Inhibition of Type IV Secretion Activity and Growth of by Cisplatin and Other Platinum Complexes.

Inhibi ion o Type IV Sec e ion Ac i i y and G ow h o Helicobac e pylo i by Cispla in and O he Pla inum Complexes Cla a Le l 1,2† , F anziska Schindele 1,2† , Giamba is a Tes olin 3,4 , Alexande Bä 5 , Tobias Rehm 5 , Ma k B öns up 3,4 , Raine Schobe 5 , U sula Bili ewski 3,4 , Raine Haas 1,2 and Wol gang Fische 1,2 * 1 Max on Pe enko e Ins i u e o Hygiene and Medical Mic obiology, Facul y o Medicine, LMU Munich, Munich, Ge many, 2 Ge man Cen e o In ec ion Resea ch (DZIF), Munich Si e, Munich, Ge many, 3 Depa men o Chemical Biology, Helmhol z Cen e o In ec ion Resea ch, B aunschweig, Ge many, 4 Ge man Cen e o In ec ion Resea ch (DZIF), Hanno e - B aunschweig Si e, B aunschweig, Ge many, 5 O ganic Chemis y Labo a o y, Uni e si y Bay eu h, Bay eu h, Ge many Type IV sec e ion sys ems a e p o ein sec e ion machine ies ha a e equen ly used by pa hogenic bac e ia o injec hei i ulence ac o s in o a ge cells o hei espec i e hos s. In he case o he human gas ic pa hogen Helicobac e pylo i, he cy o oxin- associa ed gene (Cag) ype IV sec e ion sys em is conside ed a majo cause o se e e disease, such as gas ic cance , and hus cons i u es an a ac i e a ge o specific ea men op ions agains H. pylo i in ec ions. He e, we ha e used a Cag ype IV sec e ion epo e assay o sc eening a epu posing compound lib a y o inhibi o s a ge ing his sys em. We ound ha he an i umo agen cispla in, a pla inum coo dina ion complex ha kills a ge cells by o ma ion o DNA c osslinks, is a po en inhibi o o he Cag ype IV sec e ion sys em. S ikingly, we ound ha his inhibi o y ac i i y o cispla in depends on a ligand exchange eac ion which inco po a es a sol en molecule (dime hylsul oxide) in o he complex, a modifica ion which is known o be dele e ious o DNA c osslinking, and o i s an icance ac i i y. We ex ended ou analysis o se e al analogous pla inum complexes con aining N-he e ocyclic ca bene, as well as DMSO o o he ligands, and ound a ying inhibi o y ac i i ies owa d he Cag sys em which we e no cong uen wi h hei DNA- binding p ope ies, sugges ing ha p o ein in e ac ions may cause he inhibi o y e ec . Inhibi ion expe imen s unde a ying condi ions e ealed e ec s on adhe ence and bac e ial iabili y as well, and showed ha he ype IV sec e ion-inhibi o y capaci y o pla inum complexes can be inac i a ed by sul u -con aining eagen s and in complex bac e ial g ow h media. Taken oge he , ou esul s demons a e DNA binding- independen inhibi o y e ec s o cispla in and o he pla inum complexes agains di e en H. pylo i p ocesses including ype IV sec e ion. Keywo ds: Helicobac e pylo i, ype IV sec e ion sys em, Cag, cispla in, p o ein sec e ion F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029581 Edi ed by: D. Sco Me ell, Uni o med Se ices Uni e si y, Uni ed S a es Re iewed by: Ja ie To es, Mexican Social Secu i y Ins i u e (IMSS), Mexico Timo hy Co e , Vande bil Uni e si y, Uni ed S a es *Co espondence: Wol gang Fische fi[email p o ec ed] † These au ho s ha e con ibu ed equally o his wo k Special y sec ion: This a icle was submi ed o Molecula Bac e ial Pa hogenesis, a sec ion o he jou nal F on ie s in Cellula and In ec ion Mic obiology Recei ed: 07 Sep embe 2020 Accep ed: 17 No embe 2020 Published: 18 Decembe 2020 Ci a ion: Le l C, Schindele F, Tes olin G, Bä A, Rehm T, B öns up M, Schobe R, Bili ewski U, Haas R and Fische W (2020) Inhibi ion o Type IV Sec e ion Ac i i y and G ow h o Helicobac e pylo i by Cispla in and O he Pla inum Complexes. F on . Cell. In ec . Mic obiol. 10:602958. doi: 10.3389/ cimb.2020.602958 ORIGINAL RESEARCH published: 18 Decembe 2020 doi: 10.3389/ cimb.2020.602958 INTRODUCTION The human gas ic pa hogen Helicobac e pylo i is esponsible o ch onic gas i is, pep ic ulce disease, as well as gas ic adenoca cinoma and MALT lymphoma, and hus ep esen s one o he leading causes o in ec ion-associa ed mo bidi y and mo ali y wo ldwide. I has been es ima ed ha mo e han 4 billion indi iduals a e in ec ed wi h H. pylo i, albei wi h majo egional a ia ions (Hooi e al., 2017), and ha mo e han 800,000 new cases o gas ic cance pe yea can be a ibu ed o H. pylo i in ec ion (de Ma el e al., 2020). Al hough consensus ea men s a egies a e a ailable (Fallone e al., 2016), esis ance a es agains he he apeu ically used an ibio ics a e inc easing o an ala ming ex en (Sa oldi e al., 2018). Because o his, H. pylo i has been included on he WHO p io i y lis o esea ch and de elopmen o new an ibio ics (Tacconelli e al., 2018). One possible app oach owa d no el ea men op ions is o iden i y po en ial inhibi o s o impo an i ulence p ope ies, which migh be u ilized o complemen es ablished ea men egimes. One o he majo ac o s in ol ed in pa hogenici y o H. pylo i is he ype IV p o ein sec e ion sys em encoded on he cy o oxin- associa ed gene (cag) pa hogenici y island (Backe e al., 2017). The Cag ype IV sec e ion sys em builds up a ema kable mul ip o ein complex composed o oughly 20 di e en p o eins, which spans he bac e ial inne and ou e memb anes and is able o ansloca e he bac e ial CagA p o ein in o he cy oplasm o gas ic cells (Fische , 2011;Chung e al., 2019;Hu e al., 2019). Fu he mo e, he Cag ype IV sec e ion sys em is in ol ed in inducing a s ong p oinflamma o y esponse in gas ic epi helial cells ia NF-kB signaling (Zhang e al., 2020). The p esence o he cag pa hogenici y island inc eases he isk o de eloping se e e disease, pa icula ly o gas ic adenoca cinoma (W oblewski e al., 2010). Al hough many s udies ha e add essed he composi ion o he sec e ion appa a us and he unc ions o indi idual componen s, and despi e he a ailabili y o high- esolu ion s uc u es o he Cag sec e ion appa a us (F ick-Cheng e al., 2016;Chung e al., 2019;Hu e al., 2019), de ails o he ype IV sec e ion p ocess a e s ill only poo ly unde s ood. Fo example, he ole o pilus-like s uc u es associa ed wi h he ype IV sec e ion sys em, and e en hei composi ion, a e no clea (Backe e al., 2015;Chang e al., 2018). On he o he hand, i is well-es ablished ha he Cag sys em, simila o o he ype IV sec e ion sys ems, con ains h ee di e en pu a i e ATPases ha a e all essen ial o he CagA sec e ion p ocess. Se e al s udies ha e epo ed small-molecule inhibi o s wi h he po en ial o in e e ing wi h he Cag ype IV sec e ion sys em. One s udy epo ed he iden ifica ion o compounds ha we e able o educe o ma ion o sec e ion sys em-associa ed pili, and also ype IV sec e ion i sel (Sha e e al., 2016). O he s udies ha e desc ibed small-molecule inhibi o s ha a e able o inhibi one o he ATPases, Caga(Hille ingmann e al., 2006;Saye e al., 2014; A ya e al., 2019). In his s udy, we ha e used a ecen ly desc ibed CagA ansloca ion epo e assay (Schindele e al., 2016) o sc een a small-compound epu posing lib a y o molecules ha a e able o educe his ype IV sec e ion ac i i y. Apa om a numbe o molecules wi h known o suspec ed an ibac e ial ac i i ies, we iden ified wo an i-cance d ugs, cispla in and ca bopla in, o which an ibac e ial e ec s ha e been desc ibed p e iously. Howe e , we show in de ailed ollow-up s udies, also including o he pla inum complexes, ha hei ac i i y agains H. pylo i does no depend on he ypical DNA-binding p ope ies equi ed o an icance ac i i y o he pla inum complexes, and hus di e s widely om hei ac i i y owa d o he bac e ia. MATERIALS AND METHODS Bac e ial S ains, Cell Lines, and Cul u e Condi ions H. pylo i s ains P12 and P12 [TEM-1–CagA] (Schindele e al., 2016) we e g own on GC aga pla es (Oxoid) supplemen ed wi h i amin mix (1%) and ho se se um (Li e Technologies; 8%) (se um pla es), and cul u ed o 16 o 60 h in a mic oae obic a mosphe e (85% N 2 , 10% CO 2 ,5%O 2 ) a 37°C. AGS cells we e cul i a ed in RPMI (Gibco) supplemen ed wi h 10% FCS (hea - inac i a ed; Li e Technologies). Mu ine L929 fib oblas s we e cul i a ed in DMEM (Gibco) supplemen ed wi h 10% FCS, 2 mM L-glu amine (Gibco) and 1 mM sodium py u a e (Gibco) a 37°C in a 5% CO 2 incuba o . Reagen s The LOPAC 1280 lib a y was ob ained om Sigma (o de ing no. LO1280). Cispla in (no. P4394) and cis-dichlo ido-bis(DMSO) pla inum(II) (no. 767654) we e pu chased om Sigma. T anspla in was pu chased om Al a Aesa (no. 10472). S ock solu ions o all pla inum complexes we e p epa ed in DMSO a 200- old hei final concen a ions in he assays, excep whe e indica ed o he wise. The final DMSO concen a ion in he assays was hus always 0.5%. An ibodies, SDS-PAGE, Immunoblo ing, and ELISA A polyclonal an ise um agains he CagA EPIYA egion (AK299) has been desc ibed p e iously (Schindele e al., 2016). Sodium dodecyl sul a e–polyac ylamide gel elec opho esis (SDS–PAGE) and Wes e n blo ing was pe o med as desc ibed (Fische e al., 2001). Fo he de elopmen o immunoblo s, poly inylidene difluo ide (PVDF) fil e s we e blocked wi h 5% non- a milk powde in TBS (50 mM T is–HCl, pH 7.5, 150 mM NaCl), 0.1% ( / ) Tween 20 (TBS-T), and incuba ed wi h he espec i e an ise a a app op ia e dilu ions in TBS-T wi h 1% non- a milk powde . Alkaline phospha ase-conjuga ed p o ein A was used o isualize bound an ibody. S anda d in ec ions o AGS cells wi h H. pylo i s ains and subsequen p epa a ions o phospho y osine immunoblo ing we e pe o med as desc ibed p e iously (Odenb ei e al., 2000). B iefly, cells seeded in 6-well pla es (Falcon) we e in ec ed wi h bac e ia a a mul iplici y o in ec ion o 100 o 4 h a 37°C, washed h ee imes and suspended in PBS con aining 1 mM Na 3 VO 4 , 1 mM PMSF, 10mgml −1 leupep in, and 10 mgml −1 peps a in. Cells wi h adhe en bac e ia we e collec ed by cen i uga ion and esuspended in Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029582 SDS-PAGE sample solu ion. Ty osine-phospho yla ed p o eins we e analyzed by immunoblo ing wi h he phospho y osine an ibody PY99 (San a C uz Bio echnologies). P oduc ion o IL-8 by AGS cells a e in ec ion wi h H. pylo i s ains o 4 h was de e mined omco-incuba ionsupe na an sbyasandwichELISA as desc ibed elsewhe e (Fische e al., 2001). As a con ol, IL-8 p oduc ion was induced wi h 20 ng/ml ecombinan human TNF-a(Pep o ech Inc.). TEM-1–CagA T ansloca ion Assay The CagA ansloca ion epo e assay wi h H. pylo i s ains p oducing TEM-1–CagA usions was pe o med as desc ibed elsewhe e (Schindele e al., 2016). B iefly, AGS cells we e co- incuba ed wi h H. pylo i P12 [TEM-1-CagA] o 2.5 h in 384- well, o in 96-well mic o i e pla es (black, clea bo om, issue cul u e ea ed, 4 i ude) in PBS/10% FCS. A e in ec ion, cells we e loaded wi h he fluo escen subs a e CCF4-AM in a loading solu ion (Li eBLAze -FRET B/G loading ki ; In i ogen) supplemen ed wi h 1 mM p obenecid (Sigma) acco ding o he manu ac u e ´s ins uc ions. Fo fluo escence quan ifica ion by pla e eading, in ec ed cells we e incuba ed wi h his loading solu ion a oom empe a u e in he da k o 2 h, and hen di ec ly measu ed wi h a Cla ios a eade (BMG Lab ech) using an exci a ion wa eleng h o 405 nm, and emission wa eleng hs o 460 nm, o 530 nm. CagA ansloca ion was calcula ed as he a io o backg ound-co ec ed (wells con aining no cells o bac e ia, bu CCF4-AM loading solu ion) emission alues a 460 o 530 nm, which we e no malized o P12 [TEM-1– CagA] as a posi i e con ol, and P12 [TEM-1–CagA], DcagT as a nega i e con ol. Fo inhibi ion expe imen s, bac e ia we e p e- incuba ed wi h he espec i e concen a ions o compounds ob ained om he co esponding s ock solu ions o 30 min a 37°C in PBS/10% FCS, ollowed by in ec ion o 2.5 h in he p esence o he compound. Addi ionally, CagA ansloca ion epo e assays we e pe o med in b ain hea in usion (BHI, BD), B ucella b o h (BB, BD, supplemen ed wi h 10% FCS) o PBS/10% FCS mixed wi h he indica ed amoun o BB. To es he impac o amino acids on he inhibi o y e ec o cispla in, s ock solu ions o L-cys eine (Se a), L-me hionine (Me ck) and L-alanine (Ro h) we e p epa ed in ddH 2 Oandmixedwi h cispla in (6.3 mM in DMSO). Cispla in-amino acid mix u es we e s o ed a −20°C un il u he use. G ow h Assays H. pylo i g own on se um pla es we e suspended o an op ical densi y (OD 550 nm ) o 0.075 in BB/10% FCS and sub-cul u ed in 96-well mic o i e pla es (clea , fla -bo om, Cos a , Co ning Inc.). Compounds we e added o he espec i e concen a ions om co esponding s ock solu ions, and wells we e sealed wi h a gas-pe meable memb ane (B ea he-Easy®sealing memb ane, Di e sified Bio ech). Pla es we e incuba ed a 37°C, 10% CO 2 , 200 pm in a pla e eade (Cla ios a , BMG Lab ech) wi h an a mosphe ic con ol uni (BMG Lab ech). OD 550 was au oma ically measu ed e e y 5 min un il he s a iona y phase was eached. G ow h cu es we e analyzed and p ocessed using he MARS Da a Analysis so wa e 3.10 R5 (BMG Lab ech). The e ec o cispla in on H. pylo i iabili y in he p esence o AGS cells was u he assessed by incuba ing he bac e ia (OD 550 nm = 0.1) oge he wi h AGS cells in PBS/10% FCS o BB/10% FCS supplemen ed wi h 100 µM cispla in in DMSO, o wi h DMSO only. A e 2 h a 37°C, 10% CO 2 , 5 µl o bac e ial suspension we e spo ed on se um pla es, and g ow h was checked a e 24 h. Mass Spec ome y and NMR Analysis LCMS measu emen s we e pe o med using an HPLC (Agilen echnologies 1200 se ies) equipped wi h a Gemini-NX 3u C18 110A 50 × 2.0 mm column coupled o an ion ap mass spec ome e (B uke amaZon SL). High esolu ion mass spec a we e eco ded by di ec in usion in o a Q-TOF mass spec ome e (B uke maXis HD) using elec osp ay ioniza ion (ESI) in he posi i e mode. 1 H and 195 P NMR spec a we e eco ded using a B uke Ad ance-III HD 700 MHz spec ome e . Chemical shi s a e epo ed as alues in ppm, o he 1 H-NMR ela i e o esidual sol en signal as in e nal s anda d and o he 195 P -NMR ela i e o he e e ence compound Na 2 P Cl 6 . Cy o oxici y Measu emen s The e ec o compounds on euka yo ic cell p oli e a ion and iabili y was assessed using he WST-1 cell p oli e a ion assay (Roche Applied Science). B iefly, mu ine L929 fib oblas s we e seeded in o 96-well pla es (3.0 × 10 5 cells/well, clea , fla -bo om, Cos a , Co ning Inc.) using Phenol ed- ee cul u e medium. A e 24 h incuba ion a 37°C, 5% CO 2 , compounds we e added in wo- old dilu ions and incuba ion was con inued o h ee days. WST-1 eagen was added acco ding o he manu ac u e ’s p o ocol and pla es we e kep a 37°C, 5% CO 2 o 1 h. The abso bance a 450 nm and 690 nm ( e e ence wa eleng h) was eco ded in a pla e eade (Cla ios a , BMG Lab ech). Fo e alua ion, he di e ence o A 450nm and A 690nm was calcula ed, and he con ol alue (medium and WST-1 only) was sub ac ed. Pe cen al iabili y was no malized o he un ea ed con ol. Adhe ence Assays Cispla in influence on adhe ence o H. pylo i o AGS cells was essen ially measu ed as desc ibed (Könige e al., 2016). B iefly, cells we e in ec ed wi h a GFP-p oducing a ian o s ain P12 (P12 [pHel12::g p]; (Könige e al., 2016)), using an MOI o 60, and cispla in was added a he ime o in ec ion, o 30 min la e , om DMSO s ock solu ions o he co esponding concen a ions. AGS cell in ec ion was allowed o p oceed un il 1 h a e in ec ion a 37°C and 5% CO 2 .A e h ee washing s eps wi h PBS, cells wi h adhe en bac e ia we e collec ed by EDTA ea men , and analyzed in a flow cy ome e (FACS Can oII, BD Biosciences). Fo analysis, he median fluo escence in ensi y o non-in ec ed cells was sub ac ed om ha o in ec ed samples. S a is ical Analysis Quan i a i e da a se s shown a e gene ally a e age alues esul ing om a leas h ee independen expe imen s, wi h s anda d de ia ions. IC 50 alues we e calcula ed om a leas h ee independen expe imen s by nonlinea eg ession using he G aphpad P ism5 so wa e. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029583 RESULTS Iden ifica ion o Cispla in and Ca bopla in as Inhibi o s o he Cag Type IV Sec e ion Sys em In an a emp o iden i y small molecules ha inhibi he Cag ype IV sec e ion sys em, we sc eened he Lib a y o Pha macologically Ac i e Compounds (LOPAC 1280 ;Sigma) wi h a TEM-1–CagA ansloca ion epo e assay (Schindele e al., 2016) adap ed o a mic o i e pla e o ma . This assay u ilizes enginee ed H. pylo i s ains p oducing he TEM-1 b- lac amase used o he N- e minus o CagA. Type IV sec e ion o he TEM-1–CagA usion p o ein in o a ge cells is moni o ed by loading he cells wi h he fluo escen b-lac amase subs a e CCF4-AM, and CagA ansloca ion is de e mined by calcula ing blue- o-g een fluo escence a ios (see Expe imen al P ocedu es o de ails). Ou o he 1280 compounds es ed a a concen a ion o 50 µM, 151 esul ed in a educ ion o CagA ansloca ion ac i i y o less han 50%, as compa ed o in ec ion wi h an un ea ed epo e s ain. Since hese compounds included se e al subs ances wi h suspec ed o p e iously demons a ed an ibac e ial ac i i ies agains H. pylo i, such as doxycycline and minocycline (Glupczynski e al., 1988), niclosamide (Tha malingam e al., 2018), o clo imazole, we applied a coun e sc een o compounds inhibi ing H. pylo i g ow h. When all compounds ha inhibi ed g ow h o mo e han 50% we e emo ed, only 10 compounds emained (Table 1). Subsequen ly, we used he TEM-1–CagA epo e assay in a 96-well o ma o ep oduce he ype IV-inhibi o y e ec s o hese compounds a a sligh ly lowe concen a ion (35 µM). In his assay, h ee compounds exhibi ed cy o oxic ac i i ies owa d AGS cells, as indica ed by CCF4-AM loading de ec s, and he ini ial inhibi o y e ec s could no be ep oduced o ou u he compounds a his concen a ion (Table 1). Thus, only he unspecific alkyla ing agen iodoace amide, and he an i umo agen s cispla in and ca bopla in emained as po en ial ype IV sec e ion inhibi o s. The ac i i ies o he la e compounds we e confi med in an o hogonal seconda y assay, which measu es CagA y osine phospho yla ion ollowing in ec ion o AGS cells wi h H. pylo i s ain P12. While iodoace amide did no inhibi CagA y osine phospho yla ion (da a no shown), cispla in and ca bopla in we e able o s ongly educe o e en block he appea ance o y osine-phospho yla ed CagA a a concen a ion o 50 µM, sugges ing ha hey in e e e wi h ype IV sec e ion o CagA (Figu e 1A). In con as , bo h cispla in and ca bopla in appa en ly did no in e e e wi h iabili y o H. pylo i a hese concen a ions, as demons a ed by g ow h cu es ob ained in B ucella b o h con aining 10% FCS (BB/10% FCS; Figu e 1B); howe e , he ac i i y o hese compounds depends on he media used (see below). Fu he mo e, le els o he chemokine in e leukin-8 (IL-8), which is induced in AGS cells ia he Cag ype IV sec e ion sys em, we e educed in he p esence o cispla in in a dose-dependen manne (Figu e 1C). In con as , TNF-awas s ill able o induce an IL-8 esponse in AGS cells in he p esence o 100 µM cispla in (Figu e 1C, igh panel), indica ing ha he cispla in e ec on IL-8 induc ion was due o inhibi ion o he bac e ia. Taken oge he , hese obse a ions sugges ed a specific in e e ence o cispla in and ca bopla in wi h ype IV sec e ion by H. pylo i. Dependence o Cispla in Ac i i ies on he Sol en Bo h cispla in (cis-diamminedichlo idopla inum(II)) and ca bopla in (cis-diammine-(1,1-cyclobu ane-dica boxyla o)- pla inum(II)) a e plana pla inum coo dina ion complexes ha a e able o o m in as and c osslinks be ween pu ine bases on DNA (Fich inge -Schepman e al., 1985). Fo cispla in, his c osslinking ac i i y is hough o in ol e p io exchange o a leas one chlo ido ligand o a wa e molecule (aqua ion), which esul s in a ca ionic complex wi h highe a fini y o DNA (Da ies e al., 2000;Kelland, 2007). O he ligand exchange eac ions may also ake place, o example he eplacemen o a chlo ido ligand o dime hylsul oxide (DMSO), which coo dina es pla inum ia i s nucleophilic sul u a om (Figu e 2A)(Annibale e al., 1983). Since he s ock solu ions o he LOPAC 1280 lib a y compounds we e p epa ed in DMSO and hen dilu ed in aqueous bu e s, such complexes wi h DMSO ligands may ha e been p esen in ou sc eening expe imen s. To examine his possibili y, we compa ed he ac i i y o cispla in dissol ed in DMSO wi h ha o cispla in dilu ed om s ock solu ions p epa ed in dime hyl o mamide (DMF), which does no seem o eac wi h cispla in (Hall e al., 2014). Wi h cispla in dissol ed in DMSO, we ob ained a dose-dependen inhibi ion o CagA ansloca ion wi h a hal -maximal inhibi o y concen a ion (IC 50 ) o 5.98 µM, as measu ed by he TEM-1– CagA assay, whe eas DMSO had no e ec (Figu e 2B). Using cispla in om DMF s ock solu ions, howe e , we achie ed inhibi ion only a high concen a ions (>32 µM), clea ly indica ing a sol en -dependen ac i i y o cispla in (Figu e 2C). We also ob ained an inhibi ion o CagA ansloca ion wi h cis- dichlo ido-bis(DMSO)pla inum(II) dissol ed in DMSO, albei a a educed e ficiency in compa ison o cispla in (IC 50 = 29.5 µM; TABLE 1 | P ope ies o compounds iden ified du ing ini ial sc eening o he LOPAC 1280 lib a y a . Compound E ec on ype IV sec e ion (TEM-1–CagA assay) o H. pylo i g ow h (% o un ea ed con ol) TEM-1–CagA, 384 well, 50 µM G ow h, 100 µM TEM-1–CagA, 96 well, 35 µM Ca bopla in 13.7 84.9 32.8 Cispla in 3.4 88.8 7.6 Can ha idic acid 40.1 84.7 No CCF4 loading Can ha idin 32.7 74.6 No CCF4 loading Dihyd oe goc is ine me hanesul ona e 39.7 72.5 80.2 Iodoace amide 47.3 74.7 49.2 Me hyse gide malea e 5.7 83.3 89.5 NF449 oc asodium sal 42.9 102.7 103.0 P opan heline b omide 38.7 65.6 84.0 Wo mannin 35.1 94.0 Low CCF4 loading a Compounds we e ini ially selec ed o inhibi ion o TEM-1–CagA ac i i y by mo e han 50%, and H. pylo i g ow h inhibi ion by less han 50% in compa ison o un ea ed (DMSO) con ol. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029584 Figu e 2D), bu no when he same complex was dissol ed in DMF ins ead (IC 50 ≥100 µM; da a no shown). In e es ingly, when we measu ed H. pylo i g ow h in a mic opla e o ma , we did no obse e significan g ow h de ec s upon addi ion o cispla in dissol ed in DMSO (Figu e 2E), o cis-dichlo ido-bis (DMSO)pla inum(II) dissol ed in DMSO o in DMF (da a no shown). Howe e , cispla in dissol ed in DMF esul ed in a clea dose-dependen g ow h inhibi ion (Figu e 2E). This sugges s ha cispla in dissol ed in DMF exe s i s expec ed oxic ac i i y agains H. pylo i, bu does no impac s ongly on ype IV sec e ion, whe eas a DMSO ligand in he pla inum complex modula es i s ac i i y owa d Cag ype IV sec e ion inhibi ion, while educing he an ibac e ial e ec . To examine whe he cispla in was subjec o ligand exchange eac ions in ou DMSO s ock solu ions, we analyzed he complexes by mass spec ome y, and de ec ed a cispla in de i a i e wi h one chlo ido ligand eplaced by DMSO as he majo molecula species in DMSO solu ion (m/z = 342.999 Da; Figu e 3A). Upon dilu ion in wa e o ace oni ile con aining 0.1% o mic acid, his complex pa ly eac ed u he o gi e a monoammine-monochlo o-bis-DMSO complex (m/z = 403.988 Da; Figu e 3B). No ligand exchange eac ions we e obse ed wi h cispla in dissol ed in DMF (da a no shown). Consis en wi h hese esul s, 1 H-NMR and 195 P -NMR measu emen s o cispla in dissol ed in DMF-d 7 showed s able peaks a 4.17 ppm and −2090 ppm, espec i ely (Figu e 3C). In con as , cispla in dissol ed in DMSO-d 6 was subjec o decomposi ion o e ime, wi h o ma ion o a main species wi hin 24 h, as obse ed by bo h 1 H-NMR and 195 P -NMR (Figu e 3C). Pa icula ly, he signals a ime ze o a 3.94 ppm and −2090 ppm o he 1 H and 195 P nuclei, espec i ely, dec eased o e ime, wi h a concomi an inc ease o peaks a 4.73 ppm and −3138 ppm, espec i ely. Analysis o he NMR sample by mass spec ome y confi med he exchange o a chlo ido ligand o a DMSO-d 6 molecule (Figu e 3D). Keeping he DMSO-d 6 sample a oom empe a u e showed u he con e sion o he newly o med complex, which could be a oided by keeping he samples a 4°C be ween he analyses. A his o lowe empe a u es, he cispla in-DMSO complex p o ed o be s able o a leas se e al days. Taken oge he , NMR and mass spec ome y da a indica e he o ma ion o DMSO adduc de i a i es o cispla in in DMSO- con aining solu ions. AB C FIGURE 1 | E ec s o cispla in and ca bopla in on ype IV sec e ion and on H. pylo i g ow h. (A) Bac e ia we e p e-incuba ed o 30 min a 37°C, 10% CO 2 in PBS/ 10% FCS in he p esence o absence o he indica ed compounds a 50 µM, and subsequen ly used o co-incuba ion wi h AGS cells o 4 h in he same medium. T ansloca ion o CagA was de e mined by i s subsequen y osine phospho yla ion using phospho y osine (PTy ) o CagA immunoblo s. (B) H. pylo i P12 was g own in BB/10% FCS in a mic o i e pla e in he p esence o 0.5% DMSO (con ol), o o he indica ed compounds a 50 µM. (C) Supe na an s o AGS cells co-incuba ed wi h H. pylo i unde he same condi ions as in (A), bu wi h di e en amoun s o cispla in (CisP ), we e analyzed o IL-8 concen a ions by sandwich ELISA (le panel). The indica ed alues we e no malized o he alues ob ained wi h s ain P12 ea ed wi h 0.5% DMSO alone. As a con ol, AGS cells we e ea ed wi h 20 ng/ ml TNF-ain he p esence o absence o 100 µM cispla in, and IL-8 alues we e no malized o hose ob ained o TNF-awi hou cispla in ( igh panel). All da a ep esen mean alues o ou independen expe imen s wi h s anda d de ia ions. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029585 In e es ingly, inco po a ion o DMSO as a ligand in he pla inum complex has been desc ibed as a unc ional inac i a ion wi h espec o DNA in e ac ion, since DMSO ep esen s a poo lea ing g oup (Annibale e al., 1983). The e o e, dissol ing cispla in in DMSO is s ongly discou aged in o de no o lose i s cy o oxic ac i i y (Hall e al., 2014). Using a WST-1 cy o oxici y assay in L929 mu ine fib oblas s, we confi med ha cispla in dissol ed in DMF has a cy o oxic e ec , whe eas cispla in dissol ed in DMSO, o cis- dichlo ido-bis(DMSO)pla inum(II), dissol ed in ei he DMF o DMSO, showed no cy o oxic e ec s a he same concen a ions (Figu e 2F). This sugges s ha he inhibi o y e ec o cispla in on ype IV sec e ion is no a consequence o i s DNA- binding capabili y. E ec s o O he Pla inum Complexes on Type IV Sec e ion Inhibi ion and H. pylo i G ow h I has been shown be o e ha cispla in has cy o oxic e ec s on E. coli and o he bac e ia, pa icula ly i hese ha e de ec s in hei DNA damage epai pa hways (Bha acha ya and Beck, 2002), sugges ing ha he DNA-binding p ope ies o cispla in a e esponsible o his finding. To u he e alua e whe he he obse ed ype IV sec e ion-inhibi o y e ec s may depend on DNA in e ac ion, we ex ended ou expe imen s o o he pla inum complexes wi h a ious ligands, s uc u es, and oxida ion s a es o pla inum. Fi s , we made use o dichlo idopla inum(II) complexes con aining DMSO and N- he e ocyclic ca bene (NHC) ligands, which exhibi a iable DNA-binding p ope ies depending on s e ic shielding o hei lea ing g oups (Muenzne e al., 2015;Rehm e al., 2018). Di e en dichlo idopla inum(II) complexes wi h imidazol-2- ylidene o benzimidazol-2-ylidene ligands in combina ion wi h DMSO we e ound o inhibi CagA ansloca ion, as analyzed wi h he TEM-1–CagA assay (Figu e 4A). Some o hese compounds had addi ional e ec s on H. pylo i g ow h, bu only a highe concen a ions (Figu e 4B). To examine he ole o DMSO in hese complexes, we nex es ed co esponding NHC complexes in which he DMSO ligand is eplaced by a iphenylphosphane ligand, a a ia ion which esul s in slowe DNA binding due o inc eased shielding o he chlo ide lea ing g oup, and al e ed esponses o cance cells (Muenzne e al., 2015). In e es ingly, hose NHC- iphenylphosphane complexes which did no in e e e wi h H. pylo i g ow h, also did no subs an ially inhibi CagA ansloca ion (Figu e 4C), indica ing ha he DMSO ligand canno be unc ionally eplaced by a A B DE F C FIGURE 2 | Sol en influence on ype IV sec e ion inhibi ion and cy o oxici y by pla inum complexes. (A) Scheme illus a ing he po en ial chlo ido o DMSO ligand exchange eac ion upon solu ion o cispla in in DMSO. (B) H. pylo i P12 p oducing a TEM-1-CagA usion was p e-incuba ed o 30 min wi h di e en concen a ions o cispla in dilu ed om co esponding s ock solu ions in DMSO (so ha he final DMSO concen a ion was 0.5% in each case), subsequen ly co-incuba ed wi h AGS cells o 150 min, and hen analyzed ia he TEM-1-CagA epo e assay o ansloca ion o CagA. Mean alues o inhibi ion expe imen s we e subjec ed o nonlinea eg ession analysis, esul ing in he indica ed IC 50 alue o CagA ansloca ion inhibi ion. The indica ed ba s ep esen mean alues including s anda d de ia ions o six independen expe imen s no malized o P12 [TEM-1-CagA] le un ea ed, which was se o 100%. (C, D) The same expe imen s as in (B) we e pe o med wi h cispla in dilu ed om s ock solu ions in DMF (C), o wi h a cis-dichlo ido-bis(DMSO)pla inum(II) complex (cis-dcbd-pla inum(II)) dilu ed om s ock solu ions in DMSO (D). The indica ed ba s ep esen mean alues including s anda d de ia ions o ou (C), o se en (D) independen expe imen s, espec i ely, no malized o P12 [TEM-1-CagA] ea ed wi h 0.5% DMSO only, which was se o 100%. No e ha addi ion o 0.5% DMF alone educes CagA ansloca ion by abou 20%. (E) G ow h cu es unde s anda d g ow h condi ions o H. pylo i P12 in he absence (Con ol wi h 0.5% DMSO) o p esence o he indica ed compound concen a ions aken om he espec i e s ock solu ions. Rep esen a i e cu es a e shown. (F) Cy o oxici y o he indica ed compounds, dilu ed o he espec i e concen a ions om s ock solu ions in ei he DMSO o DMF, owa d L929 cells, as de e mined by a WST-1 assay. The indica ed alues ep esen mean alues including s anda d de ia ions o h ee independen expe imen s no malized o un ea ed con ol. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029586 A B D C FIGURE 3 | Analysis o pla inum complexes o med in di e en sol en s. (A) Mass spec um (MS) and high- esolu ion mass spec um (HRMS) showing he co ec iso opic pa e n o he main componen in he sample o cispla in dissol ed in DMSO. (B) MS and HRMS showing he iso opic pa e n o cispla in dissol ed in DMSO and dilu ed wi h wa e /ace oni ile 70/30 con aining 0.1% HCOOH. (C) S acked 1 H-NMR and 195 P -NMR spec a, eco ded a he indica ed ime poin s, o cispla in dissol ed ei he in DMF-d 7 o in DMSO-d 6 . In con as o he DMSO-d 6 solu ion, cispla in in DMF-d 7 is s able o e se e al days. In he ime be ween he analyses, he samples we e kep a 4°C. (D) MS and HRMS spec a o he cispla in DMSO-d6 solu ion analyzed by NMR. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029587 A B D C FIGURE 4 | Inhibi o y p ope ies o NHC-pla inum complexes owa d ype IV sec e ion and g ow h. (A) The indica ed s able pla inum benzimidazol-2-ylidene o imidazol-2-ylidene complexes wi h DMSO ligands we e added o H. pylo i P12 [TEM-1-CagA] p e-incuba ion suspensions a he indica ed concen a ions, and ype IV ansloca ion o CagA was de e mined wi h he TEM-1-CagA epo e assay. (B) The same complexes we e added o H. pylo i P12 suspensions in BB/10% FCS, and op ical densi ies (550 nm) we e eco ded o 13-14 h in a mic o i e pla e. G ow h cu es a e shown as op ical densi y di e ences. (C) Analogous complexes wi h iphenylphosphane ligands ins ead o DMSO ligands, o complexes con aining wo di e en NHC ligands, we e added a he indica ed concen a ions, and examined wi h he TEM-1-CagA epo e assay as in (A) (le panel). G ow h cu es in he p esence o absence o he same complexes a he indica ed concen a ions ( igh panel). Some compounds pa ially p ecipi a ed in g ow h medium when added a 100 µM, and a e he e o e shown a 10 µM only. (D) An oc ahed al e achlo idopla inum(IV) complex wi h wo NHC ligands, o anspla in we e analyzed wi h he TEM-1-CagA assay a he indica ed concen a ions (le panel) G ow h cu es in he p esence o he same complexes ( igh panels). Due o pa ial p ecipi a ion a 100 µM, TR425 is only shown a a concen a ion o 10 µM. All indica ed ba s ep esen mean alues including s anda d de ia ions o a leas 3 independen expe imen s no malized o P12 [TEM-1-CagA] ea ed wi h 0.5% DMSO, which was se o 100%. Rep esen a i e g ow h cu es a e shown. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029588 iphenylphosphane ligand. Howe e , when we es ed complexes in which he DMSO ligand was eplaced by a second NHC ligand in cis configu a ion (Rehm e al., 2016), we ound some wi h clea an ibac e ial e ec s owa d H. pylo i (da a no shown), bu also o he s which had simila p ope ies as cispla in o he DMSO- con aining complexes (Figu e 4C). Finally, we analyzed he e ec o an analogous pla inum(IV) complex wi h wo NHC ligands, which exhibi s educed DNA-binding p ope ies and only mode a e cy o oxici y (Rehm e al., 2019). He e, we also obse ed a subs an ial inhibi o y po en ial wi h espec o CagA ype IV sec e ion o compound TR425, while H. pylo i g ow h was no s ongly a ec ed (Figu e 4D). This esul confi ms ha he s eng h o in e ac ion wi h DNA does no co ela e wi h ype IV sec e ion inhibi ion. In he same line, we ound ha anspla in ( ans-dichlo idodiamminepla inum(II)), which does no show DNA binding-dependen cy o oxici y, bu is subjec o ligand exchange when dissol ed in DMSO (Hall e al., 2014), s ill inhibi s CagA ansloca ion, albei wi h a educed e ficiency in compa ison o cispla in (IC 50 = 10.9 µM; Figu e 4D). An ibac e ial E ec s o Cispla in on H. pylo i and Influence on Adhe ence The esul s shown so a indica ed ha some pla inum complexes wi h DMSO o o he ligands exe inhibi o y e ec s owa d ype IV sec e ion, bu do no inhibi bac e ial g ow h, whe eas o he s also ga e ise o g ow h de ec s. Since measu ing g ow h cu es equi es he use o complex media, whe eas he ype IV sec e ion epo e assay is ou inely ca ied ou in phospha e bu e supplemen ed wi h e al cal se um, in which he bac e ia do no g ow eadily, we nex es ed whe he hese di e en media migh influence he obse ed pla inum complex e ec s, in addi ion o he sol en used o gene a ing s ock solu ions. When we examined H. pylo i ha had been ea ed wi h cispla in/DMSO in PBS/10% FCS du ing co-incuba ion wi hAGScells o 2h, o hei iabili ybypla ingon s anda d media, we no iced only weak g ow h (Figu e 5A). This an ibac e ial e ec o cispla in/DMSO was no de ec ed a e ea men and co-incuba ion in BB/10% FCS (Figu e 5A). The e o e, we ca ied ou he TEM-CagA ansloca ion assay in A B D C FIGURE 5 | Ac i i ies o pla inum complexes in di e en media and du ing in ec ion o AGS cells. (A) H. pylo i P12 was co-incuba ed wi h AGS cells o 2 h wi h 100 µM cispla in o 0.5% DMSO in he indica ed media, and subsequen ly spo ed on se um aga pla es. G ow h was moni o ed a e 24 h. (B) H. pylo i P12 [TEM-1-CagA] was p e-incuba ed and co-incuba ed wi h AGS cells in he p esence o 100 µM cispla in in PBS/10% FCS, BB/10% FCS, o mix u es o bo h, as indica ed, and CagA ansloca ion was de e mined by he TEM-1-CagA epo e assay. The same es s we e also pe o med wi h BB o BHI medium alone. (C) NHC-pla inum complexes TR425 and TR382 we e added a he indica ed concen a ions o bac e ial suspensions in ei he PBS/10% FCS, o BB/10% FCS, and CagA ansloca ion was de e mined a e AGS cell in ec ion as in (B).(D) H. pylo i P12 [TEM-1-CagA] was incuba ed in PBS/10% FCS oge he wi h 32 µM cispla in/DMSO, o addi ionally wi h he indica ed concen a ions o cys eine, me hionine, o alanine, and CagA ansloca ion was de e mined as in (B). All ba s ep esen mean alues including s anda d de ia ions o a leas 3 independen expe imen s no malized o P12 [TEM-1-CagA] ea ed wi h 0.5% DMSO, which was se o 100%. Le l e al. Pla inum Complex Ac i i y Agains H. pylo i F on ie s in Cellula and In ec ion Mic obiology | www. on ie sin.o g Decembe 2020 | Volume 10 | A icle 6029589