Effects of Rifampin and Ketoconazole on Pharmacokinetics of Morinidazole in Healthy Chinese Subjects

Abstract

Morinidazole, a 5-nitroimidazole antimicrobial drug, has been approved for the treatment of amoebiasis, trichomoniasis, and anaerobic bacterial infections in China. It was reported that drug-drug interaction happened after the coadministration of ornidazole, an analog of morinidazole, and rifampin or ketoconazole. Therefore, we measured the plasma pharmacokinetics (PK) of morinidazole and its metabolites in the healthy Chinese volunteers prior to and following the administration of rifampin or ketoconazole using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The area under the concentration-time curve from time 0 to time t (AUC_0-t) and maximum concentration in serum (C_max) of morinidazole were decreased by 28% and 23%, respectively, after 6 days of exposure to 600 mg of rifampin once daily; the C_maxs of N⁺-glucuronides were increased by 14%, while their AUC_0-ts were hardly changed. After 7 days of exposure to 200 mg of ketoconazole once daily, the AUC_0-t and C_max of the parent drug were not affected significantly. C_maxs of N⁺-glucuronides were decreased by 23%; AUC_0-ts were decreased by 14%. The exposure of sulfate conjugate was hardly changed after the coadministration of rifampin or ketoconazole. Using recombinant enzyme of UGT1A9 and human hepatocytes, the mechanism of the altered PK behaviors of morinidazole and its metabolites was investigated. In human hepatocytes, ketoconazole dose dependently inhibited the formation of N⁺-glucuronides (50% inhibitory concentration [IC₅₀], 1.5 μM), while rifampin induced the mRNA level of UGT1A9 by 28% and the activity of UGT1A9 by 53%. In conclusion, the effects of rifampin and ketoconazole on the plasma exposures of morinidazole and N⁺-glucuronide are less than 50%; therefore, rifampin and ketoconazole have little clinical significance in the pharmacokinetics of morinidazole.

INTRODUCTION

Morinidazole is developed as a 5-nitroimidazole antimicrobial injection with potent activities against anaerobic Gram-negative sporeless bacilli and Gram-positive cocci (1). Its efficacy against Clostridium perfringens, Bacteroides fragilis, Veillonella parvula, Bacteroides distasonis, Bacteroides ovatus, Bacteroides vulgatus, and Bacteroides melaninogenicus is equal to that of ornidazole and superior to those of metronidazole and tinidazole. It has been approved for the treatment of amoebiasis, trichomoniasis, and anaerobic bacterial infections in China. However, its antibacterial spectrum is still limited; thus, there is the possibility of concomitant administration with other antibacterial agents in the treatment of mixed infections. It was reported that drug-drug interaction (DDI) happened after the coadministration of ornidazole, an analog of morinidazole, with rifampin or ketoconazole. The pharmacokinetic parameters of ornidazole in healthy volunteers—area under the curve (AUC), peak concentration (C_max), elimination half-life (t_1/2), and clearance (CL)—were decreased by 21.16%, 20.43%, 18.11%, and 32.14%, respectively, by rifampin. The altered pharmacokinetic parameters were also observed in the healthy subjects with the coadministration of ketoconazole (2, 3). Metronidazole, another analog of morinidazole, led to warfarin-caused hypoprothrombinemia in healthy subjects through increasing the serum level of warfarin (4). However, morinidazole had no significant effects on the pharmacokinetics and pharmacodynamics of warfarin (5). It is still unknown whether DDI between morinidazole and rifampin or ketoconazole occurs.

The primary metabolic pathways of morinidazole are N⁺-glucuronidation mediated by UGT1A9 and sulfation. Morinidazole as a racemate is regio- and stereoselectively metabolized to R-form and S-form N⁺-glucuronides and is detectable in plasma and urine. After intravenous infusion of 500 mg of morinidazole to healthy volunteers, approximately 70% of the dose was recovered in urine, of which there was 21.2% parent drug, 28.4% R-form N⁺-glucuronide, 6.6% S-form N⁺-glucuronide, and 13% sulfate conjugate. In the renally impaired patients, the AUCs of N⁺-glucuronide and the sulfate conjugate were dramatically augmented and their renal clearance was reduced by more than 85%. Impaired function of renal transporters OAT1 and OAT3 in the patients with renal impairment is responsible for significant changes in the pharmacokinetic behaviors of morinidazole and conjugated metabolites (6).

Rifampin is an inducer widely involved in the upregulation of phase I enzymes such as CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, and CYP3A4, phase II enzymes such as UDP-glucuronosyltransferases (UGTs) and glutathione S-transferases (GSTs), and transporters such as p-glycoprotein (P-gp) via pregnane X receptor (PXR) and constitutive androstane receptor (7, 8). In addition, rifampin-mediated DDI via UGTs is common in many cases (9). Ketoconazole, a nonspecific inhibitor of CYP3A4, also can inhibit UGTs (10, 11). DDI caused by the induction or inhibition of phase II enzyme is also of the great concern for both scientists and the pharmaceutical industry. Therefore, we assessed the effects of multiple doses of rifampin or ketoconazole on the pharmacokinetics of morinidazole after a single intravenous infusion of 500 mg of morinidazole in healthy subjects.

MATERIALS AND METHODS

Study participants.

Healthy men (n = 6) and women (n = 6), aged 18 to 40 years, were enrolled in each study after providing written informed consent. The protocols for each study were approved by the institutional ethics committee, and each study was conducted in the accordance with good clinical practice and the Declaration of Helsinki. All subjects abstained from xanthine-containing food and drinks, Seville oranges, grapefruit and grapefruit juice, and alcohol for 36 h before admission and for the duration of the trial. No medication or herbal supplement was permitted to be taken 14 days prior to and during the trial. All subjects were nonsmokers or light smokers. Tobacco products were discontinued during the trial.

Study design.

A randomized, two-way crossover study was conducted at General Hospital of Chengdu Military Region (Chengdu, China). It was designed to assess the drug interaction between rifampin and morinidazole. Twelve healthy subjects were enrolled and assigned randomly into two groups equally. Subjects received 600 mg of rifampin (150 mg/tablet; Shanghai Xinyi Jiufu Pharmaceutical Co. Ltd., Shanghai, China) once daily for 6 days, followed by intravenous infusion of 500 mg of morinidazole. On the last day of the first and second treatment sequences, blood samples were collected over 36 h. The washout period was 14 days.

To assess the effect of ketoconazole on the pharmacokinetics of morinidazole, we conducted a nonrandomized, self-controlled clinical study at Tongji Hospital, affiliated with Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China). Twelve healthy subjects were enrolled and received 500 mg of morinidazole by continuous intravenous infusion on day 1. Participants received a second dose of 500 mg of morinidazole on day 8, followed by a 7-day course of oral dosing of 200 mg of ketoconazole (200 mg/tablet; Xian-Janssen Pharmaceutical Ltd., Xi'an, China) once daily.

Subjects in both studies underwent medical evaluations within 4 weeks before starting and 2 weeks after the completion of the study. Subjects were confined to a clinical research unit from the evening before dosing with morinidazole until the last plasma sample was taken.

Pharmacokinetic sampling and sample preparation.

For both studies, blood samples (4 ml) were collected and placed in heparinized tubes predosing, at 0.167, 0.333, and 0.667 h after the initiation of the infusion, and at 0.25, 0.50, 0.75, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0, 12, 24, and 36 h after the end of the infusion. Plasma samples were stored at −20°C until analysis.

The method of sample preparation was the same as that previously reported (12). Briefly, a 50-μl aliquot of plasma sample was mixed with 50 μl of 500-ng/ml metronidazole solution (internal standard), 50 μl of methanol-water (50:50, vol/vol), and 150 μl of acetonitrile. The mixture was centrifuged at 11,000 × g for 5 min. The supernatant was dried under nitrogen (N₂), and the residue was reconstituted with the mobile phase for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

LC-MS/MS analysis.

The concentrations of morinidazole, M7 (S-form sulfate conjugate), M8-1 (S-form N⁺-glucuronide), and M8-2 (R-form N⁺-glucuronide) were measured by a validated LC-MS/MS method previously published (12). Briefly, chromatography was carried out using a Phenomenex Synergi Hydro-RP column (particle size, 4 μm; 50 by 2 mm [inside diameter]; Phenomenex, CA) held at room temperature. Elution of the analytes was performed using a gradient elution. Detection was performed by MS/MS in the positive mode on an API 4000 triple quadrupole mass spectrometer with multiple-reaction monitoring of m/z 271 → 144 for morinidazole, m/z 351 → 271/144 for M7, m/z 447 → 144/320/100 for M8-1/M8-2, and m/z 172 → 82 for metronidazole. The lower limits of quantification (LLOQ) for morinidazole, M7, M8-1, and M8-2 were 10.0, 2.50, 3.00, and 10.0 ng/ml, respectively.

Inhibition of ketoconazole on UGT1A9 assay.

The inhibitory effect of ketoconazole on morinidazole glucuronidation was evaluated in the recombinant UGT1A9 and human hepatocyte system. To evaluate the 50% inhibitory concentration (IC₅₀) of ketoconazole for UGT1A9, a premix containing 0.5 mg/ml of UGT1A9, 50 mM Tris-HCl (pH 7.5), 8 mM MgCl₂, 25 μg/ml of alamethicin, and 50 μM morinidazole with or without series concentrations of ketoconazole was incubated at 37°C for 5 min, and then 2 mM UDPGA was added to initiate the reaction. After incubation at 37°C for 60 min, an equal volume of chilled acetonitrile was added to terminate the reaction.

A 200-μl aliquot of terminal mixtures were centrifuged at 14,000 rpm for 5 min. The supernatant was taken out and dried with N₂. The residue was reconstituted with the mobile phase for LC-MS/MS analysis.

Effects of rifampcin and ketoconazole on the metabolism of morinidazole in human hepatocytes.

Human hepatocytes were purchased from Xenotech (KS) and In Vitro Technologies (Victoria, Australia) for inhibition and induction studies, respectively.

In the inhibition study, 2 × 10⁶ hepatocytes/ml were preincubated with 0, 1, and 10 μM ketoconazole at 37°C and 80 rpm for 30 min before the addition of 50 μM morinidazole. The incubation was performed for further 2 h and terminated with ice-cold acetonitrile.

In the induction study, 7 × 10⁵ hepatocytes/ml were seeded in the 48-well plates and treated with 20 μM rifampin for 3 days. On the fourth day, 50 μM morinidazole was incubated with hepatocytes for 2 h, followed by the addition of 200 μl of ice-cold acetonitrile. RNAs of hepatocytes were extracted using TRIzol reagent (Invitrogen, CA).

The productions of M8-1, M8-2, and M7 were quantitated by LC-MS/MS as described above. The transcription level of UGT1A9 was evaluated by quantitative real-time PCR (qPCR). cDNA was synthesized from 1 μg of total RNA using a SuperScript III reverse transcriptase kit (Invitrogen) [synthesis primers were oligo(dT)]. The reaction was performed in a 10-μl volume of 25-ng/μl cDNA, 250 nM each primer, and 5 μl of QuantiFast SYBR green PCR master mix (Qiagen, CA) on an Applied Biosystems ViiA7 real-time PCR system. qPCR conditions were 95°C for 10 min, 95°C for 15 s, and 60°C for 60 s for 40 cycles. The sequences of primers for UGT1A9 were as follows: forward, 5′-CCAAACACCTGTTACGGAGTA-3′, and reverse, 5′-AGGCTTCAAATTCCATAGGCA-3′. The sequences of primers for PPIA were as follows: forward, 5′-CACCGTGTTCTTCGACATTG-3′, and reverse, 5′-TCCTTTCTCTCCAGTGCTCAG-3′. The levels of cDNA were quantitated by the comparative threshold cycle method using PPIA as an internal standard.

Pharmacokinetic and statistical analysis.

The main pharmacokinetic parameters, AUC, CL, and t_1/2, were calculated by noncompartmental methods using WinNonlin 5.3 software (Pharsight, Mountain View, CA). C_max and peak time (T_max) were obtained from the plasma concentration-time curve of morinidazole and main metabolites. T_max was expressed as median ± interquartile range (IQR); other parameters were expressed as means ± standard deviations (SD).

The statistical difference between control group and coadministration group was analyzed by the statistical method used for the bioequivalence analysis. A 90% two-sided confidence interval (CI) was determined to evaluate the C_max/AUC_0-t ratio of morinidazole coadministration to administration alone (where 0-t indicates time 0 to time t). If the ratio is in the range of 80% to 125%, it indicates that the inhibitor or inducer does not have an effect on the pharmacokinetics of morinidazole. The evaluation on the potency of induction or inhibition of rifampin or ketoconazole was based on the FDA Guidance for Industry Drug Interaction Studies. Specifically, strong inhibitors or inducers are defined as those drugs that increase the AUC more than 5-fold or decrease it by more than 80%, respectively, moderate inhibitors or inducers are defined as those drugs that increase the AUC 2- to 5-fold or decrease it by between 50% and 80%, and weak inhibitors or inducers are defined as those drugs that increase the AUC between 0.25- to 2-fold or decrease it between 20% and 50%. T_max differences between morinidazole alone and coadministration were examined using a nonparametric test (Mann-Whitney).

RESULTS

Demographics.

In the rifampin-morinidazole study, 12 subjects were enrolled and completed the study. The mean age (±SD) age was 22.5 ± 1.2 years, the mean body weight was 56.9 ± 7.3 kg, and the mean body mass index was 20.6 ± 1.4.

In the ketoconazole-morinidazole study, 12 subjects were enrolled and completed the study. The mean (± SD) age was 23.9 ± 1.2 years, the mean body weight was 59.3 ± 6.5 kg, and the mean body mass index was 21.3 ± 1.6.

Pharmacokinetic study.

The mean concentration-time profiles of morinidazole, M7, M8-1, and M8-2 with or without rifampin are shown in Fig. 1. The corresponding pharmacokinetic parameters, including AUC, CL, C_max, T_max, and t_1/2, are summarized in Table 1. The mean plasma concentrations of morinidazole in the combination treatment were lower than that in the single treatment (Fig. 1). The kinetic behavior changes of morinidazole accompanied the change of M8-1. The geometric mean ratios (morinidazole/rifampin versus morinidazole alone) of morinidazole C_max and AUC_0-∞ were 0.77 and 0.72, respectively, with 90% CIs of 0.66 to 0.91 and 0.69 to 0.76, respectively. Rifampin increased the clearance 1.4-fold. The half-life decreased to 4.02 min from 5.55 min in the combination treatment group compared with the morinidazole-alone group. The C_max of M8-1 increased 1.3-fold in the presence of rifampin. There was a decreasing trend of T_max of M7 and M8-1. There was no significant change in the other kinetic parameters of M8-2 and M7.

FIG 1.

Mean plasma concentration-time profiles of morinidazole (A), M7 (B), M8-1 (C). and M8-2 (D) after an intravenous infusion of 500 mg of morinidazole to healthy Chinese subjects with or without pretreatment with rifampin. Data are expressed as means ± SD (n = 12).

TABLE 1.

Pharmacokinetic parameters of morinidazole and its metabolites following an intravenous infusion of 500 mg of morinidazole to healthy volunteers with or without treatment with rifampin^a

Analyte	Treatment	Cmax, μg · ml−1	AUC0-t, μg · h · ml−1	AUC0-∞, μg · h · ml−1	Tmax, h	t1/2, h	CL, liter/h
Morinidazole	Morinidazole	11.7 ± 2.46	65.7 ± 7.86	66.4 ± 7.90	0.667 ± 0.000	5.55 ± 0.88	7.63 ± 0.90
	Morinidazole + rifampin	9.08 ± 2.01	47.6 ± 7.34	47.9 ± 7.33	0.667 ± 0.250	4.02 ± 0.47	10.7 ± 1.85
	Ratio (90%)	0.77 (0.66–0.91)	0.72 (0.69–0.76)
M7	Morinidazole	0.202 ± 0.048	1.75 ± 0.39	1.80 ± 0.38	1.17 ± 0.25	6.22 ± 0.97
	Morinidazole + rifampin	0.211 ± 0.056	1.50 ± 0.36	1.55 ± 0.36	0.917 ± 0.253	4.64 ± 0.75
	Ratio (90%)	1.04 (0.96–1.13)	0.85 (0.80–0.92)
M8-1	Morinidazole	0.220 ± 0.071	2.37 ± 0.42	2.45 ± 0.42	1.92 ± 1.50	6.38 ± 1.48
	Morinidazole + rifampin	0.282 ± 0.046	2.48 ± 0.53	2.56 ± 0.52	1.30 ± 0.75	5.04 ± 0.64
	Ratio (90%)	1.32 (1.17–1.48)	1.04 (0.93–1.16)
M8-2	Morinidazole	1.90 ± 0.57	15.1 ± 2.31	15.3 ± 2.29	1.67 ± 0.94	4.13 ± 0.59
	Morinidazole + rifampin	2.14 ± 0.44	14.7 ± 2.73	14.9 ± 2.69	1.67 ± 1.38	3.28 ± 0.48
	Ratio (90%)	1.15 (1.06–1.23)	0.97 (0.91–1.03)

^aValues other than ratios are means ± SD. CI, confidence intervals; ratio, geometric least-squares mean ratio (90% CI) for AUC_0-t and C_max. T_max is expressed as median ± IQR.

The mean concentration-time profiles of morinidazole, M7, M8-1, and M8-2 after the intravenous infusion of 500 mg of drug to healthy subjects with or without ketoconazole are shown in Fig. 2. The corresponding pharmacokinetic parameters are listed in Table 2. Unlike the effect of rifampin on the morinidazole, the pharmacokinetic behavior of morinidazole and M7 was hardly changed by ketoconazole. The AUC and C_max of N⁺-glucuronides were decreased because of the influence of ketoconazole on the metabolism of morinidazole. Specifically, the AUC and C_max of M8-1 decreased by 20% and 29%, respectively; T_maxs tend to decrease by 18%. The AUC and C_max of M8-2 decreased by 13% and 24%, respectively, accompanied by about a 24% reduction trend in T_max.

FIG 2.

Mean plasma concentration-time profiles of morinidazole (A), M7 (B), M8-1 (C), and M8-2 (D) after an intravenous infusion of 500 mg of morinidazole to healthy Chinese subjects with or without pretreatment with ketoconazole. Data are expressed as means ± SD (n = 12).

TABLE 2.

Pharmacokinetic parameters of morinidazole and its metabolites following an intravenous infusion of 500 mg of morinidazole to healthy volunteers with or without treatment with ketoconazole^a

Analyte	Treatment	Cmax, μg · ml−1	AUC0-t, μg · h · ml−1	AUC0-∞, μg · h · ml−1	Tmax, h	t1/2, h	CL, liter/h
Morinidazole	Morinidazole	10.9 ± 2.25	74.9 ± 12.7	76.0 ± 12.7	0.667 ± 0.000	6.15 ± 0.59	6.74 ± 1.03
	Morinidazole + ketoconazole	11.2 ± 1.98	77.9 ± 11.3	78.8 ± 11.4	0.667 ± 0.000	5.86 ± 0.418	6.46 ± 0.91
	Ratio (90%)	1.04 (0.93–1.17)	1.04 (1.01–1.07)
M7	Morinidazole	0.17 ± 0.051	1.70 ± 0.45	1.75 ± 0.46	1.04 ± 0.25	7.04 ± 0.56
	Morinidazole + ketoconazole	0.164 ± 0.044	1.67 ± 0.42	1.71 ± 0.43	0.917 ± 0.253	6.87 ± 0.54
	Ratio (90%)	0.97 (0.93–1.02)	0.98 (0.95–1.02)
M8-1	Morinidazole	0.248 ± 0.063	2.97 ± 0.75	3.05 ± 0.78	1.42 ± 0.88	6.96 ± 0.81
	Morinidazole + ketoconazole	0.177 ± 0.049	2.36 ± 0.53	2.45 ± 0.55	1.17 ± 1.25	7.26 ± 0.90
	Ratio (90%)	0.71 (0.66–0.76)	0.80 (0.76–0.85)
M8-2	Morinidazole	2.06 ± 0.59	17.2 ± 4.08	17.4 ± 4.12	1.54 ± 0.44	4.23 ± 0.45
	Morinidazole + ketoconazole	1.59 ± 0.54	14.9 ± 3.66	15.2 ± 3.56	1.17 ± 0.69	4.58 ± 0.51
	Ratio (90%)	0.76 (0.72–0.80)	0.87 (0.84–0.90)

^aValues other than ratios are means ± SD. CI, confidence intervals; ratio, geometric least-squares mean ratio (90% CI) for AUC_0-t and C_max. T_max is expressed as median ± IQR.

Clinical safety.

Morinidazole alone and the coadministration with rifampin and ketoconazole were well tolerated by volunteers. No serious adverse event or early withdrawals were reported during the trial. Mild elevation of the serum level of aspartate transaminase (AST) and alanine aminotransferase (ALT) occurred in only one subject, who was allowed to continue the trial. Another subject was reported to have a mild diarrhea for 3 days and continued the study after treatment with levofloxacin hydrochloride (250 ml), 0.9% sodium chloride injection, and injection of 500 ml of 5% dextrose.

The effect of rifampin on morinidazole in human hepatocytes.

The mRNA expression of UGT1A9 was measured by quantitative reverse transcription-PCR (qRT-PCR), and results are shown in Fig. 3B. The enzyme activities induced by rifampin were determined by the measurement of the production of metabolites (Fig. 3A). After a 3-day induction with rifampin, the mRNA level of UGT1A9 was increased 1.3-fold and the production of M8-2 was increased around 1.5-fold. There was no significant change in the production of other metabolites.

FIG 3.

Effects of rifampin and ketoconazole on the metabolism of morinidazole in the human hepatocyte. The mRNA level of UGT1A9 was quantitated by qRT-PCR after 3-day treatment with rifampin. The enzyme activities influenced by rifampin and ketoconazole were evaluated based on the productions of metabolites. Data are expressed as means + SD (n = 3).

Effect of ketoconazole on the metabolism of morinidazole.

The effect of ketoconazole on the metabolism of morinidazole was determined by the quantitation of the productions of metabolites of morinidazole, as shown in Fig. 3C. The biotransformations of morinidazole to M8-1 and M8-2 were reduced by ketoconazole dose dependently, while the production of M7 was only slightly decreased at 0.1 μM and 10 μM ketoconazole.

The inhibitory potency of ketoconazole on the transformation of morinidazole to N⁺-glucuronides was further analyzed using recombinant enzyme of UGT1A9, as shown in Fig. 4. The IC₅₀ values of ketoconazole for glucuronidation were both 1.5 μM, which indicated that ketoconazole is a moderate inhibitor of UGT1A9.

FIG 4.

Effects of ketoconazole on morinidazole glucuronidation after 60 min of incubation with recombinant enzyme of UGT1A9. Each point represents the mean value from triplicate experiments. “% control UGT1A9 activity” means the ratio of the production of glucuronide with ketoconazole to that without ketoconazole.

DISCUSSION

A previous study revealed that morinidazole is hardly metabolized by phase I enzymes when incubated with human liver microsomes for 1 h, while it was mainly metabolized via the phase II pathway mediated by UGT1A9. Although the UGT-mediated drug interaction is confirmed in many in vitro studies, there is no direct clinical evidence because of the lack of the quantitation of glucuronide metabolites in the clinical study (9). In this study, morinidazole, its two glucuronide metabolites, and sulfate conjugate in the plasma were simultaneously measured after the concomitant administration of morinidazole and rifampin or ketoconazole, and their pharmacokinetic parameters—AUC, CL, C_max, T_max, and t_1/2—were compared. The in vitro study for the assessment of UGT1A9 induction and inhibition supported in part the in vivo results.

Rifampin is a common interacting drug involved in glucuronidation-mediated DDI (13–18). In our study, the C_max and AUC of morinidazole in healthy subjects were decreased significantly, by around 20% to 30%, after pretreatment with rifampin, while only the C_max of M8-1was increased 1.3-fold, without a change of AUC. To explain the PK behavior of morinidazole, we employed human hepatocytes to investigate the effect of rifampin on the metabolism of morinidazole. The production of M8-1 was increased 1.5-fold, within the expected range, while the level of parent drug was not changed. It was reported that rifampin can increase UGT1A9-mediated glucuronidation of propofol 1.7-fold (19). In addition, the upregulation of PXR in colon cancer cells can promote the glucuronidation of SN38 to its glucuronide, accompanied by an increase in mRNA level of UGT1A9 (20), indicating the regulation of the expression of UGT1A9 by PXR. In our study, the mRNA level of UGT1A9 was slightly induced, 1.3-fold, in the presence of rifampin in human hepatocytes. Therefore, the effect of rifampin on the mRNA level and activity of UGT1A9 in vitro is consistent with the results in vivo in some extent. Rifampin also can induce the expression of P-gp via PXR; thus, we wonder whether the urinary excretion of morinidazole could be facilitated through the induction of P-gp in the kidney. However, the efflux ratio of morinidazole was less than 2 using the P-gp-overexpressing cell line MDCKII, indicating that morinidazole was not the substrate of P-gp (data not shown). Besides CYP3A4, rifampin is also an inducer widely involved in the upregulation of phase I enzymes such as CYP1A2, CYP2B6, CYP2C8, CYP2C9, and CYP2C19 and phase II enzymes such as UGTs and GSTs (7). It may be possible that the other unknown disposition pathway mediated the lower exposure in the combined use of morinidazole and rifampin.

The fold change of AUC caused by the inhibition of the glucuronidation-mediated DDI is usually less than two (21). In our study, the C_max and AUC of morinidazole in healthy volunteers were hardly changed, while a <2-fold increase of C_max and AUC of N⁺-glucuronides was observed in the presence of ketoconazole. In the in vitro experiment, ketoconazole dose dependently inhibited the transformation of morinidazole to N⁺-glucuronides with moderate potency, correlating with the results in vivo. It was reported that the glucuronidation of SN38 mediated by UGT1A9 also can be reduced by ketoconazole (22). The inhibition of ketoconazole on the production of N⁺-glucuronides did not decrease the exposure of parent drug, morinidazole, in plasma. Concentrations of morinidazole in plasma are much higher than that of N⁺-glucuronides, so it is not likely that the minor changes of N⁺-glucuronides would affect the exposure of the parent drug. There are four factors related to the low ratio of AUC in the presence of inhibitor to that in the absence of inhibitor, including the multiple enzymes involved in the metabolism, high K_m value, the low ratio of hepatic inhibitor concentration to K_i value, and low hepatic extraction (21). Our previous study suggests that morinidazole is primarily metabolized into three metabolites through two pathways, UGT1A9 and sulfatase. Sulfation only accounts for a small part of the transformation of morinidazole and is not inhibited by ketoconazole in healthy subjects. Thus, UGT1A9 is considered to be primarily responsible for the metabolism of morinidazole. However, considering that in vitro K_m values for N⁺-glucuronides by recombinant UGT1A9 are around 3 mM and the therapeutic concentration range of morinidazole is an order of magnitude lower (23), UGT1A9-catalyzed glucuronidation of morinidazole would not likely to be saturated by the inhibition of ketoconazole in vivo; therefore, the exposures of N⁺-glucuronides would not be affected dramatically even though the in vitro experiment showed the moderate inhibitory effect of ketoconazole on morinidazole glucuronidation by UGT1A9. Furthermore, since the renal clearance of morinidazole and its metabolites accounts for around 70% of the dose, and ketoconazole is a moderate inhibitor of P-gp (24) and UGT1A9, it would be speculated that ketoconazole may reduce the exposure of N⁺-glucuronides through the interruption of urinary excretion. However, the unchanged accumulative urinary excretion of morinidazole and its metabolites indicates that ketoconazole has no effect on their renal elimination (data not shown). Therefore, the inhibitory effect of ketoconazole only slightly reduced the exposure of N⁺-glucuronides without the interruption of the parent drug through hepatic metabolism.

In conclusion, this study demonstrated that coadministration of rifampin or ketoconazole had little effect on the systemic exposure of morinidazole or its main metabolites in humans at a single intravenous dose of 500 mg, although the in vitro or in vivo evidence indicated that rifampin was a weak inducer and ketoconazole was a moderate inhibitor of UGT1A9. Hence, the effect of ketoconazole or rifampin on the pharmacokinetics of morinidazole has little clinical significance.