Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Quinine is one of the Cinchona alkaloids used in the treatment of severe forms of malaria (White, 1992). Although the drug has been used for more than 300 years, its metabolism and elimination in humans is not well elucidated. Even though several metabolites have been identified (Bannon et al., 1998; Mirghani et al., 2001), the relative roles of these metabolites in quinine elimination and their antimalarial activity have not been investigated.

The major metabolite of quinine has been reported to be 3-hydroxyquinine just by comparing the peak heights with that of other metabolites in the chromatograms without quantification (Wanwimolruk et al., 1995). Cytochrome P450 3A4 (CYP3A4) is the most abundant isoform in human liver (up to 30%) and involved in the metabolism of more than 50% of clinically used drugs (Kuehl et al., 2001). CYP3A4 is reported to catalyze the formation of 3-hydroxyquinine both in vitro (Zhang et al., 1997) and in vivo (Mirghani et al., 1999). In a previous in vivo study, we found that about 30% of a single oral dose was excreted as quinine and 3-hydroxyquinine in human urine (Mirghani et al., 1999). Since quinine is extensively metabolized in the liver, elimination via other metabolic pathways may partly explain this low recovery. Four of the quinine metabolites, 3-hydroxyquinine, 2'-quininone, (10S)-11-dihydroxydihydroquinine, and (10R)-11-dihydroxydihydroquinine have so far been identified in human plasma and urine (Mirghani et al., 2001).

The aims of this study were to investigate the relative importance of CYP3A4 in the biotransformation of quinine in human liver microsomes and recombinant-expressed CYP3A4 and quantify the relative contribution of the different metabolites in the overall metabolism of quinine.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Human Liver Microsomes. Human liver microsomes were prepared essentially as described (von Bahr et al., 1980) from five different liver tissues belonging to a donor liver bank established at the Division of Clinical Pharmacology (Huddinge University Hospital) with the approval of the hospital ethics committee. The microsomes were stored in potassium phosphate buffer (50 mM, pH 7.4) at 80°C until used. The protein content was estimated according to Lowry et al. (1951).

Quinine Metabolism in Human Liver Microsomes and Human Recombinant CYP3A4. The incubations were performed in duplicates in 50 mM potassium phosphate buffer, pH 7.4, at 37°C in a shaking water bath. The reaction was initiated by the addition of NADPH (0.5 mM) in the final volume of 0.5 ml and terminated after 15 min by the addition of 500 µl of methanol and rapid freezing.

Enzyme Kinetic Studies. The formation kinetics of the four metabolites were studied in microsomes prepared from five different human livers. The incubations were performed with nine different quinine concentrations in the range of 15 to 400 µM, which covers the clinically relevant concentrations (20-60 µM) (Krishna and White, 1996). The quinine concentrations used for incubations with recombinant-expressed CYP3A4 were 10 to 750 µM.

Inhibition Studies. Inhibition experiments with the CYP3A4 inhibitors (Gascon and Dayer, 1991; Chang et al., 1994) ketoconazole (Janssen Biotech NV, Olen, Belgium) at 0.25, 0.5, 1, and 10 µM; troleandomycin (Solvay Duphar BV, Weesp, Netherlands) at 10, 20, 40, and 80 µM; and the CYP1A2 inhibitor (Brosen et al., 1993) fluvoxamine (Sigma-Aldrich, St. Louis, MO) at 10, 25, 50 and 100 µM were performed in microsomes from four different human livers. The inhibitors were dissolved in methanol that was evaporated to dryness before addition of the incubation mixture. For incubations with troleandomycin, the samples were preincubated together with NADPH for 10 min at 37°C in a shaking water bath. The reactions were started by the addition of quinine at a final concentration of 20 and 50 µM, and the remaining part of the procedure was performed as described above.

Correlation Studies. Formation rates of the metabolites were studied in microsomes from 20 livers at 35 and 200 µM of quinine. The measured activities were then correlated with the various probe activities commonly used to characterize phase I metabolism (see Table 2). The different probe drugs and their metabolites were analyzed as described (G. Tybring, manuscript in preparation).

Data Analysis. The enzyme kinetic parameters, the Michaelis-Menten constant (K_m), the maximal formation rate (V_max), and the intrinsic clearance (V_max/K_m) for the formation of the different quinine metabolites were determined using the Michaelis-Menten (one-enzyme) kinetic module in the software GraFit 4.03 (Erithacus Software Ltd., Surrey, UK). The correlation coefficients were determined by linear regression analysis using the software STATISTICA 5.0 (StatSoft Inc., Tulsa, Oklahoma).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Enzyme Kinetics. The mean enzyme kinetic parameters for the formation of the four metabolites are shown in Table 1. Around 2-fold interindividual variations in K_m, V_max, and V_max/K_m were observed for the formation of the different metabolites. The 3-hydroxyquinine metabolite had the highest formation rate (V_max), which was 30-, 8-, and 7-fold higher than that of (10R)-11-dihydroxydihydroquinine, (10S)-11-dihydroxydihydroquinine, and 2'-quininone, respectively. The formation rates of the metabolites fitted the Michaelis-Menten model with one enzyme kinetics. The intrinsic clearance for the formation of 3-hydroxyquinine was 22-, 10-, and 8-fold higher than that for (10R)-11-dihydroxydihydroquinine, (10S)-11-dihydroxydihydroquinine, and 2'-quininone, respectively. The interindividual variability in the intrinsic clearance was about 3-fold for 3-hydroxyquinine and 2-fold for the other metabolites.

Inhibition Studies. The effects of ketoconazole, troleandomycin, and fluvoxamine on the formation of quinine metabolites are shown in Fig. 1. Inhibitors of CYP3A4 inhibited the formation of quinine metabolites in a concentration-dependent manner to different extent. Ketoconazole (1 µM) inhibited the formation of 3-hydroxyquinine, 2'-quininone, (10R)- and (10S)-11-dihydroxydihydroquinine by 93, 92, 11, and 58%, respectively. At 80 µM of troleandomycin, the formation of 3-hydroxyquinine, 2'-quininone, (10R)-11-dihydroxydihydroquinine, and (10S)-11-dihydroxydihydroquinine was inhibited by 98, 73, 15, and 57%, respectively. At 100 µM of fluvoxamine the formation of 3-hydroxyquinine, 2'-quininone, (10R)-11-dihydroxydihydroquinine, and (10S)-11-dihydroxydihydroquinine was inhibited by 40, 17, 52, and 59%, respectively. The degree of inhibition by the various inhibitors was about the same at 20 and 50 µM of quinine.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Mean effect of different concentrations of (A) ketoconazole, (B) troleandomycin, and (C) fluvoxamine on the formation of 3-hydroxyquinine, 2'-quininone, (10R)-11-dihydroxydihydroquinine and (10S)-11-dihydroxydihydroquinine in human liver microsomes (n = 4) at 20 ìM of quinine.

Correlation Studies. Correlations between the formation rates of quinine metabolites and the activity of the different marker probes for CYP1A2, 2C9, 2C19, 2D6, and 3A4 are shown in Table 2. There were significant correlations between the formation of the four metabolites of quinine and the reactions reflecting the activity of CYP3A4: midazolam -hydroxylation, testosterone 6-hydroxylation, and R-omeprazole sulfoxidation. No correlation was observed between the 3-hydroxylation of quinine and the CYP2C19 activity measured as S-mephenytoin hydroxylation. However, a correlation was observed when the CYP2C19 catalyzed R-omeprazole 5-hydroxylation was measured at about V_max (75 µM) but not at the lower concentration, about K_m (5 µM). A significant correlation was found between the rate of formation of (10R)-11-dihydroxydihydroquinine and losartan oxidation at both high and low concentrations. No significant correlations were observed between the formation of the four metabolites and the activities of CYP1A2 and CYP2D6 measured by caffeine and dextromethorphan, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This is the first report in which the relative contribution of four metabolites of quinine in the overall metabolism of the drug in human liver microsomes was studied. The enzyme kinetic parameters clearly show, for the first time that the 3-hydroxyquinine metabolite is indeed the major metabolite of quinine in human liver microsomes, as shown in Table 1. The formation of 3-hydroxyquinine and 2'-quininone by recombinant human CYP3A4 demonstrated the involvement of this enzyme. As shown in Fig. 1, A and B, the almost complete inhibition of the formation of 3-hydroxyquinine by ketoconazole and troleandomycin indicates that CYP3A4 is the major enzyme of importance for this metabolic pathway. Also the formation of 2'-quininone is primarily dependent on CYP3A4 as shown by the extensive inhibition by ketoconazole and troleandomycin. This is consistent with previously reported data for 3-hydroxyquinine, both in vitro (Zhao and Ishizaki, 1997) and in vivo (Mirghani et al., 1999). Partial inhibition was observed by ketoconazole and troleandomycin (58 and 57%, respectively) on the formation of (10S)-11-dihydroxydihydroquinine indicating the involvement of CYP3A4. On the other hand, the role of CYP3A4 was negligible in the formation of (10R)-11-dihydroxydihydroquinine since it was neither inhibited by CYP3A4 inhibitor nor observed in recombinant CYP3A4. CYP3A4 may play a minor role in the formation of (10S)-11-dihydroxydihydroquinine, which may explain the partial inhibition observed by CYP3A4 inhibitors.

Fluvoxamine inhibits the activity of CYP2C19 (Jeppesen et al., 1997), CYP1A2 (Brosen et al., 1993) and at high concentrations also CYP3A4 activity (Fleishaker and Hulst, 1994). The incomplete selectivity limits the possibility of drawing conclusions when using this drug in inhibition experiments. One reason for using fluvoxamine in this in vitro inhibition study was that it had been used in low and presumably CYP1A2-selective doses in a previously reported in vivo study of quinine metabolism (Mirghani et al., 1999). The extent of inhibition by fluvoxamine on the formation of (10R)- and (10S)-11-dihydroxydihydroquinine (59 and 52%) was higher than that on the formation of 3-hydroxyquinine and 2'-quininone (40 and 17%), respectively. Its effect on the formation of (10R)- and (10S)-11-dihydroxydihydroquinine may be rationalized by the involvement of CYP2C19 and/or CYP1A2 in these metabolic pathways, which is in agreement with the lack of inhibition or incomplete inhibition by the two CYP3A4 inhibitors. Since no correlation was observed between the formation of the latter metabolites and caffeine N-demethylation, the involvement of CYP1A2 may be excluded. On the other hand, the lack of effect of fluvoxamine on the formation of 2'-quininone and the poor effect on the formation of 3-hydroxyquinine excludes any significant contribution of CYP2C19 and CYP1A2 in the formation of these two metabolites and supports the conclusion that CYP3A4 is the major enzyme involved. The slight effect seen on the formation of 3-hydroxyquinine might be due to the fact that fluvoxamine has some inhibitory effect on CYP3A4 at high concentrations. The formation of 3-hydroxyquinine was not inhibited by fluvoxamine in vivo (Mirghani et al., 1999), which is in agreement with the results in the present study.

The involvement of CYP3A4 in the formation of the three quinine metabolites is also indicated by the significant correlations between the formation of the metabolites and midazolam -hydroxylation, testosterone 6-hydroxylation, and R-omeprazole sulfoxidation (Table 2). The significant correlation observed between the formation of (10R)-11-dihydroxydihydroquinine and losartan hydroxylation may indicate the involvement of CYP2C9 in this metabolic pathway. The lack of correlation between the formation of quinine metabolites and caffeine 3-demethylation and dextromethorphan O-demethylation excludes the involvement of CYP1A2 and CYP2D6.

This is the first report on the role of metabolites other than 3-hydroxyquinine in the biotransformation of quinine in vitro. We conclude that formation of 3-hydroxyquinine is the major route of quinine metabolism of those we studied in vitro. The formation of 3-hydroxyquinine and 2'-quininone is entirely catalyzed by CYP3A4, and the formation of (10S)-11-dihydroxydihydroquinine may partly be catalyzed by this isoenzyme. The 2'-quininone, (10R)- and (10S)-11-dihydroxydihydroquinine have minor roles in the biotransformation of quinine in vitro in human liver microsomes. The roles of these metabolites in the metabolism and elimination of quinine in vivo will remain to be investigated. Results from this in vitro study and the earlier published in vivo study (Mirghani et al., 1999) suggest that quinine might be a suitable probe for the activity of CYP3A4 in man, which is to be further evaluated.