Identification and biological activity of the active metabolite of clopidogrel.

Like ticlopidine, the ADP receptor antagonist clopidogrel is inactive in vitro and must be administered i.v. or orally to exhibit antiaggregatory and antithrombotic activities. We have previously shown that hepatic metabolism is necessary for activity. This study demonstrates that an active metabolite can be generated from human liver microsomes incubated with clopidogrel. Using several analytical methodologies (LC/MS, NMR, chiral supercritical fluid chromatography), we have identified its structure. In vitro, this highly unstable compound, different from that formed from ticlopidine, exhibited all the biological activities of clopidogrel observed ex vivo: Irreversible inhibition of the binding of 33P-2MeS-ADP to washed human platelets (IC50) = 0.53 microM), selective inhibition of ADP-induced platelet aggregation (IC)50 = 1.8 microM) and ADP-induced adenylyl cyclase down-regulation. The irreversible modification of the ADP-receptor site which is responsible for the biological activity could be explained by the formation of a disulfide bridge between the reactive thiol group of the active metabolite and a cysteine residue of the platelet ADP receptor.

Development of a radioimmunoassay for SR 31747A, a new sigma ligand, in human plasma.

A specific and sensitive radioimmunoassay was developed for SR 31747A, a new sigma ligand, using a monoclonal antibody. This antibody was produced from spleen lymphocytes of a mouse immunized with SR 31747A coupled to bovine serum albumin via a peptide bond using SR 120684A, a succinamic acid derivative of SR 31747A. Negligible binding occurred when metabolites, obtained by chemical synthesis or by "in-vitro" incubation with hepatic microsomal fraction, were tested for cross-reactivity. A quantitative recovery in serum of the exogenous analyte was obtained for all the concentrations tested and the quantification limit was found to be 0.25 ng ml-1 of SR 31747 (the non-salified derivative). Intra- and inter-assay relative standard deviations ranged from 6.3-10.9 and from 5.3% to 15.4% respectively. Furthermore, comparison of results from samples which were assayed by radioimmunoassay and gas chromatography demonstrated an excellent correlation (r = 0.984).

The antiaggregating activity of clopidogrel is due to a metabolic activation by the hepatic cytochrome P450-1A.

Clopidogrel and ticlopidine are two well known selective anti-ADP agents which are inactive in vitro and must be administered in vivo to fully exhibit their antiaggregating and antithrombotic effects. Since previous studies have clearly demonstrated that the activation steps take place in the liver, we examined the effect of specific induction or inhibition of cytochrome P450 subfamilies on the antiaggregating activity of clopidogrel. SKF 525-A, a global cytochrome P450 inhibitor, dramatically decreased the antiaggregating effect of clopidogrel, therefore indicating that cytochrome P450 enzymes are involved in the hepatic activation of clopidogrel. The efficacy of clopidogrel was increased in animals pretreated with 3-methylcholanthrene and beta-naphthoflavone, indicating that the cytochrome P450-1A subfamily pathway was mainly involved in the activating metabolism of clopidogrel. The use of specific antibodies directed against the various cytochrome P450 subfamilies ascertained this observation.

Importance of hepatic metabolism in the antiaggregating activity of the thienopyridine clopidogrel.

The thienopyridine clopidogrel is not active in vitro and must be administered i.v. or orally, suggesting that metabolism is necessary for activity. To verify whether the effect after i.v. administration was consecutive to recycling by hepatic bile secretion of clopidogrel or its metabolite(s) in the digestive tract, a catheter was implanted in the choledocus of rats, preventing bile and pancreatic secretions from being excreted into the digestive tractus. Two hours after clopidogrel administration (10 mg/kg, i.v.), blood was withdrawn and platelet-rich plasma aggregation was measured after the addition of 5 microM ADP. Clopidogrel treatment was equally efficient for sham-operated and catheterized animals (% inhibition of platelet aggregation: 76% and 59%, respectively) suggesting that the i.v. effect of clopidogrel was independent of re-absorption of biliary-excreted products and consequently that enteric metabolism is not necessary for activity. The antiaggregating activity of clopidogrel in rats before and after functional hepatectomy by a porto-jugular shunt was then studied. A great difference between treated animals was observed 30 min after i.v. administration of 25 mg/kg of clopidogrel. Per cent inhibition of platelet aggregation was 76% and 6% (P less than 0.001) for sham-operated and hepatectomized animals, respectively. Similar results were obtained after intraduodenal administration of clopidogrel, showing that the treatment was completely ineffective in hepatectomized animals. In isolated, blood-perfused rat livers, clopidogrel inhibited ADP-induced platelet aggregation, thereby supporting the theory that the activity of clopidogrel is highly dependent on hepatic metabolism.

Human hepatocytes as a key in vitro model to improve preclinical drug development.

Over past decades, numerous in vitro and/or ex vivo models have been developed to investigate drug metabolism. In the order of complexity we found the isolated perfused liver, hepatocytes in co-culture with epithelial cells, hepatocytes in suspension and in primary culture and subcellular hepatic microsomal fractions. Because they can be easily prepared from both animals (pharmacological and toxicological species) and humans (whole livers as well as biopsies obtained during surgery) hepatocytes in primary culture provide the most powerful model to better elucidate drug behavior at an early stage of preclinical development such as: the characterization of main biotransformation reactions, the identification of phase I and phase II isozymes involved in such reactions, the evaluation of inter-species differences allowing the selection of a second toxicological animal species more closely related to man on the basis of metabolic profiles, the detection of the inducing and/or inhibitory effects of a drug on metabolic enzymes, the prediction of drug interactions, the estimation of inter-individual variability in biotransformation reactions. The use of hepatocytes, and in particular those obtained from humans, at an early stage of drug development allows the obtention of more predictive preclinical data and a better knowledge of drug behavior in humans before the first administration of the drug in healthy volunteers.

Resolution of rabbit liver cytochrome P450 3b and 3c (P450 IIC3 and IIIA4) from microsomal membrane by two-dimensional gel electrophoresis.

Two-dimensional gel electrophoresis was applied to the separation of microsomal proteins and P450 isozymes. The experimental conditions of the first-dimensional isoelectric focusing were optimized. The best results were obtained under the following conditions: (i) a mixture of two detergents (Tergitol NP 10 and CHAPS) in the focusing gel and sample buffer, (ii) a carrier ampholyte gradient between pH 6.5 and 9.0, (iii) sample loading at the anodic end of the gel, and (iv) low protein loading in the sample (below 50 micrograms). Four P450 forms, including P450 3b and 3c, could be resolved under these conditions.

Ontogenesis of rabbit liver cytochrome P450. Evidence for a cytochrome P450-IIE (3a)-related form prevailing during the post-natal period.

The liver hydroxylating system, mainly composed of cytochromes P450, is not highly active during foetal life. If develops after birth and reaches the adult level several weeks post-partum. We have studied the ontogenesis of rabbit cytochrome P450 during the post-natal period. Total P450 as well as isozymes 2, 3b, 3c, 4 and 6 were measured. The evolution of these proteins with ageing, together with qualitative modification of an electrophoretic profile, produced evidence of an early developing P450 prevailing from one week to three weeks after birth. We isolated and characterized a cytochrome, called P450 2y, from two-week liver microsomes. It is closely related to P450 3a, an adult form of rabbit P450 induced by ethanol. They have similar molecular masses, the same lambda max of CO-reduced spectrum and exhibit immunological cross-reactivity. However, we cannot conclude that the two proteins are identical from N-terminal amino acid analysis or the two-dimensional gel electrophoresis pattern. These results, as well as the recent evidence of two different genes coding for the P450 3a family, strengthen the idea that P450 2y and 3a are distinct proteins. P450 2y seems to be an early developing form abundant soon after birth, while P450 3a is a delayed form appearing like most P450 isozymes during the fourth post-natal week. Besides the quantitative development during perinatal life, there is an important qualitative modification of liver cytochrome P450 content.

The increase in urinary excretion of 6 beta-hydroxycortisol as a marker of human hepatic cytochrome P450IIIA induction.

1. Urinary excretion of 6 beta-hydroxycortisol, hepatic microsomal cortisol 6 beta-hydroxylase and the specific content of several forms of cytochrome P450 were measured in 8 to 14 patients before and after treatment with rifampicin (600 mg orally per day for 4 days). 2. Rifampicin treatment produced an average five fold increase in daily excretion of urinary 6 beta-hydroxycortisol. 3. Cortisol 6 beta-hydroxylase activity increased from 15 +/- 6 pmol min-1 mg-1 in organ donors (considered as 'control subjects') to 87 +/- 31 pmol min-1 mg-1 in rifampicin treated patients. 4. Among three forms of human P450 (P450IA, IIC and IIIA), (1), (2), measured by Western blots, only P450IIIA was significantly induced by the antibiotic. 5. Only antibodies against P450IIIA selectively inhibited cortisol 6 beta-hydroxylase in human liver microsomes. 6. Cortisol 6 beta-hydroxylase was correlated with P450IIIA specific content. 7. The urinary level of 6 beta-hydroxycortisol correlated with liver microsomal cortisol 6 beta-hydroxylase and P450IIIA specific content. 8. We conclude that P450IIIA is predominantly responsible for cortisol 6 beta-hydroxylase activity in human liver microsomes and that urinary 6 beta-hydroxycortisol is a marker of the induction of this cytochrome P450.

The interaction between cyclosporin A and erythromycin studied by means of an isolated perfused rat liver model.

The isolated perfused rat liver model was successfully used to study the clinical interaction observed between cyclosporin A (CsA) and erythromycin (ER). Three experimental studies were performed. In a "Control group", isolated rat livers were perfused with 60 micrograms 3H-CsA for 20 min. In an "ER group", isolated rat livers were simultaneously perfused with 60 micrograms 3H-CsA and 6.0 mg ER. In an "ER ind group", rats were treated with ER for 5 days, conditions under which the P450IIIA gene subfamily, principally involved in CsA metabolism, was specifically induced, and isolated rat livers were perfused with 60 micrograms 3H-CsA for 20 min. Biological parameters of the liver model such as biliary flow, oxygen consumption and pH were similar whatever group of animals. Biliary excretion of total radiolabel (sum of CsA and all the metabolites = CsAT) was similar in both the "Control" and "ER" groups but significantly higher in the "ER ind" group with respectively 8.05 +/- 0.81%, 8.45 +/- 1.93% and 14.54 +/- 2.12% of the entire amount of infused drug excreted during 2 hr. Using a high performance liquid chromatography method which specifically resolves unchanged CsA from its metabolites, we demonstrated that, although similar amounts of radiolabelled material were excreted in both the "Control" and "ER" groups, the CsA/metabolite ratio was lower than 1.0 in the "Control group" and higher than 1.0 in the "ER group", suggesting an inhibition of CsA metabolism by ER in the "ER group". These data could explain the increase in CsA blood levels observed in the clinic following administration of high ER doses and the reversing effect when ER administration is stopped.

Metabolism of cyclosporin A. IV. Purification and identification of the rifampicin-inducible human liver cytochrome P-450 (cyclosporin A oxidase) as a product of P450IIIA gene subfamily.

A cytochrome P-450 involved in the metabolism of cyclosporin A (CsA) was isolated and purified to electrophoretic homogeneity from human liver microsomes of renal transplant donors. This cytochrome, designated P-450(CsA), exhibited a type I binding spectrum in the presence of CsA with a Ks(app) of 25 microM, a molecular weight of 52 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and a maximal absorbance at 449 nm when reduced in the presence of carbon monoxide. The N-terminal sequence of P-450(CsA), determined by Edman degradation reaction, was 63% homologous with that of the rabbit liver CsA oxidase P-450 3c and 100% homologous with that of the human liver isozyme P-450(HLp/NF), recently identified as the human nifedipine (NF) oxidase. Polyclonal and monoclonal antibodies directed against P-450 3c and P-450(HLp/NF), respectively, recognized native microsomal and highly purified P450(CsA). As observed in the rabbit, human liver microsomes were shown to generate mono- and dihydroxy, as well as dihydroxy and/or monohydroxy N-demethylated, derivatives of CsA. Production of these metabolites was shown to be specifically inhibited by anti-P-450 3c polyclonal antibodies. CsA oxidase, NF oxidase, and erythromycin demethylase were shown to be closely correlated with the level of P-450(CsA) determined from Western blot or enzyme-linked immunosorbent assay. Moreover, these monoxygenase activities and the hepatic level of P-450(CsA) were simultaneously increased in the liver of patients treated for 4 days with 600 mg of rifampicin per day. Finally, NF was shown to be a competitive inhibitor of CsA oxidation and vice versa. We conclude that P-450(CsA) is responsible for most (80%) of CsA oxidase activity in human liver, is encoded by gene P450IIIA3, as is NF oxidase, or a very closely related gene, and is strongly inducible by rifampicin pretreatment.

Characterization of midazolam metabolism using human hepatic microsomal fractions and hepatocytes in suspension obtained by perfusing whole human livers.

Isolated human hepatocytes provide a useful model for studying xenobiotic metabolism. However, in vitro studies using human hepatocytes are scarce due to the limited availability of this material. A new methodology is described for obtaining hepatocytes from a whole adult human liver. This procedure is based on (i) the rapid and intense in situ washing step of the organ with Eurocollins then glucose supplemented HEPES buffer (10 mM, pH 7.4) at 4 degrees in order to both minimize the warm ischemic period and remove erythrocytes, and (ii) a perfusion of collagenase solution (0.05% in 10 mM HEPES buffer at 37 degrees) throughout the portal vein according to a recirculated model. All perfused buffers are oxygenized. Hepatocyte viability averaged 85% as determined by Trypan Blue dye exclusion. The ability of these hepatocytes to catalyze certain metabolic transformations such as Phase I and Phase II reactions has been particularly investigated using the benzodiazepine drug, midazolam, as a substance probe. Freshly isolated human hepatocytes in suspension retained the ability to metabolize midazolam to its different hydroxylated derivatives--mainly the 1-hydroxy-midazolam--which was further conjugated with glucuronic acid. For a better understanding of the cytochrome P-450 mediated reactions, we studied the metabolism of midazolam in microsomal fractions prepared from twelve human livers. It was concluded that human microsomes (i) exhibited a Type I binding spectrum upon midazolam addition (Ks = 3.3 microM) and (ii) intensively metabolized the drug to its different derivatives. Furthermore, and since we demonstrated that midazolam was predominantly transformed by a single cytochrome P-450 enzyme, we could attribute the large inter-individual variations in midazolam metabolism to differences in human liver cytochrome P-450 content.