Published Ahead of Print 28 January 2013. 10.1128/AAC.02282-12.

2013, 57(4):1743. DOI:Antimicrob. Agents Chemother. and Xiaoyan ChenPan Deng, Dafang Zhong, Kate Yu, Yifan Zhang, Ting Wang HumansExcretion of the Antiviral Drug Arbidol in Pharmacokinetics, Metabolism, and

http://aac.asm.org/content/57/4/1743Updated information and services can be found at:

These include:

REFERENCEShttp://aac.asm.org/content/57/4/1743#ref-list-1This article cites 20 articles, 5 of which can be accessed free at:

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

Pharmacokinetics, Metabolism, and Excretion of the Antiviral DrugArbidol in Humans

Pan Deng,a Dafang Zhong,a Kate Yu,b Yifan Zhang,a Ting Wang,c Xiaoyan Chena

Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai, Chinaa; Waters Corporation, Milford, Massachusetts, USAb; First Affiliated Hospital ofLanzhou University, Lanzhou, Chinac

Arbidol is a broad-spectrum antiviral drug that is used clinically to treat influenza. In this study, the pharmacokinetics, metabo-lism, and excretion of arbidol were investigated in healthy male Chinese volunteers after a single oral administration of 200 mgof arbidol hydrochloride. A total of 33 arbidol metabolites were identified in human plasma, urine, and feces. The principal bio-transformation pathways included sulfoxidation, dimethylamine N-demethylation, glucuronidation, and sulfate conjugation.The major drug-related component in the plasma was sulfinylarbidol (M6-1), followed by unmetabolized arbidol, N-demethyl-sulfinylarbidol (M5), and sulfonylarbidol (M8). The exposures of M5, M6-1, and M8, as determined by the metabolite-to-parentarea under the plasma concentration-time curve from 0 to t (AUC0-t) ratio, were 0.9 � 0.3, 11.5 � 3.6, and 0.5 � 0.2, respectively.In human urine, glucuronide and sulfate conjugates were detected as the major metabolites, accounting for 6.3% of the dose ex-creted within 0 to 96 h after drug administration. The fecal specimens mainly contained the unchanged arbidol, accounting for32.4% of the dose. Microsomal incubation experiments demonstrated that the liver and intestines were the major organs thatmetabolize arbidol in humans. CYP3A4 was the major isoform involved in arbidol metabolism, whereas the other P450s andflavin-containing monooxygenases (FMOs) played minor roles. These results indicated possible drug interactions between arbi-dol and CYP3A4 inhibitors and inducers. Further investigations are needed to understand the importance of M6-1 in the efficacyand safety of arbidol, because of its high plasma exposure and long elimination half-life (25.0 h).

Arbidol {ethyl-6-bromo-4-[(dimethylamino)methyl]-5-hydroxy-1-methyl-2-[(phenylthio)methyl]-indole-3-

carboxylate} is a broad-spectrum antiviral compound. It was firstmarketed in 1993 for prophylaxis and treatment of influenza virusA and B infections (1). Recently, Teissier et al. reported that arbi-dol could inhibit viral glycoprotein conformational changes dur-ing membrane fusion by interacting with the phospholipid mem-brane and protein motifs enriched in aromatic residues (2).Clinical trials indicated that 200 mg of arbidol taken three timesdaily for 5 to 10 days reduces the duration of influenza by 1.7 to2.65 days (3). Recent studies have extended the inhibitory activityof arbidol to other human viruses, such as hepatitis B and C vi-ruses, rhinovirus 14, bird viruses, chikungunya virus, and respira-tory syncytial virus (4, 5).

Circulating metabolites have been known to contribute to oralter the pharmacological activities of the parent drug. The safetyof circulating metabolites should be considered. Furthermore,identifying drug metabolic pathways is also important for predict-ing drug-drug interactions (DDIs). However, the current under-standing of arbidol metabolism in humans is incomplete, and onlya single study has been reported on the identification of humanurinary metabolites after oral drug administration. Glucuronidearbidol (M18) and glucuronide sulfinylarbidol (M20-1 andM20-2) were detected as the major metabolites in human urine(6). The limited pharmacokinetics of arbidol in healthy humanvolunteers has been described, which showed its rapid absorption(time to maximum concentration of drug in plasma [Tmax], 1.6 to1.8 h) and slow elimination (half-life [t1/2], 16 to 21 h) (1, 7, 8). Nocirculating arbidol metabolites have been detected or character-ized thus far.

The objectives of this study were to investigate the metabolicprofile, the routes of excretion, and the pharmacokinetics of arbi-dol in healthy male volunteers after a single oral dose of 200 mg of

arbidol hydrochloride. The metabolites in the circulation and ex-creta were identified using ultra-performance liquid chromatog-raphy– quadrupole time-of-flight mass spectrometry (UPLC–Q-TOF MS), and the major metabolites were synthesized forstructure confirmation. The pharmacokinetic profiles of arbidoland its primary metabolites were characterized. The enzymes re-sponsible for the principal metabolic pathways were identifiedusing recombinant P450s and flavin-containing monooxygenases(FMOs), human liver microsomes (HLMs), human intestine mi-crosomes (HIMs), and human kidney microsomes (HKMs) withand without P450 inhibitors and heat inactivation of FMOs.

MATERIALS AND METHODSMaterials. Arbidol hydrochloride capsules (100 mg/capsule, equivalent to89 mg base/capsule) were purchased from Shijiazhuang No. 4 Pharma-ceutical Co., Ltd. (Hebei, China). Arbidol hydrochloride was obtainedfrom the National Institute for the Control of Pharmaceutical and Bio-logical Products (Beijing, China). Arbidol derivatives, including oxidativeS-dealkylated arbidol, N-demethylsulfinylarbidol (M5), sulfinylarbidol(M6-1), 4=-hydroxylated arbidol, N-demethylsulfonylarbidol (M7), andsulfonylarbidol (M8), were synthesized by Hebei University of Scienceand Technology (Hebei, China) and served as synthetic standards to con-firm the metabolites’ structures. These compounds were �99% pure.Pooled HLMs and HKMs, recombinant P450 enzymes (CYP1A1,CYP1A2, CYP1B1, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19,

Received 13 November 2012 Returned for modification 21 December 2012Accepted 20 January 2013

Published ahead of print 28 January 2013

Address correspondence to Xiaoyan Chen, xychen@mail.shcnc.ac.cn.

Copyright © 2013, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AAC.02282-12

April 2013 Volume 57 Number 4 Antimicrobial Agents and Chemotherapy p. 1743–1755 aac.asm.org 1743

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

CYP2D6, CYP2E1, CYP3A4, CYP3A5, and CYP4A11), and FMO iso-forms (FMO1, FMO3, and FMO5) were purchased from BD Gentest(Woburn, MA). Pooled HIMs were purchased from Xenotech LLC (Le-nexa, KS). The following chemicals were purchased from Sigma-Aldrich(St. Louis, MO): �-glucuronidase from Helix pomatia (type H-1), 1-ami-nobenzotriazole (1-ABT), benzydamine, NADPH, �-naphthoflavone,ticlopidine, quinidine, clomethiazole, ketoconazole, and all solvents usedfor liquid chromatography-mass spectrometry (LC-MS) analysis. Puri-fied water was generated using a Milli-Q Gradient system (Molsheim,France).

Study design, dosing, and sample collection. This was an open-label,nonrandomized, single-dose study. Four male subjects with a mean age of24 years (range, 22 to 26 years) and in good physical health, as shown bymedical examination and history, vital signs, 12-lead electrocardiogram,and laboratory tests, were recruited. The body mass index range was 21 to23 kg/m2. The study protocol was approved by the Ethics Committee ofthe First Affiliated Hospital of Lanzhou University (Lanzhou, China).Written informed consent was obtained from all subjects before enroll-ment. The subjects received a single oral dose of 200-mg arbidol hydro-chloride capsules. Blood samples (4.5 ml) were collected into heparinizedtubes predose and at 0.125, 0.25, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0, 12, 24,36, 48, and 72 h postdose. Plasma was harvested by centrifugation andstored at �20°C until analysis. Urine samples were collected predose andat 0 to 12 h, 12 to 24 h, 24 to 48 h, 48 to 72 h, and 72 to 96 h postdose. Fecalsamples were collected predose and up to 96 h postdose. Each portion wasdiluted with 5 volumes of methanol and homogenized. The urine andhomogenized feces were stored at �20°C until analysis. The subjects wereprovided with standard meals at approximately 4 and 10 h after drugdosing.

Metabolite profiling. (i) Sample preparation and �-glucuronidasehydrolysis. Representative pooled samples were prepared for metabolite-profiling experiments. The plasma samples were segregated by samplingtime, and equal volumes of plasma samples from all subjects were pooled.The urine samples and fecal homogenates from all subjects were pooled bycombining volumes proportional to the total volume or weight excretedby each subject for each collection interval. To a 50-�l aliquot of pooledplasma, urine, and fecal-homogenate samples was added 200 �l of meth-anol. After being vortex mixed and centrifuged at 11,000 � g for 5 min, thesupernatant was transferred into a glass tube, evaporated to dryness undera stream of nitrogen at 40°C, and then reconstituted in 100 �l of methanoland 5 mM ammonium acetate (1:1 [vol/vol]). A 10-�l aliquot of thereconstituted solution was injected onto a UPLC–Q-TOF MS for analysis.For enzymatic incubation, a 50-�l aliquot of the urine sample was mixedwith 50 �l of �-glucuronidase (in 1 M citrate buffer solution at pH 5.0).The mixture was incubated at 37°C for 16 h. The effect of the glucuroni-dase was studied by comparing the LC-MS peak intensities for com-pounds of interest before and after enzymatic incubation. The com-pounds of interest included glucuronide conjugates and their hydrolyzedforms.

(ii) UPLC–Q-TOF MS analysis. Chromatographic separation for me-tabolite profiling was achieved using an Acquity UPLC system (WatersCorp., Milford, MA) on an Acquity UPLC BEH column (1.7 �m; 2.1 mmby 50 mm; Waters Corp.). The mobile phase was a mixture of 0.05%formic acid in 5 mM ammonium acetate (A) and methanol (B). Thegradient elution was started from 10% B, maintained for 1 min, increasedlinearly to 57% B over 24 min, and then increased linearly to 100% B overthe next 2 min and finally decreased to 10% B to reequilibrate the column.The column temperature was set at 35°C, and the flow rate was 0.4 ml/min. The eluent was monitored by UV detection at 316 nm. The MSdetection was conducted using a Synapt Q-TOF high-resolution massspectrometer (Waters Corp., Milford, MA) operated in positive ion elec-trospray (ES-positive) mode. A mass range of m/z 80 to 1,000 was ac-quired. Nitrogen and argon were employed as the desolvation gas andcollision gas, respectively. The desolvation temperature was set at 350°C,and the source temperature was set at 100°C. Leucine enkephalin was used

as a lock mass compound ([M � H]� m/z 556.2771) for accurate massmeasurements and was infused into the LockSpray ion source via a sepa-rate ionization probe. Data acquisition was performed using the MSE scanfunction, which was programmed with two independent collision ener-gies (CE). At low collision energy, the transfer CE and trap CE were 2 eVand 3 eV, respectively. At high collision energy, the transfer CE and trapCE were 4 eV and ramped from 15 eV to 30 eV, respectively. In thismanner, the precursor ions and fragmentation information were ob-tained in a single run. The mass spectrometer was operated using Mass-Lynx 4.1 software. The metabolites were mined from the data using themass defect filtering function (40-mDa tolerance window) with theMetaboLynx XS subroutine of the MassLynx software.

Quantification of arbidol and its metabolites in human plasma. Theconcentrations of arbidol, M5, M6-1, and M8 in plasma were determinedusing a validated LC-tandem mass spectrometry (MS-MS) method. De-butyldronedarone was used as the internal standard (IS). A 50-�l aliquotof plasma containing the IS was treated with methanol (200 �l) and cen-trifuged. The supernatant was diluted with an equal volume of the mobilephase, and 5-�l aliquots were injected onto the LC–MS-MS system. Chro-matographic separation was achieved on a Gemini C18 column (5 �m; 50mm by 2.0 mm) using methanol-5 mM ammonium acetate containing0.1% formic acid as the mobile phase with gradient elution. Mass detec-tion was carried out on an API 4000 mass spectrometer (Applied Biosys-tems, Concord, Ontario, Canada) using an ES-positive detection mode.The multiple reaction-monitoring transitions selected were m/z 479 ([M � H� 2]�) to 281 for arbidol, m/z 481 ([M � H � 2]�) to 356 for N-dem-ethylsulfinylarbidol, m/z 495 ([M � H � 2]�) to 370 for sulfinylarbidol,m/z 511 ([M � H � 2]�) to 466 for sulfonylarbidol, and m/z 501 ([M �H]�) to 114 for the IS. The assay was linear over the concentration rangeof 2.00 to 2,000 ng/ml for arbidol and sulfinylarbidol and 0.50 to 500ng/ml for N-demethylsulfinylarbidol and sulfonylarbidol.

Pharmacokinetic analysis. The pharmacokinetic parameters were de-termined using the WinNonlin noncompartmental analysis computerprogram (version 5.3; Pharsight, Mountain View, CA). The peak concen-tration (Cmax) and the time to reach it (Tmax) were determined directlyfrom the experimental data. The terminal elimination phase rate constant(Ke) was estimated using the least-squares regression analysis of theplasma concentration-time data obtained during the terminal log-linearphase. The terminal-phase half-life (t1/2) was calculated as 0.693/Ke. Thearea under the plasma concentration-time curve (AUC) was calculatedaccording to the linear trapezoidal method from 0 h to 72 h (AUC0-72) orto infinity (AUC0-�). The oral clearance (CL/F) was calculated as dose/AUC0-�.

Quantification of arbidol and its metabolites in human urine andfeces. The concentrations of arbidol, M5, M6-1, and M8 in urine and feceswere determined using the LC–MS-MS method. The urine and feces ho-mogenate were prepared in the same way as the plasma, except the fecalextraction supernatant was diluted 100-fold with the mobile phase beforeanalysis. The semiquantification of conjugate metabolites in urine wasperformed by the UPLC-UV method. The gradient elution program wasthe same as that for metabolite identification, and the UV detector was setat 316 nm. The glucuronide conjugates (M18, M20-1, and M20-2) andsulfate conjugates (sulfate arbidol [M10], sulfate N-demethylsulfinylarbi-dol [M11-2], sulfate sulfinylarbidol [M14-1], and sulfate N-demethylsul-fonylarbidol [M15]) were semiquantified using arbidol as a calibrationstandard. The percentage of the dose excreted over 0 h to 96 h was thencalculated as follows: excretion (percent) (moles of metabolites ex-creted in human urine or feces/moles of arbidol dosed to humans) � 100.

Incubation with HLM, HIM, HKM, and cDNA-expressed P450s andFMOs. Arbidol was incubated in triplicate with pooled HLM, HIM,HKM, human cDNA-expressed P450s (CYP1A1, CYP1A2, CYP1B1,CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1,CYP3A4, CYP3A5, and CYP4A11), or FMOs (FMO1, FMO3, andFMO5). The incubation mixtures (200 �l) contained potassium phos-phate buffer (0.1 M, pH 7.4); individual P450 enzymes (50 pmol/ml);

Deng et al.

1744 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

FMOs (4 pmol/ml); HLM (1 mg/ml), HIM (1 mg/ml), or HKM (1 mg/ml); arbidol (5.0 �M or 50 �M); and NADPH (2.0 mM). Arbidol stocksolution was prepared in dimethyl sulfoxide (DMSO), and the finalDMSO concentration in the incubation was 0.1% (vol/vol). The reactionswere initiated with the addition of NADPH, and the incubations wereperformed at 37°C in a water bath. After 60 min of incubation, the reac-tions were terminated with an equal volume of ice-cold acetonitrile. Con-trol samples without NADPH or substrate were included. The sampleswere analyzed using UPLC–Q-TOF MS.

For the microsomal stability assay, arbidol, M5, M6-1, and M8 (each at5.0 �M) were individually incubated with HLMs in the presence ofNADPH, using the same incubation conditions described above. The re-actions were terminated at 0 (T0), 5.0, 15, 30, and 60 min. The disappear-ance of test compound was monitored using UPLC–Q-TOF MS.

Inhibition of P450s or FMOs in HLMs and HIMs. The contributionsof P450s and FMOs to arbidol metabolism were distinguished by selec-tively inhibiting each enzyme system. The chemical inhibitors used were1-ABT (1 mM) for all P450 enzymes, �-naphthoflavone (2 �M) forCYP1A2, ticlopidine (24 �M) for 2C19, quinidine (8 �M) for 2D6,clomethiazole (24 �M) for CYP2E1, and ketoconazole (2 �M) forCYP3A4/5. The incubation mixtures (200 �l, in triplicate) containedphosphate buffer (0.1 M, pH 7.4), HLM (1 mg/ml) or HIM (1 mg/ml),arbidol (5.0 �M or 50 �M), NADPH (2.0 mM), and a single P450 chem-ical inhibitor. The FMOs were inhibited by heating the HLM and HIM at45°C for 5 min (without NADPH). The effect of heat inactivation on FMOactivity was confirmed by the incubation of benzydamine (a selectiveFMO substrate) with preheated microsomes and by determining the sub-sequent decrease in benzydamine N-oxide (m/z 326.187) production. Theformation of metabolites was analyzed by UPLC–Q-TOF MS as describedabove. A comparison was made to the controls without inhibitor, and theenzyme activity was expressed as a percentage of the activity of the control.

Data analyses. In the determination of the in vitro t1/2 of arbidol, M5,M6-1, and M8, the peak areas were converted to the percentage of thecompound that remained, using the T0 peak area values as 100%. The lnpercentage of the remaining test compound was plotted against incuba-tion time, and the slope of the linear regression (�k) was used in theconversion to in vitro t1/2. The intrinsic clearance (CLint, in ml/min/kg)was calculated using the following formula (9, 10): CLint (0.693/in vitrot1/2) � (ml incubation/mg microsome) � (48.8 mg microsome/gm liver) �(25.7 gm liver/kg body weight).

To estimate the contributions of the different P450 and FMO isoformsto the formation of M5, M6-1, and M8 in human liver, the percentages ofthe total normalized rate (percent TNR) were calculated as reported pre-viously (11). This was done by multiplying the reaction rate of each iso-form by the specific protein content in HLMs to yield the normalized rate(NR), which was expressed in pmol/min/mg microsomes. The NRs ob-tained were then summed up as the TNR. The percent TNR for eachisoform was then calculated according to the following equation: percentTNR (NR/TNR) � 100 {[pmol/min/pmol (P450 or FMO) pmol(P450 or FMO)/mg]/[(pmol/min/pmol P450 or FMO pmol P450 orFMO)/mg)]} � 100.

Statistical analysis to investigate the effects of inhibitors was carriedout in Microsoft Excel (version 2010) using a two-sided t test for indepen-dent samples. All results are presented as mean and standard deviation(SD).

RESULTSMass spectral fragmentation of arbidol and metabolite stan-dards. We previously reported the identification of arbidolmetabolites using high-performance liquid chromatography(HPLC)-ion trap mass spectrometry, and the MS fragmentationsequences of arbidol and its metabolites were proposed (6). In thepresent study, similar fragment pathways were observed for arbi-dol and the metabolite standards using Q-TOF MS. Typically,under the high collision energy, arbidol, N-demethylsulfonylarbi-

dol, sulfonylarbidol, and 4=-hydroxylated arbidol produced abun-dant fragments by sequential loss of dimethylamine (45.058 Da)or methylamine (31.039 Da), acetaldehyde (44.026 Da), and thephenylthio radical (109.011 Da), phenylsulfonyl radical, or 4=-hydroxylphenylthio radical (141.001 Da). For sulfinylarbidol andN-demethylsulfinylarbidol, the major fragment ions were pro-duced by a sequential loss of phenylsulfiny radical (125.006 Da),dimethylamine or methylamine, and acetaldehyde. The high-col-lision-energy mass spectra and chromatographic behaviors of thedetected metabolites were compared with those of the parentcompound and the available authentic standards to characterizethe structural modification.

Metabolic profiles in human plasma, urine, and feces. Table 1lists the possible arbidol metabolites, including their proposedelemental compositions and chemical structures, the retentiontime of each chromatographic peak, and the characteristic massspectral fragmentation ions. The metabolic profiles of arbidol inplasma, urine, and feces are shown in Fig. 1. The identified meta-bolic pathways of arbidol in humans are shown in Fig. 2.

(i) Urine. A total of 32 chromatographic peaks were observedin the urinary metabolic profile. The peak at 23.0 min was assignedto unchanged arbidol because the retention time and mass spec-tral fragmentation patterns were identical to those of arbidol.Likewise, the metabolite peaks at 10.7, 14.9, 14.8, and 14.5 min(which contained two coeluted metabolites) were identified asoxidative S-dealkylation metabolite (M1), N-demethylsulfinyl-arbidol (M5), sulfinylarbidol (M6-1), N-demethylsulfonyl-arbidol (M7), and sulfonylarbidol (M8), respectively. Anotherphase I metabolite eluted at 22.6 min was assigned as M3-2, and itsstructure was proposed as dimethylamine N-demethylated arbi-dol. All the other metabolites detected were phase II conjugates.The sulfate conjugation of M0, M1, M3-1, M3-2, M5, M6-1, M7,and M8 produced metabolites M10, M4, M9-1, M9-2, M11-2,M14-1, M15, and M16, respectively. Glucuronide conjugation ofM0, M1, M3-1, M3-2, M5, M6, M7, and M8 yielded the metabo-lites M18, M13 (two isomers), M17-1, M17-2, M19 (two diaste-reomers), M20 (two diastereomers and two isomers), M21, andM22, respectively. These glucuronide conjugates were readily hy-drolyzed by �-glucuronidase, and the chromatographic peaks forthe corresponding aglycones significantly increased after hydroly-sis. These results supported the structural assignment for glucu-ronide conjugate metabolites. The sulfate conjugates of N-des-methyl M1 (M2), 1-methylindole N-desmethyl arbidol (M11-1),and di-N-demethylsulfonylarbidol (M12) and the glucuronideconjugate of 4=-hydroxylarbidol (M14-2) were detected at tracelevels in the urine. Compared with the previously reported results(6), the newly identified metabolic pathways included N-dem-ethylation of 1-methylindole, 4=-hydroxylation, oxidative S-deal-kylation, and di-N-demethylation. The most abundant urinarymetabolites were glucuronide arbidol (M18) and glucuronide sul-finylarbidol (M20-1 and M20-2), which accounted for 1.5% and2.1% of the dose, respectively. The mean urinary excretion of sul-fate arbidol (M10), sulfate N-demethylsulfinylarbidol (M11-2),sulfate sulfinylarbidol (M14-1), and sulfate N-demethylsulfonyl-arbidol (M15) amounted to 0.3%, 0.4%, 1.3%, and 0.7% of thedose. The recovery of arbidol and its oxidative metabolites (M5,M6-1, and M8) from urine were less than 1% of the dose.

(ii) Feces. A total of 24 metabolites, along with the parent drug,were detected in the extracts of fecal homogenates. The production mass spectra and the UPLC retention times of these metabo-

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1745

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

TABLE 1 Metabolites for arbidol in human plasma, urine, and feces after oral drug administration

Metabolite Chemical formula Proposed chemical structure Rt (min)c m/z [M � H]� Fragment ion(s)

Arbidol C22H25BrN2O3S 23.0 (P, U, F) 477.084a 432.028, 387.997, 278.988

M0 477.085b

M1 C16H21BrN2O4 10.7 (P, U, F) 385.078a 322.008, 295.978

385.076b

M2 C15H19BrN2O7S 10.1 (U) 451.017a

451.018b

M3-1 C21H23BrN2O3S 20.4 (F) 463.072a 439.935, 418.011

463.069b

M3-2 C21H23BrN2O3S 22.6 (P, U, F) 463.072a 432.025, 388.000, 323.017, 293.976,278.990

463.069b

M4 C16H21BrN2O7S 5.3 (U) 465.032a 385.076

465.033b

M5 C21H23BrN2O4S 14.9 (P, U, F) 479.062a 354.057, 323.017

479.064b

M6-1 C22H25BrN2O4S 14.8 (P, U, F) 493.079a 368.074

493.080b

M6-2 C22H25BrN2O4S 19.6 (F) 493.078a 448.021, 403.996, 323.018, 293.979,278.991

493.080b

M7 C21H23BrN2O5S 14.5 (P, U, F) 495.059a 464.017, 419.991

495.059b

M8 C22H25BrN2O5S 14.5 (P, U, F) 509.069a 464.014, 419.988

509.075b

M9-1 C21H23BrN2O6S2 16.9 (U) 543.030a 463.067, 418.013

543.026b

(Continued on following page)

Deng et al.

1746 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

TABLE 1 (Continued)

Metabolite Chemical formula Proposed chemical structure Rt (min)c m/z [M � H]� Fragment ion(s)

M9-2 C21H23BrN2O6S2 22.6 (P, U, F) 543.025a 463.067, 432.024

543.026b

M10 C22H25BrN2O6S2 19.4 (P, U, F) 557.042a 477.088

557.042b

M11-1 C21H23BrN2O7S2 8.3 (U) 559.021a 354.057, 309.003

559.021b

M11-2 C21H23BrN2O7S2 14.3 (P, U, F) 559.020a 354.057, 434.011, 402.972

559.021b

M12 C20H21BrN2O8S2 13.3 (U, F) 560.998a 543.973, 464.015, 419.990

561.000b

M13-1 C22H29BrN2O10 6.3 (U, F) 561.106a 385.078, 340.018

561.108b

M13-2 C22H29BrN2O10 9.6 (U) 561.108a

561.108b

M14-1 C22H25BrN2O7S2 10.3 (P, U, F) 573.035a 493.076, 368.076

573.037b

M14-2 C22H25BrN2O7S2 16.7 (U, F) 573.039a 493.083, 403.995

573.037b

M15 C21H23BrN2O8S2 14.0 (P, U, F) 575.014a 495.056, 464.020, 419.989

575.016b

M16 C22H25BrN2O8S2 10.8 (P, U, F) 589.030a 464.013, 419.990

589.031b

M17-1 C27H31BrN2O9S 17.9 (U) 639.103a 418.008

639.101b

(Continued on following page)

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1747

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

lites were compared with those detected in human urine. Thenewly identified metabolites were M3-1 and M6-2, which wereeluted at 20.4 and 19.6 min, respectively. Their structures wereproposed as 1-methylindole N-demethylated arbidol and 4=-hy-droxylarbidol, respectively. Based on the high collision energymass spectra, the chemical structure of M6-2 was further con-firmed by comparison with a reference standard. Arbidol was the

predominate component excreted in feces, accounting for 32.4%of the dose, and the major metabolite in feces was the sulfateconjugate of arbidol (M10), accounting for 3.0% of the dose. Theother oxidative metabolites (M5, M6-1, and M8) represented�2% of the dose.

(iii) Plasma. A total of 16 metabolites were detected in thehuman plasma extracts, and M6-1 was the major drug-related

TABLE 1 (Continued)

Metabolite Chemical formula Proposed chemical structure Rt (min)c m/z [M � H]� Fragment ion(s)

M17-2 C27H31BrN2O9S 21.7 (U, F) 639.101a 432.024

639.101b

M18 C28H33BrN2O9S 20.2 (P, U, F) 653.114a 477.086

653.11b

M19-1 C27H31BrN2O10S 12.1 (U, F) 655.096a 530.087, 323.016

655.096b

M19-2 C27H31BrN2O10S 13.0 (U) 655.093a 530.085, 323.014

655.096b

M20-1 C28H33BrN2O10S 9.7 (P, U, F) 669.108a 544.104

669.112b

M20-2 C28H33BrN2O10S 11.0 (P, U, F) 669.110a 544.105

669.112b

M20-3 C28H33BrN2O10S 16.9 (U) 669.110a

669.112b

M20-4 C28H33BrN2O10S 18.1 (U) 669.109a

669.112b

M21 C27H31BrN2O11S 12.7 (U, F) 671.089a 495.059

671.091b

M22 C28H33BrN2O11S 11.2 (P, U, F) 685.105a 509.071, 464.011, 419.988

685.107b

a Measured m/z values for protonated ions.b Exact m/z values for protonated ions.c Rt, retention time; P, plasma; U, urine; F, feces.

Deng et al.

1748 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

component, followed by the unmetabolized arbidol. The othermetabolites that presented at relatively high levels in humanplasma were M5 and M8. Metabolites M1, M3-2, and M7 wereobserved in plasma as minor phase I metabolites. These oxidativemetabolites could be further metabolized to form phase II conju-gates (M9-2, M11-2, M14-1, M15, M16, M20-1, M20-2, and

M22). The glucuronide conjugate (M18) and the sulfate conjugate(M10) of the parent drug were also detected in plasma. All of theseconjugates were detected as minor metabolites.

Pharmacokinetics of arbidol and metabolites M5, M6-1, andM8. Table 2 presents the pharmacokinetic parameters determinedby noncompartmental analysis. The mean plasma concentration-

FIG 1 Metabolic profiles of arbidol after a single oral administration of 200-mg arbidol hydrochloride capsules. (A) Pooled plasma samples collected at 2 hpostdose. (B) Pooled urine samples collected at 0 to 24 h postdose. (C) Pooled feces samples collected at 24 to 96 h postdose. On the left are shown metaboliteprofiles by MS detection, and on the right are metabolite profiles by UV detection at 316 nm.

FIG 2 Identified metabolic processes of arbidol in humans. The major metabolic pathway was sulfoxidation. Details of the chemical structures are shown inTable 1.

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1749

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

versus-time profiles for arbidol, M5, M6-1, and M8 are shown inFig. 3. After oral drug administration, arbidol was rapidly ab-sorbed, with a mean tmax of 1.38 h. M5 had a comparable tmax at1.50 h, whereas the maximum plasma concentrations of M6-1 andM8 were reached much later, at 13.0 h and 19.0 h, respectively.The mean t1/2 values of the metabolites were 26.3, 25.0, and 25.7 h,which were clearly longer than that of arbidol (15.7 h). The sul-fone metabolite M8 had the lowest concentration among the three

major circulating metabolites, with a mean Cmax of 22.7 ng/ml anda mean metabolite-to-parent AUC0-t ratio (AUCm/AUCp) of0.5 � 0.2. The N-demethylsulfinyl compound M5 was the secondmost abundant metabolite in circulation, with a Cmax of 80.5ng/ml and a moderate AUCm/AUCp ratio of 0.9 � 0.3. The sulfinylmetabolite M6-1 was the major circulating species, with a Cmax of525 ng/ml and the highest AUCm/AUCp ratio at 11.5 � 3.6. In themean plasma concentration-time profiles of arbidol, a secondpeak was observed at about 3 h; it was not a result of enterohepaticrecirculation but of the interindividual variability in Tmax.

In vitro metabolism of arbidol. (i) Human microsomes. Bio-transformation of arbidol (at concentrations of 5.0 �M and 50�M) was investigated in HLMs, HIMs, and HKMs. Five promi-nent metabolites, namely, M3-2, M5, M6-1, M7, and M8, wereformed in HLMs and HIMs (Fig. 4), whereas only trace amountsof M6-1 were detected in HKMs. At the arbidol concentration of5.0 �M in HLMs, the parent drug was extensively metabolized,and the amounts of metabolites formed followed the order ofM3-2 � M5 � M7 � M6-1 � M8. At the arbidol concentration of50 �M, the yield of M6-1 increased, and the order was changed to

TABLE 2 Pharmacokinetic parameters of arbidol and its three majormetabolites in plasma of four healthy male subjects after a single oraladministration of 200 mg of arbidol hydrochloride

Parameter

Value

Arbidol M5 M6-1 M8

Tmax (h) 1.38 � 1.11 1.50 � 1.00 13.0 � 8.2 19.0 � 14.0Cmax (ng/ml) 467 � 174 80.5 � 37.5 525 � 147 22.7 � 9.8AUC0-t (ng · h · ml�1) 2,103 � 614 1,743 � 466 23,104 � 4,829 1,040 � 483AUC0-� (ng · h · ml�1) 2,203 � 691 2,121 � 546 28,399 � 7,656 1,315 � 561t1/2 (h) 15.7 � 3.8 26.3 � 5.9 25.0 � 5.4 25.7 � 8.8CL/F (liters/h) 99 � 34

FIG 3 Mean plasma concentration-time profiles of arbidol, M5, M6-1, and M8 after a single oral administration of 200-mg arbidol hydrochloride capsules to4 healthy male subjects. (A) Linear scale. (B) Semilogarithmic scale. The error bars indicate SD.

Deng et al.

1750 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

M3-2 � M6-1 � M5 � M7 M8. Similar trends of metaboliteformation were observed in HIMs. Omission of NADPH com-pletely abolished the metabolism of arbidol, which indicated thatthese metabolic processes were NADPH dependent.

To determine the human liver microsomal stability of arbidoland its major circulating metabolites, the parent drug, M5, M6-1,and M8 were separately incubated with HLMs in the presence ofNADPH. It was found that M5, M6-1, and M8 were metabolizedmainly via N-demethylation and/or sulfoxidation. The CLint val-ues were calculated to be 116, 23.6, 28.9, and 46.6 ml/min/kg forarbidol, M5, M6-1, and M8, respectively (Table 3). According tothe classification criteria (12, 13), M5 and M6-1 were categorizedas medium-clearance compounds, whereas arbidol and M8 werehigh-clearance compounds, and the most labile compound wasarbidol, which had a CLint value of 109 ml/min/kg.

(ii) Recombinant P450 and FMO isoforms. A series of cDNA-expressed P450s and FMOs were used to evaluate the contribu-tions of individual enzymes to arbidol metabolism. Each isoformshowed varying degrees of efficiency in the production of oxida-tive metabolites (Fig. 5). Among those tested, FMOs were respon-sible only for sulfoxidation, whereas P450s were involved in theformation not only of M6-1 and M8, but also of M5 and M7.

Isoforms that showed elevated levels of M6-1 production includedCYP1A2, CYP2C19, CYP2D6, CYP2E1, CYP3A4, CYP3A5,FMO1, FMO3, and FMO5, with FMO1 being the most efficientisoform. The same array of P450s also contributed to the forma-

FIG 4 Metabolic profiles of arbidol in human liver (A) and intestine (B) microsome incubations at 5.0 and 50 �M arbidol. The incubations were for 60 min at1 mg/ml protein and 37°C.

TABLE 3 In vitro t1/2 and CLint values for arbidol, M5, M6-1, and M8 inHLMs

Compound t1/2 (min) CLint (ml/min/kg)

Arbidol 8.20 � 0.29 116 � 3.7M5 37.0 � 2.0 23.6 � 1.3M6-1 30.3 � 3.2 28.9 � 2.9M8 18.7 � 0.6 46.6 � 1.5

FIG 5 Formation of M5, M6-1, M7, and M8 in incubations of arbidol (5.0 or 50�M) with human cDNA-expressed P450s and FMOs. The incubations were per-formed at 50 pmol of P450/ml and at 4 pmol of FMO/ml. The error bars indicate SD.

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1751

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

tion of M5, M7, and M8. P450 isoforms, such as CYP1A1,CYP1B1, CYP2A6, CYP2B6, CYP2C8, CYP2C9, and CYP4A11,were not involved in the metabolism of arbidol. In adult humans,FMO1 is primarily found in the kidney and intestines (14). Basedon the microsomal incubation experiments, the FMO1 contribu-tion to the formation of M6-1 was mainly attributed to the FMO1expressed in the intestine, not in the kidney. The kinetic parame-ters for arbidol metabolism by P450s and FMO3 were scaled to atypical human liver, and it was calculated that CYP3A4 showedthe highest percent TNR in the formation of sulfoxidation andN-demethylation metabolites (Table 4).

(iii) Inhibition studies. The effects of CYP inhibitors and heatinactivation on the formation of oxidative metabolites (M5,M6-1, M7, and M8) at 5.0 �M and 50 �M arbidol in HLMs andHIMs are shown in Fig. 6 and 7. In HLMs, when the substrateconcentration was set at 5.0 �M, 1-ABT inhibited the formationof M5, M7, and M8 by �90% and that of M6-1 by 39%. Little orno significant inhibition was observed with other inhibitors andheat treatment, with the exception of ketoconazole, which po-tently inhibited the formation of M5, M7, and M8 by 81%, 97%,and 93%, respectively. In contrast, the formation of M6-1 in-creased by 185%. At 50 �M arbidol, 1-ABT inhibited the forma-tion of the four metabolites by �90% in HLMs. Heat treatmentinhibited the formation of M5, M7, and M8 by approximately25%. Under the same conditions, the metabolism of benzy-damine, an FMO probe substrate, was inhibited by 82%. Amongthe five specific P450 inhibitors, ketoconazole significantly re-duced the formation of M6-1 by 69%, whereas �-naphthoflavone,ticlopidine, quinidine, and clomethiazole inhibited the formationof M6-1 by 46% to 51%. Similar inhibitory effects were observedfor M5, M7, and M8, with ketoconazole as the most potent inhib-itor. In HIM incubations, 1-ABT inhibited the formation of fourmetabolites by approximately 90% regardless of the arbidol con-centration. Heating the HIMs for 5 min at 45°C prior to incuba-tion decreased M6-1 formation by about 20% and the formationof M5, M7, and M8 by 30 to 65%.

DISCUSSION

In the present study, the metabolism and excretion of arbidol wereinvestigated in healthy male volunteers after a single oral admin-istration of 200 mg of arbidol hydrochloride. This study is the firstto report on the arbidol metabolites in human plasma and feces. A

total of 31 metabolites were found and identified in urine, 24 infeces, and 16 in plasma. Arbidol was mainly excreted in urine asphase II conjugates. The major urinary metabolites were glucu-ronide arbidol (M18) and glucuronide sulfinylarbidol (M20-1and M20-2), which accounted for approximately 3.6% of the dose.The urinary excretion of sulfate conjugates (M10, M11-2, M14-1,and M15) was estimated to be 2.7% of the dose. The urinary ex-cretion of the parent drug was less than 0.1% of the dose. Similarto preclinical species (1), about 32.4% of the total intake dose ofarbidol was excreted unchanged within 0 to 96 h via feces. Thesulfate conjugate of arbidol (M10) was the major metabolite inhuman feces and accounted for 3.0% of the dose, and other me-tabolites (M5, M6-1, and M8) represented �2% of the dose. Be-cause the reference materials for the conjugated metabolites werenot available, the amounts of these phase II metabolites excretedwere semiquantified using arbidol as a calibration standard. Fur-ther experiments should be carried out using radiolabeled com-pound to accurately evaluate the excretion of arbidol in humansafter oral drug administration.

It has long been known that circulating metabolites may con-tribute to drug efficacy and side effects, and attention should bepaid to metabolites that are formed at greater than 10% of parentdrug systemic exposure (http://www.fda.gov/cder/guidance/). Af-ter the oral administration of arbidol hydrochloride to healthyvolunteers, M6-1 was detected as the major circulating species inthe plasma, followed by the parent drug, M5, and M8. The meanelimination half-lives of the three metabolites were longer thanthat of the parent drug. Although the average plasma concentra-tions of M5 and M8 were much lower than that of arbidol, theirAUC0-t values were equivalent or comparable to that of the parent,and the AUCm/AUCp values were 0.9 � 0.3 and 0.5 � 0.2 for M5and M8, respectively. The exposures of M6-1 in human plasma asdetermined by AUC0-� and Cmax were 12.9- and 1.12-fold greaterthan those of unchanged arbidol. Multiple factors could contrib-ute to the pharmacokinetic properties of M6-1, including its for-mation, stability in the circulation, tissue binding, and the effectsof transporters in the liver and intestine. Assessing the safety andefficacy of M6-1 is important because of its high exposure andlong elimination half-life. A search of the literature resulted in asingle patent, which reported that M6-1 could inhibit protein ki-nase C (50% inhibitory concentration [IC50] 7.78 �M) (15). Itwas speculated that during arbidol treatment, metabolite M6-1

TABLE 4 Total normalized rates of formation of M5, M6-1, and M8 by individual P450 and FMO isoforms

Isoform

Mean content inhuman livera

(pmol isoform/mg protein)

Rate

Formation (pmol/min/mg) Normalized (pmol/min/mg) Total normalized (%)

M5 M6-1 M8 M5 M6-1 M8 M5 M6-1 M8

CYP1A2 45.0 0.11 0.31 5.01 14.0 13.1 8.30CYP2C19 19.0 0.04 0.18 0.73 3.50 1.90 2.10CYP2D6 10.0 0.21 0.75 0.01 2.10 7.51 0.15 5.50 4.50 5.00CYP2E1 49.0 0.08 0.31 3.76 15.36 9.80 9.10CYP3A4 108 0.20 0.57 0.02 21.2 61.1 2.04 55.3 36.3 69.6CYP3A5 1.00 0.09 0.34 0.01 0.09 0.34 0.01 0.20 0.20 0.40FMO1 NA 2.66 0.03FMO3b 80.0 0.57 45.6 27.1FMO5 NA 0.15a Mean content data were obtained from Rodrigues (11). NA, not available.b Mean content data for FMO3 were obtained from BD Gentest (2003 product catalog).

Deng et al.

1752 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

may account for some of the pharmacological activities associatedwith arbidol, and measuring arbidol concentrations alone mayunderestimate the potency and duration of effect for the agent. Inaddition, the recommended dosage of arbidol for the treatment ofinfluenza is 200 mg three times daily for 5 to 10 days, and theextended half-life of M6-1 suggested that it would accumulate onrepeated daily dosing of arbidol. Further investigation is needed tounderstand the importance of M6-1 in terms of safety and effi-cacy.

A comprehensive set of in vitro experiments was conducted toinvestigate the biotransformation of arbidol (at 5.0 �M and 50�M). It was found that arbidol was metabolized by HLMs andHIMs, but not by HKMs, which suggested that the liver and intes-tines could be the major organs that metabolize arbidol in hu-mans. HLM stability analysis revealed that the CLint of arbidol wasmuch higher than those of M5, M6-1, and M8, which may par-tially explain the longer plasma t1/2 of M5, M6-1, and M8 than ofthe parent drug.

It was reported that the major monooxygenases that catalyzethe formation of aliphatic sulfoxides are the P450s and FMOs(16–21). Identification of the human P450 and FMO enzymes

FIG 6 Inhibition of formation of M5, M6-1, M7, and M8 in the incubation of arbidol (5.0 or 50 �M) in HLMs by P450 inhibitor(s) and heat treatment.Remaining activity % is metabolic activity in the presence of inhibitor divided by metabolic activity in microsomes. Two-sided t test analyses: *, P � 0.01, and **,P � 0.001 compared with the control value. The error bars indicate SD.

FIG 7 Inhibition of formation of M5, M6-1, M7, and M8 in the incubation ofarbidol (5.0 or 50 �M) in HIMs by the P450 inhibitor 1-aminobenzotriazoleand heat treatment. Remaining activity % is metabolic activity in the presenceof inhibitor divided by metabolic activity in microsomes. Two-sided t testanalyses: *, P � 0.01, and **, P � 0.001, compared with the control value. Theerror bars indicate SD.

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1753

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

involved in arbidol metabolism was carried out using isoform-screening assays. The results indicated that multiple enzymes,including CYP1A2, CYP2C19, CYP2D6, CYP2E1, CYP3A4,CYP3A5, FMO1, FMO3, and FMO5, were capable of metaboliz-ing arbidol. However, FMOs were involved only in arbidol sul-foxidation. To better estimate the contribution of each enzyme tothe overall metabolism of arbidol in HLMs, the activities of P450sand FMO3, the major FMO isoform in the human liver (22), werenormalized to the content of each enzyme in HLMs (Table 4). Theresults indicated that CYP3A4 was the most active enzyme in-volved in arbidol metabolism, followed by FMO3 (which cata-lyzed only the formation of M6-1), CYP2E1, CYP1A2, CYP2D6,CYP2C9, and CYP3A5. 1-ABT and heat pretreatment of HLMsand HIMs were used to differentiate the contributions of P450sand FMOs to the metabolism of arbidol in humans. At a lowarbidol concentration (5.0 �M), incubation of HLMs in the pres-ence of 1-ABT decreased the formation of M5, M7, and M8 by�90% and that of M6-1 by 39%. In contrast, mild heat treat-ments, known to significantly reduce FMO activity, weakly af-fected arbidol metabolism in HLMs. These results indicated thatarbidol metabolism was predominantly P450 driven comparedwith FMOs. Furthermore, P450 chemical inhibition studies re-vealed that inhibition of CYP3A4 with ketoconazole decreased theproduction of M5, M7, and M8 by �80%, and this could be thereason for the corresponding increase of M6-1 (to 185%), becausethe secondary metabolism of M6-1 was inhibited by ketoconazole,and other P450s and FMOs catalyzed the formation of M6-1 com-pensatory. At a high arbidol concentration (50 �M), similartrends were observed in the inhibitory effects of 1-ABT and heattreatments (90% versus 25%). Among the five P450 inhibitors,ketoconazole showed the most potent inhibitory effect. In con-trast to the results obtained at low substrate concentration, otherP450 chemical inhibitors exhibited inhibitory effects to a certaindegree. This result could be attributed to the saturation of multi-P450-mediated arbidol metabolism at high substrate concentra-tion, and consequently, blocking the function of a single enzymeobviously reduced metabolite formation. When arbidol was incu-bated with HIMs, both 1-ABT and heat treatment inhibited me-tabolite formation, and the inhibitory effect of 1-ABT was alwaysgreater than that of heat treatment (Fig. 7). Therefore, P450s arethe major enzymes involved in arbidol metabolism in human in-testines, with FMOs contributing to a lesser degree. Overall, the invitro studies indicated that P450s mainly catalyzed arbidol metab-olism by HLMs and HIMs compared with FMOs, with CYP3A4 asthe most active enzyme. It could be predicted that there was met-abolic DDI potential between arbidol and coadministered drugsthat are CYP3A4 inhibitors or inducers.

In conclusion, arbidol undergoes extensive phase I and phase IImetabolism in humans. New metabolic pathways have been pro-posed, including N-demethylation of 1-methylindole, 4=-hy-droxylation, oxidative S-dealkylation, and di-N-demethylation.The study extended the available information on the pharmaco-kinetics of arbidol metabolites in the circulation. Sulfinylarbidol(M6-1) is the most abundant component in human plasma, withan extended Tmax, prolonged elimination half-life, and higher ex-posure than the parent drug. The in vitro studies indicated thatCYP3A4 is the major enzyme involved in arbidol metabolism inthe liver and intestines, with minor contributions from other P450and FMO enzymes. Arbidol may potentially interact with CYP3A4inhibitors and inducers. Overall, the in vivo and in vitro findings

provided new insights into the metabolism and disposition of ar-bidol in humans.

ACKNOWLEDGMENTS

This work was supported in part by the National Science and TechnologyMajor Project Key New Drug Creation and Manufacturing Program,China (grant 2009ZX09301-001).

We thank Cen Xie from Shanghai Institute of Materia Medica forhelpful discussions on in vitro experiments.

REFERENCES1. Boriskin YS, Leneva IA, Pecheur EI, Polyak SJ. 2008. Arbidol: a broad-

spectrum antiviral compound that blocks viral fusion. Curr. Med. Chem.15:997–1005.

2. Teissier E, Zandomeneghi G, Loquet A, Lavillette D, Lavergne JP,Montserret R, Cosset FL, Bockmann A, Meier BH, Penin F, Pecheur EI.2011. Mechanism of inhibition of enveloped virus membrane fusion bythe antiviral drug arbidol. PLoS One 6:e15874. doi:10.1371/journal.pone.0015874.

3. Leneva IA, Russell RJ, Boriskin YS, Hay AJ. 2009. Characteristics ofarbidol-resistant mutants of influenza virus: implications for the mecha-nism of anti-influenza action of arbidol. Antiviral Res. 81:132–140.

4. Pecheur EI, Lavillette D, Alcaras F, Molle J, Boriskin YS, Roberts M,Cosset FL, Polyak SJ. 2007. Biochemical mechanism of hepatitis C virusinhibition by the broad-spectrum antiviral arbidol. Biochemistry (Mos-cow) 46:6050 – 6059.

5. Delogu I, Pastorino B, Baronti C, Nougairede A, Bonnet E, de Lam-ballerie X. 2011. In vitro antiviral activity of arbidol against Chikungunyavirus and characteristics of a selected resistant mutant. Antiviral Res. 90:99 –107.

6. Wang Y, Chen X, Li Q, Zhong D. 2008. Metabolite identification ofarbidol in human urine by the study of CID fragmentation pathways usingHPLC coupled with ion trap mass spectrometry. J. Mass Spectrom. 43:1099 –1109.

7. Liu X, Huang YW, Li J, Li XB, Bi KS, Chen XH. 2007. Determinationof arbidol in human plasma by LC-ESI-MS. J. Pharm. Biomed. Anal. 43:371–375.

8. Metz R, Muth P, Ferger M, Kukes VG, Vergin H. 1998. Sensitivehigh-performance liquid chromatographic determination of arbidol, anew antiviral compound, in human plasma. J. Chromatogr. A 810:63– 69.

9. Naritomi Y, Terashita S, Kimura S, Suzuki A, Kagayama A, SugiyamaY. 2001. Prediction of human hepatic clearance from in vivo animal ex-periments and in vitro metabolic studies with liver microsomes from an-imals and humans. Drug Metab. Dispos. 29:1316 –1324.

10. Davies B, Morris T. 1993. Physiological parameters in laboratory animalsand humans. Pharm. Res. 10:1093–1095.

11. Rodrigues AD. 1999. Integrated cytochrome P450 reaction phenotyping:attempting to bridge the gap between cDNA-expressed cytochromes P450and native human liver microsomes. Biochem. Pharmacol. 57:465– 480.

12. Reichman M, Gill H. 2009. Automated drug screening for ADMET prop-erties, p 150. In Nassar AF, Hollenberg PF, Scatina J (ed), Drug metabo-lism handbook: concepts and applications. John Wiley & Sons, Inc.,Hoboken, NJ.

13. Wilkinson GR, Shand DG. 1975. Commentary: a physiological approachto hepatic drug clearance. Clin. Pharmacol. Ther. 18:377–390.

14. Yeung CK, Lang DH, Thummel KE, Rettie AE. 2000. Immunoquanti-tation of FMO1 in human liver, kidney, and intestine. Drug Metab. Dis-pos. 28:1107–1111.

15. Demin AV, Martianov VA, Shuster AM. 2010. Protein kinase C inhibi-tors exhibiting an anti-inflammatory, anti-allergic and anti-asthma effect.Russia patent WO/2010/064958.

16. Washio T, Arisawa H, Kohsaka K, Yasuda H. 2001. Identification ofhuman drug-metabolizing enzymes involved in the metabolism of SNI-2011. Biol. Pharm. Bull. 24:1263–1266.

17. Usmani KA, Karoly ED, Hodgson E, Rose RL. 2004. In vitro sulfoxida-tion of thioether compounds by human cytochrome P450 and flavin-containing monooxygenase isoforms with particular reference to theCYP2C subfamily. Drug Metab. Dispos. 32:333–339.

18. Pike MG, Martin YN, Mays DC, Benson LM, Naylor S, Lipsky JJ. 1999.Roles of FMO and CYP450 in the metabolism in human liver microsomesof S-methyl-N,N-diethyldithiocarbamate, a disulfiram metabolite. Alco-hol Clin. Exp. Res. 23:1173–1179.

Deng et al.

1754 aac.asm.org Antimicrobial Agents and Chemotherapy

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from

19. Madan A, Parkinson A, Faiman MD. 1993. Role of flavin-dependent mo-nooxygenases and cytochrome P450 enzymes in the sulfoxidation of S-methylN,N-diethylthiolcarbamate. Biochem. Pharmacol. 46:2291–2297.

20. Leoni C, Buratti FM, Testai E. 2008. The participation of human hepaticP450 isoforms, flavin-containing monooxygenases and aldehyde oxidasein the biotransformation of the insecticide fenthion. Toxicol. Appl. Phar-macol. 233:343–352.

21. Virkel G, Lifschitz A, Sallovitz J, Pis A, Lanusse C. 2004. Comparativehepatic and extrahepatic enantioselective sulfoxidation of albendazoleand fenbendazole in sheep and cattle. Drug Metab. Dispos. 32:536 –544.

22. Haining RL, Hunter AP, Sadeque AJ, Philpot RM, Rettie AE. 1997.Baculovirus-mediated expression and purification of human FMO3: cat-alytic, immunochemical, and structural characterization. Drug Metab.Dispos. 25:790 –797.

Biotransformation of Arbidol in Humans

April 2013 Volume 57 Number 4 aac.asm.org 1755

on May 11, 2014 by U

niv of Massachusetts A

mherst

http://aac.asm.org/

Dow

nloaded from