Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Pimobendan, 4,5-dihydro-6-(2-(4-methoxyphenyl)-1H-benzimidazol-5-yl)-5-methyl-3(2-H)-pyridazinone, has both inotropic and peripheral vasodilator properties, and these properties are attributed to the selective inhibition of phosphodiesterase III (Brankhorst et al., 1989) and sensitization of cardiac myofilaments to intracellular Ca²⁺ (Fujino et al., 1988; Fraker et al., 1997). The drug has been used for the treatment of patients with heart failure. Pimobendan is well absorbed after oral administration and is metabolized mainly to its O-demethylated form (Fig. 1) in the liver (Fitton and Brogden, 1994). The drug and its metabolite are both predominantly excreted in the feces as glucuronide forms (Hagemeijer, 1993). It has also been demonstrated that the O-demethylated metabolite is a more potent phosphodiesterase III inhibitor than pimobendan in vitro and augments Ca²⁺ sensitivity via a different Ca²⁺ mechanism from pimobendan (Fraker et al., 1997). These results suggest that changes in the capacity to metabolize pimobendan may influence the pharmacological effects of the drug. Our preliminary experiments indicated that cytochrome P-450 (CYP)² is responsible for O-demethylation of pimobendan in rat liver microsomes (S.K., S.O., M.H., C.S., K. Sakai, T.I., and M.K., submitted). The purpose of this study is to identify the CYP isoform(s) that catalyzes the O-demethylation of pimobendan in human liver microsomes.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Oxidative biotransformation of pimobendan in human liver.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Chemicals. Pimobendan and its metabolite (UD-CG212) were provided by Boehringer Ingelheim Pharma KG (Ingelheim, Germany). Cyclosporin A and 2-hydroxydesipramine were kindly donated by Novartis Co. (Basel, Switzerland). Furafylline and 4-hydroxytolbutamide were obtained from Daiichi Pure Chemicals Co. (Tokyo, Japan). Phenacetin, 7-hydroxycoumarin, tolbutamide, 4-nitrophenol, dimethylhydantoinmonomethylol and HPLC grade solvents were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan), Sulfaphenazole, desipramine hydrochloride, 4-nitrocatechol, and -naphthoflavone were obtained from Sigma Chemical Co. (St. Louis, MO). S-Mephenytoin and 4'-hydroxymephenytoin were from Sumika Chemical Analysis Service (Tokyo, Japan). Coumarin and erythromycin were purchased from Kanto Chemical Co. (Tokyo, Japan) and Merck (Darmstadt, Germany), respectively. NADP⁺, glucose 6-phosphate, and glucose 6-phosphate dehydrogenase were purchased from Oriental Yeast Co., Ltd. (Tokyo, Japan). All other chemicals and solvents used were of reagent or analytical grade.

Materials. Human livers were obtained from 16 Japanese autopsy samples (14 male and 2 female, ages 15-68) from the Department of Legal Medicine, School of Medicine, Chiba University, with the approval of the ethics committee of the School of Medicine, Chiba University. We do not have any information on medication history of these autopsy samples. Liver specimens were stored at 80°C until preparation of microsomes. Liver microsomes were prepared by differential centrifugation as described previously (Ohmori et al., 1993). Total CYP content was measured by the method of Omura and Sato (1964) in the presence of 20% glycerol and 0.2% Emulgen 911. Protein was determined as described by Lowry et al. (1951) using BSA as the standard.

Enzyme Assays.

Reaction mixture A typical reaction mixture for the measurement of phenacetin O-deethylase, coumarin 7-hydroxylase, or tolbutamide hydroxylase activity was as follows. The reaction mixture contained 100 mM potassium phosphate (pH 7.4), 0.1 mM EDTA (pH 7.4), 0.5 to 1 mg of microsomal protein, an NADPH-generating system (0.33 mM NADP⁺, 0.1 U of glucose 6-phosphate dehydrogenase, 8 mM glucose 6-phosphate, and 6 mM MgCl₂), and substrate in a final volume of 1 ml. After 10 µl of pimobendan solution in methanol (0.1 or 10 mM) was transferred to reaction tubes, methanol was evaporated under reduced pressure. Then the reaction mixture components, except for microsomes and an NADPH-generating system, were added to the reaction tubes, and the mixture was sonicated to dissolve pimobendan. The low and high concentrations of pimobendan and that of phenacetin were 1, 100, and 10 µM, respectively. Coumarin and tolbutamide were used at a final concentration of 20 µM and 1 mM, respectively. The reaction was initiated by the addition of the NADPH-generating system and carried out for an appropriate time (pimobendan, 20 min; phenacetin, 20 min; coumarin, 15 min; tolbutamide, 90 min) at 37°C with shaking under aerobic conditions. When recombinant CYPs were used as an enzyme source, reaction mixture and experimental conditions were essentially the same as in the case of microsomes except that the final volume of reaction mixture was 0.5 ml.

Pimobendan O-demethylation. We measured the O-demethylated metabolite of pimobendan by the external standard method as reported elsewhere (S.K., S.O., M.H., C.S., K. Sakai, T.I., and M.K., submitted). Briefly, The reaction was terminated by the addition of 1 ml of stop solution (0.2 M potassium phosphate, pH 2.0/2 M HCl, 1:1), and the mixture was centrifuged. The supernatant was filtered through a Chromatodisk (0.45 µm; GL Science, Tokyo, Japan) and 50 µl of the filtrate was injected into the HPLC system. This consisted of two pumps (Hitachi 655 and 655A-11; Tokyo, Japan), a Hitachi D-2000 chromatointegrator, a Shimadzu fluorescence detector RF530 (excitation 338 nm, emission 405 nm), a system controller, and an Inertsil ODS-2 column (4.6 × 150 mm, GL Science, Tokyo, Japan). The mobile phase, which consisted of methanol/acetonitrile/6% ammonium acetate, 3:1:4 (v/v/v), was delivered at a flow rate of 1.3 ml/min at 40°C. To increase the fluorescence intensity of the metabolite, sensitization solution (methanol/H₂O/phosphoric acid, 3:1:1) was added at a flow rate of 0.3 ml/min to the column effluent through a T-connector.

Phenacetin O-deethylation. The phenacetin O-deethylase activity was measured by the method of Tassaneeyakul et al. (1993a) with some modifications. The reaction was terminated by the addition of 0.12 ml of 1 N NaOH. The internal standard, 2-acetoamidophenol (0.2 ml, 5 mg/ml), and 5 ml of chloroform were added to the mixture. After centrifugation, 0.6 ml of aqueous layer was transferred to another tube and acidified with 1 M HCl. The metabolite and internal standard were extracted with 10 ml of diethyl ether. The ether extract was evaporated to dryness under reduced pressure, the residue was taken up in 0.1 ml of the mobile phase, and 50 µl of this solution was injected into the HPLC system. This consisted of a Hitachi L-6000 HPLC pump, Hitachi AS-2000 autosampler, Hitachi L4200 UV/VIS detector, Hitachi D-2500 chromatointegrator, and Inertsil ODS-3 column (4.6 × 150 mm; GL Science, Tokyo, Japan). The mobile phase consisted of methanol/H₂O, 18:82 (pH 3.5) pumped at 1 ml/min, and the detector was set at 254 nm.

Coumarin 7-hydroxylation. Coumarin 7-hydroxylase activity was measured by the method of Greenlee and Poland (1978) with modifications. Briefly, the reaction was terminated by the addition of 1 ml of ice-cold 5% trichloroacetic acid, and precipitated protein was removed by centrifugation (2500 rpm, 10 min). The supernatant (0.5 ml) was then mixed with 2.5 ml of 0.5 M sodium-potassium phosphate (pH 7.5). The 7-hydroxycoumarin formed was measured fluorometrically (excitation, 380 nm; emission, 460 nm) using a Hitachi F-2000 fluorescence spectrophotometer.

Tolbutamide hydroxylation. Because Miners et al. (1988) reported that the reaction rate was linear with time up to 180 min, incubation was performed for 90 min. The reaction was terminated by the addition of 50 µl of 1 M HCl. After the addition of 20 µl of 0.1 mM carbamazepine as an internal standard, the metabolite and internal standard were extracted with 7 ml of ethyl acetate. The organic phase was transferred to another tube and evaporated to dryness. The residue was dissolved in 0.1 ml of acetonitrile, and 10 µl of the solution was injected into the HPLC. The system used was the same as that for the measurement of phenacetin O-deethylase activity except for the use of a TSK-gel ODS-80TM column (4.6 × 150 mm; Tosoh Co. Ltd., Tokyo, Japan).

Other assays. Activity of erythromycin N-demethylase was assayed in terms of the amount of formaldehyde liberated, which was measured spectrophotometrically with Nash reagent (Nash, 1953) as described previously (Ohmori et al., 1993). S-Mephenytoin 4'-hydroxylase, desipramine 2-hydroxylase, testosterone 6-hydroxylase, and 4-nitrophenol hydroxylase activities were measured according to the procedures described by Chiba et al. (1993), Sakamoto et al. (1995), Hayashi et al. (1986), and Tassaneeyakul et al. (1993c), respectively.

Correlation Analysis. The pimobendan O-demethylase activities were compared with marker enzyme activities in human liver microsomes. The marker enzyme reactions for CYP1A2, CYP2A6, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4 used in this study were phenacetin O-deethylation (Tassaneeyakul et al., 1993a), coumarin 7-hydroxylation (Yun et al., 1991), tolbutamide hydroxylation (Relling et al., 1990), S-mephenytoin 4'-hydroxylation (Goldstein et al., 1994), desipramine 2-hydroxylation (Von Moltke et al., 1994), 4-nitrophenol hydroxylation (Tassaneeyakul et al., 1993b), and erythromycin N-demethylation (Watkins et al., 1985), respectively. Pimobendan was used at a final concentration of 100 µM, and other substrate concentrations used for correlation analysis were described in Reaction mixture. The ranges of the marker enzyme activities (picomoles of product formed per milligram per minute) in the microsomal preparations used (n = 16, CYP contents, 0.10-0.38 nmol CYP/mg) were as follows: phenacetin O-deethylation, 3.83 to 205; coumarin 7-hydroxylation, 1.48 to 59.9; tolbutamide hydroxylation, 7.70 to 137; S-mephenytoin 4'-hydroxylation, 1.49 to 141; desipramine 2-hydroxylation, 0.22 to 25.7; 4-nitrophenol hydroxylation, 20.1 to 201.4; and erythromycin N-demethylation, 25.4 to 463.

Inhibition Study. In chemical inhibition experiments, furafylline, sulfaphenazole, and cyclosporin A were used for CYP1A2 (Sesardic et al., 1990), CYP2C9 (Baldwin et al., 1995), and CYP3A4 (Nakasa et al., 1998) inhibitor, respectively. -Naphthoflavone was used for CYP1A inhibitor (Chang et al., 1994). The inhibitors except furafylline were added into the incubation mixture before the reaction was initiated. On the other hand, furafylline was preincubated with microsomes and an NADPH-generating system for 15 min at 37°C before adding the pimobendan to initiate the reaction. Concentrations of the inhibitors used in the experiments were as specified in Table 2.

Contribution of CYP1A2 and CYP3A4 to Pimobendan O-Demethylation in Human Liver Microsomes. Percent contributions of CYP1A2 and CYP3A4 to pimobendan O-demethylation in human liver microsomes were estimated by application of relative activity factor (RAF; Crespi, 1995). The following determination procedure of the contribution of specific CYP isoform to some reaction was described in detail by Kobayashi et al. (1997).

recCYP1A2

recCYP3A4

recCYP1A2

recCYP3A4

Data Analysis. The Michaelis-Menten kinetic parameters for pimobendan O-demethylation were estimated by fitting the following equation:
where V = velocity of pimobendan O-demethylation, S = substrate concentration in the incubation mixture, K_m1 and V_max1 = affinity constant and maximum catalytic activity in high-affinity components, and K_m2 and V_max2 = affinity constant and maximum catalytic activity in low-affinity components.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Kinetic Parameters for Pimobendan O-Demethylation in Human Liver Microsomes. Eadie-Hofstee plots for the O-demethylation of pimobendan in microsomes from three human liver samples are shown in Fig. 2. The plots indicate that at least two CYP isoforms are involved in the reaction. The enzyme kinetic parameters were calculated from these data. K_m and V_max values of pimobendan O-demethylation for the high-affinity component were 1.96 to 5.31 µM and 30.2 to 68.9 pmol/mg/min, respectively, whereas those for the low-affinity component were 20.0 to 51.3 µM and 70.7 to 343.4 pmol/mg/min.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Eadie-Hofstee plots for pimobendan O-demethylation in human liver microsomes. The CYP contents of HL114 (), HL214 (), and HL215 () used in this study were 0.38, 0.35, and 0.10 nmol/mg, respectively.

Correlation Analysis between Pimobendan O-Demethylase Activity and Activities for Marker Reactions. To characterize the CYP isoform(s) that catalyzes the O-demethylation of pimobendan in human liver microsomes, we studied the correlation between microsomal pimobendan O-demethylase activity and marker enzyme activities (Table 1). Pimobendan O-demethylase activity could be detected in all samples used in this study at the substrate concentration of 100 µM. The correlation study indicated that the activity of pimobendan O-demethylase was significantly correlated with the activities of phenacetin O-deethylase (probe for CYP1A2, P < .001), tolbutamide 4-hydroxylase (probe for CYP2C9, P < .001), erythromycin N-demethylase (probe for CYP3A4, P < .001), and desipramine 2-hydroxylase (probe for CYP2D6, P < .05). In contrast, S-mephenytoin, coumarin, and 4-nitrophenol hydroxylations, which are catalyzed by CYP2C19 (Goldstein et al., 1994), CYP2A6 (Yun et al., 1991), and CYP2E1 (Tassaneeyakul et al., 1993b), respectively, showed no correlation with pimobendan O-demethylase activity.

Catalytic Properties of Recombinant Human CYPs for Pimobendan O-Demethylation. To identify more conclusively the CYP isoforms involved in O-demethylation of pimobendan, we then undertook studies using recombinant CYPs. Pimobendan (1 or 100 µM) was incubated with microsomes expressing human CYP (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9/Arg144, CYP2C9/Cys144, CYP2C19, CYP2D6, CYP2E1, or CYP3A4). Among these 10 different recombinant human CYPs, only CYP1A2 showed detectable catalytic activity (22.2 pmol/nmol of CYP/min) for the O-demethylation of pimobendan when the substrate concentration was 1 µM. At a high concentration (100 µM) of pimobendan, CYP3A4 also showed the activity (206.9 pmol/nmol of CYP/min), in addition to CYP1A2 (34.6 pmol/nmol of CYP/min). The other CYPs failed to demethylate pimobendan under these incubation conditions.

Inhibition Studies. The effects of various inhibitors were shown in Table 2. As expected, the inhibitory effects of CYP1A2 inhibitors, furafylline (Sesardic et al., 1990) and -naphthoflavone (Chang et al., 1994) on pimobendan O-demethylation at a low substrate concentration were greater than those at a high substrate concentration. On the other hand, cyclosporin A showed an inhibitory effect, whereas -naphthoflavone acted as an activator of demethylation when 100 µM pimobendan was used; these effects are characteristic of CYP3A-catalyzed reactions. The CYP2C9 inhibitor, sulfaphenazole (Baldwin et al., 1995), had no effect on the activity of pimobendan O-demethylase in human liver microsomes.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Immunoinhibition of pimobendan O-demethylase in human liver microsomes by anti-CYP1A2, anti-CYP2C9, anti-CYP2D6, or anti-CYP3A4 antibody. In the immunoinhibition studies with anti-CYP1A2 and CYP3A4 antibodies, the control activities of human liver microsomes (HL114) in the presence of 1 and 100 µM pimobendan were 5.4 and 47.3 pmol/mg/min, respectively. When we examined the effects of anti-CYP2C9 and anti-CYP2D6 antibodies, the control activities of HL114 in the presence of 1 and 100 µM pimobendan were 5.2 and 45.8 pmol/mg/min, respectively. The reaction was performed in a final volume of 0.5 ml. Anti-CYP1A2 IgG (), anti-CYP2C9 IgG (), anti-CYP2D6 IgG(). or anti-CYP3A4 IgG () and preimmune IgG were added in various ratios to maintain a 10-mg total IgG protein/nmol of CYP.

Contributions of CYP1A2 and CYP3A4 to Pimobendan O-Demethylation in Human Liver Microsomes. The contributions of CYP1A2 and CYP3A4 to pimobendan O-demethylation in human liver microsomes were estimated by application of the RAF method as described in Experimental Procedures. Phenacetin O-deethylase (substrate concentration 10 µM) and testosterone 6-hydroxylase activities were selected as specific catalytic activities for CYP1A2 (Tassaneeyakul et al., 1993a) and CYP3A4 (Waxman et al., 1991), respectively. The activities of phenacetin O-deethylation by CYP1A2 and of testosterone 6-hydroxylation by CYP3A4 were 0.78 and 6.85 nmol/nmol of CYP/min, respectively. The activities of phenacetin O-deethylation and testosterone 6-hydroxylation in microsomes from four human liver samples were ranged from 71.3 to 834 pmol/mg/min and 0.81 to 1.86 nmol/mg/min, respectively. Calculated percent contributions of CYP1A2 and CYP3A4 to pimobendan O-demethylation in these liver samples are shown in Table 3. The contribution of CYP1A2 ranged from 18.3 to 76.3% and that of CYP3A4 ranged from 3.5 to 6.5%. The values of percent contribution of CYP1A2 in these four samples were well correlated with those of percent inhibition of pimobendan O-demethylase activity by CYP1A2 antibody.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In this study, we have identified the CYP isoforms responsible for pimobendan O-demethylation in human liver microsomes. The following results suggest that CYP1A2 is primarily responsible for pimobendan O-demethylation under clinically relevant conditions: 1) pimobendan O-demethylase activity in human liver microsomes was significantly correlated with phenacetin O-deethylase activity, which is specific to CYP1A2 (Tassaneeyakul et al., 1993a); 2) CYP1A2 antibody and specific inhibitors for CYP1A2, furafylline, and -naphthoflavone, strongly inhibited the metabolism of pimobendan; and 3) recombinant human CYP1A2 was the only catalyst of pimobendan O-demethylation, at a low drug concentration, among 10 species of CYP examined.

Because a kinetic study suggested that plural CYP isoforms contribute to pimobendan O-demethylation in human liver microsomes (Fig. 2), pimobendan O-demethylase activity was measured at both 1 and 100 µM pimobendan. Although the activities toward specific substrates of CYP2C9 and CYP2D6 were positively correlated with pimobendan O-demethylase activity in human liver microsomes (Table 1), inhibitors of CYP2C9 and CYP2D6 and antibodies against these CYPs did not inhibit the demethylase activity. Furthermore, recombinant CYP2C9 and CYP2D6 did not catalyze the reaction. From these results, we conclude that CYP2C9 and CYP2D6 are not responsible for pimobendan O-demethylation in human liver. However the demethylase activity was inhibited by furafylline and -naphthoflavone at a substrate concentration of 1 µM. Although -naphthoflavone inhibits CYP2C8 and CYP2C9 (Chang et al., 1994), we thought the inhibitory effect of -naphthoflavone was caused by the inhibition of CYP1A2, because recombinant CYP2C8 and CYP2C9 failed to catalyze the reaction and antibody to CYP2C9 did not inhibit the activity in human liver microsomes (Fig. 3). Thus we consider that the pimobendan O-demethylation reaction is catalyzed by CYP1A2 in human liver microsomes at a low substrate concentration.

On the other hand, cyclosporin A was an effective inhibitor and -naphthoflavone activated the activity at a high substrate concentration (Table 2), suggesting that CYP3A contributes to the reaction because it is well known that cyclosporin A is a CYP3A-specific inhibitor (Relling et al., 1994) and several CYP3A-mediated reactions are activated by -naphthoflavone (Shou et al., 1994). These results were consistent with the results of the recombinant isoform study and immunoinhibition study (Fig. 3). At the high substrate concentration, recombinant CYP3A4 catalyzed the reaction and antibody to CYP3A4 inhibited it.

The contributions of the two isoforms to the pimobendan O-demethylation reaction in human liver microsomes were studied. Crespi (1995) has proposed the use of RAF to extrapolate data obtained with recombinant CYPs to human liver microsomes. The contribution percentages of CYP1A2 and CYP3A4 were calculated according to the method of Kobayashi et al. (1997) using the RAFs (Crespi, 1995). We found that the contribution of CYP1A2 was varied from 18 to 67%. It has been demonstrated that the CYP1A2 level is influenced by environmental factors, and there are large interindividual differences in the CYP1A2 content (Clarke, 1998). In fact, an approximate 10-fold difference in the phenacetin O-deethylase activity was observed in our samples. Nevertheless, the percent inhibition by CYP1A2 antibody was well correlated with the percent contribution of the isoform (Table 3). On the other hand, less than 10% of the pimobendan O-demethylase activity was estimated to be catalyzed by CYP3A4. Inhibitions of pimobendan O-demethylation by CYP1A2 antibody, and inhibitors were more potent than those by CYP3A4 antibody and inhibitors (Fig. 3, Table 2). However, the contributions of CYP1A2 and CYP3A do not fully account for the extent of pimobendan O-demethylation in human liver microsomes. To date, more than 30 species of human CYP have been reported (Nelson et al., 1996), and the contents of only some of these in human liver have been quantified (Shimada et al., 1994); about 25% of total CYP in human liver microsomes remains unidentified. Therefore, the possibility that CYP isoforms other than CYP1A2 and CYP3A4 might be responsible in part for the pimobendan O-demethylation reaction in human liver microsomes cannot be excluded.

From our findings, we can conclude that CYP1A2 is one of the major isoforms responsible for the O-demethylation of pimobendan, and CYP3A is a minor contributor at clinically relevant concentrations of pimobendan.