Optimization of spirocyclic proline tryptophan hydroxylase-1 inhibitors

Optimization of spirocyclic proline tryptophan hydroxylase-1inhibitorsDaniel R.Goldberg a ,⇑,Stéphane De Lombaert a ,Robert Aiello a ,Patricia Bourassa a ,Nicole Barucci a ,Qing Zhang a ,Vishwas Paralkar a ,Adam J.Stein b ,Melissa Holt b ,Jim Valentine a ,William Zavadoski aa Karos Pharmaceuticals,401Winchester Ave.,5Science Park,New Haven,CT 06511,United States bCayman Chemical,5025Venture Dr.,Ann Arbor,MI 48108,United Statesa r t i c l e i n f o Article history:Received 7November 2016Revised 19December 2016Accepted 20December 2016Available online 23December 2016Keywords:Tryptophan hydroxylase-1Serotonin TPH-15-HTa b s t r a c tAs a follow-up to the discovery of our spirocyclic proline-based TPH1inhibitor lead,we describe the opti-mization of this scaffold.Through a combination of X-ray co-crystal structure guided design and an in vivo screen,new substitutions in the lipophilic region of the inhibitors were identified.This effort led to new TPH1inhibitors with in vivo efficacy when dosed as their corresponding ethyl ester prodrugs.In particular,15b (KAR5585),the prodrug of the potent TPH1inhibitor 15a (KAR5417),showed robust reduction of intestinal serotonin (5-HT)levels in mice.Furthermore,oral administration of 15b generated high and sustained systemic exposure of the active parent 15a in rats and dogs.KAR5585was selected for further pharmacological evaluation in disease models associated with a dysfunctional peripheral 5-HT system.Ó2016Elsevier Ltd.All rights reserved.Over the years,a growing number of diseases have been associ-ated with a dysfunctional peripheral serotonin system.1These include for example carcinoid syndrome,2gastro-intestinal dis-eases,3–6some fibrotic diseases,7–10obesity associated-fatty liver disease,11atherothrombotic diseases,12–14mastocytosis,15some cancers 16–18and immune diseases,19,20and pulmonary arterial hypertension.21Although it is not always clear if serotonin (5-HT)represents a mere biomarker of these diseases or if it plays an active pathophysiological role,5-HT displays well-known mito-genic and vasoconstrictive properties that may account for its effects.22,23Other mechanisms such as serotonylation processes have also been invoked.24Thus,modulation of peripheral 5-HT tone and 5-HT overproduction in diseased tissues presents a new attractive opportunity to treat these diseases.Intestinal ente-rochromaffin cells located in the gut are responsible for most of the peripheral 5-HT production.5-HT is stored in platelets,with very little amount circulating free,and it does not cross the blood-brain barrier.Therefore,5-HT levels in the periphery and in the CNS are independently regulated.Tryptophan hydroxylase 1(TPH1)is the rate limiting enzyme in the biosynthesis of periph-eral 5-HT,while the centrally located TPH2controls 5-HT produc-tion in the brain.Since reducing central 5-HT levels has been associated with undesirable psychiatric effects,TPH inhibitors,either highly selective for the TPH1isoform or devoid of CNS penetration would be useful to modulate 5-HT peripheral levels.Gut-selective phenylalanine-derived TPH inhibitors have been reported and some of them decrease 5-HT levels in the gut.3,25,26The systemic agent Telotristat ethyl (aka LX1032and telotristat etiprate),a TPH1inhibitor prodrug,showed efficacy in carcinoid syndrome patients unresponsive to somatostatin analog therapy when administered at 250mg three times per day orally during a 12-week period.27Using a combination of a structure-based design and an in vivo efficacy screen,we have discovered several distinct series of TPH1inhibitors.28,29In particular,the spirocyclic proline series provided an attractive starting point to further optimize the inhibitors’efficacy and pharmacokinetic properties,given it already displayed a combination of high in vitro potency and significant mucosal 5-HT reduction in vivo .For example,when dosed orally as its ester prodrug 1b ,the potent TPH1inhibitor 1a elicited a significant 5-HT decrease in gut and platelet rich plasma (PRP)in rats (Scheme 1).28Pharmacokinetic studies showed an 18-fold increase in the plasma exposure (AUC)of 1a after oral administration of 1b vs.1a ,indicating that this prodrug strategy was effective at increasing the systemic delivery of the active inhibitor in rodents.Aiming for an ideal target development candidate profile supportive of a once-a-day regimen,we sought to identify new compounds with more sustained TPH1inhibition compared to 1a .A qualitative metabolite ID study of 1b(d)(diastereomer of 1b )in rat hepatocytes revealed that,in addition to the primary ester hydrolysis pathway,the methyl group of the pyrazole substituent/10.1016/j.bmcl.2016.12.0530960-894X/Ó2016Elsevier Ltd.All rights reserved.⇑Corresponding author.E-mail address:dgoldberg@ (D.R.Goldberg).Bioorganic &Medicinal Chemistry Letters 27(2017)413–419Contents lists available at ScienceDirectBioorganic &Medicinal Chemistry Lettersjournal homepage:www.else v i e r.c o m /l o c a t e /b m clTable1Enzymatic activity of biarylderivatives.Cpd R TPH1IC50(nM)Cpd R TPH1IC50(nM) 2a Cl4607a323a258a32 4a269a37414 D.R.Goldberg et al./Bioorganic&Medicinal Chemistry Letters27(2017)413–419underwent further hydroxylation.Further processing led to the carboxylic acid and sulfate metabolites (Scheme 2).Although these metabolites were not quantified,this study sug-gested that modifications of the pyrazole group could lead to phar-macokinetic improvements in rats,and perhaps ultimately in humans.The known plasticity of the hydrophobic pocket of TPH1apparent in published X-ray structures supported the notion that a broad structural diversity in the corresponding lipophilic region of the inhibitor should be tolerated to maintain high potency.30Since complete removal of the pyrazole group resulted in a sig-nificant loss of potency,a focused library was produced to identify aryl alternatives to the 4-chloro substituent able to restore potency.Within this set,a small number of bicyclic aromatics led to compounds with a level of potency in the range of 1a (Table 1).Despite their intrinsic potency,these compounds failed to demonstrate robust 5-HT reduction in vivo when administered orally at 50mg/kg to mice,31either by themselves or as their cor-responding ethyl ester prodrugs (data not shown).Potency and efficacy of some biaryl derivatives could be restored once the pyra-zole group was re-introduced (Table 2).Since the prodrug 13b displayed a robust reduction of mucosal 5-HT in mice after oral administration,its PK in rats was deter-mined (Table 3).Interestingly,despite the observed in vivo efficacy,the overall oral exposure of the active inhibitor was low,raising the possibility that active metabolites may be generated.In particular,the two mono-and the di-hydroxymethyl derivatives of 13a were determined to be potent TPH1inhibitors enzymatically too (data not shown).Therefore,we focused on modifying the pyrazole directly to overcome the potential metabolic issues associated with this group.Replacements of this heterocycle with a simple phenyl group provided a potent TPH1inhibitor (15a ).A high resolution crystal structure (1.9–2.0Å)of 15a bound to TPH1was obtained (pdb code:5L01),revealing key interactions (Fig.1).The proline NH forms a hydrogen bond with T265,while the carboxylate engages R257.The chloro phenyl group p -stacks with the flexible Y235residue.While similar interactions had also been observedTable 2Enzymatic and in vivo activity of compounds with aryl alternatives to chloro substituent in 1a .Cpd R TPH1IC 50(nM)aEthyl ester prodrug%5-HT Reduction (50mg/kg PO)b1a261b 6112a2212b6013a 2013b 7014a2314b 46a IC 50values reported as the mean of at least two determinations.bData expressed as mean changes relative to vehicle control.All changes relative to vehicle significant by 1-Way ANOVA (P <0.001).Table 1(continued )aaaIC 50values reported as the mean of at least two determinations.Table 3Plasma rat PK for 13a measured after 13b po dosing.PK parameter a Result iv t 1/2(h) 3.6±0.71po C max (ng/mL)240±60.9po AUC 0-last (h*ng/mL)1634±321aProtocols:25mg/kg po in 0.5%MC in water;1mg/kg iv in 1%DMSO/5%Solutol HS15/94%saline;all values expressed as mean ±SD.D.R.Goldberg et al./Bioorganic &Medicinal Chemistry Letters 27(2017)413–419415Fig.1.(A)Binding conformation of15a(brown,grey)in TPH1(pdb code5L01).Dotted lines depict H-bond interactions between the inhibitor and TPH1residues R257 sidechain and T265backbone(clearly resolved).The orange sphere is the Fe+2atom.(B)Overlay between15a(brown,grey)and the phenyl alanine based TPH1inhibitor LP-533401(pdb code:3HF8)(magenta).30Table4Enzymatic activity of compounds with aryl and heteroaryl alternatives to pyrazole in1a.a a aCl18aO1926aClF3534aN73 19aFFF2127aFF3535aNO77416 D.R.Goldberg et al./Bioorganic&Medicinal Chemistry Letters27(2017)413–419with the known phenylalanine-derived TPH1inhibitors,30R257undergoes a conformational change to accommodate the proline moiety while maintaining the conformation of S265and S337.As anticipated,the three dimensional spirocyclic proline uniquely fills the amino acid binding pocket more fully than the phenylalanine-based chemotypes.The new phenyl substituent is not pointing outwards to the solvent,but resides in a wide but rather short lipo-philic pocket of TPH1capped by a hydrophobic triad consisting of Y312,F241and L242.This binding conformation suggested that further substitution in the meta position of the R-group phenyl ring would be tolerated,while only small groups would be allowed in the para position.Therefore,a focused library was prepared to identify tolerated substituted aryl or heteroaryl alternatives to the pyrazole (Table 4).Table 4(continued )aaaaIC 50values reported as the mean of at least two determinations.parison between 17a (cyan)and 15a (grey,brown).(A)Y235re-orientation allows better p interaction with the chloro phenyl ring.(B)One oxygen of the sulfonyl group engages in a hydrogen bond with a conserved water molecule.Table 5Mucosal 5-HT reduction after oral dosing of ethyl ester prodrugs of potent TPH1inhibitors in mice.Ethyl ester prodrugdosed poActive parent generated%5-HT reduction a(50mg/kg)(10mg/kg)15b 15a 71⁄⁄⁄55⁄⁄⁄16b 16a 42⁄⁄15⁄17b 17a 38⁄⁄ND b 18b 18a 46⁄⁄⁄19⁄19b 19a 64⁄⁄⁄31⁄⁄20b 20a 37⁄⁄30⁄⁄21b 21a 59⁄⁄⁄33⁄⁄22b 22a 54⁄⁄⁄24⁄⁄25b25a61⁄⁄⁄38⁄⁄aData expressed as mean changes relative to vehicle control.All changes relative to vehicle significant by 1-Way ANOVA:P <0.001⁄⁄⁄,P <0.01⁄⁄,P <0.1⁄.bNot determined.D.R.Goldberg et al./Bioorganic &Medicinal Chemistry Letters 27(2017)413–419417Phenyl groups substituted in the meta position with a wide variety of functional groups,provided potent TPH1inhibitors. Large alkyl groups,such as tBu(21a)were well tolerated.Methyl sulfone(17a)and unsubstituted sulfonamides(16a)were particu-larly active,in particular when compared to carboxamide(28a)or reversed-sulfonamide analogs(29a).Even a larger pyrrolidinone substituent retained some potency(35a),confirming the rather wide space available in the corresponding enzymatic pocket. Substitution of the sulfonamide nitrogen led to a decrease in potency(31a).As predicted,para-substitution,in particular with large groups,resulted in a loss of activity.Basic pyridine groups (33a,34a)were less favorable than aryl groups for activity.A high resolution X-ray crystal structure of17a(KAR5489,1.50–1.55Å)bound to TPH1(pdb code:5TPG),confirmed the expected binding conformation(Fig.2).While15a and17a adopt a similar binding pose in the TPH1active site,the side-chain of Y235adopts a slightly different orientation.In17a,Y235rotates closer to the chloro phenyl ring to engage in a stronger p–p inter-action with the ligand’s aromatic ring.As the sulfonyl group sits further into the lipophilic pocket,a water-mediated hydrogen bond is also observed.Given that a large number of compounds were potent TPH1 inhibitors in vitro,we next sought to assess their effects in vivo.A set of potent inhibitors was evaluated orally as their corresponding ethyl esters in mice,at doses of50and10mg/kg and mucosal5-HT levels were determined(Table5).While most orally administered prodrugs robustly decreased intestinal5-HT concentrations at50mg/kg,their efficacy dropped off significantly at the lower10mg/kg dose.Interestingly,the unsubstituted analog15b,remained efficacious at the low dose. Therefore,the PK of15b was elucidated in rat,a species used in various pharmacodynamics models(Table6).Compared to1b,15b generated the active parent15a with a lower C max and larger AUC,a PK profile more attractive to achieve sustained inhibition with less peak-to-trough variations.Although15a(and the other spirocyclic proline analogs)also showed potent inhibition of TPH2(IC50=7nM),an enzyme responsible for5-HT production in the CNS,15a and15b were found to be essentially confined to peripheral tissues in rats (data not shown).To verify that effective bioconversion of the prodrug also occurred in higher species,15b was administered orally to beagle dogs(Table7).High exposure of15a was achievable in dogs after administra-tion of the ethyl ester prodrug15b.The synthesis of15b and15a is representative of the method used to prepare the compounds described herein(Scheme3).In brief,(R)-1-(2-bromo-4-chlorophenyl)-2,2,2-trifluoroethanol(16) was added to2-amino-4,6-dichloropyrimidine to provide interme-diate17,to which the suitable doubly protected(S)-spirocyclic proline ester reagent was added.32Regioselective Pd-catalyzed arylation of18with phenylboronic acid afforded the biaryl inter-mediate19.To obtain15b,the Cbz protecting group was removed from19via in situ production of TMSI,after having found that deprotection by hydrogenation led to the formation of an unsepa-rable by-product lacking the chloro substituent.Final chemical saponification afforded the active TPH1inhibitor15a.This syn-thetic route provided15a and15b in good yield and high chemical and optical purities(>99%ee).In summary,we have described herein an extension of the SAR of our spirocyclic proline-based TPH1inhibitors.15In particular,we have identified15a(KAR5417),a compound when masked as its corresponding ethyl ester prodrug15b(KAR5585),produces robust5-HT reduction in mice intestinal mucosa.Oral administra-tion of KAR5585generated substantial and sustained systemic exposure of the parent TPH1inhibitor in rat and dog.Therefore, this compound was advanced in disease-relevant models of dysfunctional peripheral5-HT.This work and the resulting therapeutic potential of KAR5585will be reported in subsequent publications.Table6Plasma rat PK for15a and15b(iv and po),and comparison with1a generated from1b po dosing.Compound dosed15b iv b15b po c15b po c15a iv d1b po c Compound measured a15b15b15a15a1a PK ParameterCl(ml/min/kg)[%hepatic bloodflow]106.2±22.1[>100]0.901±0.07[1.2]–iv t1/2(h) 3.18±0.2712.1±1.7–iv Vss(L/kg)12.7±2.5–7.8±1.33–po C max(ng/mL)111±21.8722±93.0–2,253±47 po AUC0-last(h*ng/mL)809±1365,374±866–2,924±600a All values expressed as mean±SD.b Protocol for15b iv dosing:1mg/kg in1%DMSO/5%Solutol HS15/94%saline.c Protocol for15b and1b po dosing:25mg/kg in0.5%MC in water,8time points(0.083–24h).d Protocol for15a iv dosing:1mg/kg in2.5%DMSO/5%solutol HS15/92.5%saline.Table7Plasma dog PK for15a and15b(iv and po).Compound dosed15b iv b15b po c15b po c15a iv d Compound measured a15b15b15a15a PK parameterCl(ml/min/kg)[%hepatic bloodflow]0.83±0.15[46]–– 1.2±0.01[4%] iv t1/2(h) 5.05±0.347.59±0.28 4.8±0.76iv Vss(L/kg) 2.34+0.29–0.24±0.03 po C max(ng/mL)744±2182,797±713–po AUC0-last(h*ng/mL)4,564±89932,398±11,975–a All values expressed as mean±SD.b Protocol for15b iv dosing:1mg/kg in1%DMSO/5%Solutol HS15/94%saline.c Protocol for15b po dosing:25mg/kg in0.5%MC in water,8time points(0.083–24h).d Protocol for15a iv dosing:1mg/kg in2.5%DMSO/5%solutol HS15/92.5%saline.418 D.R.Goldberg et al./Bioorganic&Medicinal Chemistry Letters27(2017)413–419AcknowledgementsThe authors thank Scynexis and Santai Labs for their assistance in preparing synthetic intermediates and/orfinal compounds and ChemPartner for their help with DMPK work in this paper. References1.Lesurtel M,Soll C,Graf R,Clavien P-A.Cell Mol Life Sci.2008;65:940–952.2.Kulke MH,O’Dorisio T,Phan A,et al.Endocr Relat Cancer.2014;21:705–714.3.Liu Q,Yang Q,Sun W,et al.J Pharmacol Exp Ther.2008;325:47–55.4.Ghia J-E,Li N,Wang H,et al.Gastroenterology.2009;137:1649–1660.5.Mawe GM,Hoffman JM.Nat Rev Gastroenterol Hepatol.2013;10:473–486.6.Margolis KG,Stevanovic K,Li Z,et al.Gut.2014;63:928–937.7.Königshoff M,Dumitrascu R,Udalov S,et al.Thorax.2010;65:949–955.8.Dees C,Akhmetshina A,Zerr P,et al.J Exp Med.2011;208:961–972.9.Mann Da,Oakley F.Biochim Biophys Acta-Mol Basis Dis.2013;1832:905–910.10.Janssen W,Schymura Y,Novoyatleva T,et al.Biomed Res Int.2015;2015:1–8.11.Haub S,Ritze Y,Ladel I,et al.J Pharmacol Exp Ther.2011;339:790–798.12.Ban Y,Watanabe T,Miyazaki A,et al.Atherosclerosis.2007;195:153–159.13.Hirowatari Y,Hara K,Shimura Y,Takahashi H.J Atheroscler Thromb.2011;18:874–882.14.Sugiura T,Dohi Y,Yamashita S,Hirowatari Y,Fujii S,Ohte N.Atherosclerosis.2016;246:157–160.15.Kushnir-Sukhov NM,Brittain E,Scott L,Metcalfe DD.Eur J Clin Invest.2008;38:953–958.16.Pai VP,Marshall AM,Hernandez LL,Buckley AR,Horseman ND.Breast CancerRes.2009;11:R81.17.Soll C,Jang JH,Riener M-O,et al.Hepatology.2010;51:1244–1254.18.Liang C,Chen W,Zhi X,et al.Mol Cancer.2013;12:1–14.19.Ahern GP.Curr Opin Pharmacol.2011;11:29–33.20.Arreola R,Becerril-Villanueva E,Cruz-Fuentes C,et al.Immunol Res.2015;2015:1–21.21.Thomas M,Ciuclan L,Hussey MJ,Press NJ.Pharmacol Ther.2013;138:409–417.22.Paoletti R,Vanhoutte PM,Brunello N,Maggi FM.Serotonin:From Cell Biology toPharmacology and herlands:Springer;1990.23.Welford RWD,Vercauteren M,Trébaul A,et al.Sci Rep.2016;6.http://dx.doi.org/10.1038/srep30059.24.Penumatsa KC,Fanburg BL.Am J Physiol-Lung Cell Mol Physiol.2014;306:L309–L315.25.Shi Z-C,Devasagayaraj A,Gu K,et al.J Med Chem.2008;51:3684–3687.26.Jin H,Cianchetta G,Devasagayaraj A,et al.Bioorg Med Chem Lett.2009;19:5229–5232.27.Kulke MH,Hörsch D,Caplin ME,et al.J Clin Oncol.2016;1–14.28.Goldberg DR,De Lombaert S,Aiello R,et al.Bioorg Med Chem Lett.2016;26:1124–1129.29.Goldberg DR,De Lombaert S,Aiello R,et al.Bioorg Med Chem Lett.2016;26:2855–2860.30.Cianchetta G,Stouch T,Yu W,et al.Curr Chem Genomics.2010;4:19–26.31.Reduction of mucosal5-HT levels in mice after single daily screening dose ofethyl ester prodrugs(vehicle:0.5%methyl cellulose).32.(S)-2-Benzyl3-ethyl2,8-diazaspiro[4.5]decane-2,3-dicarboxylate was obtainedin two steps from commercially available(S)-8-tert-butyl3-ethyl2,8-diazaspiro[4.5]decane-3,8-dicarboxylate.D.R.Goldberg et al./Bioorganic&Medicinal Chemistry Letters27(2017)413–419419。