SQ109 药代动力学研究

Interspecies pharmacokinetics and in vitro metabolism of SQ109 *,1Lee Jia,2Patricia E.Noker,2Lori Coward,2Gregory S.Gorman,3Marina Protopopova&1Joseph E.Tomaszewski1Developmental Therapeutics Program,National Cancer Institute,NIH,6130Executive Blvd.,Rm8042,Rockville,MD20852, U.S.A.;2Southern Research Institute,Birmingham,AL,U.S.A.and3Sequella Inc.,Rockville,MD,U.S.A.1This study aimed at characterizing the interspecies absorption,distribution,metabolism andelimination(ADME)profile of N-geranyl-N0-(2-adamantyl)ethane-1,2-diamine(SQ109),a newdiamine-based antitubercular drug.2Single doses of SQ109were administered(intravenously(i.v.)and per os(p.o.))to rodents and dogsandbloodsamples were analyzedby liquidchromatography tand em mass spectrometry(LC/MS/MS).Based on i.v.equivalent body surface area dose,the terminal half-life(t1/2)of SQ109in dogs waslonger than that in rodents,reflected by a larger volume of distribution(V ss)anda higher clearancerate of SQ109in dogs,compared to that in rodents.The oral bioavailability of SQ109in dogs,rats andmice were2.4–5,12and3.8%,respectively.3After oral administration of[14C]SQ109to rats,the highest level of radioactivity was in the liver,followedby the lung,spleen andkid ney.Tissue-to-bloodratios of[14C]SQ109were greater than1.Fecal elimination of[14C]SQ109accountedfor22.2%of the total d ose of[14C]SQ109,while urinaryexcretion accountedfor only5.6%.The bind ing of[14C]SQ109(0.1–2.5m g mlÀ1)to plasma proteinsvaried from6to23%depending on the species(human,mouse,rat and dog).4SQ109was metabolizedby rat,mouse,d og andhuman liver microsomes,resulting in22.8,48.4,50.8or58.3%,respectively,of SQ109remaining after a10-min incubation at371C.The predominantmetabolites in the human liver microsomes gave intense ion signals at195,347and363m/z,suggestingthe oxidation,epoxidation and N-dealkylation of SQ109.P450reaction phenotyping usingrecombinant cDNA-expressedhuman CYPs in conjunction with specific CYP inhibitors ind icatedthat CYP2D6andCYP2C19were the pred ominant CYPs involvedin SQ109metabolism.British Journal of Pharmacology(2006)147,476–485.doi:10.1038/sj.bjp.0706650;publishedonline23January2006Keywords:Antituberculosis;diamine;SQ109;ADME;pharmacokineticsAbbreviations:ADME,absorption distribution metabolism elimination;CL,clearance rate;ESI,electrospray ionization;d.p.m., disintegrations per min;LC/MS/MS,liquid chromatography tandem mass spectrometry;V ss,volume ofdistribution;t1/2,terminal half-lifeIntroductionIdentification of a promising new compound for the treatment of tuberculosis is exciting,andits d evelopment is very challenging(Barry III et al.,2000).Current regimens require combinedtreatment of tuberculosis with multiple d rugs, including isoniazid,rifampin,ethambutol and pyrazinamide for the full dosing period of6months(American Thoracic Society Documents,2003).It is difficult to comply completely with this complex andprolongedregimen,andconsequently, there is a substantial rate of treatment failure,even among patients with drug-sensitive disease.Thus,the availability of a more potent antibiotic that couldclear infection more rapid ly wouldbe very valuable.Many approaches to screening andid entifying antibiotics start with a search for biochemical inhibitors of potential targets andthen focus on those that inhibit microbial growth (Rubin,2005).Lee et al.(2003)startedwith a d iverse library of 63238ethambutol analogs with1,2-diamine pharmacophore andtestedtheir activity against Mycobacterium tuberculosis using high-throughput screening.The initial rationale was to identify diamine analogs with enhanced efficacy over etham-butol to improve the existing treatment of tuberculosis,and hopefully make therapeutic implementation easier andof shorter duration.After the lead optimization and screening, three analogs were foundto have an enhancedantitubercular activity comparedto ethambutol.Basedon the results obtainedfrom comparisons of effectiveness andhigh-through-put pharmacokinetic screening using cassette dosing combined with liquidchromatography tand em mass spectrometry (LC/MS/MS)(Jia et al.,2005a),we selected N-geranyl-N0-(2-adamantyl)ethane-1,2-diamine(SQ109,MW330.2;see Figure5for SQ109structure)as a leadcompoundfor advanced drug testing.SQ109exhibited both in vitro anti-microbial activity against M.tuberculosis strain H37Rv grown inside the host murine macrophage cells and in vivo antimicrobial activity on the mouse model inoculated with the H37Rv strain.Oral administration of SQ109to mice for 28consecutive days resulted in significant reductions in mycobacterial burden in both the lungs and spleen of the mice.Monitoring SQ109levels in mouse vital tissues in the course of28-day oral administration showed that the potential sites of action(e.g.,lungs andspleen)containedSQ109at*Author for correspondence;E-mail:jiale@British Journal of Pharmacology(2006)147,476–485&2006Nature Publishing Group All rights reserved0007–1188/06$30.00/bjpconcentrations that were at least10times higher than its minimal inhibitory concentration(1.56m M).SQ109displayed a large volume of distribution to various tissues.Despite its low oral bioavailability,the targetedtissue concentrations of SQ109were at least120-foldhigher than that in the plasma (Jia et al.,2005b).Fingerprinting of SQ109pharmacopro-teomics indicates that treatment of M.tuberculosis H37Rv strain with SQ109causedsignificant upregulation of secreted antigenic proteins ESAT-6-,CFP-10-and ATP-dependent DNA/RNA helicase,andd ownregulation of b-ketoacyl-acyl carrier protein synthase(Jia et al.,2005a).The purpose of the present studies was to characterize systematically andcomprehensively the preclinical profile of absorption,distribution,metabolism and elimination(ADME) for SQ109in a number of species,including its pharmacoki-netics,plasma protein binding;tissue distribution and urinary andfecal elimination in rats;as well as its in vitro metabolism, metabolism phenotyping andpossible metabolic pathways. MethodsAnimalsMale CD2F1mice(23–27g),male Fischer rats(271–289g)and beagle dogs(7.5–8.9kg,two males and two females per dose group)were handled in accordance with the Guide for the Care andUse of Laboratory Animals(National Research Council,1996),or the Principles of Laboratory Animal Care (/laws).Analytical method for determining SQ109in biomatrices The analytical methodfor d etermining SQ109in various biomatrices was similar to that described previously(Jia et al., 2005b).Briefly,separation of SQ109from biomatrices was achievedon a b-basic C18analytical column preceded by a C18 guardcolumn(Keystone Scientific,Bellefonte,PA,U.S.A.)at 251C anda flow rate of0.6ml minÀ1.SQ109andterfenad ine (internal standard)were eluted using a mobile phase composed of buffer A(5m M CH3COONH4with0.1%trifluoroacetic acid,pH6.8)and buffer B(methanol with0.1%trifluoroacetic acid)according to the following gradient program:50%buffer A and50%buffer B were heldfor0.5min,andthen buffer A was linearly decreased to20%over3min and remained constant for1min while the analytes were eluted.The column was re-equilibratedto initial cond itions via a step gradient for 3min.A PE Sciex API3000triple quadrupole mass spectro-meter equippedwith a Turbo on spray source andoperating at4501C in the positive ion mode was used for the analysis of SQ109.The lower limit of quantitation of SQ109in the plasma was determined to be1.95ng mlÀ1.Rat and dog pharmacokinetic studies with SQ109Rats with an indwelling jugular vein catheter were used for the pharmacokinetic studies.Rats were given either a single intravenous(i.v.)bolus dose of1.5mg kgÀ1(9mg mÀ2)or an oral dose of13mg kgÀ1(78mg mÀ2)of SQ109(n¼8per dose group);rats were divided into subgroups consisting of four rats per subgroup.Rat blood(0.7ml)was withd rawn from the jugular vein catheter at alternating time points from individual rats in each subgroup.Bloodsamples were collectedat2,5,10, 15and30min and1,3,6,10and24h after a single i.v. ad ministration,or5,15and30min and1,2,4,6,10and24h after a single oral administration.Each blood sample was centrifugedto separate plasma,which was then storedat À701C until analysis.Beagle dogs were dosed by gavage at either 3.75or 15mg kgÀ1(75or300mg mÀ2),or intravenously at either 0.45or4.5mg kgÀ1(9and90mg mÀ2).Dog blood(0.7ml)was withdrawn from the jugular vein at2,5,10,20and30min and 1,2,4,8,12,18and24h after a single i.v.ad ministration,or 10,20and30min and1,2,4,8,12,18and24h after a single oral administration.Each bloodsample was collectedinto a tube containing EDTA andcentrifuged(2000Âg,10min)to separate plasma andredbloodcells.To each200m l of plasma sample,10m l of internal standard solution(10m g mlÀ1)was added.SQ109was then separatedandanalyzedby the LC/MS/MS method according to the previously described procedures(Jia et al., 2005b).Peak area ratios of SQ109to the internal standard were plottedagainst theoretical concentrations.Drug concen-trations in the plasma samples were calculatedfrom the standard calibration curves.Pharmacokinetic parameters were calculatedusing the computer program WinNonlin(Pharsight Co.,Mountain View,CA,U.S.A.),andbioavailability was calculatedas(AUC p.o.AUC i.v.À1)Â(dose i.v.dose p.o.À1)Â100%.Tissue distribution and elimination of[14C]SQ109in rats[14C]SQ109(5.8mg mlÀ1)was diluted4.4-fold with0.9%sterile saline to yielda formulation containing1.3mg mlÀ1SQ109 (225m Ci mlÀ1).Male Fischer rats(271–289g)were individually housedin metabolism cages from which urine andfeces were cumulatively collectedto d etermine[14C]SQ109excretion rate. The rats were orally dosed by gavage with13mg kgÀ1of [14C]SQ109.Rats were killedat0.5,5,10and24h after d osing (n¼3per time point)in order to collect blood,tissues and organs for quantitative analysis.Tissues,intestinal tract contents andfeces were homogenizedin10volumes of water. Duplicate aliquots of whole blood,homogenates of tissues, intestinal contents andfeces were d igestedwith tissue solubilizer Soluene350,and decolorized with30%hydrogen peroxide to eliminate chemiluminescence.The samples were radioassayed after mixing with glacial acetic acid and the scintillation cocktail.Duplicate aliquots of urine and cage rinses were radioassayed after mixing with scintillation cocktail.Pieces of carcasses were digested with650ml of10N sodium hydroxide,maintained at371C for3days,and then at room temperature until complete dissolution of the carcasses(B2 weeks).Quadruplicate aliquots of each dissolved carcass were diluted with water(1:20,v vÀ1);portions of each diluted sample were radioassayed after the addition of an appropriate scintillator.After correction for volume by dilution and volume assayed,the radioactivity expressed as disintegrations per min(d.p.m.)of[14C]SQ109in each sample was determined. Plasma protein bindingThe percent binding of[14C]SQ109to mouse,rat,dog and human plasma proteins was determined by using ultracen-trifugation(Barre et al.,1985;Boulton et al.,1998).Briefly,L.Jia et al Interspecies pharmacokinetics of SQ109477British Journal of Pharmacology vol147(5)spiking solutions of[14C]SQ109were preparedby d iluting the stock[14C]SQ109(1mCi of 5.9mgÀ1mlÀ1)with absolute ethanol to yieldspiking solutions containing5,25or 125m g mlÀ1of[14C]SQ109.For each species,a10.8ml aliquot of plasma was mixedwith0.22ml of the appropriate spiking solution of[14C]SQ109to yieldfinal SQ109concentrations of 0.1,0.5or2.5m g mlÀ1.The plasma mixtures were placedinto individual polycarbonate ultracentrifuge tubes and centrifuged at100,000Âg for24h at41C.At the endof the centrifugation period,the upper chylomicron layer,middle aqueous layer and lower protein pellet were separatedandthe volume of each layer was determined.The protein pellets were dissolved in Soluene350 tissue solubilizer.The radioactivity of duplicate portions of the chylomicron andaqueous layers as well as the solubilized protein layer was determined,and the total amount of radioactivity in each layer was calculated.The percentage of radioactivity in each layer was determined by comparing the amount of radioactivity in each layer with the sum of the total amount of radioactivity in all three postcentrifugation plasma samples.The percent of the total radioactivity in the aqueous layer was considered to represent the unbound fraction of SQ109,while the sum of the radioactivity in the chylomicron andlower(pellet)protein layers was consid eredto represent the boundfraction of the d rug.Radioactivity determinationThe radioactivity of all samples was measured in the Tri-Carb 2100TR liquidscintillation analyzer.All counts were con-verted to absolute radioactivity(d.p.m.)by automatic chemiluminescence andquench correction.Samples having radioactivity(d.p.m.)less than or equal to twice background d.p.m.were considered to be below the limit of quantitation, and therefore the reading was considered zero d.p.m.for calculation purposes.SQ109equivalents in biological samples were determined by dividing the sample d.p.m.by the specific activity of[14C]SQ109in d.p.m.per microgram,and expressed in microgram per gram of tissue.[14C]SQ109equivalents were also expressedas a percentage of[14C]SQ109amount in organs or tissues over the administered total[14C]SQ109amount per animal.The radioactivity in rat urine and feces was expressed as a percentage of the administered dose for each time interval andas a cumulative percentage.Microsomal metabolism of SQ109The microsomal assay was similar to that described previously (Jia et al.,2003).Briefly,SQ109(10m M)was incubatedwith mouse,rat,dog and human liver microsomes,respectively, in an NADPH-generating system containing1.3m M NADP, 3.3m M glucose-6-phosphate,0.4U/ml glucose-6-phosphate dehydrogenase and 3.3m M MgCl2in100m M potassium phosphate buffer(pH7.4).Reaction mixtures were prepared in duplicate and were preincubated for5min at371C.The reactions were then initiated by the addition of microsomes (30m l of a20mg mlÀ1solution in250m M sucrose,yielding a final protein concentration of0.5mg mlÀ1).The final volume of each reaction mixture was1.2ml.Negative control reactions were preparedby incubating mixtures that exclud edeither microsomes or SQ109from the mixture.For negative control incubation where microsomes were excluded,they were added back to the reaction mixture after quenching with acetonitrile. Samples were removedat0,10,20,40or80min andvortex-mixedwith coldacetonitrile to stop the reaction.After centrifugation,a portion of each resulting supernatant was analyzedby mass spectrometry for unchangedSQ109.Metabolism of SQ109by cDNA-expressed recombinant human CYPsSQ109metabolism was also evaluatedin microsomes prepared from insect cells transfectedwith cDNAs encod ing for human CYP1A2,CYP2A6,CYP3A4,CYP2B6,CYP2C8,CYP2C9, CYP2C19or CYP2D6.The recombinant enzymes and microsomes from untransfectedinsect cells were usedin parallel as a control.SQ109(10m M)was preincubatedin duplicate with the above-mentioned NADPH-generating system for5min at371C.The reactions were then initiated by the addition of the individual CYPs(final100pmol CYP mlÀ1)or corresponding untransfected cells.Samples were mixedby inversion,removedat0and30min andmixedwith ice-coldacetonitrile to stop the reactions.After centrifugation (14,000Âg for20min at41C),each extract was analyzedby using the mass spectrometry to monitor metabolite formation or SQ109depletion.Electrospray ionization(ESI)full mass scans were performedto obtain the ion chromatograms of the expected metabolites according to predicted mass gains and losses as comparedwith the molecular mass of SQ109.The ESI as a gentle ionization technique is preferredin metabolite analysis,since ESI usually does not dissociate compounds extensively.The metabolite profiling was basedon the detection of protonated,deprotonated or adduct ions,but not on the fragment ions(Kostiainen et al.,2003).Samples were assayedin both the negative andpositive ion to ensure detection of all potential metabolite(s),and define the structure(s).Metabolite quantitation was basedon percentages of peak areas of each metabolite as a function of incubation time comparedto the total area of all chromatographic peaks. Chemical inhibition studiesThe following inhibitors,at the concentrations shown,were incubatedwith the correspond ing CYP isoforms.These concentrations were basedon literature information(Parkin-son,1996;Tucker et al.,2001;Bjornsson et al.,2003): furafylline(CYP1A2;0.1,1and10m M),quinidine(CYP2D6;0.5,1and10m M),ticlopidine(CYP2C19;5,20and100m M) (Donahue et al.,1997;Tateishi et al.,1999;Ha-Duong et al., 2001)andtroleand omycin(CYP3A;0.5,1and10m M).All inhibitors were dissolved in methanol prior to addition to the incubation mixtures.Reaction mixtures containing human cDNA expressedCYP2D6or CYP2C19(100pmol P450mlÀ1 reaction mixture),the NADPH-generating system,selective CYP inhibitors and100m M potassium phosphate buffer(pH 7.4)were preincubatedfor15min at371C.Each reaction was then started by the addition of SQ109(10m M)with subsequent mixing of each sample by inversion.The samples were immediately removed and mixed with cold acetonitrile to stop reaction at0and30min of incubation.SQ109andits metabolites were identified by the LC/MS/MS method.Peak areas formedwere usedfor quantitative analyses.Control incubation mixtures included mixtures without inhibitors,and mixtures with untransfectedinsect cell microsomes usedas478L.Jia et al Interspecies pharmacokinetics of SQ109 British Journal of Pharmacology vol147(5)microsomal control,andmixtures that containedmethanol insteadof inhibitor(methanol control).Quantitative analyses were performedby comparing the peak areas of the inhibition reactions to their respective methanol controls.The total organic solvent content of the in vitro reaction mixtures was less than2%.Materials[14C]SQ109was preparedin ethanol(1mCi mlÀ1)with specific activity andrad iochemical purity(HPLC analysis)of 172m Ci mgÀ1(57mCi mmolÀ1)and99.6%,respectively.Tissue solubilizer Soluene350andTri-Carb2100TR liquidscintilla-tion analyzer were purchasedfrom Perkin-Elmer Life and Analytical Sciences(Boston,MA,U.S.A.).Scintillation cocktail(Safety SolveTM complete counting cocktail)was purchasedfrom Research Prod ucts nternational Co.(Mount Prospect,IL,U.S.A.).All liver microsomes,human CYP isoforms andspecific CYP isoform inhibitors were purchased from BD Gentest(Woburn,MA,U.S.A.).Organic solvents usedin chromatographic separation of SQ109were purchased from EM Science(Gibbstown,NJ,U.S.A.).Statistical analysisData are expressedas the mean7s.d.Statistical analysis was performedwith Stud ent’s t-test for pairedobservation(two-tailed).A probability of0.05or less was considered significant.ResultsPharmacokinetics of SQ109in rats and dogsSQ109plasma concentration–time courses after single oral and i.v.administration to rats and dogs are illustrated in Figures1 and2,respectively.Pharmacokinetic parameters calculated from noncompartmental analysis of SQ109concentrations in rat andd og plasma are presentedin Tables1and2, respectively.nterspecies pharmacokinetics are summarizedin Table3.The major pharmacokinetic parameters such as t1/2, V ss,clearance rate(CL)andoral bioavailability were comparedin Table3among mice(Jia et al.,2005b)rats and dogs at the same dose based on equivalent body surface area (9mg mÀ2).The terminal t1/2of SQ109in dogs(12–29h; Table2)was longer than in rats(7–8h;Table1),as reflected by the significantly larger volume of distribution of SQ109 in dogs,while the CL in dogs was higher than that in rats in comparison on body surface area dose.The i.v.C max andAUC in dogs increased in proportion to the increase in doses, whereas the oral C max andAUC were less proportional to the increase in doses probably because of the low oral bioavail-ability of SQ109in dogs(2.4–5%;Table2).The oral bioavailability of SQ109in rats was determined to be12%. Comparedto the pharmacokinetic profiles of rats andd ogs, male CD2F1mice given the same dose of SQ109scaled by body surface area(9mg mÀ2,i.v.;75mg mÀ2,p.o.)showeda pharmacokinetic profile(Jia et al.,2005b)similar to rats,but different from dogs in terms of t1/2,CL and V ss,probably Table1Pharmacokinetic parameters of SQ109inrats(n¼8)aRoute i.v.p.o.Dose(mg kgÀ1) 1.513(mg mÀ2)978C max(ng mlÀ1)F644T max(h)F0.5t1/2(h)7.48.2AUC0–N(ng hÀ1mlÀ1)953992CL(ml kgÀ1hÀ1)1575FV ss(ml kgÀ1)9964FBioavailability(%)F12a Noncompartmental analysis.L.Jia et al Interspecies pharmacokinetics of SQ109479British Journal of Pharmacology vol147(5)because rodents share the same allometric exponent that is different from that of dogs.Although no acute toxicity and pathological effects were identified,emesis was observed in dogs 10min after oral administration of the high dose of SQ109(750mg m À2);the emesis was dose related,and most likely a consequence of gastric irritation.[14C]SQ109tissue distribution and elimination in ratsFigure 3shows [14C]SQ109levels in various tissues of rats at 0.5,5,10and 24h after oral administration of the radioactive compound(13mg kg À1).In general,excluding the concentra-tions of [14C]SQ109in gastrointestinal tract,the highest concentrations of radioactivity were found in the liver,and the lowest in the fat.At each time point,[14C]SQ109concentrations in the lung,spleen andkid ney were similar to each other,andrankedas the tissues containing the second highest concentrations of [14C]SQ109.At 0.5h after oral administration,[14C]SQ109concentrations in the brain and whole bloodwere the same (143ng g À1).At 5,10and24h time points,[14C]SQ109concentrations in the brain were 257,187and313ng g À1,respectively,which were higher than those inTable 2Noncompartmental analysis of pharmacokinetic parameters of SQ109in beagle dogs (mean 7s.d.,n ¼4)Routes i.v.i.v.p.o.p.o.Dose(mg kg À1)0.45 4.5 3.7515(mg m À2)99075300C max (ng ml À1)2717109276271180**11.771.737.176.5**T max (h)F F0.5970.160.5170.09t 1/2(h)29.378.112.473.01**19.674.811.671.5*AUC 0–N (ng h À1ml À1)22373119427315**86.5716.0158734**CL (ml kg À1h À1)2138729524717319F F V ss (ml kg À1)75,200719,10029,20076940*F F Bioavailability (%)FF5.072.42.470.7*P o 0.05,**P o 0.01,compared with the values obtained from lower doses of the same administration route.Table 3Comparison of major pharmacokinetic parameters of SQ109administered i.v.to mice,rats and dogs at the same dose (mg m À2)aSpecies Dose (mg m À2)t 1/2(h)CL (ml m À2h À1)V ss (ml m À2)Oral bioavailability (%)bMouse c 9 5.311,36435,400 3.8Rat 97.4945059,70012Dog929.342,7601,504,0005aThe conversion factors usedfor changing a d ose expressedin terms of mg kg À1to an equivalent surface area dose mg m À1are 3(mouse),6(rat)and 20(dog),respectively,referring to the FDA Guidance /cber/gdlns/dose.pdf.bNote,the oral bioavailability was determined based on oral administration of 75mg m À2of SQ109to the animals.cThe mouse data were obtained from our previous studies (Jia et al .,2005b).480L.Jia et al Interspecies pharmacokinetics of SQ109British Journal of Pharmacology vol 147(5)whole bloodat the same time points(103,82and43ng mlÀ1), suggesting that the compoundcrosses the blood–brain barrier. Tissue-to-bloodlevel ratios were generally greater than1, indicating that the compound has the capability to move across blood vascular wall,distribute and reside in various tissues.The total equivalent concentrations of[14C]SQ109 in the stomach,small intestine andlarge intestine at0.5h after the oral administration were331771,312767and 3.471.9m g gÀ1,respectively.At24h,the concentrations declined to2.173.3,31712and2797127m g gÀ1,respectively, kinetically reflecting the gastrointestinal transit time and absorption time of[14C]SQ109.By24h after oral administration,[14C]SQ109was eliminated mainly via the feces,which accountedfor22.2%of the total dose,while urinary excretion accounted for only5.6%of the total dose(Table4).The excretion rate(m g hÀ1)of[14C]SQ109 via urine was faster than that via feces during the first10h period.Thereafter,[14C]SQ109excretion rate via feces became predominating factor.SQ109plasma protein bindingThe results of the protein binding determinations for mouse, rat,dog and human plasma are presented in Table5.In vitro metabolism of SQ109in liver microsomesThe mass spectrometric analyses of SQ109from the in vitro metabolism samples suggest that the parent compoundSQ109 is rapidly metabolized when incubated with mouse,rat,dog or human liver microsomes(Figure4).After incubation at371C for10min,the remaining SQ109accountedfor22.8,48.4,50.8 and58.3%of the total SQ109initially added to rat,mouse,dog and human liver microsomal reaction mixtures,respec-tively.The fastest rate of metabolism of SQ109occurredin rat liver microsomes,where essentially all of the added SQ109was metabolizedwithin20min.The slowest rate of metabolism occurred in human liver microsomes,where8.1%of the added SQ109remainedafter40min incubation at371C.In contrast, the percentage of SQ109remaining in the negative control sample after80-min incubation was96.5,100.0,116.5and 111.5for human,rat,mouse andd og liver microsomes, respectively,as comparedto the0min sample.The control result suggests involvement of liver microsomes in SQ109 metabolism.Liver microsomes convertedSQ109(m/z331.5)to four predominant metabolites with mass to charge ratios(m/z)at 195,347,361and363.Proposedmetabolite id entification using mass spectrometry suggests that m/z361namedM1 symbolize an N-nitrosylation product of SQ109.The m/z347 designated as M2represents two metabolites(M2-1and M2-2) with a single oxygen added to SQ109at different locations,for example,the side chain and the adamantine ring.The m/z195 represents a single metabolite obtained via an N-dealkylation reaction of an unstable intermediate and is designated M3.The m/z363(namedas M4)is consistent in mass with two d egrees of oxidation,that is,ring oxidation combined with N-hydroxylation(M4-1)or with epoxidation(M4-2)(Figure5). Metabolism by cDNA recombinant human CYPsFor the30-min reaction mixtures containing CYP2A6, CYP2C8or CYP2C9,no significant metabolite peaks were observed at the monitored masses.In addition,only negligible amounts of SQ109metabolite at m/z347were foundin incubations using individual recombinant CYP1A2,CYP3A4 or CYP2B6.Extensive metabolism of SQ109was observedin microsomes from insect cell lines transfectedwith cDNA from human CYP2D6andCYP2C19.CYP2C19showedthe capacity to catalyze metabolism of SQ109to produce metabolites of M1, M2,M3andM4,while CYP2D6primarily catalyzedSQ109to form M1andM4.Of great interest was the possible formation of N-(2-adamantyl)ethylene-1,2-diamine(M3,m/z195)cata-lyzedby CYP2C19.M3might contain the ad amantane ring after cleavage of the N–C bondof SQ109(Figure5).Basedon the degradation rate of SQ109estimated by chromatographic peak areas over time,CYP2D6was foundto play the majorTable4Cumulative excretion(%of dose)and excretion rate of[14C]SQ109following a single oral dose of[14C]SQ109(13mg kgÀ1)to ratsPeriod afterdosing(h)Excretion(%of dose)Excretion rate(m g hÀ1)Urine Faeces Urine Faeces0–5 1.1570.340.0270.048.4472.520.1670.30 5–10 1.4370.260.0470.0410.4671.920.3070.36 10–24 2.3570.3612.7173.08 6.1570.9433.2978.07 24–48 1.1770.167.8370.49 1.7970.2511.9670.75 Total(0–48h)5.6670.8722.273.33F FExcretion values are the mean7s.d.(n¼3).Table5Binding of[14C]SQ109at different concen-trations to plasma proteinSQ109(m g mlÀ1)%SQ109bound to plasma protein fromMouse Rat Dog Human0.1 6.2,7.88.4,10.310,11.414.3,16.1 0.5 5.8,8.48.1,9.19.5,11.716.3,18.5 2.58.4,8.58.4,11.810.3,12.117.8,23.8Data were obtainedfrom two tests for eachspecies.L.Jia et al Interspecies pharmacokinetics of SQ109481British Journal of Pharmacology vol147(5)。