Essays

Biphenyl/diphenyl ether renin inhibitors: Filling the S1 pocket of renin via the S3 pocket

Description
Biphenyl/diphenyl ether renin inhibitors: Filling the S1 pocket of renin via the S3 pocket
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Biphenyl/diphenyl ether renin inhibitors: Filling the S1 pocket of renin via the S3 pocket  Jing Yuan a, ⇑ , Robert D. Simpson a , Wei Zhao a , Colin M. Tice a , Zhenrong Xu a , Salvacion Cacatian a , Lanqi Jia a ,Patrick T. Flaherty a , Joan Guo a , Alexey Ishchenko a , Zhongren Wu a , Brian M. McKeever a , Boyd B. Scott a ,Yuri Bukhtiyarov a , Jennifer Berbaum a , Reshma Panemangalore a , Ross Bentley b , Christopher P. Doe b ,Richard K. Harrison a , Gerard M. McGeehan a , Suresh B. Singh a , Lawrence W. Dillard a , John J. Baldwin a ,David A. Claremon a a Vitae Pharmaceuticals, 502 West Office Center Drive, Fort Washington, PA 19034, USA b GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA a r t i c l e i n f o  Article history: Received 18 April 2011Revised 8 June 2011Accepted 10 June 2011Available online 17 June 2011 Keywords: Renin inhibitorsHypertensionBiphenylDiphenyl etherStructure-based designX-ray crystallography a b s t r a c t Structure-baseddesignledtothediscoveryofanovelclassofrenininhibitorsinwhichanunprecedentedphenyl ring filling the S1site is attached tothe phenyl ring filling theS3 pocket. Optimizationfor severalparameters including potency in the presence of human plasma, selectivity against CYP3A4 inhibitionandimprovedratoralbioavailabilityledtotheidentificationof  8d  whichdemonstratedantihypertensiveefficacy in a transgenic rat model of human hypertension.   2011 Elsevier Ltd. All rights reserved. The renin angiotensin aldosterone system (RAAS) is a centralmechanism by which mammalian blood pressure is controlled. 1 The aspartyl protease renin performs the first and rate-limitingstep in the RAAS, cleavage of the decapeptide angiotensin I fromthe N-terminus of the glycoprotein angiotensinogen (Fig. 1).Althoughreninhas longbeenviewedas a desirabletarget for anti-hypertensives, identification of orally bioavailable, low molecularweight inhibitors proved challenging. Efforts by many researchgroups during the 1980s focused on modified peptides, leading tothe identification of potent compounds such as remikiren ( 1 ). 2,3 However, none of these compounds was ultimately deemed suit-able for full development. 4 Finally in 2007, Novartis brought thefirst renin inhibitor, aliskiren ( 2 ), to market. 5,6 Comparison of the structural features of remikiren ( 1 , Fig. 1) tothe amino acid residues surrounding the cleavage site of the natu-ral substrate angiotensinogen illustrates the correspondencebetween groups occupying the S1, S2 and S3 pockets 7 in therespective structures (PDB code: 3D91). Publication of the firstX-ray structure of human renin (PDB code: 1RNE) 8 revealed thatthe S1 and S3 pockets form a contiguous superpocket, suggestingthat it might be possible to attach the group filling S3 to the groupoccupying S1. Early attempts to reduce this idea to practice led tocompounds with substantially reduced potency. 9,10 However,extensive structure-based design work at Ciba-Geigy, includingdeletion of the peptide backbone, led to the identification of  2 . 11–13 The X-ray crystal structure of   2  bound to renin (PDB code:2V0Z) confirmed that the methoxyphenyl group occupies S3 andthat the two  i -Pr groups occupy S1 and S1 0 .We recently described the discovery of potent, orally bioavail-able alkyl amine renin inhibitor  3  (Fig. 1). 14 This compound con-nects the residues filling the S1 and S3 pockets through abridging piperidine linker. Inspection of the X-ray structure of   3 boundtorenin (PDBcode: 3GW5) suggestedanalternate inhibitordesign strategy where S1 could be filled by direct attachment of asuitable group to the  ortho -position of the phenyl ring occupyingS3. In this Letter, we describe the successful realization of this de-sign strategy to afford potent, orally bioavailable renin inhibitors.Application of Contour™, a proprietary computational struc-ture-based drug design program, initially identified the phenoxygroup represented in general structure  4  as a candidate to fill theS1 site, when combined with an appropriate group at R  1 (Table 1).Introduction of the phenoxy group while retaining thecyclohexylmethyl group at R  1 ( 4a ) substantially reduced potency 0960-894X/$ - see front matter    2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.bmcl.2011.06.043 ⇑ Corresponding author. E-mail address:  jyuan@vitaerx.com (J. Yuan).Bioorganic & Medicinal Chemistry Letters 21 (2011) 4836–4843 Contents lists available at ScienceDirect Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl  comparedto 3 , asexpected. However,truncationof R  1 to i -Bu( 4b ),theP1sidechaininthenaturalsubstrate,greatlyincreasedpotency.The closely related  t  -BuCH 2  group ( 4c ) was much less potent. Fur-ther reducing the size of R  1 to Me ( 4d ) reduced potency comparedto 4b .Thegeminaldimethylsubstitutionwasalsounfavorable( 4e ).Theidentificationofpotentcompound 4b providedthefirstvalida-tion of our design.Asanextstep, wesurveyedreplacementof thesubstitutedeth-ylenediaminechainin 4 withvariouscyclicmoietiesincorporatingbasic amines to give compounds of general structure  5  (Table 2.R  3 =aminocycloalkyl, azacycloalkyl, etc.). Presentation of the basicamine as part of a cyclic structure was intended to retain the keycoulombic interactions with the catalytic aspartates and provideentropic benefit by removing torsional degrees of freedompresentintheearliercompound 4 .Fromthiseffort,weidentified(3 S  ,1 R )-3-aminocyclopentane-1-carboxylic acid derivative  5a  with modestpotency (Table 2). By contrast, the (3 R ,1 R ) and (3 R ,1 S  )-isomers 5b  and  5c  were substantially less potent. Modeling of   5a  in the re-nin binding site suggested that additional hydrophobic interac-tions with the protein could be attained by introduction of asmall lipophilic substituent at the  ortho -position of the distal ringof the diphenyl ether which occupies S1. 2-Methylphenoxy com-pound 5d provedtobe10  morepotentthan 5a .Furtherexamina-tionofour modelindicatedthatahydroxylgroupat the4-positionof the cyclopentane ring syn to the 3-amino group would enjoy afavorable interaction with the Asp32 carboxylate. Thus,  5e  exhib-ited 7   greater potency than  5d , while the epimer  5f   was onlyslightly more potent than  5d . Unfortunately, the more potent ana-logs of general structure  5  also inhibited CYP3A4 with IC 50  values 6 1 l M.Rat pharmacokinetic parameters were determined for  5e ; thecompound showed good oral bioavailability but rapid clearance(Table 3). The rat liver microsome half life (RLM  t  1/2 ) of   5e  was18min, suggesting that oxidative metabolism might well be themechanism of clearance. Introduction of an  ortho -fluorine on theproximal phenyl ring was expected to lock the diphenyl ether into Figure 1.  The N-terminal sequence of angiotensinogen is shown highlighting the amino acids in the P3–P1 0 positions. The P1, P2 and P3 groups in  1  (remikiren), the P1 0 , P1and P3 groups in  2  (aliskiren) and the P1 and P3 groups in  3  are indicated.  Table 1 SAR of diphenyl ethers  4 NHNNHMeOHOHMeOOR 1 4R 2 Compd R  1 R  2 IC 50a,b,c (nM) PRA a,b,d (nM) CYP 3A4 e ( l M) 3 c — — 0.5 13 1 4a  c  -hexCH 2  H 596 4b  i -Bu H 8.8 313 1.1 4c  t  -BuCH 2  H 420 4d  Me H 126 1340 1.9 4e  Me Me 2900 a See Ref. 14 for assay protocols. b IC 50  values are the average of at least two replicates. c Inhibition of 0.3nM of purified recombinant human renin in buffer wasmeasured. d IC 50  in the presence of human plasma. e Inhibition of CYP3A4 in human liver microsomes was measured.  J. Yuan et al./Bioorg. Med. Chem. Lett. 21 (2011) 4836–4843  4837  a favorable conformation and reduce oxidative metabolism; amodest increase in potency and RLM  t  1/2  of 31min were observed( 5g  ).Inpeptidomimeticrenininhibitors,analogswitharylringsatP1aremuchlesspotentthancompoundswithaliphaticgroups. 15 Fur-thermore, the majority of the previously described classes of po-tent, non-peptidic renin inhibitors that bind to the closed form of renin fill S1 with an aliphatic group; 16 however, workers at Sano-fi-Aventis have recently described inhibitors which deploy a phe-nyl ring in S1. 17 Modeling suggested that replacement of thediphenyl ether with a biphenyl system would be tolerated while,potentially, offering superior intrinsic metabolic stability. Unsub-stituted biphenyl analog  6a  demonstrated moderate overall po-tency (Table 4). Modeling indicated that small lipophilicsubstituents could be accomodated at the 3-position of the distalphenyl ring of the biphenyl ring system. Thus,  6c  (Y=3-Me) wasmore potent than both the unsubstituted biphenyl  6a  and the 2-and4-methylisomers 6b and 6d .Anotherimprovementinpotencywasobtainedwith3-chlorocompound 6e .Modelingindicatedthatthe chlorine interacted favorably with Val30 and Val120 in the S1pocket. Unfortunately,  6e  suffered a greater loss in potency in thepresence of plasma than  6c . However, both  6c  and  6e  were morestableinRLMthan 5e . Introductionofsmall lipophilicsubstituentsat the 3-position of the proximal phenyl ring was predicted to in-crease potency by locking the biphenyl system into a favorableconformation for binding to renin; this was borne out with fluoroanalog 6f  andchloroanalog 6i .Introductionoffluorineatthe4-po-sition( 6g  )andparticularlythe5-position( 6h )reducedpotency.Asseen with  6e , chloro analogs  6i–k  , while potent when assayed inbuffer, suffered >20   losses in potency in presence of plasma.  Table 2 SAR of 3-aminocyclopentane carboxamides  5 N R 3 OHOHMeOOXY 3456 Compd R  3 X Y IC 50a,b,c (nM) PRA a,b,d (nM) CYP 3A4 e ( l M) RLM  t  1/2  (min) 5a NH 2 H H 195 5b NH 2 H H 1437 5c NH 2 H H >5000 5d NH 2 H 2-Me 19 306 0.4 5e NH 2 OH H 2-Me 2.7 32 0.5 18 5f  NH 2 OH H 2-Me 11 233 1.0 5g  NH 2 OH 3-F 2-Me 1.2 19 0.4 31 a See Ref. 14 for assay protocols. b IC 50  values are the average of at least two replicates. c Inhibition of 0.3nM of purified recombinant human renin in buffer was measured. d IC 50  in the presence of human plasma. e Inhibition of CYP3A4 in human liver microsomes was measured.  Table 3 Rat pharmacokinetic parameters a,b Compd Oral  C  max  (ng/mL) Oral  t  max  (h) Oral AUC(0–t) ngh/mL Oral  t  1/2  (h) iv CL (mL/min.kg) Vss (L/kg)  F   (%) 3  73 3.3 472 nd 33 nd 13 5e  185 4.3 728 3.8 66 12 34 6i  60 3.3 264 4.1 39 9 8 7b c 5.6 0.2 23 50 50 39 2 8a  45 2.7 186 15.3 80 25 14 8b  71 2.3 295 4.9 58 19 13 8d  111 3.3 338 3.1 53 83 19 a All compounds were dosed as fumarate salts at 2mg/kg iv. b All compounds were dosed as fumarate salts at 10mg/kg po except  7b . c Oral dose 7mg/kg.4838  J. Yuan et al./Bioorg. Med. Chem. Lett. 21 (2011) 4836–4843  Many analogs of general structure  6  posessed submicromolar IC 50 values against CYP3A4.An X-ray structure of unsubstituted biphenyl  6a  complexed torenin (PDB code: 3Q3T) confirmed the modeled binding pose(Fig. 2). Strong electrostatic interactions of the protonated cyclo-pentylamine with the Asp32 and Asp215 side chain carboxylatesprovideakey,anchoringinteraction.Inaddition,thehydroxylsub-stituent on the cyclopentane donates a hydrogen bond to theAsp32 carboxylate. The tertiary alcohol moiety hydrogen bondswith Ser219, as observed in the crystal structure of   3  bound to re-nin (PDB code: 3GW5). The ether oxygen at the terminus of themethoxybutyl chainoccupyingtheS3subpocketformsahydrogenbond with the backbone NH of Tyr14, as seen for the analogousgroups in  2  (PDB code: 2V0Z) and  3  (PDB code: 3GW5). The S3–S1 superpocket filled by the biphenyl moiety, has the phenyl ringin the S3 pocket interacting with the positive edge of Phe117 andthe second phenyl ring in the S1 pocket interacting with Tyr75.Modeling based on 3Q3T predicted that the  m -positions of boththe phenyl rings have ample room to allow small substituents tooccupy hydrophobic spaces in the S1 and S3 pockets and provideimproved affinity as observed for  6c ,  6e–g   and  6i–k  .Inrat,  6i  demonstratedreducedivclearance, consistentwithitslonger RLM  t  1/2 , but reduced orally bioavailability, compared to  5e (Table3).Inaddition,theIC 50 valuefor 6i inthepresenceofhumanplasma (PRA) was unsatisfactory due to projected inferior efficacyin vivo. Our subsequent efforts were directed to identify analogswith improved PRA, reduced CYP3A4 inhibition and increased oralbioavailability.In earlier work with related compounds, our general approachto improving PRA and reducing CYP inhibition was to introduceadditional polar functionality into the molecule. 18 Specifically, inthese earlier compounds, removal of the hydroxyl group and Figure 2.  X-raystructureof  6a  boundtorenin(PDBcode: 3Q3T). Themolecularsurfacecorrespondingtocertainaminoacidresiduesof thebindingsitewasnotrenderedtoprovide an unobstructed view of the inhibitor and its interactions.  Table 4 SAR of biphenyls  6 NOHHMeOOHNH 2 OX Y 3456 Compd X Y IC 50a,b,c (nM) PRA a,b,d (nM) CYP 3A4 e ( l M) RLM  t  1/2  (min) 6a  H H 37 420 4.0 6b  H 2-Me 23 126 0.5 6c  H 3-Me 10 137 1.6 67 6d  H 4-Me 20 405 0.9 6e  H 3-Cl 2.1 79 0.2 73 6f   3-F 3-Me 1.4 12 0.8 6g   4-F 3-Me 7.1 50 6h  5-F 3-Me 4700 6i  3-Cl 3-Me 1.6 35 0.2 102 6j  3-F 3-Cl 1.9 47 0.5 6k   3-Cl 3-Cl 3.7 131 2.0 a See Ref. 14 for assay protocols. b IC 50  values are the average of at least two replicates. c Inhibition of 0.3nM of purified recombinant human renin in buffer was measured. d IC 50  in the presence of human plasma. e Inhibition of CYP3A4 in human liver microsomes was measured.  J. Yuan et al./Bioorg. Med. Chem. Lett. 21 (2011) 4836–4843  4839  replacement of the 4-methoxybutyl chain occupying S3 sp with anacylated 2-aminoethoxy chain greatly improved PRA inhibition.By contrast, in this series, the same modification decreased po-tency both in buffer and in the PRA ( 7a  vs  6f  ). On the other hand,retention of the hydroxyl and replacement of the 4-methoxybutylchain with a 3-(methoxycarbonylamino)propyl chain afforded  7b with improved PRA (Table 5). Replacement of fluorine with chlo-rine ( 7c)  was well tolerated. 3-(Acetylamino)propyl analogs  7d and  7e  also had good activity in the PRA and offered a modestreduction in CYP3A4 inhibition. Compound  7b  was advanced torat PK. Despite its long RLM  t  1/2 ,  7b  was subject to rapid clearancein vivo and had low oral bioavailability (Table 3).Replacementofthepiperidineringin 6 withmorpholineofferedan alternative strategy to increase overall polarity. Effecting thischange in  6f   gave  8a , with a 3   reduction in renin potency but a30   reduction in CYP3A4 inhibition (Table 6). Changing fluorineto chlorine ( 8b ) and methyl to ethyl ( 8c ) improved renin potency(Table 6). Combining, these afforded  8d  with an excellent in vitroprofile. Interestingly, replacing the ethyl group in  8d  with isopro-pyl ( 8e ) had little effect on renin potency but led to a substantialincrease in CYP3A4 inhibition. Oral bioavailability was demon-stratedfor  8a ,  8b  and 8d , but clearanceremainedhighfor all threecompounds (Table 3). Compounds  8a – f   also demonstrated excel-lent selectivity for renin over three other aspartyl proteases:  b -secretase, cathepsin D and cathepsin E (<10% inhibition at 10 l M).Theantihypertensiveefficacyof   8d  wasdemonstratedinadou-ble transgenic rat (dTGR) model of human hypertension, in whichthe animals express both human renin and human angiotensino-gen (Fig. 3). 19,20 Oral administration of 10mg/kg of   8d  led to a sta-tistically significant reduction in mean arterial blood pressure(MABP) sustained for more than 12h. At the nadir, MABP was re-duced by >20mm Hg.  Table 5 SAR of polar side chain analogs  7 NR 5 R 4 HOHNH 2 OXY Compd X Y R  4 R  5 IC 50a,b,c (nM) PRA a,b,d (nM) CYP 3A4 e ( l M) RLM  t  1/2  (min) 7a  F Me H MeOCONHCH 2 CH 2 O– 5.2 52 0.5 7b  F Me OH MeOCONH(CH 2 ) 3 – 1.6 4.4 3.8 567 7c  Cl Me OH MeOCONH(CH 2 ) 3 – 1.1 2.8 1.1 7d  F Me OH MeCONH(CH 2 ) 3 – 1.6 1.7 5.3 460 7e  Cl Me OH MeCONH(CH 2 ) 3 – 2.1 4.4 3.5 2033 a See Ref. 14 for assay protocols. b IC 50  values are the average of at least two replicates. c Inhibition of 0.3nM of purified recombinant human renin in buffer was measured. d IC 50  in the presence of human plasma. e Inhibition of CYP3A4 in human liver microsomes was measured.  Table 6 SAR of morpholine analogs  8 ONOHHOHNH 2 OMeOYX Compd X Y IC 50a,b,c (nM) PRA a,b,d (nM) CYP 3A4 e ( l M) RLM  t  1/2  (min) 8a  F Me 3.8 17 25 99 8b  Cl Me 1.8 11 18 8c  F Et 2.5 8.1 17 8d  Cl Et 1.1 4.0 10 8e  Cl  i -Pr 0.7 5.2 0.4 8f   Cl MeS 2.3 26 2 a See Ref. 14 for assay protocols. b IC 50  values are the average of at least two replicates. c Inhibition of 0.3nM of purified recombinant human renin in buffer was measured. d IC 50  in the presence of human plasma. e Inhibition of CYP3A4 in human liver microsomes was measured.4840  J. Yuan et al./Bioorg. Med. Chem. Lett. 21 (2011) 4836–4843
Search
Similar documents
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks