Characteristics of two classes of azo dye reductase activity associated with rat liver microsomal cytochrome P450.

Azo dyes are reduced to primary amines by the microsomal enzymes NADPH-cytochrome P450 reductase and cytochrome P450. Amaranth, a highly polar dye, is reduced almost exclusively by rat liver microsomal cytochrome P450 and the reaction is inhibited almost totally by oxygen or CO. Activity is induced by pretreatment with phenobarbital or 3-methylcholanthrene. In contrast, microsomal reduction of the hepatocarcinogen dimethylaminoazobenzene (DAB), a lipid soluble, weakly polar compound, is insensitive to both oxygen and CO. However, reconstitution of activity with purified NADPH-cytochrome P450 reductase and a partially purified cytochrome P450 preparation indicates that activity is catalyzed almost exclusively by cytochrome P450. Activity is induced by clofibrate but not phenobarbital, beta-naphthoflavone, 3-methylcholanthrene, isosafrol, or pregnenolone-16 alpha-carbonitrile. These observations suggest the existence of at least two classes of azoreductase activity catalyzed by cytochrome P450. To investigate this possibility, the reduction of a number of azo dyes was investigated using microsomal and partially purified systems and the characteristics of the reactions were observed. Microsomal reduction of azo dyes structurally related to DAB required a polar electron-donating substituent on one ring. Activity was insensitive to oxygen and CO if the substrates had no additional substituents on either ring or contained only electron-donating substituents. Introduction of an electron-withdrawing group into the prime ring conferred oxygen and CO sensitivity on the reaction. Substrates in the former group are referred to as insensitive and substrates in the latter group as sensitive. Inhibitors of cytochrome P450 activity depressed reduction of both insensitive and sensitive substrates. In a fully reconstituted system containing lipid, highly purified NADPH-cytochrome P450 reductase and a partially purified cytochrome P450 preparation, rates of reduction of various insensitive substrates varied several-fold, whereas rates of reduction of sensitive substrates varied by three orders of magnitude. Using purified enzymes, each of the insensitive substrates was shown to be reduced by reductase alone, but only at a fraction of the rate seen in the fully reconstituted system, implying that reducing electrons were transferred to the dyes mainly from cytochrome P450. Conversely, there was substantial, in some cases almost exclusive, reduction of sensitive substrates by purified reductase alone and almost no inhibition by CO. Their reduction, however, was inhibited by CO in microsomal systems.(ABSTRACT TRUNCATED AT 250 WORDS)

Studies on the mechanism of reduction of azo dye carcinogens by rat liver microsomal cytochrome P-450.

This laboratory has described the azoreduction of p-dimethylaminoazobenzene (1c) by rat liver microsomal cytochrome P-450. To elucidate the mechanisms involved, the reduction of structurally related azobenzenes by hepatic microsomes was investigated. High substrate reactivity was observed for 1c, its corresponding secondary (1a) and primary (1b) amines and p-hydroxyazobenzene (1d). In contrast, only negligible rates were obtained for unsubstituted azobenzene (1g), hydrazobenzene (2g), p-isopropylazobenzene (1e) and 1f, the benzoylamide derivative of 1b. These results clearly indicate that electron-donating groups, such as hydroxyl or primary, secondary and tertiary amines, are essential for binding of azo dye carcinogens to liver microsomal cytochrome P-450 and, by implication, their enzymic reduction. No inhibition of azoreduction of 1c or 1d was obtained by addition of 1e, 1g, or 2g to the reaction mixture. In the presence of hepatic microsomes, a type I binding spectrum was obtained for 1d and type II binding spectra for 1a, 1b and 1c, the reactive azo dyes. In contrast, very weak binding was observed for the unreactive compounds 1e, 1f, 1g and 2g. Thus, there is good correlation between binding and substrate reactivity. The apparent lack of binding may explain the inability of the non-reactive compounds to inhibit azoreduction. The difference in the reduction rate observed for 1g vs. 1d suggested that hydroxylation would facilitate the reduction of an otherwise non-reactive azo dye. Support for such a mechanism was obtained in two experiments. In the first, marked facilitation of azoreduction of both the inactive compounds, 2g and 2f, was seen when they were incubated with microsomes under aerobic conditions where preliminary hydroxylation can occur. In the second, azobenzene was initially incubated aerobically with microsomes from phenobarbital- or beta-naphthoflavone-induced rats. The hydroxyazobenzene formed was then readily reduced anaerobically by microsomes from untreated rats.

Identification of human liver cytochrome P450 enzymes involved in the metabolism of SCH 351125, a CCR5 antagonist.

The identification and relative contribution of human cytochrome P450 enzyme(s) involved in the metabolism of SCH 351125 were investigated. In human liver microsomes, O-deethylation was the major metabolic pathway, whereas aromatization of a piperidine ring to pyridine and the reduction of the N-oxide moiety were minor routes. Recombinant human CYP3A4 and CYP2C9 both exhibited catalytic activity with respect to the formation of rotameric O-deethylated metabolites (M12, M13), the metabolites resulting from aromatization (M22/M24) and N-oxide reduction (M31). Using the relative activity factor (RAF) approach, the relative contributions of CYP3A4 and CYP2C9 to M13 formation were estimated to be 76 and 24%, respectively. There was a high correlation (r>0.96) between the rate of formation of M12 and M13 and 6 beta-hydroxylation of testosterone catalysed by CYP3A4/5. Ketoconazole (2microM) and CYP3A4/5-specific inhibitory monoclonal antibody inhibited the formation of M12 and M13 from human liver microsomes by approximately 60 and 71%, respectively. The results demonstrate that the in vitro metabolism of SCH 351125 is mediated primarily via CYP3A4 and that CYP2C9 plays a minor role. Clinical study designs should encompass these enzymology data to address any potential drug interactions.

The mechanism of microsomal azoreduction: predictions based on electronic aspects of structure-activity relationships.

The mechanism of microsomal azoreductase is regulated by the overall Hammett sigma substituent values on each ring. A substrate dye must exhibit an overall Hammett sigma substituent value either equal or more negative than -0.37 on either ring. Dyes with Hammett sigma substituent constants less negative than -0.37 will not be reduced by microsomal cytochrome P450. Microsomal reduction of azo dyes containing only electron-donating substituents on either ring is insensitive to both oxygen and carbon monoxide. The required Hammett sigma substituent value on the opposite benzene (prime) ring for I-substrates is therefore, sigma' P < or = 0. Reduction of azo dyes containing electron-withdrawing group on opposite (prime) is sensitive to both oxygen and carbon monoxide. The required Hammett sigma substituent value on the opposite benzene (prime) ring for S-substrates is, consequently, sigma' P > 0 (Table 3). Redox Potentials. Anaerobic cyclic voltammograms of azobenzene derivatives verify the following points: A nonsubstrate azo dye will not exhibit a positive potential. (Several nonsubstrate hydrazobenzenes exhibited positive potentials, but in a low range 0.41-0.48 V. Consequently, cyclic voltammetry can distinguish between nonsubstrate azobenzenes and their nonsubstrate half-reduced hydrazo analogs.) A substrate azo dye will exhibit a positive potential in the range +1.00 to +1.50 V. I-substrate: Both negative potentials are stable in air. S-substrate: The first negative potential will immediately quench upon exposure to air. I-substrates exhibit on average potentials which are approximately 0.6 V more negative than those for S-substrates. A comparison between the oxidative and the reductive pathway of microsomal cytochrome P450 indicates a similarity in the first two steps in the reaction cycle, for example, substrate binding and uptake of the first electron by the cytochrome [76, 109, 110]. Upon reduction of the iron, ferrous cytochrome P450 may bind oxygen or carbon monoxide in a competitive manner in the oxidative cycle or may directly transfer the electrons to the substrate in a stepwise fashion in the reductive cycle [76]. Estabrook et al. [111] suggested that carbon monoxide insensitivity can occur when the formation of ferrous cytochrome P450 substrate complex is rate limiting for the overall reaction. Structure-activity relationships of azo compounds depend on (1) the electron transport component and (2) the oxidation-reduction potential of the dye, which determines its ability to accept electrons from cytochrome P-450. Nesnow et al. examined a group of 36 aryl azo dyes for their ability to be reduced by rat liver microsomal azoreductase.(ABSTRACT TRUNCATED AT 400 WORDS)

Biomimetic models for monooxygenases.

The microsomal mixed function oxidase system contains the cytochrome P-450 oxidative drug metabolizing family of enzymes. The catalytic cycle of cytochrome P-450 is believed to involve the formation of an active iron-oxygen species which is responsible for oxygen transfer to the substrate. This assumption is supported by the fact that a number of peroxidative agents can replace NADPH, the reductase, and oxygen as co-reactants in most oxidative reactions of microsomal cytochrome P-450. We have found that a mixture of either ferrous or ferric ions with hydrogen peroxide (Fenton and Ruff reagents) can serve as biomimetic models for cytochrome P-450 in hydroxylation, exposidation, sulfoxidation, and N-demethylation of various drugs. The existance of an iron-oxo active species in both Fenton and Ruff type reactions has been postulated and provides reaction cycles similar to those of cytochrome p-450. Other model systems for the hepatic hydroxylation and epoxidation using transition metal complexes with porphyrin are also discussed. The present paper reviews the various biomimetic models of the heme cytochrome P-450 and emphasizes their simulation of hepatic drug metabolism and their potential medical and industrial applications.

Simulation of the hepatic metabolism of stilbene and its tricyclic derivatives by Fenton and Ruff reagents: models for cytochrome P-450 activation of chemical carcinogens.

Reactions of trans-stilbene, cis-stilbene, 5H-dibenzo [a,d] cyclo-heptene 5-one and 5H-dibenz [b,f] azepine (iminostilbene) with Fenton reagent [Fe (II)/H2O2] clearly simulate their hepatic metabolism. Expoxidation on the corresponding ethylenic linkage was found to be a common pathway of these compounds. Epoxides of trans-stilbene, cis-stilbene, and 5H-dibenzo[a,d]cycloheptene 5-one were further oxidized to dihydrodiols, alpha-hydroxyketones, diketones, and finally cleavage of the ethylenic bonds to the formation of the corresponding aldehydes. However, the unstable epoxide of iminostilbene gave 9-acridinecarbaldehyde that is further oxidized to 9-acridone. Reaction of both trans- and cis-stilbene with Ruff reagent [Fe III)/H2O2] gave the same oxidative products to that obtained from Fenton reagent. The radical scavenger 2,6 bis (1,1-dimethylethyl)-4-methyl phenol (BHT) decreases the total yield conversion and increases the formation ratio of both cis-epoxide and d,l-hydrobenzoin from cis-stilbene.

A novel application of cyclic voltammetry for direct investigation of metabolic intermediates in microsomal azo reduction.

We have established that reduction of azo dyes structurally related to 4-(dimethylamino)-azobenzene (DAB) by rat liver microsomal cytochrome P-450 requires a polar electron-donating substituent on one ring. Reduction of azo dyes containing only electron-donating substituents is insensitive to both oxygen and CO (I substrates). However, reduction of azo dyes containing electron-withdrawing substituents as well is sensitive to both oxygen and CO (S substrates). Positive, irreversible potentials were observed by cyclic voltammetry (CV) in anhydrous solutions for both I and S substrates but not for the nonreducible azo dyes. This positive potential permits electron transfer to dyes from NADPH-cytochrome P-450 reductase and from cytochrome P-450, both of which have negative potentials. Reduction products retaining electron-donating groups (amino, phenolic) also exhibited positive potentials, implying that these groups contribute the positive potential in the dye molecule. All substrates also exhibited two negative potentials with a clear distinction between I and S substrates. The latter exhibited, on average, potentials that were less negative than the former by about 0.6 V. This is consistent with the more rapid reduction of S substrates by liver microsomes [Zbaida and Levine (1990) Biochem. Pharmacol. 40, 2415-2423]. Admitting air to the system quenched the first potential for S but not for I substrates, which is consistent with the oxygen sensitivities of their reduction. Addition of water significantly shifted the second negative potential to a more positive value, ultimately leading to single negative potential. The water permits rapid protonation of the two-electron-reduced intermediate, facilitating further reduction.(ABSTRACT TRUNCATED AT 250 WORDS)

The role of iron chelates on the selectivity of Fenton reagent in hydroxylation, N-demethylation, and sulfoxidation of cimetidine: a novel biomimetic model for the regioselectivity of cytochrome P-450.

The effect of iron chelates on the reaction of cimetidine with Fenton reagent [Fe(II)/H2O2] has been investigated. Iron chelates with high affinity to ferrous ions inhibited this reaction. However, iron chelates with high affinity to ferric ions selectively promote either hydroxylation, N-demethylation, or sulfoxidation of cimetidine. These results indicate that the oxidation of cimetidine with hydrogen peroxide activated by various chelated ferrous ions serves as a biomimetic model for the regioselectivity of multiple forms of cytochrome P-450 in the metabolism of cimetidine.

Role of electronic factors in binding and reduction of azo dyes by hepatic microsomes.

The factors which regulate the binding and reduction of azo dyes by rat liver microsomes have been investigated. Azo dyes having both electron-donating and -withdrawing substituents were reduced more readily both enzymically and chemically (dithionite) than those containing electron-donating substituents alone, which is consistent with less negative oxidation-reduction potentials of the former. A linear correlation between Vmax and Km was seen for substrates having only electron-donating substituents, suggesting a possible inverse relation between binding affinity and rate of reduction. This relationship was not apparent for substrates having both electron-donating and -withdrawing substituents. A lower Km was seen with substrates having a greater number of heteroatoms bearing nonbonding electrons in either electron-donating or -withdrawing groups. Furthermore, more basic dyes, which have a higher density of nonbonding electrons, showed an inverse correlation with both Vmax and Km. The requirement of nonbonding electrons was also observed with the binding of the fully reduced amine metabolites to microsomal cytochrome P-450. Type II binding spectra were observed for both aniline and 4-chloroaniline, but not with aniline derivatives bearing electron-withdrawing substituents such as methyl anthranilate and methyl-4-aminobenzoate. Electron-withdrawing substituents increase delocalization of nonbonding electrons on the amino residue; consequently, these are no longer available for binding to the enzyme. The binding constant of the reduced metabolite, aniline (36 microM), relative to the Km values of substrate azo dyes (range 0.31-1.73 microM) implies that the more weakly bound amine metabolites are readily released from the binding/catalytic site.(ABSTRACT TRUNCATED AT 250 WORDS)

Hepatic microsomal azoreductase activity. Reactivity of azo dye substrates is determined by their electron densities and redox potentials.

We have previously established that microsomal reduction of azo dyes requires polar electron-donating ring substituents [e.g. -OH, -NH2, -NHCH3, or -N(CH3)2]. Reduction of dyes substituted exclusively with such moieties on either phenyl ring is insensitive to both O2 and CO (I-substrates). In contrast, azoreduction of compounds containing additional electron-withdrawing substituents (e.g. -SO3H, -COOH, -COOCH3, and -AsO3H2) on the opposite (prime) phenyl ring is sensitive to both O2 and CO (S-substrates). We have recently shown that Hammett aromatic substituent constants and redox potentials of the dyes, determined by cyclic voltammetry (CV), distinguish between I- and S-substrates. Dyes that exhibit positive Hammett sigma substituent constants on 1 of the 2 rings are S-substrates, and undergo immediate quenching of their one electron-reduced intermediates upon exposure to air, as observed by CV. In contrast, dyes that have negative Hammett sigma values on one or both rings are I-substrates, and their one electron-reduced intermediates are relatively stable in air. In this study, we have investigated the susceptibility to microsomal azoreduction of monosubstituted dyes, with ring substituents possessing a wide range of Hammett sigma values. We have observed a direct correlation between the susceptibility of these compounds to microsomal azoreduction and the Hammett sigma constant of the aromatic ring substituent, and the presence of a positive potential in the dyes observed by CV. A Hammett sigma value of -0.37 or lower is required for substrate activity. Dyes with two substituents in the prime ring confirmed the previous correlation of negative and positive Hammett sigma values with I- and S-substrate activities, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Reactions of theophylline, theobromine and caffeine with Fenton's reagent--simulation of hepatic metabolism.

Theophylline and caffeine undergo N-demethylation and hydroxylation by Fenton's reagent to give uric acid derivatives; theophylline is oxidized mainly to 1-methyluric acid, and 1,3-dimethyluric acid and 1-methyluric acid are the major products obtained from caffeine. Theobromine undergoes predominantly N-demethylation to give 7-methylxanthine. The nature of the products indicate that these reactions simulate hepatic drug metabolism.

Multiple mechanisms in hepatic microsomal azoreduction.

1. Microsomal reduction of azo dyes related to dimethylaminoazobenzene (DAB) is catalysed by at least two types of cytochrome P-450. The first is selectively induced by clofibrate. The second is induced by phenobarbital, beta-naphthoflavone, isosafrole, and pregnenolone-16 alpha-carbonitrile, as well as clofibrate. 2. Azoreduction by the first type of P-450 is insensitive to both O2 and CO and involves dyes with only electron-donating substituents (I substrates). 3. Azoreduction by the second type of P-450 is inhibited by both O2 and CO and involves dyes with electron-withdrawing as well as donating substituents (S substrates). 4. All azo dye substrates exhibit two negative and one positive redox potential, as measured anaerobically by cyclic voltammetry. The negative potentials reflect one- and two-electron reductions while the positive potential permits electron transfer from microsomal P-450, the redox potential for which is reported to be negative (approximately 0.35 V). The positive potential is associated with a polar electron-donating group para to the azo linkage, which is an absolute requirement for microsomal reduction. Dyes without this functional group do not exhibit positive potentials and are not reduced. 5. The first negative potential of S substrates is quenched upon admitting air to the system, whereas this potential is unaffected in I substrates. The relative stability of the one-electron reduced state may be an explanation for the differential O2 sensitivity of I and S substrate reduction.

Substrates for microsomal azoreductase. Hammett substituent effects, NMR studies, and response to inhibitors.

In previous studies on azoreduction by microsomal cytochrome P-450, we identified two classes of substrates structurally related to 4-dimethylaminoazobenzene. Both require polar electron-donating groups for binding to enzyme and are differentiated by their structure, their redox potentials, their rates of chemical and enzymic reduction, and the influence on their metabolism of inducing agents, CO and O2. Azo compounds whose reductions are insensitive to CO and O2 (I-substrates) contain electron-donating substituents on either ring. Azo compounds whose reductions are O2- and CO-sensitive (S-substrates) also contain electron-withdrawing groups on the opposite (prime) ring. For all dyes, NMR studies revealed minor differences in the chemical shifts of the protons attached to the phenyl ring substituted with electron-donating substituents (ring A). This is consistent with the narrow range of pKa's (basicity) and KM values for all substrates. However, there are significant differences in the chemical shifts of the aromatic protons of the prime ring (ring B). The difference in chemical shifts is most pronounced for aromatic protons adjacent to the prime ring substituents, showing a clear distinction between I and S substrates. Furthermore, the Hammett sigma substituent constants on the prime ring clearly distinguish between the two classes of dyes. I- and S-substrates have negative and positive sigma Hammett values, respectively. This implies that the mechanism of microsomal azoreduction is critically dependent on the charge and redox potentials of the dyes and is exclusively determined by the nature of the substituents on the prime ring.(ABSTRACT TRUNCATED AT 250 WORDS)

Replacing the Mixed Function Oxidase N-Demethylation by Hydrogen Peroxide and either Hemoglobin or Ferrous Sulfate.

Formaldehyde liberated from N-demethylation of aminopyrine, N-methyl-piperazine and from the optically active (+)-and (-)-ephedrine was quantitated from the incubation mixtures of hepatic microsomes and from reactions of hydrogen peroxide with either hemoglobin or ferrous sulfate. Stereoselectivity was observed only in incubations with hepatic microsomes. N-Formyl derivatives were isolated from the reactions of N-ethylmorpholine, aminopyrine, N-methylpiperazine and N,N'-dimethyl-piperazine with ferrous sulfate and hydrogen peroxide (Fenton reagent). Electron-withdrawing groups (NH, NCH3, O) at the fourth position to the N-alky! group facilitate N-dealkylation with Fenton reagent.

Reaction of cimetidine with Fenton reagent. A biomimetic model for the mixed-function oxidase drug metabolism.

Cimetidine sulfoxide, N-desmethylcimetidine, N-desmethylcimetidine sulfoxide, cimetidine guanylurea, and the 5-hydroxymethylimidazole derivative of cimetidine sulfoxide were isolated from the reaction of cimetidine with Fenton reagent. The product distribution from the reaction of cimetidine with Fenton reagent clearly simulates the metabolism of cimetidine by the mixed-function oxidase.

In vitro uptake of SCH 27899 (evernimicin) by rat alveolar macrophages.

The in vitro uptake of [(14)C]evernimicin ([(14)C]SCH 27899) by primary cultures of rat alveolar macrophages and hepatocytes was determined. Both cell populations exhibited linear rates of uptake. However, the initial rate of drug uptake by alveolar macrophages was about threefold higher than that by hepatocytes. These findings demonstrate that [(14)C]evernimicin is taken up by rat alveolar macrophages, supporting the likelihood that the drug is able to reach sites of infection.

In vitro metabolism of 10-(3-chlorophenyl)-6,8,9,10-tetrahydrobenzo[b][1,8]naphthyridin-5(7H)- one, a topical antipsoriatic agent. Use of precision-cut rat, dog, monkey and human liver slices, and chemical synthesis of metabolites.

The metabolism of SCH 40120, which is the clinically effective antipsoriatic drug 10-(3-chlorophenyl)-6,8,9,10-tetrahydrobenzol[b][1,8]naphthyrid in-5(7H)-one, was determined in vitro. Rat, dog, cynomolgus monkey, and human liver slices hydroxylated the aliphatic, cyclohexenyl ring of the drug and conjugated the resulting carbinol. The identified metabolites comprised the corresponding 6-, 7-, and 9-carbinols, the glucuronide of the 6-carbinol, and the 6-ketone derived from the parent drug. Although the three carbinols appeared in the liver isolates of all species studied, the relative amounts of these metabolites varied across species. With a high, non-physiological ratio of substrate to liver, the 6-carbinol and its glucuronide were the major metabolites in human and monkey, whereas the 6-ketone was a minor metabolite in dog. Containing a stereogenic axis and center, the 6-carbinol existed as diastereomeric atropisomers. Its structure was established by 13C and 1H NMR spectroscopy, mass spectrometry, and comparison to an authentic sample.