- 1 - Systemic andcerebral exposure toand pharmacokinetics offlavonols andterpene lactones after dosingstandardizedGinkgobilobaleafextracts torats via differentroutes of administration FengChen, Li Li, FangXu, Yan Sun, Feifei Du, Xutao Ma, Chenchun Zhong, XiuxueLi, FengqingWang, Nating ZhangandChuan Li British Journal ofPharmacology -Supplemental Methods - GBE50and ShuXueNing injection GBE50, astandardized proprietaryextract ofG. biloba leaveswith aChineseSFDA ratification no.Z20000049for themanufacturer Shanghai XingLingSci. &Tech. Pharmaceuticals(Shanghai, China), wasa 200:3 solid extract, i.e., 200 kgofdried leaveswere used to make 3 kgof the final solid extract. Preparation ofGBE50 required a six-step extraction/purification process, including 60% aqueous ethanol-based extraction, concentration under reduced pressure, precipitationwith water, chromatographyon aporouspolymer column and thenonapolyamide column, further elimination ofginkgolicacidswith cyclohexane, and reduction bydry-sprayingto dryness(Xieet al.,2000).ShuXueNinginjectionwasasolutionofGBE50 availableasasterile,nonpyrogenicparenteral dosage form for intravenous or intramuscular injection, which was manufactured by Shanghai Asia Pioneer Pharmaceuticals (Shanghai, China) with a Chinese SFDA ratification no. Z20043734. Each milliliter of ShuXueNinginjectionwasprepared from3.5 mgofGBE50. For preparationofShuXueNinginjection, itrequires a multiple-step process including dissolution of GBE50, boiling, filtration, treatment with weakly basic anion-exchangeresins, ultrafiltration,sterilization, dilution with glucoseinjection and 30%ethanol,adjustment to pH 5.4–5.6, ultrafiltration, sterilization, etc. Table 1 shows the content levels of flavonols and terpene lactones present inGBE50 andShuXueNinginjection. Table 1 ContentlevelsofginkgocompoundsinShuXueNinginjectionorGBE50. Compound ShuXueNing injection (µmol∙mL−1) GBE50 (µmol∙10mg−1) Terpenelactone Bilobalide(1) 0.0002 0.954 GinkgolideA (2) 0.289 0.551 GinkgolideB (3) 0.100 0.325 GinkgolideC(4) 0.184 0.316 GinkgolideJ(5) 0.091 0.105 Flavonol Quercetin3-O-dirhamnosyl-glucoside(10) 0.072 0.270 Quercetin3-O-(p-coumaroyl)-glucosyl-O-rhamnoside(11) 0.096 0.617 Quercetin3-O-rhamnosyl-O-glucoside(12) 0.174 0.660 Quercetin3-O-glucosyl-O-rhamnoside(13) 0.065 0.354 Quercetin(16) 0.003 0.088 t-Quercetin 0.397 2.149 Kaempferol3-O-dirhamnosyl-glucoside(19) 0.089 0.425 Kaempferol3-O-(p-coumaroyl)-glucosyl-O-rhamnoside(20) 0.080 0.707 Kaempferol3-O-rhamnosyl-O-glucoside(21) 0.162 0.655 Kaempferol3-O-glucosyl-O-rhamnoside(22) 0.072 0.474 Kaempferol(26) 0.004 0.095 t-Kaempferol 0.441 2.378 Isorhamnetin3-O-dirhamnosyl-glucoside(27) 0.002 0.018 Isorhamnetin3-O-rhamnosyl-O-glucoside(29) 0.088 0.559 Isorhamnetin3-O-glucosyl-O-rhamnoside(30) 0.0003 0.004 Isorhamnetin(32) 0.001 0.014 t-Isorhamnetin 0.127 0.665 TheginkgocompoundsaregivenIDnumbers inparentheses,whichareconsistentwith thoseforGBE50described inour earlier publication (Li et al., 2012). Hydrochloric acid-based hydrolysis was applied to the sample to convert the flavonol glycosides present into their aglycone forms before measurement of the content levels of t-quercetin, t-kaempferol, and t-isorhamnetininShuXueNinginjectionandGBE50. Experimentalanimals All rat studieswereconductedin compliancewith theGuidancefor Ethical Treatment ofLaboratoryAnimals(The MinistryofScienceandTechnologyofChina,2006; www.most.gov.cn/fggw/zfwj/zfwj2006)andtheexperimental protocolswereapproved bytheInstitutional AnimalCareand Use Committeeat theShanghai InstituteofMateria Medica (Shanghai,China). Male Sprague-Dawley rats (260–300g; Sino-British SIPPR/BKLaboratoryAnimal, - 2 - Shanghai, China) were maintained in individually ventilated cages (42 × 24 × 24 cm3, three rats per cage; Suzhou-Fengshi, Suzhou, China) at 20–24°C and relativehumidity(30%–70%) with a12 h light/darkcycle. All rodentshad freeaccessto filtered tap water. Theratswere provided commercial rat chowad libitum, except for the overnight period beforedosing. Ratswereacclimated to thefacilitiesandenvironment for 7daysbeforeuse. For blood sampling, rats were anaesthetized with pentobarbital (40mg·kg−1, i.p.), and then polyurethane cannulae (BC-2P; AccessTechnologies, Skokie, IL, USA; pre-filled with 25 U·mL−1 heparinized saline) wereinserted into thefemoral arteriesafter an incision oftheleft groins. After surgery, thefemoral artery-cannulated (FAC) ratswere housed individually and allowed to regain the preoperative body weights before experimental use. Unlimited accessto foodandwater waspermitted duringtherecoveryperiod. For bilesampling, thebileductsandduodena ofratsunder pentobarbital anaesthesiawereexposed viaamidlinelaparotomyincision. Thebileduct ofeach rat wascannulated with two connected polyurethanecathetersfor bile collection and duodenuminfusion. Duringthe rat recoveryperiod (4–5 days), thedrinkingwater wassupplemented with 5%glucose, 0.9%sodiumchlorideand 0.5%potassiumchlorideand thetwo cannulaewereconnected with aunion to re-infusebileinto theduodenum. During the bile collection period, the two cannulae of thebile duct-cannulated (BDC) rat were detached and a sodiumtaurocholatesolution (pH7.4) wasinfused into theduodenumat 1 mL·h−1. All used ratswereeuthanatized with CO2 gas. PlasmaPK studies TheFAC ratswere randomlydivided into threegroupsof four ratseach toassess the systemicexposure to and plasma pharmacokinetics of flavonols and terpene lactones after a 15 min i.v. infusion dose of ShuXueNing injection at 1, 2 or 4 mL·kg−1. Serial blood samples(~110 µL; 0, 5, 15, 30 min and 1, 2, 4, 6, 8, 11 and 24 h) werecollected into heparinized tubesviafemoral arterycannula. Theblood sampleswerecentrifuged toobtain the plasmafractionsthat werefrozen at −70°C untilanalysis. A similar experiment wasalso performed onFAC rats that received asinglep.o. doseofGBE50 at 10, 30or 90 mg·kg−1 viagavage. Beforeuse, GBE50 wassuspended in 0.5%w/vsodiumcarboxymethylcellulose. ShuXueNinginjectionwasgiven to ratsthroughtherat tailvein for i.v. infusion. In addition, a 15 min i.v. infusion dose of bilobalide at 1 mg·kg−1 (0.5 mg·mL−1; dissolved in an isotonic glucose injection that contained 0.5 mg·mL−1 Cremophor EL) was given to FAC rats and the blood sampleswerecollected accordingto theprecedingtimeschedule. In addition, asubchronicPKstudywasimplemented for seven consecutivedayswith four FAC ratsreceived ShuXueNing injection at1mL·kg−1·day−1 and other fourFAC rats received GBE50 at 30mg·kg−1·day−1. Serial blood sampleswerecollected, accordingto theprecedingtimeschedule, on days1 and 7 and werecentrifuged to yield plasmafractions. Brain microdialysis Abrain microdialysisstudywasperformed insurgically-modified consciousratsaccordingtoaprotocolmodified fromour earlier method (Sun et al., 2009). In brief, FAC ratswereanaesthetized with pentobarbital (50 mg·kg−1, i.p.) and each rodent wasplaced in astereotaxicframewith incisor bar to achieveahorizontal skull. ACMA/150 temperaturecontroller (CMA/Microdialysis, Solna, Sweden) wasused toprovidesupplemental heat to therat. A CMA/11 guidecannulaholdingashaft dummywasimplanted andaimed at thehippocampus[AP −5.8,ML+5.0, DV −7.0 mm; accordingtotheatlasofrat brain (Paxinosand Watson, 1998)], and wasthen cemented to theskull surface. The surgically-modified rat was housed singly and was allowed to recover postoperatively before use. Beforemicrodialysis, theshaft dummywas replaced byaCMA/11 microdialysisprobe(cuprophanemembraneof 4 mm in length, 6 kD cut-off). The probewas perfused with an artificial cerebrospinal fluid (125 mM sodium chloride, 0.5mM sodiumdihydrogen phosphate, 2.5mM disodiumhydrogen phosphate,1.2mM calciumchloride, 2.5mM potassium chloride and 1.0mM magnesium chloride, pH 7.4) at 2 µL·min−1. The rat was allowed to equilibratefor 1 h priortoinitiationofsamplecollection. After theblankcontroldialysatesamplewascollected, therodentreceived a15min i.v. infusiondoseofShuXueNinginjection at 4 mL·kg−1 or ap.o. doseofGBE50 at 90 mg·kg−1 followed bya6 h dialysatesampling. Thedialysatesampleswerecollected in 10 min intervalsusinga CMA/142 microfraction collector, and a6min delaywas incorporated into the sampling to compensate for the dead volumebetween theactivemembraneand thesamplecollectionoutlet.Meanwhile, bloodsamplingwasalso collected at 0, 5, 15 and 30 min and 1, 2, 4 and 6 h after dosing.Theconcentrationsofginkgo terpenelactonesin brain extracellular fluid (bECF) were calculated from that in the corresponding dialysate sample using the followingequation: CbECF=Cd/Rinvivo (1) whereCbECFisthebECFconcentration;Cd isthemeasured concentration in thedialysatesample. Rinvivo isthein vivo recoverybyretrodialysis, which was16%, 13%, 11%, 9%and 11%for bilobalide, ginkgolidesA, B, C andJ respectively. Excretion studies Ratshad not undergoneanysurgerywererandomlyassigned to two groups(four rats/group) and housed singlyin Nalgene metabolic cages (Nalgene Co., Rochester, NY, USA). The urine and feces collection tubes of the metaboliccages were frozen at −15°C duringsamplecollection. Urineand feces sampleswere collected before and 0 to 8, 8 to 24 and 24 to 32h and weighed after a15 min i.v. infusion doseofShuXueNinginjection at 4 mL·kg−1 or ap.o.doseofGBE50 at90 mg·kg−1. Thefaeceswerehomogenized in ninefoldvolumesof ice-cold saline. Thefecal homogenatesand urinesampleswerestored at −70°C until analysis. ThreeBDC ratsreceived a15min i.v.infusion doseofShuXueNinginjectionat 4 mL·kg−1, whileother three - 3 - BDC ratsreceived ap.o.doseofGBE50 at 90 mg·kg−1. Bilesampleswerecollected beforeand 0 to 10, 10 to 20, 20 to 40, 40to50 min and0.8 to1.2,1.2to1.8, 1.8 to 2.2, 2.2to3, 3to3.8,3.8 to 4.2, 4.2to5.8,5.8 to 6.2,6.2 to 7.8,7.8 to 8.2, 8.2 to 10.8,10.8to 11.2, 11.2to23.8and23.8to 24.2h after dosingand weighed. Thebilesamples werestored at −70°C until analysis. Tissuedistribution study Ratsunder isofluraneanaesthesiawerekilled bybleedingfromtheabdominal aorta(~10 mL ofblood) at 0, 15 (i.v. only), 30min, 1 (i.v.only), 2 (p.o. only), 4and8h(p.o. only) after a15 min i.v. infusion doseofShuXueNing injection at 4 mL·kg−1 or ap.o. doseofGBE50 at 90 mg·kg−1. Three ratswereused per timepoint.Theheart, lungs, brain,liver and kidneyswereexcised and then rinsedinice-cold salinebeforegentlyblottingon absorbent paper.The tissue sampleswereweighed and homogenized in fourfold volumesof ice-cold saline. The resulting homogenateswerestored at −70°C until analysis. Analysisofginkgoflavonolsand terpenelactonesin biologicalsamples An LC-MS/MS system, consisted of a Thermo Fisher TSQQuantummass spectrometer (San Jose, CA, USA) interfaced viaan electrosprayionization probewith an Agilent 1100 liquidchromatograph (Waldbronn,Germany), was used to measure concentrations of the ginkgo flavonols and terpene lactones in various hydrochloric acid-hydrolyzed rat samples. Analytical assayswereperformed asdescribed byZhao etal.(2008). In brief, therat plasma, excreta, dialysateor tissuehomogenatesamples(20 µL) wereadded with 10 mM aqueousascorbicacid solution(freshlyprepared) and then treated with 20 µL of4 M hydrochloricacid at 80ºCfor 30min to releasethe flavonol aglycones from their glycosidesand conjugated metabolites. Accordingly, themeasured flavonol levels are expressed as concentrations of total quercetin (t-quercetin), total kaempferol (t-kaempferol) and total isorhamnetin (t-isorhamnetin). After the treatments, the samples were extracted with ethyl acetate. The chromatographic separation was achieved using a Phenomenex Gemini 5µm C18 column (50 × 2.0 mm i.d.; Torrance, CA, USA) with awater-acetonitrilemobilephasemodified with formicacid at 0.2 mM and delivered at 0.3mL·min−1. The gradient programmewas 0–1.5 min at 5% (v/v) acetonitrile, 1.5–1.6 min from 5% to 63% acetonitrile, 1.6–2.8 min at 63% acetonitrile, 2.8–2.9 min from 63% to 5% acetonitrile and 2.9–7.0 min at 5% acetonitrile.TheMS/MS measurement wasperformed inthenegativeionmodeandtheprecursor-to-product ion pairsused for selected reaction monitoringoft-quercetin, t-kaempferol, t-isorhamnetin, bilobalide, ginkgolidesA, B, C and Jwerem/z301→151 (theoptimal collision energy, 28 V),285→187 (37 V), 315→300 (28 V), 325→163 (26V), 453→351(27 V),423→367(20 V),439→383 (19V) and 469→423 (18 V),respectively, with ascan time of0.1 sfor each ion pair. OnlyLC eluent flowovera period of4.5 to 6.3 min wasintroduced to theion sourcefor data acquisition. Matrix-matched calibration curves (0.8–2000 ng·mL−1) were constructed for the eight analytes usingweighted (1/X) linear regressionsoftheanalyteresponse(peakarea; Y) against thecorrespondingnominal plasma concentrationsofanalytes (X, ng·mL−1), which showed good linearity (r2 >0.99). Assayvalidation was carried out according to the US Food and Drug Administration guidance on bioanalytical validation (2001; www.fda.gov/cder/guidance/index.htm)to demonstratethat thebioanalytical assayswerereliablefor theintended applications. To assessbrain penetration oftheginkgo compounds, therat braindialysatesampleswerecentrifuged for 10 min at 16 060×g.A high-speed, rapid ultrafiltrationmethoddescribedbyGuo etal. (2006)wasused to isolate unboundginkgo compoundsin plasmafor measurement oftheunboundplasmaconcentrationandplasmaprotein binding. Theglycosides and metabolites of ginkgo flavonols in rat plasma, bile, and urinewere also measured and profiled usingan AB-SCIEX API4000 QTrap massspectrometer (Toronto,Canada) interfaced viaaTurboVion sourcewith aWatersAcquityUPLC separation module(Milford, MA, USA). Thesamplescollected fromtherats received a15 min i.v. infusion doseofShuXueNinginjection at4 mL·kg−1 or ap.o.doseofGBE50at 90 mg·kg−1 and pooled accordingto thecollectingtime, i.e., 15 min or 6 h after dosingfor theplasmasamplesand 0–8 h after dosingfor theurinesampleor thebilesample. Theserat samples(100 µL; pretreated without hydrochloricacid) were precipitated with 300 µL of methanol. After centrifugation, the resulting supernatants were reduced to drynessusingacentrifugal concentrator under reduced pressure. Theresidueswerereconstitutedin 40 µL of50% methanol and were centrifuged before UPLC-MS/MS analysis. The chromatographic separation was achieved using aWaters 1.7 µmBEHC18 column (100 ×2.1 mm i.d.; Dublin, Ireland). A water-methanol mobile phase modified with 1 mM formicacid wasdelivered at 0.4 mL·min−1 and 45°C. A binarygradient methodwasused, i.e., 0–8 min from 2% to 80% methanol, 8–9 min at 80% methanol, 9–12 min at 2% methanol. The MS/MS measurement wasperformed in thenegative ion mode, and theprecursor-to-product ion pairsused for selected reaction monitoring of quercetin 3-O-dirhamnosyl-glucoside (10)/quercetin 3-O-(p-coumaroyl)-glucosyl-O- rhamnoside(11), glucuronidated quercetin 3-O-(p-coumaroyl)-glucosyl-O-rhamnoside(M10G), sulfated quercetin 3-O-(p-coumaroyl)-glucosyl-O-rhamnosides (M10S-1/M10S-2), glucuronidated quercetin 3-O-(p-coumaroyl)-glucosyl-O-rhamnosidesulfate(M10G-S), quercetin 3-O-rhamnosyl-O-glucoside(12)/quercetin 3-O-glucosyl-O-rhamnoside (13), glucuronidated quercetin 3-O-rhamnosyl-O-glucosides/quercetin 3-O-glucosyl-O-rhamnosides (M12/13G-1–M12/13G-4), sulfated quercetin 3-O-rhamnosyl-O-glucosides/quercetin 3-O-glucosyl-O-rhamnosides (M12/13S-1–M12/13S-4), glucuronidated quercetin 3-O-rhamnosyl-O- glucosides/quercetin 3-O-glucosyl-O-rhamnoside sulfates (M12/13G-S-1–M12/13G-S-3), quercetin (16), quercetin 3-O-glucuronide (M16G-1), quercetin glucuronides (M16G-2–M16G-4), quercetin sulfates (M16S-1–M16S-3), glucuronidated quercetin sulfates (M16G-S-1–M16G-S-3), quercetin diglucuronide (M162G), quercetin disulfate (M162S), sulfated quercetin diglucuronide (M162G-S), glucuronidated quercetin disulfate (M16G-2S), kaempferol - 4 - 3-O-dirhamnosyl-glucoside (19)/kaempferol 3-O-(p-coumaroyl)-glucosyl-O-rhamnoside (20), glucuronidated kaempferol 3-O-(p-coumaroyl)-glucosyl-O-rhamnoside (M20G), sulfated kaempferol 3-O-(p-coumaroyl)-glucosyl-O-rhamnosides (M20S-1/M20S-2), glucuronidated kaempferol 3-O-(p-coumaroyl)-glucosyl-O-rhamnosidesulfates(M20G-S-1–M20G-S-3), kaempferol 3-O-rhamnosyl-O-glucoside (21)/kaempferol 3-O-glucosyl-O-rhamnoside (22), glucuronidated kaempferol 3-O-rhamnosyl-O-glucosides/kaempferol 3-O-glucosyl-O-rhamnosides (M21/22G-1–M21/22G-5), sulfated kaempferol 3-O-rhamnosyl-O-glucosides/kaempferol 3-O-glucosyl-O-rhamnosides (M21/22S-1–M21/22S-4), glucuronidated kaempferol 3-O-rhamnosyl-O-glucosides/kaempferol 3-O-glucosyl-O-rhamnoside sulfates (M21/22G-S-1–M21/22G-S-3), kaempferol (26), kaempferol 3-O-glucuronide (M26G-1), kaempferol glucuronides (M26G-2/M26G-3), kaempferol sulfates(M26S-1/M26S-2), glucuronidated kaempferol sulfate(M26G-S), kaempferol diglucuronides (M262G-1–M262G-3), kaempferol disulfates (M262S-1/M262S-2), sulfated kaempferol diglucuronide (M262G-S), glucuronidated kaempferol disulfate (M26G-2S), isorhamnetin 3-O-rhamnosyl-O-glucoside (29)/isorhamnetin 3-O-glucosyl-O-rhamnoside (30), glucuronidated isorhamnetin 3-O-rhamnosyl-O-glucosides/tamarixetin 3-O-rhamnosyl-O-glucosides (M29G-1–M29G-4), sulfated isorhamnetin 3-O-rhamnosyl-O-glucosides/ tamarixetin 3-O-rhamnosyl-O-glucosides (M29S-1/M29S-2), glucuronidated isorhamnetin 3-O-rhamnosyl-O-glucosides/tamarixetin 3-O-rhamnosyl-O-glucoside sulfates (M29G-S-1/M29G-S-2), isorhamnetin (32), isorhamnetin glucuronides (M32G-1–M32G-3), isorhamnetin sulfates (M32S-1/M32S-2), glucuronidated isorhamnetin sulfates (M32G-S-1/M32G-S-2), isorhamnetin diglucuronides (M32G-1–M32G-4), isorhamnetin disulfate (M322S), sulfated isorhamnetin diglucuronide (M322G-S) andglucuronidated isorhamnetin disulfates (M32G-2S-1/M32G-2S-2) were m/z 755→300, 931→755, 835→755, 1011→755, 609→300, 785→609, 689→609, 865→609, 301→151, 477→301, 477→301, 381→301, 557→301, 653→301, 461→301, 733→301, 637→301, 739→284, 915→739, 819→739, 995→739, 593→285, 769→593, 673→593, 849→593, 285→187, 461→285, 461→285, 365→285, 541→285, 637→285, 445→285, 717→285, 671→285, 623→315, 799→623, 703→623,879→623, 315→300, 491→315,395→315, 571→315, 667→315,475→315, 747→315and 651→315, respectively, with ascan timeof 0.02 sfor each ion pair. OnlyLC eluent flowover aperiod of2 to 10 min was introduced to theion sourcefor dataacquisition. Assessmentofblood to plasma ratios Rat blood-to-plasmaratios(B/Pratios) weredetermined for terpenelactones(bilobalideand ginkgolidesA, B, C and J), flavonol glycosides(10–13, 19–22, 27, 29and 30)andtheaglyconeconjugates(M16G-1–M16G-4, M16S-3, M26G-1–M26G-3,M32G-1–M32G-3andM32S-2). Blood samples(~600 µL) werecollected fromrats15 and30 min that received a15 min i.v. infusion doseofShuXueNinginjection at 4 mL·kg−1 or 15 min, 4 and 6 h that received a p.o. doseofGBE50 at 90 mg·kg−1. Theblood sampleswerecentrifuged at 2 270×g and 6°C for 8 min to yield the plasma and erythrocyte fractions. The plasma samples (100 µL) were precipitated with 300 µL ofmethanol to measure the concentrations of ginkgo compounds. Meanwhile, equal volumes of water were added to the erythrocytesamplesbeforelysisbysonication on ice. Thelysed erythrocytesamples(100 µL) weretreated with 300 µL ofmethanol and theconcentrationsofginkgo compoundsweremeasured. TheB/P ratio wascalculated usingthefollowingequation: B/Pratio =(0.44 ×CE +0.56 ×CP)/CP (2) whereCE and CP areerythrocyte and plasmaconcentrations, respectively. Themean hematocrit valueoftherats used was0.44 ±0.02. PK data analysis PlasmaPKparameterswereestimated byanoncompartmental methodusingaThermo Kineticasoftwarepackage (Philadelphia,PA, USA). Themaximumconcentration(Cmax) and thetimetaken toachievethepeakconcentration (Tpeak) wereobtained directlyfromthedatawith nointerpolation.Theareaunder concentration-timecurveup to the last measured or measurable time point (AUC0-t) was calculated by the trapezoidal rule. The nonvascular bioavailability(F) wascalculated usingthefollowingequations: F =(AUC0-24h(p.o.) ×dosei.v.)/(AUC0-24h(i.v.) ×dosep.o.) (3) for thetotal flavonolsand F =(AUC0-∞(p.o.)×dosei.v.)/(AUC0-∞(i.v.) ×dosep.o.) (4) for theterpenelactones. Theterminal elimination half-life(t1/2) wascalculated usingtherelationship 0.693/k. The total plasmaclearance(CLtot,p) for i.v. dosingwasestimated bydividingtheadministered dosebytheAUC0-t.The apparent volumeofdistributionat steadystate(VSS) for i.v. dosingwasestimated bymultiplyingtheCLtot,pbythe mean residence time (MRT). The biliary clearance (CLB) and renal clearance (CLR) were calculated from the cumulative amount excreted (Cum.Ae) into bile and urine, respectively, divided by the plasma AUC0-t. Dose proportionality assessment of AUC0-t for total flavonols and terpene lactones from ShuXueNing injection or GBE50 was conducted by the regression of log-transformed data (the power model) with criteria that were calculated accordingto themethod described bySmith et al., (2000). To assesstheextent ofbrain penetration, aKp valuewascalculated usingthefollowingequation: Kp =AUCbECF/AUCu,p (5) whereAUCu,p istheareaunder theunboundplasmaconcentration-timecurve. All resultswereexpressed asarithmeticmean ±standard deviation(SD). Chemicalsandreagents Bilobalide, ginkgolidesA, B, C quercetin 3-O-rhamnosyl-O-glucoside (12),quercetin (16), kaempferol (26)and - 5 - isorhamnetin (32) were obtained from the National Institutes for Food and Drug Control (Beijing, China). GinkgolideJand kaempferol 3-O-glucuronide(M26G-1) werefromTauto Biotech (Shanghai,China). Isorhamnetin 3-O-rhamnosyl-O-glucoside (29) and quercetin 3-O-glucuronide (M16G-1) were from Extrasynthese (Genay, France). Quercetin 3-O-(p-coumaroyl)-glucosyl-O-rhamnoside (11), kaempferol 3-O-(p-coumaroyl)-glucosyl-O-rhamnoside (20) and kaempferol 3-O-rhamnosyl-O-glucoside (21) were kindly donated byProf. J-J. Chen (KunmingInstituteofBotany, ChineseAcademyofSciences, Kunming, China).The preceding compounds purity exceeded 98%. Organic solvents of HPLC grade, sodium carboxymethycellulose, sterilesaline, sodiumheparin andisofluranewereobtained fromSinopharmChemical Reagent Co., Ltd. (Shanghai, China).Pentobarbital wasfromShanghai WestangBiotechnologyCo. Ltd.(Shanghai,China). Cremophor EL was purchased fromSigma-Aldrich (St Louis, MO,USA). References Guo B, Li C, WangG-J, Chen L-S (2006). Rapid and direct measurement offreeconcentrationsofhighly protein-bound fluoxetineand itsmetabolitenorfluoxetinein plasma. Rapid Commun MassSpectrom20: 39–47. Li L, Zhao Y-S, DuF-F, YangJ-L, Xu F, NiuWet al.(2012). Intestinal absorption and presystemicelimination of variouschemical constituentspresent in GBE50extract, astandardized extract ofGinkgo biloba leaves. Curr Drug Metab 13: 494–509. PaxinosG, Watson C (1998).TheRat Brain inStereotaxicCoordinates. AcademicPress: San Diego,CA. Smith BP, VandenhendeFR, DeSanteKA, Farid NA, Welch PA, Callaghan JTet al. (2000).Confidenceinterval criteriafor assessment ofdoseproportionality. PharmRes17: 1278–1283. Sun Y, Dai J-Y, HuZ-Y, Du F-F,Niu W, WangF-Qet al. (2009). Oral bioavailabilityand brain penetration of (−)-stepholidine,atetrahydroprotoberberineagonist at dopamineD1 and antagonist at D2 receptors, in rats. Br J Pharmacol 158: 1302–1312. XieD-L, WangN, Gao Q, ZhangG-A, Shao B-P,Jin X-W, et al. (2000). Ginkgo bilobacomposition,method to preparethesameand usesthereof. US 006030621A. Zhao Y, Sun Y, Li C (2008).Simultaneousdetermination ofginkgo flavonoidsandterpenoidsin plasma: ammoniumformatein LC mobilephaseenhancingelectrosprayionization efficiencyand capacity. JAmSocMass Spectrom19: 445–449.