Chem. Res. Toxicol. 1988,1, 208-215
208
2,2-Dialkyl-1,2-dihydroquinolines: Cytochrome P-450 Catalyzed N-Alkylporphyrin Formation, Ferrochelatase Inhibition, and Induction of 5-Aminolevulinic Acid Synthase Activity David Lukton,? Jane E. Mackie,* Jae S. Lee,? Gerald S. Marks,*>$and Paul R. Ortiz de Montellano**t Department of Pharmaceutical Chemistry, School of Pharmacy, and Liver Center, University of California, S a n Francisco, California 94143, and Department of Pharmacology and Toxicology, Queen’s University, Kingston, Ontario, K 7 L 3N6, Canada Received April 12, 1988
Incubation of 2,4-diethyl-l,2-dihydro-2-methylquinoline (DMDQ) with hepatic microsomes from rats pretreated with phenobarbital, 3-methylcholanthrene, pregnenolone- 16a-carbonitrile, or dexamethasone results in minor loss of the cytochrome P-450 chromophore and accumulation of a hepatic pigment. The hepatic pigment consists of the four regioisomers of N-ethylprotoporphyrin I X and minor amounts of the corresponding N-methyl regioisomers. Exposure of chick embryo liver cells to DMDQ results in inhibition of their ferrochelatase activity, induction of their 5-aminolevulinic acid synthase activity, and accumulation of protoporphyrin IX. 1,2Dihydro-2,2,4-trimethylquinoline (TMDQ) causes negligible loss of cytochrome P-450 in rat liver microsomes but in vivo still produces the four N-methylprotoporphyrin I X regioisomers in low yield. Furthermore, it inhibits ferrochelatase activity, elevates 5-aminolevulinic acid synthase activity, and causes protoporphyrin IX accumulation in cultured chick embryo hepatocytes. One-electron oxidation of the 2,2-dialkyl-l,2-dihydroquinolines to radical cations is postulated to result in N-alkylation of the prosthetic heme group of cytochrome P-450. The N-alkylprotoporphyrins IX thus formed are potent inhibitors of ferrochelatase. Inhibition of ferrochelatase causes the induction of 5-aminolevulinic acid synthase and the accumulation of protoporphyrin IX. Heme alkylation and ferrochelatase inhibition may be generally associated with substrates that are subject to cytochrome P-450 mediated oxidative extrusion of alkyl radicals.
The oxidative metabolism of 3,5-dicarbethoxy-2,6-dimethyl-4-ethyl-l,4-dihydropyridine (DDEP) by rat liver microsomes results in destruction of 30-40% of the total P-450 chromophore even though chromophore destruction by the 4-methyl analogue (DDC) is negligible (1-3). Immunoquantitation of the cytochrome P-450 isozymes that are lost (4), and measurements of the abilities of purified isozymes to oxidize nifedipine, a 4-aryldihydropyridine (5), suggest that cytochromes P-45Op and P-450h are the isozymes primarily destroyed by DDEP. Administration of DDEP to rats causes the accumulation of a hepatic pigment unambiguously identified as a mixture of the four regioisomers of N-ethylprotoporphyrin IX (1, 2). The corresponding N-methylprotoporphyrin IX isomers are obtained, albeit in much lower yield, from rats treated with DDC (6). Mechanistic studies suggest that oxidation of the 4-alkyl-l,4-dihydropyridines to their radical cations by cytochrome P-450 is followed by extrusion of the 4-alkyl group as a free radical that alkylates the prosthetic hemel group (1-3, 7, 8). The cytochrome P-450 catalyzed extrusion of ethyl radicals from DDEP has been confirmed by spin-trapping experiments (2). DDC, which causes no detectable loss of the cytochrome P-450 chromophore, likewise causes no detectable radical formation but still provides low yields of heme adducts when administered in vivo. DDEP analogues with other 4-alkyl moieties have been shown to cause detectable cytochrome P-450 destruction, free radical formation, and heme adduct for+ University
of California.
* Queen’s University.
0893-228x/88/2701-0208$01.50/0
mation (2,8-11). The failure of RDC to cause significant enzyme destruction, and free radical formation is probably a reflection of the fact that the lower stability of the methyl radical makes it more difficult to eliminate than the ethyl radical. The dihydropyridine radical cation therefore aromatizes to a large extent via an alternative mechanism that does not involve methyl radical extrusion (11). Exposure of rodents or chick embryos to DDC or DDEP results in a marked decrease in their hepatic ferrochelatase activity (12-14). This decrease in activity is due to the accumulation of N-methyl- or N-ethylprotoporphyrin IX produced by alkylation of the prosthetic group of cytochrome P-450 (6, 15). N-Methylprotoporphyrin IX is a potent inhibitor of ferrochelatase (16), as are the N A ~ regioisomers (i.e., those with the N-alkyl group on the vinyl ring substituted pyrrole rings) of N-ethyl and other Nalkylprotoporphyrin IX homologues (16). Inhibition of hepatic ferrochelatase activity is followed by a rise in hepatic porphyrin levels, primarily protoporphyrin IX, due to concommitant induction of 5-aminolevulinic acid synthase, the rate-limiting enzyme of heme biosynthesis (17). Induction of 5-aminolevulinic acid synthase is caused by inhibition of ferrochelatase, which lowers cellular heme levels and thereby derepresses 5-aminolevulinic acid synthase activity, and to the lipophilicity of the dihydroAbbreviations: DDC,3,5-diethoxycarbonyl-2,4,6-trimethyl-1,4-dihydropyridine; DDEP, 3,5-diethoxycarbonyl-2,6-dimethyl-4-ethyl-l,4dihydropyridine;POBN,a-(l-oxc-4-pyridyl)-N-tert-butylni~one; DMDQ, 2,4-diethyl-2-methyl-1,2-dihydroquinoline; TMDQ,2,2,4-trimethyl-1,2dihydroquinoline; heme, iron protoporphyrin IX regardless of the iron oxidation state.
0 1988 American Chemical Society
N - Alkylporphyrins and Ferrochelatase Inhibition
pyridines, which appears to be involved in a still unknown manner in the induction of 5-aminolevulinic acid synthase (17). Although a series of 4-alkyl homologues of DDC and DDEP destroy cytochrome P-450 and inhibit ferrochelatase in a manner analogous to that for the parent 4-alkyl-1,4-dihydropyridines,only recently has it been found that other structures may have a similar mechanism of action. Thus, recent work has shown that 3-[2-((2,4,6trimethylphenyl)thio)ethyl]-4-methylsydnone(TTMS) produces protoporphyria by a comparable mechanism in which N-vinylprotoporphyrin IX is produced as the ferrochelatase inhibitory product (18, 19). In order to better define the elemental properties required for prosthetic heme alkylation and ferrochelatase inhibition, we have investigated the biochemical effects of two 2,2-dialkyl1,2-dihydroquinolines. These nitrogen heterocycles were chosen because their radical cations can only aromatize by eliminating one of the 2-alkyl groups. Indeed, evidence exists that chemical oxidation of these compounds results in elimination of one of the 2-alkyl groups as a free radical (20). We report here that oxidative metabolism of two 2,2-dialkyl-1,2-dihydroquinolinescauses minor loss of the hepatic cytochrome P-450 chromophore. This chromophore loss is associated with the formation of N-alkylporphyrins and the inhibition of ferrochelatase, induction of 5-aminolevulinic acid synthase activity, and elevation of hepatic protoporphyrin levels.
Materials and Methods Materials. NADPH and dexamethasone were purchased from Sigma Chemical Co. (St. Louis, MO), dilauroylphosphatidylcholine was purchased from Serdary Research Labs (Port Huron, MI), were purand POBN and 1,2-dihydro-2,2,4-trimethylquinoline chased from Aldrich Chemical Co. (Milwaukee, WI). The latter was distilled prior to use. Pregnenolone-16a-carbonitrile was a gift from G.D. Searle & Co. (Skokie, IL). The preparation of authentic N-methylprotoporphyrin IX, N-ethylprotoporphyrin IX, and DDEP has been reported (1,2,21). IH NMR spectra were recorded in deuteriated chloroform. Chemical shift values are reported in ppm relative to internal tetramethylsilane standard. EPR spectra were recorded on a Varian E-104 instrument. Deionized, glass-distilled water was employed for the EPR experiments. Elemental analyses were done by the Microanalytical Laboratory of the University of California, Berkeley. 2,4-Diethyl-1,2-dihydro-2-methylquinoline ( 1 , 2). The synthesis of this agent is based on the general synthetic approach of Layer (22). A mixture of 4.3 g of aniline, 14 g of butanone, and 1.7 g of p-toluenesulfonic acid in 50 mL of xylene was refluxed overnight in an apparatus fitted with a Dean-Stark trap to remove the water evolved in the reaction. The bulk of the xylene was distilled off and methylene chloride was added. The resulting solution was washed sequentially with 10% potassium carbonate and water. The methylene chloride was removed on a rotary evaporator, and the residue was fractionally distilled under vacuum. The material distilling between the aniline and the dark, viscous, final fractions was chromatographed on a 2000-pm silica gel plate developed with 25% methylene chloride in hexane. The fluorescent band was collected, rechromatographed under the same conditions, and distilled in a Kugelrohr apparatus. The oil thus obtained was stored in sealed ampules a t -20 "C: 'H NMR (240 MHz) 7.10 (d), 6.96 (t), 6.61 (t), and 6.43 (d) (aryl protons), 5.18 (9, C=CH), 3.46 (br s, NH), 2.34 (4, C=CCHzCH3), 1.45 (m, CH&H,), 1.21 (s, CH,), 1.15 (t, C=CCH2CH3), and 0.89 ppm (CH2CH3);EIMS, m / z 201 (M'), 172 (base peak). Anal. Calcd for C14H19N: C, 83.53; H, 9.51; N, 6.96. Found: C, 83.51; H, 9.55; N, 6.89. Cytochrome P-450 Destruction in Hepatic Microsomes. Rat liver microsomes were prepared from rats injected intraperitoneally once a day for 4 days with phenobarbital (80 mg/kg), 3-methylcholanthrene (40 mg/ kg), pregnenolone-16a-carbonitrile (50 mg/kg), or dexamethasone (100 mg/kg, 2 days; 80 mg/kg, 2
Chem. Res. Toxicol., Vol. 1, No. 4, 1988 209 days). The phenobarbital was injected in water, and the other three agents were injected in corn oil (23). A solution of DMDQ in tetrahydrofuran (5 mg/mL, 440 pL) was placed in a 20-mL scintillation vial, and the tetrahydrofuran was removed on a rotary evaporator. Residual traces of tetrahydrofuran were removed under a stream of nitrogen. To this was then added 10 mL of the appropriate preparation of liver microsomes (approximately 1 nmol of P-450/mL) and the mixture was incubated 3 min a t 25 "C before 600 pL of NADPH in buffer (12 mg/mL) was added. The final concentrations of DMDQ and NADPH were 1and 1.4 mM, respectively. Aliquots (1 mL) were transferred every 20 s to test tubes in an ice bath. Immediately after taking the final aliquot, the tubes were bubbled with carbon monoxide and dithionite was added. The cytochrome P-450 concentration was then determined by difference spectroscopy with respect to an unreduced sample of microsomes on an Aminco DW-2a spectrophotometer (24). Control incubations were run in the absence of DMDQ or NADPH. Assay for Inactivation of Reconstituted Cytochrome P-450b. Cytochrome P-450b was purified by the procedure of Waxman and Walsh (25)and cytochrome P-450 reductase by that of Shephard et al. from phenobarbital-pretreated rats (26). Both proteins migrated as single bands on SDS-gel electrophoresis. Incubations with the reconstituted system contained 1nmol/mL of purified cytochrome P-450b, 1nmol/mL of purified cytochrome P-450 reductase, 40 p g / d of dilauroylphosphatidylcholine,1mM NADPH, and 1mM inhibitor. The mixture was incubated at 37 "C for 20 min. Aliquots (20 pL) taken at various times during the incubation were assayed for 7-ethoxycoumarin 0-deethylase activity. The assay incubations contained 20 pmol/mL of P-450b, 20 pmol/mL of P-450 reductase, 0.08 pg/mL of dilauroylphosphatidylcholine, 0.5 mM NADPH, and 0.5 mM 7-ethoxycoumarin. The activity of the enzyme was determined by measuring the fluorescence of 7-hydroxycoumarin at 458 nm when excited a t 368 nm (27). Control experiments established that DMDQ does not quench the fluorescence of 7-hydroxycoumarin. Assay for Loss of DDEP Oxidase Activity. DMDQ (1mM) was incubated with liver microsomes from phenobarbital-pretreated rats (10 nmol of cytochrome P-450/mL), NADP (1mM), glucose 6-phosphate dehydrogenase (4 units/mL), glucose 6phosphate (3 mM), MgClz (2 mM), KC1 (150 mM), diethylenetliaminepentaacetic acid (1.5 mM), in 10 mL of phosphate buffer (50 mM, pH 7.4) at 25 "C. Control incubations were identical except for the presence of DMDQ. At the appropriate times, 900-pL aliquots of the incubation mixture were transferred to the assay mixture. The assay mixture (3-mL volume) containing 3 nmol/mL of microsomal cytochrome P-450,l mM DDEP, 1 mM NADPH, and residual 0.3 mM DMDQ was incubated 10 min a t 37 "C. To the mixture was then added 400 pL of a solution of sodium carbonate (1 M) and NaCl (2 M) and 10 pL of a 0.5 mg/mL solution of internal standard. The internal standard was the 4-phenyl analogue of DDEP (11). The mixture was extracted twice with 2-mL portions of methylene chloride, the combined extracts were evaporated to dryness, and the residue was taken up in 100 pL of 25% (v/v) tetrahydrofuran in hexane. A 10-pL sample was analyzed by high-pressure liquid chromatography on a Whatman Partisill0 column (4.6 mm X 25 cm) eluted with 25% tetrahydrofuran in hexane at a flow rate of 1mL/min. The eluent was monitored with a variable wavelength detector set a t 234 nm. Isolation and Characterization of Alkylated Porphyrins (1, 2). DMDQ (65 mg in 0.5 mL of dimethyl sulfoxide) was administered intraperitoneally on the fifth day to male Sprague-Dawley rats weighing approximately 250 g that had received an 80 mg/kg intraperitoneal dose of phenobarbital once a day for 4 days. The rats were decapitated 4 h after administration of DMDQ, and their livers were perfused with ice-cold 1.15% KC1 solution. The livers were then removed and were homogenized in a blender in cold 5% (v/v) H2S0,/methanol(100 mL/rat liver). The mixture was allowed to stand overnight at 4 "C before methylene chloride and water were added. The organic layer was washed four times with 15% potassium carbonate and was dried over anhydrous sodium sulfate. A solution of zinc acetate in methanol (3 mg/rat liver) was then added, the organic solvent removed under vacuum, and the resulting residue chromatographed on 2000-pm silica gel G plates developed with 10% (v/v) acetone/methylene chloride. The red fluorescent
Lukton et al.
210 Chem. Res. Toxicol., Vol. 1, No. 4, 1988 pigment band was extracted into acetone and was subjected to high-pressure liquid chromatography on a 25-cm Whatman PAC column eluted with a linear solvent gradient rising from 1:l (v/v) tetrahydrofuran/hexane to 100% methanol. The N-alkylporphyrins thus obtained were dissolved in methylene chloride and were demetalated by adding a few drops of 5% (v/v) H2S04 in methanol. The porphyrin solutions were then washed with water and 10% potassium carbonate and were dried over sodium sulfate. The residue obtained on solvent removal was fractionated by high-pressure liquid chromatography on a PAC column eluted isocratically with 0.75% (vjv) methanol in 1:l (v/v) tetrahydrofuranjhexane. The N-alkylporphyrinisomers thus obtained were individually characterized. The N-alkylporphyrins formed in rats treated with TMDQ were isolated by essentially the same procedure. A green, nonfluorescing, band located above the fluorescent band in the thin-layer chromatogram of the DMDQ pigment was also removed from the plate with acetone. The material thus obtained was purified by high-pressure liquid chromatography as described above. The purified material, which had an absorbance maximum at 442 nm, exhibited a parent ion at mjz 646 in the field desorption mass spectrum. Washing a chloroform solution of the sample sequentially with dilute aqueous acid, water, and potassium carbonate solution shifted the absorbance maximum to 428 nm. Addition of thiophenol, which is reported to demetalate iron N-alkylporphyrins (28), shifts the absorption maximum to 420 nm. The absorption spectrum of this material is identical with that of N-ethylprotoporphyrin IX and, on addition of zinc acetate, gives a spectrum identical with that of the zinc complex of N-ethylprotoporphyrin IX. Spin-Trapping Experiments. Incubations of DMDQ with microsomes from phenobarbital- or dexamethasone-pretreated rats (5-mL volume) were set up as described above except that POBN (20 mM) was added and the DMDQ concentration was varied between 1and 10 mM. Aliquota were withdrawn at various times and their EPR spectrum recorded. The residual mixtures were extracted with methylene chloride at the end of the incubation period, the extracts were concentrated to final volumes of approximately 100 p L (l/w of the original microsomal volume), and the EPR spectra of the concentrated samples were recorded. Control incubations were carried out in parallel in the absence of DMDQ. Determination of Ferrochelatase Activity. Livers were obtained from laday old chick embryos, and the hepatocytes were cultured by using a previously described method (10,29). The medium (Waymouth 705/1A) was changed 24 h after plating the cells onto 10-cm sterile Petri dishes containing 15 mL of medium. The drugs were added in 95% ethanol (maximum volume, 20 pL) 18 h after the medium was changed. Cells were harvested 6 h after drug administration and ferrochelatase activity was determined with a crude ferrochelatase preparation using ferrous iron and mesoporphyrin as substrates (IO, 18). The production of mesoheme was quantitated by means of the pyridine hemochromogen assay. Determination of the Porphyrin Patterns. Chick embryo hepatocyte culture was set up by using 5-cm Petri dishes containing 5 mL of media to determine the nature of accumulating porphyrins. The medium was changed 24 h after cells were plated, at which time drugs were added in 95% ethanol (maximum volume, 15 pL). Cells were harvested 24 h later, and porphyrins were extracted and quantitated with a fluorescence detector connected to an HPLC system (IO, 31) with a normal-phase silica gel column. Determination of 5-Aminolevulinic Acid Synthase Activity. Chick embryo hepatocyte culture was set up by using 10-cm Petri dishes, as described above. The medium was changed at 24 and 44 h after plating of the cells. The drugs were added a t 44 h, and the cells were harvested at various times after this. 5-Aminolevulinic acid synthase activity was determined by quantitating the formation of 5-aminolevulinic acid by using [2,3-14C2]succinic acid as the substrate (30, 31).
Results D e s t r u c t i o n of C y t o c h r o m e P-450. Incubation of hepatic microsomes from phenobarbital-pretreated r a t s
0
I
1
4
50
100
150
Time (second)
Figure 1. Destruction by DMDQ of cytochrome P-450 in hepatic microsomes from phenobarbital- or dexamethasone-pretreated rats. The losses of P-450 caused by incubation of 1mM DMDQ at 25 "C with phenobarbital-induced microsomes in the absence (e)or presence ( 8 )of NADPH or with dexamethasone-induced microsomes in the absence ( 0 )or presence (B) of NADPH are shown. The phenobarbital and dexamethasonevalues are averages of five and two independent incubations, respectively. Standard deviations are indicated by the bars for the phenobarbital data and the range of the two values by the bars for the dexamethasone data. with DMDQ causes rapid b u t limited destruction of t h e cytochrome P-450 chromophore (Figure 1). Chromophore loss is DMDQ-, NADPH-, a n d time-dependent b u t does not exceed a modest 8% of t h e total cytochrome P-450 chromophore even if t h e incubation period is extended t o 1 h. Even less (
