Regular Article

Analysis of the Pharmacokinetic Boosting Effects ofRitonavir on Oral Bioavailability of Drugs in Mice

Atsuko TOMARU, Mariko TAKEDA-MORISHITA†,*, Hirokazu BANBA and Kozo TAKAYAMA

Department of Pharmaceutics, Hoshi University, Tokyo, Japan

Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: Ritonavir dramatically increases the bioavailability of a variety of concurrently administereddrugs by inhibition of metabolic enzymes and drug transporters. The purpose of this study was to investigatethe extent to which ritonavir’s inhibition of drug transporters and/or CYP3A contributes to the increasedoral bioavailability in mice. The area under the plasma concentration-time curves (AUC) for orallyadministered saquinavir after coadministration with 50mg/kg ritonavir dramatically increased (325-fold).As a result, the bioavailability, Fa·Fg and Fh increased 75-, 38- and twofold, respectively. In addition, theincrease in the AUC predicted from the in vitro Ki value was ninefold, which was derived from the inhibitionof metabolic enzymes by ritonavir in the liver. The remaining 36-fold increase in the AUC was consideredto be derived from the inhibition in the small intestine. The AUCinf for probe substrate midazolam,fexofenadine, and pravastatin increased after the oral administration of ritonavir by only five-, 13-, andsevenfold, respectively. Moreover, the AUC0­12 for saquinavir was affected negligibly by itraconazole. Theseresults indicate ritonavir mainly affects the first-pass effect of saquinavir in the small intestine, increasingthe bioavailability of orally administered saquinavir. Furthermore, cyp isoforms other than CYP3A,which contribute to the metabolism of saquinavir in humans, are involved in the metabolism of saquinavirin mice.

Keywords: ritonavir; drug-drug interactions; P-gp; CYP3A; oral bioavailability

Introduction

Drug-drug interactions alter the pharmacokinetic profiles ofconcurrently administered drugs, and may lead to a loss of effi-cacy or induce adverse events. Therefore, it is important to under-stand the mechanisms and magnitudes of drug-drug interactions.Ritonavir (RTV) boosting is a drug therapy used in the treatment ofhuman immunodeficiency virus (HIV) that utilizes the principle ofdrug-drug interaction. RTV was first developed as an HIV type-1(HIV-1) protease inhibitor (PI) and used for a single PI treatment.However, since its potent inhibitory effect on the cytochrome P450(CYP) 3Awas clarified, RTV has been used in multidrug treatmentregimens for HIV.1,2) In antiretroviral therapies, adequate plasmaconcentrations of PIs are required for their effective anti-HIVactivity. The potent inhibition of CYP3A in the small intestine andliver by RTV improves the bioavailability of drugs and prolongstheir elimination half-lives. As a result, RTV boosting allows bothdose and dosing frequency of the concurrently administered PI tobe reduced.

Saquinavir (SQV) is also a highly selective HIV-1 PI and isadministered as one of the RTV-boosted PIs. The bioavailabilityof orally administered SQV was only 4% in healthy volunteersafter a single dose.3) The activity of the original hard-gel capsuleformulation (Inviraseμ) is limited by the drug’s poor pharmaco-kinetics, resulting in limited long-term viral suppression. Itsreplacement with a soft-gel capsule formulation (Fortovaseμ)also failed to provide adequate drug exposure.4) However, theconcurrent administration of SQV with RTV enhances the plasmaconcentration of SQV, and so SQV is used as a combinationtherapy with RTV. The poor bioavailability of SQV is attributed tothe extensive first-pass metabolism in the small intestine and liver,because SQV is a substrate of CYP3A. Therefore, the boostingeffect of RTV is considered to be mediated through its inhibitoryeffect on CYP3A.

RTV also inhibits many transporters, such as p-glycoprotein (P-gp), breast cancer resistance protein (BCRP), multidrug resistance-associated protein 2 (MRP2), and organic anion-transportingpolypeptide (OATP).5–10) P-gp, BCRP, and MRP2 are the efflux

Received June 3, 2012; Accepted August 17, 2012J-STAGE Advance Published Date: September 11, 2012, doi:10.2133/dmpk.DMPK-12-RG-057*To whom correspondence should be addressed: Mariko TAKEDA-MORISHITA, Ph.D., Faculty of Pharmaceutical Sciences, Kobe Gakuin University,1-1-3 Minatojima, Chuo-ku, Kobe 650-8586, Japan. Tel. ©81-7-8974-4816, Fax. ©81-7-8974-4816, E-mail: mmtakeda@pharm.kobegakuin.ac.jp†Present address: Faculty of Pharmaceutical Sciences, Kobe Gakuin University, 1-1-3 Minatojima, Chuo-ku, Kobe 650-8586, Japan.This study is a part of a research project for the Establishment of Evolutional Drug Development with the Use of Microdose Clinical Trialssponsored by the New Energy and Industrial Technology Development Organization (NEDO).

Drug Metab. Pharmacokinet. 28 (2): 144–152 (2013). Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

144

transporters expressed at the apical membrane in the small intestineand limit the bioavailability of their orally administered substratedrugs.11) Because SQV is also a substrate of P-gp,12) the inhibitionof P-gp by RTV in the small intestine might enhance the bio-availability of SQV. In a clinical study, the AUC for fexofenadine(FEX), a substrate of P-gp, increased twofold after its oraladministration with 100mg of RTV.13) Huisman et al. reported thatthe AUC for 50mg/kg SQV increased 20-fold after its oral admin-istration to mdr1a¹/¹/mdr1b¹/¹ mice.14) Furthermore, Washingtonet al. also reported a 6.5-fold increase in the AUC for 500mg/kgSQVafter its oral administration to mdr1a¹/¹ mice.15) These resultsindicate that the increase in the AUC for SQV can be attributed tothe inhibition of P-gp by RTV. In contrast, Richter et al. reportedthat RTV caused only a 1.4-fold increase the intestinal permeabilityin an in situ perfusion study.16) Thus, the detailed mechanism ofRTV boosting is not well understood.

Recently, it was reported that SQV is a substrate of OATP1B1and OATP1B3 because a significant increase in the accumula-tion of SQV was observed in Xenopus laevis oocytes injectedwith SLCO1B or 1B3 cDNA.17) OATP1B1 and OATP1B3 areinflux transporters and localized on the basolateral membrane ofhepatocytes. As observed in the drug-drug interaction betweencerivastatin and gemfibrozil,18) the inhibition of the influx trans-porters expressed on sinusoidal membrane increases the con-centration of drugs in the liver, causing adverse events. In anin vitro study, the half maximal inhibitory concentration values(IC50) of RTV for OATP1B1 and OATP1B3 were 1.6 and 3.6µM,10) respectively. These values are close to that for P-gp (3.8µM).5) Therefore, it is also important to investigate the RTV-boosting effect on influx transporters, such as OATP1B1 andOATP1B3.

In this study, we assessed the contribution of the RTV effecton the first-pass effect of SQV in the small intestines and liversof mice. The RTV boosting effects on several probe substrateswere also investigated. In addition, an in vitro study using mouseintestinal and liver microsomes was performed to confirm theresults obtained from the in vivo study.

Materials and Methods

Materials: RTV purchased as Norvirμ (80mg/mL) fromAbbot Laboratories (Abbott Park, IL, USA) and midazolam(MDZ) as Dormicum Injectionμ (5mg/mL) purchased fromAstellas Pharma Inc. (Tokyo, Japan) were used in the in vivostudy. In the in vitro study, RTV purchased from BiotrendChemikalien GmbH (Cologne, Germany) and MDZ purchasedfrom Wako Pure Chemical Industries, Inc. (Tokyo, Japan) wereused. SQV was purchased from Sequoia Research Products Ltd.(Pangbourne, UK) and pravastatin (PRV) sodium was purchasedfrom Wako Pure Chemical Industries, Inc. Itraconazole (ITZ)was purchased from Sigma-Aldrich Co. (St. Louis, MO, USA).Fexofenadine hydrochloride was purchased from Toronto ResearchChemicals Inc. (Ontario, Canada). All other chemicals were ofanalytical grade and are commercially available.

Animals: Female ddY mice (11–13 weeks old) were pur-chased from Sankyo Labo Service Corp (Tokyo, Japan). The micewere housed in rooms maintained at 23°C and 55 « 5% relativehumidity, and allowed free access to food and water during theacclimatization period. In the in vivo study, the mice were fastedovernight for at least 12 h, with free access to water before dosing.The animal study was performed at Hoshi University and complied

with the regulations of the Committee on Ethics in the Care and useof Laboratory Animals.

In vivo studies:Preparation of dosing solutions

The SQV dosing solution was prepared as described previ-ously.14) Briefly, SQV was dissolved in an 8% ethanol/4.2% glucosesolution for intravenous administration and in a 16.4% ethanol/3%glucose/15.6% Cremophor EL solution for oral administration.Norvirμ was diluted with ethanol and water to prepare an 8mg/mLsolution. As the control vehicle for Norvirμ, a 43%(v/v) ethanolsolution containing Cremophor EL (105mg/mL), propylene glycol(0.25mg/mL), peppermint oil (3.5mg/mL), and water-free citricacid (2.8mg/mL) was used. This control vehicle was 10-folddiluted by ethanol and water to contain the same volume of vehicleconstituents as RTV dosing solution. For the dose-dependencyof RTV inhibition effect study, Norvirμ was diluted with controlvehicle to prepare 2.4, 8 and 40mg/mL solutions. Then, thesesolutions were diluted with ethanol and water to make each RTVdosing solution (0.24, 0.8 and 4mg/mL). The dosing solutionfor each probe substrate was prepared as follows. Fexofenadinehydrochloride and pravastatin sodium were dissolved in waterto concentrations of 1.5 and 20mg/mL, respectively. DormicumInjectionμ was diluted with water to concentration of 3mg/mL andused as the MDZ dosing solution. The ITZ dosing solution wasprepared as described previously.19) Briefly, 20mg of ITZ wasdissolved with 0.1mL of 12N HCl, and 1.75mL of polyethyleneglycol 400 and 0.15mL of 8N NaOH were added to prepare a10mg/mL ITZ solution. As the control vehicle for ITZ, a solutionwhich contained the same contents of vehicle constituents was used.Effect of RTV on the pharmacokinetics of SQV

The dose-dependent effects of RTV on the pharmacokinetics ofSQV were investigated. The administered doses of RTV were set at1.5, 5, 25 and 50mg/kg. The dose of 1.5mg/kg is the clinicallyrelevant dose (RTV boosting dose: 100 or 200mg/dose, i.e., 1.5–3mg/kg for a 70 kg man). SQV (20mg/kg) was administered orally30min after the oral administration of RTV.

In addition, to investigate the contribution of RTV on the first-pass effect in the liver and small intestine, SQV was administeredorally (20mg/kg) or intravenously (2mg/kg) 30min after the oraladministration of 50mg/kg RTV. For oral administration, micewere given each solution by sonde needle into the stomach. Forintravenous administration, each solution was injected into the tailvein of mice. Thirty µL of blood was taken from the tail veinby using micro-haematocrit capillary tubes and put into the microtube at predetermined time intervals. Plasma was separated bycentrifugation at 15,400 © g for 10min. The plasma samples werestored at ¹20°C until analysis.Effect of RTV on the pharmacokinetics of probe substrates

FEX (10mg/kg) was used as the probe substrate for P-gp, MDZ(10mg/kg) for CYP3A, and PRV (100mg/kg) for OATP1B1.Each probe substrate was administered orally 30min after theoral administration of 1.5 or 50mg/kg RTV. The effect of ITZ,an inhibitor of CYP3A in humans, on the pharmacokinetics ofSQV was also investigated. SQV (20mg/kg) was administeredorally 30min after the oral administration of 50mg/kg ITZ.Thirty µL of blood was taken from the tail vein by using micro-haematocrit capillary tubes and put into the micro tube at pre-determined time intervals. Plasma was separated by centrifuga-tion at 15,400 © g for 10min. The plasma samples were stored at¹20°C until analysis.

Pharmacokinetic Boosting Effect of Ritonavir 145

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

Quantification of the plasma concentration of SQV, FEX, MDZand PRV

For the determination of each substrate concentration in plasma,100µL of acetonitrile was added to the 10 µL of plasma sample.After the protein precipitation by vortex mixing and centrifugation,80 µL of supernatant was evaporated at 40°C under nitrogen. Theresidue was reconstituted with the 100µL of mobile phase and sub-jected to liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis on an API 4000 system (AB SCIEX, Foster City,CA, USA) equipped with the Prominence LC system (ShimadzuCo., Kyoto, Japan). The LC-MS/MS conditions are summarizedin Table 1. The data were acquired with Analyst ver. 1.4.2 (ABSCIEX). The limit of quantification for all compounds was 1ng/mL.

In vitro studies:Preparation of mouse microsomes

Pooled mouse liver and intestinal microsomes were preparedaccording to a previously reported method.20) Briefly, livers andintestines were homogenized separately in three volumes of ice-cold buffer (50mM Tris-HCl, pH 7.4) containing 150mM KCl,1mM phenylmethylsulfonyl fluoride, 1mM EDTA, 1mg/mLtrypsin inhibitor, 10 µM leupeptin, 0.04 unit/mL aprotinin, 1 µMbestatin, and 20% (v/v) glycerol using a Teflon-tipped pestle. Thehomogenates were centrifuged at 9,000 © g for 20min at 4°C, andthe supernatants were then centrifuged at 105,000 © g for 60min at4°C. The microsomal pellets were resuspended in the same bufferand stored at ¹80°C. The protein concentrations were determinedusing a BCA Protein Assay Kit (Pierce Chemical, Rockford, IL,USA).Inhibition study

In preliminary experiments, we confirmed that substratedepletion increased proportionally with time (to 20min) and withprotein concentration (in the range of 0.1 to 1mg/mL protein) forboth the liver and intestinal microsomes. MDZ, SQV, RTV andITZ were dissolved in 50% methanol, 50% dimethyl sulfoxide(DMSO), methanol and DMSO, respectively. The volume oforganic solvent in the incubation mixture was under 1% and thesame volume of each solvent was added to the control sample.

The mouse liver and intestinal microsomes were incubatedwith SQV or MDZ with or without an inhibitor in an NADPH-generating system containing 2.5mM NADP, 25mM glucose 6-phosphate, 2 units of glucose-6-phosphate dehydrogenase, and10mM MgCl2 in a total volume of 0.2mL of 100mM phosphatebuffer (pH 7.4). After preincubation for 5min at 37°C, the reactionwas initiated by the addition of the NADH-generating system.After incubation, the reaction was terminated by the additionof 0.4mL of ice-cold acetonitrile and briefly vortex mixed. Thesamples were centrifuged at 15,400 © g for 15min to precipitatethe protein. The supernatant was evaporated at 40°C under nitro-gen. The residue was reconstituted with the mobile phase and thedepletion of SQV or MDZ was analyzed by high-performanceliquid chromatography with ultraviolet detection (HPLC-UV).

Quantification of SQV and MDZ concentration in the in vitrosamples

HPLC was performed with a Prominence LC system (ShimadzuCo.), equipped with a TSKgelμ ODS-100V column (2.1 © 50mm,5µm; Tosoh Co., Tokyo, Japan) for SQV and a CAPCELL PAKMGII C18 column (2.1 © 50mm, 5µm; Shiseido Co., Ltd., Tokyo,Japan) for MDZ. 10mM Na2HPO4 in 35% acetonitrile was used asthe mobile phase for SQV and 10mM phosphate buffer (pH 6) in40% acetonitrile was used as the mobile phase for MDZ. The flowrate was 0.2mL/min at 40°C and the elution was monitored atwavelengths of 254 nm for SQV and 230 nm for MDZ.

Pharmacokinetic calculations:In vivo study

The peak plasma concentration (Cmax) was obtained directlyfrom the raw data. The area under the plasma concentration-timecurve from time 0 to the last time point (AUC0–t) was calculatedusing the linear trapezoidal rule. The AUCinf, with extrapolation toinfinity was calculated by dividing the last measured concentrationby the elimination rate constant, which was determined as the slopeof regression for the terminal log-linear portion of the concen-tration versus time curve. The bioavailability (F), hepatic availa-bility (Fh), fraction absorbed (Fa), and intestinal availability (Fg)were calculated using Eqs. (1)–(4). The blood-to-plasma ratio (Rb)and the hepatic blood flow (Qh) values used were 0.621) and 5.4L/h/kg,22) respectively. Hepatic clearance was regarded as the totalclearance because the urinary excretion was negligible.3)

F ¼ ðAUCpo � DoseivÞ=ðAUCiv � DosepoÞ ð1ÞCLtot ¼ CLh ¼ Doseiv=ðAUCiv � RbÞ ð2Þ

Fh ¼ 1� CLh=Qh ð3ÞFa � Fg ¼ F=Fh ð4Þ

In vitro studyThe initial reaction rate was determined under linear conditions.

The inhibition constant (Ki) was determined by plotting the slopeof the Dixon plot versus inhibitor concentration.

The increase in the AUC (R) caused by the drug-drug inter-actions was calculated with Eq. (5) for the intravenous admin-istration and with Eq. (6) for the oral administration.23)

R ¼ 1þ ðIin,u � FhÞ=Ki ð5ÞR ¼ 1þ Iin,u=Ki ð6Þ

Iin,u ¼ ðImax þ Ka � Fa � Fg � D=QhÞ � f bwhere Imax represents the maximum concentration of the inhibitorin the systemic blood. The first-order rate constant (Ka), thefraction absorbed from the gastrointestinal tract into the portalvein (Fa), and the unbound fraction in the blood (fb) used forcalculations are summarized in Table 2.Statistical analysis

Each value is expressed as a mean « standard deviation (SD).Statistical significance was examined by one-way analysis of vari-ance (ANOVA) followed by Dunnett’s post hoc multiple-com-

Table 1. LC-MS/MS conditions for each substrate

Compounds Column Mobile phase Ionization m/z

SQV CAPCELL PAK MGII C18(2.1 © 50mm, 5 µm)(Shiseido Co., Ltd., Tokyo, Japan)

10mM acetic acid in 40% acetonitrile Positive 671 to 570FEX 10mM acetic acid in 35% acetonitrile Positive 502 to 466MDZ 10mM acetic acid in 30% acetonitrile Positive 326 to 291

PRV Hypersil Gold (2.1 © 50mm, 5 µm)(Thermo Fisher Co., Ltd., Tokyo, Japan)

10mM acetic acid in 40% acetonitrile Negative 423 to 321

Atsuko TOMARU, et al.146

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

parison test. A p value of less than 0.05 was considered significant.

Results

Effect of RTV on the pharmacokinetics of SQV: RTVincreased the Cmax and AUCinf for SQV in a dose-dependentmanner. The AUCinf for SQV was increased three-, 26-, 241-,and 325-fold by the coadministration of 1.5, 5, 25, and 50mg/kgRTV, respectively. The plasma concentration-time profiles andpharmacokinetic parameters are shown in Figure 1 and Table 3,respectively.

To assess the effect of RTV on the first-pass effect in the liverand small intestine, Fa0Fg and Fh were calculated from the phar-macokinetic parameters after the intravenous and oral adminis-

tration of SQV with and without 50mg/kg RTV. A high dose ofRTV was administered and a significant interaction was observedin this study, however, the animals tolerated it well. The plasma-concentration time profiles and pharmacokinetic parameters forSQV are presented in Figure 2 and Table 4, respectively. Afterthe intravenous administration of SQV, the coadministration ofRTV caused only a fivefold increase in the AUCinf for SQV. Incontrast, after the oral administration of SQV, the AUCinf increaseddramatically (325-fold) with the coadministration of 50mg/kgRTV. As a result, the bioavailability of orally administered SQVincreased from 0.0093 to 0.675. Fh and Fa0Fg for SQV wereincreased 1.7- and 38-fold, respectively, with the coadministrationof RTV.

Effect of RTV on the pharmacokinetics of probe substrates:The effects of RTV on the pharmacokinetics of each probesubstrate were investigated. The plasma concentration-time profilesand pharmacokinetic parameters for each probe substrate areshown in Figures 3–5 and Tables 5–7, respectively. At the clini-cally relevant dose (1.5mg/kg), the AUCinf for FEX was notaffected with the coadministration of RTV. However, the AUCinf

for MDZ and PRV increased twofold when they were coadminis-tered with 1.5mg/kg RTV. The AUCinf for MDZ, FEX and PRV

Table 2. Parameters used in the prediction of the increase in the AUC fromKi values

Parameters Units RTV ITZ

Imax µM 24 237)

Ka min¹1 0.007 0.00437)

Fa0Fg — 0.8 1Qh L/h/kg 5.422) 5.422)

Rb — 0.739) 0.638)

fb — 0.02639) 0.00319)

Imax represents the maximum concentration of the inhibitor in the systemic blood(the Cmax for RTV was obtained from our study). The Ka and Fa0Fg for RTV wascalculated using the AUCinf in this study and bioavailbility.40) The maximumvalue was used as Fa0Fg for ITZ.

Fig. 1. Plasma SQV concentration-time profiles after oral administrationof SQV with various doses of RTVSQV (20mg/kg) was given orally after oral administration of RTV at a dose of1.5, 5, 25 and 50mg/kg. Each point represents the mean « SD (n = 3–4).

Table 3. Pharmacokinetic parameters for SQVafter oral administration ofSQV with or without oral administration of RTV in mice

SQV

RTV (mg/kg) Cmax (ng/mL) AUCinf (ng0h/mL)

Control 65.6 « 43.9 165.6 « 128.71.5 266.6 « 43.5 500.7 « 78.0

Fold increase 4 35 1,969.9 « 1,829.5 4,335.0 « 3,580.6

Fold increase 30 2625 6,921.5 « 2,691.7* 39,966.6 « 21,426.0*

Fold increase 106 24150 4,779.7 « 1,151.2* 53,850.1 « 12,758.9*

Fold increase 73 325

Values are mean « SD (n = 3–4). *p < 0.01.

Fig. 2. Plasma SQV concentration-time profiles after intravenous (A) ororal (B) administration of SQV with or without RTVSQV was given intravenously (2mg/kg) or orally (20mg/kg) after oraladministration of RTV at a dose of 50mg/kg. The sampling time for oraladministration without RTV was 2.5, 5, 15, 30min, 1, 2, 4, 6, 8, 10, 12 and 24 h.The sampling time for oral administration with RTV was 2.5, 5, 15, 30min, 1, 2,4, 8, 12, 18, 24, 36 and 48 h. For intravenous administration, the sampling timewas 1, 2.5, 5, 15, 30min, 1, 2, 4, 8, 12, 18 and 24 h. Each point represents themean « SD (n = 4–6).

Pharmacokinetic Boosting Effect of Ritonavir 147

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

Table 4. Pharmacokinetic parameters of SQV after intravenous and oral administration of SQV with or without oral administration of RTV (50mg/kg)in mice

OralCmax

(ng/mL)Tmax

(h)AUC0–t

(ng0h/mL)AUCinf

(ng0h/mL)

Control 65.6 « 43.9 0.6 « 0.7 153.4 « 133.1 165.6 « 128.7with RTV 4,779.7 « 1,151.2** 7.0 « 2.0** 53,839.0 « 12,757.3** 53,850.1 « 12,758.9**

IntravenousAUC0–24

(ng0h/mL)AUCinf

(ng0h/mL)CLtot

(L/hr/kg)

Control 1,516.7 « 608.0 1,525.4 « 606.7 2.46 « 0.93with RTV 7,685.0 « 3,127.7* 7,882.0 « 3,180.2* 0.47 « 0.16*

Values are mean « SD (n = 4–6). *p < 0.05, **p < 0.01.

Fig. 3. Plasma MDZ concentration-time profiles after oral administrationof MDZ with or without RTVMDZ (10mg/kg) was given orally after oral administration of RTV at a dose of1.5 and 50mg/kg. Blood samples were obtained at 5, 15, 30min, 1, 2, 4, 6, 8, 10(RTV 50mg/kg only), 12, 18 and 24 h. Each point represents the mean « SD(n = 3).

Table 5. Pharmacokinetic parameters of MDZ after oral administration ofMDZ with or without oral administration of RTV in mice

MDZ

RTV(mg/kg)

Cmax

(ng/mL)AUCinf

(ng0h/mL)

Control 226.7 « 100.6 245.2 « 88.31.5 371.3 « 252.1 463.4 « 296.3

Fold increase 2 250 427.6 « 61.8 1,220.3 « 117.9*

Fold increase 2 5

Values are mean « SD (n = 3). *p < 0.01.

Fig. 4. Plasma FEX concentration-time profiles after oral administrationof FEX with or without RTVFEX (10mg/kg) was given orally after oral administration of RTV at a dose of1.5 and 50mg/kg. Blood samples were obtained at 5, 15, 30min, 1, 2, 4, 6, 8, 10(RTV 50mg/kg only), 12, 18 and 24 h. Each point represents the mean « SD(n = 3).

Fig. 5. Plasma PRV concentration-time profiles after oral administrationof PRV with or without RTVPRV (100mg/kg) was given orally after oral administration of RTV at a dose of1.5 and 50mg/kg. Blood samples were obtained at 5, 15, 30min, 1, 2, 4, 6, 8,12, 18 and 24 h. Each point represents the mean « SD (n = 3).

Table 6. Pharmacokinetic parameters of FEX after oral administration ofFEX with or without oral administration of RTV in mice

FEX

RTV(mg/kg)

Cmax

(ng/mL)AUCinf

(ng0h/mL)

Control 85.6 « 28.9 542.3 « 211.71.5 63.5 « 14.6 398.4 « 84.6

Fold increase 1 150 939.8 « 42.9* 7,169.2 « 586.9*

Fold increase 11 13

Values are mean « SD (n = 3). *p < 0.01.

Atsuko TOMARU, et al.148

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

increased five-, 13- and sevenfold, respectively, when they werecoadministered with 50mg/kg RTV.

Effect of ITZ on the pharmacokinetics of SQV: Theinhibitory effect of 50mg/kg ITZ on the pharmacokinetics ofSQV after its oral administration was investigated. The plasmaconcentration-time profile and pharmacokinetic parameters areshown in Figure 6 and Table 8, respectively. There was nosignificant difference in the Cmax or AUCinf after treatment with orwithout 50mg/kg ITZ and this result indicates that cyp isoformsother than cyp3a are involved in the metabolism of SQV in mice.

In vitro study: Before performing the inhibition study, theapparent Michaelis constant (Km) of SQV in the mouse liver wascalculated and the concentration of substrate in the inhibition studywas set below the Km value (data not shown). The same concen-tration was used in the inhibition study with mouse intestinalmicrosomes. Figures 7 and 8 show the Dixon plots for the inhi-bition by RTVof SQVor MDZ metabolism, respectively, in mouse

liver and intestinal microsomes. The SQVand MDZ concentrationsused were 0.25, 0.5 and 1µM. The inhibition by RTV or ITZ wasconsidered to be competitive or non-competitive. The Ki values forthe inhibition of SQV metabolism by RTV were 95 and 75 nM inthe liver and intestinal microsomes, respectively (Table 9). The Ki

value for the inhibition of MDZ metabolism was 65 nM in boththe liver and intestinal microsomes. The inhibitory effect of ITZon the metabolism of SQV was also investigated (Fig. 9). The Ki

values were 3.7 and 2.6 µM in the liver and intestinal microsomes,respectively.

The fold increases in the AUC predicted from the in vitro studyand their comparison with the values observed in the in vivo studyare shown in Table 10. A fourfold increase in the AUC for SQVwas predicted from the Ki value using Eq. (5), which is close tothe value observed after the intravenous administration of SQV.In contrast, a ninefold increase in the AUC for SQV was predictedfrom the Ki value using Eq. (6). The remaining increase in theAUC for SQV derived from the inhibition of metabolic enzymesin the small intestine was estimated about 36-fold. This resultindicates that the effect of RTV was greater in the small intestinethan in the liver, which was consistent with the results obtained inthe in vivo study.

Discussion

In this study, in vivo and in vitro experiments were performedin mice to clarify the underlying mechanisms of RTV boosting.Escalating doses of RTV had dose-dependent effects on the phar-macokinetics of SQV. There was a threefold increase in the AUCinf

for SQV when administered with 1.5mg/kg RTV, the clinically

Fig. 6. Plasma SQV concentration-time profiles after oral administrationof SQV with or without ITZSQV (20mg/kg) was given orally after oral administration of ITZ at a dose of50mg/kg. Blood samples were obtained at 5, 15, 30min, 1, 2, 4, 6, 8 and 12 h.Each point represents the mean « SD (n = 3).

Table 8. Pharmacokinetic parameters of SQV after oral administration ofSQV with or without oral administration of ITZ in mice

SQV

ITZ(mg/kg)

Cmax

(ng/mL)AUC0–12

(ng0h/mL)

Control 50.6 « 33.2 86.7 « 65.850 70.0 « 19.9 138.6 « 55.0

Fold increase 1.4 1.6

Values are mean « SD (n = 3).There were no significant differences in these parameters between the controland coadministration of 50mg/kg ITZ.

Table 7. Pharmacokinetic parameters of PRV after oral administration ofPRV with or without oral administration of RTV in mice

PRV

RTV(mg/kg)

Cmax

(ng/mL)AUCinf

(ng0h/mL)

Control 737.3 « 557.1 636.3 « 97.11.5 534.9 « 58.1 1,045.7 « 164.2

Fold increase 1 250 2,433.7 « 1,532.3 4,604.1 « 1,736.3*

Fold increase 3 7

Values are mean « SD (n = 3). *p < 0.01.

Fig. 7. Dixon plots for the inhibition of SQV metabolism by RTV in mouseliver (A) and intestinal microsomes (B)SQV (0.25, 0.5, 1 µM) was incubated at 37°C for 10min with liver microsomes(0.2mg/mL protein) (A) and 20min with intestinal microsomes (0.2mg/mLprotein) (B). The data represent the mean of n = 3.

Pharmacokinetic Boosting Effect of Ritonavir 149

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

relevant dose. In the clinical dose, it is reported that the increasein the AUC for SQV at SQV/RTV doses of 1,200/100mg is six-to sevenfold.24,25) Therefore, our results indicate the effect of RTVboosting is similar in humans and in mice.

The contribution of 50mg/kg RTV boosting to the first-passeffect on SQV in the small intestine and liver was accessed. As aresult, Fa0Fg increased 38-fold, whereas Fh increased only twofold.These results suggest that RTV mainly influences the first-passeffect in the small intestine, resulting in an increase in the bio-availability of orally administered SQV.

Factors affecting the first-pass effect in the liver are metabolicenzymes and drug transporters. Among the metabolic enzymes inthe liver, CYP3A is particularly important because it is responsiblefor the majority of phase I drug metabolism reaction, including thatof PIs. CYP3A is also an important factor in the first-pass effect inthe small intestine. In addition to the metabolic enzymes, the effluxtransporters localized in the apical membrane are also importantfactors in the small intestine. Therefore, the inhibition of P-gp/

CYP3A in the small intestine and the inhibition of CYP3A in theliver are considered to increase the oral bioavailability of SQV.However, regarding the probe substrate of CYP3A, the increase inthe AUCinf for MDZ after its oral administration with 50mg/kgRTV was only fivefold. In addition, the increase in the AUC0–12

for SQV after its oral administration with ITZ was only twofold,and did not differ significantly from the control. Yamano et al.evaluated the extent of drug-drug interactions involving metabolicinhibition in the rat liver.19) In their report, the increase in the AUCfor MDZ in the presence of ITZ was about twofold, which isconsistent with our result of the inhibition study for ITZ.

Table 10. Comparison of the fold increase in the AUC between observedvalue and predicted value

Substrates SQV MDZ SQVInhibitors RTV RTV ITZ

Oral

Fold increase in AUC(Observed)

325 5 2

Fold increase in AUC(Predicted)

9 12 1

Intravenous

Fold increase in AUC(Observed)

5 — —

Fold increase in AUC(Predicted)

4 — —

The dosage of each substrate and inhibitor are as follows. SQV: 2mg/kg forintravenous administration and 20mg/kg for oral administration, MDZ: 10mg/kg, RTV and ITZ: 50mg/kg.

Table 9. The kinetic parameters of SQV and MDZ inhibition by RTV andITZ in mouse liver and intestinal microsomes

Substrates SQV SQV MDZInhibitors RTV ITZ RTV

Liver microsomes 95 nM 3.7 µM 65 nMIntestinal microsomes 75 nM 2.6 µM 65 nM

Values are mean of n = 3.

Fig. 8. Dixon plots for the inhibition of MDZ metabolism by RTV in mouseliver (A) and intestinal (B) microsomesMDZ (0.25, 0.5, 1 µM) was incubated at 37°C for 5min with liver microsomes(0.1mg/mL protein) (A) and 15min with intestinal microsomes (0.2mg/mLprotein) (B). The data represent the mean of n = 3.

Fig. 9. Dixon plots for the inhibition of SQV metabolism by ITZ in mouseliver (A) and intestinal (B) microsomesSQV (0.25, 0.5, 1 µM) was incubated at 37°C for 10min with liver microsomes(0.2mg/mL protein) (A) and 20min with intestinal microsomes (0.2mg/mLprotein) (B). The data represent the mean of n = 3.

Atsuko TOMARU, et al.150

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

MDZ is predominantly biotransformed to the 1A-hydroxy MDZrather than to 4-hydroxy MDZ in both humans and mice.26) Inhumans, CYP3A mediates its biotransformation to both 1A-hydroxyand 4-hydroxy MDZ. On the other hand, there is a report thatcyp3a mediates only its biotransformation to 4-hydroxy MDZ inmice, and cyp isoforms other than cyp3a mediate its biotransfor-mation to 1A-hydroxy MDZ.27) Furthermore, van Waterschoot et al.even reported that MDZ metabolism was observed in liver micro-somes prepared from cyp3a¹/¹ mice.28) These data indicate aspecies difference in the MDZ metabolism in mice and humansand suggest that the cyp isoforms involved in the metabolism ofSQV in mice differ from those involved in humans.

Perloff et al. reported that the Ki values for the inhibition byketoconazole of 1A-hydroxy and 4-hydroxy MDZ biotransformationwere 1.7 and 0.066 µM, respectively, in mouse liver microsomes.In humans, these Ki values were 0.0054 and 0.039 µM, respec-tively.27) These results show the Ki value for the inhibition byketoconazole of 1A-hydroxy MDZ biotransformation in micediffered greatly from that in humans. Moreover, the Ki value forthe inhibition by ketoconazole of 1A-hydroxy MDZ biotransforma-tion was the same as our result for the inhibition by ITZ of SQVmetabolism. In addition, Yamano et al. determined the Ki valuefor the inhibition of 1A-hydroxy biotransformation by ketoconazoleand ITZ, and they found those were the same.19) In our study, theinhibitory effect of RTV on the metabolism of SQV and MDZwas the same. Therefore, RTV and ITZ inhibit the cyp isoformsinvolved in the metabolism of SQV and MDZ, and these cypisoforms may be the same or have the same affinity to RTVor ITZ.Komura and Iwaki reported that the cyp mRNA expression levelsdiffer in the livers and small intestines of mice, and that cyp3a13 ispredominantly expressed in the small intestine and is responsiblefor the first-pass metabolism of CYP3A substrates in the mousesmall intestine.26) They also evaluated the relationships of the Km

values for the CYP3A4 mediated biotransformations of 13 drugsin intestinal and liver microsomes, and a good relationship wasobserved, although the mRNA expression levels differed. Thisobservation supports our finding that the Ki values were the samefor the liver and intestinal microsomes in the in vitro study.

When we considered a substrate of P-gp, there was nosignificant difference in the Cmax or AUCinf for FEX when it wascoadministered with 1.5mg/kg of RTV. In humans, a twofoldincrease in the AUCinf for FEX was observed when given with100mg of RTV,13) so there are also species differences in theeffects of RTV on the pharmacokinetics of FEX. The oral bio-availability, Fa0Fg and Fh of FEX in humans has been reported as0.28, 0.31 and 0.9,29) respectively. Therefore, it is estimated that themaximum increase in the AUC for FEX is about threefold whenthe inhibition by RTV is completed. In our study, the increase inthe AUCinf for FEX in mice coadministered with 50mg/kg RTVwas 13-fold, which was higher than that estimated for humans. Inaddition, it has been reported the oral bioavailability, Fa0Fg and Fh

in rat were 0.014, 0.022 and 0.641,30) respectively. The maximumincrease in the AUC for FEX is estimated to be about 50-fold, withcomplete inhibition by RTV. Therefore, even 50mg/kg RTV maynot inhibit the first-pass effect in the small intestine and livercompletely in mice. Tahara et al. reported that the increase in theAUCinf for FEX in Mdr1a/1b P-gp knockout mice was aboutsixfold.31) Considering our result and the very small Fa0Fg values ofFEX in mice, efflux transporters other than P-gp may be involvedin the intestinal absorption of FEX in mice. Bcrp and Mrp2 are the

efflux transporters expressed at the apical membrane in the mousesmall intestine. It has been reported that FEX is not a substrateof BCRP in humans.32) Furthermore, Mrp2/3 are involved in thehepatic disposition of FEX in mice32,33) and MRP2 mediates theabsorption of FEX in humans.34,35) These reports suggest that effluxtransporters other than P-gp, such as Mrp2, might be involved inthe absorption of FEX in mice and that the differences in the RTV-boosting effects on these transporters may result in the speciesdifferences observed in the pharmacokinetics of FEX.

An additional experiment was performed using mouse liver andintestinal microsomes to confirm the result obtained in the in vivostudy. The fourfold increase in the AUC for SQV was predictedfrom the in vitro Ki value for intravenously administered SQV,which is consistent with the results of the in vivo study. This resultindicates that the inhibitory effect of RTV in the liver is mainlyattributable to the inhibition of metabolic enzymes with a minorcontribution on the inhibition of influx transporters in the liver. Onthe other hand, a significant increase in the AUCinf for PRV wasobserved after the oral administration of 50mg/kg RTV in thein vivo study. According to the package information for Karetraμ,which is a PI formed by the combination of lopinavir and RTV,there are no reports that Karetraμ enhances the plasma concen-tration of PRV. However, it has been reported that the AUC forrosuvastatin, a substrate of OATP1B1, is increased twofold by itscoadministration with Karetraμ.36) In this report, they concludedthis interaction might be mediated by the inhibition by lopinavirand/or ritonavir of rosuvastatin uptake at the level of absorption byBCRP or at the level of uptake into the hepatocytes by OATP1B1,by both, or by neither.

After the oral administration of SQV together with RTV, aninefold increase in the AUC for SQV was predicted from thein vitro Ki value, and this is derived from the inhibition ofmetabolic enzymes in the liver. From this result, the remaining 36-fold increase in the AUC was considered to be derived from theinhibition of metabolic enzymes in the small intestine. This resultcorresponds to the result obtained in the in vivo study, and theinhibitory effect of RTV boosting is shown to be higher in thesmall intestine than in the liver.

In conclusion, RTV mainly affects the first-pass effect in thesmall intestine, increasing the bioavailability of orally administeredSQV. The effects of RTV boosting on the pharmacokinetics ofSQV are the same in humans and in mice. However, the cypisoforms involved in the metabolism of SQV in mice differ fromthose in humans.

References

1) Kempf, D. J., Marsh, K. C., Kumar, G., Rodrigues, A. D., Denissen, J. F.,Mcdonald, E., Kukulka, M. J., Hsu, A., Granneman, G. R., Baroldi, P. A.,Sun, E., Pizzuti, D., Plattner, J. J., Norbeck, D. W. and Leonard, J. M.:Pharmacokinetic enhancement of inhibitors of the human immunodefi-ciency virus protease by coadministration with ritonavir. Antimicrob.Agents Chemother., 41: 654–660 (1997).

2) Kumar, G. N., Rodrigues, A. D., Buko, A. M. and Denissen, J. F.:Cytochrome P450-mediated metabolism of the HIV-1 protease inhibitorritonavir (ABT-538) in human liver microsomes. J. Pharmacol. Exp.Ther., 277: 423–431 (1996).

3) Lin, J. H.: Human immunodeficiency virus protease inhibitors from drugdesign to clinical studies. Adv. Drug Deliv. Rev., 27: 215–233 (1997).

4) Youle, M.: Overview of boosted protease inhibitors in treatment-experienced HIV-infected patients. J. Antimicrob. Chemother., 60:1195–1205 (2007).

5) Choo, E. F., Leake, B., Wandel, C., Imamura, H., Wood, A. J. J.,Wilkinson, G. R. and Kim, R. B.: Pharmacological inhibition of P-glycoprotein transport enhances the distribution of HIV-1 protease

Pharmacokinetic Boosting Effect of Ritonavir 151

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)

inhibitors into brain and testes. Drug Metab. Dispos., 28: 655–660 (2000).6) Kumar, S., Kwei, G. Y., Poon, G. K., Iliff, S. A., Wang, Y., Chen, Q.,

Franklin, R. B., Didolkar, V., Wang, R. W., Yamazaki, M., Chiu, S. L.,Lin, J. H., Pearson, P. G. and Baillie, T. A.: Pharmacokinetics andinteractions of a novel antagonist of chemokine receptor 5 (CCR5) withritonavir in rats and monkeys: role of CYP3A and P-glycoprotein.J. Pharmacol. Exp. Ther., 304: 1161–1171 (2003).

7) Perloff, M. D., von Moltke, L. L. and Greenblatt, D. J.: Fexofenadinetransport in Caco-2 cells: inhibition with verapamil and ritonavir. J. Clin.Pharmacol., 42: 1269–1274 (2002).

8) Gupta, A., Zhang, Y., Unadkat, J. D. and Mao, Q.: HIV proteaseinhibitors are inhibitors but not substrates of the human breast cancerresistance protein (BCRP/ABCG2). J. Pharmacol. Exp. Ther., 310: 334–341 (2004).

9) Ye, Z. W., Camus, S., Augustijns, P. and Annaert, P.: Interaction of eightHIV protease inhibitors with the canalicular efflux transporter ABCC2(MRP2) in sandwich-cultured rat and human hepatocytes. Biopharm.Drug Dispos., 31: 178–188 (2010).

10) Annaert, P., Ye, Z. W., Stieger, B. and Augustijns, P.: Interaction of HIVprotease inhibitors with OATP1B1, 1B3, and 2B1. Xenobiotica, 40: 163–176 (2010).

11) Chan, L. M. S., Lowes, S. and Hirst, B. H.: The ABCs of drug transportin intestine and liver: efflux proteins limiting drug absorption andbioavailability. Eur. J. Pharm. Sci., 21: 25–51 (2004).

12) Kim, A. E., Dintaman, J. M., Waddell, D. S. and Silverman, J. A.:Saquinavir, an HIV protease inhibitor, is transported by P-glycoprotein.J. Pharmacol. Exp. Ther., 286: 1439–1445 (1998).

13) Kharasch, E. D., Bedynek, P. S., Walker, A., Whittington, D. and Hoffer,C.: Mechanism of ritonavir changes in methadone pharmacokinetics andpharmacodynamics: II. Ritonavir effects on CYP3A and P-glycoproteinactivities. Clin. Pharmacol. Ther., 84: 506–512 (2008).

14) Huisman, M. T., Smit, J. W., Wiltshire, H. R., Hoetelmans, R. M. W.,Beijnen, J. H. and Schinkel, A. H.: P-glycoprotein limits oral availability,brain, and fetal penetration of saquinavir even with high doses ofritonavir. Mol. Pharmacol., 59: 806–813 (2001).

15) Washington, C. B., Wiltshire, H. R., Man, M., Moy, T., Harris, S. R.,Worth, E., Weigl, P., Liang, Z., Hall, D., Marriott, L. and Blaschke, T. F.:The disposition of saquinavir in normal and P-glycoprotein deficientmice, rats, and in cultured cells. Drug Metab. Dispos., 28: 1058–1062(2000).

16) Richter, M., Gyémánt, N., Molnár, J. and Hilgeroth, A.: Comparativeeffects on intestinal absorption in situ by P-glycoprotein-modifying HIVprotease inhibitors. Pharm. Res., 21: 1862–1866 (2004).

17) Hartkoorn, R. C., Kwan, W. S., Shallcross, V., Chaikan, A., Liptrott, N.,Egan, D., Sora, E. S., James, C. E., Gibbons, S., Bray, P. G., Back, D. J.,Khoo, S. H. and Owen, A.: HIV protease inhibitors are substrates forOATP1A2, OATP1B1 and OATP1B3 and lopinavir plasma concen-trations are influenced by SLCO1B1 polymorphisms. Pharmacogenet.Genomics, 20: 112–120 (2010).

18) Shitara, Y., Hirano, M., Sato, H. and Sugiyama, Y.: Gemfibrozil and itsglucuronide inhibit the organic anion transporting polypeptide 2 (OATP2/OATP1B1:SLC21A6)-mediated hepatic uptake and CYP2C8-mediatedmetabolism of cerivastatin: analysis of the mechanism of the clinicallyrelevant drug-drug interaction between cerivastatin and gemfibrozil.J. Pharmacol. Exp. Ther., 311: 228–236 (2004).

19) Yamano, K., Yamamoto, K., Kotaki, H., Sawada, Y. and Iga, T.:Quantitative prediction of metabolic inhibition of midazolam byitraconazole and ketoconazole in rats: implication of concentrativeuptake of inhibitors into liver. Drug Metab. Dispos., 27: 395–402 (1999).

20) Emoto, C., Yamazaki, H., Yamasaki, S., Shimada, N., Nakajima, M. andYokoi, T.: Characterization of cytochrome P450 enzymes involved indrug oxidations in mouse intestinal microsomes. Xenobiotica, 30: 943–953 (2000).

21) Interview form of Saquinavir. http://chugai-pharm.jp/hc/ss/pr/drug/inv_cap0200/if/index.html. Accessed 22 July (2012).

22) Davies, B. and Morris, T.: Physiological parameters in laboratory animals

and humans. Pharm. Res., 10: 1093–1095 (1993).23) Ito, K., Iwatsubo, T., Kanamitsu, S., Ueda, K., Suzuki, H. and Sugiyama,

Y.: Prediction of pharmacokinetic alterations caused by drug-druginteractions: metabolic interaction in the liver. Pharmacol. Rev., 50:387–412 (1998).

24) Kilby, J. M., Sfakianos, G., Gizzi, N., Siemon-Hryczyk, P., Ehrensing, E.,Oo, C., Buss, N. and Saag, M. S.: Safety and pharmacokinetics of once-daily regimens of soft-gel capsule saquinavir plus minidose ritonavir inhuman immunodeficiency virus-negative adults. Antimicrob. AgentsChemother., 44: 2672–2678 (2000).

25) Plosker, G. L. and Scott, L. J.: Saquinavir. A review of its use in boostedregimens for treating HIV infection. Drugs, 63: 1299–1324 (2003).

26) Komura, H. and Iwaki, M.: Species differences in in vitro and in vivosmall intestinal metabolism of CYP3A substrates. J. Pharm. Sci., 97:1775–1800 (2008).

27) Perloff, M. D., von Moltke, L. L., Court, M. H., Kotegawa, T., Shader,R. I. and Greenblatt, D. J.: Midazolam and triazolam biotransformation inmouse and human liver microsomes: relative contribution of CYP3A andCYP2C isoforms. J. Pharmacol. Exp. Ther., 292: 618–628 (2000).

28) van Waterschoot, R. A. B., van Herwaarden, A. E., Lagas, J. S.,Sparidans, R. W., Wagenaar, E., van der Kruijssen, C. M. M., Goldstein,J. A., Zeldin, D. C., Beijnen, J. H. and Schinkel, A. H.: Midazolammetabolism in cytochrome P450 3A knockout mice can be attributed toup-regulated CYP2C enzymes. Mol. Pharmacol., 73: 1029–1036 (2008).

29) Tachibana, T., Kato, M., Watanabe, T., Mitsui, T. and Sugiyama, Y.:Method for predicting the risk of drug-drug interactions involvinginhibition of intestinal CYP3A4 and P-glycoprotein. Xenobiotica, 39:430–443 (2009).

30) Gotoh, Y., Kamada, N. and Momose, D.: The advantages of the ussingchamber in drug absorption studies. J. Biomol. Screen., 10: 517–523(2005).

31) Tahara, H., Kusuhara, H., Fuse, E. and Sugiyama, Y.: P-glycoproteinplays a major role in the efflux of fexofenadine in the small intestine andblood-brain barrier, but only a limited role in its biliary excretion. DrugMetab. Dispos., 33: 963–968 (2005).

32) Matsushima, S., Maeda, K., Hayashi, H., Debori, Y., Schinkel, A. H.,Schuetz, J. D., Kusuhara, H. and Sugiyama, Y.: Involvement of multipleefflux transporters in hepatic disposition of fexofenadine. Mol. Pharma-col., 73: 1474–1483 (2008).

33) Tian, X., Zamek-Gliszczynski, M. J., Li, J., Bridges, A. S., Nezasa, K.,Patel, N. J., Raub, T. J. and Brouwer, K. L. R.: Multidrug resistance-associated protein 2 is primarily responsible for the biliary excretion offexofenadine in mice. Drug Metab. Dispos., 36: 61–64 (2008).

34) Ming, X., Knight, B. M. and Thakker, D. R.: Vectorial transport offexofenadine across Caco-2 cells: involvement of apical uptake andbasolateral efflux transporters. Mol. Pharm., 8: 1677–1686 (2011).

35) Akamine, Y., Miura, M., Sunagawa, S., Kagaya, H., Yasui-Furukori,N. and Uno, T.: Influence of drug-transporter polymorphisms on thepharmacokinetics of fexofenadine enantiomers. Xenobiotica, 40: 782–789 (2010).

36) Kiser, J. J., Gerber, J. G., Predhomme, J. A., Wolfe, P., Flynn, D. M. andHoody, D. W.: Drug/drug interaction between lopinavir/ritonavir androsuvastatin in healthy volunteers. J. Acquir. Immune Defic. Syndr., 47:570–578 (2008).

37) Yoo, S. D., Kang, E., Shin, B. S., Jun, H., Lee, S., Lee, K. C. and Lee, K.:Interspecies comparison of the oral absorption of itraconazole inlaboratory animals. Arch. Pharm. Res., 25: 387–391 (2002).

38) Product monograph of Itraconazole. http://www.janssen.ca/product/188.Accessed 22 July (2012).

39) Denissen, J. F., Grabowski, B. A., Johnson, M. K., Buko, A. M., Kempf,D. J., Thomas, S. B. and Surber, B. W.: Metabolism and disposition of theHIV-1 protease inhibitor ritonavir (ABT-538) in rats, dogs, and humans.Drug Metab. Dispos., 25: 489–501 (1997).

40) Lledó-García, R., Nácher, A., Prats-García, L., Casabó, V. G. and Merino-sanjuán, M.: Bioavailability and pharmacokinetic model for ritonavir inthe rat. J. Pharm. Sci., 96: 633–643 (2007).

Atsuko TOMARU, et al.152

Copyright © 2013 by the Japanese Society for the Study of Xenobiotics (JSSX)