Mitsukazu Kitada

Molecular cloning of monkey liver cytochrome P-450 cDNAs: similarity of the primary sequences to human cytochromes P-450.

Three cDNAs coding for monkey cytochrome P-450 (P450) 2C, 2E and 3A (MKmp13, MKj1 and MKnf2, respectively) were isolated from a lambda gt11 cDNA library of a liver from a 3-methylcholanthrene (3MC)-treated crab-eating monkey, using cDNA fragments for human P450 2C, 2E and 3A as respective probes. MKmp13 and MKnf2 were 1901 and 2032 bp long, containing entire coding regions for polypeptides of 490 and 503 residues, respectively. The deduced N-terminal amino acid sequences of MKmp13 and MKnf2 were identical with those of P450-MK1 and P450-MK2, which had been purified from liver microsomes of untreated and polychlorinated biphenyl (PCB)-treated crab-eating monkeys, respectively. MKj1 was 1508 bp long, encoding a polypeptide of 449 residues, which is presumed to lack N-terminal 45 residues as compared with the sequence for human P450 2E1. Northern blot analysis indicated that monkey P450 2C, 2E and 3A mRNAs were expressed constitutively in monkey livers. P450 2E and 3A mRNAs were induced by both 3MC and PCB, while P450 2C mRNA was induced only by PCB. The deduced amino acid sequences of four monkey cytochrome P-450 cDNAs, including P450 1A1 (MKah1) which we isolated previously, were more than 92% identical with those of corresponding human cytochrome P-450 cDNAs.

Isolation of a new human fetal liver cytochrome P450 cDNA clone: evidence for expression of a limited number of forms of cytochrome P450 in human fetal livers.

From a human fetal liver cDNA library, a new cDNA clone (lambda HFL10) was isolated using an antiserum to P450 HFLa, which has been isolated from livers of human fetuses. Cytochrome P450 cDNAs, namely lambda hPA6, lamda hP2-1, and lambda hPD4 which were highly homologous to cDNA clones, pHY13, Hp1-1, and phP450j, respectively, were also isolated from the cDNA library of human adult livers. Using these cDNA clones as probes together with Lambda HFL10, Northern blot analysis was conducted to determine whether all of these cytochromes were expressed in human fetal livers. The results clearly showed that only P450 HFL10 mRNA was detected in human fetal livers. This result supports the allegation that there is a much more limited number of forms of cytochrome P450 in human fetal livers than in adult livers.

Four Forms of Cytochrome P‐450 in Human Fetal Liver: Purification and Their Capacity to Activate Promutagens

Four forms of cytochrome P‐450 were separated and purified to electrophoretic homogeneity from human fetal livers. These forms of cytochrome P‐450, termed P‐450HFLa, P‐450HFLb, P‐450HFLc and P‐450HFLd, were distinguishable from each other in their molecular weights, spectral properties, immunochemical properties and mutagen‐producing activities from promntagens. The molecular weights of P‐450HFLa, b, c and d were estimated to be 51,500, 49,000, 51,500 and 50,000, respectively. Antibodies to P‐450HFLa recognized P‐450HFLc but not P‐450HFLb or d, and antibodies to rat P‐448‐H (P‐450IA2) cross‐reacted with P‐450HFLb but not with other forms of cytochrome P‐450. The N‐terminal amino acid sequence of P‐450HFLc was highly homologous, but not identical, to that of P‐450HFLa. Each form of cytochrome P‐450 catalyzed mutagenic activation of aflatoxin Bl (AFB1), 2‐amino‐3‐methylimidazo[4,5‐f]quinoline (IQ) and 2‐amino‐6‐methyldipyrido‐[l,2‐a:3′,2′‐d]imidazole (Glu‐P‐1) at different rates. P‐450 HFLa showed activities to produce mutagen(s) from AFB1, IQ and to a lesser extent from Glu‐P‐1. P‐450 HFLb activated IQ at a faster rate than did the other forms. P‐450 HFLc produced a mutagen from AFB1 and Glu‐P‐1 but not from IQ. P‐450 HFLd did not activate these promutagens at significant rates.

Bioactivation of monocrotaline by P-450 3A in rat liver

Monocrotaline (MCT) is bioactivated in liver cytochrome P-450s to MCT pyrrole (MCTP), which primarily injures the lung endothelium to result in the development of pulmonary hypertension (PH) in rats. However, whether there is a relation between the degree of PH and the activity of liver cytochrome P-450 to convert MCT to MCTP remains unclear. To examine the relation between these physiological and biochemical changes, we first measured the severity of MCT-induced (20 mg/kg) PH in male, female, castrated male, and phenobarbital (PB, liver P-450s inducer)-pretreated male rats. The degree of right ventricular hypertrophy was more severe in PB-pretreated male than in control male rats. It was also more severe in male than in either female or castrated male rats, suggesting that sex-specific P-450s could be involved in the metabolic pathways of MCT in the liver. Further to explore which of the isozymes (2A2, 2C11, and 3A) of P-450s in the liver is responsible for the bioactivation of MCT, we measured the rate of MCTP production in hepatic microsomes by a modified Mattocks method. Treatment of male rats with PB and pregnenolone 16alpha-carbonitrile (PCN), which is the specific inducer of P-450 3A, increased the rate of MCTP production, suggesting that P-450 3A may contribute to the conversion to pyrrole. Therefore we measured the amount of P-450 3A protein by immunoblotting and attempted to inhibit MCT metabolism by using antibodies to P-450 3A. P-450 3A was significantly induced by PCN (6.5-fold) and PB (4.6-fold) treatment and reduced by castration (0.38-fold). The amount of P-450 3A was closely correlated with the production of MCTP, and the conversion of MCT to MCTP was strongly inhibited by antibodies against P-450 3A. These results indicated that P-450 3A was predominantly responsible for the metabolism of MCT to MCTP in rat liver and suggested a tight linkage between the degree of PH and the activity of liver P-450 3A.

Stable expression of cytochrome P450IIIA7 cDNA in human breast cancer cell line MCF-7 and its application to cytotoxicity testing

A mammalian cell expression plasmid containing cytochrome P450IIIA7 complementary DNA was constructed. Breast cancer cells (MCF-7) were transfected with the plasmid and neomycin-resistant selection marker plasmid. We established three cell lines, termed M13, M21, and M27, which expressed the cytochrome P450IIIA7 as examined by RNA blot and immunoblot analyses. These cell lines showed 8- to 10-fold higher sensitivity against aflatoxin B1 compared to parental MCF-7 cells, suggesting that cytochromes P450IIIA7 expressed in the cells were responsible for the production of the cytotoxic metabolite of aflatoxin B1.

Purification of cytochrome P-450 from polychlorinated biphenyl-treated crab-eating monkeys: high homology to a form of human cytochrome P-450

Cytochrome P-450, designated as P-450-MK2, was purified to an electrophoretic homogeneity from polychlorinated biphenyl (PCB)-treated female crab-eating monkeys. P-450-MK2 catalyzed nifedipine and nilvadipine oxidations, at a rate comparable to human P-450-HM1. The N-terminal amino acid sequence of P-450-MK2 was highly homologous to those of P-450-HM1 and NF 25. The antibodies to P-450-HM1 recognized P-450-MK2 and effectively inhibited the activity of testosterone 6 beta-hydroxylase in monkey liver microsomes. These results suggest that a form of cytochrome P-450 corresponding to human P-450-HM1 or P-450NF which belongs to the P450 III gene family is also present in liver microsomes of crab-eating monkeys.

Immunochemical similarity of P-450 HFLa, a form of cytochrome P-450 in human fetal livers, to a form of rat liver cytochrome P-450 inducible by macrolide antibiotics

A protein immunochemically related to P-450 HFLa, a form of cytochrome P-450 purified from human fetal livers, was detected in rat liver microsomes. The content of the immunoreactive protein in rat liver microsomes was increased by treatments with phenobarbital, pregnenolone 16 alpha-carbonitrile (PCN), erythromycin, erythromycin estolate, and oleandomycin but not with 3-methylcholanthrene, imidazole, ethanol, isosafrole, josamycin, midecamycin, or miocamycin. The activity of erythromycin N-demethylase correlated with the content of the immunoreactive protein in rat liver microsomes (r = 0.72). In addition, anti-P-450 HFLa IgG inhibited erythromycin N-demethylase in liver microsomes from erythromycin- or oleandomycin-pretreated rats. Furthermore, the content of the immunoreactive protein highly correlated with that of P-450 PB-1, which is distinct from Waxmans terminology, and is one of the forms of PCN-inducible cytochrome P-450s (r = 0.95). From these results and the results reported so far, it seems possible that P-450 HFLa is one of the forms of cytochrome P-450 inducible by glucocorticoids.

Interspecies homology of cytochrome P-450: toxicological significance of cytochrome P-450 cross-reactive with anti-rat P-448-H antibodies in liver microsomes from dogs, monkeys and humans

The mutagenic activation of various promutagens by liver microsomes from dogs, monkeys and humans was investigated. Dog liver microsomes efficiently catalyzed the mutagenic activation of Trp-P-2 and Glu-P-1 followed by IQ and AAF. Monkey liver microsomes were most active in the activation of IQ followed by Glu-P-1, AAF and Trp-P-2. Although there were remarkable individual differences, human liver microsomes were found to be most active in the mutagenic activation of IQ followed by Trp-P-2, Glu-P-1 and AAF. Antibodies against rat P-448-H inhibited the mutagenic activation of Glu-P-1, Trp-P-2 and IQ in rat and dog liver microsomes, and Glu-P-1 and Trp-P-2 in monkey liver microsomes. The activation of Glu-P-1 and IQ in human liver microsomes was also strongly inhibited by anti-P-448-H antibodies. The amounts of cytochrome P-450 cross-reactive with anti-P-448-H antibodies in human liver microsomes highly correlated with the capacity to activate Glu-P-1, Trp-P-2 and IQ but not AAF.

Toxicological sifnificance of dog liver cytochrome P-450: examination with the enzyme expressed in Saccharomycees cerevisiae using recombinant expression plasmid

A complementary DNA (cDNA) coding for a form of beagle dog cytochrome P-450 (Dah1), which is the orthologue to the CYP1A1 cDNA of rat, mouse and human, was inserted between the alcohol dehydrogenase (ADH) promoter and terminator regions of the yeast expression vector pAAH5. On introduction of the resulting recombinant plasmid pDC-1, Saccharomyces cerevisiae strain AH22 cells synthesized up to 1.5 x 10(5) molecules per cell of cytochrome P-450 protein (P-450(Dah1)). The carbon monoxide-bound reduced form of P-450(Dah1) showed an absorption peak at 447 nm and specific content of P-450(Dah1) was about 0.1 nmole P-450 per mg of microsomal protein. P-450(Dah1) cross-reacted with antibodies to rat P-448-H (CYP1A2) and dog P-450-D2 (CYP1A2). P-450(Dah1) activated 2-amino-3-methyl-imidazo[4,5-f]quinoline (IQ) and 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ) most efficiently in the umu test and exhibited a high activity of aryl hydrocarbon hydroxylase toward benzo[a]pyrene.

The in vitro effects of lipid peroxidation on the content of individual forms of cytochrome P-450 in liver microsomes of guinea-pigs

The effect of lipid peroxidation in vitro on the amounts of several forms of cytochrome P-450 in liver microsomes from guinea-pigs was investigated. Lipid peroxide formation in liver microsomes from ascorbic acid (VC)-deficient animals was much higher than that observed in control animals. The antibodies to rat P-450IA2 (P-448-H), P-450IIB1 (P-450b) and human P-450IIIA4 (P-450NF) recognized one or two forms of cytochrome P-450 in liver microsomes of guinea-pigs. Neither cytochrome P-450 cross-reactive with anti-P-450IIB1 antibodies nor cytochrome P-450 cross-reactive with antibodies to P-450IIIA4 was virtually affected by microsomal lipid peroxidation induced by NADPH in vitro. In contrast, the forms of cytochrome P-450 immunochemically related to P-450IA2 were decreased with the increased level of lipid peroxide formation. The form-specific degradation of cytochrome P-450 due to lipid peroxidation was in agreement with our previous observation that the amounts of cytochrome P-450 cross-reactive with antibodies to P-450IA2 but not with antibodies to P-450IIIA (P-450PB-1) were predominantly decreased in VC-deficient guinea-pigs compared to control animals in vitro.