Introduction

Top Abstract Introduction Results Discussion References

Cytochrome P450 (CYP)¹ consists of a superfamily of heme-containing monooxygenases that are associated with the metabolism of substrates, such as drugs, environmental pollutants, and endogenous substrates, and have broad overlapping specificities (Guengerich, 1991). It is not unusual for a compound to be metabolized by several CYP isoforms. Identification of isoform(s) involved in the biotransformation of a compound would enable clinicians to predict and/or avoid drug interactions that might compromise therapeutic efficacy.

Amiodarone is an iodinated benzofuran derivative, which is extremely effective in the treatment of life-threatening supraventricular and ventricular arrhythmias such as ventricular fibrillation or hemodynamically unstable ventricular tachycardia (Gill et al., 1992), with predominantly Class III (Vaughan Williams' classification) antiarrhythmic effects. Amiodarone has been shown to have a long serum elimination half-life of 40 to 50 days (Kannan et al., 1982; Holt et al., 1983; Stäubli et al., 1983), which was attributed to its huge distribution. It has been reported that its primary metabolite, desethylamiodarone, is pharmacologically active and shows comparable antiarrhythmic effects to its parent compound (Pallandi and Campbell, 1987; Kato et al., 1988).

It has been suggested that a number of drugs administered concomitantly with amiodarone, such as theophylline (Hirsch et al., 1993), phenytoin (Nolan et al., 1989), warfarin (O'Reilly et al., 1987; Heimark et al., 1992), and flecainide (Funch-Brentano et al., 1994), cause drug interactions in the clinical situation. In our previous study, it was clarified that desethylamiodarone has more potent inhibitory effects than amiodarone on CYP activities (Ohyama et al., 2000). Therefore, drug interactions caused by coadministration of amiodarone have been suggested to be caused by the inhibition of CYP activities by desethylamiodarone (Ohyama et al., 2000).

It has been reported that amiodarone N-deethylation is catalyzed by CYP3A (Fabre et al., 1993; Trivier et al., 1993). However, in these in vitro assay systems, the low solubility of amiodarone was not taken into consideration. Therefore, in the present study, we improved the assay procedure to clarify the CYP isoform involved in the transformation of amiodarone to desethylamiodarone in human liver microsomes. In addition, the contributions of each CYP isoform to amiodarone N-deethylation in human liver microsomes were estimated by the relative activity factor (RAF) using microsomes from baculovirus-infected insect cells expressing human CYP. The present study is the first to clarify the significant role of human CYP2C8 in amiodarone N-deethylation.

Materials and Methods

Chemicals. Amiodarone hydrochloride (2-butyl-3-benzofuranyl 4-[2-(diethylamino)ethoxy]-3,5-diiodophenyl ketone hydrochloride) and desethylamiodarone (2-butyl-3-benzofuranyl 4-[2-(monoethylamino)ethoxy]-3,5-diiodophenyl ketone) were kindly provided by Taisho Pharmaceutical (Tokyo, Japan). 7-Ethoxyresorufin and resorufin were purchased from Sigma Chemical Co. (St. Louis, MO). Paclitaxel was kindly provided by Bristol-Myers Squibb Pharmaceutical (Tokyo, Japan). 6-Hydroxypaclitaxel, S-(+)-mephenytoin, (±)-4'-hydroxymephenytoin, (±)-bufuralol hydrochloride, and 1'-hydroxybufuralol maleate were from Ultrafine Chemicals (Manchester, UK). Testosterone, 6-hydroxytestosterone, and 11-hydroxytestosterone were from Steraloids (Wilton, NH). NADP⁺, glucose 6-phosphate, and glucose-6-phosphate dehydrogenase were purchased from Oriental Yeast (Tokyo, Japan). Anti-rat CYP1A2 rabbit serum, anti-rat CYP2C13 goat serum, anti-human CYP2D6 rabbit serum, and anti-rat CYP3A2 rabbit serum were from Daiichi Pure Chemicals (Tokyo, Japan). Other chemicals were of the highest grade commercially available.

Enzyme Preparations. Microsomes from baculovirus-infected insect cells expressing CYP1A1, CYP1A2, CYP1B1, CYP2A6+b₅, CYP2B6+b₅, CYP2C8, CYP2C8+b₅, CYP2C9, CYP2C9+b₅, CYP2C18, CYP2C19, CYP2C19+b₅, CYP2D6, CYP2E1+b₅, CYP3A4, CYP3A4+b₅, CYP3A5, CYP3A7+b₅, and CYP4A11 were obtained from Gentest (Woburn, MA). These were all coexpressed with NADPH-cytochrome P450 oxidoreductase (OR).

Amiodarone N-Deethylase Activity. The determination of amiodarone N-deethylase activity in human liver microsomes or microsomes from baculovirus-infected insect cells was performed by HPLC. A typical incubation mixture (0.2 ml of total volume) contained 50 mM potassium phosphate buffer (pH 7.4), 4% BSA, an NADPH-generating system (0.5 mM NADP⁺, 5 mM glucose 6-phosphate, 5 mM MgCl₂, 1 U/ml glucose-6-phosphate dehydrogenase), 50 µM amiodarone, and 0.25 mg/ml microsomal protein of human livers or 12.5 or 25 pmol/ml CYP content of microsomes from baculovirus-infected insect cells. The reaction was initiated by the addition of the NADPH-generating system and incubated at 37°C for 30 min after a 1-min preincubation. The reaction was terminated by the addition of 100 µl of ice-cold methanol, and -naphthoflavone (200 pmol) was added as an internal standard. The reaction mixture was extracted twice with 2 ml of dichloromethane for 1 min. The organic layer was transferred to a clean test tube and was evaporated under a gentle stream of nitrogen at 40°C. The residue was redissolved in 200 µl of mobile phase, and a 100-µl portion was subjected to HPLC.

Inhibition Studies. CYP-specific inhibitors were screened for their effects on amiodarone N-deethylation in pooled human liver microsomes at an amiodarone concentration of 50 µM. The inhibitors studied were fluvoxamine (Pastrakuljic et al., 1997), furafylline (Tassaneeyakul et al., 1993), coumarin (Yun et al., 1991), sulfaphenazole (Baldwin et al., 1995), paclitaxel (Rahman et al., 1994), quercetin (Rahman et al., 1994), S-mephenytoin (Chiba et al., 1993), quinidine (Guengerich et al., 1986; Broly et al., 1989), chlorzoxazone (Peter et al., 1990), erythromycin (Watkins et al., 1985), and ketoconazole (Watkins et al., 1985). With the exception of paclitaxel, which was dissolved in dimethyl sulfoxide with a final solvent concentration in the incubation mixture of 1%, the inhibitors were dissolved in methanol or water, and the final concentration of solvent was <0.1%. Amiodarone N-deethylation was determined as described above.

Other Assays. Ethoxyresorufin O-deethylase activity, bufuralol 1'-hydroxylase activity, and testosterone 6-hydroxylase activity were determined as described previously (Nakajima et al., 1999a). These substrate concentrations ranged from 0.1 to 2, 1 to 10, and 5 to 200 µM, respectively. Paclitaxel 6-hydroxylase activity (Yamazaki et al., 1999) and S-mephenytoin 4'-hydroxylase activity (Nakajima et al., 1999b) were determined as described previously with substrate concentrations of 0.5 to 20 and 10 to 200 µM, respectively.

Contributions of CYP1A2, CYP2C8, CYP2C19, CYP2D6, and CYP3A4 to Amiodarone N-Deethylase Activity in Human Liver Microsomes. The contributions of each CYP isoform to the amiodarone N-deethylase activity in human liver microsomes were estimated. The percentages of the contributions were estimated by the application of the RAF (Crespi, 1995) using clearance (CL) values (RAF_CL) (Nakajima et al., 1999a). The CL values were determined as the ratios of the V_max/K_M. The marker activities used in this study were ethoxyresorufin O-deethylase activity for CYP1A2, paclitaxel 6-hydroxylase activity for CYP2C8, S-mephenytoin 4'-hydroxylase activity for CYP2C19, bufuralol 1'-hydroxylase activity for CYP2D6, and testosterone 6-hydroxylase activity for CYP3A4. The RAF_CL values for CYP1A2, CYP2C8, CYP2C19, CYP2D6, and CYP3A4 were determined using the equation as follows:

(1)

The predicted activities for each CYP isoform in human liver microsomes were estimated by the equation as follows:

(2)

where V_rec-CYPx values are the amiodarone N-deethylase activity for recombinant CYP1A2, CYP2C8, CYP2C19, CYP2D6, or CYP3A4.

(3)

Data Analysis. Correlations between the amiodarone N-deethylase activity and ethoxyresorufin O-deethylase activity, paclitaxel 6-hydroxylase activity, S-mephenytoin 4'-hydroxylase activity, bufuralol 1'-hydroxylase activity, or testosterone 6-hydroxylase activity in microsomes from six human livers were determined using linear regression analysis. Kinetic parameters were estimated from the fitted curves using a computer program (KaleidaGraph, Synergy Software, Reading, PA) designed for nonlinear regression analysis.

	Results

Top Abstract Introduction Results Discussion References

Amiodarone N-Deethylase Activity in Human Liver Microsomes. In our preliminary study, amiodarone hydrochloride was insoluble in potassium phosphate buffer (pH 7.4) of more than 50 µM, although it is described in the Merck Index (1996) as being soluble in water at least up to 1 mM as a hydrochloride salt. Therefore, we improved the assay system to increase the solubility of amiodarone in potassium phosphate buffer by the addition of BSA. We confirm that amiodarone was soluble at least up to 200 µM under this condition (data not shown). The amiodarone N-deethylase activity in human liver microsomes was linear with an incubation time up to 45 min and a protein concentration up to 0.5 mg/ml. Unless specified, an incubation time of 30 min and protein concentration of 0.25 mg/ml were employed to ensure the initial rate condition for amiodarone N-deethylation. Kinetic parameters for the amiodarone N-deethylase activity were determined using microsomes from six human livers (Table 1). Apparent monophasic parameters were obtained in all samples: the mean K_M and V_max values were 31.6 ± 7.2 µM and 1.2 ± 0.7 pmol/min/pmol of CYP, respectively.

Inhibition Studies of Amiodarone N-Deethylation in Pooled Human Liver Microsomes. CYP-specific inhibitors were screened for their effects on the amiodarone N-deethylase activity in pooled human liver microsomes (Fig. 1). Fluvoxamine (an inhibitor of CYP1A2 and CYP2C19) and furafylline (an inhibitor of CYP1A2) inhibited the amiodarone N-deethylase activity by ~30% of the control at 100 µM. Paclitaxel was dissolved in dimethyl sulfoxide with a final solvent concentration in the incubation mixture of 1%. Dimethyl sulfoxide of 1% inhibited the amiodarone N-deethylase activity in pooled human liver microsomes by ~30%. Therefore, inhibitory effects of paclitaxel were determined in the presence of 1% dimethyl sulfoxide. The amiodarone N-deethylase activity was inhibited by 100 µM paclitaxel or quercetin (inhibitors of CYP2C8) by approximately 50% and 20%, respectively. S-Mephenytoin (inhibitor of CYP2C19) inhibited the activity by ~35% of the control at 100 µM. Quinidine inhibited amiodarone N-deethylation by ~40% of the control at a concentration of 100 µM a condition to inhibit CYP3A4, but had little effect at 10 µM a condition to inhibit CYP2D6. Ketoconazole (an inhibitor of CYP3A4) significantly inhibited amiodarone N-deethylation to <10% of the control at 10 µM. Erythromycin (an inhibitor of CYP3A4) also inhibited it by ~40% of the control at 100 µM. Coumarin (an inhibitor of CYP2A6), sulfaphenazole (an inhibitor of CYP2C9), and chlorzoxazone (an inhibitor of CYP2E1) had no effect on the amiodarone N-deethylase activity at 100 µM.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Inhibitory effects of CYP-specific inhibitors on amiodarone N-deethylase activity in pooled human liver microsomes. The amiodarone concentration was 50 µM. The control activity was 0.45 nmol/min/mg of protein. Data are mean of duplicate determinations.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Immunoinhibition of amiodarone N-deethylase activity in pooled human liver microsomes. Anti-rat CYP1A2 (), anti-rat CYP2C13 (), anti-human CYP2D6 (), and anti-rat CYP3A2 () sera were preincubated with human liver microsomes on ice for 1 h. Preimmune rabbit or goat sera were included in the assay condition to keep the serum concentration being 200 µl/mg of protein. Data are mean of duplicate determinations.

Correlation Study. Correlations between the amiodarone N-deethylase activity (at substrate concentrations of 5 and 100 µM) and ethoxyresorufin O-deethylase activity, paclitaxel 6-hydroxylase activity, bufuralol 1'-hydroxylase activity, S-mephenytoin 4'-hydroxylase activity, and testosterone 6-hydroxylase activity in microsomes from six human livers were examined. As shown in Fig. 3A, amiodarone N-deethylase activity at a substrate concentration of 5 µM showed a significant correlation with the paclitaxel 6-hydroxylase activity (r = 0.842, P < .05) but not with the testosterone 6-hydroxylase activity. On the other hand, amiodarone N-deethylase activity at 100 µM was significantly correlated with the testosterone 6-hydroxylase activity (r = 0.944, P < .005) but not with the paclitaxel 6-hydroxylase activity (Fig. 3, C and D). Amiodarone N-deethylase activities at substrate concentrations of 5 and 100 µM did not correlate with the ethoxyresorufin O-deethylase activity, S-mephenytoin 4'-hydroxylase activity, and bufuralol 1'-hydroxylase activity (data not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Correlations between amiodarone N-deethylase activity and paclitaxel 6-hydroxylase activity or testosterone 6-hydroxylase activity in microsomes from six human livers. Amiodarone N-deethylase activity (5 µM) and paclitaxel 6-hydroxylase activity (A) or testosterone 6-hydroxylase activity (B); amiodarone N-deethylase activity (100 µM) and paclitaxel 6-hydroxylase activity (C) or testosterone 6-hydroxylase activity (D). Data are mean of duplicate determinations.

Amiodarone N-Deethylase Activity in Microsomes from Baculovirus-infected Insect Cells Expressing Human CYP. The abilities of cDNA-expressed human CYPs for amiodarone N-deethylation were examined. As shown in Fig. 4, CYP1A1 showed the highest catalytic activity (13.0 pmol/min/pmol of CYP), followed by CYP3A4+b₅ (8.4 pmol/min/pmol of CYP). CYP1A2, CYP3A4, CYP2D6, CYP2C8+b₅, CYP2C19+b₅, CYP2C19, and CYP2C8 showed moderate amiodarone N-deethylase activity (2.7, 2.7, 2.4, 1.9, 1.3, 1.2, and 1.0 pmol/min/pmol of CYP, respectively). CYP3A5, CYP4A11, CYP3A7+b₅, CYP2E1+b₅, CYP2C9+b₅, and CYP2C9 exhibited low activity (<1.0 pmol/min/pmol of CYP). On the other hand, no activity was observed with microsomes expressing CYP1B1, CYP2A6+b₅, CYP2B6 +b₅, and CYP2C18.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Amiodarone N-deethylase activity in microsomes from baculovirus-infected cells expressing human CYP isoform. All CYP isoforms are coexpressed with OR. The concentration of amiodarone was 50 µM. Each column represents the mean of duplicate determinations. ND, not detected.

Contributions of CYP1A2, CYP2C8, CYP2C19, CYP2D6, and CYP3A4 to Amiodarone N-Deethylase Activity in Individual Human Liver Microsomes. The K_M values of ethoxyresorufin O-deethylase activity in microsomes from six human livers and recombinant CYP1A2 were 0.09 to 0.54 µM and 0.05 µM, respectively. The V_max values were 9.3 to 261.3 pmol/min/pmol of CYP and 1976.1 pmol/min/pmol of CYP, respectively. The CL values of ethoxyresorufin O-deethylase activity in the human liver microsomes and recombinant CYP1A2 were 45.4 to 919.6 µl/min/pmol of CYP and 38001.9 µl/min/pmol of CYP, respectively. Thus, RAF_CL-CYP1A2 was calculated to range from 0.001 to 0.024 (Table 3). The K_M values of paclitaxel 6-hydroxylase activity in the human liver microsomes and recombinant CYP2C8 were 2.58 to 4.55 µM and 2.85 µM, respectively. The V_max values were 0.224 to 0.583 pmol/min/pmol of CYP and 5.667 pmol/min/pmol of CYP, respectively. The CL values of paclitaxel 6-hydroxylase activity in the human liver and recombinant CYP2C8 were 0.070 to 0.191 µl/min/pmol of CYP and 1.988 µl/min/pmol of CYP, respectively. Thus, RAF_CL-CYP2C8 was estimated to range from 0.035 to 0.096. The K_M values of S-mephenytoin 4'-hydroxylase activity in microsomes from six human livers and recombinant CYP2C19 were 24.4 to 52.3 µM and 24.7 µM, respectively. The V_max values were 0.092 to 1.532 pmol/min/pmol of CYP and 9.375 pmol/min/pmol of CYP, respectively. The CL values of S-mephenytoin 4'-hydroxylase activity in the human liver microsomes and recombinant CYP2C19 were 0.003 to 0.030 µl/min/pmol of CYP and 0.380 µl/min/pmol of CYP, respectively. Thus, RAF_CL-CYP2C19 was calculated to range from 0.007 to 0.080. The K_M values of bufuralol 1'-hydroxylase activity in the human liver microsomes and recombinant CYP2D6 were 2.8 to 30.1 µM and 1.5 µM, respectively. The V_max values were 0.116 to 0.531 pmol/min/pmol of CYP and 6.382 pmol/min/pmol of CYP, respectively. The CL values of bufuralol 1'-hydroxylase activity in the human liver microsomes and recombinant CYP2C19 were 0.005 to 0.152 µl/min/pmol of CYP and 4.255 µl/min/pmol of CYP, respectively. Thus, RAF_CL-CYP2D6 was calculated to range from 0.001 to 0.036. The K_M values of testosterone 6-hydroxylase activity in the human liver microsomes and recombinant CYP3A4 were 32.9 to 46.0 µM and 34.6 µM, respectively. The V_max values were 4.0 to 23.1 pmol/min/pmol of CYP and 165.4 pmol/min/pmol of CYP, respectively. The CL values of testosterone 6-hydroxylase activity in the human liver microsomes and recombinant CYP3A4 were 0.091 to 0.585 µl/min/pmol of CYP and 4.780 µl/min/pmol of CYP, respectively. Thus, RAF_CL-CYP3A4 was calculated to range from 0.019 to 0.122. 

View larger version (37K): [in this window] [in a new window]   Fig. 5.   The percentage contribution of each CYP isoform to the amiodarone N-deethylation in microsomes from six human livers. The contributions of CYP1A2 (dark gray), CYP2C8 (light gray), CYP2C19 (black), CYP2D6 (cross-hatched), and CYP3A4 (white) were estimated at substrate concentrations of 5, 10, 20, 50, and 100 µM, respectively.

	Discussion

Top Abstract Introduction Results Discussion References

Amiodarone, a drug administered for the cure of life-threatening supraventricular and ventricular arrhythmias, has been reported to interact with a number of drugs such as theophylline (Hirsch et al., 1993), phenytoin (Nolan et al., 1989), warfarin (O'Reilly et al., 1987; Heimark et al., 1992), and flecainide (Funch-Brentano et al., 1994). In our previous study (Ohyama et al., 2000), it was revealed that these drug interactions resulted from the inhibition of CYP activities by desethylamiodarone rather than by the parent drug. It has been reported that the formation of desethylamiodarone from amiodarone is primarily catalyzed by CYP3A in humans (Fabre et al., 1993; Trivier et al., 1993). In the present study, we identified the CYP isoforms involved in amiodarone N-deethylation in human liver microsomes and evaluated the percentage contribution of each CYP isoform.

Previously, the kinetic parameters of amiodarone N-deethylase activity in human liver microsomes have been reported to be K_M = 0.33 ± 0.11 mM and V_max = 2.38 ± 0.74 nmol/min/mg of protein (Fabre et al., 1993). On the other hand, the parameters obtained in the present study were K_M = 31.6 ± 7.5 µM and V_max = 1.2 ± 0.7 pmol/min/pmol of CYP (0.52 ± 0.35 nmol/min/mg of protein). It is possible that the improvement in the solubility of amiodarone in the present assay system decreased the K_M value, because Fabre et al. (1993) did not take the insolubility into consideration.

The chemical and immunological inhibition studies using pooled human liver microsomes revealed that CYP3A4 is the primary isoform responsible for amiodarone N-deethylation at a substrate concentration of 50 µM. In addition, weak inhibition by paclitaxel, quercetin, furafylline, and fluvoxamine suggested that CYP2C8 and CYP1A2 play minor roles in this reaction. The concentration of amiodarone in human plasma has been reported to range from 0.65 to 5.7 µM in clinical situations (Stäubli et al., 1983). Because the serum binding ratio of amiodarone has been reported to be 96% (Lalloz et al., 1984), the unbound amiodarone concentrations were calculated to be 0.03 to 0.23 µM. It has been reported that amiodarone is highly accumulated in the liver because of the high lipophilicity (Berger and Harris, 1986). The total concentration of amiodarone in the liver has been reported to range from 7.1 to 1580.6 µM in postmortem and biopsy samples (Latini et al., 1984; Berger and Harris, 1986). Therefore, the correlation study was performed at substrate concentrations of 5 and 100 µM as the low and high concentrations, respectively. The results shown in Fig. 3 indicated that CYP2C8 was involved in the amiodarone N-deethylation in human liver microsomes at a low amiodarone concentration. On the other hand, CYP3A4 participates in amiodarone N-deethylation at a high substrate concentration. The findings supported the results of the predicted contribution of each CYP isoform in human liver microsomes estimated from the RAF (Table 4). The present results showing that the K_M value of recombinant CYP2C8 was the lowest (8.6 µM; Table 2) and the V_max value of CYP3A4 was the highest (18.9 pmol/min/pmol of CYP; Table 2) among the five recombinant CYP isoforms also supports the results of the prediction.

The concept of RAF to extrapolate data from recombinant CYPs to those from human liver microsomes was first proposed by Crespi (1995). In our previous study, we concluded that the RAF_CL (Nakajima et al., 1999a) is more appropriate to estimate the contribution of CYP isoforms than RAF_V (Kobayashi et al., 1997; Nakajima et al., 1998) and RAF_Vmax (Crespi, 1995). Furthermore, concerning the source of recombinant CYPs, it was suggested that microsomes from baculovirus-infected insect cells expressing CYP isoforms could be more useful for an appropriate prediction than those from B-lymphoblastoid cells (Nakajima et al., 1999a; Roy et al., 1999). In addition, it was also suggested that the coexpression of b₅ is preferable. Therefore, we used the RAF_CL to estimate the relative contribution of CYP1A2, CYP2C8, CYP2C19, CYP2D6, and CYP3A4 in human liver using microsomes from baculovirus-infected insect cells expressing each CYP isoform and OR (b₅ was coexpressed for CYP2C8 and CYP3A4).

In our estimation of the relative contributions of multiple CYP isoforms using RAF_CL, it was revealed that CYP3A4, CYP2C8, CYP1A2, CYP2D6, and CYP2C19 were the CYP isoforms catalyzing amiodarone N-deethylation. As reported previously (Fabre et al., 1993; Trivier et al., 1993), CYP3A4 is the major isoform involved in amiodarone metabolism. On the other hand, CYP2C8 was also significantly responsible for amiodarone N-deethylation at a low concentration, as supported by the correlation study (Fig. 3). The number of drugs that are metabolized by CYP2C8, such as paclitaxel (Rahman et al., 1994), rosiglitazone (Baldwin et al., 1999), zopiclone (Becquemont et al., 1999), and troglitazone (Yamazaki et al., 1999), is increasing. Therefore, for the prediction of drug interactions with amiodarone, the drugs metabolized by CYP2C8 as well as those by CYP3A4 should be taken into consideration. Although the contributions of CYP1A2, CYP2C19, or CYP2D6 were minor, these isoforms might alternatively contribute to amiodarone N-deethylation in clinical situations in which CYP3A4 or CYP2C8 is inhibited by coadministered drugs.

In conclusion, the present study suggested that CYP2C8 as well as CYP3A4 was significantly involved in amiodarone N-deethylation in human liver. In addition, CYP1A2, CYP2C19, or CYP2D6 exhibited minor contributions to amiodarone metabolism. The relevance of these CYP isoforms would be dependent on the content of each isoform and on the amiodarone concentration in the liver.