Discovery of potent iminoheterocycle BACE1 inhibitors

Discovery of potent iminoheterocycle BACE1inhibitorsJohn P.Caldwell ⇑,Robert D.Mazzola,James Durkin,Joseph Chen,Xia Chen,Leonard Favreau,Matthew Kennedy,Reshma Kuvelkar,Julie Lee,Nansie McHugh,Brian McKittrick,Peter Orth,Andrew Stamford,Corey Strickland,Johannes Voigt,Liyang Wang,Lili Zhang,Qi Zhang,Zhaoning ZhuMerck Research Laboratories,2015Galloping Hill Road,Kenilworth,NJ 07033,USAa r t i c l e i n f o Article history:Received 5September 2014Revised 29September 2014Accepted 2October 2014Available online 23October 2014Keywords:Alzheimer’s disease b -Secretase BACEAspartyl protease inhibitors Structure-based drug designa b s t r a c tThe synthesis of a series of iminoheterocycles and their structure–activity relationships (SAR)as inhibi-tors of the aspartyl protease BACE1will be detailed.An effort to access the S3subsite directly from the S1subsite initially yielded compounds with sub-micromolar potency.A subset of compounds from this effort unexpectedly occupied a different binding site and displayed excellent BACE1affinities.Select com-pounds from this subset acutely lowered A b 40levels upon subcutaneous and oral administration to rats.Ó2014Elsevier Ltd.All rights reserved.A widely pursued strategy for the potential treatment of Alzhei-mer’s disease is inhibiting the production of neurotoxic b -amyloid (A b )peptides,especially A b 40,42.The first step in generation of A b involves cleavage of amyloid precursor protein (APP)by BACE1.1Potent inhibitors of BACE1should halt A b production thereby reducing the neurotoxic A b oligomers and plaques.Non-peptidic BACE1inhibitors have been reported by our labs and others and offer the most promising avenue to the discovery of orally bioavail-able,brain penetrant inhibitors.2,3In earlier accounts,we reported on the rational design of com-pound 1from a fragment screening hit.2The amidine portion of the iminohydantoin core forms a critical hydrogen bond network with the catalytic diad (Asp 32and Asp 228)of the BACE1enzyme.Addi-tionally,the initial design strategy focused on the optimization of four subsites (S20,S1,S2,and S3)to afford a potent compound,K i =27nM (see Fig.1).4Unfortunately,this compound had low plasma exposure in an oral rat pharmacokinetic assay.Consistent with a strategy to maximize the probability of identifying inhibi-tors with high CNS exposure,we sought to explore variants of 1that would access S3directly through S1,eliminating the need for occupation of S2and inherently lowering the MW and c Log P .2d This work will detail the synthesis of these compounds and high-light the salient structure–activity relationships.Key to construction of the proposed S1–S3iminohydantoins was the use of the oxazolidinone intermediate 3(Scheme 1).2dElaboration of 3through alkylation with the appropriate allyl bromide offered many opportunities to access S3from S1.Scheme 2shows the approach to S1piperidine and pyrrolidine systems.ForBAC CE1 K i=2727nM Figure 1.X-ray of 1in the active site of BACE1./10.1016/j.bmcl.2014.10.0060960-894X/Ó2014Elsevier Ltd.All rights reserved.⇑Corresponding author.Tel.:+19087405199.E-mail address:john.caldwell@ (J.P.Caldwell).the piperidines,upon deprotonation of3with KHMDS and treat-ment with5(generated from4in a straightforward manner5,6), 1,10-oxazolidinone6was produced with retention of stereochemis-try in a manner consistent with literature precedent.7Methanoly-sis and hydrogenation of6furnished the aminoester7as a diastereomeric mixture of Boc-protected piperidines.After subse-quent cyclization to the corresponding thiohydantoins8,the diastereomerically pure compounds were obtained by silica gel chromatography.The Boc-piperidines were individually deprotec-ted,and the resulting piperidines9were elaborated through a vari-ety of standard N-derivatization chemistries(reductive amination, amidation and sulfonylation).These thiohydantoins10were subjected to oxidative amination conditions to prepare thefinal iminohydantoin targets(representative sequence shown for the (R)-isomer in Scheme2).Derivatives of both diastereomers of the S1-piperidine were explored.The S1-pyrrolidinyl analogs were prepared in an similar fashion starting with the benzyl-protected 2,5-dihydropyrrole ester11(Scheme2,bottom section).8 In a concurrent effort to explore the S1-aminocyclohexane vari-ants(Scheme3),oxazolidone3was alkylated with12,oxidized to the en-one,9and then reduced to form a mixture of alcohol diaste-reomers14.Since the position of the hydroxyl group as an allyl alcohol offered the opportunity to set the desired C1(R)-stereo-5456J.P.Caldwell et al./Bioorg.Med.Chem.Lett.24(2014)5455–5459chemistry at the cyclohexyl stereocenter,we separated the alco-hols to provide pure diastereoisomers14.Hydrogenation of the TBS-protected derivative of alcohol(R)-14proceeded cis relative to the bulky silyl protecting group,whereas hydroxyl-directed hydrogenation of(S)-14provided(S)-17.As desired,both transfor-mations provided the(R)-stereochemistry at the C1cyclohexyl.Aminoesters(R)-15and(S)-17were separately elaborated to the Boc-protected iminohydantoins16,and the alcohol moieties were then oxidized to give the corresponding cyclohexanone18. To obtain the(R)-aminocyclohexane of19,cyclohexanone18was treated with2,4-dimethoxybenzylamine(DMBNH2)under reduc-tive amination conditions.The resulting diastereomers were sepa-rated(product d.r.=(S)/(R):1/2),and the DMB group was deprotected under hydrogenation conditions.10Amine(R)-19was converted to the corresponding amides(43–45),ureas(46–53), and N-arylated targets(54–55).Once the correct diastereomer was validated through BACE1K i determination(vide infra),we developed an alternative stream-lined synthesis(Scheme3,bottom section).In this new route,oxa-zolidinone20was prepared from2and then alkylated with chiral allyl bromide21,which contained the TBS-protected-(S)-hydroxyl such that the C3-cyclcohexyl stereocenter was installed far earlier in the synthesis.After methanolysis of the oxazolidinone and TBAF deprotection,cyclohexenol22was subjected to a one-pot sequen-tial hydrogenation to(a)set the C1-cyclohexyl(R)-stereocenter through hydroxyl-directed hydrogenation,(b)deprotect the N-benzyl carbamate,and(c)reduce the phenethyl to the cyclohexyl-ethyl to produce23.Aminoester23was elaborated to the Boc-pro-tected iminohydantoin,treated with an azide transfer reagent with inversion of stereochemistry at C3,and reduced to the aminocyclo-hexane24now possessing the desired(R)-C3-stereochemistry. Amino derivatives of24were produced and Boc-deprotection afforded thefinal target compounds.Table1highlights the SAR in the S1-heterocycloalkyl series.In both the S1piperidine and S1pyrrolidine series,there is a prefer-ence in terms of BACE1potency for the(R)-stereochemistry at the tertiary center(cf.32vs33,34vs35,36vs37,and38vs39).Addi-tionally,amides appeared to be the preferred derivative when compared to substituted amines,ureas and sulfonamides(cf.29,30,and31).When amides were explored further,the n-butyl amide and cyclopentylmethyl amide showed the best potencies with several examples showing sub-micromolar BACE1K i values. The S1-pyrolidine was slightly more potent than the S1-piperidine.Table2details the results in the S1-cyclohexylamine series.In general,potencies were significantly better in this series in com-parison to the heterocycloalkyl series.Reductive amination with a variety of substrates led to compounds(40–42)with sub-micro-molar potency.From X-ray crystallographic studies(Fig.2),13com-pounds from both series(e.g.,38and41)validated the hypothesis that we can improve potency by occupying S3directly from S1.Furthermore,significant potency gains were achieved by decreasing the basicity of the exocyclic nitrogen through the amide (43–44)and urea(46–54)derivatives.The benzylamide45was five-fold less potent than the phenyl amide analog43.Once again, X-ray crystallography provided valuable insight into these dra-matic potency differences.Surprisingly,these compounds did notTable1BACE1affinities of S1-piperidine and pyrrolidine analogs12Compound R n S/R BACE-1K i(l M)Cat-D K i(l M)25Boc2S4133 26Boc2R29>100 27H2S66>100 28H2R8.5>100 29n-(CH2)4CH32R7.6>100 30–C(O)NH(CH2)2CH32R8.5 2.8 31–SO2(CH2)3CH32R 6.325.5 32–C(O)(CH2)3CH32R 3.244.0 33–C(O)(CH2)3CH32S1728.0 34–C(O)CH2-c-pentyl2R0.970.87 35–C(O)CH2-c-pentyl2S12.48.5 36–C(O)(CH2)3CH31R0.4923.0 37–C(O)(CH2)3CH31S17.4nt 38–C(O)CH2-c-pentyl1R0.64 6.0 39–C(O)CH2-c-pentyl1S37ntnt=not tested.Table2BACE1affinities of S1-aminocyclohexyl analogs12Compound R1BACE-1K i(nM)Cat-DK i(l M) 40–CH2-c-propyl48053.341–c-Pentyl430>100 42–c-Hexyl23063.543–C(O)Ph130 3.844–C(O)-4-pyridyl3919.345–C(O)Bn585 3.546–C(O)NHPh167.147–C(O)NH-2-Cl-Ph3018.548–C(O)NH-4-Cl-Ph10>100 49–C(O)NH-2-MeO-Ph276 2.450–C(O)NH-4-MeO-Ph10 2.551–C(O)NH-4-CNPh3 6.952–C(O)NH-c-hexyl138 6.0532-Quinoline–2225.0542-Quinoxaline–147.5Figure2.X-ray of compound38(orange)and41(cyan)occupying the S3subsite directly from S1.J.P.Caldwell et al./Bioorg.Med.Chem.Lett.24(2014)5455–54595457occupy the S3subsite;but,instead occupied a different subsite altogether(Fig.3).14This subsite is a narrow hydrophobic region which extends into a solvent-exposed region of the BACE1 enzyme.15Compounds occupying this subsite make favorable hydrophobic interactions with the side chains of Ile171and Lys168 as well as take advantage of a hydrogen bond interaction with the carbonyl of Phe169.To maximize the probability of identifying inhibitors with high CNS exposure,urea isosteres53and54were prepared to reduce the number of hydrogen bond donors present in the urea analogs.In Figure3,superimposition of the X-ray co-crystal structure of inhibitors44,46,and53with BACE1illustrates that the quinolinyl analog50binds in a similar manner as both the amide44and urea46analogs.Selectivities of44,46,and53for BACE1versus the closely related aspartyl protease cathepsin D (Cat-D)were494-,546-,and1111-fold,respectively.As a refer-ence,the initial lead compound1was not selective versus cathep-sin D(cathepsin D K i=31nM).The amide44and urea46failed to achieve measurable plasma levels in rats when dosed orally,however both compounds exhib-ited modest levels of plasma A b40lowering when administered sub-cutaneously(s.c.)to rats at a dose of30mg/kg(Table3).On the other hand,we were gratified when quinoline53displayed modest oral bioavailability in a rat pharmacokinetic study at a dose of10mg/kg,p.o.16Furthermore,compound53not only exhibited A b40lowering when dosed s.c.but also upon oral administration.Unfortunately,no cortical A b40reduction was observed when53 was dosed to CRND8mice.12Although53was potent in the enzy-matic assay(K i=22nM)it suffered from a pronounced cell shift when tested in the whole cell assay,WC A b40EC50=258nM.12Fur-thermore,when dosed with a known p-glycoprotein(Pgp)inhibi-tor,the cellular potency improved to110nM,the leftward shift suggesting that53was a Pgp substrate.In summary,employment of a modular synthetic strategy allowed us to access structurally distinct iminoheterocycles and enabled rapid SAR exploration of a series of small-molecule BACE1 inhibitors.Our understanding of the SAR was refined through X-ray crystallographic studies,which showed that a subset of these inhibitors serendipitously occupied a different subsite of the BACE1enzyme.Optimization studies exploiting this unique subsite led to the discovery of53,a compound possessing modest expo-sure that lowered plasma A b40levels upon subcutaneous and oral dosing.AcknowledgmentsWe would like to thank Dr.William Greenlee,Dr.Eric Parker, and Dr.John Hunter for their support and guidance of this work, Dr.Jesse Wong for preparation of intermediates,Ron Manning for genotyping and maintenance of the CRND8mice,and Dr.Jared Cumming for suggestions and helpful discussions in the prepara-tion of this e of the IMCA-CAT beamline17-ID at the Advanced Photon Source was supported by the companies of the Industrial Macromolecular Crystallography Association through a contract with the Center for Advanced Radiation Sources at the University of Chicago.References and notes1.Hardy,J.;Selkoe,D.J.Science2002,297,353.2.(a)Wang,Y.-S.;Strickland,C.;Voigt,J.H.;Kennedy,M.E.;Beyer,B.M.;Senior,M.M.;Smith,E.M.;Nechuta,T.L.;Madison,V.S.;Czarniecki,M.;McKittrick,B.A.;Stamford,A.W.;Parker,E.M.;Hunter,J.C.;Greenlee,W.J.;Wyss,D.F.J.Med.Chem.2010,53,942;(b)Zhu,Z.;Sun,Z.-Y.;Ye,Y.;Voigt,J.;Strickland,C.;Smith,E.M.;Cumming,J.;Wang,L.;Wong,J.;Wang,Y.-S.;Wyss,D.F.;Chen,X.;Kuvelkar,R.;Kennedy,M.E.;Favreau,L.;Parker,E.;McKittrick,B.A.;Stamford,A.W.;Czarniecki,M.;Greenlee,W.J.;Hunter,J.C.J.Med.Chem.2010,53,951;(c)Cumming,J.N.;Smith,E.M.;Wang,L.;Misiaszek,J.;Durkin,J.;Pan,J.;Iserloh,U.;Wu,Y.;Zhu,Z.;Strickland,C.;Voigt,J.;Chen,X.;Kennedy,M.E.;Kuvelkar,R.;Hyde,L.A.;Cox,K.;Favreau,L.;Czarniecki,M.F.;Greenlee,W.J.;McKittrick,B.A.;Parker,E.M.;Stamford,A.W.Bioorg.Med.Chem.Lett.2012,22, 2444;(d)Stamford,A.W.;Scott,J.D.;Li,S.W.;Babu,S.;Tadesse,D.;Hunter,R.;Wu,Y.;Misiaszek,J.;Cumming,J.N.;Gilbert,E.J.;Huang,C.;McKittrick,B.A.;Hong,J.;Guo,T.;Zhu,Z.;Strickland,C.;Orth,P.;Voigt,J.H.;Kennedy,M.E.;Chen,X.;Kuvelkar,R.;Hodgson,R.;Hyde,L.A.;Cox,K.;Favreau,L.;Parker,E.M.;Greenlee,W.J.ACS Med.Chem.Lett.2012,3,897.3.(a)Malamas,M.S.;Erdei;Edwards,P.D.;Albert,J.S.;Sylvester,M.;Aharony,D.;Andisik,D.;Callaghan,O.;Campbell,J.B.;Carr,R.A.;Chessari,G.;Congreve,M.;Frederickson,M.;Folmer,R.H.A.;Geschwindner,S.;Koether,G.;Kolmodin, K.;Krumrine,J.;Mauger,R.C.;Murray,C.W.;Olsson,L.L.;Patel,S.;Spear,N.;Tian,G.J.Med.Chem.2007,50,5912;(b)Barrow,J.C.;Stauffer,S.R.;Rittle,K.E.;Ngo,P.L.;Yang,Z.;Selnick,H.G.;Graham,S.L.;Munshi,S.;McGaughey,G.B.;Holloway,M.K.;Simon,A.J.;Price,E.A.;Sankaranarayanan,S.;Colussi,D.;Tugusheva,K.;Lai,M.T.;Espeseth,A.S.;Xu,M.;Huang,Q.;Wolfe,A.;Pietrak,B.;Zuck,P.;Levorse,D.A.;Hazuda,D.;Vacca,J.P.J.Med.Chem.2008,51,6259;(c)Malamas,M.S.;Erdei,J.;Gunawan,I.;Turner,J.;Hu,Y.;Wagner,E.;Fan,K.;Chopra,R.;Olland, A.;Bard,J.;Jacobsen,S.;Magolda,R.L.;Pangalos,M.;Robichaud,A.J.J.Med.Chem.2010,53,1146;(d)Rueeger,H.;Rondeau,J.M.;McCarthy,C.;Moebitz,H.;Tintelnot-Blomley,M.;Neumann,U.;Desrayaud,S.Bioorg.Med.Chem.Lett.2011,21,1942;(e)Probst,G.;Xu,Y.-Z.Expert Opin.Ther.Patents2012,22,511.4.Coordinates for the X-ray structure of compound1complexed with BACE1have been deposited in the Protein Data Bank(),and can be accessed under PDB3L5E.5.Coldham,I.;Crapnell,K.M.;Fernandez,J.-C.;Moseley,J.D.;Rabot,.Chem.2002,67,6185.6.Winkler,J.D.;Axten,J.;Hammach,A.H.;Kwak,Y.-S.;Lengweiler,U.;Lucero,M.J.;Houk,K.N.Tetrahedron1998,54,7045.7.(a)Karady,S.;Amato,J.S.;Weinstock,L.M.Tetrahedron Lett.1984,25,4373;(b)Jones,J.B.;Keitz,P.;Cheng,.Chem.1994,59,7671;(c)Eriksson,M.;Napolitano,E.;Xu,J.;Kapadia,S.;Byrne,D.;Nummy,L.;Grinberg,N.;Shen,S.;Lee,H.;Farina,V.Chimia2006,60,566.Figure3.X-ray of compounds44(cyan),46(orange),and53(yellow).Note that the substituents designed to reside in S3instead occupy a different hydrophobic region making favorable hydrophobic interactions with the Ile171and Lys168residues as well as forming a hydrogen bond with the carbonyl of Phe169.Table3In vivo evaluation of selected compoundsCompound AUC0–6h(nM-h)@10mg/kg A b40lowering plasma(30mg/kg,sc)(%)A b40lowering plasma(30mg/kg,po)44063Not determined51065Not determined533066555%5458J.P.Caldwell et al./Bioorg.Med.Chem.Lett.24(2014)5455–54598.Terao,Y.;Kotaki,H.N.;Achiwa,K.Chem.Pharm.Bull.1985,33,2766.9.Yu,J.-Q.;Corey,.Lett.2002,4,2727.10.Gartiser,T.;Selve,C.;Delpuech,J.-J.Tetrahedron Lett.1983,24,1609.11.Papeo,G.;Posteri,H.;Vianello,P.;Varasi,M.Synthesis2004,17,2886.12.Inhibitor K i values at purified human BACE1were determined using a FRET-peptide substrate hydrolysis assay.Cellular IC50values for reduction of A b40 production were determined in stably transfected HEK293-APPswe/lon cells.The protocols for these assays have been previously described in Ref.2b(Zhu et al.).All values reported are the average of a minimum of two independent determinations.CRND8-APP mice are models for early onset(familial)AD that express human APP containing both Swedish and London mutations that enhance the rate of APP cleavage by BACE1and favor production of A b42overA b40in the c-secretase cleavage step as described by Hyde,L.A.;Kazdoba,T.M.;Grilli,M.;Lozza,G.;Brussa,R.;Zhang,Q.;Wong,G.T.;McCool,M.F.;Zhang, L.;Parker,E.M.;Higgins,G.A.Behav.Brain Res.2005,160,344.13.Coordinates for the X-ray structure of compounds38and41complexed withBACE1have been deposited in the Protein Data Bank(),and can be accessed under PDB4R8Y and PDB4R9,respectively.14.Coordinates for the X-ray structure of compounds44,46,and54complexedwith BACE1have been deposited in the Protein Data Bank(),and can be accessed under PDB4R92,PDB4R93and PDB4R95,respectively. 15.This subsite was initially reported in a previous account(see Ref.2b)andtermed the‘A-site’as this subsite was occupied by‘Mode A inhibitors’.Serendipitously,we were able to occupy this subsite using‘Mode B’inhibitors.16.Korfmacher,W.A.;Cox,K.A.;Ng,K.J.;Veals,J.;Hsien,Y.;Wainhaus,S.;Broske,L.;Prelusky,D.;Nomeir,A.;White,R.E.Rapid Commun.Mass.Spectrom.2001, 15,335.J.P.Caldwell et al./Bioorg.Med.Chem.Lett.24(2014)5455–54595459。