Hydantoin analogs inhibit the fully assembled ClpXP protease without affecting the individual peptidase and chaperone domains.

Hydan oin analogs inhibi he ully assembled ClpXP p o ease wi hou a ec ing he indi idual pep idase and chape one domains Ch is ian Fe ze ,a Vadim S. Ko o ko a,b and S ephan A. Siebe a,† P o eolysis media ed by ClpXP is a c ucial cellula p ocess linked o bac e ial pa hogenesis. The de elopmen o speci ic inhibi o s has la gely ocused on ClpP. Howe e , his ocus was challenged by a ecen inding showing ha con o ma ional con ol by ClpX leads o a ejec ion o ClpP binde s. Thus, we he e ollow up on a hi molecule om a high h oughpu sc een pe o med agains he whole ClpXP complex and demons a e ha s able inhibi ion wi h high po ency is possible. Fu he in es iga ions e ealed ha he small molecule binds o ClpP wi hou a ec ing i s ac i i y. Likewise, he molecule does no inhibi ClpX and e ains he o e all oligome ic s a e o ClpXP upon binding. S uc u e ac i i y ela ionship s udies con i med s uc u al cons ain s in all h ee pa s o he molecule sugges ing binding in o a de ined s e eospeci ic pocke . O e all, he inhibi ion o ClpXP wi hou a ec ing he indi idual componen s ep esen s a no el mechanism wi h pe spec i es o u he op imiza ion o in si u applica ions. Caseinoly ic p o ease P (ClpP) is a e adecame ic enzyme belonging o he se ine hyd olase amily.1, 2 The enzyme is able o diges small pep ides bu canno ac on la ge p o ein subs a es by i sel . In he cell ClpP binds o cogna e chape ones such as ClpX ia bo h apical si es o he e adecame ic ba el o ming he p o eoly ically ac i e ClpXP complex.3 Wi hin his complex ClpX ecognizes p o eins p one o deg ada ion, e.g. du ing cellula s ess, un olds hem unde consump ion o ATP and inally pushes he pep ide chain in o he ClpP ba el o subsequen p o eoly ic diges . ClpXP ac i i y has been shown o be c ucial o cell homeos asis in cance cells as well as o he con ol o i ulence in pa hogenic bac e ia such as S aphylococcus au eus.4, 5 As a consequence o hese unc ions he gene ic dele ion as well as chemical inhibi ion led o posi i e he apeu ic e ec s in ea ed leukemia as well as bac e ial cells, espec i ely.5-7 Howe e , majo challenges in he de elopmen o inhibi o s a e he selec i i y o compounds o ClpP o ClpX, hei s abili y in plasma, esidence ime when bound o he enzyme as well as con o ma ional con ol o ClpX o e ClpP esul ing in ejec ion o compounds.8, 9 One o he easons o hese limi a ions was he p e ious ocus on solely ClpP du ing inhibi o de elopmen and jus ecen mechanis ic insigh s un a eled he impo ance o he whole ClpXP complex as mo e sui able a ge o mimic cellula ac i i y. In o de o o e come he limi a ions o p e ious co alen inhibi o s10 such as be a- lac ones, phenyl es e s and bo onic acids as well as e e sible inhibi o s such as oxazoles, a new high- h oughpu sc eening (HTS) campaign u ilized an assay moni o ing he ac i i y o whole SaClpXP complex.11 The ac i i y was moni o ed by he deg ada ion o g een luo escen p o ein (GFP) agged wi h a Ss A pep ide sequence, equi ed o ClpX media ed ecogni ion.12 In p inciple, his HTS could deli e 3 ypes o inhibi o s: 1) inhibi o s o ClpP, 2) inhibi o s o ClpX and 3) inhibi o s ha block he in e play be ween ClpX and ClpP. The HTS e eled 6 alida ed hi s which we e all able o po en ly inhibi he whole p o eoly ic complex. One inhibi o was selec ed o ollow up s udies demons a ing ha i ac ed solely on he ClpX chape one.11 Binding o his compound o ClpXP led o a dis up ion o he oligome ic complex associa ed wi h a disassembly o ClpX hexame s in o monome s. Co espondingly, he compound inhibi ed ClpX ATPase ac i i y bu had no e ec on ClpP. He e, we ollow up on hese HTS esul s and in oduce, in addi ion o al eady exis ing ClpP and ClpX inhibi o s, an unp eceden ed compound which nei he in luences ClpP no ClpX ac i i y bu po en ly blocks u no e o he whole complex. Syn hesis o se e al de i a i es e eals insigh s in o he s uc u e ac i i y ela ionship (SAR) and he p epa a ion o wo enan iome s shows a clea p e e ence o one enan iome o complex inhibi ion. Toge he wi h he mal shi assays, hese a. Technische Uni e si ä München, Cen e o In eg a ed P o ein Science Munich (CIPSM), Lich enbe gs aße 4, 85748 Ga ching (Ge many) b. Cu en add ess: Helmhol z-Zen um ü In ek ions o schung, Inho ens aße 7, 38124 B aunschweig (Ge many) † E-Mail: s ephan.siebe @ um.de esul s sugges he exis ence o a de ined, s e eospeci ic binding pocke loca ed on ClpP. Compound 1 was iden i ied as one o he mos po en inhibi o s o ClpXP ac i i y in a p e ious HTS (Figu e 1A).11 P io o u he mechanis ic s udies o his compound, we pe o med a seconda y assay o exclude a pu a i e ac i i y agains c ea ine kinase, an enzyme equi ed in he HTS o egene a e ATP om ADP in he GFP assay. In con as o a posi i e con ol wi h iodoace amide, no signi ican educ ion o ac i i y could be obse ed in he Kinase-Glo assay (Suppo ing Figu e 1). In e es ingly, 1 did nei he inhibi ClpP pep idase ac i i y, as measu ed by u no e o a luo escen subs a e (posi i e con ol wi h a p e ious oxazole inhibi o ),9 no ClpX media ed ATP hyd olysis (Figu e 1B, C). Howe e , when ClpXP ac i i y was es ed ia he GFP deg ada ion assay, he compound e ealed po en inhibi ion wi h an IC50 o 1.9 µM, a ema kable ac i i y when conside ing ha 4 µM o enzyme (ClpP monome ) is needed in his assay (Figu e 1D). To u he localize i he compound binds o ClpP o ClpX we pe o med he mal shi assays in he p esence o absence o 1 . A s ong s abiliza ion as indica ed by a 9 K shi in mel ing empe a u e was obse ed o ClpP bu no ClpX (Figu e 2A). As expec ed by he s uc u e o 1 which lacks ob ious elec ophilic moie ies, no co alen modi ica ion was obse ed on ClpP ia mass spec ome y analysis (Figu e 2B). Analy ical size exclusion ch oma og aphy showed ha 1 does nei he deoligome ize ClpP no ClpX as well as he whole complex (Figu e 2C). This indica es an unp eceden ed mode o inhibi ion o ClpXP ia e e sible binding o solely ClpP while e aining he o e all oligome ic assembly. Subsequen SAR s udies aimed o deciphe he mos impo an s uc u al elemen s o 1 esponsible o inhibi ion. Fo his he compound was dissec ed in o h ee pa s, namely he hydan oin, linke and hiophene moie ies (Figu e 3A). All analogs we e ei he syn hesized o ob ained om comme cial sou ces and hei co esponding s uc u es a e displayed in Figu e 3B. The syn hesis s a ed wi h educ i e amina ion o subs i u ed hienca baldehyde wi h alkylamines (Scheme 1). The ob ained seconda y amines we e acyla ed wi h 2- b omoace yl b omide. 5,5-Dia ylhydan oins we e syn hesized om he co esponding 1,2-dia yl-1,2-diones and u ea. Regioselec i e N-alkyla ion o 5,5-dia ylhydan oins wi h 2- 4.55.0 5.56.0 6.57.0 7.58.0 0 20 40 60 80 100 -log c [c/M] ClpP Pep idaseAc i i y[%] 286 1 DMSO nega i e 1-1 00 µM 0 20 40 60 80 100 120 ClpX ATPase Ac i i y [%] 4.04.5 5.05.5 6.06.5 7.0 0 20 40 60 80 100 -log c 1 [c/M] ClpXPP o easeAc i i y[%] A) D) B) C) 1 O N HO N N H N H OS 286 Figu e 1 In luence o compound 1 on c ea ine kinase, ClpP, ClpX and ClpXP ac i i y. A) Chemical s uc u e o ClpXP inhibi o 1. B) 1 does no al e ClpP pep idase ac i i y in a luo escen assay (1 µM ClpP; mean ± s anda d e o ). The non-co alen ClpP inhibi o 286 was used as a posi i e con ol. C) ClpX ATPase ac i i y assay is no inhibi ed by 1 (100 µM inal compound concen a ion; mean ± s anda d de ia ion). D) 1 shows a dose- dependen inhibi ion o he ClpXP p o ease (mean ± s anda d e o ). 22600 22800 23000 23200 23400 0 50 100 ClpP + 1 MW [Da] Rela i eIn ensi y[%] 22600 22800 23000 23200 23400 0 50 100 ClpP + DMSO MW [Da] Rela i eIn ensi y[%] 6810 12 14 16 0 20 40 60 80 100 V [mL] Rela i eUV Abso bance[%] 1 DMSO ClpX6ClpX2-3 6810 12 14 16 0 20 40 60 80 100 V [mL] Rela i eUV Abso bance[%] DMSO 1 ClpXP ClpP14 6810 12 14 16 0 20 40 60 80 100 V [mL] Rela i eUV Abso bance[%] DMSO 1 ClpP14 30 35 40 45 50 55 60 65 70 -1500 -1000 -500 0 500 1000 1500 Tempe a u e[°C] -d(RFU)/dT ClpP +DMSO: 46.9 ± 0.2 °C ClpP +1: 55.9 ± 0.2 °C ClpX +DMSO: 40.4 ± 0.3 °C ClpX +1: 41.0 ± 0.5 °C A) C) B) ClpP ClpX ClpXP unmodi ied ClpP unmodi ied ClpP Figu e 2 In e ac ion o 1 wi h ClpP and ClpX. A) The mal-shi assay depic s s ong s abiliza ion o ClpP while he mel ing poin o ClpX emains unal e ed. Expe imen s we e conduc ed wi h 10 µM enzyme concen a ion in PBS bu e and DMSO o 60 µM o 1. B) No co alen modi ica ions o ClpP a e de ec able by in ac -p o ein mass spec ome y upon ea men wi h 1 a 100 µM (100- old excess). C) Size-exclusion ch oma og aphy expe imen s show no subs an ial change o ClpP- e adecame , ClpX-hexame and ClpXP-complex upon ea men wi h 100 µM 1. Scheme 1 Syn hesis o 1 analogs. Reagen s and Condi ions: a) u ea, 40% aq. NaOH, E OH, e lux, 3 h, hen HCl, . ., 41 – 66%; b) R2NH2, HOAc, THF, MeOH, 0° C, 4 h, hen . ., 18 h, 92–94%; c) NaBH4, E OH, 2 h, . ., 86 – 91%; d) 2-b omoace yl b omide, CH2Cl2, 0° C o . ., 1 h, 38 – 76%; e) Cs2CO3, DMF, . ., 3 h, 25 – 57% b omoace amides using Cs2CO3 as a base led o he desi ed p oduc s.13 Al e a ions in he hiophene sec ion o he molecule by exchange o he halogen subs i u ed hiophene wi h a phenyl g oup ( 2 ) esul ed in a en old lowe IC50 alue in he ClpXP p o ease assay (Figu e 3C). P opyla ion ( 3 ) ins ead o me hyla ion ( 4 ) o he ni ogen loca ed in he linke almos comple ely p e en ed inhibi ion. Howe e , sho ening o he linke and emo al o he ni ogen ( 5 ) s ill e ained some ac i i y, al hough no ull inhibi ion could be ob ained. Replacemen o one phenyl g oup by a me hyl subs i uen a he hydan oin co e s uc u e ( 6 ) esul ed in a 40- old d op in po ency. This sugges s ha s e ically demanding g oups a e necessa y o po en inhibi ion. Hence, he ole o addi ional me hyl subs i u ed phenyl g oups was u he elucida ed. Disubs i u ed compound 1 exhibi ed he highes po ency while compound 7 , bea ing wo equally me hyl subs i u ed phenyl moie ies, showed comple e loss o ac i i y. This esul indica es ha only one enan iome o 1 is binding o ClpP as he me hyl subs i uen seems o clash wi h he binding si e in one o he wo possible con o ma ions. Consequen ly, enan iome s we e sepa a ed by chi al high-pe o mance liquid ch oma og aphy. Due o he small amoun o isola ed enan iome s he absolu e con igu a ion could no be assigned by X- ay analysis. Indi idual es ing o bo h enan iome s in he ClpXP p o ease assay e ealed ha only 1 -E1 showed inhibi ion, while he o he enan iome ( 1 -E2) was comple ely inac i e (Figu e 3C). This sugges s he exis ence o a de ined, s e eospeci ic and d uggable binding pocke enabling u he s udies o imp o e he in si u ac i i y ia pa ame e s such as up ake and s abili y. S udies in o i ulence we e pe o med ia Wes e n blo analysis and LC-MS/MS. Only a sligh educ ion o alpha-hemolysin, a majo i ulence ac o linked o ClpXP ac i i y, was obse ed o he acema e and 1 -E2 bu no o 1 -E1 ea ed samples (Suppo ing Figu e 2). Howe e , sec e ome analysis o 1 enan iome s by LC-MS/MS showed no change o alpha- hemolysin compa ed o DMSO ea ed samples (Suppo ing Table 1). We hus conclude ha physicochemical pa ame e s need o be op imized o mo e e ec i e s udies on whole cells. Conclusions In conclusion, 1 cons i u es a new class o ClpXP inhibi o s ha nei he ac on pep idase no chape one ac i i ies bu equi es he whole p o eoly ic complex o impai ing i s subs a e u no e . The compound binds s e eospeci ically in o a ye uniden i ied binding pocke on ClpP, which acco ding o ou SAR da a exhibi s a de ined shape allowing only es ic ed modi ica ions. The elucida ion o his pocke in u u e s udies may u he un a el he p ecise mechanism o ac ion and guide he design o imp o ed inhibi o s also bea ing in si u ac i i ies. Con lic s o in e es The e a e no con lic s o decla e. Acknowledgemen s This wo k was suppo ed by he Deu sche Fo schungsgemein- scha SI1096/8-1 (ClpP) and CIPSM. No es and e e ences 1. R. T. Saue and T. A. Bake , Annu. Re . Biochem., 2011, 80, 587-612. 2. K. Liu, A. Ologbenla and W. A. Hou y, C i . Re . Biochem. Mol. Biol., 2014, 49, 400–412. 3. T. A. Bake and R. T. Saue , Biochim. Biophys. Ac a, 2012, 1823, 15-28. 4. D. F ees, S. N. Qazi, P. J. Hill and H. Ingme , Mol. Mic obiol., 2003, 48, 1565-1578. 5. D. F ees, K. So ensen and H. Ingme , In ec . Immun., 2005, 73, 8100-8108. 6. A. Cole, Z. Wang, E. Coyaud, V. Voisin, M. G onda, Y. Ji ko a, R. Ma son, R. Hu en, S. Babo ic, N. Maclean, I. Res all, X. Wang, D. V. Jeya aju, M. A. Sukhai, S. P abha, S. Bashi , A. Ramak ishnan, E. Leung, Y. H. Qia, N. Zhang, K. R. Combes, T. Ke ela, F. Lin, W. A. Hou y, A. Aman, R. Al-Awa , W. Zheng, E. Wienholds, C. J. Xu, J. Dick, J. C. Wang, J. Mo a , M. D. Minden, C. J. Ea es, G. D. Bade , Z. Hao, S. M. Ko nblau, B. Raugh and A. D. Schimme , Cance cell, 2015, 27, 864-876. 7. T. Bö che and S. A. Siebe , J. Am. Chem. Soc., 2008, 130, 14400-14401. 8. M. W. Hackl, M. Lakemeye , M. Dahmen, M. Glase , A. Pahl, K. Lo enz-Baa h, T. Menzel, S. Sie e s, T. Bo che , I. An es, H. Waldmann and S. A. Siebe , J. Am. Chem. Soc., 2015, 137, 8475-8483. 9. A. Pahl, M. Lakemeye , M. T. Vielbe g, M. W. Hackl, J. Vomacka, V. S. Ko o ko , M. L. S ein, C. Fe ze , K. Lo enz- Baa h, K. Rich e , H. Waldmann, M. G oll and S. A. Siebe , Angew. Chem. In . Ed., 2015, 54, 15892-15896. 10. F. Ye, J. Li and C. G. Yang, Mol. Biosys ., 2016, 13, 23-31. 11. C. Fe ze , V. S. Ko o ko , R. Thane , K. M. Lee, M. Neuenschwande , J. P. on K ies, E. Medina and S. A. Siebe , Angew. Chem. In . Ed., 2017, 56, 15746-15750. 12. S. Go esman, E. Roche, Y. Zhou and R. T. Saue , Genes De ., 1998, 12, 1338-1347. 13. N. T. Tz e ko , H. Eule and C. E. Mülle , Beils ein J. O g. Chem., 2012, 8, 1584–1593. Figu e 3 S uc u e-ac i i y ela ionship s udies o 1 analogs in ClpXP-p o ease assay. A) Dissec ion o 1 in o 3 pa s. B) Chemical s uc u es o 1 analogs. C) Inhibi ion da a o all compounds in he ClpXP p o ease assay. A) B) O NCl S N HN O O linke hiophenehydan oin 34567 0 25 50 75 100 -log c [c/M] ClpXPP o easeAc i i y[%] 1 1 - E1 1 - E2 C) 4567 0 25 50 75 100 -log c [c/M] ClpXPP o easeAc i i y[%] 2 3 4 4567 0 25 50 75 100 -log c [c/M] ClpXPP o easeAc i i y[%] 6 5 7