver microsomal incubations with 2-chloro-4methylaniline. ( a ) ,complete incubation mixture (insert: 10 times decreased sensitivity); ( b ) ,blank incubation without NADPH; and (c), blank incubation without 2-chloro-4-methylaniline. Asterisks mark the main unidentified metabolite peaks.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1411
30
2b
10
time lminl
(ti,
I
.
L 10
0
time lminl
(4 Figure 2. Schematic representation of h.p.1.c. elution patterns of rat liver microsomal metabolites from: (a), 2-fluoro-4-methylaniline; (b), 2-chloro-4-methylaniline; and ( c ) , 2-bromo-4-methylaniline. The position of the peaks on the x-axis represent the retention time of the peak maximum in the original chromatogram. Peak heights represent peak areas in arbitrary units (a.u.). The broad peak denoted by the letter S represents the substrate peak. Asterisks mark unidentified metabalite peaks.
1412
S. Boeren
et
al.
124
100
80
-
0 40
120
80
ZLO
200
160
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
60
dc
m U c
: a 40 20.
.
111 L I I,. I I " " I '
50
100
150
200
250
Mass/Charge
151
100
80
d
60
0 m
n a
a
40
20
1 80
121)
I
100
Mass/Charge
(4
120
110
140
160
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
loor
100
I‘
SO
80.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
-
60.
U a 0
m c -0
c
2 a
&O73
r
20280
Mass/Charge
(4
100
80
-
60
al c u U
2
a
40.
20 77
lo4 230
50
100
200
150 MassAharge
(d )
80
300
1413
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
1414
S . Boeren et al.
,
LO
10020
80-
60-
403
61
11
MassAharge
(4
50
100
150
Mass/Charge
(f)
200
Rat liver microsomal metabolism of 2-halogenated 4-methylanilines
1415
181
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
104
1.’
I 160
150
200 250 MassICharge
300
350
(9) Figure 3.
Mass spectra of rat liver microsomal 2-halogenated 4-methylaniline metabolites
Mass spectra of synthetic reference compounds are shown as inserts. ( a ) , N-(4‘-amino-3’fluorobenzyl)-2-fluoro-4-methylaniline;( b ) , 3-chloro-4-aminobenzyl alcohol; (c), N-(4’-amino3’-chlorobenzyl)-2-chloro-4-methylaniline; ( d ) , 2,2’-dichloro-4,4’-dimethylhydrazobenzene; (e), 3-bromo-4-aminobenzyl alcohol; (f),3-bromo-4-aminobenzaldehyde; and (g), N-(4’-amino-3’bromobenzyl)-2-bromo-4-methylaniline.
metabolites, being more polar, should elute at shorter retention times than the reaction substrates. Furthermore, using the non-halogenated, 2-hydroxy-4methylaniline, 4-aminobenzylalcohol and 4-methylaniline, it was demonstrated that the product from aromatic ring hydroxylation of 4-methylaniline, i.e. 2-hydroxy-4methylaniline (Cheever et al. 1980), elutes after the benzyl alcohol and before the 4methylaniline parent compound (data not shown). From the results presented in figure 2 it follows that for the 2-halogenated-4-methylanilines no unidentified metabolite peak is observed in the region of the h.p.1.c. elution pattern between the peaks of the respective benzyl alcohols and the parent compounds. This also indicates that aromatic ring hydroxylation is not an important microsomal biotransformation pathway for the 2-halogenated 4-methylanilines.
QuantiJication of metabolite patterns T o quantify metabolite patterns, molar extinction coefficients of the various metabolites were determined. This was done using known amounts of the synthetic reference compounds dissolved in methanol/water mixtures that matched the percentage of methanol in water in which the respective compounds elute from the h.p.1.c. column. Results of these determinations are presented in table 2. T h e molar extinction coefficients of the N-hydroxylamines, 2-fluoro-4-methyl-nitrosobenzene, 2-chloro-4-methylnitrosobenzene and 2,2‘-dibromo-4,4’-dimethylhydrazobenzene were not determined. For the nitroso plus N-hydroxylamino derivatives quantification of the amounts formed in the microsomal incubations was carried out using a chemical assay for their detection and quantification. From the results obtained it followed that formation of these products from N-hydroxylation was below the detection limit of
1416
S . Boeren et al.
Table 2. Molar extinction coefficients of 2-halogenated 4-methylaniline metabolites at 295 nm. Molar extinction coefficients were determined using synthetic reference compounds dissolved in waterlmethanol mixtures that equalled the water/methanol mixture in which the respective compounds elute from the h.p.1.c.column. The percentageof methanol in water of these solvents is presented.
Xenobiotica Downloaded from informahealthcare.com by University of Saskatchewan on 05/16/12 For personal use only.
Compound
Methanol in water (%)
Molar extinction coefficient at 295nm (mM-lcm-’)
1 .oo
3-Fluoro-4-aminobenzyl alcohol 3-Chloro-4-aminobenzyl alcohol 3-Bromo-4-aminobenzyl alcohol
25 25 25
1.94 210
3-Fluoro-4-aminobenzaldehyde 3-Chloro-4-aminobenzaldehyde 3-Bromo-4-aminobenzaldehyde
40 40 40
7.99 10.0 9.98
N-(4’-amino-3’-fluorobenzyl)-2-fluoro-4-methylaniline N-(4’-amino-3’-chlorobenzyl)-2-chloro-4-methylaniline N-(4‘-amino-3’-bromobenzyl)-2-bromo-4-methylaniline
75 75 75
6.81 7.88 8 70
2,2’-Difluoro-4,4’-dimethylazoxybenzene 2,2’-~ichloro-4,4’-dimethylazoxybenzene
80 80
8.48 7.08
2,2’-Dichloro-4,4‘-dimethylhydrazobenzene
80
101
2,2’-Difluoro-4,4‘-dimethylazobenzene 2,2’-Dichloro-4,4’-dimethylazobenzene
80 80
6.55 5.96
the assay, which had been established to be 0 * 5 0 p ~ ,i.e. below 0.25 nmol/lOmin/nmol P450. This result is in accordance with the h.p.1.c. elution patterns (figures 1 and 2) in which no, or only very small, peaks were observed at the retention times of the synthetic hydroxylamine and nitroso derivatives. Using the peak areas in the h.p.1.c. elution patterns and the molar extinction coefficients presented in table 2, quantification of the various metabolites is possible. T h e results obtained are presented in table 3. From these data it follows that in the microsomal incubations the halogenated N-(4‘-aminobenzyl)-4-methylanilinesand the benzyl alcohols are the main metabolites formed, as together they make up about 80-90% of all metabolites formed from the three 2-halogenated-4-methylanilines. Furthermore, the results in table 3 demonstrate that, based on the formation of the benzyl alcohol and the benzaldehyde, the amount of C-hydroxylation increases going from 2-fluoro-4-methylaniline < 2-chloro-4-methylaniline < 2-bromo-4methylaniline. T h e total rate of conversion is lowest for 2-fluoro-4-methylaniIine, the other two compounds showing similar total conversion rates. Taking the formation of the hydroxylamine plus nitroso, and twice the amount of azoxy plus azo plus hydrazo derivatives to calculate the amount of N-hydroxylation, it follows that 2-chloro-4-methylaniline is N-hydroxylated to the highest extent (table 3). T h e azoxy, azo and hydrazo derivatives are counted twice, because they are believed to be formed from condensation of a nitroso and a hydroxylamino derivative (see Discussion section). Cytochrome P450 dependence of metabolite formation T o investigate the role of the microsomal cytochrone P450 system in the formation of the various metabolites, incubations were carried out in the presence of
4 1 k0.4 0.35 f 0.05
