Multiplicity of heme oxygenase isozymes. HO-1 and HO-2 are different molecular species in rat and rabbit

We report on the detection and characterization of two forms of heme oxygenase in rabbit tissues and provide data suggesting that heme oxygenases in rat and rabbit are not identical and constitute a group of heterogenous proteins. Certain molecular properties, however, are shared by the isozymes in rat and rabbit; the predominant form of the enzyme in control liver and testis is HO-2, in the liver HO-1 is the inducible form, and in the brain HO-1 is not detectable. HO-1 was purified from liver of rabbits treated with bromobenzene to near homogeneity with a specific activity of 8,270 nmol of bilirubin/mg/h and compared with a homogenous preparation of rat HO-1 with a specific activity of 6,220, also obtained from bromobenzene-treated animals. Rat and rabbit HO-1, on sodium dodecyl sulfate-polyacrylamide gel, had molecular weights of 30,000 and 30,700, respectively. Rabbit HO-2 was partially purified from testis to a specific activity of 386 nmol of bilirubin/mg/h and compared with a purified preparation of rat testis HO-2 with a specific activity of 5,700. Using Western immunoblotting, rabbit HO-2 displayed intense cross-reactivity with antibody raised in rabbit to sodium dodecyl sulfate-denatured rat HO-2, and had a substantially larger molecular weight than the rat HO-2 (42,000 versus 36,000). Rabbit HO-1 did not cross-react with antibody to rat HO-1 which was also raised in rabbit. Unlike the rat enzymes, rabbit HO-1 and HO-2 did not differ in thermolability. It is speculated that HO-1 in rat and rabbit, and possibly HO-2, have evolved from divergent evolution of a common ancestral gene(s).     

Expression and characterization of full-length human heme oxygenase-1: the presence of intact membrane-binding region leads to increased binding affinity for NADPH cytochrome P450 reductase

Heme oxygenase-1 (HO-1) is the chief regulatory enzyme in the oxidative degradation of heme to biliverdin. In the process of heme degradation, HO-1 receives the electrons necessary for catalysis from the flavoprotein NADPH cytochrome P450 reductase (CPR), releasing free iron and carbon monoxide. Much of the recent research involving heme oxygenase has been done using a 30 kDa soluble form of the enzyme, which lacks the membrane binding region (C-terminal 23 amino acids). The goal of this study was to express and purify a full-length human HO-1 (hHO-1) protein; however, due to the lability of the full-length form, a rapid purification procedure was required. This was accomplished by use of a glutathione-s-transferase (GST)-tagged hHO-1 construct. Although the procedure permitted the generation of a full-length HO-1, this form was contaminated with a 30 kDa degradation product that could not be eliminated. Therefore, attempts were made to remove a putative secondary thrombin cleavage site by a conservative mutation of amino acid 254, which replaces arginine with lysine. This mutation allowed the expression and purification of a full-length hHO-1 protein. Unlike wild type (WT) HO-1, the R254K mutant could be purified to a single 32 kDa protein capable of degrading heme at the same rate as the WT enzyme. The R254K full-length form had a specific activity of approximately 200-225 nmol of bilirubin h-1 nmol-1 HO-1 as compared to approximately 140-150 nmol of bilirubin h-1 nmol-1 for the WT form, which contains the 30 kDa contaminant. This is a 2-3-fold increase from the previously reported soluble 30 kDa HO-1, suggesting that the C-terminal 23 amino acids are essential for maximal catalytic activity. Because the membrane-spanning domain is present, the full-length hHO-1 has the potential to incorporate into phospholipid membranes, which can be reconstituted at known concentrations, in combination with other endoplasmic reticulum resident enzymes.     

Purification and characterization of the major constitutive form of testicular heme oxygenase. The noninducible isoform

Recently we reported on the presence of two isoforms of heme oxygenase in rat liver microsomes, referred to as HO-1 and HO-2, and that only HO-1 is inducible (Maines, M. D., Trakshel, G. M., and Kutty, R. K. (1986) J. Biol. Chem. 261, 411-419). Presently we report on the detection of two isoforms of the enzyme in rat testis and purification to near homogeneity of the noninducible isoform, HO-2. A comparative characterization of the liver HO-1 and the testicular HO-2 is also provided. The relative abundance of the isoforms in the two organs was dissimilar. In the testis, the predominant form was HO-2, and only minute amounts of HO-1 were detected. In the liver, however, a 1:2 ratio of HO-1 to HO-2 was noted. The activity of HO-2 in both organs was refractory to cadmium, an inducer of the hepatic HO-1. Under nondenaturing electrophoresis conditions, HO-2 showed a higher mobility than HO-1; on a sodium dodecyl sulfate-polyacrylamide gel, HO-2 displayed a higher monomeric Mr. The apparent Mr values for HO-2 and HO-1 were 36,000 and 30,000, respectively. The isoforms differed in immunochemical properties. Antiserum to the liver HO-1 did not recognize the testicular HO-2 when examined by double immunodiffusion or by Western immunoblotting. HO-2 was more sensitive to heat inactivation than HO-1. When exposed at 65 degrees C (10 min), 70% of HO-1 activity was retained; however, nearly 80% of HO-2 activity was lost. The apparent Km values for heme for HO-1 and HO-2 were 0.24 and 0.40 microM, respectively. HO-1 and HO-2 had similar requirements for cofactor and flavoprotein reductase and were inhibited by heme-ligands (CO, KCN, NaN3). HO-2 utilized as substrate, Fe-protoporphyrin, Fe-hematoporphyrin, and Fe-hematoporphyrin acetate; it did not degrade intact purified rat liver cytochromes b5 and P-450 LM2, catalase, cytochrome c, hemoglobin, or myoglobin.     

Effect of calmodulin on aldosterone synthesis by a cytochrome P-45011 beta-reconstituted system from bovine adrenocortical mitochondria

Bovine adrenocortical calmodulin was purified and its general properties were examined. The latter were similar to those of bovine brain calmodulin. When added to a cytochrome P-450(11)beta-reconstituted system in the presence of dilauroylphosphatidylcholine, calmodulin decreased the rate of aldosterone production from corticosterone from 0.8 to 0.1 nmol/(min X nmol P-450), while it increased the rate of 18-hydroxycorticosterone production from 1.8 to 4.6 nmol/(min X nmol P-450). This effect of calmodulin on steroid production was maximum at a concentration of 1 microM, when 1 microM cytochrome P-450(11)beta was used. The effect was dependent on the presence of Ca2+, and maximal response was observed at less than 1 microM Ca2+. There was essentially no difference in the effect when bovine brain calmodulin was used. Calmodulin induced a change in the activity of cytochrome P-450(11)beta in the presence of a wide concentration range of corticosterone as a substrate. As for 18-hydroxycorticosterone production, calmodulin increased both the maximal activity and the apparent Km for corticosterone, but it decreased the apparent Km for adrenodoxin. Adrenodoxin at a concentration of less than 20 microM did not fully abolish the effect of calmodulin. A small type I difference spectrum appeared when calmodulin was added to cytochrome P-450(11)beta. The difference spectrum increased significantly in the presence of both Ca2+ and adrenodoxin. These results suggest that calmodulin interacts with cytochrome P-450(11)beta in the presence of adrenodoxin and then modulates the activity of aldosterone synthesis catalyzed by cytochrome P-450(11) beta.     

Rat liver cytochrome P-450b, P-420b, and P-420c are degraded to biliverdin by heme oxygenase

In this report we provide data, for the first time, demonstrating the conversion of the heme moiety of certain cytochrome P-450 and P-420 preparations, to biliverdin, catalyzed by heme oxygenase. We have used purified preparations of cytochromes P-450c, P-450b, P-450/P-420c, or P-450/P-420b as substrates in a heme oxygenase assay system reconstituted with heme oxygenase isoforms, HO-2 or HO-1, NADPH-cytochrome c (P-450) reductase, biliverdin reductase, NADPH, and Emulgen 911. With cytochrome P-450b or P-450/P-420b preparations, a near quantitative conversion of degraded heme to bile pigments was observed. In the case of cytochrome P-450/P-420c approximately 70% of the degraded heme was accounted for as bilirubin but only cytochrome P-420c was appreciably degraded. The role of heme oxygenase in this reaction was supported by the following observations: (i) bilirubin formation was not observed when heme oxygenase was omitted from the assay system; (ii) the rate of degradation of the heme moiety was at least threefold greater with heme oxygenase and NADPH-cytochrome c (P-450) reductase than that observed with reductase alone; and (iii) the presence of Zn- or Sn-protoporphyrins (2 microM), known competitive inhibitors of heme oxygenase, resulted in 70-90% inhibition of bilirubin formation.     

Characterization of two heme oxygenase isoforms in rat spleen: comparison with the hematin-induced and constitutive isoforms of the liver

Two isoforms of heme oxygenase, designated as HO-1 and HO-2, were identified in rat spleen. The most abundant form was HO-1, wherein a relative ratio of about 5:1 of HO-1 to HO-2 was detected. The splenic HO-1 and HO-2 were immunochemically similar to the purified isoforms obtained from the liver and the testis. Moreover, the elution properties of splenic HO-1 as well as those of the constitutive liver HO-1 and the hematin-induced liver HO-1 on a DEAE-sephacel column were similar. However, the splenic HO-1 activity could not be induced by hematin. It is suggested that in the spleen heme oxygenase activity is maintained in the induced state as the result of constant exposure to hemoglobin released in the course of disruption of senescent erythrocytes.     

Variation in hepatic microsomal cytochrome P-450 1 concentration among untreated rabbits alters the efficiency of estradiol hydroxylation

The metabolism of 17 beta-estradiol was examined using both rabbit liver microsomes and highly purified forms of rabbit liver microsomal cytochrome P-450. The predominant microsomal metabolite of 17 beta-estradiol is the 2-hydroxylated product. 2-Hydroxyestradiol is also the principal metabolite in reconstitution experiments in which P-450 1 exhibits the greatest Vmax, ca. 6 mol min-1 mol P-450 1(-1), vs less than 0.6 mol min-1 mol P-450(-1) for forms 2, 3b-, 3b+, 3c, 4, and 6. In addition P-450 1 has the lowest Km, ca. 2 microM. This suggested that microsomes which differ in their content of P-450 1 would also differ in the kinetic parameters characterizing the 2-hydroxylation of 17 beta-estradiol. Microsomes containing low amounts of P-450 1, less than 0.1 nmol/mg protein, exhibit a low-efficiency (Vmax/Km) 2-hydroxylase activity. Microsomes containing elevated concentrations of P-450 1, greater than 0.3 nmol/mg protein, exhibit a substrate dependence suggestive of an additional high-efficiency enzyme. The latter is specifically inhibited by a monoclonal antibody that recognizes P-450 1. These results indicate that the elevated expression of P-450 1 in microsomes leads to a marked increase in the apparent first-order rate constant for the 2-hydroxylation of 17 beta-estradiol, as it does for the 21-hydroxylation of progesterone. This should have a marked effect on the metabolism of these two steroid hormones at concentrations that are likely to occur in vivo.     

Detection of two heme oxygenase isoforms in the human testis

This study shows heme oxygenase multiplicity is common to rat and human tissues. The isozymes in man and rat, however, are heterogenous proteins that share certain characteristics. Two forms of heme oxygenase, HO-1 and HO-2, were identified in human testis. HO-2 form was the prevalent form. Human and rat HO-1 differed in chromatographic behavior and molecular weight; human HO-1 was a larger molecule (35,400 vs 30,000). The two forms, however, were similar in that immunochemically human HO-1 exhibited reactivity toward antibody to rat HO-1. Human and rat HO-2 also were dissimilar in chromatographic behavior and showed only a weak immunological cross-reactivity. Human and rat HO-1 were essentially the same size. As in rat organs, the microsomal cytochrome P-450 content in human testis was reciprocal to heme oxygenase activity.     

Correlation of the kinetics of electron transfer activity of various eukaryotic cytochromes c with binding to mitochondrial cytochrome c oxidase.

1. A detailed study of cytochrome c oxidase activity with Keilin-Hartree particles and purified beef heart enzyme, at low ionic strength and low cytochrome c concentrations, showed biphasic kinetics with apparent Km1 = 5 x 10(-8) M, and apparent Km2 = 0.35 to 1.0 x 10(-6) M. Direct binding studies with purified oxidase, phospholipid-containing as well as phospholiptaining aid-depleted, demonstrated two sites of interaction of cytochrome c with the enzyme, with KD1 less than or equal to 10(-7) M, and KD2 = 10(-6) M. 2. The maximal velocities as low ionic strength increased with pH and were highest above ph 7.5. 3. The presence and properties of the low apparent Km phase of the kinetics were strongly dependent on the nature and concentration of the anions in the medium. The multivalent anions, phosphate, ADP, and ATP, greatly decreased the proportion of this phase and similarly decreased the amount of high affinity cytochrome c-cytochrome oxidase complex formed. The order of effectiveness was ATP greater than ADP greater than P1 and since phosphate binds to cytochrome c more strongly than the nucleotides, it is concluded that the inhibition resulted from anion interaction with the oxidase. 4mat low concentrations bakers' yeast iso-1, bakers' yeast iso-1, horse, and Euglena cytochromes c at high concentrations all attained the same maximal velocity. The different proportions of low apparent Km phase in the kinetic patterns of these cytochromes c correlated with the amounts of high affinity complex formed with purified cytochrome c oxidase. 5. The apparent Km for cytochrome c activity in the succinate-cytochrome c reductase system of Keilin-Hartree particles was identical with that obtained with the oxidase (5 x 10(-8) M), suggesting the same site serves both reactions. 6. It is concluded that the observed kinetics result from two catalytically active sites on the cytochrome c oxidase protein of different affinities for cytochrome c. The high affinity binding of cytochrome c to the mitochondrial membrane is provided by the oxidase and at this site cytochrome c can be reduced by cytochrome c1. Physiological concentrations of ATP decrease the affinity of this binding to the point that interaction of cytochrome c with numerous mitochondrial pholpholipid sites can competitively remove cytochrome c from the oxidase. It is suggested that this effect of ATP represents a possible mechanism for the control of electron flow to the oxidase.     

Enzymatic properties of ubiquinol:cytochrome c2 oxidoreductase in situ in Rhodopseudomonas palustris membranes

The presence of ubiquinol:cytochrome c2 oxidoreductase was shown in the membranes from a photosynthetic bacterium, Rhodopseudomonas palustris. Some properties of the enzyme in situ were investigated. The optimal pH of this enzyme activity was 7.0 in the intact membranes. The activity was inhibited by both antimycin and myxothiazol. Maximal activity (Vmax) was 3-4 mol cytochrome c (c2) reduced/mol cytochrome c1.s. Apparent activity of the enzyme with horse heart cytochrome c as the electron acceptor decreased as the concentration of salts in the reaction mixture increased, whereas when R. palustris cytochrome c2 was used as the electron acceptor, the activity increased as the concentration of salts increased. Moreover, the activity of the enzyme did not depend on the species or concentration of anions but on both the concentration and valency of the cations of the salts. These salt effects were thought to be due to the change of effective concentration of cytochrome molecules caused by cations near the membrane surface, which was net negatively charged. Apparent Km for ubiquinol-1 was about 80 microM irrespective of the species of cytochrome and the presence of salts.     

Antiapoptotic role of heme oxygenase (HO) and the potential of HO as a target in anticancer treatment

Heme oxygenase-1 (HO-1) is an inducible enzyme that catalyzes oxidative degradation of heme to form biliverdin, carbon monoxide (CO), and free iron. Biliverdin is subsequently reduced to bilirubin by the enzyme biliverdin reductase. Increasing evidence has indicated the critical role of HO-1 in cytoprotection and more diverse biological functions. Induction of HO-1 by various chemical inducers that are primarily cell stress inducers or by HO-1 gene transfection confers a protective capacity to cultured cells as well as to cells in several in vivo animal models. In addition, HO-1-deficient mice exhibit a significant increase in susceptibility to tissue injury. The cytoprotective action of HO-1 seems to be mainly a function of the antiapoptotic effects of the enzyme. HO-1 is believed to exert this antiapoptotic action by multiple mechanisms: (a) decreased intracellular pro-oxidant levels, (b) increased bilirubin levels, and (c) elevated CO production. CO may produce an antiapoptotic effect by inhibiting both expression of p53 and release of mitochondrial cytochrome c. HO-1 may also be a target in antitumor therapy because the growth of most tumors depends on HO-1. Our preliminary studies with an HO inhibitor showed a promising antitumor effect. This preliminary work warrants continued investigation for possible novel anticancer chemotherapy.     

Crystallization of mitochondrial ubiquinol-cytochrome c reductase.

Ubiquinol-cytochrome c reductase of beef heart mitochondria was crystallized in the presence of decanoyl-N-methylglucamide, heptanetriol, and sodium chloride with poly(ethylene glycol) as precipitant. The largest crystal has dimensions of 4 x 2 x 1 mm. The crystalline enzyme is composed of 10 subunits. It contains 2.5 nmol of ubiquinone, 8.4 nmol of cytochrome b, 4.2 nmol of cytochrome c1, 4.2 nmol of iron-sulfur cluster, and 140 nmol of phospholipid per milligram of protein. Of the last, 36% is with diphosphatidylglycerol. The crystals are very stable in the cold and show full enzymatic activity when redissolved in aqueous solution. Absorption spectra of the redissolved crystals show a Soret to UV ratio of 0.88 and 1.01 in the oxidized and the reduced forms, respectively.     

Succinobucol induces apoptosis in vascular smooth muscle cells

Probucol inhibits the proliferation of vascular smooth muscle cells in vitro and in vivo, and the drug reduces intimal hyperplasia and atherosclerosis in animals via induction of heme oxygenase-1 (HO-1). Because the succinyl ester of probucol, succinobucol, recently failed as an antiatherogenic drug in humans, we investigated its effects on smooth muscle cell proliferation. Succinobucol and probucol induced HO-1 and decreased cell proliferation in rat aortic smooth muscle cells. However, whereas inhibition of HO-1 reversed the antiproliferative effects of probucol, this was not observed with succinobucol. Instead, succinobucol but not probucol induced caspase activity and apoptosis, and it increased mitochondrial oxidation of hydroethidine to ethidium, suggestive of the participation of H(2)O(2) and cytochrome c. Also, succinobucol but not probucol converted cytochrome c into a peroxidase in the presence of H(2)O(2), and succinobucol-induced apoptosis was decreased in cells that lacked cytochrome c or a functional mitochondrial complex II. In addition, succinobucol increased apoptosis of vascular smooth muscle cells in vivo after balloon angioplasty-mediated vascular injury. Our results suggest that succinobucol induces apoptosis via a pathway involving mitochondrial complex II, H(2)O(2), and cytochrome c. These unexpected results are discussed in light of the failure of succinobucol as an antiatherogenic drug in humans.     

Functional characterization of the mitochondrial cytochrome b-c1 complex: steady-state kinetics of the monomeric and dimeric forms

The QH2:cytochrome c oxidoreductase activity of the isolated bovine heart cytochrome b-c1 complex resolved into monomeric and dimeric form was titrated with three different inhibitors of electron transfer, antimycin, myxothiazol, and 5-n-undecyl-6-hydroxy-4,7-dioxobenzothiazole (UHDBT). In all cases one inhibitor molecule per cytochrome c1 was found necessary to block completely the activity of both molecular forms of the enzyme. The antimycin-sensitive cytochrome c reduction catalyzed by the b-c1 complex was also studied as a function of increasing concentrations of either cytochrome c or quinol. Double-reciprocal plots of the activity of the monomeric enzyme were found linear either when the concentration of cytochrome c or of quinol derivatives, 2,3-dimetoxy-5-methyl-6-decyl-1,4-benzoquinol (DBH), and 2-methyl-3-undecyl-1,4-naphthoquinol (UNH), was changed. Cytochrome c reductase activity of the dimeric b-c1 complex also showed a linear Lineweaver-Burk plot as a function of cytochrome c concentrations. In contrast to the monomeric enzyme, however, dimers of the b-c1 complex express a clear nonlinear kinetic behavior toward quinol derivatives, with two apparent Km values differing approximately by one order of magnitude (about 3-4 and about 20-30 microM). At saturating quinol concentrations the activity of the dimeric enzyme becomes two to three times higher than that of monomers. The nonlinear kinetic plots were found to be the same at different temperatures and different cytochrome c concentrations. The data suggest that although the monomer of the b-c1 complex appears to be the functional unit of the enzyme, the dimer is more active. A regulatory role of the dimerization process resulting in an increase of the electrons flux through the enzyme is postulated.     

Aldrin epoxidation catalyzed by purified rat-liver cytochromes P-450 and P-448. High selectivity for cytochrome P-450

Aldrin epoxidation was studied in monooxygenase systems reconstituted from purified rat liver microsomal cytochrome P-450 or P-448, NADPH-cytochrome c reductase, dilauroylphosphatidylcholine and sodium cholate. Cytochrome P-450, purified from hepatic microsomes of phenobarbital-treated rats, exhibited a high rate of dieldrin formation. The low enzyme activity observed in the absence of the lipid and sodium cholate was increased threefold by addition of dilauroylphosphatidylcholine and was further stimulated twofold by addition of sodium cholate. The apparent Km for aldrin in the complete system was 7 +/- 2 microM. SKF 525-A, at a concentration of 250 microM, inhibited aldrin epoxidation by 65%, whereas 7,8-benzoflavone had no inhibitory effect at concentrations up to 250 microM. Addition of ethanol markedly increased epoxidase activity. The increase was threefold in the presence of 5% ethanol. When cytochrome P-448 purified from hepatic microsomes of 3-methylcholanthrene-treated rats was used, a very low rate of epoxidation was observed which was less than 3% of the activity mediated by cytochrome P-450 under similar assay conditions. Enzyme activity was independent of the lipid factor dilauroylphosphatidylcholine. The apparent Km for aldrin was 27 +/- 7 microM. The modifiers of monooxygenase reactions, 7,8-benzoflavone, SKF 525-A and ethanol, inhibited the activity mediated by cytochrome P-448. The I50 was 0.05, 0.2 and 800 mM, respectively. These results indicate that aldrin is a highly selective substrate for cytochrome P-450 species present in microsomes of phenobarbital-treated animals and is a poor substrate for cytochrome P-448. The two forms of aldrin epoxidase can be characterised by their turnover number, their apparent Km and their sensitivity to modifiers, like 7,8-benzoflavone and ethanol.     

Complex formation and electron transfer between mitochondrial cytochrome c and flavocytochrome c552 from Chromatium vinosum

Flavocytochrome c552 from Chromatium vinosum catalyzes the oxidation of sulfide to sulfur using a soluble c-type cytochrome as an electron acceptor. Mitochondrial cytochrome c forms a stable complex with flavocytochrome c552 and may function as an alternative electron acceptor in vitro. The recognition site for flavocytochrome c552 on equine cytochrome c has been deduced by differential chemical modification of cytochrome c in the presence and absence of flavocytochrome c552 and by kinetic analysis of the sulfide:cytochrome c oxidoreductase activity of m-trifluoromethylphenylcarbamoyl-lysine derivatives of cytochrome c. As with mitochondrial redox partners, interaction occurs around the exposed heme edge at the "front face" of cytochrome c. However, the domain recognized by flavocytochrome c552 seems to extend to the right of the heme edge, whereas the site of interaction with mitochondrial cytochrome c oxidase and reductase is more to the left. Km but not Vmax of the electron transfer reaction with mitochondrial cytochrome c increases with increasing ionic strength. The correlation of chemical modification and ionic strength dependence data indicates that the electrostatic interaction between the two hemoproteins involves fewer ionic bonds than that with other redox partners of cytochrome c.     

Heme oxygenase-1 protects HepG2 cells against cytochrome P450 2E1-dependent toxicity

The inducible form of heme oxygenase (HO-1) is increased during oxidative injury and HO-1 is believed to be an important defense mechanism against such injury. Arachidonic acid (AA) and l-buthionine-(S,R)-sulfoximine (BSO), which lowers GSH levels, cause cytochrome P450 2E1 (CYP2E1)-dependent oxidative injuries in HepG2 cells (E47 cells). Treatment of E47 cells with 50 microM AA or 100 microM BSO for 48 h was recently shown to increase HO-1 mRNA, protein, and activity. The possible functional significance of this increase in protecting against CYP2E1-dependent toxicity was evaluated in the current study. The treatment with AA and BSO caused loss of cell viability (40 and 50%, respectively) in E47 cells. Chromium mesoporphyrin (CrMP), an inhibitor of HO activity, significantly potentiated this cytotoxicity. ROS production, lipid peroxidation, and the decline in mitochondrial membrane potential produced by AA and BSO were also enhanced in the presence of CrMP in E47 cells. Infection with an adenovirus expressing rat HO-1 protected E47 cells from AA toxicity, increasing cell viability and reducing LDH release. HO catalyzes formation of CO, bilirubin, and iron from the oxidation of heme. Bilirubin was not protective whereas iron catalyzed the AA toxicity. The carbon monoxide (CO) scavenger hemoglobin enhanced AA toxicity in E47 cells analogous to CrMP, whereas exposure to exogenous CO partially reduced AA toxicity and the enhanced AA toxicity by CrMP. Addition of exogenous CO to the cells inhibited CYP2E1 catalytic activity, as did overexpression of the rat HO-1 adenovirus. These results suggest that induction of HO-1 protects against CYP2E1-dependent toxicity and this protection may be mediated in part via production of CO and CO inhibition of CYP2E1 activity.     

Aromatase

Aromatase catalyzes the conversion of androgens to estrogens through a series of monooxygenations to achieve the 19-desmolation and aromatization of the neutral steroid ring-A structure. We have separated two forms of aromatase, a major (P2a) and a minor (P3) form, from human term placenta through solubilization and chromatography. Partially purified aromatase in each form was immunoaffinity chromatographed to give a single band (SDS-PAGE) cytochrome P-450 of 55 kDa, utilizing a mouse monoclonal anti-human placental aromatase cytochrome P-450 IgGi (MAb3-2C2) which is capable of suppressing placental aromatase activity. The purified cytochrome P-450 showed specific aromatase activity of 25-30 nmol/min per mg with Km of 20-30 nM for androstenedione on reconstitution with NADPH-cyt P-450 reductase and dilauroyl L-alpha-phosphatidylcholine. This one step represents a higher than 100-fold purification with maintenance of the same Km. The stability analysis showed a half-life of more than 5 yr for solubilized aromatase and 2 months for the aromatase cytochrome P-450 on storage at -90 degrees C. Contrary to the recent claim that estrogen biosynthesis by reconstituted human placental cytochrome P-450 is by trans-diaxial 1 alpha,2 beta-hydrogen elimination, all of our partially purified forms and reconstituted aromatase synthesized estrogens by cis-1 beta, 2 beta-hydrogen elimination. Use of purified aromatase and [19-3H3, 4-14C]androstenedione led us to discover a metabolic switching by aromatase to 2 beta-hydroxylation of androgen. Results of the MAb3-2C2 suppression of aromatase activity in different species and tissues including human, baboons, horses, cows, pigs and rats indicated the presence of various isozymes of aromatase.