Inhibition and kinetics of cytochrome P4503A activity in microsomes from rat, human, and cdna-expressed human cytochrome P450

Midazolam (MDZ) is metabolized in human liver microsomes by the cytochrome P450 (CYP) 3A subfamily to 1'-hydroxy (1'-OH) and 4-hydroxy (4-OH) metabolites. MDZ is metabolized in the rat primarily to 4-OH MDZ, 1'-OH MDZ, and 1',4-dihydroxy (1',4-diOH) MDZ. The kinetics of 4-OH and 1'-OH metabolite formation were determined using hepatic microsomes from control, Ro 23-7637 and dexamethasone-treated male rats. KM values for the major metabolite, 4-OH MDZ, were 24.5, 43.1, and 32.8 microM, and the corresponding Vmax values were 5.9, 28.9, and 13 nmol/mg/min for the control, DEX, and Ro 23-7637-treated animals, respectively KM values for 1'-hydroxylation of MDZ (the major metabolite) after incubation with human liver microsomes from three individuals were 5.57, 2.50, and 3.56 microM, and the corresponding Vmax values were 4.38, 0.49, and 0.19 nmol/mg/min, respectively. In parallel studies using cDNA-expressed human CYP3A4 microsomes, the KM for 1'-OH formation was 1.56 microM, and the corresponding Vmax was 0.16 nmol/mg/min. MDZ was not metabolized by cDNA-expressed human CYP2D6, CYP2E1, or CYP1A2, thus confirming that these isoforms were not responsible for its biotransformation. The formation of 1',4-diOH metabolite in rat and 1'-OH formation in cDNA-expressed human CYP3A4 microsomes showed a decrease in velocity at high substrate concentrations. Inhibition studies showed that MDZ hydroxylation was strongly inhibited by ketoconazole and Ro 23-7637 in rat, human, and cDNA-expressed human CYP3A4 microsomes. alpha-Naphthoflavone stimulated 1'-OH metabolite formation in human and cDNA-expressed human CYP3A4 microsomes at low concentration (10 microM). Naringenin, a flavonoid present in grapefruit juice, also inhibited MDZ metabolism in human liver microsomes. Immunoinhibition studies revealed that polyclonal anti-rat CYP3A2 antibody inhibited MDZ metabolism 80-90% in rat, human, and cDNA-expressed human CYP3A4 microsomes, thus suggesting that members of the CYP3A4 subfamily were involved in the metabolism.

Bimodal distribution of renal cytochrome P450 3A activity in humans

It has been proposed that excessive intrarenal conversion of cortisol to 6 beta-hydroxycortisol by CYP3A may mediate increased tubular reabsorption of sodium, leading to a state of mild volume expansion and the clinical phenotype of salt-sensitive hypertension. Therefore, we characterized CYP3A activity in a bank of microsomes from human kidneys using the formation of 1'-hydroxymidazolam (1'-OHM) as a prototypical CYP3A-catalyzed reaction. Maximal rates of metabolite formation occurred at midazolam concentrations of 12.5-50 microM; higher concentrations resulted in dramatic substrate inhibition. At 12.5 microM midazolam, 4 of 27 kidneys exhibited relatively high mean +/- standard deviation 1'-OHM formation rate (184.0 +/- 14.4 pmol/hr/mg) compared with the remaining 23 samples, which had a mean formation rate of (10.1 +/- 6.4 pmol/hr/mg). Triacetyloleandomycin and anti-CYP3A antibody inhibited midazolam hydroxylation by 53% and 57%, respectively. The correlation between CYP3A5 content, determined through immunoblotting, and 1'-OHM formation rate was high (r2 = 0.84, 24 experiments). The expressions of mRNA corresponding to CYP3A3, CYP3A4, CYP3A5, and CYP3A7 were determined through polymerase chain reaction with specific oligonucleotides as primers. All kidneys examined (25 experiments) expressed CYP3A5 protein and contained the corresponding CYP3A5 mRNA. CYP3A4 mRNA was detected in 40% of the kidney samples, and 70% of those that contained detectable CYP3A4 mRNA also expressed detectable levels of the corresponding protein. Therefore, in contrast to hepatic tissue, in which CYP3A4 is universally expressed, CYP3A5 is the ubiquitously expressed member of the CYP3A family in renal tissue. The distribution of enzyme activity and protein content suggests bimodality and may represent induction of CYP3A5 in a select population and/or a genetically determined organ-specific pattern of expression.

Determination of midazolam and its metabolites in serum microsamples by high-performance liquid chromatography and its application to pharmacokinetics in rats

A single-solvent extraction step high-performance liquid chromatographic method is described for quantitating midazolam and its two hydroxy metabolites in rat serum microsamples (50 microliters). The separation used a 2 mm I.D. reversed-phase Symmetry C18 column with an isocratic mobile phase consisting of methanol-acetonitrile-14.9 mM sodium acetate in water at pH 3.0 (10:23:67, v/v). The detection limit was 10 ng/ml for all the compounds using an ultraviolet detector operated at 230 nm. The method was used to study the pharmacokinetics of midazolam after an intravenous bolus dose (0.75 mg/kg).

The aliphatic oxidation of salmeterol to alpha-hydroxysalmeterol in human liver microsomes is catalyzed by CYP3A

Salmeterol xinafoate (Serevent) is a long-acting beta2-adrenoceptor agonist, used in the treatment of asthma, that has bronchodilator and anti-inflammatory action. Salmeterol is extensively metabolized by aliphatic oxidation in humans, with the major metabolite being alpha-hydroxysalmeterol. The aim of this investigation was to identify the specific cytochrome P450 (P450) isoform or isoforms involved in the formation of alpha-hydroxysalmeterol in human liver microsomes. [14C]Salmeterol was incubated with a pooled sample (N = 19) of human liver microsomes in the absence or presence of selective chemical inhibitors of the major human P450 isoforms. One microM ketoconazole, a selective inhibitor of CYP3A, substantially inhibited the metabolism of salmeterol to alpha-hydroxysalmeterol. Disulfiram caused a small but consistent decrease in the amount of alpha-hydroxysalmeterol formed, possibly reflecting less than total selectivity for CYP2E1 under the conditions used. Other selective inhibitors had no significant effect on the metabolism of salmeterol. The rates of formation of alpha-hydroxysalmeterol in 10 individual liver microsomal samples showed an approximately 10-fold variation and were found to be highly correlated (r2 = 0.94; p < 0.001) with rates of metabolism of midazolam to 1'-hydroxymidazolam, a marker of CYP3A activity, in the same microsomal samples. No significant correlation was evident for the metabolism of salmeterol with levels of total P450 or other markers of human P450 activities in the same microsomal samples, thus indicating that the formation of alpha-hydroxysalmeterol is catalyzed predominantly by CYP3A. Insect cell microsomes that coexpressed human CYP3A and NADPH-P450 reductase were able to metabolize [14C]salmeterol to alpha-hydroxysalmeterol, thus confirming the role of CYP3A in catalyzing this reaction. The therapeutic dose of salmeterol is very low, so it is unlikely that any clinically relevant interactions will be observed as a consequence of the coadministration of salmeterol and other pharmaceutical agents that are metabolized by CYP3A.

What changes drug metabolism in critically ill patients--III? Effect of pre-existing disease on the metabolism of midazolam

Liver samples were obtained at hepatectomy from patients with end stage alcoholic liver disease (n = 5), primary biliary cirrhosis (n = 5) and chronic rejection needing retransplantation (n = 5). Normal liver material was also obtained from five organ donors. From these samples microsomes were made containing cytochrome P450 3A. The amount of this enzyme was measured by Western immunoblotting and its function assessed by measuring the rate of production of two metabolites of midazolam, 1-hydroxy midazolam and 4-hydroxy midazolam. There was a wide range in all groups for both the expression and function of this enzyme. Liver tissue affected by cirrhotic disease showed greater preservation of enzyme function than that affected by hepatocellular disease. There was a good correlation between the expression of the enzyme and production of the 1-hydroxy metabolite, but a poor correlation between production of the 4-hydroxy metabolite and expression. This poor correlation may reflect the failure to measure the specific enzyme responsible for producing 4-hydroxy midazolam.

What changes drug metabolism in critically ill patients?--II Serum inhibits the metabolism of midazolam in human microsomes

Serum samples from five critically ill patients were incubated with microsomes prepared from three human livers. The activity of cytochrome P450 3A4 was assessed by measuring the disappearance of midazolam and the appearance of 1-hydroxy midazolam in the incubates. Significant inhibition of the ability of this enzyme to metabolise midazolam was seen. This occurred in incubates containing serum samples from critically ill patients and not in those containing serum from two normal volunteers. The mechanism of this inhibition is unknown, but several possibilities are discussed.

Effect of fluoxetine, norfluoxetine, sertraline and desmethyl sertraline on human CYP3A catalyzed 1'-hydroxy midazolam formation in vitro

The ability of fluoxetine, norfluoxetine, sertraline and desmethyl sertraline to inhibit the CYP3A subfamily of cytochromes P450 was examined in vitro, using the formation of 1'-hydroxy midazolam as a probe for CYP3A catalytic activity. The inhibition observed with these four compounds was modeled using competitive, noncompetitive, uncompetitive and mixed competitive/noncompetitive relationships by nonlinear regression analysis. The best fit model of the inhibition of CYP3A-mediated 1'-hydroxy midazolam formation by all four compounds examined was determined to be mixed inhibition. The calculated Ki values were 65.7 +/- 12.0 microM for fluoxetine, 19.1 +/- 1.9 microM for norfluoxetine, 64.4 +/- 11.6 microM for sertraline and 48.1 +/- 11.6 microM for desmethyl sertraline. Steady-state plasma levels of fluoxetine and norfluoxetine can approach a concentration of 1 microM (approximately 350 ng/ml of plasma). Assuming an inhibitor concentration of 1 microM and a concentration of the substrate substantially below its Km (at least 10-fold lower), the results reported predict that fluoxetine and norfluoxetine together would inhibit CYP3A catalytic activity by less than 7% (less than 0.7% if the unbound plasma concentration of fluoxetine is considered). By using the same assumptions and concentrations for sertraline and desmethyl sertraline, these agents together would be predicted to inhibit the metabolic clearance of a coadministered CYP3A metabolized drug by less than 4%. The observations reported here suggest that fluoxetine and sertraline would have little effect on CYP3A-mediated clearance of coadministered drugs.

In vitro inhibition of midazolam and quinidine metabolism by flavonoids

Studies in humans in vivo have demonstrated that substances found in grapefruit juice may increase the bioavailability of dihydropyridine derivatives as a result of the inhibition of liver enzyme activities by flavonoids found in grapefruit. Since the metabolism of dihydropyridine drugs is mediated by cytochrome P-450 (CYP) 3A4, it has been hypothesized that flavonoids may also influence the metabolism of other drugs, such as midazolam and quinidine, which are biotransformed by the same CYP isoform. Three flavonoids, kaempferol, naringenin and quercetin, are found in grapefruit juice but not in orange juice. The effect of these substances on the metabolism of midazolam and quinidine has been investigated in human liver microsomes. In the concentration range 10-160 microM the inhibitory potential of flavonoids was the same for both of the tested drugs; it decreased in the order quercetin >> kaempferol > naringenin. The data suggest that the flavonoids found in grapefruit juice may influence the kinetics of midazolam and quinidine in man.

Prolonged sedation due to accumulation of conjugated metabolites of midazolam

Midazolam is a short-acting benzodiazepine routinely used in intensive-care medicine. Conjugates of its main metabolite, alpha-hydroxymidazolam, have been shown to accumulate in renal failure but have not previously been related to the prolonged sedative effects commonly observed in critically ill patients. We report five patients with severe renal failure who had prolonged sedation after administration of midazolam. In all five patients, the comatose state was immediately reversed by the benzodiazepine-receptor antagonist flumazenil. Serum concentration monitoring showed high concentrations of conjugated alpha-hydroxymidazolam when concentrations of the unconjugated metabolite and the parent drug were below the therapeutic range. In-vitro binding studies showed that the affinity of binding to the cerebral benzodiazepine receptor of glucuronidated alpha-hydroxymidazolam was only about ten times weaker (affinity constant 16 nmol/L) than that of midazolam (1.4 nmol/L) or unconjugated alpha-hydroxymidazolam (2.2 nmol/L). Conjugated metabolites of midazolam have substantial pharmacological activity. Physicians should be aware that these metabolites can accumulate in patients with renal failure.

Interaction between grapefruit juice and midazolam in humans

OBJECTIVE

To investigate the effects of grapefruit juice on the pharmacokinetics and dynamics of midazolam.

METHODS

Eight healthy male subjects participated in this open crossover study. Intravenous (5 mg) or oral (15 mg) midazolam was administered after pretreatment with water or grapefruit juice. We measured the pharmacokinetics and pharmacodynamics (reaction time, Digit Symbol Substitution Test [DSST], general impression judged by the investigators, and drug effect judged by the subjects) of midazolam and the pharmacokinetics of alpha-hydroxymidazolam.

RESULTS

In comparison to water, pretreatment with grapefruit juice did not change the pharmacokinetics or pharmacodynamics of intravenous midazolam. After oral administration, pretreatment with grapefruit juice led to a 56% increase in peak plasma concentration (Cmax), a 79% increase in time to reach Cmax (tmax), and a 52% increase in the area under the plasma concentration-time curve (AUC) of midazolam, which was associated with an increase in the bioavailability from 24% +/- 3% (water) to 35% +/- 3% (Grapefruit juice; mean +/- SEM, p < 0.01) After oral administration of midazolam, pretreatment with grapefruit juice was associated with a 105% increase in tmax and with a 30% increase in the AUC of alpha-hydroxymidazolam. For oral midazolam, pretreatment with grapefruit juice led to significant increases in tmax for all dynamic parameters and in the AUC values for the reaction time and DSST, whereas the maximal dynamic effects remained unchanged.

CONCLUSIONS

Pretreatment with grapefruit juice is associated with increased bioavailability and changes in the pharmacodynamics of midazolam that may be clinically important, particularly in patients with other causes for increased midazolam bioavailability such as advanced age, cirrhosis of the liver, and administration of other inhibitors of cytochrome P450.

Stereospecific transduction of behavioral effects via diazepam-insensitive GABAA receptors

Previous studies reported a positive correlation between ligand affinities at diazepam-insensitive GABAA receptors and substitution for the discriminative stimulus effects of the benzodiazepine receptor antagonist, flumazenil, in pigeons. In the present experiments, bretazenil and Ro 14-5974 (ethyl-(S)-11,12,13,13 a-tetrahydro-9-oxo-9H-imidazo[1,5-a]-pyrrolo-[2,1-c] [1,4]benzodiazepine-1-carboxylate) partially substituted for, and blocked the discriminative stimulus effects of midazolam, congruent with their actions at diazepam-sensitive GABAA receptors in vitro. In addition, bretazenil and Ro 14-5974, but not their R-enantiomers, had high affinity for diazepam-insensitive receptors and fully substituted for the discriminative stimulus effects of flumazenil. The R-enantiomers of these compounds had low affinity (Ki > 1 microM) for diazepam-sensitive and diazepam-insensitive receptors, and did not share discriminative stimulus effects with flumazenil or midazolam. Ro 19-0528 (7-chloro-3-(3-cyclopropyl-1,2,4-oxadiazol-5-yl)-4,5-dihydro-5-met hyl-6H- imidazo[1,5-a][1,4]benzodiazepin-6-one), a structurally related compound with full agonist actions at diazepam-sensitive GABAA receptors, had high diazepam-insensitive receptor affinity (Ki = 96 nM) and partially substituted for the discriminative stimulus effects of flumazenil. These results are consistent with stereospecific mediation of the discriminative stimulus effects of flumazenil through high affinity binding to diazepam-insensitive receptors in pigeons.

The effect of three H2-receptor antagonists on the disposition of midazolam in the rat in-situ perfused liver model

The rat in-situ perfused liver model was used to investigate the effect of three H2-receptor antagonists on the pharmacokinetic disposition of the short-acting benzodiazepine, midazolam. Perfusion experiments, using standard techniques, were carried out on four groups (one control and three H2-receptor antagonist-treated groups) of male Sprague-Dawley rats (300-350 g). All animals received midazolam 1 mg; the three treated groups received cimetidine (8 mg), ranitidine (3 mg) or famotidine (0.4 mg). Perfusate and bile samples were collected and assayed for midazolam using gas chromatography. The perfusate data indicated that midazolam disposition was impaired at 10, 50 and 60 min of the experimental period following the addition of cimetidine, whereas ranitidine and famotidine produced an effect at 10 min only; midazolam levels in bile were not affected by the presence of an H2-receptor antagonist. It was concluded that the limited inhibitory effect of cimetidine may be attributed to its lack of specificity for CYP3A, the isoenzyme responsible for the metabolism of midazolam.

Bioreactor for a larger scale hepatocyte in vitro perfusion

A bioreactor construction for hepatocytes and liver sinusoidal endothelial cells is described. The reactor is based on capillaries for hepatocyte immobilization. Four discrete capillary membrane systems, each serving different purposes, are woven to create a three-dimensional framework for decentralized cell perfusion with low metabolite gradients and decentralized oxygenation and CO2 removal. The biochemical performance of reactors initially seeded with 2.5 x 10(9) hepatocytes were evaluated over 3 weeks. On day 21, pig albumin synthesis was 4.7 mg/day, lidocaine metabolism was 813.7 +/- 23 micrograms/hr, galactose elimination was 210.1 +/- 3 mg/hr, and midazolam metabolism was 37.1 +/- 2 micrograms/hr. The specific construction of the reactor enables scale-up to hybrid liver support systems as extracorporeal bridging devices for liver transplantation.

Interpatient heterogeneity in expression of CYP3A4 and CYP3A5 in small bowel. Lack of prediction by the erythromycin breath test

The CYP3A subfamily of cytochromes P450 metabolize many medications and environmental contaminants. CYP3A4 and, in 25% of patients, CYP3A5 seem to be the major CYP3A genes expressed in adult liver. Hepatic levels of CYP3A4 can be estimated by the erythromycin breath test and vary at least 10-fold among patients. CYP3A4 has also been shown to be present in small bowel where it is responsible for significant "first-pass" metabolism of orally administered substrates. However, it is not known whether there is significant interindividual variability in the intestinal expression of CYP3A4, or whether the liver and intestinal catalytic activities of CYP3A4 correlate within an individual. It is also not known whether CYP3A5 is expressed in the small intestine. To address these questions, we administered the erythromycin breath test to 20 patients and obtained biopsies from their small bowel. There was a 6-fold variation in CYP3A catalytic activity (midazolam hydroxylation), an 11-fold variation in CYP3A4 protein content, and an 8-fold variation in CYP3A4 mRNA content in intestinal biopsies. There was an excellent correlation between intestinal CYP3A4 protein level and catalytic activity (r = 0.86; p = 0.0001); however, neither parameter significantly correlated with hepatic CYP3A4 activity as measured by the erythromycin breath test result (r = 0.27; p = 0.24 and r = 0.33; p = 0.15, respectively). We also found that CYP3A5 protein was readily detectable in biopsies from 14 (70%) of the patients, indicating that CYP3A5 is commonly expressed in human small intestine.(ABSTRACT TRUNCATED AT 250 WORDS)

Midazolam is metabolized by at least three different cytochrome P450 enzymes

Distribution volumes and metabolism determine the pharmacokinetics of midazolam. Cytochrome P450 3A4 has been considered a significant enzyme in its metabolism. Using heterologously expressed cytochrome P450 enzymes, we have confirmed the additional involvement of cytochromes P450 3A3 and 3A5 in the hydroxylation of the midazolam. Whereas cytochrome P450 3A3 metabolized midazolam to the same extent as cytochrome P450 3A4, cytochrome P450 3A5 increased its metabolism by a factor of 2.7. The relationship of alpha- to 4-hydroxylation of midazolam was approximately 1.3 for cytochromes P450 3A3 and 3A4, and approximately 8.8 for 3A5. The primary location of cytochromes P450 3A3 and 3A4 is the liver in contrast with cytochrome P450 3A5, which occurs predominantly in the kidney. Therefore, further in vivo study is required to prove conclusively that enzymes in the kidney are involved in the metabolism of midazolam. Nitrendipine itself is metabolized by cytochrome P450 3A enzymes and this was shown to inhibit human liver microsomal hydroxylation of midazolam and preferentially alpha-hydroxylation by about 77%. 4-Hydroxylation was inhibited to 32% of control by nitrendipine. In contrast with inhibition of 4-hydroxylation, alpha-hydroxylation would appear to be competitively inhibited. These findings may be relevant to drug interactions in combined therapy.

Use of midazolam as a human cytochrome P450 3A probe: II. Characterization of inter- and intraindividual hepatic CYP3A variability after liver transplantation

Immunosuppression therapy with cyclosporine is often hampered by significant interindividual variability in the metabolic clearance of the drug. It has been suggested that much of the variability in cyclosporine clearance is due to differences in the cytochrome P450 3A4 (CYP3A4) content in the liver and intestinal mucosa. A study was conducted in liver transplant recipients to characterize hepatic CYP3A variability during the first 10 days after surgery. The formation of 1'-hydroxymidazolam (1'-OH MDZ) was followed in the plasma after i.v. midazolam (MDZ) administration to 21 multiple-organ donors and to recipients of 10 of the 21 donor livers. Liver biopsy tissue was obtained from donors and recipients after the in vivo pharmacokinetic test. For liver donors, the plasma 1'-OH MDZ/MDZ concentration ratio 30 min after the i.v. MDZ dose was well correlated with the hepatic CYP3A4 content (r = .87, P < .001). Much of the variability in the two parameters was attributed to the administration of enzyme-inducing drugs before organ procurement. The mean hepatic CYP3A4 content and plasma 1'-OH MDZ/MDZ concentration ratio in six inducer-treated donors was 4.7-fold and 2.3-fold higher than the respective mean value for all other donors. The hepatic CYP3A4 content and plasma 1'-OH MDZ/MDZ ratio for liver recipients, studied on postoperative day 10, was negatively correlated with the respective parameter measured in donors on day 0 (r = -0.60 for CYP3A4 and r = -0.79 for 1'-OH MDZ/MDZ; P < .05 and P < .01).(ABSTRACT TRUNCATED AT 250 WORDS)

Use of midazolam as a human cytochrome P450 3A probe: I. In vitro-in vivo correlations in liver transplant patients

The clearance of midazolam (MDZ) in humans is principally due to metabolic biotransformation catalyzed by CYP3A isoforms. A study was conducted in patients who had undergone liver transplants that provides evidence that MDZ can be used as an in vivo probe of interindividual hepatic CYP3A variability. The clearance of MDZ and cyclosporine after i.v. administration were determined in 10 patients approximately 10 days after transplant surgery. Liver biopsy specimens were obtained within 24 hr of the pharmacokinetic study and CYP3A content and MDZ 1'-hydroxylation activity were measured in 13,000 x g tissue supernatants (S-13). The in vitro rate of 1'-hydroxy-MDZ formation was found to correlate significantly with the total CYP3A content in hepatic S-13 fractions (r = .84, P < .01). The total MDZ clearance measured in vivo was highly correlated with the hepatic CYP3A content measured in vitro (r = .93, P < .001) and with in vivo cyclosporine clearance (r = .81, P < .001). For five of the patients, the intrinsic clearance of midazolam to 1'-hydroxy-MDZ (Vmax/Km) in vitro measured in S-13 preparations was scaled for total liver mass and applied to the well stirred model of hepatic clearance to yield a prediction of MDZ clearance in vivo. The mean MDZ clearance predicted from in vitro 1'-hydroxylation data was identical to the mean clearance observed in vivo (0.60 +/- 0.24 versus 0.59 +/- 0.25 liter/min). Together, the results suggest that variability in hepatic CYP3A expression in liver transplant recipients, and possibly in other populations, can be determined by the measurement of MDZ metabolic clearance.

Regioselective biotransformation of midazolam by members of the human cytochrome P450 3A (CYP3A) subfamily

The capabilities of cytochrome P4503A4 (CYP3A4), CYP3A5, and fetal hepatic microsomes containing CYP3A7 to metabolize midazolam were investigated using human hepatic microsomes and purified CYP3A4 and CYP3A5. Under initial rate conditions and high substrate concentration (400 microM midazolam), variability among eighteen human liver microsomal samples was 30- and 16- fold for 1'- and 4-hydroxylation of midazolam, respectively. Exclusion of two samples isolated from patients previously administered barbiturates reduced the inter-individual variability to 10.5- and 6.0-fold for 1'- and 4-hydroxylation, respectively. Six fetal hepatic microsomal samples showed 10-fold variation in both 1'-hydroxymidazolam and 4-hydroxymidazolam formation rates. The rates of formation of 4-hydroxymidazolam and 1'-hydroxymidazolam from midazolam by adult samples containing only CYP3A4 and by fetal liver samples were highly correlated (r2 = 0.99 and 0.97, P < 0.01, respectively). The rates of formation of 1'-hydroxymidazolam and 4-hydroxymidazolam from midazolam (400 microM) by adult samples that contained only CYP3A4 were correlated significantly (P < 0.01) with the ability of the samples to N-demethylate erythromycin (r2 = 0.95 and 0.92, respectively). 6 beta-hydroxylate testosterone (r2 = 0.96 and 0.96, respectively), and the CYP3A4 content of the samples (r2 = 0.89 and 0.86, respectively). Microsomal samples containing CYP3A5 in addition to CYP3A4 exhibited a significantly greater ratio of 1'-hydroxymidazolam to 4-hydroxymidazolam compared with samples containing only CYP3A4 or CYP3A7 (P < 0.001). Purified CYP3A5 in a reconstituted system, consisting of dilauroylphosphatidylcholine, cytochrome b5, and NADPH-cytochrome P450 reductase, and an NADPH-regenerating system displayed a 2-fold greater rate of 1'-hydroxymidazolam formation and a similar rate of 4-hydroxymidazolam formation compared with a reconstituted system with CYP3A4. In conclusion, CYP3A4, CYP3A5, and fetal microsomes containing CYP3A7 catalyze 1'- and 4-hydroxylation of midazolam with the ratio of these metabolites indicative of the CYP3A form.

O-demethylation of epipodophyllotoxins is catalyzed by human cytochrome P450 3A4

We previously demonstrated that O-demethylation of the pendant dimethoxyphenol ring of epipodophyllotoxins to produce their respective catechol metabolites is catalyzed by cytochrome(s) P450 in human liver microsomes. Our objective was to identify the specific human cytochrome(s) P450 responsible for catechol formation. Using a panel of prototypical substrates and inhibitors for specific cytochromes P450, we identified substrates for CYP3A4 (midazolam, erythromycin, cyclosporin, and dexamethasone) as inhibitors of catechol formation from both etoposide and teniposide. Dexamethasone inhibition was competitive, with Ki values of 60 and 45 microM for etoposide and teniposide, respectively. In 58 human livers, the correlation coefficients for teniposide catechol formation versus 1'- and 4-hydroxymidazolam formation were 80% and 85%, respectively; for etoposide catechol formation versus 1'- and 4-hydroxymidazolam formation r2 was 83% and 79%, respectively. Teniposide and etoposide catechol formation rates were also significantly correlated with immunodetectable CYP3A (r2 = 49% and 51%, respectively) and not with immunodetectable CYP1A2, 2E1, or 2C8. Finally, cDNAs for human CYP3A4, 3A5, 2A6, 2B6, 2C8, and 2C9 were functionally expressed in HepG2 cells, using a vaccinia viral vector. Teniposide and etoposide catechol formation was catalyzed primarily by 3A4 (15.4 and 40.9 pmol/pmol/hr, respectively) and to a lesser degree by 3A5 (1.94 and 11.3 pmol/pmol/hr, respectively), whereas there was no detectable O-demethylation of epipodophyllotoxins by 2A6, 2B6, 2C8, 2C9, or the control virus alone. Moreover, the relative activities of midazolam hydroxylation, compared with O-demethylation of epipodophyllotoxins, were similar for heterologously expressed 3A4 and for human liver microsomes. We conclude that catechol formation from teniposide and etoposide is primarily mediated by human CYP3A4, making these reactions susceptible to inhibition by prototypical 3A substrates and inhibitors.

Selective inhibition of major drug metabolizing cytochrome P450 isozymes in human liver microsomes by carbon monoxide

The selectivity of carbon monoxide binding to specific human cytochrome P450 isozymes was investigated by studying its inhibition of prototype reactions for 3 major drug metabolizing P450s in liver microsomes: dextromethorphan O-demethylation and (+)-bufuralol 1'-hydroxylation (P450DB1, CYP2D6), diclofenac 4'-hydroxylation (P450TB, CYP2C subfamily), and midazolam 1'-hydroxylation (P450NF, CYP3A subfamily). The affinity of carbon monoxide is different for each P450 isozyme. Warburg partition coefficients were 0.35, 1.1 and 3.9 microM for P450DB1, P450TB and P450NF, respectively. Differential inhibition by carbon monoxide may be a useful tool to identify specific human cytochrome P450 isozymes in the early screening of drug biotransformation catalysts. Further studies involving other P450 isozymes and substrates should extend our understanding of the phenomena and their implications.

Benzodiazepine receptor equilibrium constants for flumazenil and midazolam determined in humans with the single photon emission computer tomography tracer [123I]iomazenil

This study is based on the steady state method for the calculation of Kd values recently described by Lassen (J. Cereb. Blood Flow Metab. 12 (1992), 709), in which a constant infusion of the examined nonradioactive ligand is used with a bolus injection of tracer. Eight volunteers were examined twice, once without receptor blockade and once with a constant degree of partial blockade of the benzodiazepine receptors by infusion of nonradioactive flumazenil (Lanexat) or midazolam (Dormicum). Single photon emission computer tomography and blood sampling were performed intermittently for 6 h after bolus injection of [123I]iomazenil. The tracer in plasma was determined by high-pressure liquid chromatography and also by a simple octanol extraction procedure. The free concentration of flumazenil and midazolam in plasma water averaged 52% and 3.5% of that in whole plasma. The Kd values for the entire cortical rim for flumazenil were 7.4, 10.0, 10.3 and 17.7 nmol/l plasma water and, for midazolam, 73, 76, 58 and 30 nmol/l plasma water. The variation exceeds random methodological error and is probably due to interindividual differences in receptor affinity. The Kd level of midazolam is considerably higher than expected from the results of in vitro studies.

Drug metabolism studies using "intrinsic" and "extrinsic" labels. A demonstration using 15N vs. Cl in midazolam

The chemical reaction interface for mass spectrometry (CRIMS) has been used to evaluate the ability of an "intrinsic" label (chlorine) to replace an "extrinsic" label (15N) in a study of the metabolite profile of a drug, in this case the benzodiazepine anesthetic agent midazolam. We find equally high selectivity and comparable signal/noise characteristics for chlorine as for isotopic nitrogen demonstrating that the chlorine in midazolam is itself an effective label and that special synthesis to incorporate isotopic labels is not necessary. The power of either detection mode of CRIMS is shown by detecting 14 metabolites, whereas only four had been previously determined. The ratios of 15N/Cl for each metabolite peak along with conventional mass spectra provide clues to the structures of these new metabolites.

Thawed human hepatocytes in primary culture

In drug metabolism studies, isolated and cultured human hepatocytes provide a useful model for overcoming the difficulty of extrapolating from animal data. In vitro studies with human hepatocytes are scarce because of the lack of livers and suitable methods of storage. After developing a new method for cryopreservation of human hepatocytes, we evaluated the effects of deep freezing storage on their viability, morphology, and functional and toxicological capabilities in classical culture conditions. Freshly isolated human hepatocytes were cryopreserved in medium containing 10% Me2SO and 20% fetal calf serum, using a Nicool ST20 programmable freezer (-1.9 degrees C/min for 18 min and -30 degrees C/min for 4 min). Cells were stored in liquid nitrogen. Viability of thawed human hepatocytes was 50-65% as assessed by erythrosin exclusion test prior to purification on a Percoll density gradient. Morphological criteria showed that thawed human hepatocytes require an adaptation period to the medium after seeding. Functional assessments showed that human hepatocytes which survive freezing and thawing preserve their protein synthesis capabilities and are able to secrete a specific protein, anionic peptidic fraction, which is involved in the hepatic uptake of bile-destined cholesterol. We then studied Midazolam biotransformation to test metabolic functions, and erythromycin toxicity by Neutral Red test (cell viability) and 3-(4,5-dimethylthiazol-2-yl)-diphenyl tetrazolium bromide test (cell metabolism). All of these experiments indicated that thawed human hepatocytes should be used 38 h after seeding for optimum recovery of their functions: membrane integrity, protein synthesis, and stabilization of drug metabolism enzymes.