1521-009X/43/1/82–88$25.00 http://dx.doi.org/10.1124/dmd.114.060558DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 43:82–88, January 2015Copyright ª 2014 by The American Society for Pharmacology and Experimental Therapeutics

Imperatorin Is a Mechanism-Based Inactivator of CYP2B6

Liwei Zheng, Jiaojiao Cao, Dan Lu, Lin Ji, Ying Peng, and Jiang Zheng

School of Pharmacy (L.Z., J.C., D.L., L.J., Y.P.), Key Laboratory of Structure-Based Drug Design & Discovery of Ministry of Education(J.Z.), Shenyang Pharmaceutical University, Shenyang, Liaoning, P. R. China; and Center for Developmental Therapeutics, SeattleChildren’s Research Institute, Division of Gastroenterology and Hepatology, Department of Pediatrics, University of Washington

School of Medicine, Seattle, Washington (J.Z.)

Received August 15, 2014; accepted November 5, 2014

ABSTRACT

Imperatorin (IMP) is the major active ingredient in many commonmedicinal herbs. We examined the irreversible inhibitory effect ofIMP on CYP2B6. IMP produced a time- and concentration-dependent inactivation of CYP2B6. About 70% of activity ofCYP2B6 was suppressed after its incubation with 1.5 mM IMP for9 minutes. KI and kinact were found to be 0.498 mM and 0.079 min21,respectively. The loss of CYP2B6 activity required the presence ofNADPH. Glutathione and catalase/superoxide dismutase showedlittle protection against the IMP-induced enzyme inactivation.

Ticlopidine, a substrate of CYP2B6, showed protection of theenzyme against the inactivation induced by IMP. The estimatedpartition ratio of the inactivation was approximately 4. Additionally,a g-ketoenal intermediate was identified in microsomal incubationswith IMP. CYP1A2, CYP2A6, CYP2B6, CYP2D6, CYP2E1, CYP3A4,and CYP3A5 were found to be involved in bioactivation of IMP. Inconclusion, IMP is a mechanism-based inactivator of CYP2B6. Theformation of g-ketoenal intermediate may account for the enzymeinactivation

Introduction

Imperatorin (IMP), a linear furanocoumarin compound, is the majoractive ingredient in many common umbelliferous herbs used astraditional Chinese medicines, such as Cnidium monniericuss andAngelica dahurica (Baek et al., 2000; Lia and Chen, 2004). The IMP-containing herbs are widely used in China for the treatment ofosteoporosis and relief of swelling and pain (Zhang et al., 2003). Thereported pharmacological activities of IMP include anti-inflammatoryproperties (Garcia-Argaez et al., 2000), antiosteoporosis (Tang et al.,2008), antibacterial effects (Rosselli et al., 2007), antitumor (Kim et al.,2007), b-secretase inhibition (Marumoto and Miyazawa, 2010), andinhibition of myocardial hypertrophy (Zhang et al., 2010). Recentstudies have demonstrated that IMP participates in hypertensiontreatment by inhibiting voltage-dependent calcium channels andreceptor-mediated Ca2+ influx and release (He et al., 2007). Theobserved multiple pharmacological activities of IMP, along with itsnatural abundance, have shown a great potential for its use asa pharmaceutical agent.A number of furanocoumarin compounds have been found to be

inhibitors of cytochrome P450 (P450) enzymes. For example, 69,79-dihydroxybergamottin was the first furanocoumarin compound reportedas an inhibitor of the rat CYP3A subfamily (Edwards et al., 1996).Angelicin, isopimpinellin, 8-methoxypsoralen, psoralen, isopsoralen,bergamottin, and other related compounds in this family have showninhibitory effects on CYP1A1, CYP1A2, CYP2B6, CYP3A4, andCYP3A5 (Baumgart et al., 2005; Lin et al., 2005; Zhuang et al., 2013). Inaddition, several furanocoumarin compounds in umbelliferous vegetables,

such as 8-methoxypsoralen, bergapten, and isopimpinellin, have beenfound to be the mechanism-based inactivators of CYP2B1 (Cai et al.,1993; Koenigs and Trager, 1998). We speculated that IMP may showa similar irreversible inhibitory effect on the CYP2B subfamily.CYP2B6 has been found in the liver, brain, kidney, and heart in

humans (Thum and Borlak, 2000; Miksys and Tyndale, 2004).Hepatic CYP2B6 content ranges from 2 to 10% of the total P450content (Stresser and Kupfer, 1999). Human CYP2B6 has beenproven to be a very important metabolizing enzyme, and the enzyme isresponsible for the preferential metabolism of approximately 3–8% ofwidely used pharmaceuticals, such as buprokpion, methadone,ifosfamide, efavirenz, and cyclophosphamide (Faucette et al., 2000;Huang et al., 2000; Ward et al., 2003; Gerber et al., 2004). Todetermine the role of specific P450s in the clearance of various drugsand to prevent drug-drug interactions induced by the inhibition ofP450s, prediction and identification of compounds that act asmechanism-based inactivators of P450s have become an importantissue in the drug discovery process. The objectives of this study wereto investigate the mechanism-based inactivation of CYP2B6 by IMP,to characterize the reactive metabolites responsible for the enzymeinactivation, and to identify the P450 enzymes responsible for themetabolic activation of IMP.

Materials and Methods

Imperatorin (IMP) with purity of 98% was purchased from Shanghai YuanyeBiologic Technology Co., Ltd. (Shanghai, China). Glutathione (GSH), hexylglutathione, bupropion, propranolol, ticlopidine, and NADPH were purchasedfrom Sigma-Aldrich (St. Louis, MO). Liver microsomes were prepared frommale Sprague-Dawley rats, by following the published method in Lin et al.(2007). Recombinant human P450 enzymes were acquired from BD Gentest(Woburn, MA). All organic solvents were from Fisher Scientific (Springfield,NJ). Distilled water was purchased from Wahaha Co., Ltd. (Hangzhou, China).

This work was supported in part by the National Natural Science Foundation ofChina [Grant 81373471].

dx.doi.org/10.1124/dmd.114.060558.

ABBREVIATIONS: EPI, enhanced product ion; GSH, glutathione; IMP, imperatorin; LC-MS/MS, liquid chromatography–tandem massspectrometry; P450, cytochrome P450; SOD, superoxide dismutase.

82

http://dmd.aspetjournals.org/content/suppl/2015/02/11/dmd.114.060558.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

All solvents and reagents were either analytical or high-performance liquidchromatography grade.

Time-, Concentration-, and NADPH-Dependent Inactivation of CYP2B6by IMP. The primary incubation mixtures contained CYP2B6 (0.1 mM) and IMPat concentrations of 0, 0.2, 0.5, 1, 1.5, and 2.5 mM in potassium phosphate buffer(pH 7.4). To determine the NADPH dependence in enzyme inactivation, IMP(1.5 mM) and CYP2B6 were incubated in the absence of NADPH as a negative

control. Except for the negative control, the primary reactions were initiated byaddition of NADPH, and aliquots of the primary mixtures (40 ml) were transferredat 0, 3, 6, and 9 minutes to the secondary incubation mixtures containing 100 mMbupropion and 0.45 mM NADPH in potassium phosphate buffer (pH 7.4) witha total reaction volume of 0.12 ml. After 30 minutes of incubation at 30�C, thereaction was stopped by adding equal volume ice-cold acetonitrile containingpropranolol as the internal standard, followed by centrifugation at 16,000 rpm for

Fig. 1. (A) Time- and concentration-dependent in-activation of CYP2B6 by IMP. Recombinant humanCYP2B6 was incubated with IMP at concentrations of0 (d), 0.2 (j), 0.5 (u), 1.0 (m), 1.5 (n), and 2.5 (s)mM in the presence of NADPH at 30�C for 0, 3, 6, and9 minutes. Aliquots of incubation mixtures were trans-ferred to the secondary incubation mixtures for the de-termination of residual enzymatic activity. The residualenzymatic activities at 0 minutes were normalized to100% at each concentration. (B) Wilson’s plot. Theobserved inactivation rate constant Kobs was calcu-lated from the slope of the regression lines shown in(A). (C) NADPH-dependent inactivation of CYP2B6by IMP. CYP2B6 was incubated with vehicle (d)and IMP (1.5 mM) in the presence (m) or absence(j) of NADPH. Data are mean 6 S.D. (n = 3).

Inhibition of CYP2B6 by Imperatorin 83

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

10 minutes. The supernatants were then collected and analyzed by liquidchromatography–tandem mass spectrometry (LC-MS/MS) as described below.

Substrate Protection. Substrate protection from IMP-induced inactivation ofCYP2B6 was determined by incorporating ticlopidine (50, 100, and 200 mM)with IMP (1.5 mM) in the primary reaction mixture. The reaction was initiatedwith NADPH, and aliquots (40 ml) of the mixtures containing 0.1 mM CYP2B6were transferred at 0, 3, and 9 minutes to the secondary incubation mixture forthe determination of bupropion hydroxylase activities of CYP2B6. Controlincubations were performed in the absence of IMP or ticlopidine in parallel.

Effects of GSH and Catalase/Superoxide Dismutase on the EnzymeInactivation. GSH (2.0 mM) was included in the primary reaction mixturecontaining CYP2B6 (0.1 mM), IMP (1.5 mM), and NADPH (1.0 mM).Aliquots (40 ml) were transferred to the secondary incubation mixture todetermine the remaining enzyme activities at various time points. An equalvolume of phosphate buffer was added instead of GSH solution as a controlsample. In a separate study, CYP2B6 was incubated with IMP and NADPH inthe presence or absence of a mixture of catalase and superoxide dismutase(SOD; 800 unit/ml for each). After incubation for 0, 3, and 9 minutes, theresidual activities were then detected as described below.

Partition Ratio. CYP2B6 (0.1 mM) was mixed with IMP at concentrationsof 0 (control), 0.2, 0.5, 1.0, 1.5, 2.5, 5.0, 7.5, and 10 mM. The primary reactionswere initiated by adding NADPH at a final concentration of 1.0 mM. Negativecontrol incubations were performed in the absence of NADPH. Afterincubation for 9 minutes in a water bath shaker at 30�C, aliquots (40 ml)were transferred from the primary reaction mixtures to the secondary reactionmixtures for measurement of CYP2B6 activities as follows.

Irreversibility of Inactivation. Primary incubations containing 2.5 mMIMP and 0.1 mM CYP2B6 were performed in the presence of NADPH at 30�C.The control incubations lacked IMP. Aliquots of the mixtures were withdrawnpartly at 0 minutes for enzymatic activity determination (nondialyzed sample).At 9-minute incubation, aliquots of the control and inactivated samples weredialyzed at 4�C against 1-l potassium phosphate buffer (pH 7.4, 3 � 2 hours).The dialyzed samples were brought to room temperature, and the enzymeactivities of the resulting samples were assessed as described below.

CYP2B6 Assay. CYP2B6 activity was assessed by measuring the formationof hydroxybupropion analyzed by LC-MS/MS. The LC-MS/MS systemconsists of AB Sciex Instruments 4000 QTRAP MS (Applied Biosystems,Foster City, CA) interfaced online with an ekspert ultraLC 100 system (AppliedBiosystems). The chromatographic separations were performed on an AccuoreC18 column (2.1 � 50 mm, 2.6 mm; Thermo Fisher, Pittsburgh, PA). Themobile phase consisted of 0.1% formic acid in acetonitrile (mobile phase A)and 0.1% formic acid in water (mobile phase B). The flow rate was maintainedat 0.3 ml/min and the column temperature was operated at 30�C. The gradientelution was set as follows: 0–1.0 minutes, 10% mobile phase A; 1.0–1.5minutes, 10–30% mobile phase A; 4.0–5.5 minutes, 30–10% mobile phase A;and 5.5–8.0 minutes, 10% mobile phase A. Quantification was performed bymultiple-reaction monitoring, and ion pairs of 256 → 238 for hydroxybupro-pion (product) and 260.7 → 116.3 for propranolol (internal standard) wereacquired in the positive mode.

Reactive Intermediate Trapping by GSH. IMP (100 mM) and GSH(1.0 mM) were incubated with rat liver microsomes (1.0 mg protein/ml) orindividual human recombinant P450 enzymes, including CYP1A2, CYP2A6,CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP3A4, or CYP3A5(0.1 mM for each) in the presence or absence of NADPH (1.0 mM) at 37�C. Tomaximize the quantity of the reactive intermediate to be formed, the enzymeincubations were conducted for 60 minutes. The resulting reactions werequenched by mixing with equal volume ice-cold acetonitrile containing hexylglutathione as the internal standard, followed by vortexing and centrifuging.The supernatants were harvested and evaporated to dryness. The residue wasreconstituted with 50 ml of 50% acetonitrile in water and analyzed by LC-MS/MS.The separation of GSH conjugates was achieved on an Accuore C18 col-umn (4.6 � 150 mm, 5 mm; Thermo Fisher). The solvent system consisted ofmobile phase A (acetonitrile with 0.1% formic acid) and mobile phase B(water with 0.1% formic acid), and the flow rate was set at 0.8 ml/min. Thegradient elution was set as follows: 0–2.0 minutes, 10% mobile phase A; 2.0–8.0 minutes, 10–90% mobile phase A; 8.0–10 minutes, 90% mobile phase A;10–13 minutes, 90–10% mobile phase A; and 13–15 minutes, 10% mobilephase A. The GSH conjugates were monitored in positive multiple-reaction

monitoring mode (594.0 → 465.0 for IMP-GSH conjugates; 392.2 → 246.3 forhexyl glutathione as the internal standard). The information-dependentacquisition standard was followed by triggering the enhanced product ion(EPI) scans. Information-dependent acquisition was used to initiate acquisitionof EPI spectra for ions exceeding 500 cps with exclusion of former target ionsafter three occurrences for 10 seconds. The EPI scan was run in positive modeat a scan range for product ions from m/z 50 to 650. The EPI scanningparameters are listed as follows: scan mode = profile; step size = 0.08 Da; andscan rate = 1000 Da/s (5-millisecond pause between mass ranges).

Synthesis of IMP-GSH Conjugates. IMP (2.5 mg) was dissolved inacetone (200 ml) and mixed with saturated sodium bicarbonate solution (40 ml)and oxone (6.5 mg). The mixture was stirred for 30 minutes at room temperature,followed by addition of GSH (35 mg). The mixture was further stirred for1 hour at room temperature and centrifuged. The supernatant was divided intotwo equal portions; one was directly submitted to LC-MS/MS analysis and theother portion of the supernatant was mixed with sodium borohydride (10 mg,NaBH4). The resulting mixture was gently vortexed for 5 minutes, followed byLC-MS/MS analysis.

Results

Time-, Concentration-, and NADPH-Dependent Inactivation ofCYP2B6 by IMP. The remaining CYP2B6 activity was monitored bymeasuring the amount of hydroxybupropion produced. As shown inFig. 1A, IMP produced a time- and concentration-dependent inhibitionon CYP2B6. The residual enzymatic activities at 0 minutes werenormalized to 100% at each concentration. The inactivation of CYP2B6increased progressively with increasing time and concentrations of IMPapplied. About 70% of activity of CYP2B6 was suppressed after itsincubation with 1.5 mM IMP for 9 minutes at 30�C. Nevertheless, noloss of enzyme activity was observed in the absence of IMP or NADPH(Fig. 1C). A double-reciprocal plot (Wilson’s plot) of the observed ratesof inactivation (Kobs) and IMP concentrations was used to calculate thekinetic constants KI and kinact (Fig. 1B), where KI was found to be 0.498mM, and kinact was found to be 0.079 min21.Substrate Protection. Substrate protection from IMP-dependent

inactivation of CYP2B6 was evaluated by including ticlopidine,a substrate of CYP2B6 (Talakad et al., 2011), in the primary incubationsthat contained CYP2B6 and IMP. The presence of ticlopidine was foundto slow down the CYP2B6 inactivation induced by IMP (Fig. 2). Inaddition, the protective effect of ticlopidine increased with the increasein the concentrations of ticlopidine applied.Effects of GSH and Catalase/Superoxide Dismutase on the Enzyme

Inactivation. To evaluate the protective effect of GSH against the en-zymatic inactivation induced by IMP, CYP2B6 inactivation incubation

Fig. 2. Substrate protection against inactivation of CYP2B6 by IMP. CYP2B6 wasincubated with vehicle (d) and IMP (1.5 mM) in the absence (u) or presence ofticlopidine at concentrations of 50 (♦), 100 (m), and 200 mM (j) of ticlopidine. Theresidual enzymatic activities at 0 minutes were normalized to 100% at eachconcentration. Data are mean 6 S.D. (n = 3).

84 Zheng et al.

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

was conducted in the presence or absence of GSH (2 mM), an elec-trophile trapping agent. After incubation for 9 minutes, the remainingCYP2B6 activities were 27.4 6 2.0% (with GSH) and 29.9 6 2.4%

(without GSH). This agent produced little protective effect on theenzyme from inactivation. Additionally, a mixture of catalase andSOD, scavengers of reactive oxygen species, showed little protectionagainst the inactivation of CYP2B6 by IMP, and the remaining en-zyme activities were 29.7 6 1.0% (with catalase/SOD) and 29.9 62.4% (without catalase/SOD) at 9 minutes.Partition Ratio. The partition ratio (P), a measure of the efficiency

of an enzyme inactivator, was estimated by a plot of the percentageremaining activity versus the IMP/CYP2B6 molar ratio (Fig. 3),according to a previously published method (Silverman, 1996). Theturnover number (P + 1) was about 5, and the extrapolated partitionratio of IMP was approximately 4 (Fig. 3).Irreversibility of Inactivation. CYP2B6 was incubated with IMP

(2.5 mM) in the presence of NADPH, followed by 6-hour dialysis (3 �2 hours). The remaining activities of the control samples and thosetreated with IMP were determined before and after dialysis. Only 10%of CYP2B6 activity was recovered after dialysis.Reactive Metabolite Trapping by GSH. GSH was included in the

IMP incubation systems containing rat liver microsomes or recombi-nant P450 enzymes to trap the reactive metabolites. The samplesobtained from the incubations in the presence or absence of NADPHwere analyzed by LC-MS/MS. A peak with protonated molecule ion

Fig. 3. Partition ratio determination for CYP2B6 inactivation by IMP. CYP2B6 wasincubated with IMP at various concentrations. The extrapolated P + 1 was deter-mined from the point of intersection to the abscissa. Data are mean 6 S.D. (n = 3).

Fig. 4. (A) The MS/MS spectrum, chemical structure, and origins of the characteristic fragment ions for IMP. (B) MS/MS spectrum of IMP-GSH conjugate generated inmicrosomal incubation. (C) MS/MS spectrum of synthetic IMP-GSH conjugate. (D) Extracted ion (m/z 594 → 465) chromatograms obtained from LC QTRAP MS analysisof microsomal incubations containing IMP (solid line) and chemical oxidation of IMP trapped by GSH (dashed line).

Inhibition of CYP2B6 by Imperatorin 85

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

[M + H]+ at m/z 594 at retention time of 6.8 minutes was observed(Fig. 4D); however, no such peak was found in the control sample(absence of NADPH). The MS/MS spectrum of the conjugate wasobtained by multiple-reaction monitoring/EPI scanning (ion transitionm/z 594/465). The spectrum showed the indicative characteristicfragment ions associated with the cleavage of the GSH moiety (Fig.4B). The product ions at m/z 465 and 519 were derived from theneutral loss of g-glutamyl portion (2129 Da) and glycine portion(275 Da) from parent ion m/z 594, respectively.To verify the characterization, IMP was chemically oxidized by

oxone in acetone, followed by reaction with GSH. As expected, theproduct responsible for the IMP-derived GSH conjugate showed thesame chromatographic behaviors and identical mass spectrum (Fig.4C) as that of the conjugate generated in the microsomal reactions(Fig. 4B). Unfortunately, we were unable to obtain enough of the

product for NMR characterization due to poor reaction yield.Furthermore, the other portion of the reaction solution was mixedwith sodium borohydride for reduction and then analyzed by LC-MS/MS.A new peak at retention time of 6.7 minutes was observed (Fig.5A), and the [M + H]+ was m/z 596.2, two Daltons higher than that ofthe conjugate observed before the reduction by sodium borohydride(Fig. 5B).P450 Enzymes Responsible for IMP Bioactivation. To determine

which P450 enzymes preferentially catalyze the oxidation of IMP, IMPwas incubated with individual recombinant human P450 enzymes,including CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19,CYP2D6, CYP2E1, CYP3A4, or CYP3A5 (0.1 mM for each) in thepresence of GSH as the trapping agent. The IMP-GSH conjugate wasassessed in the incubation with CYP1A2, CYP2A6, CYP2B6,CYP2D6, CYP2E1, CYP3A4, or CYP3A5. Apparently, little IMP-

Fig. 5. (A) Extracted ion (m/z 596 → 467) chromatogram obtained from LC QTRAP MS analysis of the chemical oxidation reaction followed by reaction with NaBH4. (B)MS/MS spectrum of the reduced product.

86 Zheng et al.

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

GSH adduct was found in the incubations with CYP2C9 or CYP2C19(Fig. 6). The experiments demonstrated that multiple P450 enzymes areinvolved in bioactivation of IMP, especially for CYP1A2.

Discussion

Our kinetic study clearly demonstrated that IMP produced a time-dependent inhibition of CYP2B6. The observed enzyme inactivationwas also found to be concentration dependent, and it reached about 70%inactivation at the concentration of 2.5 mM. CYP2B6 inactivationinduced by IMP was not observed unless NADPH was present in theprimary incubation mixtures. This indicates that IMP itself is not aninactivator of CYP2B6 and that the enzyme inactivation by IMP needsbiotransformation with assistance from P450 cofactor NADPH.Taken together, these results provided strong evidence that IMP is amechanism-based inactivator of CYP2B6.Little protection by catalase and SOD against enzyme inactivation

by IMP was observed. Catalase and SOD are known as enzymes

quenching hydrogen peroxide, superoxide anion, and other reactiveoxygen species that are possible agents to inactivate enzymes (Klaassenet al., 1986). The observed lacking of protection effect by catalase andSOD indicates that reactive oxygen species may not play a significantrole in CYP2B6 inactivation induced by IMP. Additionally, GSH,a nucleophilic agent, did not show protective effect on CYP2B6inactivation by IMP. This implies in part that CYP2B6 is covalentlymodified by electrophilic metabolites of IMP before escaping from theactive site. The protective effect of ticlopidine on CYP2B6 inactivationobserved in the competition experiment indicates the competitionbetween ticlopidine and IMP to reach the active site of the enzyme,which slows down the generation of the reactive metabolites of IMP.This critical finding provided evidence that bioactivation of IMP tookplace in the active site of CYP2B6.The P value is defined as the number of product molecules

produced per inactivation event and reflects the efficiency of theinactivator. The P values reported in the literature for mechanism-based inactivators of P450 enzymes vary from 2 (very highly efficientinactivators) to .1000 (inefficient) (Kent et al., 2001). Bergamottin, afuranocoumarin compound, was an efficient mechanism-based inacti-vator of CYP2B6 and CYP3A5 with partition ratios of approximately2 for CYP2B6 and 20 for CYP3A5 (Lin et al., 2005). Therefore, IMP(P = 4) may be ranked as an efficient inactivator of CYP2B6.At time 0 minutes, a decrease in enzyme activity was observed with

the increase of applied IMP concentrations. The observed enzymeinhibition may result from reversible inhibition at the beginning of theincubation.Generation of reactive metabolites is a key step for mechanism-

based enzyme inactivation. We hypothesized that IMP is metabolized tofuranoepoxide intermediate 2 (Scheme 1) and/or g-ketoenal in-termediate 3, and that g-ketoenal is also suggested to be generatedby rearrangement of the furanoepoxide. To probe the hypothesis, wesearched for the electrophilic intermediates using GSH as a trappingagent. An IMP-derived GSH conjugate in microsomal incubations

Fig. 6. Human P450 enzymes involved in the formation of IMP-conjugate. IMP wasincubated with individual recombinant human P450 enzymes in the presence ofNADPH and GSH. Data are mean 6 S.D. (n = 3).

Scheme 1. Proposed pathways for the formation of reactive intermediate(s) and GSH adducts during the metabolism of IMP.

Inhibition of CYP2B6 by Imperatorin 87

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from

with IMP was detected by LC-MS/MS (Fig. 4D). The detectedprotonated molecular ion [M + H]+ matched the molecular weight ofGSH conjugates 4–6 (Scheme 1). The MS/MS spectrum of themetabolite showed the indicative characteristic fragments of GSHand some other fragments valuable for metabolite identification (Fig.4B). A fragment at m/z 162 responsible for the molecular formula ofC9H6O3

+ was detected, indicating that the five-member ring fusedwith the coumarin ring was opened (Scheme 1). This allowed us toexclude the formation of GSH conjugate 4, since the ring opening ofconjugate 4 would produce a thio-ester. Only GSH conjugate 6 hadthe ring opening structure, and conjugate 5 could be spontaneouslyconverted to conjugate 6. To further characterize conjugate 6, wechemically reduced the conjugate with NaBH4. The reductionreaction afforded a new product with retention time at 6.7 minutesand [M + H]+ at m/z 596.2 (Fig. 5A), 2.0 Da higher than that of theone detected before the reduction reaction. The MS/MS spectrum ofthe reduced GSH conjugate demonstrated that the fragment at m/z162 was retained, indicating that no reduction took place in thecoumarin ring. In addition, the original IMP-GSH conjugate showeda fragment at m/z 177 that resulted from the elemental compositionof C10H9O3

+ of the conjugate losing the side chain, GSH, and CO2.In other words, the fragment of m/z 177 contained the aldehydegroup (Fig. 4B). However, no such fragment was observed in theMS/MS spectrum of the reduced IMP-GSH conjugate (Fig. 5B).Instead, the reduced IMP-GSH conjugate showed a fragment of m/z179, 2 Da higher than that (m/z 177) observed in the MS/MSspectrum of the original IMP-GSH conjugate (Fig. 5B). Thisprovided evidence that the reduction took place on the aldehyde noton the lactone group. An additional experiment showed that IMPwas resistant to reduction by NaBH4 (data not shown). We anticipatethat the NaBH4-based reduction of conjugate 6 gave conjugate 7.The observed conjugate 7 indicates that conjugate 6 was the primaryGSH conjugate resulting from metabolic activation of IMP.However, we are unable to tell whether conjugate 6 came fromintermediate 2 or 3, since conjugate 5 may be spontaneously convertedto conjugate 6, and vice versa. In other words, whether intermediate 2or 3 was the primary metabolic intermediate remains unknown. Wespeculate that IMP was oxidized to the corresponding g-ketoenal/epoxide (on the furan rings), which chemically attacked the hostenzyme to form covalent binding, and the enzyme modificationinitiated the process of the enzyme inactivation.The bioactivation studies with individual recombinant enzymes

demonstrated that multiple P450 enzymes catalyzed the metabolism ofIMP to the reactive intermediate (Fig. 6). CYP1A2 was found to be themajor enzyme responsible for the bioactivation of IMP. Whether IMP isa mechanism-based inactivator of CYP1A2 is under investigation. Insummary, we have shown that IMP is characterized as a mechanism-based inactivator of CYP2B6. A GSH conjugate derived from a g-ketoenalintermediate was identified in microsomal incubations with IMPtrapped with GSH.

Authorship ContributionsParticipated in research design: J. Zheng, Peng.Conducted experiments: L. Zheng, Cao, Lu, Ji.Performed data analysis: L. Zheng.Wrote or contributed to the writing of the manuscript: J. Zheng, L. Zheng.

References

Baek NI, Ahn EM, Kim HY, and Park YD (2000) Furanocoumarins from the root of Angelicadahurica. Arch Pharm Res 23:467–470.

Baumgart A, Schmidt M, Schmitz HJ, and Schrenk D (2005) Natural furocoumarins as in-ducers and inhibitors of cytochrome P450 1A1 in rat hepatocytes. Biochem Pharmacol 69:657–667.

Cai Y, Bennett D, Nair RV, Ceska O, Ashwood-Smith MJ, and DiGiovanni J (1993) Inhibitionand inactivation of murine hepatic ethoxy- and pentoxyresorufin O-dealkylase by naturallyoccurring coumarins. Chem Res Toxicol 6:872–879.

Edwards DJ, Bellevue FH, 3rd, and Woster PM (1996) Identification of 69,79-dihydroxybergamottin,a cytochrome P450 inhibitor, in grapefruit juice. Drug Metab Dispos 24:1287–1290.

Faucette SR, Hawke RL, Lecluyse EL, Shord SS, Yan B, Laethem RM, and Lindley CM (2000)Validation of bupropion hydroxylation as a selective marker of human cytochrome P450 2B6catalytic activity. Drug Metab Dispos 28:1222–1230.

García-Argáez AN, Ramírez Apan TO, Parra Delgado H, Velázquez G, and Martínez-Vázquez M(2000) Anti-inflammatory activity of coumarins from Decatropis bicolor on TPA ear micemodel. Planta Med 66:279–281.

Gerber JG, Rhodes RJ, and Gal J (2004) Stereoselective metabolism of methadone N-demethylationby cytochrome P4502B6 and 2C19. Chirality 16:36–44.

He JY, Zhang W, He LC, and Cao YX (2007) Imperatorin induces vasodilatation possibly viainhibiting voltage dependent calcium channel and receptor-mediated Ca2+ influx and release.Eur J Pharmacol 573:170–175.

Huang Z, Roy P, and Waxman DJ (2000) Role of human liver microsomal CYP3A4 and CYP2B6in catalyzing N-dechloroethylation of cyclophosphamide and ifosfamide. Biochem Pharmacol59:961–972.

Kent UM, Juschyshyn MI, and Hollenberg PF (2001) Mechanism-based inactivators as probes ofcytochrome P450 structure and function. Curr Drug Metab 2:215–243.

Kim YK, Kim YS, and Ryu SY (2007) Antiproliferative effect of furanocoumarins from the rootof Angelica dahurica on cultured human tumor cell lines. Phytother Res 21:288–290.

Klaassen CD, Amdur MO, and Doull J (1986) Biotransformation of xenobiotics, in Casarett andDoull’s Toxicology, the Basic Science of Poisons (Klaassen CD ed) pp 113–186, McGraw-Hill,New York.

Koenigs LL and Trager WF (1998) Mechanism-based inactivation of cytochrome P450 2B1 by8-methoxypsoralen and several other furanocoumarins. Biochemistry 37:13184–13193.

Lia HB and Chen F (2004) Preparative isolation and purification of bergapten and imperatorinfrom the medicinal plant Cnidium monnieri using high-speed counter-current chromatographyby stepwise increasing the flow-rate of the mobile phase. J Chromatogr A 1061:51–54.

Lin G, Tang J, Liu XQ, Jiang Y, and Zheng J (2007) Deacetylclivorine: A gender-selectivemetabolite of clivorine formed in female Sprague-Dawley rat liver microsomes. Drug MetabDispos 35:607–613.

Lin HL, Kent UM, and Hollenberg PF (2005) The grapefruit juice effect is not limited tocytochrome P450 (P450) 3A4: Evidence for bergamottin-dependent inactivation, heme de-struction, and covalent binding to protein in P450s 2B6 and 3A5. J Pharmacol Exp Ther 313:154–164.

Marumoto S and Miyazawa M (2010) b-secretase inhibitory effects of furanocoumarins from theroot of Angelica dahurica. Phytother Res 24:510–513.

Miksys S and Tyndale RF (2004) The unique regulation of brain cytochrome P450 2 (CYP2)family enzymes by drugs and genetics. Drug Metab Rev 36:313–333.

Rosselli S, Maggio A, Bellone G, Formisano C, Basile A, Cicala C, Alfieri A, Mascolo N,and Bruno M (2007) Antibacterial and anticoagulant activities of coumarins isolated from theflowers of Magydaris tomentosa. Planta Med 73:116–120.

Silverman RB (1996) Mechanism-based enzyme inactivation, in Contemporary Enzyme Kineticsand Mechanisms (Purich DL ed) pp 291–335, Academic Press, San Diego, CA.

Stresser DM and Kupfer D (1999) Monospecific antipeptide antibody to cytochrome P-450 2B6.Drug Metab Dispos 27:517–525.

Talakad JC, Shah MB, Walker GS, Xiang C, Halpert JR, and Dalvie D (2011) Comparison of invitro metabolism of ticlopidine by human cytochrome P450 2B6 and rabbit cytochrome P4502B4. Drug Metab Dispos 39:539–550.

Tang CH, Yang RS, Chien MY, Chen CC, and Fu WM (2008) Enhancement of bone morpho-genetic protein-2 expression and bone formation by coumarin derivatives via p38 and ERK-dependent pathway in osteoblasts. Eur J Pharmacol 579:40–49.

Thum T and Borlak J (2000) Gene expression in distinct regions of the heart. Lancet 355:979–983.

Ward BA, Gorski JC, Jones DR, Hall SD, Flockhart DA, and Desta Z (2003) The cytochromeP450 2B6 (CYP2B6) is the main catalyst of efavirenz primary and secondary metabolism:Implication for HIV/AIDS therapy and utility of efavirenz as a substrate marker of CYP2B6catalytic activity. J Pharmacol Exp Ther 306:287–300.

Zhang QY, Qin LP, Huang BK, Zheng HC, Wang Y, Wang H, and Chen L (2003) Study oneffects of total coumarines from the fruits of Cnidium monnieri on osteoporosis in ovariec-tomized rats. Chin Pharm J 38:101–103.

Zhang Y, Cao Y, Zhan Y, Duan H, and He L (2010) Furanocoumarins-imperatorin inhibitsmyocardial hypertrophy both in vitro and in vivo. Fitoterapia 81:1188–1195.

Zhuang XM, Zhong YH, Xiao WB, Li H, and Lu C (2013) Identification and characterization ofpsoralen and isopsoralen as potent CYP1A2 reversible and time-dependent inhibitors in humanand rat preclinical studies. Drug Metab Dispos 41:1914–1922.

Address correspondence to: Dr. Jiang Zheng, Center for DevelopmentalTherapeutics, Seattle Children’s Research Institute, Division of Gastroenterologyand Hepatology, Department of Pediatrics, University of Washington, Seattle, WA98101, or Key Laboratory of Structure-Based Drug Design & Discovery of Ministryof Education, Shenyang Pharmaceutical University, Shenyang, Liaoning, 110016,P.R. China. E-mail: jiang.zheng@seattlechildrens.org; or Dr. Ying Peng, Schoolof Pharmacy, Shenyang Pharmaceutical University, Shenyang, Liaoning, 110016,P. R. China. E-mail: yingpeng1999@163.com

88 Zheng et al.

at ASPE

T Journals on O

ctober 31, 2020dm

d.aspetjournals.orgD

ownloaded from