Advertisement
Research ArticleArticle

In Vitro Evaluation of the Disposition of A Novel Cysteine Protease Inhibitor

Wolfgang Jacobsen, Uwe Christians and Leslie Z. Benet
Drug Metabolism and Disposition November 2000, 28 (11) 1343-1351;

Abstract

K11777 (N-methyl-piperazine-Phe-homoPhe-vinylsulfone-phenyl) is a potent, irreversible cysteine protease inhibitor. Its therapeutic targets are cruzain, a cysteine protease of the protozoan parasiteTrypanosoma cruzi, and cathepsins B and L, which are associated with cancer progression. We evaluated the metabolism of K11777 by human liver microsomes, isolated cytochrome P450 (CYP) enzymes, and flavin-containing monooxygenase 3 (FMO3) in vitro. K11777 was metabolized by human liver microsomes to three major metabolites:N-oxide K11777 (apparentKm = 14.0 ± 4.5 μM and apparentVmax = 3460 ± 3190 pmol · mg−1 · min−1, n = 4), β-hydroxy-homoPhe K11777 (Km = 16.8 ± 3.5 μM and Vmax = 1260 ± 1090 pmol · mg−1 · min−1, n = 4), andN-desmethyl K11777 (Km = 18.3 ± 7.0 μM and Vmax = 2070 ± 1830 pmol · mg−1 · min−1, n = 4). All three K11777 metabolites were formed by isolated CYP3A and their formation by human liver microsomes was inhibited by the CYP3A inhibitor cyclosporine (50 μM, 54–62% inhibition) and antibodies against human CYP3A4/5 (100 μg of antibodies/100 μg microsomal protein, 55–68% inhibition). CYP2D6 metabolized K11777 to its N-desmethyl metabolite with an apparent Km (9.2 ± 1.4 μM) lower than for CYP3A4 (25.0 ± 4.0 μM) and human liver microsomes. The apparent Km forN-oxide K11777 formation by cDNA-expressed FMO3 was 109 ± 11 μM. Based on the intrinsic formation clearances and the results of inhibition experiments (CYP2D6, 50 μM bufuralol; FMO3 mediated, 100 mM methionine) using human liver microsomes, it was estimated that CYP3A contributes to >80% of K11777 metabolite formation. K11777 was a potent (IC50 = 0.06 μM) and efficacious (maximum inhibition 85%) NADPH-dependent inhibitor of human CYP3A4 mediated 6′β-hydroxy lovastatin formation, suggesting that K11777 is not only a substrate but also a mechanism-based inhibitor of CYP3A4.

K117771(N-methyl-piperazine-Phe-homoPhe-vinylsulfone-phenyl, Fig. 1) is a recently developed peptidomimetic that is a potent, irreversible inhibitor of cysteine proteases such as cathepsin B (E.C. 3.4.22.1) and L (E.C. 3.4.22.15) as well as cruzain. Cruzain is a cysteine protease of Trypanosoma cruzi, a protozoan parasite and the etiological agent of American trypanosomiasis or Chagas' disease. Chagas' disease is the leading cause of heart disease in Latin America (Filardi and Brener, 1987;Libow et al., 1991; Parada et al., 1997). The currently available therapeutic options for Chagas' disease are generally considered unsatisfactory. Due to significant toxicity, chemotherapy with nifurtimox or benznidazole must be carried out under close medical control (Gorla et al., 1988). Although both drugs significantly shorten the acute phase and decrease mortality of Chagas' disease, they fail to eradicate the parasite in 40% of the patients and are not used during the chronic phase of the disease (Kirchhoff, 1993).

Figure 1
Figure 1

Chemical structures of K11777 in comparison to K11002.

The arrow marks structural differences.

Recent chemotherapeutic approaches have focused on the development of specific inhibitors of cruzain (Cazzullo et al., 1997). Inhibition of cruzain prevented growth and differentiation of T. cruzi in cell culture models (Harth et al., 1993), showing that cruzain is a promising therapeutic target. In addition, K11777 inhibits the cystein proteases cathepsins B and L, which are significantly involved in all stages of cancer progression, including growth, angiogenesis, invasion, and metastasis (Chen, 1992; Elliott and Sloane, 1996).

K11777 is the most recent development in a line of peptidomimetic cysteine protease inhibitors (personal communication, Arris Pharmaceutical Corp., South San Francisco, CA). In comparison to its predecessor K11002, for which the oral bioavailability in Sprague-Dawley rats was 3% (Zhang et al., 1998b), K11777 possesses anN-methyl-piperazine ring instead of a morpholine-urea ring, increasing solubility in intestinal fluids and thereby oral bioavailability. Indeed, oral bioavailability of K11777 (mean of 19.9%, unpublished data) proved to be significantly greater than that of K11002.

It was our goal to evaluate the oxidative metabolism of K11777 by liver monooxygenases and to assess the influence of theN-methyl-piperazine ring on the in vitro metabolite pattern and monooxygenases involved with K11777 in comparison to K11002 (Zhang et al., 1998a).

Materials and Methods

Chemicals.

K11777 and K11002 (Arris Pharmaceutical Corp.), lovastatin (Merck, Sharp & Dohme, Rahway, NJ), itraconazole (Janssen Pharmaceutical Inc., Titusville, NJ), and cyclosporine (Novartis Pharma AG, Basel, Switzerland) were kind gifts of the manufacturers.

(±)-Bufuralol was purchased from UFC Ltd. (Manchester, UK) and NADP, isocitric acid, and isocitrate dehydrogenase for the NADPH-generating system as well as NADPH were purchased from Sigma (St. Louis, MO). Recombinant CYP and FMO3 enzymes as well as monoclonal antibodies against human CYP3A and human liver microsomes were from Gentest (Woburn, MA). All solvents were HPLC grade and were from Fisher Scientific (Santa Clara, CA). All other chemicals were reagent grade and were purchased from Sigma.

Isolation of Microsomes.

The collection of human liver tissue samples for in vitro drug metabolism studies was approved by the Committee on Human Research, University of California, San Francisco. Microsomes were isolated by differential centrifugation as described by Guengerich (1982) with the following modifications: instead of Tris buffer, 0.1 M Na+/K+ phosphate buffer (pH 7.4) was used. After ultracentrifugation, the supernatant was discarded and the residue reconstituted in 4 times its volume of a buffer solution containing 0.1 mM Na+/K+ phosphate buffer/glycerol (4:1, v/v) and stored at −80°C. Protein concentrations were determined using the bicinchoninic acid method described by Smith et al. (1985). Bovine serum albumin was used as the standard. CYP concentrations were determined using the method of Omura and Sato (1964) following the protocol of Estabrook and Werringloer (1978).

Metabolism by Human Liver Microsomes.

Microsomal protein (0.05–0.2 mg), buffer, and K11777 (in acetonitrile/water, 9:1 v/v, final concentration 5–150 μM) were preincubated for 5 min. The reaction was started by adding 0.4 ml of an NADPH-generating system, containing 5 mM EDTA, 25 mM MgCl2, 2.5 mM NADP, 45 mM isocitric acid, and 1.75 U isocitrate dehydrogenase in 0.1 M Na+/K+ phosphate buffer (pH 7.4). Inhibition studies with methionine to assess FMO3 activity (vide infra), because FMO3 is inactivated by preincubation without NADPH, were carried out as follows: the substrate, inhibitor (if applicable), NADPH-generating system, and buffer were preincubated at 37°C for 4 min. The reaction was started by addition of human liver microsomes (volume 5 μl). The final volume of the incubation samples was always 1 ml. The mixtures were incubated for 10 min and the reaction was stopped by protein precipitation after addition of 0.4 ml of acetonitrile containing 100 ng of the internal standard K11002.

Metabolism by cDNA-Expressed, Isolated CYP Enzymes and FMO3.

Microsomes from baculovirus-infected insect cells expressing the following human monooxygenases were used: CYP1A1 (lot 10), 1A2 (lot 12), 2B6 (lot 2), 2C8 (lot 2), 2C9 (Arg144) (lot 5), 2C19 (lot 3), 2D6 (lot 14), 2E1 (lot 5), 3A4 (lot 15/20), 3A5 (lot 8), and FMO3 (lot 8) (all from Gentest). CYP enzymes, except 3A5, 1A1, 1A2, and 2D6, were coexpressed with NADPH-CYP reductase and cytochromeb5. The metabolism assays (final volume of 1 ml) were carried out as described for the human liver microsomes using 100 μM K11777, except when FMO3-mediated metabolism was assessed: The assays containing the substrate, inhibitor (if applicable), NADPH-generating system, and buffer were preincubated at 37°C for 4 min. The reaction was started by addition of FMO3 (5 μl). All assays were incubated for 10 min.

Extraction and Quantification of K11777 and Its Metabolites.

After protein precipitation by addition of 0.4 ml of acetonitrile to the reaction mixtures, samples were vortexed for 10 s and centrifuged at 4°C and 10,900g for 5 min. The protein precipitation reagent acetonitrile also contained the internal standard K11002 (250 ng/ml). Supernatants were transferred into 1.8-ml HPLC vials (Hewlett-Packard, Palo Alto, CA).

K11777 and its metabolites were quantified by HPLC/electrospray-MS in combination with an on-line column switching extraction step using a Hewlett-Packard series 1100 liquid chromatograph that consisted of a G1313A autosampler, G1312A binary pump, G1322A degasser, and G1316A column thermostat, connected to a 5989B mass spectrometer (all Hewlett-Packard) equipped with an Iris hexapole ion guide (Analytica of Branford, Branford, CT) by a 59987A electrospray interface (Hewlett-Packard). Solvent for the on-line extraction column was delivered by an additional binary HPLC pump (Perkin-Elmer, Norwalk, CT), controlled via the external contacts of the HP1100 HPLC system. The HPLC/MS system was controlled and data were processed using ChemStation software, revision A04.02 (Hewlett-Packard). The column switching HPLC/MS system was set up as described in detail by Kirchner et al. (1999). In brief, 100-μl samples were injected onto a 10 × 2-mm extraction column filled with Hypersil ODS-1 of 10-μm particle size (Shandon, Chadwick, UK). Samples were washed with a mobile phase of 2 mM NH4-acetate/methanol (8:2, v/v). The flow was 6 ml/min. After 1 min, the switching valve was activated and the analytes were eluted in the backflush mode from the extraction column onto a 35 × 4.6-mm analytical column (Phenomenex, Torrance, CA) filled with Capcell PAK UG-120A CN material, 5-μm particle size (Shiseido, Tokyo, Japan). The mobile phase consisted of 2 mM ammonium acetate and methanol. The following gradient was run: 0 min, 60% methanol; 1 min, 60% methanol; 10 min, 88% methanol. For the following 2.5 min, the column was washed with 95% methanol and was reequilibrated to the starting conditions for 3 min. The flow was 0.4 ml/min and the column temperature 40°C. The electrospray interface was adjusted to the following parameters (nomenclature according to the ChemStation software): nebulizer gas: nitrogen of grade 5.0 purity, 80 psi; drying gas: nitrogen 5.0, flow, 40 (arbitrary units), 330°C; Vcap, −3500 V; Vend, −2600 V; Vcyl, −5000 V, capillary exit voltage, 88 V. The following parameters were used for mass spectrometry analysis: quadrupole temperature, 120°C; multiplier voltage, 2,275 V; X-ray, 10,000 V. Positive ions [M+H]+ of K11777 and its metabolites gave the best signal-to-noise ratio and were recorded in the single ion mode:m/z = 575.2 for K11777 (retention time 8.7 min); m/z = 591.2 for N-oxide K11777 (retention time 7.4 min) and β-hydroxy-homoPhe K11777 (retention time 8.2 min); and m/z = 561.2 for N-desmethyl K11777 (retention time 11.8 min). The dwell time for each ion was 100 ms.

The assay had the following specifications for K11777: lower limit of detection (signal/noise = 3:1), 0.5 ng/ml; lower limit of quantitation, 1.25 ng/ml; linearity, 1.25–500 ng/ml (r2 = 0.996); and interassay variability (3 days, n = 15), 6.2% for 20 ng/ml and 5.3% for 160 ng/ml. The specifications for the metabolites, as checked by spiking inactivated reaction mixtures with isolated metabolites, did not differ from those of K11777. The recovery of K11777 and its metabolites was >80%. Within-batch stability of extracted samples was established for at least 24 h and all samples were analyzed within this time period.

Determination of the Apparent Michaelis-Menten ConstantsKm and Vmax.

Liver microsomes (n = 4 for each concentration), CYP (n = 3), or FMO3 (n = 3) enzyme preparations were incubated with the following K11777 concentrations: 5, 10, 15, 20, 30, 40, 60, 80, and 100 μM (for FMO3 additionally 150 μM) as described above, for 10 min. Higher K11777 concentrations could not be tested due to limited solubility above 150 μM. Heat-denaturated microsomes, control microsomes from insect cells (Gentest), and incubation mixtures without NADPH were used as negative controls. Microsomal preparations were inactivated by incubation at 95°C for 5 min. Heat-denatured microsomal preparations were also used to establish stability of the study drug during incubation and as matrix for quality control samples during study sample analysis, but were not used to assess the relative contributions of FMO3 and CYP to K11777 metabolism. Therefore, conditions were chosen that inactivated all drug-metabolizing enzymes. After heat inactivation, microsomes or isolated enzymes did not show any detectable activity as assessed using standard substrates (CYP3A, lovastatin; CYP2D6, (±)-bufuralol; FMO3, methionine). Km andVmax were determined after data fitting using Hanes-Woolf plots. Data were fitted using the SigmaPlot software (version 5.0; SPSS Inc., San Rafael, CA).

Structural Evaluation of K11777 Metabolites by MS/MS.

Metabolites were isolated after incubation of K11777 with human liver microsomes and the NADPH-generating system using semipreparative HPLC. Extraction and setup of the HPLC system were as described for HPLC/MS analysis of K11777, except that an injection volume of 400 μl and that in place of the mass spectrometer a G1314A variable wavelength UV monitor (Hewlett-Packard, Palo Alto, CA) were used. Chromatograms were recorded at a UV wavelength of 210 nm. Metabolite fractions were collected manually and purity was established by HPLC/UV and HPLC/MS analysis. Structures of the isolated metabolites were evaluated by electrospray-MS/MS in combination with collision-induced dissociation and analysis of the fragmentation patterns using a PE SCIEX API III triple-quadrupole mass spectrometer (Perkin-Elmer/Applied Biosystems, Foster City, CA) equipped with a heated nebulizer inlet (Thornhili, Ontario, Canada). The isolated metabolite fractions were introduced into the electrospray interface by continuous infusion (5 μl/min) via a syringe pump (Harvard Apparatus, Holliston, MA). Positive ions were recorded. Scans were acquired over a mass range of m/z 70 to 600. For collision-induced dissociation, the collision energy was set to 60 eV and air was used as the collision gas.

Identification of the Monooxygenases Involved in K11777 Metabolism.

The monooxygenases involved were identified by metabolism of K11777 by isolated, recombinant CYP and FMO enzymes (vide supra). In addition, K11777 metabolism was inhibited by specific chemical inhibitors and antibodies. The effects of the specific CYP3A inhibitors itraconazole and cyclosporine (Guengerich, 1995) on K11777 metabolism were assessed. Cyclosporine was dissolved in methanol and itraconazole in acetonitrile/dimethyl sulfoxide (2:1, v/v). Inhibitor solution (10 μl) resulting in final concentrations of 0.1, 1, 5, and 20 μM (n = 4) for itraconazole and 0.1, 1, 10, and 50 μM (n = 4) for cyclosporine was added to the liver microsomal preparations containing 10 μM K11777. Microsomes to which 10 μl of the vehicle was added were used as controls. The mixtures were incubated and analyzed as described above.

To exclude that itraconazole, which like K11777 possesses a piperazine ring, interacted with FMO3, 15 μM itraconazole was incubated with 0.25 mg of isolated cDNA-expressed FMO3 and the NADPH-generating system at 37°C for 60 min. Itraconazole and its potential N-oxide metabolite were measured using HPLC/MS. Ions [M+H]+ were detected in the selected ion mode:m/z = 705.5 for itraconazole andm/z = 721.5 for N-oxide itraconazole. Under these conditions, no itraconazole metabolism was detected. In addition, 15 μM itraconazole failed to significantly inhibit N-oxide K11777 formation by isolated FMO3 enzymes, strongly suggesting that itraconazole was neither a FMO3 substrate nor an inhibitor in our assays.

For inhibition of K11777 metabolism by specific CYP antibodies, 100 μg of microsomal protein isolated from human liver (pool of four) in 20 μl of 0.1 M Tris buffer (pH 7.4) was preincubated on ice with 0, 1, 2, 5, or 10 μl of CYP3A4/5 antibody solution (1 μl ≅ 10 μg of protein) (Gentest) for 25 min. Then buffer (final volume 1 ml), K11777 (final concentration 50 μM), and the NADPH-generating system were added. Samples were incubated and extracted as described above. To exclude a nonspecific effect of high protein concentrations, control samples were incubated with rabbit preimmune serum (Gentest) at the same concentration as the highest antibody concentration used (n = 4).

To estimate the relative contributions of CYP2D6 and FMO3 to the overall metabolism of K11777, K11777 metabolism by human liver microsomes was inhibited by addition of 50 μM (±)-bufuralol (CYP2D6 inhibitor, Gut et al., 1984) or 100 mM methionine (FMO3 substrate, Ripp et al., 1999). As controls, K11777 metabolism catalyzed by isolated CYP3A, CYP2D6, or FMO3 enzymes was inhibited using 50 μM (±)-bufuralol or 100 mM methionine. The effect of the inhibitors and antibodies on K11777 metabolite formation was statistically evaluated by analysis of variance (GLM procedure, SAS, version 6.12; SAS Institute, Cary, NC).

Inhibition of the CYP3A-Mediated Lovastatin-6′β-hydroxylation by K11777 and K11002.

To assess and compare inhibition of CYP3A-mediated metabolism by K11777 and K11002, their effect on the CYP3A-dependent 6′β-hydroxylation of lovastatin (Wang et al., 1991) was evaluated. K11777 and K11002 (0, 0.012, 0.06, 0.12, 1.2, 12, and 120 μM) were incubated with 25 μg of human microsomal CYP3A4 protein with or without (control) an NADPH-generating system at 37°C. After a preincubation period of 30 min, the reaction was started by addition of 1 mmol of NADPH and lovastatin (final concentration in the assay 100 μM). For comparison, lovastatin was added together with K11777 or K11002 and the NADPH-generating system without a preincubation period. Lovastatin and its metabolite 6′β-hydroxy lovastatin were quantified by HPLC as we described previously (Jacobsen et al., 1999).

Inhibition of the CYP2D6-Mediated (±)-Bufuralol-1′-hydroxylation by K11777.

To evaluate the effect of K11777 on the CYP2D6-mediated metabolism of (±)-bufuralol (Gut et al., 1984), K11777 (0, 0.012, 0.06, 0.12, 1.2, 12, and 120 μM) was incubated with 25 μg of isolated human CYP 2D6, and with NADPH-generating system at 37°C. After a preincubation period of 30 min, (±)-bufuralol (final concentration 50 μM) and 1 mmol of NADPH were added. As controls, the inhibition assay was carried out with a 30-min preincubation period without NADPH (n= 4) and without preincubation (n = 4). (±)-Bufuralol and its 1′-hydroxy metabolite were quantified by HPLC/MS as described for K11777 with the following gradient: 0 min 70% methanol, 6 min 95% methanol. The column was washed with 95% methanol for 2 min and reequilibrated to the starting conditions within 4.5 min. Bufuralol and 1′-hydroxy bufuralol [M+H]+ were recorded in the single ion mode: m/z = 262.2, bufuralol (retention time 8.4 min); m/z = 278.2, 1′-hydroxyl bufuralol (retention time 7.0 min). The dwell time for each ion was 500 ms.

Results

In Vitro Metabolism of K11777 by Human Liver Microsomes.

After incubation of K11777 with human liver microsomes, three metabolites were detected by HPLC/UV as well as HPLC/MS in the scan and single ion mode (Fig. 2). These peaks were neither detected in control incubations containing heat-inactivated microsomes nor during incubations with microsomes without NADPH. Two metabolites with a molecular ion [M+H]+ of m/z = 591.2 were detected, which is a mass difference of +16 amu compared with K11777. Their HPLC retention times of 7.4 and 8.2 min (Fig. 2) were different, indicating oxidation of K11777 at two different positions. A third metabolite had a retention time longer than K11777 (11.8 versus 8.7 min). Its molecular ion ofm/z = 561 indicated demethylation in one position. Additional metabolites were not detected (limit of detection 0.5 ng/ml). The formation of all three metabolites was linear for incubation periods up to 15 min and microsomal protein concentrations from 0.025 to 0.2 mg/ml. Apparent Km, apparent Vmax, and intrinsic formation clearances (CLint,Vmax/Km) for the three K11777 metabolites are shown in Table1. The CLint values indicated that formation of the N-oxide of K11777 (structural identification, vide infra) was the main metabolism reaction with a mean CLint 3.7-fold higher than for β-hydroxy homoPhe K11777 and 2.5-fold higher than forN-desmethyl K11777.

Figure 2
Figure 2

Ion chromatograms after incubation of K11777 with human cDNA-expressed 3A4 supersomes.

Incubation of 100 μM K11777 with 25 pmol of human cDNA-expressed CYP3A4 and an NADPH-generating system at 37°C for 30 min. Single ions were recorded [m/z = 575, K11777 (retention time 8.7 min); m/z = 591,N-oxide K11777 (retention time 7.4 min) and β-hydroxy homoPhe K11777 (retention time 8.2 min); andm/z = 561,N-desmethyl K11777 (retention time 11.8 min)]. Ion chromatograms were overlaid. Representative ion chromatograms (total,n = 3) are shown. The run time of the ion-chromatograms (x-axis) is given in minutes.

View this table:
Table 1

Apparent Km and apparent Vmax values and intrinsic formation clearances (CLint = Vmax/Km) for the formation of K11777 metabolites by human liver microsomes

Structural Evaluation of K11777 Metabolites by MS/MS.

The mass spectra and proposed fragmentation pathways of K11777 and its isolated metabolites are shown in Fig. 3. Corresponding fragments of metabolites and K11777 are compared in Table2. Analysis of the fragmentation patterns and comparison with K11777 suggested the following structures for the three metabolites.

Figure 3
Figure 3

Tandem mass spectra of K11777 (A) and its metabolite β-hydroxy homoPhe K11777 (B).

View this table:
Table 2

Fragmentation pattern of K11777 after collision induced dissociation and comparison with corresponding fragments of its metabolites

N-oxide K11777 (retention time 7.4 min).

In comparison to K11777 (Fig. 3A), no fragment withm/z = 101, which comprises theN-methyl- piperazine ring, but two new fragments withm/z = 99 and m/z= 117 were detected (Table 2). The fragmentm/z = 117 appeared to represent hydroxylation at the N-methyl-piperazine ring. This was confirmed by the other new fragment at m/z = 99, which was most likely m/z = 117- H2O. All fragments not containing theN-methyl-piperazine ring were identical with those of K11777, whereas all fragments containing the N-methyl- piperazine ring had m/z values 16 amu greater than the corresponding K11777 fragments (Table 2). Further structural identification based solely upon MS/MS analysis was not possible.

β-Hydroxy homoPhe K11777 (retention time 8.2 min) (Fig.3B).

The molecular ion [M+H]+ of 591.2 indicated hydroxylation in one position. The fragment ofm/z = 274 showed that theN-methyl-piperazine Phe- part of the molecule was unchanged (Table 2) and hydroxylation occurred at the homoPhe-vinylsulfone-phenyl part of K11777. In comparison to K11777, two additional fragments atm/z = 331 [M+H]+ andm/z = 368 [M+K]+were present (Fig. 3, A and B). These could be explained by loss of the homoPhe group from fragment m/z = 434, suggesting that the hydroxylation had most likely occurred at the first C atom of the homoPhe residue of K11777. This fragment was probably caused by the loss of homoPhe (105 m/z) from fragment m/z = 434 (N-methyl-piperazine-Phe-homoPhe-vinyl-residue). The formation of this additional fragment could be explained by α-cleavage of the hydroxy-homoPhe residue due to hydroxylation at the β-C-atom of this residue.

N-desmethyl K11777 (retention time 11.8 min).

All fragments not containing the N-methyl-piperazine ring were identical with those of K11777, whereas all fragments containing the N-methyl-piperazine ring had m/zvalues 14 amu lower than the corresponding K11777 fragments (Table 2). Fragment 101 m/z (N-methyl-piperazine) was shifted to 87 m/z, indicating that the structure of this metabolite was most likely N-desmethyl K11777.

Identification of the Monooxygenases Involved in K11777 Metabolism.

K11777 metabolism by isolated CYP3A4 and 2D6 was linear during incubation periods up to 10 min; metabolism by FMO3 for 15 min. Reaction kinetics was linear for protein concentrations from 0.025 to 0.2 mg/ml. The K11777 metabolite profiles after incubation with human liver microsomes and cDNA-expressed human CYP3A4 were identical, as confirmed by mass spectrometry. In addition, in vitro incubations indicated that CYP2D6 and FMO3 might constitute major pathways for the formation of N-desmethyl and N-oxide K11777, respectively. The apparent Km and apparentVmax values are shown in Table3. It must be taken into account that the cDNA-expressed isolated CYP and FMO preparations varied significantly in enzyme activities, which also did not reflect the enzymes' relative in vivo activities. Therefore, apparentVmax data in Table 3 and formation rates in Table 4 only allow comparison of the formation of different metabolites by the same enzyme but not comparison of the formation of metabolites by different enzymes. In addition to CYP3A, CYP2D6, and FMO3, incubation of all other CYP enzymes tested with K11777 resulted in detectable amounts of one or more K11777 metabolites (Table 4).

View this table:
Table 3

Apparent Km, Vmax and intrinsic clearances (CLint) for the formation of K11777 metabolites by cDNA expressed CYP enzymes and cDNA expressed FMO3

View this table:
Table 4

K11777 metabolite formation rates by human cDNA-expressed CYP enzymes and cDNA-expressed FMO3.

Antibodies against human CYP3A enzymes inhibited K11777 metabolism by human liver microsomes with an IC50 of 50 μg of antibody/100 μg human liver microsomal protein (Fig.4). Compared with uninhibited controls, 100 μg of antibody/100 μg human liver microsomal protein reducedN-oxide K11777, β-hydroxy-homoPhe K11777, andN-desmethyl K11777 formation by 55 ± 10, 68 ± 8, and 61 ± 11%, respectively (Fig. 4). In the presence of rabbit preimmune serum at the same protein concentration as the highest antibody concentration used, N-oxide K11777 formation was 91.8 ± 1.4% and N-desmethyl K11777 formation was 97.1 ± 4.2% of controls without serum.

Figure 4
Figure 4

CYP3A4/5 antibody concentration-dependent immunoinhibition of K11777 metabolite formation by human liver microsomes.

Microsomal protein (100 μg) and different concentrations of antibody (Gentest) were preincubated for 25 min. The reaction was started by addition of 50 μM K11777 and NADPH-generating system. The assays were incubated at 37°C for 15 min. Concentrations of K11777 metabolites were determined by HPLC/MS as described under Materials and Methods. Each data point represents the mean ± standard deviation (n = 3).

The addition of 50 μM of the specific CYP3A substrate and inhibitor cyclosporine reduced the formation of N-oxide, β-hydroxy-homoPhe, and N-desmethyl K11777 by 54 ± 8, 61 ± 9, and 62 ± 9%, and the addition of 1 μM itraconazole by 79 ± 4, 83 ± 4, and 82 ± 7%, respectively (mean ± standard deviation, n = 4).

(±)-Bufuralol (50 μM) inhibited N-demethylation of K11777 by human liver microsomes (31.9 ± 2.5%, n = 4), whereas formation of N-oxide K11777 was not affected. The same bufuralol concentration inhibited N-demethylation of K11777 by isolated CYP2D6 enzymes by 81.2 ± 0.6% (n = 3) but had no effect on K11777-mediatedN-demethylation or N-oxidation by isolated CYP3A enzymes. Methionine (10 mM) reduced the FMO3-catalyzed formation ofN-oxide K11777 by 93.8 ± 0.35% (n = 4). In human liver microsomes, addition of 100 mM methionine decreasedN-oxidation by 27.5 ± 8.1% (n = 4). However, when 100 mM methionine was incubated with isolated CYP3A, formation of N-oxide K11777 was reduced to the same extent (26.9 ± 1.8%, n = 4), indicating nonspecific inhibition of CYP3A-mediated metabolism at the methionine concentrations used.

Inhibition of CYP3A- and CYP2D6-Mediated Metabolism by K11777 and K11002.

When preincubated with microsomes and NADPH for 30 min, K11777 was a potent (half-maximal inhibition constant IC50 = 0.06 μM) and an efficient inhibitor (maximum inhibition 85%) of the CYP3A-dependent formation of 6′β-hydroxy lovastatin (Fig.5A). This inhibition was NADPH-dependent. Incubation without NADPH and without preincubation both inhibited 6′β-hydroxy lovastatin formation with an IC50of 50 μM (Fig. 5A). The results of the controls with preincubation without NADPH and without preincubation were similar, indicating that the 30-min preincubation period did not significantly affect CYP3A activity. Like K11777, K11002 proved to be an NADPH-dependent inhibitor of 6′β-hydroxy lovastatin formation (Fig. 5B). However, its IC50 was 0.5 μM and thus 8.4-fold higher that of K11777. In contrast to CYP3A, K11777 failed to inhibit the CYP2D6-mediated bufuralol-1′-hydroxylation either with or without preincubation with NADPH. Again, the controls indicated that the 30-min preincubation period did not significantly affect CYP2D6 activity.

Figure 5
Figure 5

K11777 (A) and K11002 (B) concentration- and NADPH-dependent inhibition of 6′β-hydroxy lovastatin metabolite formation by human cDNA-expressed CYP3A4.

Isolated CYP3A (25 μg) and different concentrations of K11777 (A) or K11002 (B) were preincubated for 30 min at 37°C with and without NADPH. The reaction was started by addition of 100 μM lovastatin and NADPH. The assays were incubated at 37°C for 10 min. Concentrations of 6′β-hydroxy lovastatin were determined by HPLC/UV as described under Materials and Methods. Each data point represents the mean ± standard deviation (n = 4).

Discussion

Human liver microsomes metabolized K11777 to three metabolites. The metabolic pathways proposed on the basis of our study results are shown in Fig. 6. For two metabolites,N-desmethyl K11777 and β-hydroxy-homoPhe K11777, structures could be proposed based on the MS/MS fragmentation patterns. For the third metabolite (HPLC retention time 7.4 min), the fragmentation pattern suggested hydroxylation at theN-methyl-piperazine ring. This metabolite was generated by incubation of K11777 with FMO3. FMOs are microsomal enzymes that catalyze the NADPH- and O2-dependent oxidation of several nitrogen-, sulfur-, selenium- and phosphorus-containing compounds (Ziegler, 1993). Currently, five FMO enzymes, FMO1 to 5 have been identified, each exhibiting its own unique species- and tissue-dependent expression (Hines et al., 1994). The predominant FMO enzyme in adult human liver is FMO3, which was the reason we used it in our study. It is known that FMO-mediatedN-oxidation occurs preferentially at the N-methyl nitrogen atom of N-methyl-N′-substituted piperazines, e.g., N-oxidation of clozapine (Tugnait et al., 1997) and the N-methyl-piperazine-substituted phenothiazine drug trifluoperazine (Sofer and Ziegler, 1978). Based on these data, the most likely structure of the K11777 metabolite hydroxylated at theN-methyl-piperazine ring is N-oxide K11777.

Figure 6
Figure 6

Proposed metabolic pathways of K11777 in human liver microsomes.

Although the pH optimum of FMO3 is between 9 and 9.5 (Lang et al., 1998), these conditions are not relevant in vivo in humans and we adjusted our assays assessing FMO3 activity to pH 7.4. It must also be noted that only apparent enzyme kinetic parameters could be reported for assays based on liver microsomal preparations. Microsomes do not represent an isolated enzyme but a mixture of different enzymes and, as demonstrated by our study, more than one CYP enzyme is involved in the metabolism of K11777.

Finding that K11777 was metabolized by isolated CYP3A enzymes, and that the specific CYP3A substrate cyclosporine and antibodies against human CYP3A4/5 were effective inhibitors of K11777 metabolite formation in human liver microsomes, led us to conclude that CYP3A is primarily involved in K11777 metabolism. Although all other tested cDNA-expressed CYP enzymes were able to generate at least one of the three K11777 metabolites, the inhibition experiments with bufuralol and methionine indicated that CYP3A4 is responsible for at least 66% of the formation of N-desmethyl K11777, whereas FMO3 does not seem to be a relevant metabolic pathway. The choice of effective inhibitor concentrations was confirmed by the inhibition experiments with isolated CYP2D6 and FMO3, showing >80% inhibition of the isolated enzymes. Based on this information and the intrinsic formation clearances (CLint, Table 1) in human liver microsomes, it can be estimated that CYP3A is responsible for more than 80% of the total metabolism of K11777.

Nevertheless, it is interesting that K11777 is a substrate for such a wide variety of CYP enzymes. It is well established thatN-demethylation reactions are catalyzed by many different CYP enzymes, such as 1A2, 2C9, 2C19, 2D6, 2E1, and 3A4 (Coutts et al., 1994). For example, the N-demethylation of clozapine, a drug also containing an N-methyl-piperazine ring, is catalyzed by CYP enzymes 1A2, 2C9, 2C19, 2D6, and 3A4 (Linnet and Olesen, 1997). In our study, we used a sensitive HPLC/MS assay that enabled us to detect metabolite concentrations as low as 0.5 ng/ml.

In addition to being a substrate, K11777 was a potent inhibitor of CYP3A with an inhibition constant in the nanomolar range. Inhibition of CYP3A by K11777 was most potent and efficient after a preincubation period of 30 min in the presence of NADPH, indicating that K11777 is a mechanism-based inhibitor. In contrast, K11777 did not inhibit the catalytic activity of CYP2D6, the other major CYP enzyme involved in K11777 metabolism. The structurally related protease inhibitor K11002 was also a mechanism-based inactivator of CYP3A, indicating that the phenylalanine-homo-phenylalanine-vinylsulfone-phenyl part of the molecule rather than the morpholine-urea ring of K11002 or theN-methyl-piperazine ring of K11777 was involved.

As mentioned above, the structural difference between K11777 and K11002 is the replacement of the morpholine-urea ring of K11002 by anN-methyl-piperazine ring. This modification resulted in significant differences of the in vitro drug metabolism of K11777 in comparison to K11002. Zhang et al. (1998a) identified the three major metabolites of K11002 as hydroxy metabolites. The hydroxylation positions were at the phenylalanine, homo-phenylalanine, and the phenyl residue. The morpholine-urea ring was not involved. Introduction of theN-piperazine ring shifted the main target of metabolism reactions from the phenylalanine-homo-phenylalanine-vinylsulfone-phenyl part of the molecule to the N-methyl-piperazine ring. In contrast to K11002, two of the three K11777 metabolism sites were at the N-methyl-piperazine ring. Also, theN-methyl-piperazine ring allowed for additional metabolic pathways, e.g., via FMO3. In comparison to K11002, K11777 exhibited smaller Km values for its metabolism by human liver microsomes. Those described for K11002 ranged between 38 to 52 μM (Zhang et al., 1998a) and were more than 2-fold higher than those found for K11777 here.

In conclusion, the N-methyl-piperazine ring is the major metabolism site of K11777, and N-oxygenation,N-demethylation, and hydroxylation by CYP3A4 constitute the major metabolic pathways (Fig. 6). Intestinal CYP3A enzymes have been shown to play an important role in the low and variable oral bioavailability of drugs that are CYP3A substrates (Benet et al., 1996), and may be one of the factors responsible for the low oral bioavailability of K11777. In addition, many clinically important drugs that are CYP3A substrates, inhibitors, and/or inducers (Rendic and DiCarlo, 1997) may interact with K11777 pharmacokinetics. Vice versa, we have shown that K11777 is a potent and effective mechanism-based inhibitor of CYP3A and thus can be expected to affect the pharmacokinetics of coadministered drugs, if they are CYP3A substrates.

Footnotes

  • Send reprint requests to: Leslie Z. Benet, Ph.D., Professor, Department of Biopharmaceutical Sciences, School of Pharmacy, University of California, San Francisco, CA 94143-0446. E-mail: benet{at}itsa.ucsf.edu

  • This study was supported in part by National Institutes of Health Grant CA72006 (to L.Z.B), the Deutsche Forschungsgemeinschaft Grant Ch 95/6-2 (to U.C.), and National Institutes of Health NCRR Grant 01614.

  • Abbreviations used are::
    K11777
    N-methyl-piperazine-Phe-homoPhe-vinylsulfone-phenyl
    CYP
    cytochrome P450
    FMO3
    flavin-containing monooxygenase 3
    MS
    mass spectrometry
    HLF
    human liver (female)
    HLM
    human liver (male)
    LOD
    below limit of detection
    • Received November 11, 1999.
    • Accepted July 26, 2000.

References

  1. ↵
    1. Benet LZ,
    2. Wu CY,
    3. Hebert MF,
    4. Wacher VJ
    (1996) Intestinal drug metabolism and antitransport processes: A potential paradigm shift in oral drug delivery. J Controlled Release 39:139–143.
  2. ↵
    1. Cazzulo JJ,
    2. Stoka V,
    3. Turk V
    (1997) Cruzipain, the major cysteine protease from the protozoan parasite Trypanosoma cruzi. Biol Chem 378:1–10.
    OpenUrlPubMed
  3. ↵
    1. Chen WT
    (1992) Membrane proteases: Roles in tissue remodeling and tumor invasion. Curr Opin Cell Biol 4:802–809.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Coutts RT,
    2. Su P,
    3. Baker GB
    (1994) Involvement of CYP2D6, CYP3A cytochrome P-450 isozymes in N-dealkylation reactions. J Pharmacol Toxicol Methods 31:177–186.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Elliott E,
    2. Sloane BF
    (1996) The cysteine protease cathepsin B in cancer. Perspect Drug Discovery Des 6:12–32.
  6. ↵
    1. Estabrook RW,
    2. Werringloer J
    (1978) The measurement of difference spectra: Application to the cytochromes of microsomes. Methods Enzymol 52:212–220.
    OpenUrlPubMed
  7. ↵
    1. Filardi LS,
    2. Brener Z
    (1987) Susceptibility and natural resistance of Trypanosoma cruzi strains to drugs used clinically in Chagas' disease. Trans R Soc Trop Med Hyg 81:755–759.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Gorla NB,
    2. Ledesma GP,
    3. Barbieri,
    4. Larripa IB
    (1988) Assessment of cytogenetic damage in chagasic children treated with benznidazole. Mutat Res 206:217–220.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Hayes AW
    1. Guengerich FP
    (1982) Microsomal enzymes involved in toxicology- Analysis and separation. in Principles and Methods of Toxicology, ed Hayes AW (Raven Press, New York), pp 609–637.
  10. ↵
    1. Ortiz de Montellano PR
    1. Guengerich FP
    (1995) Human cytochrome P450 enzymes. in Cytochrome P450: Structure, Mechanism, and Biochemistry, ed Ortiz de Montellano PR (Plenum Press, New York), 2nd ed. pp 473–535.
  11. ↵
    1. Gut J,
    2. Gasser R,
    3. Dayer P,
    4. Kronbach T,
    5. Catin T,
    6. Meyer UA
    (1984) Debrisoquine-type polymorphism of drug oxidation: Purification from human liver of a cytochrome P-450 isozyme with high activity for bufuralol hydroxylation. FEBS Lett 173:287–290.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Harth G,
    2. Andrews N,
    3. Mills AA,
    4. Engel JC,
    5. Smith R,
    6. McKerrow JH
    (1993) Peptide-fluoromethyl ketones arrest intracellular replication and intercellular transmission of Trypanosoma cruzi in vitro. Mol Biochem Parasitol 58:17–24.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Hines RN,
    2. Cashman JR,
    3. Philpot RM,
    4. Williams DE,
    5. Ziegler DM
    (1994) The mammalian flavin-containing monooxygenases: Molecular characterization and regulation of expression. Toxicol Appl Pharmacol 125:1–6.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Jacobsen W,
    2. Kirchner G,
    3. Hallensleben K,
    4. Mancinelli L,
    5. Deters M,
    6. Hackbarth I,
    7. Benet LZ,
    8. Sewing KF,
    9. Christians U
    (1999) Comparison of the cytochrome P-450-dependent metabolism and drug interactions of the 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors lovastatin and pravastatin in the liver. Drug Metab Dispos 27:173–179.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Kirchhoff LV
    (1993) American Trypanosomiasis (Chagas' disease) - A tropical disease now in the United States. New Engl J Med 329:639–644.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Kirchner GI,
    2. Vidal C,
    3. Jacobsen W,
    4. Franzke A,
    5. Hallensleben K,
    6. Christians U,
    7. Sewing KF
    (1999) Simultaneous on-line analysis of sirolimus (rapamycin) and ciclosporin in blood by liquid chromatography/electrospray mass spectrometry. J Chromatogr B 721:285–294.
  17. ↵
    1. Lang DH,
    2. Yeung CK,
    3. Peter RM,
    4. Ibarra C,
    5. Gasser R,
    6. Itagaki K,
    7. Philipot RM,
    8. Rettie AE
    (1998) Isoform specificity of trimethylamine N-oxygenation by human flavin-containing monooxygenase (FMO) and P450 enzymes. Selective catalysis by FMO3. Biochem Pharmacol 56:1005–1012.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Libow LF,
    2. Beltranni VP,
    3. Silvers DN,
    4. Grossman ME
    (1991) Post-cardiac transplant reactivation of Chagas' disease diagnosed by skin biopsy. Cutis 48:7–40.
  19. ↵
    1. Linnet K,
    2. Olesen OV
    (1997) Metabolism of clozapine by cDNA-expressed human cytochrome P450 enzymes. Drug Metab Dispos 25:524–527.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Omura T,
    2. Sato R
    (1964) The carbon monoxide-binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J Biol Chem 239:2370–2378.
    OpenUrlFREE Full Text
  21. ↵
    1. Parada H,
    2. Carrasco HA,
    3. Anez N,
    4. Inglessis I
    (1997) Cardiac involvement is a constant finding in acute Chagas' disease: A clinical, parasitological and histopathological study. Int J Cardiol 60:49–54.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Rendic S,
    2. DiCarlo FJ
    (1997) Human cytochrome P450 enzymes: A status report summarizing their reactions, substrates, inducers, and inhibitors. Drug Metab Rev 29:413–580.
    OpenUrlPubMed
  23. ↵
    1. Ripp SL,
    2. Itagaki K,
    3. Philpot RM,
    4. Elfarra AA
    (1999) Species and sex differences in expression of flavin-containing monooxygenase form 3 in liver and kidney microsomes. Drug Metab Dispos 27:46–52.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Smith PK,
    2. Krohn RI,
    3. Hermanson GT,
    4. Mallia AK,
    5. Gartner FH,
    6. Provenzano MD,
    7. Fujimoto EK,
    8. Goeke NM,
    9. Olson BJ,
    10. Klenk DC
    (1985) Measurement of protein using bicinchoninic acid. Anal Biochem 150:76–85.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Sofer SS,
    2. Ziegler DM
    (1978) Microsomal mixed-function amine oxidase. Oxidation products of piperazine-substituted phenothiazine drugs. Drug Metab Dispos 6:232–239.
    OpenUrlAbstract
  26. ↵
    1. Tugnait M,
    2. Hawes EM,
    3. McKay G,
    4. Rettie AE,
    5. Haining RL,
    6. Midha KK
    (1997) N-Oxygenation of clozapine by flavin-containing monooxygenase Drug Metab Dispos 25:524–527.
  27. ↵
    1. Wang RW,
    2. Kari PH,
    3. Lu AYH,
    4. Thomas PE,
    5. Guengerich FP,
    6. Vyas KP
    (1991) Biotransformation of lovastatin. IV. Identification of cytochrome P450 3A proteins as the major enzymes responsible for oxidative metabolism of lovastatin in rat and human liver microsomes. Arch Biochem Biophys 290:355–361.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Ziegler DM
    (1993) Recent studies on the structure and function of multisubstrate flavin-containing monooxygenases. Annu Rev Pharmacol Toxicol 33:179–199.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Zhang Y,
    2. Guo X,
    3. Lin ET,
    4. Benet LZ
    (1998a) Overlapping substrate specificities of cytochrome P450 3A and P-glycoprotein for a novel cysteine protease inhibitor. Drug Metab Dispos 26:360–366.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Zhang Y,
    2. Hsieh Y,
    3. Izumi T,
    4. Lin ET,
    5. Benet LZ
    (1998b) Effects of ketoconazole on the intestinal metabolism, transport and oral bioavailability of K02, a novel vinylsulfone peptidomimetic cysteine protease inhibitor and a P450 3A, P-glycoprotein dual substrate, in male Sprague-Dawley rats. J Pharmacol Exp Ther 287:246–252.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleArticle

In Vitro Evaluation of the Disposition of A Novel Cysteine Protease Inhibitor

Wolfgang Jacobsen, Uwe Christians and Leslie Z. Benet
Drug Metabolism and Disposition November 1, 2000, 28 (11) 1343-1351;
Reddit logo Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement