Testosterone metabolism by cytochrome P-450 isozymes RLM3 and RLM5 and by microsomes. Metabolite identification

Testosterone metabolism by cytochrome P-450 isozymes RLM3 and RLM5 in a reconstituted system and by rat liver microsomes was examined. Eleven metabolites were detected. Two of these, found in spots 2 and 4 of a thin layer plate, were only formed by the rat liver microsomes and may represent reductive metabolites of testosterone. A number of monohydroxy metabolites were conclusively identified by gas chromatography-mass spectrometry. These include the 2-, 6 beta-, 7 alpha-, and 16 alpha-hydroxy isomers. Liver microsomes formed the 2 alpha- and 2 beta-epimers in a 1:2 ratio and both co-chromatographed with a third reduced metabolite in thin layer plate spot 4. In contrast with RLM5 about 90% of the 2-hydroxy isomer was the 2 alpha-epimer. RLM3 did not perform the 2-hydroxylation in detectable amounts. The 6 beta-isomer was a major metabolite of RLM3 and microsomes, but a minor product of metabolism by RLM5. In contrast, the 7 alpha-isomer was a minor metabolite of RLM3, was not formed by RLM5, and was a major microsomal metabolite. Hydroxylation at position 16 alpha was a major activity of RLM5 and the heterogeneous microsomal cytochromes, but with RLM3 it was a minor reaction. One new metabolite was found which appeared to be hydroxylated in the D-ring, had a mass spectrum different from both 16 alpha- and 16 beta-hydroxytestosterone, and was tentatively identified as a 15-hydroxy isomer. In agreement with the literature, androstene-3,17-dione was found to be an oxidative metabolite of testosterone by both microsomes and purified cytochrome P-450. It was a major metabolite of RLM5 but was not produced by RLM3. Studies with 18O2 and H218O conclusively show that oxidation of testosterone at C-17 does not involve transient incorporation of an oxygen atom in this position. A mechanism is suggested whereby cytochrome P-450 acts as a peroxidase in the formation of androstenedione.     

Responses to insulin by two forms of rat hepatic microsomal cytochrome P-450 that undergo major (RLM6) and minor (RLM5b) elevations in diabetes

Total cytochrome P-450 levels rise in diabetic rats. Two specific forms of cytochrome P-450 that are elevated have been isolated from liver microsomes of streptozotocin-induced idabetic male rats. One enzyme, termed RLM6, metabolizes aniline and acetol, but not testosterone, in a reconstituted system with NADPH-cytochrome P-450 reductase. RLM6 is isolated as a high spin cytochrome with a minimum molecular weight of 53,500. It has a unique amino-terminal amino acid sequence lacking methionine at the amino-terminal position. Polyclonal antibodies to RLM6 recognized most other forms of cytochrome P-450 in Western blots, but could be made monospecific by adsorption to cross-reacting proteins coupled to Sepharose 4B. Using the monospecific antibodies, RLM6 was estimated to be present in microsomes of untreated male rats at 0.04 nmol/mg protein (5% of total P-450). In chronically diabetic rats this level rose to 0.35 nmol/mg protein and 24% of the P-450 content. Immunoreactive protein of molecular weight identical to RLM6 was elevated in microsomes of non-diabetic rats treated with ethanol, acetone, or isoniazid as well as in rats starved for 48 h. Insulin treatment of diabetic rats for 1 week lowered the immunologically detectable levels of RLM6 to levels found in the untreated rat. The other form of cytochrome P-450, RLM5b, does not metabolize aniline and only poorly metabolizes acetol and testosterone. This 52.5-kDa protein is isolated as a predominantly (60%) high spin enzyme. It has a unique NH2-terminal amino acid sequence with methionine as the terminal residue, and is present in untreated male rat liver microsomes at 0.16 nmol/mg protein. It is elevated in diabetes, like RLM6, but treatment with insulin for 1 week does not completely restore the microsomal content to that of the non-diabetic rat.     

Participation of two structurally related enzymes in rat hepatic microsomal androstenedione 7 alpha-hydroxylation.

Rat hepatic cytochrome P-450 form 3 (testosterone 7 alpha-hydroxylase; P-450 gene IIA1) and P-450 form RLM2 (testosterone 15 alpha-hydroxylase; P-450 gene IIA2) are 88% identical in primary structure, yet they hydroxylate testosterone with distinct and apparently unrelated regioselectivities. In this study, androstenedione and progesterone were used to assess the regioselectivity and stereospecificity of these two P-450 enzymes towards other steroid substrates. Although P-450 RLM2 exhibited low 7 alpha-hydroxylase activity with testosterone or progesterone as substrate (turnover number less than or equal to 1-2 nmol of metabolite/min per nmol of P-450), it did catalyse androstenedione 7 alpha-hydroxylation at a high rate (21 min-1) which exceeded that of P-450 3 (7 min-1). However, whereas P-450 3 exhibited a high specificity for hydroxylation of these steroids at the 7 alpha position (95-97% of total activity), P-450 RLM2 actively metabolized these compounds at four or more major sites including the nearby C-15 position, which dominated in the case of testosterone and progesterone. The observation that androstenedione is actively 7 alpha-hydroxylated by purified P-450 RLM2 suggested that this P-450 enzyme might make significant contributions to microsomal androstenedione 7 alpha-hydroxylation, an activity that was previously reported to be associated with immunoreactive P-450 3. Antibody inhibition experiments were therefore carried out in liver microsomes using polyclonal anti-(P-450 3) antibodies which cross-react with P-450 RLM2, and using a monoclonal antibody that is reactive with and inhibitory towards P-450 3 but not P-450 RLM2. P-450 3 was thus shown to catalyse only around 35% of the total androstenedione 7 alpha-hydroxylase activity in uninduced adult male rat liver microsomes, with the balance attributed to P-450 RLM2. The P-450-3-dependent 7 alpha-hydroxylase activity was increased to approximately 65% of the total in phenobarbital-induced adult male microsomes, and to greater than 90% of the total in untreated adult female rat liver microsomes. These observations are consistent with the inducibility of P-450 3 by phenobarbital and with the absence of P-450 RLM2 from adult female rat liver respectively. These findings establish that P-450 RLM2 and P-450 3 can both contribute significantly to microsomal androstenedione 7 alpha-hydroxylation, thus demonstrating that the 7 alpha-hydroxylation of this androgen does not serve as a specific catalytic monitor for microsomal P-450 3.     

Cytochrome P-450 hPCN3, a novel cytochrome P-450 IIIA gene product that is differentially expressed in adult human liver. cDNA and deduced amino acid sequence and distinct specificities of cDNA-expressed hPCN1 and hPCN3 for the metabolism of steroid hormones and cyclosporine

Immunoblotting analysis of human liver microsome preparations revealed that human cytochrome P-450 PCN1 (hPCN1, Mr approximately 52,000) was expressed in each of 40 individual specimens examined. In about 10-20% of the livers, an immunologically related protein having a lower electrophoretic mobility (Mr approximately 52,500) was also detected. A single liver was found that expressed only the lower mobility protein, designated hPCN3, and RNA isolated from this liver was used to construct a lambda gt11 library. The library was screened with an hPCN1 cDNA probe resulting in the isolation of a unique full-length cDNA that was sequenced and shown to encode hPCN3. The deduced amino acid sequence of this cDNA contained 502 residues, a calculated molecular mass of 57,115 daltons, and displayed 84% similarity with hPCN1. The deduced amino-terminal sequence of hPCN3 was identical to that of HFLa, a major cytochrome P-450 expressed in human fetal liver that is immunologically cross-reactive with several family III cytochrome P-450s. hPCN1 and hPCN3 cDNAs were expressed in Hep G2 cells using a vaccinia virus expression system and shown to encode active enzymes, both characterized by reduced CO-binding spectra with lambda max at 450 nm. Enzymatic analysis revealed that both cytochrome P-450s were similarly active in catalyzing oxidation of the calcium channel blocking drug nifedipine. Both enzymes also catalyzed 6 beta-hydroxylation of the steroid hormones testosterone, progesterone, and androstenedione, although hPCN1 exhibited several-fold higher expressed activity than hPCN3. Several minor oxidation products of these steroids (e.g. 15 beta-hydroxytestosterone), comprising up to approximately 20% of the total metabolites, were formed by hPCN1 but not hPCN3, indicating that hPCN3 is a more highly regiospecific monooxygenase catalyst with steroid substrates. Clear differences were also detected in their catalytic activities toward the immunosuppressive drug cyclosporine, with two hydroxylated metabolites (M1 and M17) and one demethylated metabolite (M21) formed by hPCN1 but only one metabolite (M1) formed by hPCN3. These studies establish that hPCN3 is a newly described cytochrome P-450 that is differentially expressed in the adult human population and that has overlapping substrate specificity compared to hPCN1 for metabolism of steroid and drug substrates.     

Differences in the spectral interactions between NADPH-cytochrome P-450 reductase and a series of cytochrome P-450 enzymes

The interaction between NADPH-cytochrome P-450 reductase and a series of cytochrome P-450 isozymes was investigated using UV-visible spectrophotometry. In the absence of substrate the interactions between the reductase and RLM3, RLM5, and RLM5a were tight, exhibiting sub-micromolar dissociation constants and resulted in type I spectra of varying magnitude from which the following increases in the proportion of high spin hemoprotein were calculated; RLM3 (7%), RLM5 (36%), RLM5a (6%), LM2 (29%), RLM2 (0%). Preincubation of LM2 with its type I substrate benzphetamine increased the affinity of the cytochrome for the reductase. Using initial estimates of the P-450 spin states in the absence of reductase in conjunction with the spectral binding data and equations relating these parameters to the microequilibria for the association of reductase with high or low spin P-450, RLM3, RLM5, RLM5a and LM2 were shown to bind significantly more tightly to high spin P-450. The relevance of this data to the understanding of spin state influence on P-450 reduction is discussed.     

Evidence for functional and structural multiplicity of pregnenolone-16 alpha-carbonitrile-inducible cytochrome P-450 isozymes in rat liver microsomes

Administration of pregnenolone-16 alpha-carbonitrile (PCN) to adult female rats caused a 2-fold increase in total liver microsomal cytochrome P-450 along with 5-7-fold increases in four in vitro monooxygenase activities considered diagnostic for the major PCN-inducible cytochrome P-450 isozyme. However, upon administration of chloramphenicol to PCN-treated rats, these monooxygenase activities could be resolved into three groups. Thus, the ability of the microsomes to convert triacetyloleandomycin to a metabolite that forms a spectral complex with the reduced heme iron was decreased by 80% by chloramphenicol, whereas only a 50% decrease was observed in the rate of conversion of (R)-warfarin to its 9,10-dehydro metabolite and in the rate of 6 beta-hydroxylation of androstenedione. More strikingly, the 10-hydroxylation of (R)-warfarin was actually enhanced 2-fold by the chloramphenicol treatment. Fractionation studies were carried out on liver microsomes from PCN-treated adult male rats, and two highly purified cytochromes P-450, referred to as PCNa and PCNb, were recovered. PCNb was found to be identical in the sequence of the first 15 amino acid residues with a PCN-inducible isozyme, the complete amino acid sequence of which has recently been deduced in another laboratory [Gonzalez, F. J., Nebert, D. W., Hardwick, J. P., & Kasper, C. B. (1985) J. Biol. Chem. 260, 7435-7441]. The other isozyme, PCNa, differed in amino acid sequence in three of the first 15 positions from PCNb. Upon immunoblot analysis, polyclonal antibodies raised to PCNb also recognized PCNa. Thus, the PCN-inducible family of rat liver cytochrome P-450 comprises at least two separate proteins.     

Adult male-specific and neonatally programmed rat hepatic P-450 forms RLM2 and 2a are not dependent on pulsatile plasma growth hormone for expression

Rat hepatic cytochrome P-450 form RLM2 is a testosterone 15 alpha-hydroxylase reported to be male-specific on the basis of purification studies (Jansson, I., Mole, J., and Schenkman, J. B. (1985) J. Biol. Chem. 260, 7084-7093). The sex dependence, developmental regulation, xenobiotic induction, and hormonal control of P-450 RLM2 expression were studied using P-450 form-specific immunochemical and catalytic assays. Polyclonal antibodies raised to rat hepatic P-450 3 (P-450 gene IIA1) were found to cross-react strongly with P-450 RLM2, but not with 10 other rat P-450 forms, suggesting that P-450 3 and P-450 RLM2 are highly conserved in primary structure. Western blotting of liver microsomes under conditions where P-450s 3 and RLM2 are resolved electrophoretically revealed that P-450 RLM2 is markedly induced at puberty in male rats, with no protein detected (less than or equal to 5% of adult male levels) in adult females or immature animals of either sex. A similar developmental dependence was observed for hepatic microsomal testosterone 15 alpha-hydroxylase activity, which was found to be catalyzed primarily by P-450 RLM2. P-450 RLM2 was resistant to induction by several xenobiotics and in the case of phenobarbital and beta-naphthoflavone, was suppressed by 50-60%. Studies on the steroid hormonal regulation of P-450 RLM2 revealed that its adult male-specific expression is imprinted (programmed) in response to neonatal testosterone exposure. Ovariectomy studies demonstrated that suppression by estrogen does not contribute significantly to the absence of P-450 RLM2 in adult female rats. Although the male-specific developmental induction of P-450 RLM2 in response to neonatal testosterone is strikingly similar to that of P-450 2c (testosterone 2 alpha/16 alpha-hydroxylase; gene IIC11), P-450 RLM2 expression is not dependent on the pulsatile pituitary growth hormone secretion required for P-450 2c synthesis. Rather, hypophysectomy of adult male rats increased P-450 RLM2 and its associated testosterone 15 alpha-hydroxylase activity by 50-100%.(ABSTRACT TRUNCATED AT 400 WORDS)     

Inverse relationship between cytochrome P-450 phosphorylation and complexation with cytochrome b5

Cytochrome P-450 LM2 purified from rabbit liver microsomes has been shown to be a substrate for cAMP-dependent protein kinase. Cytochrome b5, in contrast, was a very poor substrate for cAMP-dependent protein kinase, although it stimulated the activity of the kinase toward histone. When purified rabbit cytochrome b5 was mixed with purified LM2, phosphorylation of LM2 by cAMP-dependent protein kinase was inhibited approximately 80-90%. Recently, a functional covalent complex of cytochrome b5 and LM2 was prepared and purified to homogeneity (P.P. Tamburini and J.B. Schenkman (1987) Proc. Natl. Acad. Sci. USA 84, 11-15). When present as a covalent complex with cytochrome b5, the phosphorylation of LM2 in the complex by cAMP-dependent protein kinase was also inhibited about 80-90% relative to an equivalent amount of LM2 alone. On the other hand, when the LM2 was phosphorylated prior to interaction with cytochrome b5, the ability of the latter to perturb the spin equilibrium of LM2 and oxidation of p-nitroanisole by the LM2 was diminished to an extent comparable to the degree of phosphorylation. The results suggest either that the phosphorylation site on LM2 may be within the cytochrome b5 binding site or that phosphorylation and cytochrome b5 cause mutually exclusive conformational changes in LM2. In addition, eight different forms of cytochrome P-450 from the rat (RLM2, RLM3, fRLM4, RLM5, RLM5a, RLM5b, RLM6, and PBRLM5) were examined as potential substrates for cAMP-dependent protein kinase under the same conditions. Maximal phosphorylation of about 20 mol% was obtained with LM2, and about half as much with PBRLM5. The low extent of phosphorylation of LM2 was not due to the prior presence of phosphate on the enzyme since LM2, as isolated, contains less than 0.1 mol phosphate/mol of enzyme. The other forms of cytochrome P-450 tested showed little or no phosphorylation in vitro despite the presence of a cAMP-dependent protein kinase phosphorylation sequence on at least two of them.     

Regioselective hydroxylation of prostaglandins by constitutive forms of cytochrome P-450 from rat liver: formation of a novel metabolite by a female-specific P-450

Previous studies demonstrated that liver microsomes from untreated rats catalyze the omega, omega-1, and omega-2 hydroxylation of prostaglandins [K. A. Holm, R. J. Engell, and D. Kupfer (1985) Arch. Biochem. Biophys. 237, 477-489]. The current study examined the regioselectivity of hydroxylation of PGE1 and PGE2 by purified forms of P-450 from untreated male and female rat liver microsomes. PGE1 was incubated with a reconstituted system containing cytochrome P-450 RLM 2, 3, 5, 5a, 5b, 6, or f4, NADPH-P-450 reductase, and dilauroylphosphatidylcholine in the presence or absence of cytochrome b5. Among the P-450 forms examined, only RLM 5 (male specific), 5a (present in both sexes), and f4 (female specific) yielded high levels of PGE hydroxylation. With PGE1, RLM 5 catalyzed solely the omega-1 hydroxylation and 5a catalyzed primarily the omega-1 and little omega and omega-2 hydroxylation. By contrast, f4 effectively hydroxylated PGE1 and PGE2 at the omega-1 and at a novel site. Based on retention on HPLC and on limited mass fragmentation, we speculate that this site is omega-3 (i.e., 17-hydroxylation). Kinetic analysis of PGE1 hydroxylation demonstrated that the affinity of f4 for PGE1 is approximately 100-fold higher than that of RLM 5; the Km values for f4, monitoring 19- and 17-hydroxylation of PGE1, were about 10 microM. Surprisingly, cytochrome b5 stimulated the activity of RLM 5a and f4, but not that of RLM 5. Hydroxylation of PGE2 by RLM 5 was at the omega, omega-1, and omega-2 sites, demonstrating a lesser regioselectivity than with PGE1. These findings show that the constitutive P-450s differ dramatically in their ability to hydroxylate PGs, in their regioselectivity of hydroxylation, and in their cytochrome b5 requirement.     

Chemical characterization of protein-protein interactions between cytochrome P-450 and cytochrome b5

Native cytochrome b5 interacts with either RLM5 or LM2 to form tight equimolar complexes (Kd = 250 and 540 nM, respectively) in which the content of high spin cytochrome P-450 was substantially increased. Cytochrome b5 caused 3- and 7-fold increases in the binding affinities of RLM5 and LM2 for benzphetamine, respectively, and benzphetamine decreased the apparent Kd for cytochrome b5 binding. Upon formation of the ternary complex between cytochromes P-450, b5, and benzphetamine the percentage of cytochrome P-450 in the high spin state was increased from 28 to 74 (RLM5) and from 9 to 85 (LM2). Cytochrome b5 caused 13- and 7-fold increases in the rate of RLM5- and LM2-dependent p-nitroanisole demethylation, respectively. Amino-modified (ethyl acetimidate or acetic anhydride) cytochrome b5 produced results similar to those obtained above with native cytochrome b5. In contrast, modification of as few as 5 mol of carboxyl groups/mol of amidinated cytochrome b5 resulted in both a substantial loss of the spectrally observed interactions with either cytochrome P-450 LM2 or cytochrome P-450 RLM5, and in a loss of the cytochrome b5-mediated stimulation of p-nitroanisole demethylation catalyzed by either monooxygenase. In further studies, native and fully acetylated cytochromes b5 reoxidized carbonmonoxy ferrous LM2 at least 20 times faster than amidinated, carboxyl-modified cytochrome b5 derivatives. In contrast, amidination, or acetylation of amino groups, or amidination of amino groups plus methylamidination of the carboxyl groups did not appreciably slow the rate of reduction of the cytochrome b5 by NADPH-cytochrome P-450 reductase. Collectively, the results provide strong evidence for an essential role of cytochrome b5 carboxyl groups in functional interactions with RLM5 and LM2.     

Evidence that isoniazid and ethanol induce the same microsomal cytochrome P-450 in rat liver, an isozyme homologous to rabbit liver cytochrome P-450 isozyme 3a

Cytochrome P-450j has been purified to electrophoretic homogeneity from hepatic microsomes of adult male rats administered ethanol and compared to the corresponding enzyme from isoniazid-treated rats. The enzymes isolated from ethanol- and isoniazid-treated rats have identical chromatographic properties, minimum molecular weights, spectral properties, peptide maps, NH2-terminal sequences, immunochemical reactivities, and substrate selectivities. Both preparations of cytochrome P-450j have high catalytic activity in aniline hydroxylation, butanol oxidation, and N-nitrosodimethylamine demethylation with turnover numbers of 17-18, 37-46, and 15 nmol product/min/nmol of P-450, respectively. A single immunoprecipitin band exhibiting complete identity was observed when the two preparations were tested by double diffusion analysis with antibody to isoniazid-inducible cytochrome P-450j. Ethanol- and isoniazid-inducible rat liver cytochrome P-450j preparations have also been compared and contrasted with cytochrome P-450 isozyme 3a, the major ethanol-inducible isozyme from rabbit liver. The rat and rabbit liver enzymes have slightly different minimum molecular weights and somewhat different peptide maps but similar spectral, catalytic, and immunological properties, as well as significant homology in their NH2-terminal sequences. Antibody to either the rat or rabbit isozyme cross-reacts with the heterologous enzyme, showing a strong reaction of partial identity. Antibody against isozyme 3a specifically recognizes cytochrome P-450j in immunoblots of induced rat liver microsomes. Aniline hydroxylation catalyzed by the reconstituted system containing cytochrome P-450j is markedly inhibited (greater than 90%) by antibody to the rabbit protein. Furthermore, greater than 85% of butanol or aniline metabolism catalyzed by hepatic microsomes from ethanol- or isoniazid-treated rats is inhibited by antibody against isozyme 3a. Results of antibody inhibition studies suggest that cytochrome P-450j is induced four- to sixfold by ethanol or isoniazid treatment of rats. All of the evidence presented in this study indicates that the identical cytochrome P-450, P-450j, is induced in rat liver by either isoniazid or ethanol, and that this isozyme is closely related to rabbit cytochrome P-450 isozyme 3a.     

Selective reactivation of steroid hydroxylases following dissociation of the isosafrole metabolite complex with rat hepatic cytochrome P-450

In order to elucidate the isozyme specificity of complex formation between cytochrome P-450 and the isosafrole metabolite the effect of complex dissociation on different steroid hydroxylation pathways was studied in hepatic microsomal fractions. Isosafrole induction was found to increase the 16 beta- and 7 alpha-hydroxylation of androst-4-ene-3,17-dione approximately 2.8- and 1.7-fold, respectively, whereas the 16 alpha-hydroxylation pathway was decreased to about one-quarter of control activity; 6 beta-hydroxylation was unchanged from control activity. More striking changes were apparent following dissociation of the isosafrole metabolite from its complex with ferricytochrome P-450 by the steroid substrate. Thus an approximate fourfold elevation of 16 beta-hydroxylase activity was observed after displacement and 6 beta-hydroxylation increased about twofold; 7 alpha-hydroxylase activity was decreased to 0.75-fold of undisplaced activity and 16 alpha-hydroxylase activity was unchanged. These data provide convincing evidence that at least two forms of phenobarbital-inducible cytochrome P-450 (cytochromes P-450PB-B and P-450PB/PCN-E) are present to some extent in a catalytically inactive complexed state in isosafrole-induced rat hepatic microsomes. Furthermore, there is now evidence to suggest that the constitutive isozymes cytochrome P-450UT-A and cytochrome P-450UT-F are not complexed to any degree in hepatic microsomes from isosafrole-induced rats.     

Mechanism of interaction between cytochromes P-450 RLM5 and b5: evidence for an electrostatic mechanism involving cytochrome b5 heme propionate groups

The role of cytochrome b5 heme propionate groups in the functional interactions between cytochromes P-450 RLM5 and b5 has been investigated by comparing the capacity of RLM5 to interact with both native b5 and a b5 derivative in which the native heme was replaced with ferric protoporphyrin IX dimethyl ester (DME-b5). Both forms of b5 interacted with RLM5 causing an increase in the RLM5 spin state from 28 to 68% high-spin RLM5 at saturation, as judged using uv-visible spectrophotometry. However, DME-b5 exhibited a 7-fold weaker affinity for RLM5. The apparent dissociation constant (Kd) for the interaction between RLM5 and b5 was also shown to be a strong function of ionic strength, in a manner consistent with the involvement of electrostatic attraction in complex formation. Reconstitution of b5 into an RLM5-dependent monooxygenase system stimulated the p-nitroanisole demethylase rate about 25-fold and 7-ethoxycoumarin deethylase about 6-fold. DME-b5, however, produced only 30% of the stimulation of RLM5-dependent turnover of p-nitroanisole observed at equivalent concentrations of native b5 without a change in Km. With 7-ethoxycoumarin, turnover was 50% diminished. The diminished capacity of DME-b5 to stimulate RLM5-dependent substrate turnover was shown not to be due to impairment of electron flow between NADPH-cytochrome P-450 reductase and DME-b5, since the Km of reductase for DME-b5 is 2.5-fold lower, and the Vmax is actually increased, but rather to an impairment of some aspect of functional interaction between the DME-b5 and RLM5. The data show that complex formation between cytochrome P-450 and b5 involves electrostatic attraction mediated in part by cytochrome b5 heme propionate groups.     

Purification and characterization of cytochrome P-450-dependent arachidonic acid epoxygenase from human liver

A novel human liver cytochrome P-450 isozyme (P-450-AA), which catalyzes arachidonic acid epoxidation, has been purified to electrophoretic homogeneity from human liver. As judged spectrally, the newly described isozyme is low spin in the oxidized state, with a soret band at 415 nm and an increased maximum at 451 nm in the CO-difference spectrum. Cytochrome P-450-AA appeared homogeneous as judged by the appearance of a single band on sodium dodecyl sulfate-polyacrylamide gel electrophoresis with an estimated molecular weight of 53,100. Although cytochrome P-450-AA had a relatively low specific content of 10.8 nmol/mg, it possessed a high activity of arachidonic acid epoxidation. The P-450-AA oxidized arachidonic acid in a reconstituted system into the four regioisomeric epoxyeicosatrienoic acids (EETs) (5, 6-, 8, 9-, 11, 12-, 14, 15-EETs) at a rate of 2,010 pmol/nmol/min, a rate which is 37-fold higher than that observed with the crude microsomal preparation. Moreover, the purified cytochrome P-450-AA catalyzed the de-ethylation of 7-ethoxyresorufin at the rate of 2970 pmol/nmol/min, whereas other cytochrome P-450-dependent reactions were carried out at 23-2,000-fold lower rates and ranged between 0.3-130 pmol/nmol/min. The amino acid composition is different from that of other cytochrome P-450 isozymes. The NH2-terminal sequence of 20-amino acid residues was compared to that of LM2 and PB2-B2, the phenobarbital-induced forms in rabbit and rats, respectively. Comparison was also made with two forms of human cytochrome P-450, HLc and HLd. There were 7/20 identical residues for P-450-AA and LM2 and 4/20 for P-450-AA and PB2-B2. There were 2/20 identical residues for P-450-AA and HLd, and no identical residues were found for HLc. We conclude that the biologically active EETs, are formed by a distinct and unique P-450 isozyme from human liver and that arachidonic acid can serve as a screen for detection of the novel P-450 isozyme.     

Isolation of four forms of acetone-induced cytochrome P-450 in chicken liver by h.p.l.c. and their enzymic characterization.

The purpose of this study was to purify and characterize the forms of cytochrome P-450 induced in chicken liver by acetone or ethanol. Using high performance liquid ion-exchange chromatography, we were able to isolate at least four different forms of cytochrome P-450 which were induced by acetone in chicken liver. All four forms of cytochrome P-450 proved to be distinct proteins, as indicated by their N-terminal amino acid sequences and their reconstituted catalytic activities. Two of these forms, also induced by glutethimide in chicken embryo liver, appeared to be cytochromes P450IIH1 and P450IIH2. Both of these cytochromes P-450 have identical catalytic activities towards benzphetamine demethylation. However, they differ in their abilities to hydroxylate p-nitrophenol and to convert acetaminophen into a metabolite that forms a covalent adduct with glutathione at the 3-position. Another form of cytochrome P-450 induced by acetone is highly active in the hydroxylation of p-nitrophenol and in the conversion of acetaminophen to a reactive metabolite, similar to reactions catalysed by mammalian cytochrome P450IIE. Yet the N-terminal amino acid sequence of this form has only 30-33% similarity with cytochrome P450IIE purified from rat, rabbit and human livers. A fourth form of cytochrome P-450 was identified whose N-terminal amino acid sequence and enzymic activities do not correspond to any mammalian cytochromes P-450 reported to be induced by acetone or ethanol.     

Further characterization of RLM2 and comparison with a related form of cytochrome P-450, RLM2b

We have extended the characterization of RLM2, a constitutive form of rat liver cytochrome P-450, using immunochemical means to quantitate its presence in microsomes, to follow its development in maturing male and female rats, and to determine its response to prototypical P-450 inducers. In addition, RLM2 is compared to RLM2b, a form of P-450 with similar migration on SDS-PAGE and NH2-terminal amino acid sequence. RLM2b is expressed in both sexes at a level of 0.08 nmol/mg microsomal protein at 2 weeks of age. In female rats, this level is unchanged with maturation. However, in the male, the level declined with maturation to reach 0.02 nmol/mg protein by 12 weeks of age. RLM2 is a male-specific form of cytochrome P-450. Originally absent in the 2-week-old rat, it reached a level of 0.03 nmol/mg protein in the adult male, its appearance and increase coinciding with the onset of puberty. Both phenobarbital and 3-methylcholanthrene induced microsomal levels of RLM2b in the adult male and female rat. RLM2, however, was suppressed in the male rat, 58 and 42%, respectively, by the same treatments. RLM2b and RLM2 each catalyze a unique spectrum of hydroxytestosterone metabolites. RLM2b is highly site specific. In contrast, RLM2 produces several isomeric products in the same region of the testosterone molecule. Substitution of the acetyl group of progesterone for the 17-hydroxy group of testosterone did not alter the site specificity of RLM2b, but did alter it for RLM2, indicating, further, a difference in the active site conformation of the two enzymes. Although RLM2b and RLM2 responded differently to inducers and to a changing physiology during maturation, and were functionally quite distinct, the proteins showed a high degree of immunologic relatedness which is suggestive of significant structural similarities. Structural differences do exist, however, as alpha-chymotryptic digestion formed a number of peptide fragments that differed between the two proteins.     

Studies on the rate-determining factor in testosterone hydroxylation by rat liver microsomal cytochrome P-450: evidence against cytochrome P-450 isozyme:isozyme interactions

The aim of the present study was to examine a recent proposal that inhibitory isozyme:isozyme interactions explain why membrane-bound isozymes of rat liver microsomal cytochrome P-450 exert only a fraction of the catalytic activity they express when purified and reconstituted with saturating amounts of NADPH-cytochrome P-450 reductase and optimal amounts of dilauroylphosphatidylcholine. The different pathways of testosterone hydroxylation catalyzed by cytochromes P-450a (7 alpha-hydroxylation), P-450b (16 beta-hydroxylation), and P-450c (6 beta-hydroxylation) enabled possible inhibitory interactions between these isozymes to be investigated simultaneously with a single substrate. No loss of catalytic activity was observed when purified cytochromes P-450a, P-450b, or P-450c were reconstituted in binary or ternary mixtures under a variety of incubation conditions. When purified cytochromes P-450a, P-450b, and P-450c were reconstituted under conditions that mimicked a microsomal system (with respect to the absolute concentration of both the individual cytochrome P-450 isozyme and NADPH-cytochrome P-450 reductase), their catalytic activity was actually less (69-81%) than that of the microsomal isozymes. These results established that cytochromes P-450a, P-450b, and P-450c were not inhibited by each other, nor by any of the other isozymes in the liver microsomal preparation. Incorporation of purified NADPH-cytochrome P-450 reductase into liver microsomes from Aroclor 1254-induced rats stimulated the catalytic activity of cytochromes P-450a, P-450b, and P-450c. Similarly, purified cytochromes P-450a, P-450b, and P-450c expressed increased catalytic activity in a reconstituted system only when the ratio of NADPH-cytochrome P-450 reductase to cytochrome P-450 exceeded that normally found in liver microsomes. These results indicate that the inhibitory cytochrome P-450 isozyme:isozyme interactions described for warfarin hydroxylation were not observed when testosterone was the substrate. In addition to establishing that inhibitory interactions between different cytochrome P-450 isozymes is not a general phenomenon, the results of the present study support a simple mass action model for the interaction between membrane-bound or purified cytochrome P-450 and NADPH-cytochrome P-450 reductase during the hydroxylation of testosterone.     

NH2-terminal sequence analyses of four rat hepatic microsomal cytochromes P-450

Cytochromes P-450f, P-450g, P-450h, and P-450i are four hepatic microsomal hemoproteins that have been purified from adult rats. Whereas cytochromes P-450g and P-450h appear to be male-specific hemoproteins, cytochrome P-450i is apparently a female-specific enzyme purified from untreated adult female rats. Cytochrome P-450f has been purified from adult male and female rats with equivalent recoveries. Amino-terminal sequence analyses of the first 15-20 amino acid residues of each of these cytochromes P-450 has been accomplished in the current investigation. Each protein possesses a hydrophobic leader sequence consisting of 65-87% hydrophobic amino acids, and only one charged amino acid (Asp) in the amino-terminal region. Although differences in the amino-terminal sequences of cytochromes P-450f, P-450g, P-450h, and P-450i are identified, these hemoproteins all begin with Met-Asp, and marked structural homology is observed among certain of these enzymes. Cytochromes P-450g and P-450h, two male-specific proteins, have 11-12/15 identical residues with cytochrome P-450i, a female-specific isozyme. Cytochromes P-450f and P-450h have 16/20 identical amino-terminal residues. Only limited sequence homology is observed between the amino-terminal sequences of cytochromes P-450f-i compared to rat liver cytochromes P-450a-e. The results demonstrate that cytochromes P-450f, P-450g, P-450h, and P-450i are isozymic to each other and five additional rat hepatic microsomal cytochrome P-450 isozymes (P-450a-e).     

In vivo and in vitro effects of a new macrolide antibiotic roxithromycin on rat liver cytochrome P-450: comparison with troleandomycin and erythromycin

The effects of a new macrolide antibiotic (Roxithromycin) and one of its major metabolite (RU 39001) on rat hepatic drug metabolizing enzymes were compared to those of erythromycin, erythralosamine and troleandomycin (TAO) both in vitro and in vivo. In contrast to erythromycin, erythralosamine and TAO, roxithromycin and its metabolite RU 39001 exhibit: (i) a very poor affinity for rat liver cytochrome P-450, (ii) an unability to be metabolized into a stable inhibitory metabolite-cytochrome P-450 complex and (iii) a decreased ability to induce liver cytochrome P-450 PCNE, an isozyme implicated in drug associations involving some macrolide antibiotics.     

Regioselective progesterone hydroxylation catalyzed by eleven rat hepatic cytochrome P-450 isozymes

Quantitative high-pressure liquid chromatographic assays were developed that separate progesterone and 17 authentic monohydroxylated derivatives. The assays were utilized to investigate the hydroxylation of progesterone by 11 purified rat hepatic cytochrome P-450 isozymes and 8 different rat hepatic microsomal preparations. In a reconstituted system, progesterone was most efficiently metabolized by cytochrome P-450h followed by P-450g and P-450b. Seven different monohydroxylated progesterone metabolites were identified. 16 alpha-Hydroxyprogesterone, formed by 8 of the 11 isozymes, was the only detectable metabolite formed by cytochromes P-450b and P-450e. 2 alpha-Hydroxyprogesterone was formed almost exclusively by cytochrome P-450h, and 6 alpha-hydroxyprogesterone and 7 alpha-hydroxyprogesterone were only formed by P-450a. 6 beta-hydroxylation of progesterone was catalyzed by four isozymes with cytochrome P-450g being the most efficient, and 15 alpha-hydroxyprogesterone was formed as a minor metabolite by cytochromes P-450g, P-450h, and P-450i. None of the isozymes catalyzed 17 alpha-hydroxylation of progesterone, and only cytochrome P-450k had detectable 21-hydroxylase activity. 16 alpha-Hydroxylation catalyzed by cytochrome P-450b was inhibited in the presence of dilauroylphosphatidylcholine (1.6-80 microM), while this phospholipid either stimulated (up to 3-fold) or had no effect on the metabolism of progesterone by the other purified isozymes. Results of microsomal metabolism in conjunction with antibody inhibition experiments indicated that cytochromes P-450a and P-450h were the sole 7 alpha- and 2 alpha-hydroxylases, respectively, and that P-450k or an immunochemically related isozyme contributed greater than 80% of the 21-hydroxylase activity observed in microsomes from phenobarbital-induced rats.