UNIVERZITů PůLůCKÉHO V OLOMOUCI Přírodovědecká fakulta Katedra anorganické chemie
SYNTÉZů, CHARAKTERIZACE A BIOLOGICKÁ ůKTIVITů KOMPLEX P ECHODNÝCH KOV S R ZNÝMI N-DONOROVÝMI LIGůNDY
Habilitační práce
Mgr. Pavel Štarha, Ph.D.
Olomouc, 2015
Poděkování Velmi rád bych zde upřímně poděkoval spolupracovníkům, kolegům a v mnoha případech i kamarádům z Katedry anorganické chemie, Katedry buněčné biologie a genetiky, Katedry biofyziky a Regionálního centra pokročilých technologií a materiálů (RCPTM) Přírodovědecké fakulty Univerzity Palackého v Olomouci za vynikající a podnětnou spolupráci, podporu a kvalitní pracovní zázemí, kterých se mi dostávalo po celou dobu mého působení na olomoucké Alma Mater. Speciální poděkování patří prof. RNDr. Zdeňku Trávníčkovi, Ph.D., jehož odborné a věcné vedení, stejně jako příležitost podílet se na výzkumu moderních a zajímavých tematik na půdě špičkově vybaveného pracoviště, byly, jsou a zcela nepochybně také budou pro moji profesní kariéru naprosto zásadní. Dále chci poděkovat prof. RNDr. Zdeňku Dvořákovi, DrSc. et Ph.D., prof. RNDr. Viktoru Brabcovi, DrSc. a členům jejich výzkumných týmů z Katedry buněčné biologie a genetiky resp. Katedry biofyziky, Přírodovědecké fakulty Univerzity Palackého v Olomouci, za detailní studia biologických vlastností níže komentovaných sloučenin, jejichž výsledky zásadním způsobem přispěly, a doufám, že i v budoucnu budou přispívat, do mnou studované problematiky. Díky patří i generálnímu řediteli RCPTM prof. RNDr. Radku Zbořilovi, Ph.D. za umožnění využití mnohých analytických technik z přístrojového parku centra pro potřeby mé výzkumné činnosti. Velké díky tímto posílám také do Department of Chemistry, University of Warwick, Coventry, UK, kde mi Prof. Peter J. Sadler FRS FRSE dal jedinečnou příležitost pracovat po dobu šesti měsíců ve špičkovém výzkumném týmu na jedné z nejlépe hodnocených britských univerzit a načerpat tak nové a neocenitelné vědomosti a dovednosti, ze kterých budu nepochybně čerpat při své nastávající vědecko-výzkumné činnosti. Děkuji rodičům za trpělivou podporu nejen v dobách studií na Přírodovědecké fakultě Univerzity Palackého v Olomouci, ale i na počátku mého profesního působení tamtéž. V neposlední řadě chci poděkovat své rodině, manželce Ivaně a dcerám Aničce a Emičce, za jejich lásku, trpělivost a nikdy nekončící přísun životní energie.
2
Obsah 1. Úvod ................................................................................................................................ 4 2. Chemoterapeutika na bázi komplexů přechodných kovů ............................................... 6 2.1. Protinádorová chemoterapeutika na bázi platiny ........................................................ 6 2.1.1. Historie a současnost ................................................................................................ 7 2.1.2. Mechanismus účinku ................................................................................................ 9 2.2. Chemoterapeutika jiných přechodných kovů ............................................................. 14 3. Platnaté komplexy s různými N-donorovými ligandy................................................... 17 3.1. Platnaté komplexy s deriváty 7-azaindolu ................................................................. 17 3.1.1. Cis-dichloroplatnaté komplexy s deriváty 7-azaindolu ........................................... 19 3.1.1.1. In vitro protinádorová aktivita a molekulární farmakologie ................................ 20 3.1.1.2. Protinádorová aktivita in vivo .............................................................................. 27 3.1.2. Karboxylato komplexy s deriváty 7-azaindolu ....................................................... 29 3.1.2.1. Oxalatoplatnaté komplexy s deriváty 7-azaindolu ............................................... 30 3.1.2.2. Fototoxické deriváty karboplatiny s deriváty 7-azaindolu .................................. 32 3.1.2.3. Selektivní malonato a dekanoato komplexy......................................................... 37 3.1.4. Možnosti cíleného transportu komplexů na bázi platiny ........................................ 38 3.2. Platnaté komplexy s deriváty N6-benzyladeninu ....................................................... 41 3.2.1. Protinádorová aktivita in vitro................................................................................. 42 3.2.2. Protinádorová aktivita in vivo ................................................................................. 46 4. Komplexy jiných přechodných kovů s různými N-donorovými ligandy ...................... 48 4.1. Protizánětlivá aktivita Au(I) komplexů s deriváty N6-benzyladeninu ....................... 48 4.2. Protinádorově účinné Au(I) komplexy s deriváty 7-azaindolu .................................. 52 5. Závěr.............................................................................................................................. 54 6. Literatura ....................................................................................................................... 56 7. Seznam příloh ................................................................................................................ 65 Přílohy ................................................................................................................................. 67
3
1. Úvod Léčiva na bázi koordinačních sloučenin platiny počítají svoji historii od 60. let minulého století, kdy Barnett Rosenberg, profesor chemie a biofyziky na Michigan State University, objevil protinádorové vlastnosti cis-diammin-dichloroplatnatého komplexu (cis-[Pt(NH3)2Cl2]), známého od té doby pod triviálním a posléze i obchodním názvem cisplatina [1–4]. Tato látka byla v roce 1λ78 schválena jako chemoterapeutikum protinádorové terapie a dodnes je vůdčím léčivem při léčbě různých typů rakoviny. Klinický úspěch cisplatiny byl základem rozsáhlého výzkumu mnoha vědeckých týmů, jehož výsledkem bylo a, s ohledem na nezastupitelnost metaloterapeutik na bázi platiny v onkologické praxi, stále je mnoho nových komplexů platiny, u kterých byly studovány jejich biologické vlastnosti. Některé z těchto látek pak, obdobně jako cisplatina, dosáhly klinického využití. Tyto jsou známé pod názvy karboplatina, oxaliplatina, nedaplatina, lobaplatina a heptaplatina [1–3,5]. Je obecně známým faktem, že chemoterapie léčivy na bázi platiny je spojená s negativními vedlejšími účinky, jako jsou nefrotoxicita, neurotoxicita, ototoxicita, myelosuprese nebo resistence některých typů nádorů, a to ať už vrozená nebo získaná. Přirozenou snahou bioanorganických chemiků je vývoj takových látek, které při srovnatelné nebo dokonce vyšší protinádorové aktivitě zmíněné vedlejší účinky nevykazují. Tohoto lze dosáhnout různými způsoby - vhodnou modifikací klinicky užívaných léčiv na bázi platiny užitím jiných ligandů, aktivací (např. fotoaktivace) původně neaktivních látek ve fyziologickém prostředí, zaměřením biologického účinku na jiný cíl než je jaderná DNA v případě konvenčních platnatých metaloterapeutik (např. mitochondrie a redoxní mechanismus účinku) nebo využitím principů cíleného transportu u klinicky používaných nebo nově připravených komplexů platiny (např. nanočástice oxidů železa nebo liposomy) [6–13]. Nezbytné je v této souvislosti zmínit také studium (zatím bez praktického využití v onkologické praxi) biologicky aktivních koordinačních sloučenin jiných (někdy též neplatinových) přechodných kovů, z nichž se jako medicinálně nejperspektivnější jeví komplexy ruthenia a v poslední době též osmia a iridia [11,14–20]. Některé z těchto přístupů moderní bioanorganické chemie, resp. výzkumu a vývoje nových medicinálně perspektivních koordinačních sloučenin přechodných kovů,
budou v předložené habilitační
práci
komentovány jako výsledky vědecko-výzkumné činnosti, na které se autor po ukončení doktorského studia v listopadu β010 podílel v rámci svého působení na Katedře anorganické chemie a Regionálního centra pokročilých technologií a materiálů. Za reprezentativní příklad
4
jistě mohou posloužit cis-dichloroplatnaté komplexy s halogen-deriváty 7-azaindolu, které vykazují několikanásobně vyšší in vitro cytotoxickou aktivitu na různých typech lidských nádorových buněčných linií (např. karcinom vaječníku, prsní karcinom nebo osteosarkom s hodnotami IC50 = 1,8, β,0 resp. β,5 M) ve srovnání s cisplatinou (IC50 = 12,0, 19,6 resp. γ4,β
M vůči zmíněným lidským nádorovým buněčným liniím) a jejichž in vivo
protinádorová aktivita, která je srovnatelná s cisplatinou, vyvolává ve výrazně menší míře negativní vedlejší účinky na testovaná zvířata; IC50 = letální koncentrace pro 50% testovaných buněk [21]. Předložená habilitační práce je rozdělena do dvou základních částí. V kapitole 2 jsou popsány literární poznatky o protinádorových léčivech na bázi platiny a jiných přechodných kovů se zaměřením na terapeutika na bázi platiny, jejichž deriváty byly v dosavadní vědeckovýzkumné kariéře autora jeho hlavním předmětem studia. Podkladem pro následující kapitoly 3 a 4 předložené habilitační práce je soubor jedenácti publikací, na kterých se uchazeč autorsky spolupodílel a jež byly publikované v renomovaných a v oblasti bioanorganické chemie respektovaných časopisech, jakými jsou např. Journal of Medicinal Chemistry, PLoS ONE nebo Journal of Inorganic Biochemistry. Část předložených výsledků je předmětem patentové ochrany celkem tří národních patentů a jedné národní a dvou mezinárodních přihlášek vynálezu. Závěrem lze uvést, že cíle vědecko-výzkumné činnosti uchazeče byly po ukončení doktorského studia následujícíμ 1/ design, příprava a studium fyzikálně-chemických vlastností koordinačních sloučenin platiny a jiných biologicky perspektivních přechodných kovů; 2/ studium biologické (především protinádorové) aktivity připravených sloučenin na úrovni in vitro i in vivo, se zaměřením na schopnost látek překonávat získanou rezistenci vůči klinicky používaným chemoterapeutikům a na selektivní účinek těchto látek vůči nádorovým buňkám vedoucí ke snížení negativních vedlejších účinků; 3/ studium cíleného transportu připravených biologicky aktivních koordinačních sloučenin metodou funkcionalizace magnetických nanočástic na bázi oxidů železa.
5
2. Chemoterapeutika na bázi komplex p echodných kov Bioanorganická chemie koordinačních sloučenin přechodných a vnitřně přechodných prvků je nesmírně bohatým a pestrým odvětvím chemie, které pro smysluplné naplnění svých základních cílů nutně musí spolupracovat a překrývat se s mnoha jinými chemickými (např. analytická chemie, biochemie) i nechemickými (např. biofyzika, toxikologie, farmakologie) vědními obory [2,14,19]. Její praktický význam je, s ohledem na klinické využití
mnoha
koordinačních
sloučenin
(např. protinádorová léčiva na bázi platiny, MRI kontrastní látky na bázi gadolinia), obrovský a i do dalších let tento vědní obor skýtá mnohé perspektivy, což lze dedukovat z neutuchajícího
pronikání
koordinačních
sloučenin do klinických experimentů, v jehož různých fázích se nyní nachází několik desítek komplexů různých d- a f- prvků. Nelze však nesouhlasit s obecně přijímaným tvrzením, že jestli je některá skupina látek vůdčí
mezi
biologicky
aktivními
sloučeninami na bázi komplexů přechodných kovů, jsou to, díky více než 50% podílu na celosvětové
protinádorové
Obrázek 1. Strukturní vzorce klinicky používaných metaloterapeutik na bázi platiny
chemoterapii,
komplexy platiny. 2.1. Protinádorová chemoterapeutika na bázi platiny Původním protinádorovým léčivem na bázi koordinačních sloučenin platiny je cisplatina (cis-diammin-dichloroplatnatý komplex), kterou do celosvětové klinické praxe následovaly karboplatina (diammin-cyklobutan-1,1`-dikarboxylatoplatnatý komplex [22]) a oxaliplatina (1R,2R-diaminocyklohexan-oxalatoplatnatý komplex [23,24]), na lokálních asijských trzích je dále používána nedaplatina (diammin-glykolatoplatnatý komplex [25]), lobaplatina (1,2-bis(aminomethyl)cyklobutan-laktatoplatnatý komplex [26]) a heptaplatina ((4R,5R)-4,5-bis(aminomethyl)-2-isopropyl-1,3-dioxolan-malonatoplatnatý
6
komplex
[27])
(obrázek 1) [1–3,14]. V různých fázích klinických studií se v současné době nachází několik dalších komplexů platiny, ať už v oxidačním stupni +II nebo +IV. Některé z těchto látek budou blíže okomentovány níže. 2.1.1. Historie a současnost Základy bioanorganické a medicinální chemie protinádorově aktivních koordinačních sloučenin platiny byly položeny na počátku 60. let β0. století americkým univerzitním profesorem Barnettem Rosenbergem, který jako první popsal cytotoxické účinky cis-[Pt(NH3)2Cl2] (cisplatina, obrázek 1) [4]. Zajímavostí jistě je, že původní výzkum Prof. Rosenberga nebyl orientován na biologicky aktivní komplexy platiny či jiných přechodných kovů, ale na vliv elektromagnetického (jak elektrické, tak i magnetické složky) záření na buněčné dělení bakterií. Experimenty výzkumného týmu v testovací komoře obsahující platinové elektrody sice vedly k výrazným morfologickým změnám studovaných E. coli, ale podrobnější studium tohoto procesu prokázalo, že za to nenese přímou odpovědnost
procházející
elektrický
proud,
ale
sloučeniny,
které
se
uvolňují
ze zmíněných platinových elektrod ponořených do média. Detailní analýza prokázala, že se do roztoku uvolňuje více platinu obsahujících sloučenin, které byly posléze nasyntetizovány a samostatně testovány. Proběhnuvší biologické experimenty prokázaly jako látku odpovědnou za potlačení buněčného dělení právě zmíněnou cisplatinu (nejednalo se o původní látku, byla již více než sto let známa jak tzv. Peyronova sůl [28]; pro přesnost a úplnost - cisplatina byla jedna ze dvou aktivních látek, avšak druhá z nich, cis-[Pt(NH3)2Cl4], klinické využití nenachází). Tento veskrze průlomový objev odstartoval rapidní výzkum protinádorové aktivity cisplatiny a jejího mechanismu účinku, kdy byly v nebývale krátké době léčeni první pacienti (1971) a již v roce 1978 byla cisplatina přijata Food and Drug Administration (FDA) ke klinickému využití v onkologické praxi [1–3]. Je třeba už na tomto místě zmínit, že již ve zmíněných 70. letech minulého století bylo patrné, že vynikající terapeutické schopnosti cisplatiny, která byla účinná a od počátku používaná na více typech nádorů, jsou spojeny s celou řadou negativních vedlejších účinků, jako je nefrotoxicita (toxický efekt pro ledviny), neurotoxicita (porušení normální funkce nervového systému), ototoxicita (negativní vliv na fyziologii a funkci vnitřního ucha) nebo myelosuprese (potlačení krvetvorby v kostní dřeni). V současné literatuře se lze dokonce setkat s pochybnostmi, jestli by, navzdory své aktuální vůdčí pozici mezi protinádorovými chemoterapeutiky, byla cisplatina schválena ke komerčnímu využití, pokud by klinické testování a schvalovací proces probíhal o několik desetiletí později. Neméně závažnými 7
problémy spojenými s terapií cisplatinou a jinými léčivy na bázi platiny jsou nutnost intravenózní aplikace a také rezistence (rozlišujeme vrozenou a získanou, viz níže [3,29]). Zmíněné jevy, k nimž je třeba připočítat psychické problémy onkologických pacientů, které jsou jimi vyvolány, nutně vedly k derivatizaci cisplatiny a intenzivní syntéze a studiu biologických vlastností mnoha set nových komplexů platiny (následně také komplexů jiných, neplatinových přechodných kovů). Pouze zlomek z těchto sloučenin dosáhl fáze klinického testování a pouze méně než deset látek bylo do dnešní doby přijato jako protinádorové léčivo do klinické praxe (viz výše; obrázek 1) [3]. Cisplatina je účinná především proti rakovině vaječníků a varlat, úspěšně používána je také k léčbě nádorů hlavy, krku, močového měchýře, melanomu nebo osteosarkomu. Obecně platí, že stejné nádory, jaké jsou léčeny cisplatinou, lze léčit i užitím karboplatiny (schválená celosvětově) a nedaplatiny (schválená v Japonsku). Výhodou těchto derivátů cisplatiny, které opravňují jejich klinické využití, jsou výrazně nižší nefrotoxicita, ototoxicita a neurotoxicita ve srovnání s cisplatinou. Nicméně též platí, že tyto látky jsou méně protinádorově aktivní než cisplatina. Obdobný farmakologický profil zmíněných léčiv pak souvisí s tzv. křížovou rezistencí karboplatiny, resp. nedaplatiny s cisplatinou. Z medicinálního a také, v jistém smyslu, z molekulárně-farmakologického pohledu se od výše zmíněných léčiv liší oxaliplatina [23,24]. Tato látka byla vyvinuta již v roce uvedení cisplatiny do klinické praxe (1978), pro celosvětové klinické využití byla schválena až téměř po více než dvaceti letech (2002). Oxaliplatina je také látkou, která přinesla svěží vítr do v té době již trochu unavené problematiky léčiv na bázi platiny (toho času pouze cisplatiny a karboplatiny) - je to komplex s jiným typem nosného N-donorového ligandu (obrázek 1). Jednoznačnou výhodou oxaliplatiny oproti výše uvedeným látkám je použití (v kombinované terapii s 5-fluorouracilem a leukovirinem) vůči jiným typům nádorů, především pak vůči cisplatinou a karboplatinou neléčitelné rakovině tlustého střeva. Aplikace oxaliplatiny navíc není spojena s většinou známých negativních vedlejších účinků (myelosuprese, nefrotoxicita, ototoxicita), klinicky aplikovatelná dávka oxaliplatiny je omezena jí vyvolávanou neurotoxicitou. Na základě řečeného lze předpokládat, a ve skutečnosti to také platí, že oxaliplatina nevykazuje křížovou rezistenci s cisplatinou a je účinná vůči některým nádorům se získanou resistencí vůči cisplatině [30]. Obdobně se můžeme vyjádřit i k dalším léčivům na bázi platiny, které stejně jako oxaliplatina, obsahují N-donorové ligandy odlišné od NH3 obsaženého v cisplatině, karboplatině a nedaplatině. Jsou jimi na lokálních asijských trzích používané lobaplatina (Čína; léčba chronické myelogenní leukémie, nemalobuněčného
8
karcinomu plic nebo neoperabilních karcinomů prsu) a heptaplatina (Jižní Korea; léčba rakoviny žaludku) (obrázek 1). Lze konstatovat, že hlavním cílem bioanorganických chemiků již není vývoj co nejaktivnějších nebo ve srovnání s cisplatinou méně toxických sloučenin, ale správné pochopení mechanismů rezistence nádorů vůči platinovým metaloterapeutikům a z toho vyplývající cílená syntéza látek, které překonávají schopnost nádorových buněk snížit transport léčiva z extracelulárního prostoru do prostoru intracelulárního, cytoplasmatickou detoxifikaci, schopnost buněk opravovat DNA adukty s léčivy na bázi platiny a toleranci nádorových buněk k poškození cílové molekuly jaderné DNA [7,31,32]. Příklady takových látek jsou známy jak z onkologické praxe (oxaliplatina, lobaplatina, heptaplatina), tak i mezi klinicky testovanými agens (satraplatina, bis-acetato-ammin-dichloro-cyklohexylaminplatičitý komplex; pikoplatina, cis-ammin-dichloro-2-methylpyridinplatnatý komplex [33,34]; obrázek β). Zajímavou možností je i zlepšení transportu (např. liposomální formulace známých léčiv na bázi platiny, příp. navýšení lipofility léčiva, z něhož plyne vyšší transport do buňky (lipoplatina, aroplatina [12,35])), nebo aktivní ovlivnění hladiny fyziologických látek zapojených do mechanismu rezistence nádorových buněk vůči účinkům léčiv na bázi platiny. Toto je z praxe známo např. pro glutathion (vytváří adukty s léčivy na bázi platiny v rámci a tím
intracelulární zvyšuje
jejich
z nádorové buňky) sulfoximin
inaktivace transport
a L-buthionin-
(L-BSO;
inhibitor
enzymu -glutamylcystein syntetázy, který je zapojen do fyziologické syntézy
glutathionu),
preaplikací
nebo
kdy
lze
kombinovanou
terapií účinně navýšit aktivitu léčiv i na původně necitlivých nádorových buňkách [36].
Obrázek 2. Strukturní vzorce klinicky testovaných satraplatiny a pikoplatiny
2.1.2. Mechanismus účinku Cisplatina je chemicky jednoduchý platnatý komplex obsahující dva N-donorové NH3 ligandy (tzv. carrier ligands), které zůstávají ve fyziologickém prostředí koordinovány na centrální atom, a dva chloridové ionty (tzv. leaving groups), které jsou ve fyziologickém prostředí nahrazeny ligandy jinými (viz níže) [1–3]. Z chemického úhlu pohledu je cisplatina 9
poměrně nereaktivní látka, která navíc, vzhledem k její nízké rozpustnosti ve vodě, musí být pacientům aplikována intravenózně. Základním dějem po aplikaci léčiva je jeho transport do cílové buňky, resp. buněčná akumulace (obrázek γ). Mechanismy transportu léčiv na bázi platiny sice stále nejsou zcela jednoznačně objasněny, je však známo, že je mnoho vlivů na transport do buňky (např. koncentrace Na+/K+ iontů, pH), a především, že kromě pasivní difúze
existuje
minimálně
jeden
aktivní
transport
přes
plazmatickou
membránu
z extracelulárního prostoru směrem do buňky, a to přes měďnaté transportéry CTR1 [37]. V buňce, kde je koncentrace chloridových iontů nižší (~10 mM) ve srovnání s extracelulárním prostředím (~100 mM), jsou chloridové ionty cisplatiny nahrazovány molekulami vody (obrázek γ) [38–40]. Z mechanistického hlediska chápeme tento proces jako nezbytnou aktivaci léčiva, jelikož jsou to právě kladně nabité mono- nebo diaqua komplexy (příp. produkty jejich protolytických reakcí), které se mohou vázat na cílovou molekulu jaderné DNA. Vysoce labilní Pt–O vazba (platina–voda) je v jádře nahrazena stabilnější vazbou Pt–N (platina– nukleobáze),
kterou
aktivované
léčivo
vytváří
odštěpením zmíněných aqua ligandů a následnou koordinací na dusíkové atomy nukleobází, především pak guaninu, v menší míře adeninu (obrázek 3). Ve většině případů se jedná o bifunkcionální adukty na jednom vlákně DNA (tzv. intrastrand adukty), které podle toho, na jakou nukleobázi sahají, rozlišujeme na adukty mezi dvěma sousedními guaniny (1,2-d(GpG); 60–65% aduktů), mezi sousedním guaninem a adeninem (1,2-d(ApG); 20–β5%) nebo mezi dvěma guaniny, mezi nimiž je ještě jedna nukleobáze (1,3-d(GpXpG); 2%; X = adenin, cytosin nebo uracil). Dále byly detekovány monofunkcionální adukty na guanin (2%) a také adukty mezi dvěma guaniny obou vláken DNA (tzv. interstrand adukty; 2%). Takto vzniklé adukty tvořené cisplatinou a DNA mění sekundární strukturu DNA a vedou k řadě procesů, jako je rozpoznání a oprava poškození DNA, zastavení buněčného cyklu nebo apoptóza [41].
Obrázek 3. Schematické znázornění buněčné akumulace a následných vnitrobuněčných procesů cisplatiny (nahoře) a typy Pt–DNA aduktů tvořených cisplatinou (dole). Převzato z
[40]. 10
Z terapeutického pohledu toto způsobuje inhibici DNA a RNA syntézy, jinými slovy je potlačeno buněčné dělení, což vede nádorovou buňku k buněčné smrti, z makroskopického pohledu tedy k léčbě nádorového onemocnění. Také ostatní léčiva na bázi platiny musí nezbytně podstoupit aktivaci nahrazením leaving group (což jsou u všech klinických analogů cisplatiny bidentátně koordinované karboxylátové dianionty; obrázek 1). Ačkoli tento hydrolytický proces téměř neprobíhá ve vodném prostředí, v prostředí fyziologickém jsou zmíněné karboxylato ligandy nahrazeny, podobně jako u cisplatiny, dvěma molekulami vody, přičemž do tohoto děje aktivně vstupují HCO3– a H2PO4– ionty nebo jiné fyziologicky přítomné látky (např. guanosin-5'-monofosfát, GMP) [42–44]. Aktivované diaqua komplexy opět atakují molekulu jaderné DNA, jak je popsáno výše pro cisplatinu. Je určitě vhodné zde zmínit, že v případě karboplatiny a nedaplatiny jsou aktivované částice chemickými analogy aktivovaných částic cisplatiny o složení cis-[Pt(NH3)2(H2O)2]2+, což je vysvětlením obdobného terapeutického profilu a vzájemné křížové rezistence těchto tří metaloterapeutik. Naproti tomu aktivace oxaliplatiny, lobaplatiny a heptaplatiny poskytuje jiné typy částic, jelikož tato léčiva obsahují ve svých molekulách jiné typy N-donorových ligandů (obrázek 1). Toto způsobuje rozsáhlejší změny na cílovém místě v souvislosti s přítomností objemného N-donorového ligandu, jako jsou třeba sterické efekty ve velkém žlábku DNA, které mohou vést k účinnější inhibici DNA syntézy. Zmíněné bylo prokázáno v případě oxaliplatiny, jejíž schopnost tvořit kovalentní adukty s jadernou DNA je sice nižší ve srovnání s cisplatinou, i tak je ale oxaliplatina více cytotoxická než cisplatina, protože stejný počet DNA aduktů oxaliplatiny efektivněji inhibuje buněčné dělení ve srovnání s cisplatinou [45,46]. Asi nejzávažnějším problémem chemoterapie léčivy na bázi platiny, a to jak z pohledu klinické aplikace, tak i optikou bioanorganických chemiků zabývajících se vývojem nových látek tohoto typu, je rezistence. Jak již bylo uvedeno výše, rozlišujeme rezistenci vrozenou (chápejme ji tak, že některé typy nádorů, jako jsou např. karcinomy prostaty, tlustého střeva a konečníku nebo prsu, odolávají biologickému účinku cisplatiny a karboplatiny; jinak řečeno, těmito léčivy mohou být léčeny pouze některé typy nádorů) a získanou (některé typy původně léčitelných nádorů, známo je to např. pro nádory vaječníků léčené cisplatinou a karboplatinou, jsou vůči opakované chemoterapii těmito léčivy na bázi platiny necitlivé, jinak řečeno recidiva některých nádorů vyžaduje použití jiného, často méně účinného chemoterapeutika) [3,29]. V zásadě platí, že rezistentní nádorové buňky vykazují jiné mechanismy akumulace léčiva, rozpoznání a opravy vzniklých Pt–DNA aduktů a apoptózy [47,48]. Jinak řečeno, za rezistencí stojí dva základní mechanismy - omezení množství léčiva 11
atakujícího
cílovou
molekulu
DNA
a
omezení
schopnosti vzniklých Pt–DNA aduktů vyvolat buněčnou smrt. Bylo experimentálně prokázáno, že rezistentní buňky přijímají méně léčiva, než jejich analoga před vypěstováním rezistence [47]. Jak tomuto mechanismu rezistence předejít ukazuje výše zmíněná satraplatina. Tento platičitý komplex přechází ve fyziologickém prostředí ve vícero látek, z nichž jedna je cis-ammindichloro-cyklohexylaminplatnatý komplex, který, ačkoli je chemicky nepříliš odlišný od cisplatiny, výrazně lépe proniká do nádorových buněk [49]. Jiný přístup k navýšení transportu léčiv na bázi platiny jsou výše zmíněné liposomální formulace s konvenčními (lipoplatina [12]; obrázek 4) nebo novými lipofilními (ProLindac [50]) komplexy nebo funkcionalizované
nanočástice
[6–8,51–54].
Pro
názornost uveďme, že enkapsulace léčiva (cisplatiny v případě lipoplatiny) do liposomové formulace vede
Obrázek 4. Schematické znázornění lipoplatiny, liposomální formulace nesoucí jako léčivo cisplatinu (vlevo) a porovnání její aktivity (▲) s cisplatinou (■), směsí cisplatiny a liposomů (□) a samotnými liposomy (čárkovaná čára). Převzato z regulon.com a z [55].
k výrazně vyšší účinnosti cisplatiny, čehož přímým důsledkem je nižší toxicita vůči nenádorovým buňkám, a tedy i nižší negativní vedlejší účinky [55]. V případě výše zmíněného ProLindac, derivátu oxaliplatiny s odstupujícím ligandem na bázi hydroxypropylmethakrylamidu, je z ekvitoxické dávky do cílové nádorové buňky účinně dopraveno více než desetkrát vyšší množství platiny ve srovnání s oxaliplatinou [56]. Již výše v práci byl zmíněn glutathion (GSH) jako jedna ze síru obsahujících biomolekul
(můžeme
zmínit
ještě
metalothioneiny
[57]),
které
jsou
přítomné
v intracelulárním prostoru, kde se díky vysoké afinitě síry k platině ochotně koordinují na zmíněný kov, a tím usnadňují jeho transport ven z buňky (eflux) skrze GS-X pumpy. Roli glutathionu při rezistenci nádorových buněk potvrzují i četné publikace, kde je popsána vyšší hladina GSH u rezistentních buněk ve srovnání s buňkami původními [36]. Tuto schopnost nádorových buněk chránit se cytotoxickému účinku léčiv na bázi platiny překonává, pravděpodobně díky sterickému bránění koordinace síru obsahujících biomolekul, pikoplatina (obrázek β). Tato látka odvozená od cisplatiny náhradou jedné NH3 molekuly 2-methylpyridinem (2-pikolin; na tomto místě je vhodné upozornit na chemickou podobnost zmíněného β-methylpyridinu s níže diskutovaným 7-azaindolem a jeho deriváty, jejichž 12
koordinační sloučeniny byly hlavním předmětem studia uchazeče) vede ke schopnosti této látky působit na nádorové buňky necitlivé vůči účinku jak cisplatiny, tak i oxaliplatiny. Nádorová buňka má dále celou řadu možností, jak se vyrovnat s tím, že se léčivo na bázi platiny dostalo do buňky a v buňce do jádra, kde se kovalentně navázalo na molekulu DNA. Pro příklad možností oprav Pt–DNA aduktů lze uvést MMR (mismatch repair), což je opravný systém, v rámci kterého jeho proteiny rozpoznávají poškození DNA a stimulují buňku k opravným procesům [58]. Narušením této dráhy lze naopak navyšovat citlivost léčených nádorů vůči účinku aplikovaných léčiv. Nelze v této souvislosti opomenout ani fakt, že i přes velmi dobře známé negativní vedlejší účinky léčiv na bázi platiny se je stále nedaří nahradit méně toxickými léčivy, ať už na bázi neplatinových přechodných kovů nebo látkami organickými. Jednou z alternativ, jak omezit toxicitu léčiv na bázi platiny, je kombinovaná terapie, jak ji známe u oxaliplatiny podávané od počátku jejího klinického využití s jiným protinádorovým chemoterapeutikem 5-fluoruracilem.
Obdobně
je
v klinické
praxi
karboplatina
podávaná
při
léčbě
nemalobuněčného karcinomu plic s paklitaxelem a bevacizumabem [59]. Celá problematika rezistence nádorových buněk vůči cytotoxickému ataku chemoterapeutik na bázi komplexů platiny je sice stále více a více prozkoumanou a objasněnou oblastí, nicméně celková vzájemná provázanost jednotlivých faktorů (viz obrázek 5), které do ní přispívají, je i nadále z velké části zahalena rouškou tajemství, a to
Obrázek 5. Schematické znázornění drah zapojených do zprostředkování biologického účinku vyvolaného léčivy na bázi platiny (vlevo) a mechanismy inhibující cytotoxický účinek (apoptózu) léčiv na bázi platiny v rezistentních liniích (vpravo). Převzato z [48].
13
především z toho úhlu pohledu, který se týká cíleného designu a následného biologického účinku nových léčiv. Dosavadní poznatky poukazují na to, že cílení na jeden libovolný faktor buněčné rezistence není cestou, protože léčená buňka dokáže ztrátu/omezení jednoho mechanismu rezistence efektivně nahradit mechanismy jinými. 2.2. Chemoterapeutika jiných p echodných kov Mnoho koordinačních sloučenin přechodných (Ti, V, Cr, Mn, Fe, Co, Cu, Zn, Y, Zr, Mo, Tc, Ru, Pd, Ag, La, Ta, Os, Pt, Au) a vnitřně přechodných (Ce, Sm, Gd, Ho, Lu, Ac) prvků nachází v současné době praktické medicinální využití nebo k němu směřuje v různých fázích klinických experimentů [2,14,19]. Z uvedených souvisí s prací autora protinádorově aktivní komplexy platiny (viz výše) a vybraných neplatinových prvků VIII.B. skupiny periodického systému, dále pak protizánětlivé komplexy zlata v oxidačním stupni +I. Ačkoli klinický úspěch chemoterapeutik na bázi platiny je i z dnešního pohledu neoddiskutovatelný, již v době uvedení těchto léčiv na trh bylo patrné, že je nutné vyvinout takové látky, které rozšíří spektrum léčitelných tumorů, budou méně toxické (omezení negativních vedlejších účinků metaloterapeutik na bázi platiny používaných v onkologické praxi) a schopné překonávat rezistenci vůči klinicky používaným chemoterapeutikům. Tyto požadavky vedly bioanorganické chemiky nejen k syntéze nových, vhodně strukturně modifikovaných komplexů platiny (viz výše), ale také k vývoji a studiu biologických vlastností u koordinačních sloučenin jiných přechodných prvků. Mezi těmito se jako velmi vhodní kandidáti pro klinické využití jeví komplexy ruthenia [14,60–62]. Historii těchto látek počítáme již od počátku 80. let minulého století, kdy byla publikována vysoká aktivita jednoduchého Ru(III) komplexu, fac-[RuCl3(NH3)3], vůči EMT-6 sarkomu u myší [60,63]. Fáze klinických testů dosáhly až rozpustnější iontové komplexy ruthenia KP1019/NKP-1339 nebo NAMI-A (obrázek 6) [64]. Z výhod těchto látek uveďme pro příklad schopnost KP1019/NKP-1339 snížit objem kolonorektálního karcinomu myší o 95 %, aniž by byla u testovaných zvířat pozorována jakákoli mortalita, a pouze zanedbatelný úbytek hmotnosti (6%), komplex NAMI-A je zajímavý svojí výraznou antimetastatickou aktivitou [65]. Nicméně je nutné zmínit, že mechanismus účinku těchto látek není stále zcela objasněný. Předpokládá se in vivo redukce z Ru(III) na Ru(II), ovlivnění redoxní rovnováhy léčených nádorových buněk a interakce komplexů s molekulou jaderné DNA [66]. Významnou skupinou látek, které navazují na výše uvedené Ru(III) komplexy, jsou organokovové koordinační sloučeniny s arenem navázaným na centrální atom [67,68], které se od výše uvedených komplexů mechanisticky liší a jejichž biologická aktivita (schopnost překonávat 14
rezistenci nádorů vůči cisplatině, selektivní účinek) je velice perspektivní pro následné klinické využití [14,69]. V posledních letech je stále větší pozornost věnována komplexům
osmia,
především
pak
organokovovým
sloučeninám se složením odvozeným od výše uvedených komplexů ruthenia. Obecně lze konstatovat, že komplexy osmia jsou kineticky inertnější ve srovnání s analogickými komplexy ruthenia, mechanisticky jsou si však obě skupiny podobné [např. 69,70]. Zajímavý je jistě i fakt, že izostrukturní komplexy osmia v mnoha případech převyšují analogické komplexy ruthenia, jak lze demonstrovat na komplexech [Os(µ6-p-cym)(Impy)Cl]PF6,
Obrázek 6. Strukturní vzorec komplexních aniontů sloučenin NAMI-A (vlevo) a KP1019/NKP-1339 (vpravo)
[Os(µ6-p-cym)(Impy)I]PF6,
[Ru(µ6-p-cym)(Impy)Cl]PF6 a [Os(µ6-p-cym)(Impy)I]PF6, jejichž in vitro cytotoxicita (IC50, tj. letální koncentrace pro 50% testovaných buněk) je rovná γ,0, 1,β, 16,β a γ,0 µM [69]; p-cym = p-cymen, Impy = N1,N1-dimethyl-N4-[(2-pyridyl)methylen]anilin. Zmiňme zde také podobné organokovové komplexy iridia, které jsou v posledních letech v literatuře zmiňovány jako látky výrazně protinádorově aktivní a vysoce klinicky perspektivní [např. 18,71]. Také tyto komplexy mají, obdobně jako předešlé komplexy ruthenia a osmia, na centrální atom koordinovaný aromatický kruh, kterým je v tomto případě různě derivatizovaný cyklopentadien (např. 1,β,γ,4,5-pentamethylcyklopentadien nebo 1,2,3,4-tetramethyl-5-bifenyl-cyklopentadien
(Cpbiph)).
Také
tyto
komplexy
svojí
cytotoxicitou překonávají klinicky používanou cisplatinu, současně je jejich účinek selektivní vůči nádorovým buňkám a jsou schopné působit na nádorové buňky rezistentní vůči cisplatině.
Příkladem
uveďme
alespoň
komplexy
[(η5-Cpbiph)Ir(phen)Cl]PF6
a [(η5-Cpbiph)Ir(bpy)Cl]PF6 a jejich vysokou in vitro cytotoxicitu (IC50 = 0,72, resp. 0,57 µM); phen = 1,10-fenanthrolin; bpy = 2,2`-bipyridin [72]. Historii protizánětlivě aktivních koordinačních sloučenin zlata počítáme od β0. let 20. století, kdy byla do léčby revmatoidní artritidy, jednoho ze zánětlivých onemocnění, přijata zlatná thiolátová léčiva. Příkladem takových léčiv je i dodnes používaný orálně podávaný
auranofin
(triethylfosfin-(2,3,4,6-tetra-O-acetyl-1-thio-β-D-glukopyranosato-
S)zlatný komplex (obrázek 7), který je klinicky používaný od roku 1λ85 [73]. Mezi další klinicky používané protizánětlivé Au(I) komplexy patří vedle zmíněného auranofinu ještě myocrisin
(aurothiomalát
sodný),
sanochrysin
(aurothiosulfát
sodný),
solganol
(aurothioglukosa) nebo allochrysin (aurothiopropanolsulfonát sodný) [2]. Auranofin je 15
lipofilnější než ostatní zmíněné komplexy. Je podáván orálně a v menší míře vyvolává negativní vedlejší účinky spojené s aplikací Au(I) komplexů do organismu, především pak poškození ledvin. Samotný mechanismus účinku auranofinu není zcela objasněn, je ale prokázáno, že ještě před vstupem do buňky dochází k odštěpení S-donorového ligandu, resp. k jeho nahrazení
jiným
síru
obsahujícím
ligandem
Obrázek 7. Strukturní vzorec auranofinu
(např. glutathionem) [74]. Postupně je ve vnitrobuněčném prostoru nahrazen i druhý původní ligand auranofinu (triethylfosfan) a vzniká konjugát Au(I) komplexu a proteinu (albuminu). Takto se dostává do cílového místa, kde je pravděpodobně transformován na [Au(CN)2]–, který působí proti zánětu [75].
16
Výsledková část habilitační práce
3. Platnaté komplexy s r znými N-donorovými ligandy Vývoj a studium protinádorových vlastností nových komplexů platiny byl v prvních dekádách po objevu cisplatiny orientován do dvou základních směrů - oba berou jako základní molekulu právě zmíněnou cisplatinu, jež je derivatizována buď záměnou odstupujících chloridových iontů nebo N-donorových ligandů výchozí cisplatiny. Z výše uvedeného textu je patrné, že oba přístupy byly úspěšné z pohledu uvedení léčiv do klinické praxe. Zopakujme zde ještě jednou, že to bylo právě zavedení jiných N-donorových ligandů do struktury protinádorově aktivních komplexů platiny, které vlilo novou energii do vývoje metaloterapeutik na bázi platiny, a to tím, že bylo prokázáno (oxaliplatina) a následně potvrzeno (lobaplatina, heptaplatina, nebo klinicky studovaná pikoplatina), že právě typ N-donorového ligandu má vliv na typ léčitelných a léčených nádorů. Současný přístup hledání nových potenciálních léčiv na bázi platiny lze pak shrnout tak, že je snahou výzkumných skupin
připravit
komplexy
platiny
s novými
nosnými
N-donorovými
ligandy
(např. cis-dichloroplatnaté komplexy, představující přímé deriváty cisplatiny), které prokáží dostatečný, pokud možno vyšší ve srovnání s cisplatinou, biologický účinek, který pak je následně modifikován, ať už derivatizací samotného N-donorového ligandu nebo zavedením jiných odstupujících skupin místo zmíněných chloridových iontů. Neopomenutelným přístupem pak je také studium biologických vlastností nově vyvinutých komplexů platiny v různých nadmolekulárních formulacích (liposomy, funkcionalizované nanočástice atd.). Heterocyklické sloučeniny, které byly použity jako N-donorové ligandy různých typů platnatých komplexů, jejichž studiem se uchazeč zabýval nebo se na něm podílel, jsou v zásadě dvojího typu - jedná se o deriváty 7-azaindolu a N6-benzyladeninu. Obě uvedené skupiny organických látek vykazují určité společné vlastnosti. Jsou odvozené od biomolekul (indolu a adeninu), jsou to polydentátní ligandy koordinačních sloučenin a, v neposlední řadě, mnohé organické látky obsahující uvedené základní motivy vykazují výrazné biologické účinky včetně účinků protinádorových. 3.1. Platnaté komplexy s deriváty 7-azaindolu Přehled o rozsahu chemie koordinačních sloučenin platiny se 7-azaindolem (obrázek 8) a jeho deriváty si můžeme udělat z relevantních chemických databází. Krystalografická
17
strukturní databáze (CSD, [76]) čítá ke dni 2. 3. 2015 celkem 53 molekulových struktur, ve kterých je na centrální atom platiny koordinován přes atom/atomy dusíku 7-azaindol [např. 77] nebo jeho derivát [např. 78–80]. Pokud by poznatky z krystalografické databáze měly být zobecněny, jedná se ve většině případů o jednojaderné organokovové komplexy s bidentátně koordinovaným N-donorovým ligandem tvořeným
Obrázek 8. Strukturní vzorec 7-azaindolu zobrazený se schématem číslování jeho uhlíkových a dusíkových atomů.
dvěma 7-azaindolovými motivy spojenými přes oba N1 atomy organickým linkerem (ten se v některých případech sám koordinuje přes atom uhlíku) [79,80]. Z těch několika málo odlišných struktur jmenujme alespoň látky popsané v níže uvedených pracích, na kterých se uchazeč podílel (obsahují monodentátní ligandy na bázi 7-azaindolu), nebo zajímavé struktury komplexů o složení [NBu4][Pt(C6F5)2(aza)(aza–)] (obsahuje jednu molekulu 7-azaindolu koordinovanou typicky přes atom N7 a jeden deprotonizovaný 7-azaindol
koordinovaný
přes
atom
N1)
a
[NBu4]2[{Pt(C6F5)2}2(µ-OH)(µ-aza–)]
(s deprotonizovaným 7-azaindolem koordinovaným na dvě kovová centra přes atomy N1 a N7) [77], příp. komplexy se zajímavými bidentátními deriváty 7-azaindolu (např. 1-(2pyridyl)-7-azaindolem), které se na centrální atom koordinují přes N7 atom 7-azaindolového kruhu a pyridylový dusík, což dohromady dává stabilní, šestičetný, centrální atom obsahující cyklus [78]. V databázi SciFinder® bylo nalezeno celkem 1γ6 látek obsahujících 7-azaindolový motiv a platinu, 111 z nich představuje koordinační sloučeniny s Pt–N vazbou mezi centrálním atomem a některým z dusíků 7-azaindolového cyklu. U většiny z těchto látek nacházíme atom platiny koordinovaný na dusíkový atom N7, pět látek obsahuje můstkující ligand na bázi deprotonizovaného 7-azaindolu [77,81,82], ve struktuře tří komplexů se 7-azaindolový kruh koordinuje na centrální atom výhradně přes N1 dusík pyrolového kruhu [77]. Následující text bude zaměřen na diskuzi výsledků dosažených v rámci studia platnatých komplexů s N-donorovými deriváty na bázi 7-azaindolu publikovaných v pracích, na kterých se uchazeč experimentálně a autorsky podílel. Je vhodné už na tomto místě zmínit fakt, že vysoká medicinální perspektiva níže diskutovaných cis-dichloroplatnaých a cis-dijodoplatnaých komplexů s deriváty 7-azaindolu byla zohledněna podáním jedné mezinárodní (EP13001033.3) a tří národních přihlášek vynálezu, na jejichž základě byly prozatím uděleny národní patenty CZ303417 (Dichlorido komplexy platiny s halogenderiváty 7-azaindolu, způsob jejich přípravy a použití těchto komplexů jako léčiv v protinádorové 18
terapii. Původciμ Z. Trávníček, P. Štarha, I. Popa. Majitelμ Univerzita Palackého v Olomouci. Datum uděleníμ 26. 7. 2012) a CZ303560 (Použití dichlorido komplexů platiny s halogenderiváty 7-azaindolu pro přípravu léčiv pro léčbu nádorových onemocnění. Původciμ Z. Trávníček, P. Štarha, Z. Dvořák. Majitelμ Univerzita Palackého v Olomouci. Datum uděleníμ 24. 10. 2012). 3.1.1. Cis-dichloroplatnaté komplexy s deriváty 7-azaindolu Syntéza platnatých dichloro komplexů s deriváty 7-azaindolu je zajímavá pro svoji jednoduchost, rychlost a možnost poskytnout vysoce biologicky aktivní látky jednokrokovou syntézou, což neplatí pro cisplatinu, od které lze zmíněné komplexy formálně odvozovat. Pokud se ve stručnosti podíváme na průmyslovou syntézu cisplatiny, tak jejím základním krokem je výroba tetrajodoplatnatanu z výchozího tetrachloroplatnatanu, následuje interakce s amoniakem resp. vhodnou amonnou solí za vzniku cis-diammin-dijodoplatnatého komplexu, který podléhá další reakci se stříbrnou solí (např. AgNO3), čímž vzniká komplexní cis-diammin-diaquaplatnatý kationt, který je následně převeden na cisplatinu reakcí s nadbytkem alkalického chloridu. Tuto obecně platnou (např. i pro karboxylato komplexy, viz níže pro oxaliplatinu) a široce používanou metodu přípravy různých typů platnatých komplexů známe z literatury jako tzv. Dharovu metodu [83]. V práci Štarha et al., Polyhedron,
2012
(příloha
1)
je
popsána
syntéza
dichloroplatnatého
komplexu
se 7-azaindolem, která pak byla základem pro syntézu série analogů s četnými halogenderiváty 7-azaindolu (Štarha et al., J. Inorg. Biochem., 2012 (příloha β), Muchová et al., J. Biol. Inorg. Chem., 2013 (příloha γ), Štarha et al., PLoS ONE, 2014 (příloha 4)). Nespornou výhodou pro nově vznikající sérii dichloro komplexů bylo to, že k syntéze cis-dichloroplatnatého
komplexu
se
7-azaindolem v chemické a izomerické čistotě odpovídající
minimálním
požadavkům
na biologické testování (≥λ5%) nebylo potřeba sáhnout k výše rozepsané Dharově metodě - tyto látky byly připraveny přímou jednokrokovou syntézou z K2[PtCl4] (viz obrázek 9). Výhody jsou nasnadě - méně reakčních kroků zkracuje reakční čas a zvyšuje výtěžek, navíc tím, že odpadá nutnost použití Ag(I) soli, je současně
19
Obrázek 9. Signály N1–H vodíkových atomů cis- a trans-izomeru komplexu [PtCl2(3Claza)2] prokazující ≥λ5% izomerickou čistotu studovaného cis-izomeru.
vyřešen problém kontaminace produktu stříbrem, jak je známo z průmyslové výroby cisplatiny. 3.1.1.1. In vitro protinádorová aktivita a molekulární farmakologie Zmíněný komplex cis-[PtCl2(aza)2], popsaný v práci Štarha et al., Polyhedron, 2012 (příloha 1), nevykazoval žádnou in vitro cytotoxickou aktivitu (IC50 > 1,0 µM) na lidském prsním karcinomu MCF7 a osteosarkomu HOS, toto lze ovšem připočíst omezené rozpustnosti studovaného komplexu v médiu biologického experimentu. Koncept studia platnatých komplexů s N-donorovými ligandy na bázi 7-azaindolu nicméně od počátku nestál pouze na samotném 7-azaindolu, ale především na jeho derivátech. Z více důvodů (komerční dostupnost, různé typy substituentů, různé substituované polohy, možnost dalších chemických derivatizací) se jako velmi vhodné jevily halogenderiváty 7-azaindolu. Celkem bylo pro syntézu cis-dichloroplatnatých komplexů použito osm takových derivátů, konkrétně se jedná o monosubstituované deriváty γ-chlor-7-azaindol (3Claza), 3-brom-7-azaindol (3Braza), 3-jod-7-azaindol (3Iaza), 4-chlor-7-azaindol (4Claza), 4-brom-7azaindol (4Braza) a 5-brom-7-azaindol (5Braza), a disubstituované deriváty γ-chlor-5-brom7-azaindol (3Cl5Braza) a 3-jod-5-brom-7-azaindol (3I5Braza). Detailně prostudované, tzn. in vitro cytotoxicita vůči panelu lidských nádorových buněčných linií, studium molekulárně-farmakologických vlastností a dokonce také in vivo studium protinádorových vlastností na myším modelu leukémie, jsou prozatím komplexy s 3Claza, 3Iaza a 5Braza, což je detailně popsáno a diskutováno v pracích Štarha et al., J. Inorg. Biochem., 2012 (příloha 2), Muchová et al., J. Biol. Inorg. Chem., 2013 (příloha γ) a Štarha et al., PLoS ONE, 2014 (příloha 4). U komplexů s ostatními jmenovanými ligandy je známa jejich in vitro cytotoxicita na sérii vybraných lidských nádorových buněčných linií (viz níže). V práci Štarha et al., J. Inorg. Biochem., 2012 (příloha β) je popsána jednokroková syntéza tří cis-dichloroplatnatých komplexů s 3Claza, 3Iaza nebo 5Braza, jejíž výchozí platnatou solí je výše zmíněný tetrachloroplatnatan draselný. Produkty byly detailně fyzikálně-chemicky charakterizované např. multinukleární NMR spektroskopií, hmotnostní spektrometrií a, v případě komplexu cis-[PtCl2(3Claza)2], také monokrystalovou rentgenovou strukturní analýzou (viz příloha β, obr. 2 na str. 60). Komplexy byly testovány in vitro vůči třem vybraným lidským nádorovým buněčným liniím, konkrétně vůči MCF7, HOS a karcinomu prostaty LNCaP. Získané výsledky ukázaly, že nové látky jsou výrazně (ve většině případů statisticky významně na p < 0,05 hladině významnosti) aktivnější ve srovnání s cisplatinou (viz příloha β, tabulka 2 a obr. 3 na str. 61). Nejaktivnějším byl 20
komplex s 5Braza, který byl na zmíněných lidských nádorových buněčných liniích přibližně λ,8×, 1γ,7× resp. β,5× aktivnější ve srovnání s cisplatinou, což jsou rozhodně výsledky velice perspektivní a předurčující danou látku pro studium dalších aspektů protinádorové aktivity. Již v práci Štarha et al., J. Inorg. Biochem., 2012 (příloha β) byla studována schopnost výše uvedených komplexů a pro srovnání také cisplatiny vytvářet adukty s fragmentem DNA a schopnost těchto vzniklých aduktů inhibovat RNA syntézu T7 RNA polymerázou. Na základě získaných výsledků (viz příloha β, obr. 4 na str. 61) lze konstatovat, že studované cis-dichloroplatnaté komplexy se váží na DNA podobným způsobem jako cisplatina a efektivně inhibují syntézu RNA výše zmíněnou RNA polymerázou, což je mechanismus známý z literatury jako jeden z těch, které přivádí buňku k její smrti, příp. apoptóze [84]. Hlubší molekulárně farmakologické poznatky o cis-dichloroplatnatých komplexech s deriváty 7-azaindolu (byly publikovány pouze výsledky s komplexy obsahující 3Claza a 3Iaza, výsledky získané na komplexu s 5Braza byly v některých experimentech ovlivněny jeho nižší rozpustností ve srovnání se zbylými dvěma analogy), resp. o faktorech souvisejících s jejich výraznou in vitro cytotoxicitou, jsou rozepsány a diskutovány v práci Muchová et al., J. Biol. Inorg. Chem., 2013 (příloha γ). Studie probíhaly na lidské nádorové linii karcinomu vaječníku A2780, která v oblasti protinádorově aktivních komplexů platiny platí za modelovou linii standardně používanou nejen pro stanovení cytotoxicity, ale i jako model pro studium resistence (u linie Aβ780 lze opakovaným přidáváním daného léčiva (např. cisplatiny) vypěstovat rezistenci) či mnohých mechanistických dějů. Studium prokázalo, že nahrazení obou NH3 molekul ve struktuře cisplatiny halogenderiváty 7-azaindolu výrazně ovlivnilo základní mechanistický krok, kterým je akumulace působícího terapeutika cílovou buňkou. Konkrétně bylo prokázáno, že obsah platiny v Aβ780 buňkách atakovaných komplexy se 7-azaindoly je po 24 h přibližně 45× vyšší, než je tomu u buněk, které interagovaly s cisplatinou. S tímto úzce souvisí i platinace DNA, vypovídající jednak o množství platiny uvnitř nádorové buňky, resp. jejího jádra, druhak o schopnosti léčiva interagovat s cílovou molekulou jaderné DNA. I zde byl po β4 h odhalen nezanedbatelný, zhruba čtrnáctinásobný rozdíl mezi novými komplexy a cisplatinou (viz příloha γ, tabulka 1 na str. 585). Protože nelze pro všechny komplexy platiny automaticky předpokládat molekulu jaderné DNA jako jediný možný cíl, je vhodné přesvědčit se o jejich schopnosti vázat se na DNA experimentálně. V diskutované práci (Muchová et al., J. Biol. Inorg. Chem., 2013) je tedy, kromě výše uvedeného pokusu popisujícího míru platinace jaderné DNA, použit ještě model interakce studovaných látek s DNA v bezbuněčném prostředí. Z výsledků experimentů plyne, že se studované komplexy efektivně váží na molekulu DNA a že ca λ0 % molekul 21
komplexů se 7-azaindoly je po β4 h vázáno na DNA bifunkcionálně. Bylo kvantifikováno množství interstrand aduktů (viz příloha γ, obr. 4 na str. 585), které pro nové komplexy činilo ca 5%, pro cisplatinu to bylo procent šest. V souladu s předpoklady pak zbytek připadá na intrastrand adukty. Je známo, že cisplatina (stejně jako mnohé další komplexy platiny) vyvolává apoptózu a v menší míře také nekrózu a že tyto látky zastavují buněčný cyklus v S a/nebo Gβ fázi. Pro komplexy se 7-azaindoly bylo zjištěno, že ve studovaných ovariálních buňkách ve srovnání s cisplatinou výrazně více indukují apoptózu, která navíc u komplexů se 7-azaindoly významně převyšuje nekrózu (viz příloha γ, obr. β na str. 58γ). Analýza průtokovou cytometrií prokázala, že se velká část buněk atakovaných komplexy se 7-azaindoly, na rozdíl od cisplatiny, nachází v sub-G1 fázi buněčného cyklu, zato poměr v S a Gβ fázích je opačný, tedy v těchto fázích buněčného cyklu nacházíme především buňky, k nimž byla jako testovaná látka přidána cisplatina (viz příloha γ, obr. γ na str. 58γ). Zajímavý je také nárůst procentového zastoupení buněk léčených komplexy se 7-azaindoly v subG1 fázi při zvýšení koncentrace studovaných látek z 3,0 na 5,0 µM, a to na úkor buněk v G0/G1 fázi. Z uvedeného je patrný odlišný terapeutický mechanismus účinku nových komplexů s halogenderiváty 7-azaindolu od cisplatiny. I po zmíněném navýšení koncentrace testovaných látek se 7-azaindoly zůstává poměr buněk v G2 fázi nižší ve srovnání s cisplatinou léčenými buňkami a dokonce i s kontrolní neléčenou skupinou buněk. Toto je zajímavé s ohledem na výsledný cytotoxický účinek studovaných látek, který je výrazně vyšší ve srovnání s cisplatinou. Lze z toho usuzovat, že změny vyvolané vytvořením Pt–DNA aduktů s komplexy se 7-azaidoly mohou být pro buňku daleko závažnější, hůře opravitelné a efektivněji vedoucí buňku k její smrti (apoptóze). I z tohoto důvodu byly na studovaných látkách, a pro srovnání i na cisplatině, provedeny experimenty na uvedenými komplexy platiny modifikovaném pUC1λ plasmidu, které poskytly informace o efektivitě oprav poškození DNA studovanými komplexy platiny. Jak je patrné z dosažených výsledků (viz příloha γ, obr. 6 na str. 587), množství produktů oprav poškozené DNA, přeneseně tedy schopnost buňky působit mechanismy nukleotidové excisní reparace (NER) vůči terapeutikem vyvolanému poškození DNA, je pro nové platnaté komplexy s halogenderiváty 7-azindolu nižší ve srovnání s cisplatinou. Uvedené lze chápat jako naprosto zásadní poznatek, který je jednou z hlavních příčin vyššího cytotoxického účinku komplexů se 7-azaindoly ve srovnání s konvenční cisplatinou a může souviset i se schopností testovaných látek působit na nádorové buňky rezistentní vůči cisplatině. Výše je uvedeno, že buňky ovariálního karcinomu jsou vhodným a často používaným modelem pro studium schopnosti testovaných látek překonávat získanou rezistenci tohoto typu nádoru vůči 22
farmakologickému
účinku
cisplatiny.
Výsledkem tohoto experimentu je porovnání hodnot IC50 získaných na cisplatina-senzitivní (A2780)
a cisplatina-rezistentní
(Aβ780R)
buněčné linii lidského karcinomu vaječníku (obrázek
Poměr
10).
IC50(A2780R)
ku IC50(Aβ780), známý z literatury jako tzv. rezistenční faktor (RF; též, faktor rezistence nebo rezistenční index) je pak pro testované komplexy
po
72
cis-[PtCl2(3Claza)2]
h
roven a
0,38
0,44
pro pro
Obrázek 10. Graf zobrazující výsledky in vitro cytotoxicity cis-[PtCl2(3Claza)2], cis-[PtCl2(3Iaza)2] a cisplatiny vůči cisplatina-senzitivní (Aβ780) a cisplatina-rezistentní (Aβ780R) buněčné linii lidského karcinomu vaječníku.
cis-[PtCl2(3Iaza)2], což jsou hodnoty zásadně odlišné od cisplatiny s RF = 3,9. Z celkového pohledu tedy lze konstatovat, že studované cis-dichloroplatnaté
komplexy
s
halogenderiváty
7-azaindolu
vykazují
výrazné
farmakologické výhody ve srovnání s konvenčním protinádorovým chemoterapeutikem cisplatinou, které jsou důležitým faktorem, na jehož základě mohly být tyto látky přijaty do návazných biologických studií zahrnujících také in vivo experimenty na myších. V přímé návaznosti na výše diskutované práce Štarha et al., J. Inorg. Biochem., 2012 (příloha β) a Muchová et al., J. Biol. Inorg. Chem., 2013 (příloha γ) byla v práci Štarha et al., PLoS ONE, 2014 (příloha 4) u výše zmíněných cis-dichloroplatnatých komplexů s halogenderiváty 7-azaindolu 3Claza, 3Iaza a 5Braza popsána jejich in vitro cytotoxicita na širokém panelu lidských nádorových buněčných linií a především in vivo protinádorová aktivita na myších (viz níže). In vitro cytotoxicita byla studována na lidských nádorových buněčných liniích A2780, A2780R, karcinomu plic A549, karcinomu děložního čípku HeLa a maligního melanomu G-γ61 (viz příloha 4, tabulka 1 a obr. 4 na str. 7), čímž doplnily MCF7, HOS a LNCaP (Štarha et al., J. Inorg. Biochem., 2012, příloha β) na celkový počet osmi nádorových linií. Stejně jako v práci Štarha et al., J. Inorg. Biochem., 2012 (příloha β) byl i v této práci tím nejaktivnějším komplexem ten, který obsahuje 5-brom-7-azaindol. Tato látka byla na uvedených liniích použitých v diskutované práci Štarha et al., PLoS ONE, 2014 (příloha 4) přibližně 5.γ× (A54λ), β.γ× (HeLa), 5.7× (G-γ61), 6.7× (Aβ780) a 1β.λ× (Aβ780R) cytotoxičtější (statisticky významně na hladině významnosti p < 0,05), než použitý standard cisplatina. Také v této práci byl z poměru aktivit na Aβ780R a Aβ780 vypočítán rezistenční faktor, který je pro studované komplexy přibližně β× nižší ve srovnání s cisplatinou, což lze v rámci diskutované práce dát do přímé souvislosti s neochotou nových 23
platnatých komplexů interagovat se síru obsahujícími biomolekulami, pro které je z literatury znám jejich přímý vliv na rezistenci nádorových buněk vůči cisplatině a jiným protinádorově aktivním platnatým komplexům. Za zmínku jistě stojí i další cis-dichloroplatnaté komplexy s odlišně substituovanými halogenderiváty 7-azaindolu, které byly v návaznosti na výše uvedené práce připraveny, charakterizovány a testovány na jejich in vitro cytotoxickou aktivitu vůči panelu lidských nádorových buněčných linií (manuskript zaslaný do MedChemComm). V konkrétním přiblížení se jedná o komplexy s 4Claza, 3Braza, 4Braza, 3Cl5Braza a 3I5Braza. Složení bylo prokázáno kombinací hmotnostní spektrometrie (byly detekovány molekulové píky studovaných látek; obrázek 11) a multinukleární NMR spektroskopie, která jednoznačně potvrdila přítomnost organických ligandů ve struktuře studovaných koordinačních sloučenin (v 1H,
13
C a
15
N NMR spektrech byly detekovány všechny signály příslušných derivátů
7-azaindolu) a jejich koordinaci na centrální atom přes atom dusíku N7, což lze konstatovat na základě porovnání koordinačních posunů (Δδ, ppm) vypočtených jako rozdíl chemických posunů příslušných signálů ve spektrech komplexu a volného ligandu (Δδ = δkomplex – δligand). Pro atom N1 je Δδ = 2.5–γ.6 ppm, zatímco pro N7 je hodnota koordinačního posunu –114.6– (–101.0) ppm. In vitro cytotoxicita byla studována vůči celkem osmi lidským nádorovým buněčným liniím (jmenovitě Aβ780, Aβ780R, HOS, MCF7, Gγ61, A549, HeLa a LNCaP) a linii zdravých lidských fibroblastů MRC5 (tabulka 1). Z výsledků testování in vitro cytotoxicity shrnutých v tabulce 1 je patrné, že všechny komplexy s monosubstituovanými deriváty 7-azaindolu významně cisplatina
(A–C) (p < 0,05)
byly
statisticky
aktivnější
než
vůči HOS, MCF7, Aβ780,
Aβ780R, A54λ a HeLa. Poněkud odlišně se chovaly
komplexy
D
a
E
s disubstituovanými deriváty 7-azaindolu, jejichž aktivita se vzájemně řádově liší a vyšší in vitro cytotoxicitu tak ve srovnání s cisplatinou vykazuje pouze komplex D, a to na liniích HOS, Aβ780 a Aβ780R. Uvedené statisticky významné rozdíly ve srovnání s konvenčním chemoterapeutikem
Obrázek 11. ESI– hmotnostní spektrum studovaného komplexu cis-[PtCl2(3Braza)2] s označeným (*) píkem {[PtCl2(3Braza)2]–H}– a s vloženým porovnáním jeho experimentálně zjištěného a teoretického izotopového rozložení
24
cisplatinou poukazují na schopnost studovaných komplexů překonávat vrozenou rezistenci některých typů nádorů vůči biologickému účinku zmíněného léčiva. Aplikovaný model testování schopnosti studovaných látek překonávat získanou rezistenci buněk karcinomu vaječníku prokázal, že všechny nové komplexy se 7-azaindoly tuto schopnost mají (viz hodnoty RF v tabulce 1). Studium vlivu testovaných látek na zdravé lidské fibroblasty prokázalo, že s výjimkou komplexu D je vliv všech nových platnatých komplexů s deriváty 7-azaindolu na nenádorové buňky výrazný a ve srovnání s cisplatinou vyšší (viz hodnoty SF v tabulce 1). Jedním dechem je však nutné dodat, že toto nemusí korelovat s celkovým vlivem testovaných látek na organismus a vyvolávání negativních vedlejších účinků, jak je popsáno v práci Štarha et al., PLoS ONE, 2014 (příloha 4). Co se týká porovnání in vitro cytotoxicity těchto pěti cis-dichloroplatnatých komplexů s výše uvedenými komplexy s jinými deriváty 7-azaindolu (3Claza, 3Iaza a 5Braza), lze konstatovat, že s výjimkou HeLa buněčné linie je nejaktivnější látkou cis-[PtCl2(5Braza)2].
Tabulka 1. Výsledky testování in vitro cytotoxicity cis-dichloroplatnatých komplexů s 4-chlor-7-azaindolem (4Claza; komplex A), 3-brom-7-azaindolem (3Braza; B), 4-brom-7-azaindolem (4Braza; C), 3-chlor-5-brom-7azaindolem (3Cl5Braza; D) a 3-jod-5-brom-7-azaindolem (3I5Braza; E) a cisplatiny vůči lidským nádorovým buněčným liniím prsního karcinomu MCF7, osteosarkomu HOS, cisplatina-senzitivního karcinomu vaječníku A2780, cisplatina-rezistentního karcinomu vaječníku Aβ780R, karcinomu plic A54λ, karcinomu děložního čípku HeLa, maligního melanomu Gγ61 a karcinomu prostaty LNCaP, a buněčné linii lidského zdravého fibroblastu MRC5. Buňky interagovaly s testovanými látkami β4 h. Data jsou uvedena jako IC50 ± SD ( M). Linie / faktor A B C D E Cisplatina MCF7
γ.5±1.0*
5.γ±0.8*
β.7±1.β*
>10.0a
>50.0a
18.1±5.1
HOS
7.5±β.4*
5.γ±β.1*
4.5±β.7*
4.γ±0.5*
46.7±4.0
β5.4±8.5
A2780
γ.8±0.β*
4.1±0.4*
γ.8±0.5*
4.8±1.0*
γγ.4±γ.γ
β1.8±γ.λ
A2780R
γ.6±0.7*
γ.6±0.7*
γ.5±1.1*
γ.8±1.0*
β4.7±1.0
γβ.0±λ.6
A549
11.6±4.β*
1λ.0±5.6*
11.1±0.γ*
>10.0
HeLa
11.λ±1.β*
17.1±0.8*
λ.β±β.0*
>10.0a
G361
γ.β±0.5
β.λ±0.4
β.7±0.4
γ.0±0.5
LNCaP
γ.λ±0.1
4.λ±0.1
4.0±0.6
>10.0
MRC5
γ.5±0.γ
4.0±0.γ
γ.6±0.γ
RFb
0.95
0.88
c
0.92
0.98
SF
a
>50.0
a
>50.0a
>50.0a γλ.λ±4.6
>50.0
a
5.8±β.4
>50.0
a
γ.8±1.5
17.0±4.8
>20.0
a
>50.0a
0.92
0.79
0.74
1.47
0.95
3.54
>0.60
>2.29
a
* značí statisticky významný rozdíl hodnot od standardu cisplatiny na p < 0,05 hladině významnosti (ANOVA) a ) IC50 nebylo dosaženo do koncentrace dané rozpustností příslušného komplexu v testovacím médiu b ) RF = rezistenční faktor vypočítaný jako IC50(A2780R)/IC50(A2780) c ) SF = faktor selektivity vypočítaný jako IC50(MRC5)/IC50(A2780)
25
Obrázek 12. ESI– hmotnostní spektra směsi studovaného komplexu cis-[PtCl2(3Braza)2] s glutathionem zaznamenaná ihned (0 h), 1 nebo β4 h po smíchání ve směsi voda/methanol (1:1, v/v) s šedým polem vyznačujícím oblast, ve které byl po β4 h detekován adukt o složení {[PtCl(3Braza)2(GSH)]–2H}– a s vloženým porovnáním jeho experimentálně zjištěného a teoretického izotopového rozložení
Obrázek 13. 1H NMR spektra komplexu cis-[PtCl2(4Braza)2] v DMF-d7 a ve směsi DMF-d7/H2O (4:1, v/v) bez nebo s glutathionem (GSH) měřená po dobu dvou dnů v intervalu 24 h. ○ = N1–H, cis-[PtCl2(4Braza)2]; * = N1–H, trans-[PtCl2(4Braza)2]; ● = N1–H, hydrolytický produkt; ♦ = N1–H, cis-[PtCl(4Braza)2(GSH)]; ^ = C6–H, cis-[PtCl2(4Braza)2]; # = C6–H, trans-[PtCl2(4Braza)2]; ◊ = N–H glycinu a cysteinu obsažených v GSH
Stejně jako v případě předchozí série cis-dichloroplatnatých komplexů s deriváty 7-azaindolu byla i v případě těchto látek, konkrétně u reprezentativního komplexu cis-[PtCl2(3Braza)2], studována optikou ESI hmotnostní spektrometrie schopnost těchto látek interagovat se, z mechanistického pohledu aktivity protinádorově aktivních komplexu platiny, stěžejní síru obsahující biomolekulou glutathionem (GSH). Na rozdíl od experimentů popsaných v práci Štarha et al., PLoS ONE, 2014 (příloha 4) byly v tomto případě interakční experimenty provedeny na delší časové škále (β4 h). Po uplynutí zmíněné doby byl v ESI–
26
hmotovém spektru detekován pík odpovídající aduktu {[PtCl(3Braza)2(GSH)]–2H}– (obrázek 12). Z tohoto lze usuzovat, že u těchto komplexů je jeden chloridový aniont nahrazován molekulou glutathionu.1H NMR spektroskopie byla užita ke studiu výše zmíněné interakce s glutathionem a hydrolytické stability ve směsném rozpouštědle DMF-d7/H2O (4:1, v/v), tentokráte u komplexu cis-[PtCl2(4Braza)2]. Spektra byla zaznamenávána v intervalu β4 h po dobu pěti dnů. Také tyto NMR experimenty jednoznačně prokázaly schopnost těchto komplexů interagovat s glutathionem a také podléhat hydrolytickým dějům za vzniku nových částic
předpokládaného
složení
cis-[Pt(4Braza)2Cl(H2O)]+,
cis-[Pt(H2O)2(4Braza)2]2+,
případně produktům jejich protolytických dějů (obrázek 13). U komplexů (aplikovány v koncentraci β0 µM) byla dále na purifikovaném proteasomu získaném z lidských nádorových buněk karcinomu vaječníku Aβ780 studována jejich schopnost inhibovat aktivitu β0S proteasomu. Bylo zjištěno, že žádný z komplexů neinhiboval katalytickou aktivitu zmíněného proteasomu. 3.1.1.2. Protinádorová aktivita in vivo V práci Štarha et al., PLoS ONE, 2014 (příloha 4) jsou, kromě výše zmíněného, popsány výsledky in vivo protinádorové aktivity u cis-dichloroplatnatých komplexů s 3Claza, 3Iaza a 5Braza na myším modelu lymfocytární leukémie L1210 (samcům myší byla injekcí aplikována suspenze β∙106 buněk myší leukémie L1β10; po desetidenní indukční době byly u myší makroskopicky patrné sekundární tumory; následujících sedm dnů byla zvířata léčena intraperitoneálně injekčně podávanými komplexy v dávce γ mg/kg). U testovaných myší byla denně zaznamenávána jejich hmotnost a zdravotní stav jako takový. Zvířata byla utracena, pokud hmotnost zvířete klesla pod 80% - rakovinné tkáně byly následně odebrány k histologickému
a
histochemickému
studiu
(např.
imunohistochemická
detekce
transkripčního faktoru p5γ, kaspáz 3 a 8, a zánětlivého cytokinu TNF-α). Za základní výsledek považujeme stanovení průměrné doby přežití vztažené vůči neléčené skupině myší (kontrola), což je v práci popsáno jako %T/C vypočtené jako podíl průměrné doby přežití léčené a neléčené skupiny myší (viz příloha 4, tabulka β na str. 8). Je patrné, že terapeutický účinek nových platnatých komplexů se 7-azaindoly (%T/C = 97,1–100,0) neznamenal pro léčená zvířata žádnou výraznou změnu v porovnání s neléčenou skupinou, což ale nelze brát jako negativní výsledek, přihlédneme-li k hodnotě %T/C = λγ,γ, která je výsledkem pro skupinu zvířat léčených cisplatinou. Už jen z tohoto důvodu nelze brát hodnotu %T/C jako jediný výsledek proběhnuvšího in vivo testování, ale je nutné pohlédnout na výsledky komplexněji se zvážením více experimentálních parametrů. Takovými parametry, 27
stanovenými post mortem, byly hmotnost zvířat, hmotnost nádorové tkáně a její procentuální poměr k celkové tělesné hmotnosti (viz příloha 4, tabulka β na str. 8). Hodnoty těchto parametrů v zásadě korelovaly s výše uvedenými hodnotami %T/C, jelikož pro cisplatinu byl sice pozorován nejvýraznější úbytek hmotnosti nádorové tkáně, který byl ovšem doprovázen také výrazným snížením celkové hmotnosti testovaných zvířat, což v důsledku vedlo k vysoké mortalitě zvířat této skupiny a tedy k nízké hodnotě %T/C. Na druhou stranu vliv platnatých komplexů se 7-azaindoly na hmotnost nádorové tkáně nebyl tak výrazný, tyto látky však v daleko menší míře působily na celkovou hmotnost testovaných zvířat. S tímto souvisí také hodnocení celkového zdravotního stavu testovaných zvířat v průběhu experimentu, který u nových komplexů se 7-azaindolovými deriváty nevykazoval žádné rozdíly oproti normálnímu chování, zato zvířata léčená cisplatinou vykazovala, kromě zmíněného úbytku na hmotnosti, ještě další negativní příznaky spojené s aplikací metaloterapeutik na bázi platiny, jako je únava, nechutenství nebo nenormální projevy chování. Vše zmíněné přispělo ke konečnému pozitivnímu hodnocení cis-dichloroplatnatých komplexů s halogenderiváty 7-azaindolu ve srovnání s cisplatinou, kdy nižší protinádorovou aktivitu na úrovni in vivo převyšovala absence negativních vedlejších účinků spojená s aplikací zmíněných nových komplexů do těla léčených myší. V lyzátech nádorových tkání byly stanoveny vybrané proteiny, jmenovitě kaspáza 8 (Casp-8), kaspáza 3 (Casp-3) a p53. Je známo (viz výše), že vznik aduktů terapeutických komplexů na bázi platiny a DNA (Štarha et al., J. Inorg. Biochem., 2012 (příloha β) a Muchová et al., J. Biol. Inorg. Chem., 2013 (příloha γ) v případě zde studovaných komplexů s deriváty 7-azaindolu) vede k výrazným změnám ve struktuře DNA, které v konečném důsledku vedou k pozastavení buněčného cyklu, resp. k apoptóze. Poškození DNA bývají, mimo jiných mechanismů, rozpoznávány také transkripčním faktorem p53, který v takových případech spouští transkripci genu pβ1 a tím děje zastavující buněčný cyklus, což může být v případě přetrvávajícího poškození DNA následováno ději apoptotickými. Logicky s uvedeným tvrzením pak zní konstatování, že většina nádorových buněk má zmutovaný gen pro p5γ, a naopak že u efektivně léčených buněk jsou hodnoty p5γ vyšší. Na procesu apoptózy se pak aktivně podílí další z proteinů stanovovaných v diskutované práci Štarha et al., PLoS ONE, 2014 (příloha 4), kterými byly kaspázy γ a 8 (v obecném přiblížení řečeno, vyšší hodnoty těchto proteinů jsou důsledkem efektivního cytotoxického účinku testovaných látek). Z výsledků experimentů popsaných v diskutované práci vyplynuly rozdíly mezi testovanými komplexy cis-[PtCl2(3Claza)2] a cis-[PtCl2(3Iaza)2] ve srovnání s komplexem cis-[PtCl2(5Braza)2] (viz příloha 4, obr. 7 na str. 10). V případě kaspázy γ byl 28
komplex obsahující ve své struktuře 5-brom-7-azaindol tím, který jako jediný výrazně navyšoval hladinu tohoto proteinu (hladinu kaspázy γ nepatrně navyšovaly ještě komplex cis-[PtCl2(3Iaza)2] a cisplatina). Obdobně také v případě kaspázy 8 a p53 byla pouze u komplexu s 5Braza (a cisplatiny) zjištěna vyšší hladina zmíněných proteinů. Společně s výše uvedeným byl stanoven také mitotický index (MI; viz příloha 4, tabulka γ na str. λ), jehož nízké hodnoty indikují cytotoxický účinek studovaných komplexů na nádorovou tkáň (regresi tumoru) - také z hodnot MI je patrný rozdíl mezi cis-[PtCl2(5Braza)2] a jeho dvěma analogy s odlišně substituovanými deriváty 7-azaindolu. Dalším z hodnotících parametrů, který v sobě zahrnuje většinu z výše uvedených výsledků, je tzv. index reaktivity. Je dán semikvantitativním vyhodnocením výsledků provedených histologických a histochemických metod (viz příloha 4, tabulka γ na str. λ) a lze jej brát jako parametr vypovídající o míře reakce nádorové tkáně na chemoterapeutický účinek aplikovaných látek nebo o stupni regrese tumoru. V případě cisplatiny je její nízký index reaktivity (1,γ5) dán jejím výrazným cytotoxickým účinkem (viz úbytek nádorové tkáně), který je spojený s nízkou mitotickou aktivitou a naopak výraznou nekrózou a krvácením. Z pohledu nepříliš výrazné in vivo protinádorové aktivity (ve smyslu redukce hmotnosti nádorové tkáně a průměrné doby přežití) byla poněkud překvapivá hodnota indexu reaktivity (1,71) získána u komplexu cis-[PtCl2(5Braza)2]. Tato sice nekoreluje, jako v případě výše zmíněné cisplatiny, s úbytkem hmotnosti nádorové tkáně, i tak je ale nutné ji brát jako hodnotu perspektivní, protože nám indikuje regresi tumoru nezávisle od jejího pomalejšího nástupu u zmíněného komplexu ve srovnání s cisplatinou. Hodnoty indexu reaktivity pro další dva testované komplexy s deriváty 7-azaindolu jsou výrazně vyšší než u zmíněného komplexu s 5Braza. Z mnoha zásadních poznatků práce Štarha et al., PLoS ONE, 2014 (příloha 4) lze rozhodně vyzdvihnout fakt, že komplex cis-[PtCl2(5Braza)2] obstál ve studiu protinádorové aktivity na úrovni in vivo provedeném na myším modelu leukémie L1β10, což jej favorizuje do dalších stádií základního farmakologického testování (metabolismus látky a lékové interakce na úrovni enzymů P450, farmakokinetika atd.) s výhledem na následný preklinický výzkum této látky. 3.1.2. Karboxylato komplexy s deriváty 7-azaindolu Je zajímavým faktem, že většina komplexů platiny, které v návaznosti na cisplatinu vstoupily do fáze klinických testů, obsahuje ve své molekule jeden bidentátní nebo dva monodentátní karboxylato ligandy (např. karboplatina, oxaliplatina, heptaplatina atd.; viz obrázek 1) [85]. Různorodost těchto O-donorových ligandů a možnost jejich derivatizace 29
umožňují bioanorganickým chemikům vhodně modifikovat farmakologicky důležité vlastnosti výsledných komplexů platiny (např. hydrolytická stabilita nebo lipofilita). Je také nutné zdůraznit, že karboxylato ligandy obsažené v klinicky testovaných nebo používaných léčivech na bázi platiny představují vhodné a mnoha studiemi prověřené odstupující skupiny. Z toho důvodu byly některé z nich vybrány pro přípravu nových komplexů platiny s deriváty 7-azaindolu, jinak řečeno k modifikaci chemických a farmakologických vlastností zjištěných u základní série cis-dichloroplatnatých komplexů s uvedenými N-donorovými ligandy. 3.1.2.1. Oxalatoplatnaté komplexy s deriváty 7-azaindolu Studium cis-dichloroplatnatých komplexů se 7-azaindolem a jeho halogenderiváty provázelo v pracích Štarha et al., Polyhedron, 2012 (příloha 1) a Štarha et al., J. Inorg. Biochem., 2012 (příloha β) také studium oxalatoplatnatých komplexů s analogickými N-donorovými ligandy. Syntéza platnatých oxalato komplexů s deriváty 7-azaindolu je, stejně jako v případě výše diskutovaných dichloro komplexů, atraktivní svojí jednoduchostí a rychlostí - i tyto látky totiž lze získat v jednom reakčním kroku. Uvedené je zajímavé, přihlédneme-li k výrobě oxaliplatiny, od které opět zde diskutované oxalatoplatnaté komplexy s deriváty 7-azaindolu formálně odvozujeme. Její výroba probíhá podle výše zmíněné Dharovy metody [83], čili reakcí 1R,2R-diaminocyklohexan-dichloroplatnatého (nebo jiného dihalogeno) komplexu s alespoň dvěma molárními ekvivalenty stříbrné solí (např. AgNO3) za vzniku komplexního diaquaplatnatého kationtu. Přidáním kyseliny šťavelové nebo její alkalické soli pak dostáváme výslednou oxaliplatinu. Příprava oxalatoplatnatého komplexu se 7-azaindolem je popsána v práci Štarha et al., Polyhedron 2012 (příloha 1). Komplex byl připraven v jednom reakčním kroku syntézou vycházející z K2[Pt(ox)2]∙βH2O (tento je komerčně dostupný nebo jednoduše připravitelný reakcí K2[PtCl4] s nadbytkem alkalického šťavelanu, např. K2(ox)∙H2O [86]) interagujícím se stechiometrickým množstvím 7-azaindolu (aza) za vzniku požadovaného produktu, [Pt(ox)(aza)2]. Stojí za zmínku, že jednokroková syntéza oxalatoplatnatých komplexu vycházející z alkalického bis(oxalato)platnatanu byla v literatuře poprvé popsána autorem této práce [87] a je jedinou známou alternativou výše popsané Dharovy metody přípravy oxalatoplatnatých komplexů. Výhody jsou opět zřejmé, můžeme zde zopakovat výše uvedené pro dichloro komplexy - jeden reakční krok znamená kratší reakční čas a vyšší výtěžek, není také nutné řešit kontaminaci produktu stříbrem, jak je známo z průmyslové výroby zde zmíněné oxaliplatiny [např. 88]. Ale zpět ke zmíněnému komplexu [Pt(ox)(aza)2], jenž byl, stejně jako jeho dichloro analog, prokázán jako in vitro cytotoxicky neaktivní do koncentrace dané jeho rozpustností 30
v testovacím médiu (IC50 > 0,1 µM vůči MCF7 a HOS), jak je popsané v práci Štarha et al., Polyhedron, 2012 (příloha 1, str. 407). V navazující práci Štarha et al., J. Inorg. Biochem., 2012 (příloha β) jsou, vedle výše diskutovaných cis-dichloroplatnatých komplexů, popsány také tři komplexy oxalatoplatnaté s 3Claza, 3Iaza a 5Braza. Také tyto komplexy byly připraveny
zmíněnou
jednokrokovou
syntézou
vycházející
z bis(oxalato)platnatanu
draselného. U těchto komplexů byla studována jejich in vitro cytotoxicita vůči lidským nádorovým buněčným liniím MCF7, HOS a LNCaP, avšak závěr z těchto biologických experimentů je obdobný jako pro [Pt(ox)(aza)2], tedy že jsou necytotoxické do koncentrace dané jejich rozpustností v použitém médiu (viz příloha β, tabulka β na str. 61). Na tato zjištění bylo navázáno prací P. Štarha et al., Molecules, 2014, 10832 (příloha 5), která si kladla za cíl studium vlivu zavedení odlišně substituovaných derivátů 7-azaindolu do molekuly oxalatoplatnatého komplexu na in vitro protinádorovou aktivitu. Do série ke čtyřem výše uvedeným komplexům s aza, 3Claza, 3Iaza a 5Braza tak přibyl také komplex s bromem v poloze 3 7-azaindolového kruhu (3Braza) a navíc komplexy se substituenty v poloze 4 (4Claza a 4Braza). Všechny tři nově připravené komplexy byly testovány na jejich in vitro cytotoxicitu vůči lidským nádorovým buněčným liniím HOS a MCF7. Stejně jako v předchozích případech nebyla ani u komplexů [Pt(ox)(4Claza)2] a [Pt(ox)(4Braza)2] zjištěna žádná aktivita až do koncentrace dané jejich rozpustností (IC50 > 1,0 M). Ovšem v případě komplexu [Pt(ox)(3Braza)2], který byl v testovacím médiu velmi dobře rozpustný (>50,0 M), byla na zmíněných dvou liniích zjištěna zajímavá in vitro cytotoxicita srovnatelná s cisplatinou a vyšší než u oxaliplatiny (viz příloha 5, tabulka 1 na str. 10838 a obr. 3 na str. 10839). Na základě těchto výsledků byl tento komplex testován na šesti dalších lidských nádorových buněčných liniích, kterými byly Aβ780, Aβ780R, Gγ61, HeLa, A54λ a LNCaP. Z výsledků (viz příloha 5, tabulka 1 na str. 10838 a obr. 3 na str. 10839) plyne, že [Pt(ox)(3Braza)2] je na liniích Aβ780 a HeLa více in vitro cytotoxický než cisplatina a na liniích Aβ780, HeLa a Gγ61 více než oxaliplatina. Na liniích Aβ780R, A54λ a LNCaP nebyla zjištěna žádná aktivita až do limitní testované koncentrace (IC50 > 50,0 M). U in vitro protinádorově aktivního komplexu [Pt(ox)(3Braza)2] bylo studováno 1
( Ha
195
Pt NMR, ESI hmotnostní spektrometrie) jeho chování ve vodu obsahujících
rozpouštědlech (směs DMF-d7/H2O, 9:1 v/v; směs voda/methanol, 1:1 v/v) a schopnost interagovat s vybranými biomolekulami (cystein, GSH, GMP). Komplex byl ve směsi DMF-d7/H2O stabilní a nepodléhal hydrolýze - NMR spektra neobsahovala ani po pěti dnech žádné nové signály (viz příloha 5, obr. 1 na str. 108γ6). V souladu s tímto zjištěním není překvapivé, že ani hmotnostní spektra nevykazovala v čase žádné změny spojitelné 31
se vznikem nových platinu obsahujících částic (viz příloha 5, obr. β na str. 108γ7). Co se týká schopnosti studovaného komplexu interagovat s uvedenými biomolekulami, která byla studována ESI hmotnostní spektrometrií, tak ani zde nebyla nedetekována žádná nově vzniklá platinu obsahující částice, která by byla prokazatelným produktem interakce některé z biomolekul se studovaným komplexem. Jedinou výjimkou je pík velmi nízké intenzity odpovídající cystein (cys) obsahující částici {[Pt(cys)(ox)(3Braza)2]+H}+, obsahující pravděpodobně monodentátně koordinovaný oxalátový aniont jako důsledek rozštěpení jedné Pt–O vazby a otevření pětičetného PtO2C2 kruhu (viz příloha 5, obr. β na str. 108γ7). 3.1.2.2. Fototoxické deriváty karboplatiny s deriváty 7-azaindolu Další ze studovaných typů platnatých komplexů,
Pt(cbdc)(naza)2]
(cbdc
=
cyklobutan-1,1-dikarboxylátový
aniont; naza = 3Claza, 3Braza, 3Iaza, 4Claza, 4Braza a 5Braza), lze odvodit od karboplatiny nahrazením obou NH3 molekul dvěma deriváty 7-azaindolu (obrázek 14). Tyto komplexy byly připraveny podle výše zmíněné Dharovy metody, s klíčovým meziproduktem o složení cis-PtI2(naza)2]. Práce popisující zmíněné komplexy a jejich in vitro cytotoxicitu a fototoxicitu byla 16. 3. 2015 akceptována k publikaci v časopise PLoS One
Obrázek 14. Strukturní vzorec studovaných derivátů karboplatiny
(manuskript je k habilitační práci přiložen jako příloha 6). Zde si dovolím malou odbočku, jelikož i u zmíněných dijodo komplexů byla studována jejich in vitro cytotoxicita vůči lidským nádorovým buněčným liniím Aβ780, A2780R, MCF7, HOS, A549, G361, HeLa a karcinomu prostaty 22Rv1 (tabulka 2). Poněkud Tabulka 2: In vitro cytotoxicita cis-[PtI2(naza)2] komplexů vůči lidským nádorovým buněčným liniím karcinomu vaječníku Aβ780, cisplatin-resistentního karcinomu vaječníku Aβ780R, prsního karcinomu MCF7, osteosarkomu HOS, karcinomu plic A54λ, maligního melanomu Gγ61, karcinomu děložního čípku HeLa a karcinomu prostaty 22Rv1 a jejich srovnání s cisplatinou. Výsledky jsou dány jako hodnoty IC50 ( M). 2Me4Claza = 2-methyl-4-chloro-7-azaindol A2780 A2780R MCF7 HOS A549 G361 HeLa 22Rv1 cis-[PtI2(aza)2] γ,5±0,7* γ,γ±0,γ* 1,7±0,8* 0,8±0,4* 1β,γ±1,1 β,λ±0,6* 7,0±0,6* 4,6±1,β* cis-[PtI2(3Claza)2] 4,1±0,8* γ,γ±0,5* 1,5±0,4* 1,γ±0,8* 6,4±1,4 β,γ±1,0* 4,8±0,4* 4,8±β,4* cis-[PtI2(3Iaza)2] β,8±0,γ* γ,β±0,β* 1,λ±0,5* β,β±1,2* 8,8±β,β γ,1±0,β* 4,7±0,8* 4,β±1,β* cis-[PtI2(3Braza)2] β,γ±1,1* β,6±0,8* 1,8±0,γ* β,8±1,0* λ,8±1,γ 1,6±0,7* 6,β±0,7* 4,5±1,6* cis-[PtI2(4Claza)2] γ,7±1,1* γ,4±0,β* 1,5±0,5* 0,5±0,β* 4,7±0,γ γ,β±0,β* γ,8±0,γ* 4,β±0,1* cis-[PtI2(4Braza)2] γ,7±1,1* 3,3±0,4* 1,0±0,4* 0,4±0,1* 4,γ±0,λ γ,β±0,β* γ,8±0,β* γ,8±0,1* cis-[PtI2(5Braza)2] γ,4±0,β* γ,5±0,5* 1,6±0,8* 1,4±1,1* 7,γ±1,6 γ,4±0,γ* 5,4±1,β* 5,1±0,8* cis-[PtI2(2Me4Claza)2] 1,7±1,1* 1,0±0,4* β,1±1,0* 0,7±0,β* γ,5±0,β 1,7±1,γ* γ,8±0,1* γ,5±0,β* Cisplatina β1,6±0,6 β0,β±4,7 17,λ±γ,5 18,λ±1,7 >50,0 5,γ±0,7 γ0,4±11,0 β6,λ±γ,5
32
překvapivě, vzhledem k faktu, že dijodoplatnaté komplexy byly už v prvních letech po objevu cisplatiny postulovány jako její necytotoxické a tedy farmakologicky neperspektivní analogy [89] a že není v literatuře popsáno příliš dijodoplatnatých komplexů vykazujících výraznou in vitro cytotoxicitu [např. 90–94], byla u dijodo komplexů se 7-azaindoly zjištěna výrazná in vitro cytotoxicita, která je mnohonásobně a statisticky významně (p < 0,05) vyšší ve srovnání s cisplatinou a ve většině případů i s dichloro analogy uvedených dijodo komplexů. Farmakologická perspektiva se odrazila v podání přihlášky vynálezu, jež chrání využití těchto látek jako léčiv v protinádorové terapii (PV β014-275, 04/2014). Bylo také zjištěno, že dijodo komplexy pravděpodobně působí jiným mechanismem, než jaký je prokázán pro jejich dichloro analogy (viz výše), protože v experimentu na Aβ780 buňkách má
cis-[PtI2(4Braza)2]
odlišnou
(vyšší)
IC50
(tj.
nižší
cytotoxicitu)
než
směs
cis-[PtCl2(4Braza)2] a dvou ekvivalentů KI. Navíc byla aktivita zmíněného dijodo komplexu modifikována (navýšena) přídavkem 5,0 µM L-BSO, a to konkrétně z hodnoty IC50 = γ,7 µM na IC50 = 1,0 µM. Na základě tohoto experimentu samozřejmě nelze rozhodnout, zdali je to dáno neochotou studovaných dijodoplatnatých komplexů podléhat inaktivačním reakcím s glutathionem (jehož fyziologickou syntézu L-BSO inhibuje), nebo významným ovlivněním redoxních dějů v buňce (glutathion je znám svojí schopností regulovat vnitrobuněčnou hladinu volných radikálů zasahujících do redoxních dějů) [11,95,96]. Toto vyžaduje další studium, jehož výsledky budou součástí připravované publikace. Vrátíme-li se ke zmíněným derivátům karboplatiny, lze konstatovat, že z výsledků in vitro cytotoxicity je patrná poměrně výrazná aktivita na A2780 ovariálním karcinomu (až 4,6× vyšší než u cisplatiny), nižší cytotoxicita pak byla zjištěna v případě lidských Tabulka 3. Výsledky testování in vitro cytotoxicity derivátů karboplatiny s deriváty 7-azaindolu a standardů karboplatiny a cisplatiny vůči lidským nádorovým buněčným liniím cisplatina-senzitivním karcinomu vaječníku A2780 a karcinomů prostaty LNCaP a PC-3. Buňky interagovaly s testovanými látkami 24 h. Data jsou uvedena jako hodnoty IC50 ± SD (μM). Komplex A2780 LNCaP PC-3 [Pt(cbdc)(3Claza)2] 11,8±6,2* 30,8±3,6 36,3±2,3 [Pt(cbdc)(3Iaza)2] 10,3±4,3* >20,0a 18,5±0,9 [Pt(cbdc)(3Braza)2] 14,4±6,0 >50,0a 42,3±0,8 [Pt(cbdc)(4Claza)2] 5,3±0,9* 18,7±5,1 17,6±8,8 [Pt(cbdc)(4Braza)2] 5,1±0,9* 23,5±3,8 26,6±4,1 [Pt(cbdc)(5Braza)2] 4,7±1,9* 22,1±1,4 29,6±9,4 Karboplatina >50,0a >50,0a >50,0a Cisplatina 21,8±3,9 >50,0a >50,0a * značí statisticky významný rozdíl hodnot od standardu cisplatiny na p < 0,05 hladině významnosti a) IC50 nebylo dosaženo do uvedené koncentrace
33
nádorových buněčných linií karcinomu prostaty LNCaP a PC-3 (tabulka 3). U komplexu s 3Braza bylo dále experimentálně stanoveno, že po přidání 5,0 µM L-BSO statisticky významně (p < 0,05) vzrostla (ca 4,4×) cytotoxicita na IC50 = γ,γ±0,γ µM. S ohledem na fakt, že pro karboplatinu, od které jsou zde studované komplexy odvozeny, je známo, že její in vitro cytotoxicita může být efektivně navýšena aktivací UVA zářením [97], rozhodli jsme se i pro její, zde popsané deriváty se 7-azaindoly tuto problematiku prostudovat. Studium bylo zaměřeno na vliv UVA záření na složení reprezentativního komplexu a na změny biologických vlastností spojených s ozářením
komplexů. Jak je patrné z obrázku 15, viabilita buněk byla výrazně nižší (p 0,05; ANOVA), pokud byly testované buňky Aβ780 a LNCaP po akumulaci komplexu ozářeny UVA zářením (20 min, 365 nm). V souladu s výše uvedeným vlivem UVA záření na viabilitu nádorových buněk se studované komplexy ve výrazně vyšší míře vázaly na CT DNA, jinými slovy, míra platinace DNA po UVA ozáření byla ve srovnání s neozářenými vzorky ca β–γ× vyšší.
Obrázek 15. In vitro viabilita lidských nádorových buněk cisplatina-senzitivního karcinomu vaječníku A2780 a karcinomu prostaty LNCaP pro komplexy [Pt(cbdc)(3Iaza)2] (A), [Pt(cbdc)(3Braza)2] (B), [Pt(cbdc)(4Claza)2] (C) a [Pt(cbdc)(4Braza)2] (D) aplikovaných v 10 µM koncentraci ve tmě a po ozáření UVA zářením. * značí statisticky významný rozdíl hodnot na ozářených a neozářených buňkách (p < 0,05)
Vzhledem k tomu, že biologické experimenty poskytly odlišné výsledky pro UVA ozářené a neozářené komplexy, bylo nutné pokusit se zjistit, jaký vliv má zmíněné záření na složení komplexů. 1H NMR studiem bylo zjištěno, že komplexy (bez UVA ozáření) jsou v DMF-d7 stabilní po dobu minimálně 14 dní, během kterých nebyly v protonových spektrech detekovány žádné změny. U reprezentativního komplexu [Pt(cbdc)(4Braza)2] byla ve směsném rozpouštědle DMF-d7/H2O (1:1, v/v) po 24 h (bez UVA ozáření) detekována nová sada signálů (např. N1–H signál při 11,94 ppm, obrázek 16) odpovídající monofunkcionální částici o složení [Pt(4Braza)2(cbdc´)(H2O)] (cbdc´ = monodentátně koordinovaný cbdc) [42–44]. Ve spektrech komplexů (rozpuštěných také ve směsi
DMF-d7/H2O, 1:1, v/v) po působení UVA záření (20 min; max = 365 nm, 4,3 mW cm–2) byly 34
detekovány tři N1–H signály 4Braza (1γ,0β, 1β,γ5 a 11,8β ppm) ve srovnání s jedním signálem ve spektru původního komplexu (13,02 ppm) (obrázek 16). Po β4 h stání při laboratorní teplotě a laboratorním osvětlení nebyly detekovány žádné změny ve zmíněném protonovém NMR spektru.
Obrázek 16. 1H NMR spektra (N1–H oblast ligandu 4Braza) zaznamenaná v čase (A a C - čerstvé roztoky; B a D - po β4 h stání za laboratorní teploty a laboratorního osvětlení) pro vzorky před (A a B) a po (C a D) UVA iradiaci (β0 min, max = 365 nm) pro komplex [Pt(cbdc)(4Braza)2] ve směsi DMF-d7/H2O (1:1, v/v).
Obrázek 17. 1H NMR spektra (N1–H oblast ligandu 4Braza) zaznamenaná v čase (po 24 (A) nebo 96 h (B a C) stání za laboratorní teploty a laboratorního osvětlení) pro vzorky po UVA iradiaci (β0 min, max = 365 nm) pro komplex [Pt(cbdc)(4Braza)2] ve směsi DMF-d7/H2O (1:1, v/v) (A a B) nebo pro tento komplex s přídavkem roztoku volného 4Braza (C)
Následné 1H NMR experimenty (tzv. spikování volným 4Braza) prokázaly, že jeden (11,8β ppm) ze zmíněných N1–H signálů náleží volnému 4Braza uvolněnému z původního komplexu (viz obrázek 17). Dále získané výsledky poukazovaly na to, že druhý nový N1–H signál (1β,γ5 ppm) odpovídá molekule 4Braza obsažené v nově vzniklé platinu obsahující částici o pravděpodobném složení cis-[Pt(H2O)2(cbdc´)(4Braza)]. U této látky se pak cbdc ligand vyskytuje s největší pravděpodobností jako monodentátní (značeno jako cbdc´), na což poukazuje nový kvintet (β,08 ppm) charakteristický pro CH2–CH2–CH2 vodíky tohoto ligandu (1,89 ppm v původním komplexu). Fakt, že dochází k rozštěpení jedné z Pt–O vazeb a tím k otevření šestičlenného PtO2C3 cyklu dokládá i zjištění, že signál detekovaný při 12,35 ppm přechází v čase v signály dva (1β,γ5 a 1β,18 ppm), což je pravděpodobně spojeno s izomerací uvedené částice. Uvolňování 4Braza ze struktury původního komplexu prokázalo také ESI-MS studium, protože ve spektru UVA ozářeného vzorku byla několikanásobně vyšší
35
intenzita píku částic {4Braza+H}+ a {[Pt(cbdc)(4Braza)]–H}–, než ve spektru vzorku před ozářením. Analogické
1
H NMR a ESI-MS experimenty byly provedeny také pro směs
studovaného komplexu s GSH nebo GMP v DMF-d7/H2O (1:1, v/v). V případě GSH nebyla uvedenými technikami prokázána žádná interakce. V případě směsi s GMP byly v protonovém NMR spektru po ozáření rozpoznány signály uvolněného 4Braza a navíc byl u původního C8–H signálu GMP (8,37 ppm) detekován signál nový při 8,γ1 ppm, což indikuje interakci studovaného komplexu s GMP. V hmotových spektrech pak byly nalezeny píky
{[Pt(cbdc)(4Braza)(GMP)]–Na+2H}+,
a {[Pt(cbdc)(4Braza)(GMP)]–2Na+H}–
při
{[Pt(cbdc)(4Braza)(GMP)]+H}+
920,2,
942,3
a 896,3 m/z, což domněnku o interakci komplexu s GMP potvrdilo. Transkripční mapování Pt–DNA aduktů, provedené pro
vybrané
komplexy
[Pt(cbdc)(3Iaza)2]
a [Pt(cbdc)(4Braza)2] metodou studia in vitro RNA syntézy T7 RNA polymerázou na pSP73KB/HpaI DNA fragmentu, prokázalo, že místa inhibice jsou pro tyto komplexy jak s UV ozářením tak i bez něj podobné cisplatině (tedy především GG nebo AG místa; obrázek 18). Nicméně, výrazný je rozdíl v efektivitě inhibice mezi ozářeným a neozářeným komplexem, konkrétně UVA ozářené komplexy inhibovaly, při stejné míře platinace DNA, RNA syntézu výrazně efektivněji než komplexy neozářené. Toto souvisí se známým rozdílem mezi schopností mono- a bifunkcionálních Pt–DNA aduktů inhibovat RNA syntézu [98–100]. Jinak řečeno, komplexy po ozáření UVA zářením se na DNA vážou jiným způsobem
(bifunkcionálně)
než
komplexy
Obrázek 18. Inhibice RNA syntézy T7 RNA polymerázou na pSP73KB/HpaI fragmentu modifikovaném komplexem [Pt(cbdc)(4Braza)2]
neozářené
(monofunkcionálně). Výsledky experimentů provedených na stejných komplexech a lineární pSP73KB/EcoRI DNA, které byly zaměřené na kvantifikování interstrand a intrastrand aduktů komplexů s DNA, pak prokázaly, že UVA ozářené komplexy vytvářely oba typy aduktů, zatímco neozářené komplexy vytvářely pouze intrastrand adukty. Toto velmi zajímavé zjištění týkající se odlišné povahy (mono- vs. bifunkcionální) vzniklých Pt–DNA aduktů v závislosti na tom, jestli byl platnatý komplex ozářen UVA zářením nebo ne, bylo podpořeno výsledky fluorescenčních experimentů provedených na 36
DNA s ethidium bromidem (EtBr). Je známo, že vznik bifunkcionálních aduktů (např. u cisplatiny) efektivně brání interkalaci EtBr do DNA, což je pozorováno jako pokles fluorescence [101–103]. Uvedené ale neplatí pro adukty monofunkcionální (např. chloro-diethylentriaminplatnatý komplex, dienplatin), které způsobují pouze nepatrný pokles intenzity fluorescence. Z výsledků získaných pro výše uvedené komplexy plyne stejný závěr, jako je uvedený pro modelové sloučeniny cisplatinu a dienplatinu. Jinými slovy, modifikace DNA UVA ozářenými komplexy vedla k výraznému poklesu u cisplatiny,
fluorescence zatímco
podobnému
jako
modifikace
DNA
neozářenými komplexy způsobila ne tak výrazný pokles intenzity pozorované fluorescence (obrázek
Obrázek 19. A. Vznik interstrand (ICL) aduktů pro UVA ozářený (linie 1–5) a neozářený (linie 6–10) komplex [Pt(cbdc)(4Braza)2]. B. Závislost EtBr fluorescence na rb pro DNA modifikovanou UVA ozářeným (■) a neozářeným (▲) komplexem [Pt(cbdc)(4Braza)2]. Data jsou uvedena jako průměr ± SD. Data pro cisplatinu (čárkovaně) a dienplatinu (tečkovaně) jsou pro srovnání vložena také.
19). Souhrnně řečeno, tato práce prokázala deriváty karboplatiny se 7-azaindoly jako in vitro cytotoxické látky, jejichž aktivita může být různými mechanismy (L-BSO, fotoaktivace) dále navyšována. Rozdíl mezi aktivními částicemi vzniklými z původních komplexů za normálních podmínek a po ozáření UVA zářením byl ve výborné shodě prokázán chemickými i biologickými metodami. Velmi zajímavé je také zjištění, že studované komplexy uvolňují pod vlivem UVA záření N-donorový ligand. Toto by mohlo být velice zajímavé z biologického a farmakologického pohledu, protože pokud by takový uvolňující se N-donorový ligand byl sám o sobě určitým způsobem protinádorově aktivní, synergicky by tím přispíval k aktivitě platinu obsahující částice s perspektivou dosažení vyššího biologického efektu. 3.1.2.3. Selektivní malonato a dekanoato komplexy U aktuálně studovaných malonato (mal) a dekanoato (dec) komplexů platiny obecného složení [Pt(mal)(naza)2] a cis-[Pt(dec)2(naza)2] (obrázek 20) byla prozatím testována jejich in vitro cytotoxicita vůči cisplatin senzitivní (Aβ780) a rezistentní (Aβ780R) lidské nádorové linii karcinomu vaječníku, linii normálního lidského fibroblastu (MRC5) a primární kultuře lidských hepatocytů (Hep). Mezi těmito komplexy výrazně vynikly látky [Pt(mal)(3Iaza)2] 37
Obrázek 20. Strukturní vzorce studovaných malonato (nahoře) a dekanoato (dole) komplexů s deriváty 7-azaindolu.
(IC50 = 26,6±8,λ µM vůči A2780 a 28,λ±6,7 µM vůči A2780R) a cis-[Pt(dec)2(3Braza)2] (14,5±0,6 µM vůči Aβ780 a 14,5±γ,8 µM vůči A2780R), nikoli uvedenými hodnotami in vitro cytotoxicity (tabulka 4), ale především svojí selektivitou, tedy schopností působit na nádorové buňky a současně nevykazovat žádný biologický účinek na buňkách zdravých. V konkrétním přiblížení byly tyto komplexy na nenádorových buněčných liniích neaktivní až do nejvyšší testované koncentrace (vůči MRC5: IC50 > 50,0 µM pro [Pt(mal)(3Iaza)2] a IC50 > 25,0 µM pro cis-[Pt(dec)2(3Braza)2]; vůči Hep: IC50 > 250,0 µM pro [Pt(mal)(3Iaza)2] i cis-[Pt(dec)2(3Braza)2]). Tabulka 4. Výsledky testování in vitro cytotoxicity malonato a dekanoato komplexů s deriváty 7-azaindolu a cisplatiny vůči lidským nádorovým buněčným liniím Aβ780 a Aβ780R, nenádorové linii MRC5 a primární kultuře lidských hepatocytů (Hep). Buňky interagovaly s testovanými látkami β4 h. Data jsou uvedena jako hodnoty IC50 ± SD ( M). Komplex A2780 A2780R MRC5 Hep [Pt(mal)(3Braza)2] >50,0a β7,γ±8,λ >50,0a 32,8 [Pt(mal)(3Iaza)2] β6,6±8,λ β8,λ±6,7 >50,0a >250,0a [Pt(mal)(4Braza)2] β4,4±1,β ββ,4±1,8 β1,λ±γ,1 nt [Pt(dec)(3Braza)2] 14,5±0,6* 14,5±γ,8 >25,0a >250,0a a [Pt(dec)(3Iaza)2] 1γ,0±1,1* 15,8±4,β >25,0 nt a a [Pt(dec)(4Claza)2] >50,0 β4,6±1,β >50,0 105,1 [Pt(dec)(4Braza)2] 14,γ±γ,7* 1λ,5±β,1 β6,β±4,8 nt Cisplatina β6,γ±γ,6 >50,0a >50,0a >250,0a * značí statisticky významný rozdíl hodnot od standardu cisplatiny na p < 0,05 hladině významnosti nt, nebylo testováno a) IC50 nebylo dosaženo do uvedené koncentrace
3.1.4. Možnosti cíleného transportu komplexů na bázi platiny Jak je uvedeno výše v práci, jednou z možností, jak navýšit aktivitu metaloterapeutik a/nebo snížit jejich negativní vedlejší účinky, je jejich cílený transport za využití různých typů nanočástic jako transportních systémů. Jednou z nejperspektivnějších oblastí je pak studium funkcionalizovaných magnetických nanočástic na bázi oxidů železa pro magnetický transport 38
léčiv. Do této oblasti zapadá i práce Štarha et al., Molecules, 2014, 1622 (příloha 7), ve které je popsána příprava a charakterizace maghemitových core-shell nanočástic obalených 4-aminobenzoovou kyselinou, která váže biologicky aktivní částici, cis-[Pt(H2O)2(naza)]2+. Studované nanočástice byly připravené srážecí metodou ze směsi FeCl3∙6H2O, FeCl2∙4H2O a zmíněné organické karboxylové kyseliny a následným přídavkem vodného roztoku NH3. Syntéza nebyla provedena v inertní atmosféře, ale v atmosféře oxidační (vzduch), která zajistila oxidaci vznikajícího Fe3O4 (magnetit) na -Fe2O3 (maghemit). Vzniklé nanočástice následně interagovaly s aktivovaným platnatým komplexem (vznikl aktivací stříbrnou solí z původních cis-dichloroplatnatých komplexů s deriváty 7-azaindolu 3Claza a 5Braza) za vzniku výsledné funkcionalizované nanočástice (viz příloha 7, obr. 1, str. 1624). Přítomnost obalující organické kyseliny i funkcionalizujícího komplexu byla potvrzena FTIR spektroskopií a TG/DTA termickou analýzou (viz příloha 7, obr. 2 na str. 1625). Platina pak byla jednoznačně detekovaná v EDS spektrech výsledných systémů. U vzniklých nanokompozitů předpokládáme vazbu Pt–N mezi funkcionalizujícím komplexem a obalující karboxylovou kyselinou, což bylo nepřímo potvrzeno detailním FTIR studiem. Připravené nanočástice mají sférický tvar s průměrnou velikostí 1γ,0±β,1 nm, jak je patrné z TEM snímků
(viz
příloha
7,
obr.
3
na
str.
1627).
Makroskopicky
pozorovatelné
superparamagnetické vlastnosti (viz příloha 7, obr. 1 na str. 1624) byly u studovaných nanočástic experimentálně prokázány detailním studiem Mössbauerovou spektroskopií (viz příloha 7, tabulka 1 a obr. 4 na str. 1628). V diskutované práci nejsou popsány biologické vlastnosti (např. stabilita v roztoku, uvolňování léčiva v různých podmínkách, toxicita a cytotoxicita výchozích i výsledných funkcionalizovaných systémů atd.) uvedených nanosystémů. Výzkum uvedeného nyní probíhá a jeho výsledky budou popsány a diskutovány v navazující publikaci. Na práci Štarha et al., Molecules, 2014, 1622 (příloha 7) bylo navázáno studiem odvozených
typů
nanočástic,
jejichž
základem
je
kompozit
zlata
navázaného
na maghemitovém jádře, u kterých je na zmíněné zlato navázána síru obsahující karboxylová kyselina
(např.
kyselina
lipoová
nebo
merkaptopropionová),
která
je
pak
dále
funkcionalizovaná (Štarha et al., Int. J. Mol. Sci., 2015 (příloha 8)). V případě naší aktuální práce byly nanočástice s kyselinou lipoovou funcionalizovány aktivovanou cisplatinou (příprava
a charakterizace
analogických
systémů
s platnatými
komplexy
s deriváty
7-azaindolu v současné době probíhá) (obrázek β1). Výhody kompozitu zlato/maghemit oproti výše zmíněným maghemitovým nanočásticím jsou takové, že kombinují vlastnosti obou komponent, tady maghemitu (superparamagnetismus, potenciál při MRI zobrazování) 39
a zlata (chemická odolnost, fototermální [104].
efekt)
Zlato
navíc
chrání
maghemitové jádro před chemickou a enzymatickou
degradací
ve
fyziologickém prostředí a je vhodné pro chemickou (např. zmíněnými
derivatizaci síru
obsahujícími
molekulami) [105,106]. Z literatury jsou známé
podobné
systémy
zlatem
obalených nanočástic na bázi oxidu železa a jejich rozmanité využití [např.
Obrázek 21. Předpokládané složení studovaných mag/Au– LA–CDDP* nanočástic (vlevo), část XPS jejich spektra (vpravo nahoře) a jejich TEM snímek (vpravo dole)
107–111]. Avšak je prozatím známa pouze jedna práce, která popisuje tyto nanočástice funkcionalizované modifikovanou cisplatinou nebo jinou biologicky aktivní sloučeninou na bázi platiny [112]. Naše práce přináší ve srovnání se zmíněnou publikací jednodušeji připravitelné systémy, které pak účinněji vážou aktivní komplex platiny (cisplatinu). Studované nanočástice jsou připraveny následujícím postupem. Ze směsi Fe(III) a Fe(II) solí (molární poměr βμ1) jsou srážecí metodou připraveny magnetitové (Fe3O4) nanočástice [113], které jsou následně oxidovány zředěnou kyselinou dusičnou na nanočástice maghemitové ( -Fe2O3), symbolizované dále jako mag [114]. Následné navázání zlata na maghemit (mag/Au nanočástice) je provedeno postupným přidáváním 1% HAuCl4 [112]. Dalším přídavkem je pak na zlato navázána kyselina lipoová (mag/Au–HLA) [115]. mag/Au– HLA nanočástice jsou v případě připravované práce funkcionalizovány aktivovanou cisplatinou, cis-[Pt(NH3)2(H2O)2]2+ (CDDP*), čímž je získán finální produkt mag/Au–LA– CDDP*. Připravené mag/Au–LA–CDDP* nanočástice (pro srovnání a jednoznačnější interpretaci také syntetické meziprodukty) byly charakterizovány rozličnými technikami. Z HRTEM a TEM snímků je patrné, že se jedná o kulovité nanočástice dobré velikostní distribuce, jejichž průměrná velikost je 12,2±1,9 nm (viz příloha 8, obr. 2 na str. 2037). Přítomnost platiny (přeneseně aktivované cisplatiny) byla prokázána výsledky XPS a EDS spektroskopie a ICP-MS. V XPS spektru mag/Au–LA–CDDP* nanočástic (viz příloha 8, obr. 4 a 5 na str. 2039 resp. 2040) byly, kromě Fe2p, O1s, C1s, Au4d, Au4f, Au5p a S2p píků detekovaných také v XPS spektru meziproduktu mag/Au–HLA, nalezeny fotoelektronové píky jednoznačně přiřaditelné aktivované cisplatině (N1s, Pt4d a Pt4f). Stejné výsledky poskytla také zmíněná EDS spektroskopie, kterou byla také detekována platina. Velmi 40
důležitým poznatkem přispěla do aktuálního studia TG/DTA termická analýza, z jejichž výsledků lze usuzovat, že aktivovaná cisplatina se na mag/Au–HLA nanočástice váže jednou Pt–O vazbou. Obsah Pt v mag/Au–LA–CDDP* nanočásticích byl studován metodou ICP-MS. Molární poměr PtμAu je 1:2,6 a platina představuje 4,0% celkové hmotnosti studovaných nanočástic. Studium stability mag/Au–LA–CDDP* nanočástic prokázalo, že výrazně závisí na pH. V kyselém prostředí je uvolňování aktivované cisplatiny zanedbatelné. V neutrálním prostředí dosahuje hodnot dostačujících pro následné testování in vitro cytotoxicity nebo jiných biologických vlastností. Nejnižší stabilitu, tedy nejvyšší obsah uvolňované platinu obsahující částice, prokázaly studované nanočástice v zásaditém prostředí. Ačkoli dosavadní výsledky představují prozatím pouze chemickou část studované problematiky, jeví se diskutované nanosystémy jako slibné i pro následné biologické studium (in vitro a in vivo protinádorová aktivita, toxicita, MRI zobrazování atd.). Je zde také rozsáhlý prostor modifikace studovaných nanočástic ve smyslu jiné síru obsahující látky, jiné platinu obsahující funkcionalizující látky (např. zmíněné platnaté komplexy se 7-azaindoly) nebo třeba navázání specifických látek pro selektivní rozpoznání cílových nádorových buněk. Uvedená témata budou předmětem dalšího studia. 3.2. Platnaté komplexy s deriváty N6-benzyladeninu N6-benzyladenin (6-benzylaminopurin) a jeho deriváty jsou biologicky aktivní sloučeniny, které v závislosti na míře substituce purinového kruhu vykazují buď cytokininovou aktivitu (cytokininy jsou jednou ze základních skupin rostlinných hormonů) nebo aktivitu protinádorovou vycházející ze schopnosti derivátů vhodně substituovaných na purinovém kruhu inhibovat cyklin-dependentní kinázy (CDK) živočišných (včetně lidských) buněk. CDK jsou proteiny přirozeně regulující buněčný cyklus, jejichž inhibicí je tento děj pozastaven [116–118]. Nejznámějším CDK inhibitorem na bázi derivátů N6-benzyladeninu
je
2-(1-ethyl-2-hydroxyethylamino)-N6-benzyl-9-isopropyladenin
(roskovitin), který je pod jménem Seliciclib klinicky testován na pacientech trpících nemalobuněčným karcinomem plic [117,119]. Optikou koordinačního chemika jsou deriváty N6-benzyladeninu, tedy dusíkaté heterocyklické sloučeniny, vhodnými vícedentátními N-donorovými ligandy pro syntézu komplexů přechodných kovů. V literatuře jsou popsány komplexy Fe(II/III) [120], Co(II) [121], Ni(II) [122], Cu(II) [123], Zn(II) [124], Ru(III) [125], Pd(II) [126], Au(I) (viz práce Trávníček et al., J. Med. Chem., 2012, příloha 10), Au(III) [127], Pt(IV) [128] a v neposlední řadě také Pt(II) dichloro komplexy [129–134].
41
3.2.1. Protinádorová aktivita in vitro V rámci kontinuity výzkumu biologicky aktivních koordinačních sloučenin platiny na Katedře anorganické chemie Přírodovědecké fakulty Univerzity Palackého v Olomouci se v návaznosti na výše zmíněné dichloroplatnaté komplexy s deriváty N6-benzyladeninu uchazeč věnoval studiu biologicky aktivních komplexů platiny - tentokrát se jednalo o oxalato komplexy - v rámci své disertační práce [87,135,136], kde jsou také detailně okomentovány způsoby jejich přípravy a jejich protinádorová aktivita, která byla v mnoha případech výrazně vyšší
ve
srovnání
s cisplatinou.
Příprava
oxalatoplatnatých
komplexů
s deriváty
N6-benzyladeninu a jejich využití v protinádorové terapii je předmětem patentové ochrany národního patentu CZγ0β6βγ (udělen βγ. 6. β011) a mezinárodní přihlášky vynálezu WOβ010/1β1575 (publikováno β8. 10. β010). Jediná práce, která prozatím navázala na postgraduální studium oxalatoplatnatých komplexů s deriváty N6-benzyladeninu,
popisuje u série
čtyř
oxalato komplexů
s roskovitinem a jeho benzyl substituovanými deriváty jejich schopnost ovlivňovat lidský jaterní mikrosomální cytochrom P450 [137]. Studium aktivity devíti CYP enzymů zmíněného mikrosomálního systému (CYPs 1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, 3A4), prokázalo, že studované oxalato komplexy výrazně ovlivňují CYP3A4, a to na různých koncentračních úrovních. Toto pravděpodobně souvisí se schopností tohoto enzymu interagovat s objemnými molekulami. Vzhledem k tomu, že CYPγA4 je enzymem metabolizujícím většinu léčiv, je nutné uvedené výsledky brát jako nevýhodu ve smyslu jejich možného dalšího farmakologického využití. Oxalatoplatnaté
komplexy
jsou
popsány také v práci Štarha et al., Molecules, 2014, 3832. Tentokráte ovšem byly jako N-donorové ligandy použity jiné deriváty N6-benzyladeninu, a to benzyl substituované deriváty N6-benzyladenosinu (obrázek ββ). Obrázek 22. Strukturní vzorec studovaných oxalatoplatnatých komplexů s deriváty N6-benzyladenosinu.
Komplexy byly připraveny výše popsanou jednokrokovou
syntézou
z bis(oxalato)platnatanu
vycházející draselného
a charakterizovány širokou škálou fyzikálně-chemických technik včetně ESI hmotnostní spektrometrie a multinukleární (1H, prokázaly
složení
a
čistotu
13
C, 15N,
Pt) NMR spektroskopie (obrázek βγ), které
195
komplexů, 42
včetně
koordinace
použitých
derivátů
Obrázek 23. Výsledky 1H–15N gs-HMBC experimentů pro N6-(2-chlorobenzyl)adenosin (2ClL; vlevo) a komplex (vpravo) obsahující 2ClL jako N7-koordinovaný ligand, prokazující zmíněný typ koordinace přes atom N7 adeninového kruhu.
N6-benzyladenosinu přes dusíkový atom N7. Komplexy byly testovány na in vitro cytotoxicitu vůči dvěma lidským nádorovým buněčným liniím (HOS a MCF7), avšak nebyl u nich zjištěn žádný biologický efekt až do koncentrace dané jejich rozpustností v testovacím médiu. Příčinou cytotoxické inaktivity byla nízká (ca 50× nižší ve srovnání s cisplatinou) buněčná akumulace studovaných oxalatoplatnatých komplexů zjištěná na linii MCF7. S uvedeným souvisí také fakt, že, jak plyne z výsledků ESI-MS a 1H NMR, komplexy nepodléhají hydrolýze. Analogické N-donorové ligandy, tedy deriváty N6-benzyladenosinu substituované v tomto případě nejen na benzenovém kruhu, ale v některých případech i na purinovém skeletu, byly použity také v práci Štarha et al., Molecules, 2013 popisující a diskutující přípravu, charakterizaci a in vitro cytotoxicitu trans-dichloroplatnatých komplexů (syntéza cis-izomerů nevedla k izolování produktů o akceptovatelné izomerické čistotě) (obrázek 24). Cíle práce lze ztotožnit s těmi uvedenými výše pro práci Štarha et al., Molecules, 2014, 3832, nicméně ani v tomto případě nebyla u testovaných
komplexů
zjištěna
žádná
in vitro cytotoxicita vůči HOS a MCF7
43
Obrázek 24. Strukturní vzorec derivátů N6-benzyladenosinu použitých jako ligandy pro studované trans-dichloroplatnaté komplexy.
Obrázek 25. ESI– hmotnostní spektra studovaných komplexů 12 (a) a 7 (b) v methanolu (nahoře) a jejich směsí s L-methioninem měřené ihned (uprostřed) a 12 h (dole) po přípravě studované směsi.
lidským nádorovým buněčným liniím, a to až do maximální testované koncentrace (IC50 > 50,0 µM). Jedním z možných mechanismů odpovídajícím za inaktivitu v uvedeném biologickém experimentu je interakce s jednou z hlavních fyziologických síru obsahujících biomolekul L-methioninem, jak bylo v práci prokázáno ESI-MS studiem (viz obrázek 25).
44
Jiný typ Pt(II) komplexů s deriváty N6-benzyladeninu, odvozený od karboplatiny nahrazením obou NH3 molekul v její struktuře zmíněnými různě substituovanými deriváty N6-benzyladeninu (viz příloha 9, obr. 1 na str. 653), je popsán v práci Dvořák et al., Toxicol. in Vitro, 2011 (příloha 9), kde je detailně rozebrána in vitro cytotoxická aktivita sedmi Pt(II) komplexů vůči celkem sedmi lidským nádorovým buněčným liniím, včetně jejího srovnání s klinicky užívanými metaloterapeutiky cisplatinou, karboplatinou a oxaliplatinou. Výrazná aktivita byla pozorována na liniích karcinomu vaječníku Aβ780, karcinomu vaječníku rezistentního vůči cisplatině Aβ780R, prsního karcinomu MCF7 a maligního melanomu Gγ61, na kterých byla pro nejaktivnější komplex [Pt(cbdc)(4CF 3L)2] (4CF 3L = 2-chloro-N6-(4-trifluormethylbenzyl)-9-isopropyladeninu)
určena
aktivita
odpovídající hodnotám IC50 rovnajícím se 6,4±0,1, 5,6±1,7, 1λ,0±6,6, resp. 11,γ±β,8 µM (viz příloha 9, obr. 2–4 na str. 654 a 655). Zmíněné látky účinně překonávaly získanou rezistenci nádorových buněk vůči cisplatině (studováno na modelu karcinomu vaječníku A2780) - jejich rezistenční faktor se rovná 0,λ–1,4 (pro cisplatinu je to β,6). Nebyl také pozorován žádný vliv studovaných Pt(II) komplexů na zdravé lidské hepatocyty. Molekulární
farmakologií
cis-dichloroplatnatých
komplexů
s deriváty
N6-benzladeninu ze skupiny CDK inhibitorů a jejich syntetických prekurzorů (viz výše) se zabývají práce [134,138]. Ve druhé zmíněné, na které se uchazeč autorsky spolupodílel, je popsán a s cisplatinou srovnán mechanismus účinku potenciálního protinádorového léčiva o složení cis-[PtCl2(boh)2], kde boh symbolizuje CDK inhibitor ze skupiny derivátů N6-benzyladeninu bohemin, 2-(3-hydroxypropylamino)-N6-benzyl-9-isopropyladenin [129]. Bylo zjištěno, že se komplex efektivně váže na CT-DNA (t50% = 640±27 min), avšak ve významně menší míře v porovnání s cisplatinou (t50% = ca 120 min). Kvantifikace monofunkcionálních aduktů s DNA (experiment s [14C]thiomočovinou, inkubace 48 h) prokázala, že v případě studovaného komplexu je takových aduktů 10% (2% pro cisplatinu), jinak řečeno, λ0% ze vzniklých aduktů komplexu s boheminem je po 48 h vázáno bifunkcionálně. Studovaný komplex také účinně rozvolňoval supercoiled DNA, podobný efekt byl pozorován na oligodeoxyribonukleotidovém duplexu, se kterým, s ohledem na složení
použitého
oligodeoxyribonukleotidu,
komplex
s boheminem
vytváří
1,2-GG intrastrand adukt. Vliv studovaného komplexu na DNA byl zkoumán také chemickou metodou založenou na schopnosti některých agens (KMnO4, diethylpyrokarbonát) reagovat pouze s nukleobázemi jednovláknové nebo rozvolněné dvouvláknové DNA, nikoli s intaktní dvouvláknovou DNA. Také v tomto případě byl cis-dichloroplatnatý komplex s boheminem tím, který více rozvolňuje DNA, ve srovnání s cisplatinou. Z dosažených výsledků je patrné, 45
že komplex s boheminem vykazuje poněkud odlišnou aktivitu, z mechanistického úhlu pohledu, ve srovnání s cisplatinou. Proto byl studován, a s cisplatinou porovnán, vliv interakce komplexu s boheminem s DNA na rozpoznání vzniklých aduktů proteiny HMG (high-mobility group) domény. Je zajímavé, že ačkoli je vazba HMG proteinů na DNA poškozenou cisplatinou nebo jinými cytotoxickými léčivy na bázi platiny důležitým dějem zodpovídajícím za vlastní cytotoxický účinek, nebyla, na rozdíl od cisplatiny, pro komplex s boheminem provedeným experimentem detekována. Na druhou stranu to byl komplex s boheminem, který účinněji inhiboval polymerizaci DNA a transkripci DNA RNA polymerázou II ve srovnání s cisplatinou. U komplexu s boheminem byla dále zjištěna, v porovnání s cisplatinou, vyšší míra syntézy spojené s opravami DNA, jinými slovy účinnější oprava Pt–DNA aduktů vytvořených mezi komplexem s boheminem, než u aduktů s cisplatinou. UV-Vis elektronová spektroskopie pak byla použitá jako nástroj ke studiu interakce obou výše uvedených cis-dichloroplatnatých komplexů s glutathionem. Bylo zjištěno, že komplex s boheminem interaguje s uvedenou biomolekulou výrazně méně ve srovnání s cisplatinou. Závěrem lze k této práci říci, že komplex s CDK inhibitorem ze skupiny vícesubstituovaných derivátů N6-benzyladeninu boheminem a cisplatina mají některé mechanistické procesy společné, bez výrazných rozdílů (např. vazebná místa na DNA, rozvolnění DNA atd.), jiné procesy jsou odlišné, ať už v prospěch jednoho nebo druhého protinádorově aktivního platnatého komplexu. Potvrdit nebo vyvrátit farmakologický potenciál studované látky tedy musí další studie. 3.2.2. Protinádorová aktivita in vivo Perspektivní hodnoty in vitro cytotoxické aktivity výše zmíněných oxalatoplatnatých komplexů, které ve své struktuře obsahovaly deriváty N6-(substituovaný benzyl)-9isopropyladeninu různě substituované na atomu Cβ purinového skeletu, a to buď chlorem (tj. syntetické prekurzory roskovitinových derivátů [87]; např. zde studovaný komplex [Pt(ox)(2,4diOMeL)2],
kde
2,4diOMeL
=
2-chlor-N6-(2,4-dimethoxybenzyl)-9-
isopropyladeninu) nebo 2-aminobutan-1-olem (vlastní deriváty roskovitinu [135]; např. zde studovaný komplex [Pt(ox)(4OMeros)2], kde 4OMeros je derivát roskovitinu 2-(1-ethyl-2hydroxyethylamino)-N6-(4-methoxybenzyl)-9-isopropyladenin), byly základem pro vstup těchto látek do fáze in vivo testování, kdy byl pro studium vybrán model lymfocytární leukémie L1β10 na myších (připravovaná práce; metodika je popsána v práci Štarha et al., PLoS ONE, 2014, příloha 4).
46
Z výsledků plyne, že se in vivo biologický účinek obou látek výrazně liší a je přesně opačný, než tomu bylo ve fázi in vitro experimentů (pro příklad zde uveďme hodnoty IC50, na modelové lidské nádorové buněčné linii karcinomu vaječníku Aβ780, která je pro [Pt(ox)(2,4diOMeL)2] rovna
4,0±1,0
µM
a
pro
[Pt(ox)(4OMeros)2] je to 14,4±1,4 µM [135,136]). Za hlavní výsledek
Obrázek 26. Výsledky testování in vivo protinádorové aktivity vybraných oxalatoplatnatých komplexů s deriváty N6-benzyladeninu. 2,4diOMeL = 2-chlor-N6-(2,4dimethoxybenzyl)-9-isopropyladenin, 4OMeros = 2-(1-ethyl-2hydroxyethylamino)-N6-(4-methoxybenzyl)-9-isopropyladenin.
in vivo experimentu lze pokládat průměrnou délku prodloužení života ve srovnání s kontrolní (neléčenou) skupinou a cisplatinou. Cisplatina vyvolávala u testovaných zvířat výrazné nežádoucí účinky, které se následně projevovaly na celkovém zdravotním stavu, což vedlo, vedle snižování objemu nádorových tkání, k celkovému úbytku hmotnosti, apatii, nechutenství a nakonec i systémovému
selhání.
Obdobné
účinky,
ale
v menší
míře,
vyvolával
komplex
[Pt(ox)(2,4diOMeL)2], nejméně toxických projevů bylo pozorováno u [Pt(ox)(4OMeros)2]. Uvedené se u cisplatiny a komplexu [Pt(ox)(2,4diOMeL)2] projevilo ve snížení průměrné doby přežití testovaných myší ve srovnání s neléčenou kontrolou. Naopak aplikace komplexu [Pt(ox)(4OMeros)2] výrazně, ve srovnání s cisplatinou dokonce statisticky významně (na hladině významnosti p < 0,05; ANOVA), prodloužila průměrnou délku života myší (obrázek 26). Perspektivu do dalšího studia vnáší i výsledky histologického hodnocení prokazující toxické účinky (např. cirhóza jater, poškození střevního endotelu, nefrotoxicita), které sice byly detekovány po aplikaci všech testovaných sloučenin, avšak s výrazným rozdílem mezi zvířaty léčenými cisplatinou (irreparabilní poškození tkání) a komplexy s deriváty N6-benzyladeninu (reverzibilní poškození tkání).
47
4. Komplexy jiných p echodných kov s r znými N-donorovými ligandy Již na předešlých stranách této práce je uvedeno, že moderní bioanorganická chemie biologicky aktivních komplexů přechodných kovů má čím dál širší záběr, alespoň pokud se na problematiku díváme optikou typů přechodných kovů, jež jsou v posledních letech klinicky nebo preklinicky studovány [2,9–11,13–16,18–20]. Je sice nutné jedním dechem dodat, že to jsou stále komplexy platiny, které v těchto počtech nad ostatní přechodné kovy vyčnívají, nicméně z pohledu klinického využití si s nimi v mnohém nezadají zlatné komplexy používané jako jedna z možností v terapii zánětlivých onemocnění, jako je např. revmatoidní artritida. Také naše pracoviště v poslední době pravidelně přispívá do problematiky protizánětlivě aktivních komplexů zlata pracemi popisujícími látky na bázi komplexů zlata s různými N-donorovými ligandy, které v četných případech překonávají protizánětlivé účinky různých konvenčních terapeutik, ať už na bázi zlata nebo jiného typu. 4.1. Protizánětlivá aktivita ůu(I) komplex s deriváty N6-benzyladeninu V práci Trávníček et al., J. Med. Chem., 2012 (příloha 10) se zabýváme studiem protizánětlivé aktivity zlatných komplexů s benzyl substituovanými deriváty N6-benzyladeninu (obrázek β7), a to na úrovni in vitro i in vivo. U látek byla studována (a s klinicky užívaným protizánětlivým léčivem na bázi zlata auranofinem porovnávána) jejich schopnost regulovat hladinu proa protizánětlivých
cytokinů
spojených
s projevy
zánětlivých onemocnění, jako je TNF-α (faktor nádorové nekrózy α), IL-1
Obrázek 27. Obecný strukturní vzorec studovaných komplexů zlata s deriváty N6-benzyladeninu.
(interleukin-1beta), HMGB1 (high-mobility group protein B1), IL-10
(interleukin-10) a IL-1RA (receptorový antagonista interleukinu-1), a to na modelu lipopolysacharidem (LPS) stimulovaných makrofágů buněčné linie lidské akutní monocytické leukémie (THP-1). In vivo experimenty byly založeny na studiu změn objemu otoku tlapek u potkanů po aplikaci látky vyvolávající zánět (karagenanu). Složení celkem osmi komplexů obecného složení [Au(nL)(PPh3)] (viz příloha 10, obr. 1 na str. 4569) obsahujících 2F, 3F, 2Cl, 3Cl, 4Cl, 3Me, 4Me nebo 3OMe deriváty N6-beznyladeninu (nL) bylo prokázáno ESI+ hmotnostní spektrometrií, kterou byly detekovány molekulové píky [Au(nL)(PPh3)+H]+, a 1H a 13C NMR spektroskopií, kterou byly
48
detekovány všechny vodíkové a uhlíkové atomy připravených látek, s výjimkou N9–H vodíkových atomů použitých derivátů N6-benzyladeninu, které se ve struktuře studovaných komplexů nachází deprotonizované (během přípravy v bazickém prostředí odštěpují zmíněný N9–H vodík). Zmíněný Nλ atom pak byl na základě zjištěných hodnot koordinačních posunů atomů purinového skeletu prokázán jak atom, kterým se tyto ligandy koordinují na centrální atom. ESI-MS byla také použita jako nástroj ke studiu interakce studovaných komplexů se síru obsahujícími biomolekulami cysteinem a glutathionem. Z výsledků plyne, že uvedené biomolekuly (především cystein) s komplexy interagují ve smyslu nahrazení N-donorových ligandů na bázi N6-benzyladeninu. Ještě před vlastním studiem protizánětlivých vlastností byla u komplexů studována a prokázána (LD50 ca β M) in vitro cytotoxicita vůči THP-1 buňkám (viz příloha 10, obr. 2 na str. 4570). Určitě stojí za zmínku, že při nižších koncentracích studované komplexy nejen že THP-1 buňky neinhibovaly, ale dokonce jejich buněčné dělení stimulovaly. Na modelu prozánětlivého cytokinu TNF-α bylo zjištěno, že po 2 a 4 h studované komplexy různou měrou, avšak více ve srovnání s auranofinem, snižují genovou transkripci projevující se snížením hladiny mRNA spojené s TNF-α (viz příloha 10, obr. 3 na str. 4571). Tyto komplexy také ca β,5× snižují sekreci TNF-α. Obdobné výsledky byly získány také na dalším použitém prozánětlivém cytokinu, kterým tentokráte je IL-1
(viz příloha 10, obr. 4 na
str. 4571). Fyziologickým antagonistou zmíněného cytokinu IL-1 je protizánětlivý cytokin IL-1RA, přičemž důležitým parametrem při hodnocení protizánětlivých vlastností studovaných látek je jejich vzájemný poměr. Pro tento cytokin platí, že ačkoli studované zlatné komplexy s deriváty N6-benzyladeninu snižovaly jeho transkripci, nevykazovaly žádný výrazný efekt na sekreci IL-1RA (viz příloha 10, obr. 5 na str. 4572). Pro mnoho zánětlivých onemocnění včetně revmatoidní artritidy platí, že jsou doprovázeny zvýšenou produkcí prozánětlivého IL-1 a/nebo naopak sníženou produkcí protizánětlivého IL-1RA, což způsobuje, v konečném důsledku stejně se projevující, absolutní nebo relativní zvýšení hladiny IL-1. Z tohoto důvodu byl pro studované látky stanoven poměr zmíněných cytokinů, tedy IL-1 /IL-1RA (viz příloha 10, obr. 6 na str. 4572). Z výsledků vyplynulo, že dva ze tří studovaných komplexů (komplexy s 2F a 2Cl deriváty N6-benzyladeninu) mají výrazně nižší hodnotu poměru IL-1 /IL-1RA ve srovnání se třetím studovaným komplexem (komplex s 3Me derivátem N6-benzyladeninu), auranofinem i prednisonem (kortikosteroid pro léčbu zánětlivých onemocnění) a naopak, že se tato hodnota neliší od kontrolní skupiny buněk, u kterých nebyla lipopolysacharidem vyvolána zánětlivá reakce. V případě prozánětlivého HMGB1 studované zlatné komplexy během prvních čtyř hodin neovlivňovaly jeho transkripci 49
odlišně od auranofinu, avšak po osmi hodinách už mezi nimi lze pozorovat významný rozdíl (viz příloha 10, obr. 7 na str. 4573). Kromě výše zmíněného IL-1RA byl studován také jiný protizánětlivý cytokin, a to IL-10, pro který bylo zjištěno, že jeho transkripci studované látky ovlivňují pouze nepatrně a víceméně srovnatelně s auranofinem a prednisonem (viz příloha 10, obr. 8 na str. 4573). Na úrovni in vitro byl tedy jednoznačně prokázán vliv studovaných zlatných komplexů s deriváty N6-benzyladeninu na transkripci a expresi pro- a protizánětlivých cytokinů, což lze chápat jako výborný základ pro navazující studium protizánětlivé aktivity in vivo. Tyto experimenty byly provedeny tak, že testovaným zvířatům byl podán karagenan, kterým byl vyvolán zánět projevující se jako otok jejich tlapky. Následně byla zvířata léčena jak studovanými komplexy s N6-benzyladeniny, tak také auranofinem a indomethacinem (standardy), které byly intraperitoneálně aplikovány v dávce 10 mg/kg (5 mg/kg pro indomethacin).
Změny
objemu
otoku
tlapek
testovaných
potkanů
byly
měřeny
pletysmometricky po dobu 6 h po podání karagenanu. Jako nejaktivnější ze studie vyšel komplex [Au(2FL)(PPh3)], který významně (více než auranofin a indomethacin) snižoval objem otoku potkaních tlapek (viz příloha 10, obr. 9 na str. 4573). Další dva studované komplexy [Au(2ClL)(PPh3)] a [Au(3MeL)(PPh3)] objem otoku tlapek nesnižovaly. Pro komplex [Au(3MeL)(PPh3)] platí, že byl obdobně aktivní jako indomethacin a aktivnější než auranofin, zatímco pro nejméně aktivní [Au(2ClL)(PPh3)] lze konstatovat, že jeho aktivita byla podobná té zjištěné pro auranofin. Pro lepší pochopení mechanismů in vivo protizánětlivé aktivity a rozdílů mezi jednotlivými komplexy a použitými standardy byla provedena cytologická a histochemická analýza zvířecích tkání izolovaných po ukončení pletysmometrických pokusů. Vedle klasického hodnocení histologických změn prokazujících zánětlivé procesy v hodnocených tkáních (viz příloha 10, obr. 10 na str. 4574) byla provedena také imunohistochemická analýza kaspázy 3, TNF-α, IL-6 a E selektinu (CD62E) (viz příloha 10, tabulka 1 na str. 4574). Semikvantitativním hodnocením byla zjištěna následující míra poškození tkáníμ [Au(3MeL)(PPh3)] > [Au(2ClL)(PPh3)] > [Au(2FL)(PPh3)] = auranofin. I z těchto výsledků tedy plyne schopnost studovaných zlatných komplexů snižovat projevy zánětlivých procesů. Celkově vzato, z výsledků prezentovaných v práci Trávníček et al., J. Med. Chem., 2012 (příloha 10) plyne, že studované komplexy jsou srovnatelně nebo více in vitro a in vivo protizánětlivě aktivní ve srovnání s konvenčním protizánětlivým léčivem na bázi zlata auranofinem.
50
Na výsledky publikované v práci Trávníček et al., J. Med. Chem., 2012 (příloha 10) pak bylo navázáno dalšími pracemi našeho výzkumného týmu, které se zabývají zlatnými komplexy a jejich biologickou (nejen protizánětlivou, ale také protinádorovou) aktivitou [127,139,140]. Jednou z prací, která na uvedenou studii navazuje, je Hošek et al., PLoS ONE, 2013 (příloha 11), ve které je popsána analogická studie podobných komplexů zlata, které se od výše diskutovaných liší pouze tím, že obsahují deriváty N6-benzyladeninu substituované nejen na benzenovém kruhu, ale také na atomu Cβ purinového skeletu (obrázek β8). Tři reprezentativní komplexy, konkrétně to jsou
Obrázek 28. Obecný strukturní vzorec studovaných komplexů zlata s deriváty 2-chlorN6-benzyladeninu.
[Au(nL)(PPh3)], kde nL symbolizuje deprotonizovaný β-chlor-N6-benzyladenine (L) nebo jeho 2-methoxy (2OMeL) a 4-methyl (4MeL) deriváty, byly podrobeny studiu jejich in vitro a in vivo protizánětlivé aktivity. Také v případě těchto sloučenin byla pozorována in vitro cytotoxicita na THP-1 buňkách (LD50 ~ 1.5 µM) a při nižších koncentracích naopak navýšení metabolické aktivity nad úroveň kontrolních buněk (viz příloha 11, obr. 2 na str. 4). Na prozánětlivých cytokinech TNF-α a IL-1 bylo zjištěno navýšení jejich exprese závislé na koncentraci komplexů (viz příloha 11, obr. 3 na str. 5 a obr. 4 na str. 6), které navíc bylo výraznější ve srovnání s auranofinem. Ve srovnání s předchozí prací Trávníček et al., J. Med. Chem., 2012 (příloha 10) byl nově studován také vliv komplexů zlata na metaloproteinázy (MMP), který byl výrazný a ve srovnání s auranofinem vyšší (viz příloha 11, obr. 5 na str. 7). S ohledem na poměrně vysokou in vitro protizánětlivou aktivitu bylo poněkud překvapivé zjištění, že na úrovni in vivo studované komplexy žádnou protizánětlivou aktivitu nevykazují. Po jejich aplikaci nebyl pozorován žádný pozitivní vliv na velikost vnějším zásahem vyvolaného edému na tlapkách testovaných potkanů. Histologické studium edematické tkáně ukázalo obdobné projevy zánětlivých reakcí u vzorků všech komplexů (viz příloha 11, obr. 7 na str. 9). Dále byly imunohistochemicky stanoveny kaspáza 3, TNF-α, IL-6 a E selektin, ovšem toto studium bylo ztíženo faktem, že všechny izolované vzorky si byly vzájemně velice podobné bez výrazných rozdílů. Lze ovšem konstatovat, že histologické a histochemické změny korelovaly s výsledky pletysmometrických měření. Stejně jako u předchozích komplexů (viz výše) byla i u zde diskutovaných komplexů prokázána (ESI-MS studium) jejich schopnost interagovat s jednou ze základních síru obsahujících biomolekul
51
cysteinem, který ve struktuře původního komplexu nahrazuje příslušný derivát β-chlor-N6benzyladeninu (viz příloha 11, obr. 8 na str. 10). Celkově vzato, výsledky práce Hošek et al., PLoS ONE, 2013 (příloha 11) získané pro studované komplexy zlata s deriváty 2-chlor-N6-benzyladeninu na úrovni in vitro jsou v kontrastu
s výsledky
in
vivo
experimentů
a
následných
histologických
a imunohistochemických experimentů. Vysvětlení tohoto rozdílu vyžaduje další studium, možným vysvětlením je nižší stabilita studovaných komplexů ve fyziologických podmínkách, než je tomu v případě komplexů s odlišně substituovanými deriváty N6-benzyladeninu (Trávníček et al., J. Med. Chem., 2012, příloha 10), u kterých výsledky in vitro, in vivo a ex vivo experimentů vzájemně korelovaly. 4.2. Protinádorově účinné ůu(I) komplexy s deriváty 7-azaindolu U zlatných komplexů není studována pouze jejich protizánětlivá aktivita, jak je popsáno výše pro Au(I) komplexy s deriváty N6-benzyladeninu, ale i jiné typy biologických aktivit. Jednou z takových, kde Au(I) komplexy často vykazují farmakologicky perspektivní hodnoty, je aktivita protinádorová [141,142]. S cílem připravit takové Au(I) komplexy, které by mohly být vysoce protinádorově aktivní, byly nasyntetizovány komplexy obecného složení [Au(PPh3)(naza)], které obsahují
Obrázek 29. Obecný strukturní vzorec studovaných komplexů zlata s deriváty 7-azaindolu.
deriváty 7-azaindolu (obrázek β9). Z koordinačně-chemického pohledu je jistě zajímavé, že se deriváty 7-azaindolu koordinují na centrální atom přes dusíkový atom N1 deprotonizovaný přídavkem báze (u výše zmíněných komplexů platiny tomu bylo přes atom dusíku N7), jak bylo prokázáno detailním NMR studiem. Látky byly testovány vůči cisplatina-senzitivnímu karcinomu vaječníku (A2780), cisplatina-rezistentnímu karcinomu vaječníku (Aβ780R) a buněčné linii zdravého lidského fibroblastu (MRC5) (tabulka 5). Z výsledků je patrné, že studované Au(I) komplexy jsou aktivnější na obou nádorových liniích (u většiny látek to lze interpretovat jako schopnost překonávat rezistenci nádorových buněk vůči účinku cisplatiny), ale i na linii nenádorové. Toto se logicky projevuje do nepříliš vysokého a farmakologicky neperspektivního indexu selektivity většiny studovaných látek. Výjimkami jsou komplexy s 3Iaza, 5Braza a 2Me4Claza, jejichž index selektivity je vyšší než 8, čímž tyto látky předurčuje pro další biologická studia (více nádorových buněčných linií, lidské hepatocyty, mechanistická studia, příp. in vivo experimenty). 52
Tabulka 5. Výsledky testování in vitro cytotoxicity trifenylfosfan-zlatných komplexů s deriváty 7-azaindolu a cisplatiny vůči lidským nádorovým buněčným liniím cisplatina-senzitivního karcinomu vaječníku A2780, cisplatina-rezistentního karcinomu vaječníku A2780R a buněčné linii lidského zdravého fibroblastu MRC5. Buňky interagovaly s testovanými látkami 24 h. Data jsou uvedena jako IC50 ± SD (μM). Komplex A2780 A2780R MRC5 RF SF [Au(PPh3)(aza)] 3,8±1,1* 4,4±0,8 7,7±1,7 1,16 2,03 [Au(PPh3)(3Claza)] 22,4±5,7 21,7±0,8 31,4±4,9 0,97 1,40 [Au(PPh3)(3Braza)] 23,γ±γ,4 21,γ±0,8 27,γ±1,7 0,91 1,17 [Au(PPh3)(3Iaza)] 3,5±0,4* 11,0±4,7 29,β±5,9 3,14 8,34 [Au(PPh3)(5Braza)] 3,1±0,5* 14,4±γ,3 26,0±4,2 4,65 8,39 [Au(PPh3)(3Cl5Braza)] 22,λ±β,3 13,8±β,1 26,5±γ,7 0,60 1,16 [Au(PPh3)(3I5Braza)] 20,β±β,5 22,1±0,3 27,γ±β,6 1,09 1,35 [Au(PPh3)(2Me4Claza)] 2,8±0,7* 8,λ±γ,8 β4,6±0,4 3,18 8,78 Cisplatina 20,γ±β,3 >50,0a >50,0a * značí statisticky významný rozdíl hodnot od standardu cisplatiny na p < 0,05 hladině významnosti a) IC50 nebylo dosaženo do uvedené koncentrace
53
5. Závěr V předložené habilitační práci jsou po teoretickém úvodu do uchazečem studované problematiky biologicky aktivních komplexů přechodných kovů předloženy a diskutovány výsledky uchazečovy vědecko-výzkumné činnosti (celkem je okomentováno jedenáct příloh a několik dalších prací, ať už publikovaných nebo v různých fázích přípravy), kterou se od ukončení doktorského studia na podzim roku β010 zabýval na půdě Katedry anorganické chemie a Regionálního centra pokročilých technologií a materiálů, Přírodovědecké fakulty Univerzity Palackého v Olomouci. Tuto činnost lze rozdělit na dva vzájemně nepoměrné tematické celky, kterými jsou protinádorově aktivní komplexy platiny s různými N-donorovými ligandy (deriváty 7-azaindolu, deriváty N6-benzyladeninu) a protizánětlivě aktivní komplexy zlata, které ve své struktuře obsahují stejné typy N-donorových ligandů. V rámci obou zmíněných tematických celků se podařilo připravit látky, které svými biologickými účinky mnohonásobně převyšují konvenční chemoterapeutika na bázi platiny (cisplatina), resp. zlata (auranofin). Za všechny zde bude zmíněn alespoň cis-[PtI2(4Braza)2] komplex obsahující 4-brom-7-azaindol, který vykazuje téměř padesátkrát vyšší in vitro cytotoxicitu
vůči
nádorovým
buňkám
lidského
osteosarkomu
(HOS),
komplex
[Pt(ox)(4OMeros)2] s 4-methoxy derivátem roskovitinu, který v in vivo experimentech probíhajících na myších s implementovanou lymfocytární leukémií efektivně prodlužoval (na rozdíl od cisplatiny) průměrnou délku života testovaných zvířat a jehož aplikace navíc nevyvolávala negativní vedlejší účinky spojené s terapií léčivy na bázi platiny, nebo protizánětlivě aktivní komplex [Au(2FL)(PPh3)] s N6-(2-fluorbenzyl)adeninem, který během in vivo experimentů snižoval objem vnějším zásahem vyvolaného zánětlivého otoku potkaních
tlapek,
což
nedokázala
ani
klinicky
používaná
léčiva
auranofin
(metaloterapeutikum na bázi zlata) a indomethacin (kortikosteroid). U většiny připravených farmakologicky perspektivních látek byly v rámci širšího autorského kolektivu (kolegové z Katedry buněčné biologie a genetiky a z Katedry biofyziky resp. Biofyzikálního ústavu AV ČR) na různých úrovních studovány aspekty jejich mechanismu účinku. Rozhodně lze konstatovat, že se v souladu s cíli uchazečovy vědecko-výzkumné práce vytyčenými v Úvodu předložené habilitační práce podařilo připravit biologicky perspektivní koordinační sloučeniny vybraných přechodných kovů, které se ukázaly býti vysoce biologicky (především protinádorově) aktivními, a které, v případě protinádorově aktivních komplexů platiny, účinně překonávaly získanou rezistenci nádorových buněk vůči klinicky
54
používanému chemoterapeutiku cisplatině a působily selektivně na nádorové buňky. Mnohé z těchto látek mají potenciál pro to být dále studovány směrem k budoucím preklinickým a klinickým studiím. Obdobně byly připraveny a fyzikálně-chemicky prostudovány magnetické nanočástice na bázi oxidů železa funkcionalizované biologicky aktivními komplexy platiny, které plní požadavky pro magnetický cílený transport zmíněných terapeutik. Uvedená tématika, tedy vývoj nových biologicky aktivních koordinačních sloučenin přechodných kovů, i v současnosti je, a s výhledem do blízké budoucnosti také bude, hlavním vědecko-výzkumným zaměřením uchazeče. Z hlavních směrů výzkumu, ať už probíhajících nebo plánovaných, lze uvést např. studium karboxylato komplexů platiny s biologicky perspektivními N-donorovými ligandy pro účinnější akumulaci cílovou nádorovou buňkou, studium fotoaktivovatelných komplexů platiny s protinádorově aktivním N-donorovým ligandem odstupujícím po aplikaci vhodného záření a následně synergicky biologicky působícím pro účinnější chemoterapii bez negativních vedlejších účinků, organokovové komplexy biologicky perspektivních neplatinových přechodných kovů (např. Ru, Ir, Os) s vybranými N-donorovými ligandy vykazující jiný typ protinádorové aktivity ve srovnání s komplexy na bázi platiny nebo již výše v práci zmíněné studium funkcionalizovaných nanočástic na bázi oxidů železa pro magneticky asistovaný cílený transport léčiv potenciálně spojený také s diagnostickou funkcí.
55
6. Literatura 1.
L.R. Kelland, N.P. Farrell: Platinum-Based Drugs in Cancer Therapy. Humana: Totowa, 2000.
2.
M. Gielen, E.R.T. Tiekink: Metallotherapeutic drugs and metal-based diagnostic agents. John Wiley & Sons, Ltd.: Chichester, Emgland, 2005.
3.
L. Kelland: The resurgence of platinum-based cancer chemotherapy. Nature Rev. Cancer 7 (2007) 573.
4.
B. Rosenberg et al.: Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature 205 (1965) 698.
5.
L.M. Pasetto et al.: The development of platinum compounds and their possible combination. Crit. Rev. Oncol. Hemat. 60 (2006) 59.
6.
H.S. Oberoi et al.: Nanocarriers for delivery of platinum anticancer drugs. Adv. Drug Delivery Rev. 65 (2013) 1667.
7.
J.S. Butler, P.J. Sadler: Targeted delivery of platinum-based anticancer complexes. Curr. Opin. Chem. Biol. 17 (2013) 175.
8.
N.P.E. Barry, P.J. Sadler: Challenges for metals in medicine: how nanotechnology may help to shape the future. ACS Nano 7 (2013) 5654.
9.
N.A. Smith, P.J. Sadler: Photoactivatable metal complexes: from theory to applications in biotechnology and medicine. Philos. Trans. R. Soc. London, Ser. A 371 (2013) 20120519.
10.
P.J. Bednarski et al.: Photoactivatable platinum complexes. Anti-Cancer Agents Med. Chem. 7 (2007) 75.
11.
I. Romero-Canelon, P.J. Sadler: Next-generation metal anticancer complexes: multitargeting via redox modulation. Inorg. Chem. 52 (2013) 12276.
12.
T. Boulikas: Clinical overview on Lipoplatin: a successful liposomal formulation of cisplatin. Expert Opin. Investig. Drugs 18 (2009) 1197.
13.
D. Liu et al.: Organoiridium complexes: anticancer agents and catalysts. Int. J. Nanomed. 8 (2013) 3309.
14.
N.P.E. Barry, P.J. Sadler: Exploration of the medical periodic table: towards new targets. Chem. Commun. 49 (2013) 5106.
15.
E.S. Antonarakis, A. Emadi: Ruthenium-based chemotherapeutics: are they ready for prime time? Cancer Chemother. Pharmacol. 66 (2010) 1.
16.
A. Bergamo et al.: Approaching tumour therapy beyond platinum drugs status of the art and perspectives of ruthenium drug candidates. J. Inorg. Biochem. 106 (2012) 90.
17.
Z. Liu, P.J. Sadler: Organoiridium complexes: anticancer agents and catalysts. Acc. Chem. Res. 47 (2014) 1174.
56
18.
C.H. Leung et al.: Bioactive iridium and rhodium complexes as therapeutic agents. Coord. Chem. Rev. 257 (2013) 1764.
19.
N. Muhammad, Z. Guo: Metal-based anticancer chemotherapeutic agents. Curr. Opin. Chem. Biol. 19 (2014) 144.
20.
N. Cutillas et al.: Anticancer cyclometalated complexes of platinum group metals and gold. Coord. Chem. Rev. 257 (2013) 2784.
21.
P. Štarha et al.: Pharmacological and molecular effects of platinum(II) complexes involving 7azaindole derivatives. PLoS ONE 9 (2014) e90341.
22.
K.R. Harrap: Preclinical studies identifying carboplatin as a viable cisplatin alternative. Cancer Treat. Rev. 12 (1985) 21.
23.
Y. Kidani et al.: Antitumor activity of 1,2-diaminocyclohexaneplatinum complexes against Sarcoma-180 ascites form. J. Med. Chem. 21 (1978) 1315.
24.
A. Stein, D. Arnold: Oxaliplatin: a review of approved uses. Expert Opin. Pharmacother. 13 (2012) 125.
25.
H. Akaza et al.: Phase II study of cis-diammine(glycolato)platinum, 254-S, in patients with advanced germ-cell testicular cancer, prostatic cancer, and transitional-cell carcinoma of the urinary tract. Cancer Chemoth. Pharm. 31 (1992) 187.
26.
M.J. McKeage: Lobaplatin: a new antitumour platinum drug. Expert Opin. Inv. Drugs 10 (2001) 119.
27.
D.K. Kim et al.: Synthesis and antitumor activity of a series of [2-substituted-4,5bis(aminomethyl)-1,3-dioxolane]platinum(II) complexes. J. Med. Chem. 37 (1994) 1471.
28.
F.A. Cotton et al.: Advanced Inorganic Chemistry, Wiley; New York, 1999.
29.
D.W. Shen et al.: Cisplatin resistance: a cellular self-defense mechanism resulting from multiple epigenetic and genetic changes. Pharmacol. Rev. 64 (2012) 706.
30.
E. Raymond et al.: Cellular and molecular pharmacology of oxaliplatin. Mol. Cancer Ther. 1 (2002) 227.
31.
J.C. Zhang et al.: Status of non-classical mononuclear platinum anticancer drug development. Mini-Rev. Med. Chem. 9 (2009) 1357.
32.
B.W. Harper et al.: Advances in platinum chemotherapeutics. Chem. Eur. J. 16 (2010) 7064.
33.
L.R. Kelland et al.: Preclinical antitumor evaluation of bis-acetato-ammine-dichlorocyclohexylamine platinum(IV): an orally active platinum drug. Cancer Res. 53 (1993) 2581.
34.
J. Holford et al.: In vitro circumvention of cisplatin resistance by the novel sterically hindered platinum complex AMD473. Br. J. Cancer 77 (1998) 366.
35.
T. Dragovich et al.: A Phase 2 trial of the liposomal DACH platinum L-NDDP in patients with therapy-refractory advanced colorectal cancer. Cancer Chemoth. Pharm. 58 (2006) 759.
57
36.
P. Mistry et al.: The relationships between glutathione, glutathione-S-transferase and cytotoxicity of platinum drugs and melphalan in eight human ovarian carcinoma cell lines. Br. J. Cancer 64 (1991) 215.
37.
S. Ishida et al.: Uptake of the anticancer drug cisplatin mediated by the copper transporter Ctr1 in yeast and mammals. Proc. Natl Acad. Sci. USA 99 (2002) 14298.
38.
M.S. Davies et al.: Rates of platination of -AG- and -GA- containing double-stranded oligonucleotides: effect of chloride concentration. J. Inorg. Biochem. 79 (2000) 167.
39.
A.M.
Fichtinger-Schepman
et
al.:
Adducts
of
the
antitumor
drug
cis-
diamminedichloroplatinum(II) with DNA: formation, identification, and quantitation. Biochemistry 24 (1985) 707. 40.
D. Esteban-Fernández et al.: Analytical methodologies for metallomics studies of antitumor Ptcontaining drugs. Metallomics 2 (2010) 19.
41.
V. Brabec: DNA modifications by antitumor platinum and ruthenium compounds: their recognition and repair. Prog. Nucleic Acid Res. Mol. Biol. 71 (2002) 1.
42.
U. Frey et al.: Ring-opening reactions of the anticancer drug carboplatin: NMR characterization of cis-[Pt(NH3)2(CBDCA-O)(5'-GMP-N7)] in solution. Inorg. Chem. 32 (1993) 1333.
43.
R.W. Hay, S. Miller: Reactions of platinum(II) anticancer drugs. Kinetics of acid hydrolysis of cis-diammine(cyclobutane-1,1-dicarboxylato)platinum(II) “Carboplatin”. Polyhedron 17 (1998) 2337.
44.
M. Pavelka et al.: On the hydrolysis mechanism of the second-generation anticancer drug Carboplatin. Chem.-Eur. J. 13 (2007) 10108.
45.
M.M. Jennerwein et al.: Characterization of adducts produced in DNA by isomeric 1,2diaminocyclohexaneplatinum(II) complexes. Chem. Biol. Interact. 70 (1989) 39.
46.
J.M. Woynarowski et al.: Oxaliplatin-induced damage of cellular DNA. Mol. Pharmacol. 58 (2000) 920.
47.
L.R.
Kelland
et
al.:
Mechanism-related
circumvention
of
acquired
cis-
diamminedichloroplatinum(II) resistance using two pairs of human ovarian carcinoma cell lines by ammine/amine platinum(IV) dicarboxylates. Cancer Res. 52 (1992) 3857. 48.
Z.H. Siddik: Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 22 (2003) 7265.
49.
G. Samimi, S.B. Howell: Modulation of the cellular pharmacology of JM118, the major metabolite of satraplatin, by copper influx and efflux transporters. Cancer Chemother. Pharmacol. 57 (2006) 781.
50.
D.P. Nowotnik, E. Cvitkovicμ ProLindac™ (AP5γ46)μ A review of the development of an HPMA DACH platinum Polymer Therapeutic. Adv. Drug Deliv. Rev. 61 (2009) 1214.
51.
Y. Zhang et al.: Advanced materials and processing for drug delivery: The past and the future. Adv. Drug Deliv. Rev. 65 (2013) 104.
58
52.
A.S. Goncalves et al.: Therapeutic nanosystems for oncology nanomedicine. Clin. Transl. Oncol. 14 (2012) 883.
53.
J. Chomoucka et al.: Magnetic nanoparticles and targeted drug delivering. Pharmacol. Res. 62 (2010) 144.
54.
F.M. Kievit, M. Zhang: Surface engineering of iron oxide nanoparticles for targeted cancer therapy. Acc. Chem. Res. 44 (2011) 853.
55.
K.N. Burger et al.: Nanocapsules: lipid-coated aggregates of cisplatin with high cytotoxicity. Nature Med. 8 (2002) 81.
56.
J.R. Rice et al.: Preclinical efficacy and pharmacokinetics of AP5346, a novel diaminocyclohexane-platinum tumor-targeting drug delivery system. Clin. Cancer Res. 12 (2006) 2248.
57.
J. Holford et al.: Mechanisms of drug resistance to the platinum complex ZD0473 in ovarian cancer cell lines. Eur. J. Cancer 36 (2000) 1984.
58.
J. Helleman et al.: Mismatch repair and treatment resistance in ovarian cancer. BMC Cancer 6 (2006) 201.
59.
A. Sandler et al.: Paclitaxel–Carboplatin alone or with bevacizumab for Non–Small-Cell Lung cancer. N. Engl. J. Med. 355 (2006) 2542.
60.
M.J. Clarke: Ruthenium metallopharmaceuticals. Coord. Chem. Rev. 236 (2003) 209.
61.
N. Cutillas et al.:Anticancer cyclometalated complexes of platinum group metals and gold. Coord. Chem. Rev. 257 (2013) 2784.
62.
S. Medici et al.: Noble metals in medicine: Latest advances. Coord. Chem. Rev. 284 (2015) 329.
63.
M.J. Clarke et al.: Reduction and subsequent binding of ruthenium ions catalyzed by subcellular components. J. Inorg. Biochem. 12 (1980) 79.
64.
R. Trondl et al.: NKP-1339, the first ruthenium-based anticancer drug on the edge to clinical application. Chem. Sci. 5 (2014) 2925.
65.
A. Vacca et al.: Inhibition of endothelial cell functions and of angiogenesis by the metastasis inhibitor NAMI-A. Br. J. Cancer 86 (2002) 993.
66.
G. Süss-Fink: Arene ruthenium complexes as anticancer agents. Dalton Trans. 39 (2010) 1673.
67.
F. Wang et al.: Kinetics of aquation and anation of ruthenium(II) arene anticancer complexes, acidity and X-ray structures of aqua adducts. Chem.–Eur. J. 9 (2003) 5810.
68.
P.J. Dyson and G. Sava: Metal-based antitumour drugs in the post genomic era. Dalton Trans. (2006) 1929.
69.
I. Romero-Canelon et al.: The contrasting activity of iodido versus chlorido ruthenium and osmium arene azo- and imino-pyridine anticancer complexes: control of cell selectivity, crossresistance, p53 dependence, and apoptosis pathway. J. Med. Chem. 56 (2013) 1291.
70.
Y. Fu et al.: The contrasting chemical reactivity of potent isoelectronic iminopyridine and azopyridine osmium(II) arene anticancer complexes. Chem. Sci. 2 (2012) 2485.
59
71.
Z. Liu and P.J. Sadler: Organoiridium Complexes: Anticancer Agents and Catalysts. Acc. Chem. Res. 47 (2014) 1174.
72.
Z. Liu et al.: Organometallic Half-Sandwich Iridium Anticancer Complexes. J. Med. Chem. 54 (2011) 3011.
73.
B.M. Sutton et al.: Oral gold. Antiarthritic properties of alkylphosphinegold coordination complexes. J. Med. Chem. 15 (1972) 1095.
74.
C.F. Shaw III: Gold-based therapeutic agents. Chem Rev. 99 (1999) 2589.
75.
G.G. Graham et al.: The cellular metabolism and effects of gold complexes. Metal-Based Drugs 1 (1994) 395.
76.
F.A. Allen: The Cambridge Structural Database: a quarter of a million crystal structures and rising. Acta Crystallogr., Sect. B: Struct. Sci. 58 (2002) 380.
77.
J. Ruiz et al.: New 7-azaindole palladium and platinum complexes: crystal structures and theoretical calculations. In vitro anticancer activity of the platinum compounds. Dalton Trans. 39 (2010) 3290.
78.
S.B. Zhao et al.μ Intramolecular C−H activation directed self-assembly of an organoplatinum(II) molecular square. J. Am. Chem. Soc. 129 (2007) 3092.
79.
D. Song et al.: Syntheses and structures of new luminescent cyclometalated palladium(II) and platinum(II) complexesμ M(Bab)Cl, M(Br−Bab)Cl (M = Pd(II), Pt(II)), and Pd 3Cl4(Tab)2 (Bab = 1,3-bis(7-azaindolyl)phenyl, Br−Bab = 1-bromo-3,5-bis(7-azaindolyl)phenyl, Tab =1,3,5tris(7-azaindolyl)phenyl). Organometallics 20 (2001) 4683.
80.
S.B. Zhao et al.: Regioselective C−H activation of toluene with a 1,2-bis(N-7azaindolyl)benzene platinum(II) complex. Organometallics 24 (2005) 3290.
81.
S.W. Lai et al.: Synthesis of organoplatinum oligomers by employing N-donor bridges with predesigned geometryμ structural and photophysical properties of luminescent cyclometalated platinum(II) macrocycles. Organometallics 18 (1999) 3991.
82.
J.A. Bailey et al.: Spectroscopic and structural properties of binuclear platinum-terpyridine complexes. Inorg. Chem. 32 (1993) 369.
83.
S.C. Dhara: A rapid method for the synthesis of cis-[Pt(NH3)2Cl2]. Indian J. Chem. 8 (1970) 193.
84.
Y. Jung, S.J. Lippard: Multiple states of stalled T7 RNA polymerase at DNA lesions generated by platinum anticancer agents. J. Biol. Chem. 278 (2003) 52084.
85.
N.J. Wheate et al.: The status of platinum anticancer drugs in the clinic and in clinical trials, Dalton Trans. 39 (2010) 8113.
86.
K. Krogmann and P. Dodel: Uber die isomeric der dioxalatoplatinate. 1. Die salze. Chem. Ber. Recl. 99 (1966) 3402.
87.
P. Štarha et al.: Platinum(II) oxalato complexes with adenine-based carrier ligands showing significant in vitro antitumor activity. J. Inorg. Biochem. 104 (2010) 639.
60
88.
F. Žák, A. Czajka-Poulováμ Oxaliplatina s nízkým obsahem doprovodných nečistot a způsob její výroby. Patent CZ297703 (2007).
89.
M.J. Cleare and J.D. Hoeschele: Studies on the antitumor activity of group VIII transition metal complexes. Part I. Platinum(II) complexes. Bioinorg. Chem. 2 (1973) 187.
90.
L. Messori et al.: Cytotoxic profile and peculiar reactivity with biomolecules of a novel “rulebreaker” iodidoplatinum(II) complex. ACS Med. Chem. Lett. 1 (2010) 381.
91.
F. Huq et al.: Antitumour activity of cis-bis{imidazo(1,2-α)-pyridineplatinum(II). Int. J. Cancer Res. 2 (2006) 367.
92.
I. Lakomska et al.: Platinum(II) complexes with 5,7-disubstituted-1,2,4-triazolo [1,5a]pyrimidines: Spectroscopical characterization and cytotoxic activity in vitro. Spectrochim. Acta A, 91 (2012) 126.
93.
M. Ravera et al.: Synthesis, characterization, structure, molecular modeling studies and biological activity of sterically crowded Pt(II) complexes containing bis(imidazole) ligands. J. Inorg. Biochem. 105 (2011) 400.
94.
Q.S. Ye et al.: Synthesis, characterization and cytotoxicity of dihalogeno-platinum(II) complexes with L-histidine ligand. Chem. Pharm. Bull. 57 (2009) 424.
95.
B.A.J. Jansen et al.: Glutathione induces cellular resistance against cationic dinuclear platinum anticancer drugs. J. Inorg. Biochem. 89 (2002) 197.
96.
I. Romero-Canelón et al.: The contrasting activity of iodido versus chlorido ruthenium and osmium arene azo- and imino-pyridine anticancer complexes: control of cell selectivity, crossresistance, p53 dependence, and apoptosis pathway. J. Med. Chem. 56 (2013) 1291.
97.
J. Mlcouskova et al.: Antitumor carboplatin is more toxic in tumor cells when photoactivated: enhanced DNA binding. J. Biol. Inorg. Chem. 17 (2012) 891.
98.
M.A. Lemaire et al.: Interstrand cross-links are preferentially formed at the d(GC) sites in the reaction between cis-diamminedichloroplatinum(II) and DNA. Proc. Natl. Acad. Sci. USA 88 (1991) 1982.
99.
V. Brabec et al.: Monofunctional adducts of platinum(II) produce in DNA a sequencedependent local denaturation. Biochemistry 33 (1994) 1316.
100. V. Brabec and M. Leng: DNA interstrand cross-links of trans-diamminedichloroplatinum(II) are preferentially formed between guanine and complementary cytosine residues. Proc. Natl. Acad. Sci. USA 90 (1993) 5345. 101. J.L. Butour and J.P. Macquetμ Differentiation of DNA · platinum complexes by fluorescence. Eur. J. Biochem. 78 (1977) 455. 102. J.L. Butour et al.: Effect of the amine non-leaving group on the structure and stability of DNA complexes with cis-[Pt(R-NH2)2(NO3)2]. Eur. J. Biochem. 202 (1991) 975.
61
103. F.J. Ramos-Lima et al.: Structural characterization, DNA interactions, and cytotoxicity of new transplatin analogues containing one aliphatic and one planar heterocyclic amine ligand. J. Med. Chem. 49 (2006) 2640. 104. C. Huang et al.: Trapping iron oxide into hollow gold nanoparticles. Nanoscale Res. Lett. 6 (2011) 1. 105. L. Wang et al.: Monodispersed core-shell Fe3O4@Au nanoparticles. J. Phys. Chem. B 109 (2005) 21593. 106. L. Zhang et al.: Efficient purification of His-tagged protein by superparamagnetic Fe3O4/Au– ANTA–Co2+ nanoparticles. Mater. Sci. Eng. C 33 (2013) 1989. 107. M. Kumagai et al.: Enhanced in vivo magnetic resonance imaging of tumors by PEGylated ironoxide–gold core–shell nanoparticles with prolonged blood circulation properties. Macromol. Rapid Commun. 31 (2010) 1521. 108. H. Yang et al.: Feasibility of MR imaging in evaluating breast cancer lymphangiogenesis using Polyethylene glycol-GoldMag nanoparticles. Clin. Radiol. 68 (2013) 1233. 109. Z. Fan et al.: Multifunctional plasmonic shell–magnetic core nanoparticles for targeted diagnostics, isolation, and photothermal destruction of tumor cells. ACS Nano 6 (2012) 1065. 110. E.S. Krystofiak et al.: Multiple morphologies of gold–magnetite heterostructure nanoparticles are effectively functionalized with protein for cell targeting. Microsc. Microanal. 19 (2013) 821. 111. J. Salado et al.: Functionalized Fe3O4@Au superparamagnetic nanoparticles: in vitro bioactivity. Nanotechnology 23 (2012) 315102. 112. A.J. Wagstaff et al.: Cisplatin drug delivery using gold-coated iron oxide nanoparticles for enhanced tumour targeting with external magnetic fields. Inorg. Chim. Acta 393 (2012) 328. 113. D. Maity et al.: Surface design of core–shell superparamagnetic iron oxide nanoparticles drives record relaxivity values in functional MRI contrast agents. Chem. Commun. 48 (2012) 11398. 114. J. Wang et al.: A platinum anticancer theranostic agent with magnetic targeting potential derived from maghemite nanoparticles. Chem. Sci. 4 (2013) 2605. 115. C.K. Lo et al.: Homocysteine-protected gold-coated magnetic nanoparticles: synthesis and characterisation. J. Mater. Chem. 17 (2007) 2418. 116. L. Havlíček et al.: Cytokinin-derived cyclin-dependent kinase inhibitorsμ synthesis and cdcβ inhibitory activity of Olomoucine and related compounds. J. Med. Chem. 40 (1997) 408. 117. L. Meijer et al.: Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur. J. Biochem. 243 (1997) 527. 118. V. Kryštof et al.: Synthesis and biological activity of olomoucine II. Bioorg. Med. Chem. Lett. 12 (2002) 3283. 119. C. Benson et al.: Clinical anticancer drug development: targeting the cyclin-dependent kinases. Br. J. Cancer 92 (2005) 7.
62
120. Z. Trávníček et al.: The first iron(III) complexes with cyclin-dependent kinase inhibitors: Magnetic, spectroscopic (IR, ES+ MS, NMR,
57
Fe Mössbauer), theoretical, and biological
activity studies. J. Inorg. Biochem. 104 (2010) 405. 121. A. Klanicová et al.: Synthesis, characterization and in vitro cytotoxicity of Co(II) complexes with N6-substituted adenine derivatives: X-ray structures of 6-(4-chlorobenzylamino)purin-diium diperchlorate dihydrate and [Co6(µ-L6)4Cl8(DMSO)10]∙4DMSO. Polyhedron 25 (2006) 1421. 122. Z. Trávníček et al.: Preparation and cytotoxic activity of nickel(II) complexes with 6-benzylaminopurine derivatives. Transition Met. Chem. 27 (2002) 918. 123. A. Klanicová et al.: Dinuclear copper(II) perchlorate complexes with 6-(benzylamino)purine derivatives: Synthesis, X-ray structure, magnetism and antiradical activity. Polyhedron 29 (2010) 2582. 124. Z. Trávníček et al.: Zinc(II) complexes with potent cyclin-dependent kinase inhibitors derived from 6-benzylaminopurine: Synthesis, characterization, X-ray structures and biological activity. J. Inorg. Biochem. 100 (2006) 214. 125. Z. Trávníček et al.: In vitro and in vivo biological activity screening of Ru(III) complexes involving 6-benzylaminopurine derivatives with higher pro-apoptotic activity than NAMI-A. J. Inorg. Biochem. 105 (2011) 937. 126. Z. Trávníček et al.: Synthesis, characterization and assessment of the cytotoxic properties of cis and trans-[Pd(L)2Cl2] complexes involving 6-benzylamino-9-isopropylpurine derivatives. J. Inorg. Biochem. 101 (2007) 477. 127. R. Křikavová et al.: Diverse in vitro and in vivo anti-inflammatory effects of trichloridogold(III) complexes with N6-benzyladenine derivatives. J. Inorg. Biochem. 134 (2014) 92. 128. Z. Trávníček et al.: The first platinum(IV) complexes involving aromatic cytokinins or cyclindependent kinase inhibitors derived from 6-benzylaminopurine: X-ray structures of (BohH22+)[PtCl6]∙H2O and (RosH22+)2[PtCl6]Cl2∙4H2O. Polyhedron 26 (2007) 5271. 129. Z. Trávníček et al.: Mixed ligand complexes of platinum(II) and palladium(II) with cytokininderived compounds Bohemine and Olomoucine: X-ray structure of [Pt(BohH+-N7)Cl3]∙λ/5H2O {Boh = 6-(benzylamino)-2-[(3-(hydroxypropyl)-amino]-9-isopropylpurine, Bohemine}. J. Inorg. Biochem. 94 (2003) 307. 130. M. Maloň et al.: Synthesis, spectral study and cytotoxicity of platinum(II) complexes with 2,9disubstituted-6-benzylaminopurines. J. Inorg. Biochem. 99 (2005) 2127. 131. L. Szüčová et al.: Novel platinum(II) and palladium(II) complexes with cyclin-dependent kinase inhibitors: Synthesis, characterization and antitumour activity. Bioorg. Med. Chem. 14 (2006) 479.
63
132. Z.
Trávníček
et
al.:
trans-Bis{2-chloro-6-[(3-hydroxybenzyl)amino]-9-isopropylpurine-
κN7}platinum(II) dimethylformamide disolvate. Acta Crystallogr. Sect. E-Struct. Rep. Online E62 (2006) m1482. 133. L. Szüčová et al.: Preparation and cis-to-trans transformation study of square-planar [Pt(Ln)2Cl2] complexes bearing cytokinins derived from 6-benzylaminopurine (Ln) by view of NMR spectroscopy and X-ray crystallography. Polyhedron 27 (2008) 2710. 134. B. Liskova et al.: Cellular response to antitumor cis-dichlorido platinum(II) complexes of CDK inhibitor Bohemine and its analogues. Chem. Res. Toxicol. 25 (2012) 500. 135. Z. Trávníček et al.: Roscovitine-based CDK inhibitors acting as N-donor ligands in the platinum(II) oxalato complexes: Preparation, characterization and in vitro cytotoxicity. Eur. J. Med. Chem. 45 (2010) 4609. 136. R. Vrzal et al.: Evaluation of in vitro cytotoxicity and hepatotoxicity of platinum(II) and palladium(II) oxalato complexes with adenine derivatives as carrier ligands. J. Inorg. Biochem. 104 (2010) 1130. 137. V. Masek et al.: Interaction of selected platinum(II) complexes containing roscovitine-based CDK inhibitors as ligands with human liver microsomal cytochrome P450. Biomed. Pap.Olomouc (2014), v tisku. 138. O. Novakova et al.: Conformation and recognition of DNA damaged by antitumor cis-dichlorido platinum(II) complex of CDK inhibitor bohemine. Eur. J. Med. Chem. 78 (2014) 54.
139. Z. Trávníček et al.: Komplexy zlata s -substituovanými deriváty 6-alkyloxy-9-deazapurinů a deriváty fosfanu a použití těchto komplexů pro přípravu léčiv k terapii zánětlivých a nádorových onemocnění. Užitný vzor CZ27030 (2014). 140. Z. Trávníček et al.: Komplexy zlata s deriváty hypoxantinu a s deriváty fosfanu a jejich použití pro přípravu léčiv v protizánětlivé a protinádorové terapii. Užitný vzor CZ27032 (2014). 141. J. Vančo et al.: Gold(I) complexes of 9-deazahypoxanthine as selective antitumor and antiinflammatory agents. PLoS ONE 9 (2014) e109901. 142. R. Křikavová et al.: Gold(I)-triphenylphosphine complexes with hypoxanthine-derived ligands: in vitro evaluations of anticancer and anti-inflammatory activities. PLoS ONE 9 (2014) e107373.
64
7. Seznam p íloh [1]
Štarha, P.; Marek, J.; Trávníček, Z. Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations. Polyhedron 33 (2012) 404–409.
[2]
Štarha, P.; Trávníček, Z.; Popa, A.; Popa, I.; Muchová, T.; Brabec, V. How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole. J. Inorg. Biochem. 115 (2012) 57–63.
[3]
Muchová, T.; Prachařová, J.; Štarha, P.; Olivová, R.; Vrána, O.; Benešová, B.; Kašpárková, J.; Trávníček, Z.; Brabec, V. Insight into the toxic effects of cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives in tumor cells. J. Biol. Inorg. Chem. 18 (2013) 579–589.
[4]
Štarha, P.; Hošek, J.; Vančo, J.; Dvořák, Z.; Suchý, P.; Popa, I.; Pražanová, G.; Trávníček, Z. Pharmacological and molecular effects of platinum(II) complexes involving 7-azaindole derivatives. PLoS ONE 9 (2014) e90341.
[5]
Štarha, P.; Trávníček, Z.; Popa, I.; Dvořák, Z. Synthesis, characterization and in vitro antitumor activity of platinum(II) oxalato complexes involving 7-azaindole derivatives as coligands. Molecules 19 (2014) 10832–10844.
[6]
Štarha, P.; Trávníček, Z.; Dvořák, Z.; Radošová-Muchová, T.; Prachařová, J.; Vančo, J.; Kašpárková, J. Potentiating effect of UVA irradiation on anticancer activity of carboplatin derivatives involving 7-azaindoles. PLoS ONE (2015) akceptovaný manuskript.
[7]
Štarha, P.; Stavárek, M.; Tuček, J.; Trávníček, Z. 4-Aminobenzoic acid-coated maghemite nanoparticles as potential anticancer drug magnetic carriers: a case study on highly cytotoxic cisplatin-like complexes involving 7-azaindoles. Molecules 19 (2014) 1622–1634.
[8]
Štarha, P.; Smola, D.; Tuček, J.; Trávníček, Z. Efficient synthesis of a maghemite/gold hybrid nanoparticle system as a magnetic carrier for the transport of platinum-based metallotherapeutics. Int. J. Mol. Sci. 16 (2015) 2034–2051 .
[9]
Dvořák, Z.; Štarha, P.; Trávníček, Z. Evaluation of in vitro cytotoxicity of 6-benzylaminopurine carboplatin derivatives against human cancer cell lines and primary human hepatocytes. Toxicol. in Vitro 25 (2011) 652–656.
65
[10] Trávníček, Z.; Štarha, P.; Vančo, J.; Šilha, T.; Hošek, J.; Suchý, P.; Pražanová, G. Anti-inflammatory active gold(I) complexes involving 6-substituted-purine derivatives. J. Med. Chem. 55 (2012) 4568–4579. [11] Hošek, J.; Vančo, J.; Štarha, P.; Paráková, L.; Trávníček, Z. Effect of 2-chlorosubstitution of adenine moiety in mixed-ligand gold(I) triphenylphosphine complexes on anti-inflammatory activity: the discrepancy between the in vivo and in vitro models. PLoS ONE 8 (2013) e82441.
66
PŘÍLOHA 1 Štarha, P.; Marek, J.; Trávníček, Z. Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations Polyhedron 33 (2012) 404–409
Polyhedron 33 (2012) 404–409
Contents lists available at SciVerse ScienceDirect
Polyhedron journal homepage: www.elsevier.com/locate/poly
Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations Pavel Štarha a, Jaromír Marek b, Zdeneˇk Trávnícˇek a,⇑ a
´ University, 17. listopadu 12, Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacky CZ-771 46 Olomouc, Czech Republic b ´ University, 17. listopadu 12, CZ-771 46 Olomouc, Czech Republic Department of Inorganic Chemistry, Faculty of Science, Palacky
a r t i c l e
i n f o
Article history: Received 31 October 2011 Accepted 30 November 2011 Available online 13 December 2011 Keywords: Platinum(II) complexes 7-Azaindole X-ray structure Multinuclear NMR In vitro cytotoxicity
a b s t r a c t Platinum(II) dichlorido and oxalato (ox) complexes of the composition cis-[PtCl2(Haza)2] (I) and [Pt(ox) (Haza)2]H2O (IIH2O) were prepared by a one-step syntheses from K2[PtCl4] and K2[Pt(ox)2]2H2O, respectively (Haza symbolises 7-azaindole). The complexes were characterised by a set of spectroscopic methods (IR, Raman and multinuclear and 2D NMR), as well as by mass spectrometry, elemental (C, H, and N) and thermal (TG/DTA) analyses. Single-crystal X-ray analysis was performed for cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II), obtained by recrystallization of the complexes I and IIH2O from N,N0 -dimethylformamide (DMF). Both complexes, of a square-planar geometry, involve a tetracoordinated central Pt(II) atom with two monodentate N7-coordinated 7-azaindole molecules in cis-positions, and with the coordination sphere completed by two chloride anions (I; PtN2Cl2 donor set) or one bidentatecoordinated oxalate dianion (II; PtN2O2 donor set). The complexes did not show any in vitro cytotoxic effect, as evaluated by an MTT assay against breast adenocarcinoma and osteosarcoma human cancer cell lines up to a concentration of 1.0 lM (for I) and 0.1 lM (for IIH2O), given by their limited solubility in the medium used. Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Up until now, several platinum(II) complexes involving 7-azaindole (Haza) have been reported [1–7]; those whose structure contains derivatives of Haza are described in the literature [8–18]. From a coordination chemistry point of view, Haza may serve as a monodentate or bidentate N-donor ligand, which can be coordinated to the metal centre(s) through its N1 and/or N7 atoms. As for the platinum(II) complexes, the following coordination modes of the 7-azaindole moiety to the Pt(II) atom were determined by crystallographic studies. [NBu4][Pt(C6F5)2(Haza)(aza)]Haza involves the Haza molecule coordinated through the N7 atom, however its deprotonated form (aza) is coordinated through the N1 atom [3]. The same work also reported the X-ray structures of [NBu4]2[{Pt(C6F5)2}2(l-OH)(l-aza)] (aza coordinates through N1 and N7), [Pt(aza)(dmba)(PPh3)] and [Pt(aza)(dmba)(dmso)] (both with aza coordinated through N1); dmba stands for N,C-bidentate coordinated N,N0 -dimethylbenzylamine. A deprotonated form of 7-azaindole coordinates through N1 and N7 to Pt(II) and Ag(I) in the structure of [NBu4][{(C6F5)3Pt}(l-aza){Ag(Me2CNH)}] (Me2CNH = acetone imine) [5]. 1-Diphenylphosphino-7-azaindole (dppaza) is bound to the Pt(II) atom through N7 in [PtCl2(dppaza)] ⇑ Corresponding author. Tel.: +420 585 634 352; fax: +420 585 634 954. E-mail address:
[email protected] (Z. Trávnícˇek). 0277-5387/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.poly.2011.11.059
[8]. Zhao et al. prepared platinum complexes involving 1-(20 -pyridyl)-7-azaindole (paza), which is bidentate-coordinated through the N7 atom of the 7-azaindole ring and the nitrogen atom of pyridyl in the case of [Pt(paza)(Me)2], while in the structure of an organoplatinum tetrameric macrocycle, [Pt(paza)(Me)]4, it is also bound through the C2 atom of the 7-azaindole moiety [11]. A group of numerous complexes with two 7-azaindole molecules linked together by various linkers bound to the N1 atoms and coordinated to platinum through N7, such as [Pt(Ph)2(1,3-bam)], have been reported in the literature [12] [bam = N,N-bis(7-azaindolyl)methane, Ph = phenyl]. Similar compounds have also been described in the literature, as follows: [PtCl(1,3-bab)] and [PtCl(Brbab)] with 1,3-bis(N7-azaindolyl)benzene (1,3-bab) and 1-bromo-3,5-bis(N7-azaindolyl)benzene (Brbab), coordinated through the mentioned N7 atoms as well as through one carbon atom of the aromatic ring [13]; [Pt(1,2-bab)Me2], [Pt(1,2bab)Ph(SMe2)][BAr0 4], [Pt(1,2-bab)(Bz)(SMe2)][BAr0 4], [Pt(1,20 bab)(Bz)(MeCN)][BAr 4] were reported in [15], where Bz = benzyl and Ar0 = 3,5-bis(trifluoromethyl)phenyl; Zhao et al. prepared [Pt(L)(CHMePh)(MeCN)][BAr0 4] (L stands for 1,2-bab and bam) and [Pt(1,2-bab)(l3-CHMePh)][BAr0 4] [16]. A different coordination mode of the 7-azaindole moiety was described for the complexes with bam and its isomer linked through the N1 and N7 atoms (L0 ), which are involved in [Pt(L0 )Me2] and [Pt(L0 )Ph(SMe2)][BAr0 4], where the azaindole-based ligands are coordinated via the
P. Štarha et al. / Polyhedron 33 (2012) 404–409
nitrogen atoms not substituted by a CH2 linker [14]. A different linker, namely Me3Si-substituted methylene, connects both 7-azaindole moieties within the bam derivative (Me3Si-bam) structure of [Pt(Me3Si-bam)Me2] and [PtI2(Me3Si-bam)][{PtIMe3}4] [17]. As it is known from the literature, as well as from the recently reported results of our research team, platinum(II) complexes involving an organic compound, which is physiologically potent itself, as a ligand coordinated to the metal centre show significant biological effects, such as in vitro cytotoxicity. With respect to the fact that the derivatives of 7-azaindole are known to be physiologically active substances, which show various types of activities in dependence on their structure [19–25], we decided to prepare simple platinum(II) dichlorido and oxalato complexes with these substances (concretely halogeno derivatives of 7-azaindole) and to study their in vitro cytotoxicity. Preliminary results regarding one of the mentioned complexes showed that its in vitro cytotoxicity was evaluated as several times higher than that of cisplatin (unpublished data). The herein described cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2] (IIH2O) were prepared and characterised as model compounds with respect to syntheses optimizations and physicochemical properties. It has to be stated that the title complexes were, together with cis-[PtI2(Haza)2], reported by Harrison et al. [1] and their in vivo toxicity and antitumor activity against Yoshida Sarcoma, Osteosarcoma, ADJ/PC6A tumour and P388 Lymphocytic leukaemia were studied on rats. Interestingly, neither [1], nor any of the articles cited therein, nor any other literature source have described the synthesis and characterisation of these compounds as yet.
405
2.3. Physical methods
2. Experimental
Elemental analyses were carried out on a Flash 2000 CHNS Elemental Analyzer (Thermo Scientific). Infrared spectra were recorded on a Nexus 670 FT-IR (Thermo Nicolet) in the 400–4000 cm 1 (KBr pellets) and 150–600 cm 1 (Nujol technique) regions. Raman spectra were recorded using an NXR FT-Raman Module (Thermo Nicolet) between 150 and 3750 cm 1. The reported IR and Raman signal intensities have been defined as w = weak, m = middle, s = strong and vs = very strong. Mass spectra of acetonitrile solutions of the complexes were obtained by LCQ Fleet ion trap mass spectrometry using the ESI technique in both the positive and negative modes (Thermo Scientific). All the observed isotopic distribution representations were compared with the theoretical ones (QualBrowser software, version 2.0.7, Thermo Fischer Scientific). Simultaneous TG/DTA analyses were performed using an Exstar TG/DTA 6200 thermal analyzer (Seiko Instruments Inc.); 100 mL min 1 dynamic air atmosphere, 25–700 °C (2.5 °C min 1). 1H, 13C and 195Pt NMR spectra and two dimensional correlation experiments (1H–1H gs-COSY, 1H–13C gs-HMQC, 1H–13C gs-HMBC) of the DMF-d7 solutions were measured at 300 K on a Varian 400 device. 1H and 13C spectra were adjusted against the signals of tetramethylsilane (Me4Si). 195Pt spectra were calibrated against K2[PtCl6] in D2O, found at 0 ppm. 1 H–15N gs-HMBC experiments were obtained at natural abundance and calibrated against the residual signals of N,N0 -dimethylformamide (DMF) adjusted to 8.03 ppm (1H) and 104.7 ppm (15N). The splitting of proton resonances in the reported 1H spectra is defined as s = singlet, d = doublet, t = triplet, br = broad band, dd = doublet of doublets, m = multiplet.
2.1. Starting materials
2.4. Single crystal X-ray analysis
K2[PtCl4], K2(ox)H2O, 7-azaindole and solvents were purchased from Sigma–Aldrich Co., Acros Organics Co. and Lachema Co. and they were used as received. K2[Pt(ox)22H2O was prepared according to the literature [26].
In efforts to obtain crystals suitable for a single-crystal X-ray analysis, the compounds cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2]H2O (IIH2O) were recrystallized from DMF, which led to the formation of cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II). The X-ray data of IDMF and II were collected on an Xcalibur™2 diffractometer (Oxford Diffraction Ltd.) with a Sapphire2 CCD detector and with Mo Ka radiation (Monochromator Enhance, Oxford Diffraction Ltd.). Data collection and reduction were performed using the CrysAlis software [27]. The same software was used for data correction for absorption effects by the empirical absorption correction using spherical harmonics, implemented in the SCALE3 ABSPACK scaling algorithm. The structure was solved by direct methods using SHELXS-97 and refined on F2 using the full-matrix least-squares procedure (SHELXL-97) [28]. Non-hydrogen atoms were refined anisotropically and hydrogen atoms were located in difference maps, fixed in their theoretical geometrical positions with distances of N–H = 0.88 Å, C–H = 0.95 Å and
2.2. Synthesis of cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2]H2O (IIH2O) 7-Azaindole (1.0 mmol) and 0.5 mmol of K2[PtCl4] (for I) or K2[Pt(ox)2]2H2O (for IIH2O) were separately dissolved in a minimum volume of hot ethanol and hot distilled water, respectively (Scheme 1). The solutions were mixed together and the reaction mixture was stirred at 50 °C for 2 days. The obtained products were filtered off and washed with distilled water and ethanol, and then dried at a temperature of 40 °C. The results of the elemental analysis, mass spectrometry and IR, Raman and NMR spectroscopy are given in Supplementary data.
Scheme 1. Schematic pathway for the preparation of cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2]H2O (IIH2O) given together with the atom numbering and colour labelling of ring A (Haza moiety involving C and N atoms), ring B (Haza moiety involving C0 and N0 atoms) and ring C (atoms of the PtX2N2 coordination environment, where X symbolises Cl for I and O for IIH2O). (Colour online.)
406
P. Štarha et al. / Polyhedron 33 (2012) 404–409
Cmethyl–H = 0.98 Å, and refined isotropically by using the riding model with Uiso(H) = 1.2Ueq(C, N) [1.5Ueq(C) for methyl atoms]. The crystal data and structure refinements are given in Table 1. Molecular graphics as well as additional structural calculations were drawn and interpreted using DIAMOND [29] and Mercury [30]. 2.5. In vitro cytotoxicity In vitro cytotoxic activity was determined by an MTT assay [MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] in human breast adenocarcinoma (MCF7; ECACC No. 86012803) and human osteosarcoma (HOS; ECACC No. 87070202) cancer cell lines purchased from the European Collection of Cell Cultures (ECACC). For more details see Supplementary data. 3. Results and discussion 3.1. General and spectroscopic properties We prepared I and IIH2O using the simple one-step syntheses starting from K2[PtCl4] and K2[Pt(ox)2]2H2O, respectively, which reacted with two molar equivalents of 7-azaindole (see Scheme 1). The contents of C, H and N were proven by elemental analysis, showing the composition and purity of the prepared complexes. The results of thermal analysis proved complex I to be non-solvated, while IIH2O was found to be monohydrated (Dm between 26 and 132 °C: calc. 3.3% for H2O; found 3.7%); the TG and DTA curves of IIH2O are given in Supplementary data (Fig. S1). ESI- mass spectrometry of I (see Fig. S2 in Supplementary data) detected the molecular peak of [PtCl2(Haza)(aza)] , its [{PtCl2(aza)2}+Na] adduct with the sodium cation, as well as Table 1 Crystal data and structure refinements for cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II).
Empirical formula Formula weight T (K) Wavelength (Å) Crystal system Space group a (Å) b (Å) c (Å) a (°) b (°) c (°) V (Å3) Z Dcalc (g cm 3) Absorption coefficient (mm 1) Crystal size (mm) F (0 0 0) h Range for data collection (°) Index ranges (h, k, l)
Reflections collected/unique (Rint) Data/restraints/parameters Goodness-of-fit (GOF) on F2 Final R indices [I > 2r(I)] R indices (all data) Largest peak and hole (e Å
3
)
IDMF
II
C14H12Cl2N4PtC3H7NO 575.36 120(2) 0.71073 orthorhombic P212121 8.92437(12) 10.63464(16) 20.2948(3) 90 90 90 1926.13(5) 4 1.984 7.579 0.4 0.3 0.25 1104 2.98 6 h 6 25.05 10 6 h 6 10 12 6 k 6 12 24 6 l 6 20 18 048/3409 (0.0213)
C16H12N4O4Pt 519.39 120(2) 0.71073 monoclinic P21/c 9.0200(3) 16.0498(6) 11.1413(4) 90 103.249(4) 90 1569.98(10) 4 2.197 8.968 0.35 0.15 0.1 984 2.92 6 h 6 25.05 10 6 h 6 9 19 6 k 6 15 13 6 l 6 13 9856/2778 (0.0368) 2778/0/226 0.995 R1 = 0.0222, wR2 = 0.0452 R1 = 0.0342, wR2 = 0.0466 1.618, 0.747
3409/0/237 1.127 R1 = 0.0118, wR2 = 0.0298 R1 = 0.0120, wR2 = 0.0298 0.746, 0.345
peaks whose m/z values and isotopic distribution corresponded to the [PtCl(aza)2] and [PtCl2(aza)] fragments (see Supplementary data). Similarly, the mass spectrum of IIH2O contains the molecular peaks of the [Pt(ox)(Haza)(aza)] and [{Pt(ox)(aza)2}+Na] adducts (Fig. S2). Fragments of the compositions [Pt(ox)(aza)] and [aza] were also observed in the mass spectrum of the studied oxalato complex. Further, mass spectrometry performed in the positive mode (ESI+) detected the [{PtCl2(Haza)2}+K]+ adduct of complex I with the potassium cation at 540.0 m/z and fragments whose m/z values of 430.2 and 118.9 correspond to the composition of [{Pt(Haza)(aza)}]+ and [Haza+H]+, respectively. The spectrum also contains the peaks of [{PtCl2(Haza)2}2+K]+, [{PtCl2(Haza)2}2+Na]+ and [PtCl(Haza)3]+, which are formed by the recombinations of the molecules of the studied complex with other ones or with the free Haza molecule. The ESI+ mass contains the spectrum of IIH2O [{Pt(ox)(Haza)2}+Na]+ adduct at 542.0 m/z, its fragments of [{Pt(Haza)(aza)}]+, [{Pt(ox)(Haza)}+Na]+ and [Haza+H]+, and [{Pt(ox)(Haza)2}2+K]+, [{Pt(ox)(Haza)2}2+Na]+ and [{Pt(ox)(Haza)2}2+H]+ adducts. The Raman and IR spectra of I and IIH2O (Supplementary data) contain all the vibrations of 7-azaindole [31], whose maxima positions were slightly shifted and their intensities changed as well, due to the coordination of Haza to the Pt(II) atom. However, several facts should be discussed in more details: (1) the Raman vibrations of free Haza at 1586 and 1603 cm 1 (IR: 1584 and 1600 cm 1) were found as a single vibration in this region for I and IIH2O; (2) the band in the IR spectrum of free Haza detected at 1499 cm 1 was split in the spectra of both complexes; (3) the Raman vibration of Haza at 1044 cm 1 was not detected in the spectra of the complexes; (4) the Raman vibration at 1338 cm 1 (IR: 1343 cm 1) remained unchanged in the case of IIH2O, but it split into two peaks in the spectra of I; (5) the Raman spectrum of I contains, contrary to Haza, a vibration of a medium intensity at 338 cm 1 assignable to the Pt–Cl bond, and another additional band was observed at 1090 cm 1; (6) additional bands at 1128, 484 and 467 cm 1 were detected in the IR spectrum of I as compared to free Haza; (7) the maximum of Haza of the strong intensity at 765 cm 1 was not found in the IR spectra of the complexes; (8) as for the difference between the spectra of free Haza and IIH2O, several more bands were detected in the Raman (values in the text) and IR (values in parentheses) spectra of this complex, those at 933, 1383 (1393), 1654 (1654), (1671) and 1689 (1705) cm 1 belong to the vibrations of the oxalate dianion, and the vibration with a maximum at 483 (484) cm 1 can be assigned to a five-membered ring (PtO2C2) deformation [32]. All the signals in the 1H, 13C and 1H–15N gs-HMBC spectra of 7azaindole were found in the corresponding spectra of both complexes (Table 2). The 1H–15N gs-HMBC experiments proved the coordination via the N7 atom of Haza, since the coordination shifts (calculated as Dd = dcomplex dligand) of this atom equal 102.8 ppm for I and 116.2 ppm for IIH2O, while the Dd values of the second nitrogen atom involved in the structure of Haza (N1) are 2.34 ppm (I) and 1.72 ppm (IIH2O). The coordination of Haza to the metal centre through N7 causes an electron density redistribution, which leads to the upfield shift of the C7a atom, which neighbours the coordination site. Other carbon atoms, including the C6 atom situated, as well as the above mentioned C7a, next to N7, are shifted downfield. Surprisingly, the |Dd| values of the carbons neighbouring the coordination site of the Haza moiety are not the highest ones (Table 2): |Dd|(C6) in the case of both studied complexes is exceeded by |Dd|(C3a), and |Dd|(C4) and |Dd|(C7a) are exceeded by all the carbon atoms of I and IIH2O, except for the C5 atom of I. The 13C NMR spectrum of IIH2O involves one more signal (166.1 ppm) as compared with that of free Haza and I, which belongs to the carbon atoms (C11, C12) of the oxalate
407
P. Štarha et al. / Polyhedron 33 (2012) 404–409 Table 2 The results of NMR spectroscopy given as 1H, 13C and 15N NMR chemical shifts d/coordination shifts (Dd = dcomplex complexes cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2]H2O (IIH2O). 1
I IIH2O
dligand) [ppm] obtained and calculated for the prepared
13
H NMR
15
C NMR
N NMR
N1H
C2H
C4H
C5H
C6H
C2
C3
C3a
C4
C5
C6
C7a
N1
N7
13.13/ 1.43 13.07/ 1.37
7.81/ 0.26 7.86/ 0.31
8.06/ 0.08 8.20/ 0.22
7.13/ 0.06 7.19/ 0.12
8.88/ 0.61 8.74/ 0.47
127.7/ 1.6 128.1/ 2.0
101.9/ 1.8 102.2/ 2.1
123.4/ 3.2 123.5/ 3.3
131.3/ 3.1 132.0/ 3.8
116.9/ 1.2 117.1/ 1.4
145.1/ 2.3 145.8/ 3.0
147.7/ 1.4 148.1/ 1.0
142.3/ 2.3 141.7/ 1.7
169.9/ 102.8 156.5/ 116.2
Fig. 1. The molecular structure of cis-[PtCl2(Haza)2]DMF (IDMF) with the nonhydrogen atoms depicted as thermal ellipsoids at the 50% probability level and given with the atom numbering scheme (except for C3A0 and C7A0 ).
Fig. 3. Part of the crystal structure of cis-[PtCl2(Haza)2]DMF (IDMF) showing selected non-covalent interactions of the N–H O, N–H C, C–H C and C–H Cl types depicted as dashed green (hydrogen bonds) and orange lines; hydrogen atoms not involved in the depicted interactions are omitted for clarity. (Colour online.)
d = 2119.5 ppm for I, and 1783.8 ppm for IIH2O. It corresponds with our previously obtained 195Pt NMR spectroscopy results of cisdichlorido [33] and oxalato [26,34] complexes involving monodentate-coordinated N-donor N6-benzyladenine-based ligands, whose signals were found at ca. 2020 and 1685 ppm, respectively.
Fig. 2. The molecular structure of [Pt(ox)(Haza)2] (II) with the non-hydrogen atoms depicted as thermal ellipsoids at the 50% probability level and given with the atom numbering scheme (except for C3A0 and C7A0 ).
dianion, bidentately-coordinated to the metal centre. The highest changes of the chemical shift in the 1H NMR spectra were found for the N1H and C6H protons, which in the case of N1H is probably given by the steric hindrance caused by the coordination of the Haza molecules to the central Pt(II) atom and by the involvement of the N1H proton in the non-covalent bond system. The coordination shift of the C6H proton relates to the coordination through the neighbouring N7 atom. The 195Pt NMR spectra showed one signal, however, it was detected at very different positions:
3.2. Single crystal X-ray analysis of cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II) The complexes show a slightly distorted square-planar geometry in the vicinity of the tetracoordinated central Pt(II) atom. The dichlorido complex I (Fig. 1) adopts a PtCl2N2 donor set with the chlorine atoms mutually arranged in cis positions, while a PtO2N2 coordination environment of II (Fig. 2) involves a bidentate-coordinated oxalate anion with two Haza ligands in a cis fashion. The bond lengths and angles formed by the atoms of both coordination environments can be found in Table 3. The Pt–N bond lengths of both structures are somehow shorter as compared with the average value of this bond determined for platinum(II) dichlorido [2.013(25) Å] and oxalato [1.997(14) Å] complexes with two N-donor imine ligands deposited in Crystallographic Structural
408
P. Štarha et al. / Polyhedron 33 (2012) 404–409
3.3. In vitro cytotoxicity As was already mentioned above, Harrison et al. [1] reported I and II as substances showing no in vivo toxic and antitumor effect against Yoshida Sarcoma, Osteosarcoma, ADJ/PC6A tumour and P388 Lymphocytic leukaemia on rats. On the other hand, these complexes were prepared as model compounds of platinum(II) dichlorido and oxalato complexes involving various derivatives of 7-azaindole whose in vitro cytotoxic effects were found to be significantly higher as compared with the platinum-based anticancer drug cisplatin (unpublished results). Because of these facts, we decided to evaluate the in vitro cytotoxicity of I and IIH2O, however, it could not be exactly determined owing to the limited solubility of the complexes in the used medium, and thus the obtained IC50 values can be expressed only as >1.0 lM for I and >0.1 lM for IIH2O against both cell lines (IC50 of cisplatin, formerly determined by the same method, equals 19.6 lM against MCF7 and 34.2 lM against HOS; [26]). Fig. 4. Part of the crystal structure of [Pt(ox)(Haza)2] (II) showing selected noncovalent interactions of the N–H O, C–H O, C–H C and C C types depicted as dashed green (hydrogen bonds) and orange lines; hydrogen atoms not involved in the depicted interactions are omitted for clarity. (Colour online.)
Table 3 Selected bond lengths (Å) and angles (°) of cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II). Bond lengths
IDMF
II
Bond angles
IDMF
II
Pt N7 Pt N70 Pt–Cl1 Pt–Cl2 Pt–O1 Pt–O2
2.023(2) 2.009(2) 2.2913(7) 2.2966(7)
2.001(4) 2.009(4)
N7–Pt–N70 N7–Pt–X1 N7–Pt–X2 N70 –Pt–X1 N70 –Pt–X2 X1–Pt–X2
90.94(9) 89.16(7) 177.02(7) 177.77(7) 87.18(7) 92.81(3)
87.8(2) 92.12(14) 174.66(14) 178.93(14) 97.4(2) 82.68(13)
2.012(3) 2.031(3)
X = Cl for IDMF and O for II.
Database (CSD, ver. 5.32.1, August 2011 update [35]). Similarly, the Pt–Cl [2.296(13) Å] and Pt–O [2.009(18) Å] bond lengths of the previously mentioned reported structures found in CSD can be compared with the parameters determined for our complexes, and it can be said that the values correlate well, except for the Pt–O2 bond which is significantly longer. The differences of the angles around the central atom are caused by the presence of the bidentate oxalate anion in the structure of II, resulting in significantly different values of analogous X1–Pt–X2 angles (X = Cl for IDMF and O for II), as well as in different values for the other angles given in Table 2. The planes (see Scheme 1) fitted through atoms of both Haza moieties (ring A involves C and N atoms, ring B involves C0 and N0 atoms) and atoms of the PtX2N2 coordination environment (ring C; X symbolises Cl for IDMF and O for II) form the following dihedral angles: \AB = 82.23(7)° (IDMF) and 86.60(12)° (II), \AC = 66.81(5)° (IDMF) and 68.49(10)° (II), \BC = 88.82(4)° (IDMF) and 89.41(7)° (II). The crystal structures of IDMF (see Fig. 3) and II (see Fig. 4) contain N–H O hydrogen bonds with N O distances and N– H O angles ranging from 2.803(3) to 3.067(6) Å, and 163.9(2)° to 174.4(3)°, respectively, which fall to the interval of values typical for hydrogen bonds of medium strength (D A 2.5–3.2 Å and \NHO 130–180°; see Ref. [36]). The other non-covalent contacts detected in the crystal structures are of the types N–H C, C–H C and C–H Cl (for IDMF; Fig. 3), and C–H C, C–H O and C C (for II; Fig. 4), whose parameters are summarised in Table S1 in Supplementary data.
4. Conclusions The cis-[PtCl2(Haza)2] (I) and [Pt(ox)(Haza)2]H2O (IIH2O) complexes with 7-azaindole were synthesised and thoroughly characterised. The crystallographic study of the title complexes showed a slightly distorted square-planar geometry with two monodentate 7-azaindole molecules coordinated through the N7 atoms to the metal centre. The in vitro cytotoxicity of both I and IIH2O was studied, but the complexes were inactive up to a concentration of 1.0 lM, and 0.1 lM, respectively, given by their limited solubility in the medium used. For all that, we believe in the biological potential of such types of complexes, which motivates us for future work. Now we are studying structural analogues of the herein presented compounds, involving halogeno-derivatives of 7-azaindole. Based on the preliminary results we can confirm the generally known fact that only an insignificant structural modification of the molecule may significantly improve its biological interest because the changes lead to the preparation of biologically attractive platinum(II) compounds with in vitro cytotoxicity several times better than that of cisplatin. After the completion of these results, they will become the subject of a future publication.
Acknowledgements The authors gratefully thank the Grant Agency of the Czech Republic (P207/11/0841), the Operational Program Research and Development for Innovations – European Regional Development Fund (CZ.1.05/2.1.00/03.0058), the Operational Program Education for Competitiveness – European Social Fund (CZ.1.07/2.3.00/ 20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic, and Palacky´ University (PrF_2011_014) for financial support. The authors also thank Dr. Radim Vrzal for carrying out the cytotoxicity tests.
Appendix A. Supplementary data CCDC 848957 and 848958 contain the supplementary crystallographic data for cis-[PtCl2(Haza)2]DMF (IDMF) and [Pt(ox)(Haza)2] (II). These data can be obtained free of charge via http:// www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
[email protected]. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.poly.2011.11.059.
P. Štarha et al. / Polyhedron 33 (2012) 404–409
References [1] R.C. Harrison, C.A. McAuliffe, M.E. Friedman, Inorg. Chim. Acta 92 (1984) 43. [2] E.J. New, C. Roche, R. Madawala, J.Z. Zhang, T.W. Hambley, J. Inorg. Biochem. 103 (2009) 1120. [3] J. Ruiz, V. Rodríguez, C. de Haro, A. Espinosa, J. Pérez, C. Janiak, Dalton Trans. 39 (2010) 3290. [4] J.M. Casas, B.E. Diosdado, J. Forniés, A. Martín, A.J. Rueda, A.G. Orpen, Inorg. Chem. 47 (2008) 8767. [5] J.M. Casas, J. Forniés, A. Martín, A.J. Rueda, Organometallics 21 (2002) 4560. [6] S.W. Lai, M.C.W. Chan, K.K. Cheung, S.M. Peng, C.M. Che, Organometallics 18 (1999) 3991. [7] J.A. Bailey, V.M. Miskowski, H.B. Gray, Inorg. Chem. 32 (1993) 369. [8] H.L. Milton, M.V. Wheatley, A.M.Z. Slawin, J.D. Woollins, Inorg. Chem. Commun. 7 (2004) 1106. [9] A.D. Burrows, M.F. Mahon, M. Varrone, Dalton Trans. (2003) 4718. [10] Z.M. Hudson, S.B. Zhao, R.Y. Wang, S. Wang, Chem. Eur. J. 15 (2009) 6131. [11] S.B. Zhao, R.Y. Wang, S. Wang, J. Am. Chem. Soc. 129 (2007) 3092. [12] S.B. Zhao, G.H. Liu, D. Song, S. Wang, Dalton Trans. (2008) 6953. [13] D. Song, Q.G. Wu, A. Hook, I. Kozin, S.N. Wang, Organometallics 20 (2001) 4683. [14] D. Song, S. Wang, Organometallics 22 (2003) 2187. [15] S.B. Zhao, D.T. Song, W.L. Jia, S.N. Wang, Organometallics 24 (2005) 3290. [16] S.B. Zhao, G. Wu, S. Wang, Organometallics 25 (2006) 5979. [17] S.B. Zhao, R.Y. Wang, S. Wang, Organometallics 28 (2009) 2572. [18] S.B. Zhao, Q. Cui, S. Wang, Organometallics 29 (2010) 998. [19] P.J. Cox, T.N. Majid, S. Amendola, S.D. Deprets, C. Edlin, J.Y.Q. Lai, A.D. Morley, US Patent 20040198737, 2004. [20] M. Klein, US Patent 20110166175, 2011.
409
[21] H.J. Dyke, S. Price, K. Williams, US Patent 20100216768, 2010. [22] G.N. Anilkumar, S.B. Rosenblum, S. Venkatraman, F.G. Njoroge, J.A. Kozlowski, US Patent 20100239527, 2010. [23] D. Simard, Y. Leblanc, C. Berthelette, M.H. Zaghdane, C. Molinaro, Z. Wang, M. Gallant, S. Lau, T. Thao, M. Hamel, R. Stocco, N. Sawyer, S. Sillaots, F. Gervais, R. Houle, J.F. Lévesque, Bioorg. Med. Chem. Lett. 20 (2010) 7212. [24] S. Hong, S. Lee, B. Kim, H. Lee, S.-S. Hong, S. Hong, Bioorg. Med. Chem. Lett. 21 (2011) 841. [25] C. Marminon, A. Pierré, B. Pfeiffer, V. Pérez, S. Léonce, P. Renard, M. Prudhomme, Bioorg. Med. Chem. 11 (2003) 679. [26] P. Štarha, Z. Trávnícˇek, I. Popa, J. Inorg. Biochem. 104 (2010) 639. [27] CrysAlis CCD and CrysAlis RED, Version 1.171.33.52, Oxford Diffraction Ltd., England, 2009. [28] G.M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr. 64 (2008) 112. [29] K. Brandenburg, DIAMOND, Release 3.1f, Crystal Impact GbR, Bonn, Germany, 2006. [30] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, J. Appl. Crystallogr. 41 (2008) 466. [31] B. Karthikeyan, Spectrochim. Acta, Part A 64 (2006) 1083. [32] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, fifth ed., Wiley-Interscience, New York, 1997. [33] L. Szücˇová, Z. Trávnícˇek, I. Popa, J. Marek, Polyhedron 27 (2008) 2710. [34] Z. Trávnícˇek, P. Štarha, I. Popa, R. Vrzal, Z. Dvorˇák, Eur. J. Med. Chem. 45 (2010) 4609. [35] F.A. Allen, Acta Crystallogr., Sect. B: Struct. Sci. 58 (2002) 380. [36] G. Gilli, P. Gilli, The Nature of the Hydrogen Bond: Outline of a Comprehensive Hydrogen Bond Theory, Oxford, New York, 2009.
PŘÍLOHA 2 Štarha, P.; Trávníček, Z.; Popa, A.; Popa, I.; Muchová, T.; Brabec, V. How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole J. Inorg. Biochem. 115 (2012) 57–63
Journal of Inorganic Biochemistry 115 (2012) 57–63
Contents lists available at SciVerse ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole Pavel Štarha a, Zdeněk Trávníček a,⁎, Alexandr Popa a, Igor Popa a, Tereza Muchová b, c, Viktor Brabec c a b c
Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacký University, 17. listopadu 12, CZ-77146 Olomouc, Czech Republic Department of Biophysics, Faculty of Sciences, Palacký University, 17. listopadu 12, CZ-77146 Olomouc, Czech Republic Institute of Biophysics, Academy of Sciences of the Czech Republic, v.v.i., Kralovopolská 135, CZ-61265 Brno, Czech Republic
a r t i c l e
i n f o
Article history: Received 13 April 2012 Accepted 17 May 2012 Available online 24 May 2012 Keywords: Platinum(II) complexes 7-azaindole derivatives X-ray structure Multinuclear NMR In vitro cytotoxicity DNA binding
a b s t r a c t The platinum(II) dichlorido and oxalato complexes of the general formula cis-[PtCl2(nHaza)2] (1–3) [Pt(ox)(nHaza)2] (4–6) involving 7-azaindole halogeno-derivatives (nHaza) were prepared and thoroughly characterized. A single-crystal X-ray analysis of cis-[PtCl2(3ClHaza)2]·DMF (1·DMF; 3ClHaza symbolizes 3-chloro-7-azaindole) revealed a distorted square-planar arrangement with both the 3ClHaza molecules coordinated through their N7 atoms in a cis fashion. In vitro cytotoxicity of the complexes was evaluated by an MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay against the HOS (osteosarcoma), MCF7 (breast adenocarcinoma) and LNCaP (prostate adenocarcinoma) human cancer cell lines. The dichlorido complexes 1–3 (IC50 = 3.8, 3.9, and 2.5 μM, respectively) showed significantly higher in vitro anticancer effect against HOS as compared with cisplatin, whose IC50 = 37.7 μM. The biological effect of cisplatin against MCF7 (IC50 = 24.5 μM) and LNCaP (IC50 = 3.8 μM) was also exceeded by 1–3 (except for 2 against LNCaP), but the difference can be classified as significant only in the case of 1 (IC50 = 3.4 μM) and 3 (IC50 = 2.0 μM) against MCF7. The molecular pharmacological studies (RNA synthesis by T7 RNA polymerase in vitro) proved that 1–3 bind to DNA in a similar manner as cisplatin, since the RNA synthesis products of 1–3 and cisplatin showed a similar sequence profile of major bands. © 2012 Elsevier Inc. All rights reserved.
1. Introduction The 7-azaindole (Haza) derivatives have been utilized and patented as biologically effective organic compounds with various types of activities, such as anti-proliferative [1], protein-kinase inhibition [2] and treatment of kinase-induced diseases [3], anticancer and anti-inflammatory [4], and antiviral [5]. Other works such as those of Simard et al. [6], Marminon et al. [7] and Hong et al. [8] are also noteworthy, since they describe Haza derivatives of various structural types showing high biological activity, namely anti-proliferative activity [6,7] or CRTH2 (chemoattractant receptor-homologous molecule expressed on Th2 cells) receptor antagonistic activity [8]. With respect to these facts it is quite surprising that only three works have dealt with platinum(II) complexes involving 7-azaindolebased ligands and described a biological study of such compounds. The cis-[PtCl2(Haza)(NH3)] and trans-[PtCl2(Haza)(NH3)] complexes
⁎ Corresponding author. Tel.: + 420 585 634 352; fax: +420 585 634 954. E-mail address:
[email protected] (Z. Trávníček). 0162-0134/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2012.05.006
were studied for their cytotoxic activity in vitro against A2780 human ovarian carcinoma cells and they showed lower activity as compared with cisplatin (see Section 3.2) [9]. J. Ruiz et al. [10] prepared several mono- and dinuclear complexes of platinum with 7-azaindole, such as [Pt(dmba)(aza)(dmso)] showing submicromolar cytotoxic activity against A2780, its cisplatin-resistant analog A2780R and human breast cancer cells T47D; dmba = N,N′-dimethylbenzylamine, aza = a deprotonated form of 7-azaindole. The complexes cis-[PtCl2(Haza)2], [Pt(ox)(Haza)2] and cis-[PtI2(Haza)2] were studied in vivo for their antitumor activity against Yoshida sarcoma, osteosarcoma, ADJ/PC6A tumor and P388 lymphocytic leukemia on rats, but surprisingly no activity was observed [11]. One more conclusion of the literature research regarding 7-azaindole derivatives (nHaza) has to be mentioned — 3-chloro-7azaindole (3ClHaza), 3-iodo-7-azaindole (3IHaza) and 5-bromo-7azaindole (5BrHaza) have not been reported as ligands of transition metal complexes to date. In other words, these heterocyclic compounds can be taken into account as novel ligands from the coordination chemistry point of view. One of our latest works reported synthesis, characterization and cytotoxicity in vitro, determined by an MTT [3-(4,5-dimethylthiazol2-yl)-2,5-diphenyltetrazolium bromide] assay against breast adenocarcinoma (MCF7) and osteosarcoma (HOS) human cancer cell lines, of
58
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63
two of the above-mentioned substances [12]. The cis-[PtCl2(Haza)2] (IC50 >1.0 μM against both cell lines) and [Pt(ox)(Haza)2] (IC50 >0.1 μM against both cell lines) complexes showed very limited solubility in the medium used (0.1% DMF in water; DMF = N,N′-dimethylformamide), which disallowed exact evaluation of their cytotoxicity in vitro. Thus, one of the topics of the present work was how to modify the Haza molecule to increase both solubility and cytotoxicity of the final platinum(II) complexes. One of the important early phases of the mechanism by which platinum compounds exert their anticancer activity is the formation of adducts on nuclear DNA by these agents [13–15]. Hence, data on the DNA binding mode of platinum complexes are of interest. In the present study, we have also applied a methodology based on transcription mapping of DNA adducts in a cell-free medium previously developed for cisplatin and its analogs to investigate the reaction of 1–3 with natural DNA. 2. Materials and methods 2.1. Preparation of the platinum(II) dichlorido (1–3) and oxalato (4–6) complexes K2[PtCl4] (0.5 mmol) was dissolved in a minimum of distilled water and poured into the ethanolic solution of 3ClHaza, 3IHaza, and 5BrHaza (1.0 mmol), respectively. The mixture was stirred at 50 °C for two days and then the product was filtered off and washed by distilled water (5 mL) and ethanol (5 mL). The obtained complexes cis-[PtCl2(3ClHaza)2] (1), cis-[PtCl2(3IHaza)2]·0.75EtOH (2) and cis-[PtCl2(5BrHaza)2]·0.5EtOH (3) were dried at the temperature of 40 °C (Fig. 1). K2[Pt(ox)2]·2H2O (0.5 mmol) dissolved in a minimum volume of distilled water was added to 1.0 mmol of 3ClHaza, 3IHaza or 5BrHaza dissolved in a minimum volume of ethanol. The obtained products of the composition [Pt(ox)(3ClHaza)2]·0.2EtOH (4), [Pt(ox)(3IHaza)2]·0.75H2O (5), [Pt(ox)(5BrHaza)2]·0.75H2O (6) were filtered off after two days of
stirring at 50 °C, washed by 5 mL of distilled water and 5 mL of ethanol, and dried at the temperature of 40 °C (Fig. 1). 1: Anal. calc. for C14H10N4Cl4Pt: C, 29.44%; H, 1.76%; N, 9.81%. Found: C, 29.02%; H, 1.65%; N, 9.33%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.51 (br, N1H, 1H), 9.02 (d, 5.6, C6H, 1H), 8.07 (d, 6.5, C4H, 1H), 8.05 (s, C2H, 1H), 7.29 (m, C5H, 1H). 13C NMR (DMF-d7, SiMe4, ppm): δ 146.82 (C6), 146.00 (C7a), 129.17 (C4), 125.17 (C2), 120.95 (C3a), 117.70 (C5), 104.24 (C3). 15N NMR (DMF-d7, ppm): δ 140.9 (N1), 173.4 (N7). 195Pt NMR (DMF-d7, K2PtCl6, ppm): δ −2126.2. 2: Anal. calc. for C14H10N4Cl2I2Pt·0.75EtOH: C, 23.61%; H, 1.85%; N, 7.10%. Found: C, 23.32%; H, 1.68%; N, 6.73%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.59 (br, N1H, 1H), 8.98 (dd, 5.7, 1.1, C6H, 1H), 8.09 (d, 2.6, C2H, 1H), 7.82 (d, 7.6, C4H, 1H), 7.28 (m, C5H, 1H). 13 C NMR (DMF-d7, SiMe4, ppm): δ 147.35 (C7a), 146.50 (C6), 132.51 (C2), 131.57 (C4), 125.66 (C3a), 117.84 (C5), 55.58 (C3). 15N NMR (DMF-d7, ppm): δ 149.8 (N1), 172.6 (N7). 195Pt NMR (DMF-d7, K2PtCl6, ppm): δ − 2121.6. 3: Anal. calc. for C14H10N4Cl2Br2Pt·0.5EtOH: C, 26.37%; H, 1.91%; N, 8.20%. Found: C, 25.93%; H, 1.74%; N, 8.27%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.47 (br, N1H, 1H), 9.25 (d, 1.7, C6H, 1H), 8.31 (d, 1.6, C4H, 1H), 7.91 (t, 3.0, C2H, 1H), 6.65 (m, C3H, 1H). 13C NMR (DMF-d7, SiMe4, ppm): δ 146.84 (C7a), 145.19 (C6), 133.66 (C4), 129.74 (C2), 124.74 (C3a), 110.17 (C5), 101.94 (C3). 15N NMR (DMF-d7, ppm): δ 143.5 (N1), 174.9 (N7). 195Pt NMR (DMF-d7, K2PtCl6, ppm): δ − 2112.4. 4: Anal. calc. for C16H10N4Cl2O4Pt·0.2EtOH: C, 32.97%; H, 1.89%; N, 9.38%. Found: C, 33.25%; H, 1.78%; N, 9.35%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.35 (br, N1H, 1H), 8.82 (dd, 5.8, 1.0, C6H, 1H), 8.22 (dd, 7.9, 1.0, C4H, 1H), 8.09 (s, C2H, 1H), 7.32 (m, C5H, 1H). 13 C NMR (DMF-d7, SiMe4, ppm): δ 165.89 (C11, C12), 147.59 (C6), 146.42 (C7a), 129.87 (C4), 125.53 (C2), 121.04 (C3a), 117.87 (C5), 104.47 (C3). 15N NMR (DMF-d7, ppm): δ 139.9 (N1), 159.9 (N7). 195Pt NMR (DMF-d7, K2PtCl6, ppm): δ − 1785.7. 5: Anal. calc. for C16H10N4I2O4Pt·0.75H2O: C, 24.49%; H, 1.48%; N, 7.14%. Found: C, 24.18%; H, 1.20%; N, 6.65%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.48 (br, N1H, 1H), 8.79 (dd, 5.9, 1.1, C6H, 1H), 8.11 (s, C2H, 1H), 7.96 (dd, 8.0, 1.2, C4H, 1H), 7.30 (m, C5H, 1H). 13C NMR (DMF-d7, SiMe4, ppm): δ 165.94 (C11, C12), 147.79 (C7a), 147.25 (C6), 132.89 (C2), 132.26 (C4), 125.76 (C3a), 118.00 (C5), 55.80 (C3). 15 N NMR (DMF-d7, ppm): δ 148.7 (N1), 159.4 (N7). 195Pt NMR (DMF-d7, K2PtCl6, ppm): δ −1783.0. 6: Anal. calc. for C16H10N4Br2O4Pt·0.75H2O: C, 27.82%; H, 1.68%; N, 8.11%. Found: C, 27.86%; H, 1.39%; N, 7.61%. 1H NMR (DMF-d7, SiMe4, ppm): δ 13.35 (br, N1H, 1H), 9.16 (d, 1.9, C6H, 1H), 8.47 (d, 1.9, C4H, 1H), 7.96 (d, 3.5, C2H, 1H), 6.74 (d, 3.4, C3H, 1H). 13C NMR (DMF-d7, SiMe4, ppm): δ 165.80 (C11, C12), 147.46 (C7a), 145.66 (C6), 134.38 (C4), 130.14 (C2), 124.83 (C3a), 110.22 (C5), 102.21 (C3). 15N NMR (DMF-d7, ppm): δ 143.0 (N1), 160.8 (N7). 195 Pt NMR (DMF-d7, K2PtCl6, ppm): δ −1777.6. The results of thermogravimetric (TG) and differential thermal (DTA) analyses, electrospray ionization (ESI) mass spectrometry, and IR and Raman spectroscopy are given in the Supplementary data. 2.2. Materials and physical measurements
Fig. 1. Synthesis and schematic representation of the cis-[PtCl2(3ClHaza)2] (1), cis[PtCl2(3IHaza)2]·0.75EtOH (2), cis-[PtCl2(5BrHaza)2] (3), [Pt(ox)(3ClHaza)2]·0.2EtOH (4), [Pt(ox)(3IHaza)2]·0.75H2O (5) and [Pt(ox)(5BrHaza)2]·0.75H2O (6) complexes given together with the 7-azaindole derivatives involved in their structures.
K2[PtCl4], K2(ox)·H2O, 3-chloro-7-azaindole (3ClHaza), 3-iodo-7azaindole (3IHaza) and 5-bromo-7-azaindole (5BrHaza) and solvents were purchased from Sigma-Aldrich Co., Acros Organics Co., Lachema Co. and Fluka Co. and they were used as received. K2[Pt(ox)2]·2H2O was prepared according to the literature [16]. Elemental analyses were carried out on a Flash 2000 CHNS Elemental Analyzer (Thermo Scientific). Infrared spectra were recorded on a Nexus 670 FT-IR (Thermo Nicolet) in the 400–4000 cm − 1 (ATR technique) and 150–600 cm− 1 (Nujol technique) regions. Raman spectra of 1, 2, 4 and 6 (the complexes 3 and 5 burnt under laser beam) were recorded using an
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63
NXR FT-Raman Module (Thermo Nicolet) between 150 and 3750 cm− 1. Mass spectra of the acetonitrile solutions of complexes were obtained by an LCQ Fleet ion trap mass spectrometer using the ESI-technique (Thermo Scientific). All the observed isotopic distribution representations were compared with the theoretical ones (QualBrowser software, version 2.0.7, Thermo Fischer Scientific). Simultaneous TG/DTA analyses were performed using an Exstar TG/DTA 6200 thermal analyzer (Seiko Instruments Inc.); ceramic crucible, 100 mL min− 1 dynamic air atmosphere, 25–700 °C temperature range and temperature gradient of 2.5 °C min− 1. 1H, 13C and 195Pt NMR spectra and two dimensional correlation experiments (1H–1H gsCOSY, 1H–13C gs-HMQC, 1H–13C gs-HMBC; 1H– 15N gs-HMBC; gs = gradient selected, COSY = correlation spectroscopy, HMQC = heteronuclear multiple quantum coherence, HMBC = heteronuclear multiple bond coherence) of the DMF-d7 solutions were measured at 300 K on a Varian 400 device at 400.00 MHz ( 1H), 100.58 MHz (13C), 86.00 MHz (195Pt) and 40.53 MHz ( 15N). 1H and 13C spectra were adjusted against the signals of tetramethylsilane (Me4Si). 195Pt spectra were calibrated against potassium hexachloroplatinate (K2PtCl6) in D2O found at 0 ppm. 1H–15N gs-HMBC experiments were obtained at natural abundance and calibrated against the residual signals of DMF adjusted to 8.03 ppm (1H) and 104.7 ppm (15N). The splitting of proton resonances in the reported 1H spectra is defined as s = singlet, d = doublet, t = triplet, br = broad band, dd = doublet of doublets, m = multiplet. The single crystal X-ray data of the selected crystal of cis-[PtCl2(3ClHaza)2]·DMF (1·DMF), obtained by a slow evaporation of a DMF solution, was obtained using an Xcalibur™2 diffractometer (Oxford Diffraction Ltd.) with a Sapphire2 CCD detector, and with Mo Kα (Monochromator Enhance, Oxford Diffraction Ltd.). Data collection and reduction were performed using the CrysAlis software [17]. The same software was used for data correction for an absorption effect by the empirical absorption correction using spherical harmonics, implemented in SCALE3 ABSPACK scaling algorithm. The structure was solved by direct methods using SHELXS-97 and refined on F2 using the full-matrix least-squares procedure (SHELXL-97) [18]. Non-hydrogen atoms were refined anisotropically and hydrogen atoms were located in a difference map and refined by using the riding model with C\H = 0.95 and 0.98 Å, and N\H = 0.88 Å, with Uiso(H) = 1.2 Ueq(CH, NH) and 1.5 Ueq(CH3). The crystal data and structure refinements are given in Table S1. The molecular graphics as well as additional structural calculations were drawn and interpreted using DIAMOND [19] and Mercury [20]. Plasmid pSP73KB (2455 bp) was isolated according to standard procedures. Restriction endonucleases NdeI and HpaI were obtained from New England Biolabs (Beverly, MA). T7 RNA polymerase and RNasin ribonuclease inhibitor were purchased from Promega (Mannheim, Germany). Ribonucleotide triphosphates were from Roche Diagnostics, GmbH (Mannheim, Germany). [α- 32P]rCTP was from MP Biomedicals, LLC (Irvine, CA). 2.3. In vitro cytotoxic activity testing In vitro cytotoxic activity was determined by an MTT assay in human breast adenocarcinoma (MCF7; ECACC no. 86012803), human osteosarcoma (HOS; ECACC no. 87070202) and prostate adenocarcinoma (LNCaP; ECACC no. 89110211) cancer cell lines purchased from European Collection of Cell Cultures (ECACC). The cells were cultured according to the ECACC instructions and they were maintained at 37 °C and 5% CO2 in a humidified incubator. Human cancer cells were treated with 1–6 and cisplatin (applied up to 50 μM) for 24 h, using multi-well culture plates of 96 wells. In parallel, the cells were treated with vehicle (DMF; 0.1%, v/v) and Triton X-100 (1%, v/v) to assess the minimal (i.e. positive control) and maximal (i.e. negative control) cell damage, respectively. The MTT assay was measured spectrophotometrically at 540 nm (TECAN, Schoeller Instruments LLC).
59
2.4. Statistical evaluation The data were expressed as the percentage of viability, when 100% and 0% represent the treatments with DMF and Triton X-100, respectively. The cytotoxicity data from the cancer cell lines were acquired from three independent experiments (conducted in triplicate) using cells from different passages. The IC50 values were calculated from viability curves. The results are presented as arithmetic mean ± SD. The significance of the differences between the results was assessed by ANOVA analysis with p b 0.05 considered to be significant (QC Expert 3.2, Statistical software, TriloByte Ltd.). 2.5. DNA transcription by RNA polymerase in vitro A double-stranded DNA template was prepared by digesting the pSP73KB plasmid (2455 bp) with NdeI and HpaI restriction endonucleases. The resulting two fragments, 212- and 2243-bp long, were separated on a 1% agarose gel in the buffer containing 40 mM Tris–acetate (pH 8), 1 mM EDTA and 0.5 mg mL− 1 ethidium bromide. The 212-bp fragment was isolated from the gel, purified using Wizard_SV and PCR Clean-Up System and incubated with the complexes 1, 2, 3 or cisplatin in NaClO4 (10 mM) for 24 h at 37 °C in the dark. At the end of the incubation, the samples were precipitated by ethanol and dissolved in the TE buffer (10 mM Tris–Cl, pH 7.4, 1 mM EDTA). The level of platination (rb, the number of molecules of the platinum complex bound per nucleotide residue) in aliquots of these samples was checked by flameless atomic absorption spectrometry. In this way, the analyses of DNA transcription were performed in the absence of unbound (free) platinum complexes. Transcription of the 212-bp (NdeI/HpaI) restriction fragment with DNA-dependent T7 RNA polymerase and electrophoretic analysis of transcripts were performed according to the protocols recommended by Promega (Promega Protocols and Applications, 43–46 (1989/90)) and previously described in detail [21]. 3. Results and discussion 3.1. Preparation and structural characterization The pale yellow platinum(II) complexes of the composition cis[PtCl2(nHaza)2]·xSolv (1–3; xSolv stands for 0.75EtOH for 2 and 0.5EtOH for 3) were prepared from K2PtCl4 in the mixture of water and ethanol at 50 °C with a yield of 70–80% (Fig. 1). The oxalato complexes, [Pt(ox)(nHaza)2]·xSolv (4–6; xSolv = 0.2EtOH for 4 and 0.75H2O for 5 and 6), were isolated as the pale yellow (4, 6) or pale gray (5) powders with yields of ~ 90% (Fig. 1). The empirical formulas, including the water (5 and 6) and ethanol (2, 3 and 4) molecules of crystallizations, resulted from the elemental (C, H, N) and thermal (TG/DTA) analyses (see Supplementary data, Fig. S1). Negativemode electrospray mass spectrometry detected molecular peaks of all the studied complexes at m/z = 570 (1), 752 (2), 658 (3), 586 (4), 770 (5) and 676 (6) (see Supplementary data, Fig. S2). All the mass spectra of Pt(II)-dichlorido complexes (1–3) also contain a peak whose mass corresponds to the fragment of the composition [PtCl2(naza)] −. Analogical fragments, [Pt(ox)(naza)] −, were also detected in the spectra of the oxalato complexes (4–6). All the 1H and 13C signals of free nHaza molecules were detected in the NMR spectra of 1–6 and their positions were, due to their coordination to the PtCl2 (for 1–3) or Pt(ox) (for 4–6) motif, somehow shifted as compared with the free nHaza molecules (Table 1). The 13C NMR signals of 4–6 observed at 165.80–165.94 ppm are assignable to the carbon atoms (C11, 12) of the oxalate dianion. The 1H– 15N gs-HMBC experiments of the complexes showed both the nitrogen atoms of the 7azaindole moiety, however, they were found to be significantly differently shifted (Table 1), which proved the N7-coordination mode of nHaza molecules to the Pt(II) ion, as it was crystallographically determined for 1·DMF (see below). The 195Pt chemical shifts of Pt(II)-
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63
60 Table 1 The 1H, 13C and 1
1 2 3 4 5 6
15
N NMR coordination shifts (Δδ = δcomplex − δligand; ppm) of the 7-azaindole moiety atoms and 13
H NMR
195
Pt chemical shifts (δ; ppm) of 1–6. 15
C NMR
N NMR
N1H
C2H
C3H
C4H
C5H
C6H
C2
C3
C3a
C4
C5
C6
C7a
N1
N7
1.81 1.41 1.56 1.65 1.30 1.44
0.32 0.28 0.28 0.36 0.30 0.33
– – 0.15 – – 0.24
0.10 0.08 0.11 0.25 0.22 0.27
0.09 0.08 – 0.12 0.10 –
0.67 0.67 0.95 0.47 0.48 0.86
1.75 1.51 1.53 2.11 1.89 1.93
1.56 1.70 1.95 1.79 1.93 2.11
3.27 3.12 2.63 3.36 3.22 2.27
3.34 3.29 3.40 4.04 3.98 4.12
1.28 1.08 −0.93 1.45 1.24 −0.88
2.53 2.30 2.29 3.30 3.05 2.76
−1.42 −1.29 − 0.64 −1.00 − 0.85 0.01
2.7 1.6 5.6 1.7 0.4 2.1
− 101.5 −102.4 −102.7 − 115.0 −115.6 − 116.7
dichlorido (1–3) (~−2120 ppm) and oxalato (4–6) (~−1780 ppm) complexes differ (see also Table 1) similarly to the platinum(II) dichlorido and oxalato complexes with unsubstituted 7-azaindole [12]. Although the studied complexes involve differently halogenosubstituted 7-azaindole derivatives, their IR and Raman spectra are very similar and involve the bands (recently reported for 7-azaindole [22]) detected at 1491–1522 and 1580–1598 cm− 1, which are assignable to the C\N, and C\C stretching vibrations, respectively, supporting the presence of the nHaza molecules within the structures of the complexes. The IR and Raman spectra of the studied dichlorido complexes (1–3) showed the bands of the ν(Pt\Cl) vibration at ca. 340 cm− 1, while the bands assignable to ν(Pt\N) were observed at ca. 510 cm− 1 [23]. The spectra of the oxalato complexes (4–6) contain the peaks of the oxalate dianion vibrations, such as νas(C_O) at 1650–1700 cm− 1 and νs(C\O) at 1380–1400 cm− 1. The bands of a PtO2C2 ring deformation appeared in the IR and Raman spectra of 4–6 at ca. 450 cm− 1 (see Supplementary data for more detailed information about IR and Raman spectra). A single crystal X-ray analysis of 1·DMF proved the tetracoordinated Pt(II) atom within a square-planar PtCl2N2 donor set formed by two chlorine ions and by two N7 nitrogen atoms of 3ClHaza molecules, through which they are coordinated to the metal center, while the second nitrogen of the 7-azaindole moiety, i.e. N1, binds a hydrogen atom (Fig. 2). The distortion of the square-planar geometry of 1·DMF can be clearly demonstrated by the values of the N7\Pt1\Cl1 and N7A\Pt1\Cl2 bond angles, which were found to be 176.29(10)° and 177.50(10)°, respectively. The angle formed by the planes fitted through the non-hydrogen atoms of both 3ClHaza molecules equals 88.41(8)°. These planes form the angles of 80.22(6)° and 82.18(6)° with the plane created through the atoms of the PtCl2N2 chromophore. The crystal structure of 1·DMF is stabilized by N\H⋯O hydrogen bonds and other non-covalent contacts of the C\H⋯Cl, C\H⋯C, N\H⋯C, C⋯Cl and C⋯C types (see Supplementary data, Fig. S3 and Table S2).
195
Pt
−2126.2 −2121.6 −2112.4 −1785.7 − 1783.0 −1777.6
evaluation (p b 0.05) showed complex 3 as significantly more active than 1 and 2 against all the MCF7, HOS and LNCaP cells, and complex 1 was found to be significantly more effective against MCF7 as compared with complex 2. It has to be stated that only the dichlorido complexes 1–3 were evaluated as in vitro cytotoxic against the MCF7 and HOS cancer cells. The oxalato analogs of 1–3, i.e. 4–6, were found to be inactive up to the concentration of 10.0 (4), 25.0 (5) and 0.5 (6) μM, primarily because of their limited solubility in the used water/DMF mixture. However, the oxalate dianion was recently predicted [24] and several times reported (e.g. lit. [25–27]) in a biological perspective as suitable leaving group of cytotoxic platinum(II) complexes and the substances involving this group were found to be more effective against cancer cells as compared with compounds involving different leaving groups including chlorine ions. With respect to this fact, it is quite a surprise that the oxalato complexes 4–6 were found to be at least three-times (for 4) and even ten-times (for 6) less effective as compared with their highly in vitro cytotoxic dichlorido analogs 1, and 3, respectively.
3.2. In vitro cytotoxic activity The complexes 1–6 and clinically used platinum-based drug cisplatin (applied within the concentration range of 0.01–50.0 μM, unless their solubility is lower) were studied by an MTT assay for their in vitro cytotoxicity against the MCF7 breast adenocarcinoma, HOS osteosarcoma and LNCaP prostate adenocarcinoma human cancer cell lines. The obtained IC50 values (μM) of 1–3 and cisplatin are given in Table 2 and graphically depicted in Fig. 3. The figure does not contain the results of the remaining substances (4–6), which were found to be cytotoxic inactive in the studied concentration ranges given by their solubility in the used water/DMF mixture, thus their cytotoxicity can be evaluated as >10.0 (for 4), >25.0 (for 5) and >0.5 (for 6) μM as it is given in Table 2. The prepared complexes 1–3 (except for 2 against MCF7) are significantly more cytotoxic in vitro (ANOVA, p b 0.05) against MCF7 and HOS as compared with cisplatin is (see Fig. 3), with the IC50 values (Table 2) at least 2.5-times lower than cisplatin. In the case of the LNCaP cancer cell line, none of the complexes showed significantly higher biological effect as compared to cisplatin. Analogical statistical
Fig. 2. The molecular structure of cis-[PtCl2(3ClHaza)2]·DMF (1·DMF) with the nonhydrogen atoms depicted as thermal ellipsoids at the 50% probability level and given with the atom numbering scheme. The DMF molecule of crystallization is omitted for clarity. Pt1\N7 = 2.020(3) Å, Pt1\N7A = 2.012(3) Å, Pt1\Cl1 = 2.2946(10) Å, Pt1\Cl2 = 2.2982(10) Å, N7\Pt1\N7A = 90.10(12)°, N7\Pt1\Cl2 = 88.91(9)°, N7A\Pt1\Cl1 = 87.76(9)° and Cl1\Pt1\Cl2 = 93.35(4)°.
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63
61
Table 2 The results of the in vitro cytotoxic activity testing of 1–6 and cisplatin against human breast adenocarcinoma (MCF7), osteosarcoma (HOS) and prostate adenocarcinoma (LNCaP) cell lines: cells were treated with tested compounds for 24 h; measurements were performed in triplicate, and cytotoxicity experiment was repeated in three different cell passages; data are expressed as IC50 ± SD (μM). Complex
MCF7
HOS
LNCaP
1 2 3 4 5 6 Cisplatin
3.4 ± 0.3* 8.0 ± 0.9 2.0 ± 0.2* >10.0a >25.0a >0.5a 19.6 ± 4.3
3.8 ± 0.1* 3.9 ± 0.2* 2.5 ± 0.1* >10.0a >25.0a >0.5a 34.2 ± 6.4
3.3 ± 0.7 3.8 ± 1.3 1.5 ± 0.4 nt >25.0a nt 3.8 ± 1.5
nt, not tested. Asterisks (*) indicate significantly different values (p b 0.05) between 1–6 and cisplatin. a IC50 values were not reached due to the limited solubility and they can be expressed as >0.5, >10.0 and >25.0 μM.
In the case of 2 and 5, which both involve the 3IHaza carrier ligands, the oxalato complex is several times less soluble in the used water/ DMF mixture than the dichlorido one. A comparison with the above-mentioned platinum(II) complexes involving 7-azaindole as N-donor ligand, whose biological activity was determined (see Section 1), is not relevant, because they were tested against different cells and they also differ in composition and the type of ligands involved in their structures. Concretely, the dichlorido complex cis-[PtCl2(Haza)(NH3)] (IC50 = 3.6 μM), which is structurally related to 1–3, and its trans isomer (IC50 = 6.0 μM) showed 2.6- and 4.3-times lower biological effect against A2780, respectively, than cisplatin (IC50 = 1.4 μM) [9]. The cytotoxicity in vitro of structurally different [Pt(dmba)(aza)(dmso)] complex, which involves only one 7-azaindole molecule within its structure, was found to be submicromolar against A2780, A2780cis and T47D with IC50 values equalled 0.34 μM, 0.45 μM, and 0.53 μM, respectively, which is 2.6-times (A2780), 24.4-times (A2780cis) and even 69.8times (T47D) more active than cisplatin [10]. Finally, the most active complex 3 involves 5-bromo-7-azaindole within its structure, which allows us to carry out another conclusion regarding the in vitro cytotoxicity of the prepared compounds. It seems that the substitution of the C5 position of the 7-azaindole moiety is more convenient for the resulting biological, particularly in vitro antitumour, activity, but this assumption has to be proved by testing
Fig. 4. Inhibition of RNA synthesis by T7 RNA polymerase on the NdeI/HpaI fragment of pSP73KB plasmid modified by complexes 1, 2, 3 or cisplatin. (A) Autoradiogram of 6% polyacrylamide/8 M urea sequencing gel. Lanes: control — nonmodified template; cisPt, 1, 2, 3 — the template modified by cisplatin, complex 1, 2, and 3 at rb = 0.01, respectively; A, U, G, C — chain terminated marker RNAs. (B) Schematic diagram showing the portion of the nucleotide sequence of the template of the NdeI/HpaI fragment used to monitor inhibition of RNA synthesis by PtII complexes. The arrow indicates the start of the T7 RNA polymerase. (o), major stop signals (from A). The numbers correspond to the nucleotide numbering in the sequence map of pSP73KB plasmid.
of the complexes with identically substituted 7-azaindole ring in its C3 and C5 positions. 3.3. Transcription mapping of DNA adducts
Fig. 3. The results of the in vitro cytotoxicity testing obtained by an MTT assay against human MCF7 breast adenocarcinoma, HOS osteosarcoma and LNCaP prostate adenocarcinoma cell lines for complexes 1–3 (the results determined for 4–6 are not depicted, because their IC50 values were not reached due to the limited solubility). The cells were exposed to the compounds for 24 h. Measurements were performed in triplicate and each cytotoxicity experiment was repeated three times. The given IC50 ± S.D. (μM) values represent an arithmetic mean. The asterisk (*) denotes significant difference (p b 0.05) between the studied complexes and cisplatin.
The mechanism of action of clinically used anticancer platinum drugs, such as cisplatin and its analogs, involves coordination to purine DNA bases. The resulting DNA damages are then encountered and processed by specific cellular proteins, events that determine the ultimate outcome of DNA damage [15,28]. Among these proteins, RNA polymerase has become a major focus of study because of its various roles in processing damaged DNA. Several types of DNA lesions, including cisplatin cross-links, inhibit transcription by blocking RNA polymerase [29,30]. Arrested RNA polymerase not only functions as a damage recognition factor, eliciting transcription-coupled repair, but also triggers programmed cell death, or apoptosis [31]. In order to determine binding of 1, 2 and 3 to natural DNA, sequence specificity of this binding and the effects on RNA synthesis,
62
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63
we also employed in the present work a method which consists in RNA synthesis by T7 RNA polymerase in vitro. This method was used in the same way as in several previous studies of the sequence specificity of various DNA-damaging agents including platinum drugs [21,32–34]. T7 RNA polymerase was chosen to initiate these investigations because it is well characterized, its promoter is clearly defined, and the purified enzyme is commercially available. RNA synthesis by various RNA polymerases including T7 RNA polymerase on DNA templates containing several types of bifunctional adducts of platinum complexes can be prematurely terminated at the level or in the proximity of adducts [21,32]. Importantly, monofunctional DNA adducts of several platinum complexes including cisplatin are unable to terminate RNA synthesis [21,35]. Cutting of pSP73KB DNA [12] by NdeI and HpaI restriction endonucleases yielded a 212-bp fragment (a substantial part of its nucleotide sequence is shown in Fig. 4B). This fragment contained T7 RNA polymerase promoter in the upper strand close to its 3′-end (Fig. 4B). The experiments were carried out using this linear DNA fragment, randomly modified by cisplatin, 1, 2, or 3 at rb = 0.01, for RNA synthesis by T7 RNA polymerase (Fig. 4A, lanes cisPt, 1, 2, and 3, respectively). RNA synthesis on the template modified by the platinum complexes yielded fragments of defined sizes, which indicates that RNA synthesis on these templates was prematurely terminated. These results suggest that new Pt complexes 1, 2 and 3 were able to bind DNA forming adducts capable to stall RNA polymerase. The sequence analysis revealed that the major bands resulting from termination of RNA synthesis by the adducts of all platinum(II) dichlorido complexes involving 7-azaindole halogeno-derivatives were similar to those produced by cisplatin, i.e. they appeared mainly at GG and AG sites. This indicates that 1, 2, and 3 bind to DNA preferentially at the sites similar to preferential DNA binding sites of cisplatin and that they also form similar types of DNA adducts with a similar frequency as cisplatin. These initial data might be also interpreted to mean that the enhanced activity of the analogs of cisplatin 1, 2, or 3 is associated with some features of the damaged DNA and/or its cellular processing not perceptible by the method used in the present work or to an event that is not related to DNA binding. There are many biochemical factors affecting activity of platinum compounds in tumor cells [15,28,36,37]. Further studies are, therefore, warranted to reveal a relative contribution of all potential factors contributing to the enhanced potency of analogs of cisplatin containing 7-azaindole substituted by halogeno-substituents. 4. Conclusions The synthesized platinum(II) dichlorido and oxalato complexes of the composition cis-[PtCl2(3ClHaza)2] (1), cis-[PtCl2(3IHaza)2]·0.75EtOH (2), cis-[PtCl2(5BrHaza)2]·0.5EtOH (3), [Pt(ox)(3ClHaza)2]·0.2EtOH (4), [Pt(ox)(3IHaza)2]·0.75H2O (5) and [Pt(ox)(5BrHaza)2]·0.75H2O (6) involving 7-azaindole-based N-donor ligands (nHaza) were thoroughly characterized (e.g. multinuclear and 2D NMR spectroscopy, ESI mass spectrometry, X-ray analysis). Their geometry was determined as the distorted square-planar with the appropriate nHaza molecules coordinated through their N7 atoms, as proved by a singlecrystal X-ray analysis of the 1·DMF representative. The complexes were screened for their in vitro cytotoxicity against the MCF7, HOS and LNCaP human cancer cell lines. The complexes 1–3 showed significantly higher biological effect (p b 0.05) against MCF7 and HOS as compared with commercially used platinum-based drug cisplatin (except for 2 against MCF7). The IC50 values of the most active complex 3 equalled 2.0 μM (MCF7) and 2.5 μM (HOS), which is about 10times, and 14-times, respectively, higher cytotoxic activity than cisplatin. The mechanism of action of 1–3 was studied by means of transcription inhibition by DNA adducts of 1–3, which is in the case of cisplatin considered to be one of the major routes by which this anticancer drug kills cancer cells [31]. Thus, the results of the
present work also suggest that active platinum complexes 1–3 bind to major pharmacological target of platinum antitumor drugs, DNA, and that the resulting DNA adducts effectively inhibit DNAdependent-RNA-synthesis. As one of the key factors that is important for platinum drug-mediated cytotoxicity is the arrest of RNA synthesis by Pt-DNA adducts [38] then the latter observation may also imply that inhibition of transcription by DNA adducts of 1–3 might significantly contribute to their cytotoxic effects. The reported complexes with high cytotoxicity in vitro (i.e. 1–3) involve halogeno-substituted derivatives of 7-azaindole within their structures in contrast to previously reported platinum(II) dichlorido complex with unsubstituted 7-azaindole, cis-[PtCl2(Haza)2], which was found to be both in vitro [12] and in vivo [11] cytotoxic inactive. The slight modification of 7-azaindole molecule by the halogeno substituents improved the solubility and bio-availability of the platinum(II) complexes with these ligands involved in their structure. With respect to this fact, our future work dealing with the platinum complexes involving 7-azaindole derivatives will be focused on the complexes with a 7-azaindole moiety substituted by more halogeno-substituents or by substituent/s of different types, which is believed to lead to even more biologically effective platinum(II) complexes. Acknowledgments The authors (PS, AP, IP, ZT) gratefully thank the Czech Science Foundation (GAČR P207/11/0841), Operational Program Research and Development for Innovations — European Regional Development Fund (CZ.1.05/2.1.00/03.0058), Operational Program Education for Competitiveness — European Social Fund (CZ.1.07/2.3.00/20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic and by Palacký University in Olomouc (PrF_2012_009). The research of TM and VB was supported by the Czech Science Foundation (GAČR P301/10/0598) and by Palacký University in Olomouc (PrF_2012_026). Appendix A. Supplementary data CCDC 859555 contains the supplementary crystallographic data for cis-[PtCl2(3ClHaza)2]·DMF (1·DMF). Data can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
[email protected]. The results of thermal analysis, ESI-mass spectrometry and IR and Raman spectroscopy, Figure S1 (TG/DTA thermal analysis of 2 and 5), Figure S2 (ESI mass spectrum of 4), Figure S3 (a part of the crystal structure of 1·DMF showing the packing of the molecules within the unit cell and non-covalent contacts), Table S1 (crystal data and structure refinement for 1·DMF) and Table S2 (interatomic parameters of the non-covalent contacts of 1·DMF) are given in Supplementary data. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.jinorgbio.2012.05.006. References [1] L.D. Arnold, X. Chen, H. Dong, A. Garton, M.J. Mulvihill, C.P.S. Smith, G.H. Thomas, T.M. Krulle, J. Wang, US Patent 20070208053 (2007). [2] P.J. Cox, T.N. Majid, S. Amendola, S.D. Deprets, C. Edlin, J.Y.Q. Lai, A.D. Morley, US Patent 20040198737 (2004). [3] M. Klein, US Patent 20110166175 (2011). [4] H.J. Dyke, S. Price, K. Williams, US Patent 20100216768 (2010). [5] G.N. Anilkumar, S.B. Rosenblum, S. Venkatraman, F.G. Njoroge, J.A. Kozlowski, US Patent 20100239527 (2010). [6] D. Simard, Y. Leblanc, C. Berthelette, M.H. Zaghdane, C. Molinaro, Z. Wang, M. Gallant, S. Lau, T. Thao, M. Hamel, R. Stocco, N. Sawyer, S. Sillaots, F. Gervais, R. Houle, J.F. Lévesque, Bioorg. Med. Chem. Lett. 20 (2010) 7212–7215. [7] C. Marminon, A. Pierré, B. Pfeiffer, V. Pérez, S. Léonce, P. Renard, M. Prudhomme, Bioorg. Med. Chem. 11 (2003) 679–687.
P. Štarha et al. / Journal of Inorganic Biochemistry 115 (2012) 57–63 [8] S. Hong, S. Lee, B. Kim, H. Lee, S.-S. Hong, S. Hong, Bioorg. Med. Chem. Lett. 21 (2011) 841–845. [9] E.J. New, C. Roche, R. Madawala, J.Z. Zhang, T.W. Hambley, J. Inorg. Biochem. 103 (2009) 1120–1125. [10] J. Ruiz, V. Rodríguez, C. de Haro, A. Espinosa, J. Pérez, C. Janiak, Dalton Trans. 39 (2010) 3290–3301. [11] R.C. Harrison, C.A. McAuliffe, M.E. Friedman, Inorg. Chim. Acta 92 (1984) 43–46. [12] P. Štarha, J. Marek, Z. Trávníček, Polyhedron 33 (2012) 104–409. [13] E.R. Jamieson, S.J. Lippard, Chem. Rev. 99 (1999) 2467–2498. [14] V. Brabec, Prog. Nucleic Acid Res. Mol. Biol. 71 (2002) 1–68. [15] L. Kelland, Nat. Rev. Cancer 7 (2007) 573–584. [16] P. Štarha, Z. Trávníček, I. Popa, J. Inorg. Biochem. 104 (2010) 639–647. [17] CrysAlis CCD and CrysAlis RED, Version 1.171.33.52, Oxford Diffraction Ltd., England, 2009. [18] G.M. Sheldrick, Acta Crystallogr. A: Found Crystallogr. 64 (2008) 112–122. [19] K. Brandenburg, DIAMOND, Release 3.1f, Crystal Impact GbR, Bonn, Germany, 2006. [20] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock, L. Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, J. Appl. Crystallogr. 41 (2008) 466–470. [21] V. Brabec, M. Leng, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 5345–5349. [22] B. Karthikeyan, Spectrochim. Acta A 64 (2006) 1083–1087. [23] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, fifth ed. Willey-Interscience, New York, 1997. [24] M.J. Cleare, Coord. Chem. Rev. 12 (1974) 349–405.
63
[25] E. Monti, M. Gariboldi, A. Maiocchi, E. Marengo, C. Cassino, E. Gabano, D.J. Osella, J. Med. Chem. 48 (2005) 857–866. [26] V.D. Sen, V.A. Golubev, L.M. Volkova, N.P. Konovalova, J. Inorg. Biochem. 64 (1996) 69–77. [27] U. Mukhopadhyay, J. Thurston, K.H. Whitmire, Z.H. Siddik, A.R. Khokhar, J. Inorg. Biochem. 94 (2003) 179–185. [28] M.A. Fuertes, J. Castilla, C. Alonso, J.M. Perez, Curr. Med. Chem. 10 (2003) 257–266. [29] S. Tornaletti, S.M. Patrick, J.J. Turchi, P.C. Hanawalt, J. Biol. Chem. 278 (2003) 35791–35797. [30] C. Cullinane, S.J. Mazur, J.M. Essigmann, D.R. Phillips, V.A. Bohr, Biochemistry 38 (1999) 6204–6212. [31] Y. Jung, S.J. Lippard, J. Biol. Chem. 278 (2003) 52084–52092. [32] O. Novakova, O. Vrana, V.I. Kiseleva, V. Brabec, Eur. J. Biochem. 228 (1995) 616–624. [33] N. Sasaki, M. Izawa, M. Watahiki, K. Ozawa, T. Tanaka, Y. Yoneda, S. Matsuura, P. Carninci, M. Muramatsu, Y. Okazaki, Y. Hayashizaki, Proc. Natl. Acad. Sci. U. S. A 95 (1998) 3455–3460. [34] L.C. Perrin, C. Cullinane, W.D. McFadyen, D.R. Phillips, Anticancer Drug Des. 14 (1999) 243–252. [35] M.A. Lemaire, A. Schwartz, A.R. Rahmouni, M. Leng, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 1982–1985. [36] D. Wang, S.J. Lippard, Nat. Rev. Drug Discov. 4 (2005) 307–320. [37] Y. Jung, S.J. Lippard, Chem. Rev. 107 (2007) 1387–1407. [38] R.C. Todd, S.J. Lippard, Metallomics 1 (2009) 280–291.
PŘÍLOHA 3 Muchová, T.; Prachařová, J.; Štarha, P.; Olivová, R.; Vrána, O.; Benešová, B.; Kašpárková, J.; Trávníček, Z.; Brabec, V. Insight into the toxic effects of cis-dichloridoplatinum(II) complexes containing 7azaindole halogeno derivatives in tumor cells J. Biol. Inorg. Chem. 18 (2013) 579–589
J Biol Inorg Chem (2013) 18:579–589 DOI 10.1007/s00775-013-1003-7
ORIGINAL PAPER
Insight into the toxic effects of cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives in tumor cells Tereza Muchova • Jitka Pracharova • Pavel Starha • Radana Olivova • Oldrich Vrana • Barbora Benesova • Jana Kasparkova • Zdenek Travnicek • Viktor Brabec
Received: 5 March 2013 / Accepted: 27 April 2013 / Published online: 15 May 2013 Ó SBIC 2013
Abstract The cisplatin analogues cis-[PtCl2(3ClHaza)2] (1) and cis-[PtCl2(3IHaza)2] (2) (3ClHaza and 3IHaza are 3-chloro-7-azaindole and 3-iodo-7-azaindole, respectively) are quite toxic to ovarian tumor cells, with moderately better IC50 values than for cisplatin in the cisplatin-sensitive cell line A2780. We investigated potential factors which might be involved in the mechanism underlying the cytotoxic effects of 1 and 2 and compared these factors with those involved in the mechanism underlying the effects of conventional cisplatin. Our data indicate that the higher cytotoxicity of 1 and 2 originates mainly from their efficient cellular accumulation, different effects at the level of cell cycle regulation, and reduced propensity for DNA adduct repair. Studies of their reactivity toward cellular components reveal efficient binding to DNA, which is typically required for an active platinum drug. Further results suggest that 1 and 2 are
Electronic supplementary material The online version of this article (doi:10.1007/s00775-013-1003-7) contains supplementary material, which is available to authorized users. T. Muchova J. Pracharova R. Olivova B. Benesova J. Kasparkova Department of Biophysics, Faculty of Sciences, Palacky University, 17. listopadu 12, 77146 Olomouc, Czech Republic
capable of circumventing resistance to cisplatin induced by alterations in cellular accumulation and DNA repair. Hence, the latter two factors appear to be responsible for differences in the toxicity of 1 or 2, and cisplatin in tumor cells. The results of this work reinforce the idea that direct analogues of conventional cisplatin-containing halogenosubstituted 7-azaindoles offer much promise for the design of novel therapeutic agents. Keywords Platinum drugs Cytotoxicity Cellular uptake Cell cycle DNA damage DNA repair Abbreviations 3ClHaza 3-Chloro-7-azaindole 3IHaza 3-Iodo-7-azaindole CT Calf thymus DMF N,N0 -Dimethylformamide EtBr Ethidium bromide FAAS Flameless atomic absorption spectrometry GSH Glutathione Compound concentration that produces 50 % IC50 cell growth inhibition ICP-MS Inductively coupled plasma mass spectroscopy MTT 3-(4,5-Dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide SD Standard deviation
P. Starha Z. Travnicek Department of Inorganic Chemistry, Faculty of Science, Regional Centre of Advanced Technologies and Materials, Palacky University, 17. listopadu 12, 77146 Olomouc, Czech Republic
Introduction
O. Vrana V. Brabec (&) Institute of Biophysics, Academy of Sciences of the Czech Republic, v.v.i., Kralovopolska 135, 61265 Brno, Czech Republic e-mail:
[email protected]
The design of new potential platinum drugs to overcome tumor resistance to conventional cisplatin is an active area of bioinorganic chemistry and molecular pharmacology. This is because clinical use of platinum-based drugs is
123
580
limited by side effects and poor activity in certain types of cancer resulting from acquired or intrinsic resistance [1, 2]. These limitations have given impetus to the design and synthesis of new platinum-based chemotherapeutics with improved pharmacological properties. Hence, the search continues for platinum compounds with novel preclinical properties, such as activity in cisplatin-resistant cells or a pattern of cytotoxicity significantly different from that of cisplatin and its clinically used analogues against a panel of cell lines of various origins [3]. One approach that we have used to circumvent the shortcomings of conventional classic bifunctional platinum-based drugs, such as cisplatin, carboplatin, and oxaliplatin, is to prepare and characterize dichloridoplatinum(II) and oxalatoplatinum(II) complexes of the general formulas cis-[PtCl2(nHaza)2] (Fig. 1) and [Pt(ox)(nHaza)2] (H2ox is oxalic acid) involving 7-azaindole halogeno derivatives (nHaza) [4]. These new complexes obey the originally devised structure–activity relationships for antitumor candidates in the platinum family: such compounds should be neutral PtII species with two am(m)ine ligands or one bidentate chelating diamine, and should have leaving ligands that can be replaced during aquation reactions [5]. The dichloridoplatinum(II) complexes, cis-[PtCl2(nHaza)2] [nHaza is 3-chloro-7-azaindole (3ClHaza) for complex 1, 3-iodo-7-azaindole (3IHaza) for complex 2 or 5-bromo-7azaindole], previously demonstrated [4] promising in vitro cytotoxicity against the HOS (osteosarcoma), MCF7 (breast adenocarcinoma), and LNCaP (prostate adenocarcinoma) human cancer cell lines, i.e., against the tumor cell lines inherently resistant to cisplatin. Also importantly, it was the first time that 7-azaindole halogeno derivatives had been used as ligands of transition metal complexes. Preliminary results have also suggested that the cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives effectively bind, in a cell-free medium, to DNA, which is considered a major pharmacological target of antitumor platinum-based drugs [6, 7], and that some features of the DNA binding mode of
Fig. 1 The PtII complexes used in this work
123
J Biol Inorg Chem (2013) 18:579–589
these new PtII complexes are similar to those of cisplatin [4]. The mode of action of cisplatin and its direct analogues is a multistep process which includes (1) cell accumulation, (2) drug activation or inactivation by sulfur-containing compounds, (3) DNA binding, and (4) cellular responses to the DNA damage, including DNA adduct repair, perturbed cell cycle, and the inhibition of apoptosis or necrosis [2, 8]. The cellular uptake, DNA binding in cells, reactivity with glutathione (GSH), DNA repair synthesis, perturbation of the cell cycle, and inhibition of apoptosis and necrosis were investigated to provide insight into the potency of cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives. Complexes 1 and 2 were selected for this study.
Materials and methods Starting materials and reagents Cisplatin, N,N0 -dimethylformamide (DMF), propidium iodide, and glutathione (GSH) were obtained from SigmaAldrich (Prague, Czech Republic). Compounds 1 and 2 were synthesized and characterized as described previously [4]. Stock solutions of 1, 2, and cisplatin were prepared at a concentration of 5 9 10-2 M in DMF, stored at 4 °C in the dark, and diluted with water to the appropriate concentration just before use. The final concentration of DMF was less than 0.5 %. Stock solutions of platinum complexes for the cytotoxicity and cellular uptake studies were also prepared in DMF and were used immediately after dissolution. The concentrations of platinum in the stock solutions and after dilution with water were determined by flameless atomic absorption spectrometry (FAAS). Calf thymus (CT) DNA (42 % G ? C, mean molecular weight approximately 2 9 107) was also prepared and characterized as described previously [9, 10]. Plasmid pUC19 (2,686 bp) was isolated according to standard procedures. Restriction endonucleases, the Klenow fragment from DNA polymerase I, and plasmid pBR322 (4,361 bp) were purchased from New England Biolabs. Ethidium bromide (EtBr) was from Merck (Darmstadt, Germany). Agarose was from FMC BioProducts (Rockland, ME, USA). Proteinase K and ATP were from Boehringer (Mannheim, Germany). 3-(4,5Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was from Calbiochem (Darmstadt, Germany). Radioactive products were obtained from MP Biomedicals (Irvine, CA, USA). RPMI 1640 medium, fetal bovine serum, and trypsin/EDTA were from PAA (Pasching, Austria). Gentamicin was from Serva (Heidelberg, Germany). The cell-free extract was prepared from the repairproficient HeLa S3 cell line as reported previously [11].
J Biol Inorg Chem (2013) 18:579–589
In vitro growth inhibition assay Human ovarian carcinoma cisplatin-sensitive A2780 cells and cisplatin-resistant A2780cisR cells (cisplatin-resistant variant of A2780 cells) were kindly supplied by B. Keppler, University of Vienna (Austria). The A2780cisR cells were grown in RPMI 1640 medium supplemented with gentamicin (50 lg mL-1) and heat-inactivated fetal bovine serum (10 %). The acquired resistance of A2780cisR cells was maintained by supplementing the medium with cisplatin (1 lM) every second passage. The cells were cultured in a humidified incubator at 37 °C in a 5 % CO2 atmosphere and subcultured two or three times a week with an appropriate plating density. Stock solutions were freshly prepared in DMF immediately prior to testing. The cells were seeded in 96-well tissue culture plates at a density of 104 cells per well in 100 lL of medium. After overnight incubation (16 h), the cells were treated with the compounds at final concentrations in the range of 0–100 lM in a final volume of 200 lL per well. The final DMF concentration in all wells was 0.1 %, which was shown not to affect cell growth. The cell lines were incubated for 24 or 72 h with the platinum compounds, and cell death was evaluated using a system based on the tetrazolium compound MTT in the same way as described previously [12, 13]. Briefly, 10 lL of a freshly diluted MTT solution (2.5 mg mL-1) was added to each well, and the plate was incubated at 37 °C in a humidified 5 % CO2 atmosphere for 4 h. At the end of the incubation period, the medium was removed, and the formazan product was dissolved in 100 lL of dimethyl sulfoxide. Cell viability was evaluated by measurement of the absorbance at 570 nm, using a Tecan Infinite M200 absorbance reader (Schoeller). Compound concentrations that produce 50 % cell growth inhibition (IC50) were calculated from curves constructed by plotting cell survival (%) versus drug concentration (lM). All experiments were done in triplicate. The reading values were converted to the percentage of the control (percentage cell survival). Cytotoxic effects were expressed as IC50. Detection of apoptosis and necrosis The cell death detection ELISA plus kit (Roche Molecular Biochemicals, Mannheim, Germany) was used as an indicator of apoptosis and necrosis [14]. In this assay, internucleosomal DNA fragmentation was quantitatively assayed by antibody-mediated capture and detection of cytoplasmic mononucleosome- and oligonucleosomeassociated histone–DNA complexes. Briefly, after centrifugation (200g), 20 lL of the supernatant was used in the ELISA for detection of necrosis. A2780 cells were resuspended in 200 lL of the lysis buffer supplied by the manufacturer and incubated for 30 min at room
581
temperature. After the nuclei (200g, 10 min) had been pelleted, 20 lL of the supernatant (cytoplasmic fraction) was used in the ELISA for detection of apoptosis following the manufacturer’s standard protocol. Following incubation with peroxidase substrate for 20 min, the absorbance was determined at 405–490 nm (reference wavelength) with a microplate reader (Tecan Sunrise absorbance reader, Schoeller). Signals from wells containing the substrate only were subtracted as background. Other parts of this assay and data analysis were performed according to the manufacturer’s instructions. Cell cycle analysis After 24 h, floating cells (A2780) were collected, and attached cells were harvested by trypsinization (trypsin/ EDTA in phosphate-buffered saline). Total cells (floating and attached) were washed twice in phosphate-buffered saline (4 °C), fixed in 70 % ethanol, and stored at -20 °C. Cell pellets were subsequently rinsed with phosphate-buffered saline and sediment was stained with a solution of propidium iodide (50 lg mL-1, Sigma) supplemented with RNase A (10 lg mL-1, Qiagen) for 30 min at 25 °C in the dark. DNA content was measured using flow cytometry (Cell Lab Quanta SC MPL, Beckman Coulter). The percentages of cells in the individual cell cycle phases were analyzed using Multicycle AV for Windows (Verity Software House). Cellular platinum complex accumulation Cellular accumulation of 1, 2, and cisplatin was measured in A2780 cells. The cells were seeded in 100-mm tissue culture dishes (30,000 cells per square centimeter). After overnight incubation, the cells were treated with the PtII complex (10 lM) for 5 or 24 h [the concentration was verified by the measurement of platinum in the growing medium by FAAS or inductively coupled plasma mass spectroscopy (ICP-MS)]. The attached cells were washed twice with phosphate-buffered saline (4 °C), and the pellet was stored at -80 °C. The pellets were digested by a highpressure microwave digestion system (MARS5, CEM) with HCl to give a fully homogenized solution, and final platinum content was determined by FAAS. The cellular platinum uptake values were corrected for adsorption effects [15]. For other details, see ‘‘Results.’’ All experiments were performed in triplicate. DNA platination in cells exposed to platinum complexes A2780 cells grown to near confluence were exposed to 10 lM complex 1, 2, or cisplatin for 5 or 24 h. After the
123
582
incubation, the cells were trypsinized and washed twice in ice-cold phosphate-buffered saline. Cells were then lysed in DNAzol (MRC) supplemented with RNase A (100 lg mL-1). The genomic DNA was precipitated from the lysate with ethanol, dried, and resuspended in water. The DNA content in each sample was determined by UV spectrophotometry. To avoid the effect of high DNA concentration on ICP-MS detection of platinum in the samples, the DNA samples were digested in the presence of hydrochloric acid (11 M) using a high-pressure microwave mineralization system (MARS5, CEM). Experiments were performed in triplicate, and the values are the mean ± the standard deviation (SD). Interstrand DNA cross-linking in a cell-free medium Complex 1, complex 2, and cisplatin were incubated for 24 h with 0.5 lg of a linear 2,686-bp fragment of pUC19 plasmid linearized by EcoRI. The linear fragment was first 30 -end-labeled by means of the Klenow fragment of DNA polymerase I in the presence of [a-32P]dATP. The platinated samples were analyzed for DNA interstrand crosslinks by previously published procedures [16, 17]. The number of interstrand cross-links was analyzed by electrophoresis under denaturing conditions on alkaline agarose gel (1 %). After the electrophoresis had been completed, the intensities of the bands corresponding to single strands of DNA and interstrand cross-linked duplex were quantified. The frequency of interstrand cross-links was calculated as ICL/Pt (%) = XL/5372rb (the DNA fragment contained 5,372 nucleotide residues), where ICL/Pt (%) is the number of interstrand cross-links per adduct multiplied by 100, and XL is the number of interstrand cross-links per molecule of the linearized DNA duplex, and was calculated assuming a Poisson distribution of the interstrand crosslinks as XL = -ln A, where A is the fraction of molecules running as a band corresponding to the non-cross-linked DNA. Characterization of DNA adducts by thiourea Incorporation of [14C]thiourea into DNA under controlled conditions (8.7 9 10-4 M, 10 min, 25 °C [18]) was used to quantitate platinum–DNA monofunctional adducts of 1, 2, or cisplatin. The measurements were performed in 10 mM NaClO4 at 25 °C. The molar ratio of free PtII complex to nucleotide phosphates at the onset of incubation with CT DNA, ri, was 0.05 [the DNA concentration was 0.24 mg mL-1 (0.75 mM related to the phosphorus content), and the concentration of PtII complexes was 3.75 9 10-5 M]. Samples (120 lL) were withdrawn after 24 h, and each sample was divided into two aliquots (60 lL). In one aliquot, conversion of the monofunctional
123
J Biol Inorg Chem (2013) 18:579–589
adducts to bifunctional cross-links was blocked by addition of NaCl (the final concentration was 0.15 M) and quick cooling to -20 °C. The unbound PtII complex was removed from these samples by centrifugation through a Sephadex G50 coarse column, and the molar ratio of covalently bound molecules of PtII complexes per nucleotide residue, rb, was determined by FAAS. In the other aliquot, the conversion of the adducts was blocked by addition of [14C]thiourea (the final concentration was 8.7 9 10-4 M, and the specific activity was 2.5 mCi mmol-1) and NaCl (the final concentration was 0.15 M) so that the final volume of these samples was 1.0 mL. The samples were further incubated at 25 °C for 10 min and subsequently were layered on Millipore filters (the diameter of pores was 0.1 lm); the unreacted thiourea and complexes formed between unbound PtII complexes and thiourea were removed by washing the filters with 15 mL of 5 % (v/v) trichloroacetic acid.The filters were dried under an infrared lamp and transferred to glass tubes, to which 5 mL of toluene scintillator was added. The radioactivity was measured with a TriCarb 2800 TR liquid scintillation analyzer (PerkinElmer) (2 9 2 min). The content of free coordination sites in DNA adducts of PtII complexes not involved in the binding to DNA was determined as the amount (%) of radioactive thiourea bound to platinated DNA; the concentration of thiourea corresponding to twofold concentration of PtII complexes (having two potential DNA binding sites) bound to DNA in each sample determined by FAAS (rb) was taken as 100 %. It was also verified that, at 8.7 9 10-4 M thiourea, complete saturation of monofunctional or bifunctional adducts was obtained with no apparent reversal of platination [19, 20]. Reactions of PtII complexes with glutathione Reactions of 1, 2, and cisplatin were investigated by monitoring the UV absorbance at 260 nm of solutions containing the platinum complex and GSH as described in previous work [21, 22] with minor modifications. The PtII complexes (30 lM) were mixed with GSH (15 mM) at 37 °C in 4.6 mM NaCl plus 100 mM NaClO4, pH 6.0 and 50 % DMF. Reactions were initiated by mixing the PtII complex with the buffer, followed by immediate addition of GSH. The experiments were performed in triplicate. The absorbance at 260 nm reflects the presence of platinum–sulfur and disulfide bonds. The kinetic data were fitted by nonlinear regression (GraphPad Prism) to one-phase or two-phase exponential association. The decision that a fit to two-phase exponential association was more appropriate for each dependence was made by comparing the fits of two equations by using an F test (GrahPad Prism).
J Biol Inorg Chem (2013) 18:579–589
DNA repair synthesis by human cell extract Repair DNA synthesis of cell-free extracts was assayed using pUC19 plasmid. Each reaction mixture of 50 lL contained unmodified pBR322 (500 ng) and unmodified or platinated pUC19 (500 ng), ATP (2 mM), KCl (30 mM), creatine phosphokinase (rabbit muscle) (0.05 mg mL-1), dGTP (20 mM), dCTP (20 mM), dTTP (20 mM), dATP (8 mM), [a-32P]dATP (74 kBq) in a buffer composed of N-(2-hydroxyethyl)piperazine-N-ethanesulfonic acid–KOH (40 mM, pH 7.5), MgCl2 (5 mM), dithiothreitol (0.5 mM), creatine phosphate (22 mM), bovine serum albumin (1.4 mg mL-1), and cell-free extract (150 lg). Reaction mixtures were incubated for 3 h at 30 °C and the reactions were terminated by adding EDTA to a final concentration of 20 mM, sodium dodecyl sulfate to 0.6 %, and proteinase K to 250 lg mL-1 and then incubating the mixture for 30 min at 45 °C. The products were extracted with 1 vol of 1:1 phenol–chloroform. The DNA was precipitated from the aqueous layer by the addition of 0.02 vol NaCl (5 M), glycogen (5 mg), and 2.5 vol ethanol. After 20 min of incubation on dry ice and centrifugation at 12,000g for 30 min at 4 °C, the pellet was washed with 0.5 mL of 70 % ethanol and dried in a vacuum centrifuge. DNA was finally linearized by SspI before electrophoresis on a 1 % agarose gel. Gels were stained with EtBr for photodocumentation, and the radioactivity associated with the bands was quantitated using a FUJIFILM BAS 2500 bioimaging analyzer with AIDA Image Analyzer (raytest Isotopenmessgera¨te, Straubenhardt, Germany). Experiments were performed in duplicate. Other physical methods Absorption spectra were measured with a Beckman 7400 DU spectrophotometer equipped with a thermoelectrically controlled cell holder. The FAAS measurements were conducted with a Varian AA240Z Zeeman atomic absorption spectrometer equipped with a GTA 120 graphite tube atomizer. The analysis with the aid of ICP-MS was performed using an Agilent 7500 instrument (Agilent, Japan). The gels were visualized with a FUJIFILM BAS 2500 bioimaging analyzer, and the radioactivity associated with the bands was quantified with AIDA Image Analyzer. Statistical evaluation of the untreated control cells and drug-treated cells was performed using Student’s t test. If not stated otherwise, a probability of 0.05 or less was deemed statistically significant.
583
ovarian carcinoma cell line, which is commonly used to test the cytotoxic activity of cisplatin analogues. The tumor cell line was incubated for 24 or 72 h with the platinum compounds and cell survival in cultures treated with the platinum compounds was evaluated as described in ‘‘Materials and methods.’’ The IC50 values (mean ± SD of three independent experiments) obtained for 1, 2, and cisplatin after 24 h were 3.1 ± 0.3, 4.2 ± 0.9, and 14.9 ± 1.1 lM, respectively; the IC50 values obtained after 72 h were 1.6 ± 0.3, 1.6 ± 0.3, and 3.8 ± 0.5 lM, respectively. These results show that both complex 1 and complex 2 were markedly more cytotoxic than cisplatin in cisplatin-sensitive A2780 cells after 24 h treatment, whereas the longer treatment (72 h) resulted in only moderately higher cytotoxicity of complexes 1 and 2. Cell death detection To analyze the characteristics of the cell death induced by complexes 1 and 2 and to identify whether cell death is related to apoptotic or necrotic processes, the level of apoptosis and necrosis induced by these drugs and cisplatin at a concentration of 3 lM over 24 h was quantified by a specific ELISA kit. This analysis allows the appearance and relative amounts of cytoplasmic histone-associated DNA fragments (mononucleosomes and oligonucleosomes) to be measured after the induction of apoptosis or when these fragments are released from necrotic cells. Figure 2a shows DNA fragmentation induced by 1, 2, and cisplatin in A2780 cells as a result of apoptotic processes. These results demonstrate that treatment with these antitumor agents led to apoptotic events in the A2780 cell line. Importantly, complexes 1 and 2 induced a significantly higher level of DNA fragmentation due to apoptosis in comparison with cisplatin in A2780 cells.
Results Cytotoxicity The cytotoxic activity of complexes 1 and 2 was determined against the A2780 (cisplatin-sensitive) human
Fig. 2 Effects of 1, 2, and cisplatin on activation of the apoptotic pathway (a) and necrosis (b) in A2780 cells determined by DNAfragmentation ELISAs. Cells were treated for 24 h with 1, 2, or cisplatin at a concentration of 3 lM. The results are expressed as the mean of two independent experiments with duplicate runs
123
584
Fig. 3 Effects of 1, 2, and cisplatin on cell cycle distribution. Untreated (control) A2780 cells or A2780 cells treated at a concentration of 3 lM (a) or 5 lM (b) for 24 h were harvested, fixed, stained with propidium iodide, and assessed for cell cycle distribution by fluorescence-activated cell sorting. The estimated percentages of A2780 cells in different phases of the cell cycle are indicated. The results are expressed as the mean ± the standard deviation (SD) of three independent experiments with duplicate runs. An asterisk denotes a significant difference (p \ 0.05) from the untreated control, and a hash denotes a significant difference (p \ 0.05) between 1 or 2 and cisplatin
A similar procedure was used to detect the extent of necrosis induced by 1, 2, or cisplatin. Importantly, the level of necrosis was significantly lower compared with that of apoptosis triggered by 1, 2, and cisplatin (Fig. 2b). Cell cycle analysis The status of the cell cycle for cells treated with complexes 1 and 2 and for comparative purposes also with cisplatin was analyzed. The analysis of cell cycle perturbation was performed using A2780 cells exposed to 1, 2, or cisplatin at concentrations of 3 or 5 lM for 24 h (Fig. 3). An evaluation of the effects on A2780 cells produced by complexes 1 and 2 in comparison with untreated control A2780 cells showed several significant differences in cell cycle modulation and these became more pronounced with increased
123
J Biol Inorg Chem (2013) 18:579–589
concentration of the drug. Exposure of A2780 cells to 1, 2, or cisplatin caused the appearance of a population in the sub-G1 region of the profile, where apoptotic and necrotic cells are found. The appearance of a sub-G1 peak is consistent with the onset of internucleosomal DNA cleavage in late apoptosis [23]. Importantly, the efficiency of the compounds tested to cause the appearance of a population in the sub-G1 region of the profile was different: the sub-G1 population for cells treated with complexes 1 and 2 was significantly greater than that for cells treated with cisplatin (Fig. 3). This result is consistent with the morphological changes of A2780 cells observed by light microscopy (Figs. S1, S2). Treatment of these cells with complexes 1 or 2 for 24 h resulted in a high number of detached (apoptotic/ necrotic) cells, whereas most of the cells treated with cisplatin under the same conditions were still growing and dividing. Also, importantly, treatment with complex 1 or complex 2 affected the G0/G1 populations much less than treatment with cisplatin, which markedly decreased the G0/ G1 populations. Similarly, treatment with complexes 1 and 2 somewhat decreased the peaks corresponding to the S phase in contrast to the peaks corresponding to this phase in case of treatment with cisplatin, which were only affected slightly (Fig. 3). In addition, cisplatin caused an accumulation of cells in the G2 phase in contrast to treatment with complexes 1 and 2, which decreased the G2 populations considerably less (see Fig. 3). Thus, different effects were observed for the cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives and cisplatin. The fact that complexes 1 and 2 and cisplatin had different effects on cell cycle progression suggests that substitution of the NH3 groups in cisplatin by 3ClHaza or 3IHaza changed the mechanism of action of the parent platinum drug. Cellular accumulation The factor that is usually thought to contribute to drug cytotoxicity is cellular accumulation. To examine the accumulation of complexes 1 and 2, the cellular levels of these compounds were measured after 5 and 24 h exposure of the A2780 cells to the drugs at a concentration of 10 lM. The accumulation of complexes 1 and 2 in A2780 cells was, after 5 and 24 h exposure, approximately 12-fold to 15-fold and 45-fold to 46-fold greater, respectively, than that of cisplatin (Table 1). DNA-bound platinum in cells exposed to the platinum complexes Platinum levels on nuclear DNA were determined after the exposure of A2780 cells to 10 lM 1, 2, or cisplatin for 5 or 24 h. The levels of platinum on DNA were determined by
J Biol Inorg Chem (2013) 18:579–589
585
Table 1 Accumulation of platinum complexes in A2780 cells and platinum content of DNA isolated from A2780 cells exposed to complex 1, complex 2, or cisplatin Complex 1
Complex 2
Cisplatin
Accumulation after 5 ha
0.705 ± 0.130
0.842 ± 0.105
0.056 ± 0.002
Accumulation after 24 ha
3.732 ± 0.096
3.836 ± 0.104
0.083 ± 0.007
Pt content of DNA after 5 hb
0.72 ± 0.14
0.75 ± 0.20
0.16 ± 0.04
Pt content of DNA after 24 hb
4.70 ± 0.24
5.03 ± 0.13
0.34 ± 0.08
For the structures of the complexes, see Fig. 1. Cells were exposed to 1, 2, or cisplatin (10 lM) for 5 or 24 h. The results are expressed as the mean ± the standard deviation of three independent experiments. a
The accumulation of Pt complexes is in nanomoles of Pt per 1 9 106 cells.
b
The Pt content of DNA is in picmoles of Pt complex per microgram of DNA.
ICP-MS (Table 1). The platinum content of DNA for the A2780 cells treated with 1 or 2 for 5 or 24 h was approximately 4.5-fold to 4.7-fold or 13.8-fold to 14.8-fold greater, respectively, than that for the cells treated with cisplatin. DNA binding in cell-free media The platinum complexes were incubated with CT DNA at various ri values (ri is defined as the molar ratio of free platinum complex to nucleotide phosphate) in 10 mM NaClO4 at 37 °C in the dark, aliquots withdrawn after 24 h were quickly cooled in an ice bath, and then the free (unbound) platinum compound was removed by gel filtration chromatography using Sephadex G50 columns. The content of platinum and the concentration of DNA in these DNA samples was determined by FAAS and absorption spectrophotometry. Hence, it was possible to prepare samples of DNA modified by these PtII complexes at a preselected value of rb (the number of molecules of the platinum complex bound per nucleotide residue). Samples of DNA modified by 1, 2, or cisplatin and analyzed further by biophysical or biochemical methods were prepared in NaClO4 (10 mM) at 37 °C. After 24 h of the reaction of DNA with the complex, the samples were precipitated in ethanol and dissolved in the medium necessary for a particular analysis, and the rb value in an aliquot of this sample was checked by FAAS. In this way, all analyses performed in cell-free media described in this article were performed in the absence of unbound (free) platinum complex. Bifunctional platinum compounds, which coordinate base residues in DNA, form various types of interstrand
and intrastrand cross-links and monofunctional adducts. Considerable evidence suggests that the antitumor efficacy of bifunctional platinum compounds is the result of the formation of these lesions, but their relative efficacy remains unknown. Therefore, we decided to quantitate first the interstrand cross-linking efficiency of complexes 1 and 2 in linearized pUC19 plasmid. This plasmid DNA was linearized by EcoRI (EcoRI cuts only once within pUC19 plasmid) and globally modified by 1, 2, or cisplatin. The samples were analyzed for the interstrand cross-links by agarose gel electrophoresis under denaturing conditions [17]. The intensity of the more slowly migrating band increased with increasing level of the platination (Fig. 4). The radioactivity associated with the individual bands in each lane was measured to obtain estimates of the fraction of non-cross-linked or crosslinked DNA under each condition. The DNA interstrand cross-linking efficiency of complexes 1 and 2 was 5 %. This implies that interstrand cross-links represent only a minor portion of the adducts formed in DNA by complexes 1 and 2, i.e., in this respect, complexes 1 and 2 behave like the parent cisplatin, whose interstrand crosslinking efficiency is 6 % under identical conditions (Fig. 4) [17]. Previous studies [18–20, 24] have shown that under the appropriate conditions thiourea quantitatively reacts with DNA monoadducts of bifunctional PtII complexes without displacing platinum from the DNA. In other words, during the initial period of the reaction of DNA with bifunctional PtII complexes, when a significant proportion of the molecules are bound monofunctionally, the other coordination site can be blocked by thiourea [20]. This is so because, although the Pt–N bond has higher thermodynamic stability than the Pt–S (thioether) bond, thioether sulfur is kinetically more favorable than the guanine nitrogen when binding to PtII drugs. Thus, incorporation of [14C]thiourea into DNA under controlled conditions [18] (see ‘‘Materials
Fig. 4 The formation of interstrand cross-links by 1, 2, and cisplatin in pUC19 DNA. Autoradiograms of the denaturing 1 % agarose gels of linearized DNA which was 30 -end-labeled; the interstrand crosslinked DNA appears as the top bands (ICLs) migrating on the gels more slowly than the single-stranded DNA (ss) (contained in the bottom bands). Plasmid linearized by EcoRI was not modified (lane C), modified by complex 1 (lane 1), modified by complex 2 (lane 2), or modified by cisplatin (lane 3). The rb value was 0.001
123
586
and methods’’) was used to quantitate platinum–DNA monofunctional adducts of complexes 1 and 2. The adducts in which complexes 1 and 2 still possessed free coordination sites not involved in binding to DNA (which could be blocked by thiourea, presumably in monofunctional lesions) were quantitated after 24 h incubation of complex 1 or complex 2 with CT DNA at rb = 0.075 by incorporation of [14C]thiourea. The complete protocol is described in ‘‘Materials and methods,’’ but it should be emphasized that these experiments were performed in the absence of any free (unbound to DNA) platinum complex. The proportions of free coordination sites in 1, 2, and cisplatin not involved in the binding to DNA calculated in this manner were 12, 10, and 2 %, respectively. The resulting data indicate that after 24 h of the reaction of DNA with complexes 1 and 2, approximately 88 and 90 % of adducts were already bifunctional cross-linked adducts, whereas the reactions of DNA with cisplatin resulted in 98 % bifunctional adducts, in agreement with previously published results [7]. Reactions with glutathione PtII compounds have a strong thermodynamic preference for binding to sulfur-donor ligands, such as thiolates [25, 26]; hence, before PtII drugs reach the DNA in the nucleus of tumor cells, or even after they bind to DNA, they may still react with various sulfur-containing compounds [27, 28]. These reactions are generally believed to play a role in mechanisms underlying tumor resistance to platinum compounds, their inactivation, and side effects. Therefore, interest in the interactions of platinum antitumor drugs with sulfur-containing molecules of biological significance has recently increased markedly [28]. In this work, we investigated, using UV absorption spectrophotometry, irreversible binding of GSH to complexes 1 and 2 in comparison with cisplatin following the procedure outlined by Dabrowiak et al. [21], which was slightly modified as described in ‘‘Materials and methods.’’ Complexes 1 and 2 were incubated with GSH at a ratio of thiol to drug of 500:1, which represents the physiologically relevant value [21]. Figure 5 shows the UV absorbance (at 260 nm) of the platinum complexes and GSH as a function of time, with the absorbances of GSH and the platinum complex alone subtracted. To establish the rate of the initial reaction with respect to the platinum complex, each difference curve was fitted by nonlinear regression (GraphPad Prism) to the following equation: Id = C ? A1exp(-b1t) ? A2exp(-b2t) (A1, A2, b1, b2, and C are constants, and t is the time of the reaction). The initial slope (Sin) was calculated as -(A1b1 ? A2b2) [21]. Sin values of 0.00226 ± 0.00009 and 0.00222 ± 0.00009 min-1 were calculated for reaction of
123
J Biol Inorg Chem (2013) 18:579–589
Fig. 5 UV absorbance associated with the reaction of 1, 2, and cisplatin with glutathione (GSH). The absorbance at 260 nm is shown as a function of time for a 240-min incubation at 37 °C of the platinum complexes at a concentration of 30 lM with GSH (15 mM) in a medium of NaCl (4.6 mM) plus NaClO4 (100 mM, pH 6.0) in the dark at 37 °C. The curves represent the absorbances (260 mm) of solutions containing the platinum complex plus GSH from which absorbances of GSH and the platinum complex alone were subtracted. Curve 1 complex 1, curve 2 complex 2, curve 3 cisplatin
GSH with complexes 1 and 2, respectively, and 0.00225 ± 0.00004 min-1 was calculated for the reaction of GSH with cisplatin. Thus, complexes 1 and 2 reacted with GSH with a similar rate as cisplatin. DNA repair synthesis by human cell extract DNA repair is another key factor which significantly reduces the number of platinum adducts on DNA in cells and thereby may significantly contribute to their biological activity [29]. Therefore, the DNA repair efficiency in pUC19 plasmid (2,686 bp) globally modified by 1, 2, or cisplatin at rb = 0.07 was tested using cell-free extract of repair-proficient HeLa S3 cells. Untreated plasmid pBR322 (4,361 bp) was also included in each reaction to monitor damage-independent nucleotide incorporation. Repair activity was monitored by measurement of the amount of radiolabeled nucleotide incorporated . The incorporation of radioactive material was corrected for the relative DNA content in each band, determined by densitometry of EtBrstained gel. As illustrated in Fig. 6, damage-induced DNA repair synthesis determined for the plasmid modified by complex 1 or complex 2 was lower than that found for cisplatin at the same level of modification (60 or 45 %, respectively, of that found for the plasmid modified by cisplatin; Fig. 6b). Thus, DNA repair appears to be a factor contributing to the elevated toxicity in tumor cells of 1 and 2 compared with cisplatin. Cytotoxicity in cisplatin-resistant tumor cells The results described so far indicate that cellular accumulation and DNA repair may be important factors
J Biol Inorg Chem (2013) 18:579–589
587
A2780cisR human ovarian carcinoma cells (with acquired cisplatin resistance), which are resistant to cisplatin through a combination of decreased uptake, enhanced DNA repair/tolerance, and elevated GSH levels [30, 31]. The tumor cell line was incubated for 24 or 72 h with the platinum compounds and the cell survival in the culture treated with the platinum compounds was evaluated as described in ‘‘Materials and methods.’’ The IC50 values (mean ± SD of three independent experiments) obtained for 1, 2, and cisplatin after 24 h were 2.0 ± 0.4, 3.6 ± 0.6, and 26.0 ± 3.1 lM, respectively; the IC50 values obtained after 72 h were 0.6 ± 0.1, 0.7 ± 0.1, and 14.8 ± 0.8 lM, respectively. These results show that both complex 1 and complex 2 were markedly more active than cisplatin in the cisplatin-resistant cell line A2780cisR.
Discussion
Fig. 6 In vitro DNA repair synthesis assay. DNA repair of pUC19 modified with cisplatin or complexes 1 and 2 at rb = 0.07 was mediated by the extract from repair-proficient HeLa S3 cells. Unmodified pBR322 was also used as an internal control. Repair efficiency is given by the radioactivity of incorporated [a-32P]dATP normalized to the relative amount of DNA in the band determined from the ethidium bromide (EtBr)-stained gel. a Results of a typical experiment. Top panel: autoradiographic image of the same gel showing radiolabel incorporation. Bottom panel: EtBr-stained gel showing migration of undamaged control plasmid (pBR322, top bands) and platinated pUC19 (bottom bands). C control, unmodified pUC19 plus unmodified pBR322; cisPt pUC19 modified with cisplatin plus unmodified pBR322; 1 pUC19 modified with complex 1 plus unmodified pBR322; 2 pUC19 modified with complex 2 plus unmodified pBR322. b Incorporation of [a-32P]dATP into unmodified or platinated pUC19 plasmid. For all quantifications representing the mean values of two separate experiments, incorporation of radioactive material is normalized to the relative DNA content in each band determined from the relative densities of the bands on the EtBrstained gel. The radioactivity associated with incorporation of [a-32P]dATP into DNA modified with cisplatin was arbitrarily set to 100 %. Values shown in the graph are the means (± SD) of two separate experiments, each conducted in duplicate
responsible for differences in toxicity of 1, 2, and cisplatin in tumor cell lines. To further support this thesis, we tested the effects of complexes 1 and 2 on viability of cancer cells which acquired resistance to cisplatin also owing to reduced cellular accumulation and enhanced DNA repair. Such tumor cells should be more sensitive to complexes 1 and 2 in comparison with cisplatin. Therefore, the cytotoxicity of complexes 1 and 2 was also determined in
We have demonstrates in this work that complexes 1 and 2 are quite toxic to ovarian tumor cells, with IC50 values that were significantly better than those observed for cisplatin in the cisplatin-sensitive cell line A2780. Thus, these results complement those obtained in a previous study [4] showing that complexes 1 and 2 exhibit promising cytotoxicity with regard to the HOS (osteosarcoma), MCF7 (breast adenocarcinoma), and LNCaP (prostate adenocarcinoma) human cancer cell lines. In this work we investigated factors which might be involved in the mechanism underlying the cytotoxic effects of complexes 1 and 2 and compared these factors with those involved in the mechanism underlying the cytotoxic effects of the most frequently studied anticancer metallodrug, cisplatin. Cisplatin is known to exert its cytotoxic effect by inducing apoptosis [32–34]. To investigate further the mechanism of cell death induced by complexes 1 and 2, the levels of apoptosis and necrosis were quantified. The results (Fig. 2) indicate that complexes 1 and 2 induce cell death by apoptosis in A2780 cells with considerably higher efficiency than cisplatin and that apoptosis markedly prevailed over necrosis. Activation of cell cycle checkpoints is a general cellular response after exposure to cytotoxic agents. Previous studies have indicated that cisplatin and other platinum agents predominantly inhibit cell cycle progression at the S and/or G2 phase [13, 35–37]. Our studies performed in the cell line with wild-type p53 status show differences between complexes 1 and 2 and cisplatin at the level of cell cycle regulation (Fig. 3). Nuclear debris from apoptotic or necrotic cells is observed as a sub-G1 population for cells treated with complexes 1 and 2, but the sub-G1 population is markedly lower for cisplatin-treated cells (Fig. 3), and this is consistent with a significantly higher level of DNA
123
588
fragmentation due to apoptosis in comparison with cisplatin in A2780 cells (Fig. 2a). Moreover, although cisplatin blocks A2780 cells in the G2 phase at concentration as low as 3 lM, complexes 1 and 2 induced a somewhat smaller block in A2780 cells only at 5 lM concentration. The differences between complexes 1 and 2 and cisplatin at the level of cell cycle regulation imply a molecular mechanism of action of the cis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives distinct from that of cisplatin. The results of this work also show that using bulkier ligands such as 7-azaindole halogeno derivatives gives rise to increased cellular accumulation and higher cytotoxicity. In addition, the amount of platinum found on the DNA of A2780 cells incubated with complexes 1 and 2 for 5 or 24 h was higher than that found in cells treated with cisplatin (Table 1), which correlates with their cytotoxicity (see ‘‘Cytotoxicity’’) and cellular accumulation (Table 1). DNA may therefore be a potential target also for these cytotoxic cisplatin analogues containing halogeno-substituted 7-azaindoles, although we cannot rule out the possibility that nuclear DNA may not be the only target. The finding that complexes 1 and 2 are capable of delivering platinum to DNA in the cell nucleus prompted us to examine the binding of complexes 1 and 2 to DNA in a cell-free medium. The resulting DNA damage triggers downstream effects, including the inhibition of replication and transcription, and cell cycle arrest [1, 2]. The DNA binding experiments performed in this work in a cell-free medium indicated that modification reactions result in the irreversible coordination of complexes 1 and 2. In addition, transcription mapping experiments [4], determination of monofunctional adducts, and determination of the interstrand cross-linking efficiency of complexes 1 and 2 (Fig. 4) suggest that several aspects of the DNA binding mode of complexes 1 and 2 are similar to those of the parent cisplatin. However, it cannot be excluded that identical types of DNA adducts of 1, 2, and cisplatin can distort the DNA conformation differently and can be processed by cellular components differently. Thus, the distinct differences in cell killing observed for complexes 1 and 2 and cisplatin may be associated with processes at the DNA level. DNA adducts of cisplatin inhibit DNA replication to an extent that slows cell cycle progression through the S phase but allows cells to accumulate in the G2 phase [38]. Efficient repair of the cross-links and replicative bypass across the adducts by translesion DNA polymerases contribute to the survival of cisplatin-treated cells [39]. However, the low accumulation of cells in the G2 phase and the still significant cell killing observed in A2780 cells treated with complex 1 or complex 2 suggests that inhibition of DNA synthesis by adducts of
123
J Biol Inorg Chem (2013) 18:579–589
these agents is more lethal to the cells than are DNA adducts of cisplatin. The results of this work also demonstrate that complexes 1 and 2 react with GSH with a rate similar to that of cisplatin (Fig. 5). Both clinical and preclinical studies have shown that cells with an elevated level of GSH (more than 10 mM) may be resistant to cisplatin and its analogues [40–43]. Further studies are warranted to determine whether the different cytotoxicity of complexes 1 and 2 and cisplatin is attributable to different reactions with sulfurcontaining compounds. Recent clinical studies suggest that high levels of expression of proteins associated with nucleotide excision repair of cisplatin–DNA adducts result in tumor resistance and, ultimately, are responsible for the low efficacy of classic platinum-based regimens [44, 45]. Nucleotide excision repair system most efficiently recognizes and removes irreversible DNA adducts that severely distort and destabilize double-helical DNA. Thus, on the basis of the findings of our experiments in cell-free systems, the adducts produced by complexes 1 and 2 should be poorer substrates for nucleotide excision repair than the adducts of cisplatin. The relative resistance to DNA repair would explain why complexes 1 and 2 show major pharmacological advantages over cisplatin in the ovarian cancer cell line. The repair synthesis assay in randomly modified plasmid performed in this study (Fig. 6) demonstrates that the presumably most cytotoxic and major adducts formed by complexes 1 and 2 are repaired considerably less efficiently than the damage caused by cisplatin, which may potentiate toxic effects of this class of PtII compounds in tumor cells. The toxicity of 1, 2, and cisplatin in A2780cisR cells suggests that complexes 1 and 2 are capable of circumventing resistance to the parent metallodrug induced by alterations in cellular accumulation and DNA repair. Hence, the latter two factors appear to be responsible for differences in the toxicity of complexes 1 and 2 and cisplatin in tumor cells. The work reported here demonstrates that rational chemical design can be applied to cis-dichloridoplatinum(II) complexes to achieve potent cancer cell cytotoxicity. It is notable that halogeno-substituted 7-azaindole nonleaving groups can play a major role in controlling the chemical and biological properties, such as cellular accumulation, effects at the level of cell cycle regulation, propensity for DNA adduct repair, binding to DNA, and reactivity toward sulfur-containing nucleophiles. The results of this work reinforce the idea that direct analogues of conventional cisplatin-containing halogeno-substituted 7-azaindoles offer much promise for the design of novel therapeutic agents.
J Biol Inorg Chem (2013) 18:579–589 Acknowledgments This research was supported by the Czech Science Foundation (grant P301/10/0598) and the Ministry of Education of the Czech Republic (grant LH13096). The research of T.M., and J.P. was also supported by the student project of Palacky University in Olomouc (grant PrF 2013 017). J.K.’s research was also supported by the Operational Program Education for Competitiveness–European Social Fund (CZ 1.07/2.3.00/20.0057) of the Ministry of Education, Youth and Sports of the Czech Republic. P.S. and Z.T. acknowledge funding from the Operational Program Research and Development for Innovations–European Regional Development Fund (CZ.1.05/2.1.00/03.0058) and the Operational Program Education for Competitiveness–European Social Fund (CZ.1.07/2.3.00/20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic.
References 1. Kelland L (2007) Nat Rev Cancer 7:573–584 2. Wang D, Lippard SJ (2005) Nat Rev Drug Discov 4:307–320 3. Fojo T, Farrell N, Ortuzar W, Tanimura H, Weinstein J, Myers TG (2005) Crit Rev Oncol Hematol 53:25–34 4. Starha P, Travnicek Z, Popa A, Popa I, Muchova T, Brabec V (2012) J Inorg Biochem 115:57–63 5. Reedijk J (2003) Proc Natl Acad Sci USA 100:3611–3616 6. Johnson NP, Butour J-L, Villani G, Wimmer FL, Defais M, Pierson V, Brabec V (1989) Prog Clin Biochem Med 10:1–24 7. Jamieson ER, Lippard SJ (1999) Chem Rev 99:2467–2498 8. Fuertes MA, Castilla J, Alonso C, Perez JM (2003) Curr Med Chem 10:257–266 9. Brabec V, Palecek E (1970) Biophysik 6:290–300 10. Brabec V, Palecek E (1976) Biophys Chem 4:76–92 11. Reardon JT, Vaisman A, Chaney SG, Sancar A (1999) Cancer Res 59:3968–3971 12. Bugarcic T, Novakova O, Halamikova A, Zerzankova L, Vrana O, Kasparkova J, Habtemariam A, Parsons S, Sadler PJ, Brabec V (2008) J Med Chem 51:5310–5319 13. Kisova A, Zerzankova L, Habtemariam A, Sadler PJ, Brabec V, Kasparkova J (2011) Mol Pharm 8:949–957 14. Moser C, Lang SA, Kainz S, Gaumann A, Fichtner-Feigl S, Koehl GE, Schlitt HJ, Geissler EK, Stoeltzing O (2007) Mol Cancer Ther 6:2868–2878 15. Egger AE, Rappel C, Jakupec MA, Hartinger CG, Heffeter P, Keppler BK (2009) J Anal At Spectrom 24:51–61 16. Farrell N, Qu Y, Feng L, Van Houten B (1990) Biochemistry 29:9522–9531 17. Brabec V, Leng M (1993) Proc Natl Acad Sci USA 90: 5345–5349
589 18. Boudny V, Vrana O, Gaucheron F, Kleinwa¨chter V, Leng M, Brabec V (1992) Nucleic Acids Res 20:267–272 19. Eastman A (1983) Biochemistry 22:3927–3933 20. Eastman A (1986) Biochemistry 25:3912–3915 21. Dabrowiak JC, Goodisman J, Souid AK (2002) Drug Metab Dispos 30:1378–1384 22. Hagrman D, Goodisman J, Dabrowiak JC, Souid AK (2003) Drug Metab Dispos 31:916–923 23. Clodi K, Kliche KO, Zhao SR, Weidner D, Schenk T, Consoli U, Jiang SW, Snell V, Andreeff M (2000) Cytometry 40:19–25 24. Page JD, Husain I, Sancar A, Chaney SG (1990) Biochemistry 29:1016–1024 25. Lempers ELM, Inagaki K, Reedijk J (1988) Inorg Chim Acta 152:201–207 26. Lempers ELM, Reedijk J (1990) Inorg Chem 29:217–222 27. Reedijk J (1999) Chem Rev 99:2499–2510 28. Wang X, Guo Z (2007) Anticancer Agents Med Chem 7:19–34 29. Brabec V, Kasparkova J (2009) In: Hadjiliadis N, Sletten E (eds) Metal complex—DNA interactions. Wiley, Chichester, pp 175–208 30. Kelland LR, Barnard CFJ, Mellish KJ, Jones M, Goddard PM, Valenti M, Bryant A, Murrer BA, Harrap KR (1994) Cancer Res 54:5618–5622 31. Kelland LR, Sharp SY, ONeill CF, Raynaud FI, Beale PJ, Judson IR (1999) J Inorg Biochem 77:111–115 32. Wang GD, Reed E, Li QQ (2004) Oncol Rep 12:955–965 33. Perez J-M, Montero EI, Quiroga AG, Fuertes MA, Alonso C, Navarro-Ranninger C (2001) Metal Based Drugs 8: 29–37 34. Sedletska Y, Giraud-Panis M-J, Malinge J-M (2005) Curr Med Chem Anticancer Agents 5:251–265 35. Ormerod M, O’Neill C, Robertson D, Kelland L, Harrap K (1996) Cancer Chemother Pharmacol 37:463–471 36. Siddik ZH (2003) Oncogene 22:7265–7279 37. Liskova B, Zerzankova L, Novakova O, Kostrhunova H, Travnicek Z, Brabec V (2012) Chem Res Toxicol 25:500–509 38. Sorenson CM, Eastman A (1988) Cancer Res 48:4484–4488 39. Kartalou M, Essigmann JM (2001) Mutat Res 478:23–43 40. Akiyama S, Chen ZS, Sumizawa T, Furukawa T (1999) Anticancer Drug Des 14:143–151 41. Chen G, Hutter KJ, Zeller WJ (1995) Cell Biol Toxicol 11: 273–281 42. Brabec V, Kasparkova J (2005) Drug Resist Updates 8:131–146 43. Kelland LR (2000) Drugs 59:1–8 44. Weaver D, Crawford E, Warner K, Elkhairi F, Khuder S, Willey J (2005) Mol Cancer 4:18 45. Fujii T, Toyooka S, Ichimura K, Fujiwara Y, Hotta K, Soh J, Suehisa H, Kobayashi N, Aoe M, Yoshino T, Kiura K, Date H (2008) Lung Cancer 59:377–384
123
PŘÍLOHA 4 Štarha, P.; Hošek, J.; Vančo, J.; Dvořák, Z.; Suchý, P.; Popa, I.; Pražanová, G.; Trávníček, Z. Pharmacological and molecular effects of platinum(II) complexes involving 7-azaindole derivatives PLoS ONE 9 (2014) e90341
Pharmacological and Molecular Effects of Platinum(II) Complexes Involving 7-Azaindole Derivatives Pavel Sˇtarha1, Jan Hosˇek1, Ja´n Vancˇo1, Zdeneˇk Dvorˇa´k2, Pavel Suchy´ Jr3, Igor Popa1, Gabriela Prazˇanova´3, Zdeneˇk Tra´vnı´cˇek1* 1 Department of Inorganic Chemistry, Regional Centre of Advanced Technologies and Materials, Faculty of Science, Palacky´ University, Olomouc, Czech Republic, 2 Department of Cell Biology and Genetics, Regional Centre of Advanced Technologies and Materials, Faculty of Science, Palacky´ University, Olomouc, Czech Republic, 3 Department of Human Pharmacology and Toxicology, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic
Abstract The in vitro antitumour activity studies on a panel of human cancer cell lines (A549, HeLa, G-361, A2780, and A2780R) and the combined in vivo and ex vivo antitumour testing on the L1210 lymphocytic leukaemia model were performed on the cis[PtCl2(naza)2] complexes (1–3) involving the 7-azaindole derivatives (naza). The platinum(II) complexes showed significantly higher in vitro cytotoxic effects on cell-based models, as compared with cisplatin, and showed the ability to avoid the acquired resistance of the A2780R cell line to cisplatin. The in vivo testing of the complexes (applied at the same dose as cisplatin) revealed their positive effect on the reduction of cancerous tissues volume, even if it is lower than that of cisplatin, however, they also showed less serious adverse effects on the healthy tissues and the health status of the treated mice. The results of ex vivo assays revealed that the complexes 1–3 were able to modulate the levels of active forms of caspases 3 and 8, and the transcription factor p53, and thus activate the intrinsic (mitochondrial) pathway of apoptosis. The pharmacological observations were supported by both the histological and immunohistochemical evaluation of isolated cancerous tissues. The applicability of the prepared complexes and their fate in biological systems, characterized by the hydrolytic stability and the thermodynamic aspects of the interactions with cysteine, reduced glutathione, and human serum albumin were studied by the mass spectrometry and isothermal titration calorimetric experiments. Citation: Sˇtarha P, Hosˇek J, Vancˇo J, Dvorˇa´k Z, Suchy´ Poˆ Jr, et al. (2014) Pharmacological and Molecular Effects of Platinum(II) Complexes Involving 7-Azaindole Derivatives. PLoS ONE 9(3): e90341. doi:10.1371/journal.pone.0090341 Editor: Heidar-Ali Tajmir-Riahi, University of Quebect at Trois-Rivieres, Canada Received December 16, 2013; Accepted January 31, 2014; Published March 6, 2014 Copyright: ß 2014 Sˇtarha et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The authors acknowledge funding from the Czech Science Foundation (GACˇR P207/11/0841), Operational Program Research and Development for Innovations - European Regional Development Fund (CZ.1.05/2.1.00/03.0058), Operational Program Education for Competitiveness - European Social Fund (CZ.1.07/2.3.00/20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic, Palacky´ University in Olomouc (PrF_2013_015), and University of Veterinary and Pharmaceutical Sciences Brno (IGA VFU 82/2012/FaF). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail:
[email protected]
side effects (nephrotoxicity, neurotoxicity, myelosuppression etc.) [1,2,9]. Since 1978, the development of new cisplatin-inspired bioactive complexes seemed many times as a lost cause, nevertheless the preparation of several novel highly-active compounds proved that there is still a room for the improvement of pharmacological properties of antitumour platinum complexes [10]. One of the possible future research directions was shown in the case of picoplatin [11]. A rational replacement of one NH3 molecule in the cisplatin molecule by one relatively simple heterocyclic N-donor ligand (2-methylpyridine) led to the lower interaction with sulphurcontaining biomolecules (e.g. glutathione) resulting in lower inactivation of the substance and, as a consequence of this, in a notable ability to overcome resistance of various tumour types to the action of cisplatin and oxaliplatin [11–13]. Although picoplatin failed in the clinical trials on non-small-cell lung carcinoma due to the continued progression of the disease and showing several drawbacks (e.g. neutropenia, thrombocytopenia or vomiting), it is currently undergoing clinical trials as therapeutic for colorectal and prostate cancer [14]. Bearing this in mind, we aimed to find a simple, planar and well-coordinating N-donor heterocycle, whose incorporation into
Introduction Cisplatin is a simple platinum(II) coordination compound that is used world-wide for the treatment of various types of cancer [1,2]. The discovery of its antitumour effect on human cancer cells in 1960s [3,4] and its consequent approval by the Food and Drug Administration for the therapeutic use in 1978 represent an important milestone in the field of both bioinorganic and medicinal chemistry. Representing a relatively uncomplicated leading compound, hundreds and hundreds of platinum(II) complexes were prepared using diverse strategies how to modify the structure and biological action of cisplatin. Two basic approaches focused either on the substitution of the leaving groups, i.e. two chlorides (e.g. in carboplatin [5] or nedaplatin [6]), or on the substitution of two NH3 molecules within the cisplatin molecule by different N-donor ligands (e.g. in oxaliplatin [7] or lobaplatin [8]) represent the most promising ways leading towards clinically useful platinum(II) compounds. Nevertheless, none of these compounds avoided completely both of the main disadvantages related with the platinum-based drugs application, i.e. the resistance (acquired or intrinsic), and negative and dose-limiting
PLOS ONE | www.plosone.org
1
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
of Animals Against Cruelty at the University of Veterinary and Pharmaceutical Sciences Brno (Permit Number: 42/2011). To minimize the suffering of laboratory animals, the number of pharmacological interventions was limited to the necessary minimum and the animals were observed regularly for any signs of unnecessary suffering from the symptoms of the tumor progression. All animals showing at least one of the following symptoms - weight loss higher than 50% of the initial weight, symptoms of acute toxicity caused by the tested compounds, and excessive volume of the tumors hindering the mobility or social communication of the animals, had to be sacrificed immediately by cervical dislocation. However, no such cases occurred within the whole 21 days of pharmacological testing. The animal tissues for the ex vivo experiments were taken post mortem, immediately after all animals were sacrificed by the cervical dislocation.
the cisplatin molecule, instead of one or both NH3 molecules, could bring in the similar effect on the antitumour properties as in the case of picoplatin. Thus, we chose 7-azaindole and its halogeno derivatives, which fulfil the mentioned requirements [15–17]. Moreover, the biological perspective of 7-azaindole moiety, as recently proved on various 7-azaindole derivatives reported as having notable biological properties, such as anticancer activity [18], inhibition of kinases (e.g. tropomyosin-related [19] or Abl and Src kinases [20]) or antiviral effect [21], has to be taken into account as well. Recently, we prepared and thoroughly characterized the platinum(II) dichlorido complexes involving 7-azaindole or its halogeno derivatives 3-chloro-7-azaindole (the complex 1 in this work), 3-iodo-7-azaindole (2 in this work) and 5-bromo-7azaindole (3 in this work) (see Figure 1) and screened them for their in vitro antitumour activity against HOS osteosarcoma, MCF7 breast carcinoma and LNCaP prostate carcinoma human cancer cell lines with IC50 equalled 1.5–8.0 mM [22,23]. In addition, the results of mechanistic studies confirming their analogous mechanism of action to cisplatin and significantly higher cell-uptake, intracellular transport and DNA platination, resulting in the higher in vitro effectiveness as compared with cisplatin were recently reported [23,24]. Following the previous promising results of in vitro studies, we were determined to perform an advanced study of in vitro cytotoxicity on an extended panel of human cancer cell lines (A549, HeLa, G-361, A2780 and cisplatin-resistant A2780R), together with the in vivo and ex vivo studies on L1210 lymphocytic leukaemia model complemented by the histological and immunohistochemical investigation on the cancerous tissues and studies of expression of caspases 3 and 8, p53 and VEGF-A, i.e. the proteins associated with the tumour growth progression and induction of apoptosis. In this paper, we present the results of thorough biological testing, extended by the data regarding the applicability of the compounds and the stability of their solutions in water containing media.
Chemicals and Biochemicals K2[PtCl4], 3-chloro-7-azaindole (3Claza), 3-iodo-7-azaindole (3Iaza), 5-bromo-7-azaindole (5Braza) and solvents (acetone, methanol, ethanol, diethyl ether, N,N9-dimethylformamide, water) were purchased from the following commercial sources - Sigma– Sigma-Aldrich Co. (Praha, Czech Republic), Acros Organics Co. (Pardubice, Czech Republic) and Fisher-Scientific Co. (Pardubice, Czech Republic). The complexes cis-[PtCl2(3Claza)2] (1), cis[PtCl2(3Iaza)2] (2) and cis-[PtCl2(5Braza)2] (3) (Figure 1) were synthesized by the reactions of water solution of K2[PtCl4] with two molar equivalents of naza dissolved in ethanol, and characterized as reported previously [23]. The RPMI 1640 medium and penicillin-streptomycin mixture were purchased from Lonza (Verviers, Belgium). Phosphatebuffered saline (PBS), fetal bovine serum (FBS), phorbol myristate acetate (PMA), Auranofin (98%#), erythrosin B, and Escherichia coli 0111:B4 lipopolysaccharide (LPS) were purchased from SigmaAldrich (Steinheim, Germany). Cell Proliferation Reagent WST-1 was obtained from Roche (Mannheim, Germany). Instant ELISA Kits (eBioscience, Vienna, Austria) were used to evaluate the production of TNFa and IL-1b.
Materials and Methods Ethics Statement
NMR Spectroscopy
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health. The protocol was approved by the Expert Committee on the Protection
1
H, 13C and 195Pt NMR spectra and two dimensional correlation experiments (1H–1H gs-COSY, 1H–13C gs-HMQC, 1 H–13C gs-HMBC; 1H–15N gs-HMBC; gs = gradient selected, COSY = correlation spectroscopy, HMQC = heteronuclear multiple quantum coherence, HMBC = heteronuclear multiple bond coherence) of the DMF-d7 solutions were measured at 300 K on a Varian 400 device at 400.00 MHz (1H), 100.58 MHz (13C), 86.00 MHz (195Pt) and 40.53 MHz (15N). 1H and 13C spectra were adjusted against the signals of tetramethylsilane (Me4Si). 195 Pt spectra were calibrated against potassium hexachloroplatinate (K2PtCl6) in D2O found at 0 ppm. 1H–15N gs-HMBC experiments were obtained at natural abundance and calibrated against the residual signals of DMF adjusted to 8.03 ppm (1H) and 104.7 ppm (15N). The splitting of proton resonances in the reported 1H spectra is defined as s = singlet, d = doublet, t = triplet, br = broad band, dd = doublet of doublets, m = multiplet. Stability study: the DMF-d7 solutions of 1–3 were studied by 1H NMR after 6 h, 24 h and 1 week and by all the above-mentioned NMR experiments after two weeks of standing at laboratory temperature, and by 1H NMR after 15 min and 1 h of heating at 100uC. Hydrolysis study: the complexes 1 and 3 (ca. 20 mg) were dissolved in 0.5 ml of DMF-d7 and 0.25 ml of distilled water were added. The mixture was heated to 50 or 100uC for 3 h and white
Figure 1. General structural formula of the studied platinum(II) complexes. R3 = Cl for 1, I for 2 and H for 3; R5 = H for 1, H for 2 and Br for 3. doi:10.1371/journal.pone.0090341.g001
PLOS ONE | www.plosone.org
2
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
37uC and 5% CO2 in a humidified incubator. The human cancer cells were treated with 1–3 and cisplatin (applied up to 50 mM) for 24 h, using multi-well culture plates of 96 wells. In parallel, the cells were treated with vehicle (DMF; 0.1%, v/v) and Triton X100 (1%, v/v) to assess the minimal (i.e. positive control) and maximal (i.e. negative control) cell damage, respectively. The MTT assay was measured spectrophotometrically at 540 nm (TECAN, Schoeller Instruments LLC). The data were expressed as the percentage of viability, when 100% and 0% represent the treatments with DMF and Triton X100, respectively. The cytotoxicity data from the cancer cell lines were acquired from three independent experiments (conducted in triplicate) using cells from different passages. The IC50 values were calculated from viability curves. The results are presented as arithmetic mean6SD.
solid formed. The product was removed and dissolved in DMF-d7. The purity of the hydrolyzed products was .95%.
The Interaction and Stability Studies of 1 Evaluated by the ESI-MS The 10 mM methanolic solution (final concentration) of the selected complex cis-[PtCl2(3Claza)2] (1) was mixed with the equivalent volume of L-cysteine (Cys) and reduced glutathione (GSH) dissolved in water in the physiological concentrations (the final concentration of 260 mM for Cys, and 6 mM for GSH, respectively) [25]. The mixture was sealed and kept at 25uC for a month. During this period (immediately after the preparation, 1 h, 12 h and 1 month after the preparation), the small amounts of the reacting mixture (usually 20 mL) were analyzed by the FIA-ESIMS method in both the positive and negative ionization mode. The mobile phase composed of 90% methanol and 10% of 10 mM ammonium acetate was pumped at the rate of 0.2 mL/ min by the quaternary pump of Dionex Ultimate 3000 HPLC System. The samples were injected (50 mL) into the flow of the mobile phase by the autosampler of the HPLC from the same make. The parameters of Thermo LCQ Fleet mass spectrometer were set as follows: API Source voltage 4.5 kV, sheath gas flow rate 40 (arb units), aux gas flow rate 20 (arb units), capillary temperature 275uC, RF lens voltage 22.91 V, lens0 25.89 V, lens1 28.89 V, gate lens 243.94 V, multipole1 offset 211.87 V, front lens 270.71 V. The similar FIA-ESI-MS experiment was arranged to test the stability of the selected complex 1 in a water/ methanol solution (10 mM, 1:1 v/v) over the same time period as mentioned above.
In Vivo Antitumour Activity The testing of in vivo antitumour activity was carried out according to the previously published procedure using the female DBA/2 SPF mice as experimental animals [26]. In this specific case, the animals were housed in the Sealsafe NEXT – IVC Blue Line Housing System (Tecniplast, Italy) to ensure the best possible experimental conditions and eliminate the risk of possible intergroup cross-contamination. Due to the lack of toxicological data and in order to eliminate the excessive use of laboratory animals, the experimental setup, involving the use of the same dose for all the platinum(II) complexes (3 mg/kg) was used. The animals were weighed daily and observed several times a day for the signs of tumour progression, changes in behaviour, sudden death, and sacrificed if their body weight decreased below 50% of the starting weight or if other severe toxicological problems were seen. The experimental data, describing the survival of the experimental animals, were expressed as the percentage of mean survival time, %T/C defined as the ratio of the mean survival time of the treated animals (T) divided by the mean survival of the untreated control group (C). There were no deaths attributable to acute toxicity of the tested compounds.
ITC Experiments All the isothermal titration calorimetry (ITC) measurements were performed at 30uC using a VP-ITC device (MicroCal Inc.). The studied solutions (2.5 mM HSA, 100.0 mM GSH, 100.0 mM Cys, 1.0 mM cisplatin and 1.0 mM 1) in a water/DMF mixture (1:1 v/v) were degassed prior to the titration. The experiments (titration of 100.0 mM Cys with 1.0 mM solution of 1, titration of 100.0 mM GSH with 1.0 mM solution of 1, titration of 2.5 mM HSA with 1.0 mM solution of 1, titration of 100.0 mM Cys with 1.0 mM solution of cisplatin, titration of 100.0 mM GSH with 1.0 mM solution of cisplatin and titration of 2.5 mM HSA with 1.0 mM solution of cisplatin) were carried out uniformly: the biomolecule (HSA, GSH or Cys) was titrated with 1 (or cisplatin) by 25 injections of 10 mL each with interval between injections being 300 s. The blank experiments were performed using the same conditions without appropriate biomolecule in cell, where only the water/DMF (1:1 v/v) mixture was poured. The blank experiment data were subtracted and the data were fitted (one-site, two-site or sequential binding model) using the MicroCal Origin software version 7.0.
Histological and Histochemical Evaluations The tissue samples, obtained by the dissection of the sacrificed animals were divided into two parts, the first one was kept below 280uC for further evaluation by the methods of molecular biology, and the second one was processed by the histological procedures as described previously [26]. The paraffin embedded preparations were stained by the standard hematoxylin and eosin staining. The immunohistochemical detection of the Caspases 3 and 8, tumour necrosis factor-alpha (TNF-a) and transcription factor p53 expression were achieved by the use of appropriate antibodies, selective for the mouse species, obtained from AbCam (Cambridge, UK). The quantification of the tissue expression of different proteins in the samples was semi-quantitative, based on the percentage of the areas containing the detected proteins in the view field of at least three different samples. The scale from 0 to 4 was used, where 0 = protein was not detected, 1 = up to 25%, 2 = up to 50%, 3 = up to 75% and 4 = up to 100% of the area contains the detected protein within the view field. The median value from at least three observations was used to evaluate the expression of selected proteins and other histological observations.
Cell Culture and In Vitro Cytotoxicity Testing In vitro cytotoxic activity of 1–3 and the clinically used platinumbased drug cisplatin was determined by an MTT assay in Human Negroid Cervix Epitheloid Carcinoma (HeLa; ECACC No. 93021013), Human Caucasian Malignant Melanoma (G361; ECACC No. 88030401), Human Ovarian Carcinoma (A2780; ECACC No. 93112519), Human Ovarian Carcinoma Cisplatin-resistant Cell Line (A2780R; ECACC No. 93112517) and Human Caucasian Lung Carcinoma (A549; ECACC No. 86012804) cancer cell lines purchased from European Collection of Cell Cultures (ECACC). The cells were cultured according to the ECACC instructions and they were maintained at PLOS ONE | www.plosone.org
Protein Expression Analysis Frozen tumour samples were homogenized by a bench blender IKA DI25 (IKA-Werke, Germany) in the presence of a lysis buffer [50 mM Tris-HCl pH 7.5, 1 mM EGTA, 1 mM EDTA, 1 mM sodium orthovanadate, 50 mM sodium fluoride, 5 mM sodium 3
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
pyrophosphate, 270 mM sucrose, 0.5% (v/v) Triton X-100]. Subsequently, the blended samples were centrifuged at 7.000 g at 4uC for 15 min. The supernatants were collected and the protein concentration was measured by the Brandford’s method using the Brandford reagent (Amresco, USA) according to a manufacturer’s manual. The measured samples were stored at 280uC for the following experiments. The concentration of VEGF-A was evaluated by ELISA using VEGF-A Platinum ELISA (eBioscience, Austria). The amount of produced VEGF-A was normalized according to the total protein concentration. The expression of caspase 3, caspase 8 and p53 were evaluated by Westernblot and immunodetection. Lysates were mixed with a denaturing loading dye [250 mM Tris-HCl pH 6.8, 5% (v/v) bmercaptoethanol, 10% (w/v) SDS, 30% (v/v) glycerol, 0.04% (w/ v) bromophenol blue], heated for 5 min at 70uC and loaded into a 12% polyacrylamide denaturing gel. The final protein amount was 50 mg per well. After protein separation in the gel, they were transferred on the supported nitrocellulose membrane 0.2 mm (Bio-Rad, USA) and the proteins were visualized by Ponceau S dye (Sigma-Aldrich, USA) for loading control. Then, the membrane was blocked by 5% (w/v) BSA at TBST [10 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% (v/v) Tween-20]. The following primary and secondary antibodies [conjugated with horseradish peroxidase (HRP)] were used for immunodetection: caspase 3 (eBioscience, USA) in the dilution of 1:1000, caspase 8 (Abcam, UK) in the dilution of 1:1000 and p53 (Abcam, UK) in the dilution of 1:1000, goat anti-mouse antibody in the dilution of 1:3000 (Sigma-Aldrich, USA), goat anti-rabbit antibody in the dilution of 1:3000 (Sigma-Aldich, USA). HRP was detected by Amplified Opti-4CN kit (Bio-Rad, USA). One sample from the control group was loaded into all gels to avoid inter-gel variability. The amount of the proteins was determined by densitometric analysis using AlphaEasy FC 4.0.0 software (Alpha Innotech, USA) and was correlated according to the control sample.
Stability Studies The NMR spectra of the studied complexes in DMF-d7 did not show any difference after two weeks at laboratory temperature and even after 15 min at 100uC. We observed a slight increase of transisomer portion (see the data in File S1 labelled as 1t, 2t, and 3t, which correspond to the trans-isomers of the complexes 1, 2, and 3, respectively; the NMR data for the corresponding cis-isomers were already published in our previous work [23]) after 1 h at 100uC (Figure 2). In addition, the hydrolytic stabilities of the complexes 1 and 3 were investigated by means of 1H NMR spectroscopy in the mixture of DMF-d7/H2O (2:1 v/v). We did not detect any 1H NMR spectra changes even after one week at laboratory temperature. However, the set of new signals was detected at the spectra of the complexes heated in the mentioned DMF-d7/ H2O mixture to 50uC or 100uC (see the data in File S1 labelled as 1 h and 3 h for the hydrolytic products of the complexes 1, and 3, respectively). These changes are most probably connected with the hydrolysis of the studied complexes in the mentioned watercontaining system, since the chemical shifts of the new signals do not correspond to either cis- (NMR data given in [23] or transisomer (the NMR data are given in File S1) of the studied complexes or to the signals of free naza molecules. Probably, these signals may be associated with the formation of the hydrolytic products/species, such as [Pt(naza)2(OH)2] or [Pt(naza)2(H2O)2]2+. In addition, the similar ESI mass spectrometry experiment was arranged to test the stability of the complex 1 in water-containing solution, i.e. in a water/methanol mixture (10 mM, 1:1 v/v) in this case. The attention was drawn towards the possible formation of intermediates which could (at least theoretically) facilitate the interaction of the prepared complexes with target biomolecules (e.g. parts of the proteasome or DNA), and indeed, a very slow time-dependent accrual in the relative intensity of ions corresponding to hydrolysis products (Figure 3), dominantly the species at m/z2 533.05, together with m/z2 564.85 corresponding to the ionic species [M-H-2Cl+2OH]2, and [M-H-2Cl+2OH+ CH3OH]2, respectively, was evident in the mass spectra obtained in the negative ionization mode. In comparison to that, the inverse dependence of intensity on the time was found for another type of intermediate observed at m/z2 416.10, corresponding (according to mass and isotopic distribution) to the ionic species [M-L]2; L = 3Claza.
Zymography Samples were prepared by the same way as for immunodetection. Zymography of MMPs was performed in the gelatin impregnated gel as was described previously [27]. For the analysis, 30 mg of proteins obtained from lysates were loaded in a gel. The intensity of the digested regions was calculated by AlphaEasy FC 4.0.0 software (Alpha Innotech, USA) for densitometric analysis. The results were normalized according to 1% (v/v) fetal bovine serum (FBS) (Sigma-Aldrich, USA), which was used in each gel as a standard control.
ESI-MS Interaction Studies of 1 with Cysteine and Reduced Glutathione In an effort to describe the reactivity of the prepared complexes with the major sulphur-containing biomolecules found in the human plasma, the methanolic solution of the complex 1, as a representative example, was mixed with the mixture of L-cysteine (Cys) and reduced glutathione (GSH) in water in the physiological concentrations. The macroscopic appearance (a colourless solution without any traces of precipitation) of the solution stayed the same during the whole duration of the experiment. The analysis of mass spectra (and that applies to all measured data) did not reveal the evidence that the prepared complexes could be able to interact with physiological levels of main sulphur-containing biomolecules (Cys and GSH). We did not detect any species whose mass corresponds to the adduct of the fragments or hydrolysis products of the studied platinum(II) complex with Cys and/or GSH even one month after the preparation of the mixture. In fact, only the primal pseudomolecular ion at m/z 619.91, corresponding (according to mass and isotopic distribution) to the ionic species [M-H-2Cl+CH3OH+5H2O]+, was clearly present in the mass spectra of the analysed mixture (see Figure S1 in File S1).
Statistical Evaluation The significance of the differences between the results was assessed by the ANOVA analysis, followed by Tukey’s post-hoc test for multiple comparisons, with p,0.05 considered to be significant (QC Expert 3.2, Statistical software, TriloByte Ltd.).
Results and Discussion Synthesis and General Properties The studied complexes cis-[PtCl2(3Claza)2] (1), cis[PtCl2(3Iaza)2] (2) and cis-[PtCl2(5Braza)2] (3) (Figure 1) were prepared by the one-step reactions of 7-azaindole halogenoderivatives (naza) with K2[PtCl4] at 50uC as described previously [23], yielding the products of high chemical purity as demonstrated by 1H, 13C, 15N and 195Pt NMR data, with isomeric purity . 95% (based on 1H NMR data; see Figure 2A). PLOS ONE | www.plosone.org
4
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
Figure 2. Part of the 1H NMR spectra of cis-[PtCl2(5BrHaza)2] (3). The spectra show the N1–H signal of cis- (left) and trans-isomers (right) as observed at different times and temperatures: (A) fresh solution at laboratory temperature; (B) after two weeks at laboratory temperature; (C) after 15 min at 100uC; (D) after 1 h at 100uC. doi:10.1371/journal.pone.0090341.g002
Figure 3. The ESI-MS spectra of the complex 1. The studied complex was dissolved in water/methanol solution (10 mM, 1:1 v/v) and measured 12 h (A) and 1 month after the preparation (B) using the negative ionization mode. doi:10.1371/journal.pone.0090341.g003
PLOS ONE | www.plosone.org
5
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
eters: n1 = 33.461.70, K1 = (1.3760.18)6104 M21, DH1 = 2 20.062.3 kcal/mol, DS1 = 247.2 cal mol21 K21, n2 = 28.46 1.87, K2 = (1.6760.20)6107 M21, DH2 = 210.860.0 kcal/mol, DS2 = 22.49 cal mol21 K21 (for cisplatin the data K1 = (2 1.0660.52)6103 M21, DH1 = 229.561.4 kcal/mol, K2 = (7.89 63.20)6104 M21, DH2 = 31.861.3 kcal/mol, DS2 = 1.07 kcal mol21 K21, K3 = (24.4660.21)6103 M21, DH3 = 227.86 0.7 kcal/mol were obtained by a three-site sequential binding model). In other words, the ITC results indicated two different binding sites of the complex 1 on albumin and different type of non-covalent interaction of the complex 1 in comparison with cisplatin. Similar results, but with only one-fold difference between the binding constants K1 and K2, were reported for (to the best of our knowledge) the only platinum complex, whose interaction with HSA was studied by ITC [34]. The presumption that the difference in Kn could be caused by the conformation changes of HSA after binding of the complex to the first binding site, has to be taken into account as well.
ITC Interaction Studies with Cysteine, Reduced Glutathione and Human Serum Albumin The ITC experiments for the representative complex 1 and cisplatin were performed to demonstrate their ability to interact with Cys, GSH and human serum albumin (HSA) as well as differences between their interaction. The interaction with biomolecules, such as the above-mentioned reduced glutathione or cysteine, is one of the crucial features of antitumour effective platinum(II) complexes [28]. Since any covalent interactions of the complex 1 with GSH and Cys were not observed by means of ESI-MS, some kind of interactions of the studied complexes with GSH were detected by Uv-Vis spectroscopy in our previous work [24]. Therefore, we decided to use ITC as a very sensitive tool for the investigation and thermodynamic characterization of the interaction (unnecessary covalent from the principle of ITC) with various biomolecules [29,30]. We performed analogical experiments for the representative complex 1 and cisplatin to demonstrate the differences between both platinum(II) dichlorido complexes. An interaction of the complex 1 with both the cysteine and GSH was observed by ITC (see Figure S2 in File S1). Low solubility of the complex 1 in the medium used did not allow us to improve the data quality (in the case of more soluble cisplatin, we used the same conditions to make the obtained results on both compounds comparable). The data were fitted by two- (GSH) and three-site (cysteine) sequential binding model to give the following results: K1 = (1.046 0.10)6105 M21, DH1 = 4.460.5 kcal/mol, DS1 = 37.6 cal mol21 K21, K2 = (9.0660.60)6104 M21, DH2 = 26.360.2 kcal/mol, DS2 = 1.87 cal mol21 K21, K3 = (9.9360.63)6104 M21, DH3 = 25.462.0 kcal/mol, DS3 = 107 cal mol21 K21 for titration of cysteine and K1 = (4.3260.38)6104 M21, DH1 = 20.860.1 kcal/ mol, DS1 = 18.4 cal mol21 K21, K2 = (6.9960.59)6104 M21, DH2 = 4.460.2 kcal/mol, DS2 = 36.7 cal mol21 K21 for GSH. In the case of cisplatin, the best-fitted results were obtained by three- (GSH) and four-site (cysteine) sequential binding model with the following results: K1 = (3.7160.20)6105 M21, DH1 = 7.66 0.3 kcal/mol, DS1 = 50.7 cal mol21 K21, K2 = (1.1860.10) 6105 M21, DH2 = 2.360.2 kcal/mol, DS2 = 30.9 cal mol21 K21, K3 = (3.9760.35)6104 M21, DH3 = 44.969.4 kcal/mol, DS3 = 169 cal mol21 K21, K4 = (4.9060.47)6104 M21, DH4 = 2 88.8615.4 kcal/mol, DS4 = 2271 cal mol21 K21 for titration of cysteine and K1 = (3.8260.34)6105 M21, DH1 = 23.260.1 kcal/ mol, DS1 = 14.9 cal mol21 K21, K2 = (3.6260.36)6104 M21, DH2 = 2.760.5 kcal/mol, DS2 = 29.6 cal mol21 K21, K3 = (6.7260.67)6104 M21, DH3 = 29.660.7 kcal/mol, DS3 = 2 9.5 cal mol21 K21 for GSH. A comparison of the data obtained on 1 and cisplatin indicated different non-covalent interactions of these platinum(II) dichlorido complexes with cysteine and GSH (see Figure S2 in File S1), but due to the above-mentioned reasons regarding solubility we were unable to compare the systems from thermodynamic point of view. Serum proteins, including human serum albumin (HSA), are well-known to perform transport intracellular and delivery of the platinum(II) metallodrugs to the tumour tissues [31,32]. It has been also proved that an interaction of some platinum(II) complexes with albumin could result in enhancement of antitumour activity [33]. Therefore the study of a complex (1 and cisplatin in the case of this work) interaction with HSA should be carried out for complete description of biological properties of the studied substance. Again, we used ITC as a sensitive and capable thermodynamic characterization method suitable for the description of a platinum complex interaction with serum albumin (see Figure S2 in File S1). The data for the complex 1 were fitted to a two-site model resulting in the following thermodynamic paramPLOS ONE | www.plosone.org
In Vitro Antitumour Activity The complexes 1–3 were studied for their antitumour activity in vitro against lung carcinoma A549, cervix epithelial carcinoma HeLa, malignant melanoma G-361, ovarian carcinoma A2780 and cisplatin-resistant ovarian carcinoma A2780R human cancer cell lines, commonly used in the antitumour activity testing of platinum(II) complexes [35–38]. The results are given in Figure 4 and Table 1. The complexes 1–3 exceeded the antitumour activity in vitro of cisplatin against all the employed cancer cell lines as they were found to be 3.6-, 2.5- and 5.3-times (A549), 2.2-, 2.0- and 2.3times (HeLa), 1.7-, 1.1- and 5.7-times (G-361), 4.6-, 5.0- and 6.7times (A2780) and 10.0-, 9.6- and 12.9-times (A2780R) more effective than cisplatin. The complex 3 was the most active substance determined by the in vitro experiments with IC50 values lower than those of cisplatin as well as both the complexes 1 and 2. The complexes 1 and 3 were significantly more antitumour active in vitro (ANOVA, p,0.05) against all the cell lines as compared with cisplatin, while the same conclusion can be made for the complex 2 only against the A549, A2780 and A2780R cell lines (Figure 4, Table 1). These results indirectly proved that the studied platinum(II) complexes with 7-azaindoles are able to overcome intrinsic resistance to cisplatin on the A549, A2780 and A2780R (1–3), and HeLa and G-361 (1, 3) human cancer cell lines in vitro. To support this statement, we used a comparison of logIC50 (see Table S1 in File S1) of the complexes 1–3 and cisplatin, particularly the differences between the mean logIC50 observed for individual substances on all cancer cell lines reported in this work and in [23], and logIC50 of the individual substances against the concrete cell line (Figure 5A), and the differences between the mean logIC50 obtained on individual human cancer cell lines and logIC50 of the individual complex (Figure 5B), to demonstrate the sensitivity or resistance of the cancer cell to the action of the studied complexes in comparison with the others including cisplatin [39,40]. The antitumour activity in vitro of the complexes 1–3 can be evaluated also by means of the resistance factors, since the substances were tested on both cisplatin-sensitive (A2780) and resistant (A2780R) ovarian carcinoma cell lines (Figure 4). The resistance factors, calculated as IC50(A2780R)/IC50(A2780), equal 1.04 (1), 1.17 (2), 1.17 (3) and 2.25 (cisplatin), which show on the ability of the complexes 1–3 to avoid also the acquired type of cancer cell resistance against cisplatin. This feature of the studied complexes is in good agreement with the above-discussed ability of the studied complexes to overcome an interaction with the sulphur-containing biomolecules, which is well-known as one of 6
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
Table 1. In vitro antitumour activity given as IC50 6 S.D. in mM of the complexes 1–3 and cisplatin.
Cell line
1
2
3
Cisplatin
Reference
A549
7.262.3 *
10.364.3 *
4.961.7 *
25.867.1
This work
HeLa
4.562.9 *
5.061.9
4.361.8 *
10.062.6
This work
G-361
2.060.6 *
3.061.7
0.660.2 *
3.460.1
This work
A2780
2.660.4 *
2.460.7 *
1.860.7 *
12.060.8
This work
A2780R
2.760.7 *
2.860.8 *
2.160.8 *
27.064.6
This work
MCF7
3.460.3 *
8.060.9
2.060.2 *
19.664.3
[23]
HOS
3.860.1 *
3.960.2 *
2.560.1 *
34.266.4
[23]
LNCaP
3.360.7
3.861.3
1.560.4
3.861.5
[23]
Asterisks (*) symbolize significant difference (p,0.05) in in vitro antitumour activity of 1–3 as compared to cisplatin. doi:10.1371/journal.pone.0090341.t001
Figure 4. The in vitro antitumour activity testing results. The data were obtained by the MTT assay against human lung carcinoma A549, cervix epithelial carcinoma HeLa, malignant melanoma G-361, ovarian carcinoma A2780 and cisplatin-resistant ovarian carcinoma A2780R cell lines for 1–3 and cisplatin. The cells were exposed to the compounds for 24 h. Measurements were performed in triplicate and each cytotoxicity experiment was repeated three times. The given IC506S.D. (mM) values represent an arithmetic mean. The asterisk (*) denotes a significant difference (p,0.05) between the studied complexes and cisplatin (A); the plot of resistance factors calculated as IC50(A2780R)/IC50(A2780) for 1–3 and cisplatin (B). doi:10.1371/journal.pone.0090341.g004
the crucial factors decreasing the resistance of various tumours to the action of platinum-based drugs [11–13]. To conclude the in vitro antitumour activity testing of the complexes 1–3 on the human cancer cell lines, it has to be stated that they were recently tested for their antitumour activity in vitro against three others human cancer cell lines (i.e. eight human cancer cell lines in total), namely breast adenocarcinoma MCF7 (IC50 = 3.4, 8.0, and 2.0 mM for 1–3, respectively), osteosarcoma HOS (IC50 = 3.8, 3.9, and 2.5 mM for 1–3, respectively) and prostate adenocarcinoma LNCaP (IC50 = 3.3, 3.8, and 1.5 mM for 1–3, respectively) [23]. Two of the studied complexes (1, 3) have significantly higher in vitro antitumour activity (p,0.05) than cisplatin against all eight human cancer cell lines, while the complex 2 only on the A549, A2780, A2780R and HOS cells. Although the testing of the platinum complexes against five or more human cancer cell lines is quite common in the literature [40–42], the markedly higher antitumour effect than control (cisplatin in this work as well as in most of similarly focused paper) obtained on all the cell lines may be considered as unique to the best of our knowledge.
Figure 5. The results of in vitro antitumour activity on logarithmic scale. The plot representing the differences from the mean logIC50 obtained for the individual compounds (1, 2, 3 and cisplatin) on eight human cancer cell lines reported in this work (A549, HeLa, G-361, A2780 and A2780R) and in [23] (MCF7, HOS and LNCaP), where the positive values show the sensitivity of the particular cell line, while the negative ones relate to the resistance of the cancer cell line to the action of the platinum(II) complex (A) and the differences from the mean logIC50 obtained on the individual human cancer cell lines reported in this work (A549, HeLa, G-361, A2780 and A2780R) and in [23] (MCF7, HOS and LNCaP) (B). doi:10.1371/journal.pone.0090341.g005
the cisplatin group and the others experimental groups (1–3), none of these results proved to be significantly different (p,0.01). This is not a surprising result, because to reach the similar results of %T/ C between two antitumour active compounds, they usually have to share also a portion of similarity in toxicological, pharmacodynamic, and pharmacokinetic parameters. It seems likely, that this was not a case in the described experiment. Even if the tested complexes showed no significant extension of the mean survival time in comparison to the control group, the decision about the overall antitumour activity of tested complexes is much more complex process. That is why we chose a multiparametric semiquantitative approach to evaluate the overall antitumour
In vivo Antitumour Activity The in vivo antitumour activity of the complexes 1–3 and the reference drug cisplatin was tested on the mouse model of lymphocytic leukemia (L1210) using the complexes as therapeutic agents in the 7-day dosing schedule (3 mg/kg, i.p.) following the previous 10-day initiation period in which the primary tumours formed (examined by palpation). The percentages of mean survival time, %T/C defined as the ratio of the mean survival time of the treated animal groups (T) divided by the mean survival time of the untreated control group (C), were calculated and are shown in Table 2 and Figure 6. It should be noted that despite the relatively large differences of the absolute %T/C values between PLOS ONE | www.plosone.org
7
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
determined by the histological and histochemical methods, induced by the tested compounds are summarized in Table 3. As a parameter that indicated the overall state of tumour regression, the reactivity index was calculated as the ratio of atypical mitoses to the apoptoses in one view field of a microscope at 2006 magnification. The reactivity index demonstrates quite well the reactivity of the tumour tissue to the chemotherapy and its possible outcomes, i.e. the probability of further regression of the tumour tissues or, on the other hand, the possibility to restore the earlier rates of its proliferation leading to the impending death. The semi-quantitative evaluation of the tumour reactivity revealed, as expected from the lowering of the tumour tissue weight, the highest antiproliferative activity in the cisplatin-treated group of animals with the reactivity index of 1.35. The relatively low mitotic activity was accompanied by massive necroses and hemorrhage, which might be responsible for the induction of the inflammatory response documented by the higher detection of the pro-inflammatory cytokine TNF-a. The induction of the proinflammatory cellular responses was recently associated also with its nephrotoxicity [45]. The reactivity index calculated for the complex 3-treated group, which did not correlate proportionally with the decrease of tumour tissue mass, was found to be very positive and perspective. Nevertheless, its absolute value of 1.71 indicates that the complex 3 stimulates all the cellular signs of tumour regression, even if its onset of action is in comparison with cisplatin significantly slower. In the complex 1- and 2-treated groups, there were also found signs of tumour regression (reactivity index was 3.69, and 3.67, respectively), in this case, however, their impact on the reduction of the tumour tissue mass was insignificant. Other parameters semi-quantitatively determined in the tissue samples, are in agreement with the above mentioned overall evaluation on the basis of the reactivity index and their incidence correlate quite well with its values. These preliminary results could serve as the basis for further in vivo studies of the most promising compound (i.e. the complex 3) on the solid tumour models.
activity of both the prepared complexes and cisplatin, used as a standard. Other parameters, which were determined post mortem, comprise the weight of the tumour tissue and its weight ratio to the whole body weight (see Table 2). Both these parameters proved the aggressive cytotoxic action of cisplatin (elimination of more than 50% of tumour tissues, as compared to control group), while the effect of the complexes 1–3 was much more gradual (the complexes 2 and 3 caused the 2%, and 8% decrease of tumour tissue weight, respectively). The overall health status observations made during the whole time of the experiment showed that the animals in the groups treated by the complexes 1–3 showed no signs of toxicity or changes of normal behavior up to the 18th day of the experiment (in average the last day of the experiment), while the animals treated with cisplatin showed all known side effects of this therapy, i.e. loss of weight, fatigue, loss of appetite, various types of aberrant behavior. With respect to the given results we can conclude that the lower toxicity of the complexes 1–3 and less aggressive effect against tumour cells have to be considered as more beneficial for the treated animals than far more aggressive action of cisplatin. In addition to the macroscopic observations, much more information was obtained from the methods that evaluated the mechanisms of cytotoxicity, starting from the histological and histochemical observations, followed by the ELISA and Westernblot determination of the expression of selected proteins.
Macroscopic and Histological Observations All the animals implanted with the cell line L1210 expressed significant tumour infiltration of the abdominal cavity, peritoneal infiltration and well circumscribed tumours in the place of application, which were identified as anaplastic lymphoma. In all the animals, these tumours proliferated in the visceral fatty tissue and in the gonadal area, and formed well circumscribed focuses. Tumour cell dissemination of the diffuse type was found in GALT (gut-associated lymphoid tissue), the proliferation of the tumour tissues was found in the ileocaecal area and mesocolon. This type of tumour was identified according the WHO classification as diffuse B-lymphoma [43]. The infiltration by tumour cells was observed in several different organ preparations, but no metastatic focuses were identified. This observation is in agreement with the unchanged level of MMPs (see Figure S3 in File S1), which are responsible, at least in part, for metastasis formation [44].
ELISA and Westernblot Determination of the Expression of Selected Proteins A tumour blood supplementation is very important for its growth. Hence, the vascular endothelial growth factor A (VEGFA) was quantified in tumour lysates. No significant change was observed for this cytokine, only the complex 1 slightly decreased its amount in comparison with the untreated samples (see Figure S4 in File S1). A level of the initiator caspase 8 (Casp-8), effector caspase 3 (Casp-3), and regulatory protein p53 was also measured. The complex 1 significantly decreased the amount of the active form of Casp-8 (p18) by the factor of 1.8 in comparison with the untreated control tumours (Figure 7A). The complex 2 also reduced p18, but only 1.2-times without any statistical significance.
Evaluation of Cellular Processes and Semiquantitative Immunohistochemical Detection of Selected Proteins in Tumour Tissue Samples The results of antiproliferative or pro-apoptotic activity, and necrosis-induction and other physiopathological processes, as
Table 2. Selected parameters obtained from the pharmacological part of the in vivo L1210 antitumour activity model.
Compound
Control
1
2
3
Cisplatin
Body weight 6 S.E. (g)
24.7661.28
25.0660.85
23.0860.76
22.6061.49
14.8060.43
Weight of the cancer tissues 6 S.E. (g)
5.0660.58
5.0960.18
4.7560.15
4.2360.40
1.5060.01 10.1360.05
% of tumour tissues 6 S.E.
20.4562.35
20.3160.72
20.5960.63
18.7461.78
Mean survival time 6S.E. (days)
18.260.3
17.760.5
18.260.2
18.160.3
17.060.5
% T/C
100.0
97.1
100.0
99.4
93.3
doi:10.1371/journal.pone.0090341.t002
PLOS ONE | www.plosone.org
8
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
Figure 6. The Kaplan-Meier curves depicting the percentage of survival of animals in the individual groups. doi:10.1371/journal.pone.0090341.g006
sion, in comparison with untreated control, in the order of its absolute value: cisplatin ,3,2,1. In other words, the obtained results showed on different action of the complexes 1 and 2 as compared with the complex 3 and cisplatin in accordance with the above-mentioned results of both in vitro and in vivo studies. It is a well-established fact that antitumour active platinum(II) complexes covalently bind to the DNA molecule [46]. This mode of action was, together with notably higher cell-uptake and DNA platination as compared with cisplatin, also proved within the mechanistic studies performed very recently on the herein reported complexes with 7-azaindoles [23,24]. Platination of DNA molecule induces DNA damage which results in a cell cycle arrest or in apoptosis [47]. This effect is usually caused by the stabilization and activation of the transcription factor p53, which recognizes damaged DNA and is able to trigger the intrinsic (mitochondrial) apoptotic pathway in the case of massive DNA damage [48]. This mechanism is in concordance with our results, since we detected non-significantly higher amounts of p53 in tumours treated by both the most in vivo active compounds cisplatin and the complex 3. On the other hand, the complex 1 significantly decreased the p53 expression, but the amount of active effector Casp-3 remained unchanged. This effect could be caused by the activation of Casp-3 by p53-independent pathways, e.g. via p73 activation [49]. The complexes 1 and 2 also attenuated activation of the initiator Casp-8, which might indicate the triggering of apoptosis via an intrinsic pathway rather than extrinsic pathway [50]. Higher amounts of active Casp-3 after the complex 3 and cisplatin treatment correspond with TNF-a positivity observed in histological preparations. It has been described that macrophages surrounding cancer cells are able to produce antitumour cytokines, e.g. TNF-a and FasL, after cisplatin treatment in vitro [51]. These cytokines subsequently trigger apoptosis via activation of TNF-a and FasL receptors. The higher production of TNF-a, and the related inflammatory reaction related with it, is probably responsible for significantly elevated necrosis and hemorrhage after cisplatin and the complex 3 treatments.
On the other hand, the complex 3 and cisplatin slightly increased the level of active Casp-8 by the factor of 1.3. In comparison to cisplatin, the complexes 1 and 2 showed significant diminution of the Casp-8 level by the factor of 2.4, and 1.5, respectively. The complex 3 was the only compound which was able to significantly increase the level of the active form of the effector Casp-3 (Figure 7B), concretely 1.5-times in comparison to the untreated tumours. The complex 2 and cisplatin only weakly raised its amount by the factor of 1.2 and the complex 1 did not have any effect. The expression pattern for p53 was similar to Casp-8. The complex 1 significantly decreased the level of p53 by the factor of 1.8 in comparison to the untreated tumours (Figure 7C). The moderate diminution of p53 level was also detected for the complex 2 (1.4times). An increase of the p53 level was observed in the case of the complex 3 (1.4-times) and cisplatin (1.6-times). Finally, all the groups treated by the platinum(II) complexes showed lower mitotic index (MI; lower MI indicates tumour-static or tumour-toxic activity of the given complexes), which indicates tumour regresTable 3. Selected parameters obtained from the histological and immunohistochemical analyses of the in vivo antiproliferative assay.
Compound
Control
1
2
3
Cisplatin
Necrosis
2
2
2
3
4
Hemorrhage
1
3
3
4
3
Mitotic Index
56
48
44
36
23
Caspase 8
3
2
2
4
3
Caspase 3
1
4
3
4
1
p53
2
2
2
4
4
TNF-a
2
0
0
1
0
Total Score
11
13
12
20
15
Reactivity Index
5.60
3.69
3.67
1.71
1.35
doi:10.1371/journal.pone.0090341.t003
PLOS ONE | www.plosone.org
9
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
studies, which strengthened the evidence in the perspective of advanced testing of biological activities, were followed by in vivo and ex vivo trials based on the mouse L1210 lymphocytic leukemia model. The pharmacological observations have been complemented by the histological and immunohistochemical evaluations of isolated cancerous tissues and the expression of selected key proteins, associated with the tumour growth progression and induction of apoptosis (i.e. caspases 3 and 8, p53, VEGF-A were determined by ELISA and Western blot methods). The complexes 1–3 (applied at the same dose as cisplatin) showed that their effect on the reduction of cancerous tissues volume is markedly lower than that of cisplatin, however, they also proved to cause less serious adverse effects on the healthy tissues and the health status of the treated mice. The results of ex vivo assays revealed that above all others, the complex 3 was also able to effectively modulate the levels of active forms of caspases 3 and 8, and the transcription factor p53, and thus activate the intrinsic (mitochondrial) pathway of apoptosis. These results, unequivocally confirming the highest antitumour effect of the complex 3, were supported both by the histological and immunohistochemical observations. The overall positive results of antitumour activity testing were supported by the results of stability and interaction studies, which indicated the ability of the tested complexes to resist the formation of covalent adducts with low molecular sulfur-containing biomolecules in the biologically applicable media. The presented results unambiguously demonstrate that there is still a room for improvement of pharmacological properties of antitumour cis-dichloridoplatinum(II) complexes using the synthetic approach, based on the substitution of carrier ligands and that the 7-azaindole might serve as a viable basis for such studies.
Figure 7. The results of the studies of Casp-8, Casp-3 and p53 expression. (A): The amount of the active form of Casp-8 (p18) was detected by Westernblot and immunodetection. The results are presented as mean 6 S.E. ** significant difference in comparison to untreated cells (p,0.01), # significant difference in comparison to cisplatin-treated cells (p,0.05), ### significant difference in comparison to cisplatin-treated cells (p,0.001). (B): the amount of the active form of Casp-3 (p17) was detected by Westernblot and immunodetection. The results are presented as mean 6 S.E. ** significant difference in comparison to untreated cells (p,0.01). (C): Amount of p53 was detected by Westernblot and immunodetection. The results are presented as mean 6 S.E. ** significant difference in comparison to untreated cells (p,0.01), ### significant difference in comparison to cisplatin-treated cells (p,0.001). doi:10.1371/journal.pone.0090341.g007
Supporting Information Supporting information. The 1H, 13C, 15N and 195Pt NMR data assigned to trans-[PtCl2(naza)2] complexes (1t, 2t, 3t) detected as an impurity of the studied cis-[PtCl2(naza)2] complexes, the 1H, 13C, 15N and 195Pt NMR data for the products of hydrolysis of 1 and 3 in DMF-d7/H2O mixture (1 h and 3 h). Figure S1. The ESI-mass spectrum of the mixture of the complex 1 with cysteine and glutathione measured one month after the mixing of the components. Figure S2. ITC results showing the heat released during the titration of cysteine, GSH and HSA by 1 or cisplatin and the binding isotherms. Figure S3. The effect of 1– 3 and cisplatin on MMPs secretion in isolated tumours. Figure S4. The effect of 1–3 and cisplatin on VEGF-A secretion in isolated tumours. Table S1. The values of logIC50 (mM) for the complexes 1–3 and cisplatin. (DOCX) File S1
Conclusions In conclusion, this work represents a substantial contribution to the broad and systematic study of cisplatin-derived complexes involving the 7-azaindole derivatives (naza) for the identification of notable biological activities and molecular mechanisms interconnected with these activities. The presented data elaborate previous promising results of in vitro cytotoxicity screening carried out on a series of cis-[PtCl2(naza)2] complexes (1–3) and underlying mechanisms based on the interaction of the complexes with DNA. In this paper, we extended the scope of in vitro cytotoxicity testing to the expanded panel of human cancer cell lines (A549, HeLa, A2780 and cisplatin-resistant A2780R, and G-361) and found significant antitumour activity in vitro (the IC50 values were as low as 1 mM level), which surpassed that of cisplatin considerably. The complexes 1–3 also proved to be able to avoid the acquired resistance of the A2780R cell line to cisplatin. The in vitro
Acknowledgments We thank Mr. Alexandr Popa for his assistance with the preparation of the complexes and Mrs. Katerˇina Kubesˇova´ for help with in vitro cytotoxicity testing.
Author Contributions Conceived and designed the experiments: PSˇ JH JV ZD ZT. Performed the experiments: PSˇ JH JV PS IP GP. Analyzed the data: PSˇ JH JV ZD ZT. Wrote the paper: PSˇ JH JV ZT.
References 1. Kelland L (2007) The resurgence of platinum-based cancer chemotherapy. Nature Rev Cancer 7: 573–584.
PLOS ONE | www.plosone.org
2. Kelland LR, Farrell NP (2000) Platinum-Based Drugs in Cancer Therapy. Humana: Totowa.
10
March 2014 | Volume 9 | Issue 3 | e90341
Anticancer Pt(II) Complexes with 7-Azaindole
3. Rosenberg B, VanCamp L, Krigas T (1965) Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature 205: 698–699. 4. Rosenberg B, VanCamp L, Trosko JE, Mansour VH (1969) Platinum compounds - a new class of potent antitumour agents. Nature 222: 385–386. 5. Harrap KR (1985) Preclinical studies identifying carboplatin as a viable cisplatin alternative. Cancer Treat Rev 12: 21–33. 6. Akaza H, Togashi M, Nishio Y, Miki T, Kotake T, et al. (1992) Phase II study of cis-diammine(glycolato)platinum, 254–S, in patients with advanced germ-cell testicular cancer, prostatic cancer, and transitional-cell carcinoma of the urinary tract. Cancer Chemother Pharmacol 31: 187–192. 7. Kidani Y, Inagaki K, Iigo M, Hoshi A, Kuretani K (1978) Anti-tumor activity of 1,2-diaminocyclohexaneplatinum complexes against sarcoma-180 ascites form. J Med Chem 21: 1315–1318. 8. McKeage MJ (2001) Lobaplatin: a new antitumour platinum drug. Expert Opin Inv Drugs 10: 119–128. 9. Shen DW, Pouliot LM, Hall MD, Gottesman MM (2012) Cisplatin Resistance: A Cellular Self-Defense Mechanism Resulting from Multiple Epigenetic and Genetic Changes. Pharmacol Rev 64: 706–721. 10. Galanski M, Yasemi A, Slaby S, Jakupec MA, Arion VB, et al. (2004) Synthesis, crystal structure and cytotoxicity of new oxaliplatin analogues indicating that improvement of anticancer activity is still possible. Eur J Med Chem 39: 707– 714. 11. Holford J, Sharp SY, Murrer BA, Abrams M, Kelland LR (1998) In vitro circumvention of cisplatin resistance by the novel sterically hindered platinum complex AMD473. Br J Cancer 77: 366–373. 12. Holford J, Raynaud F, Murrer BA, Grimaldi K, Hartley JA, et al. (1998) Chemical, biochemical and pharmacological activity of the novel sterically hindered platinum coordination complex, cis-[amminedichloro(2-methylpyridine)] platinum(II) (AMD473). Anticancer Drug Design 13: 1–18. 13. Sharp SY, O’Neill CF, Rogers PM, Boxall FE, Kelland LR (2002) Retention of activity by the new generation platinum agent AMD0473 in four human tumour cell lines possessing acquired resistance to oxaliplatin. Eur J Cancer 38: 2309– 2315. 14. Wheate NJ, Walker S, Craig GE, Oun R (2010) The status of platinum anticancer drugs in the clinic and in clinical trials. Dalton Trans 39: 8113–8127. 15. Pogozheva D, Baudrona SA, Rogez G, Hosseini MW (2013) From discrete tricyanovinylene appended 7-azaindole copper(II) paddlewheel to an infinite 1D network: Synthesis, crystal structure and magnetic properties. Polyhedron 52: 1329–1335. 16. Tsoureas N, Nunn J, Bevis T, Haddow MF, Hamilton A, et al. (2011) Strong agostic-type interactions in ruthenium benzylidene complexes containing 7azaindole based scorpionate ligands. Dalton Trans 40: 951–958. 17. Ruiz J, Rodrı´guez V, de Haro C, Espinosa A, Pe´erez J, et al. (2010) New 7azaindole palladium and platinum complexes: crystal structures and theoretical calculations. In vitro anticancer activity of the platinum compounds. Dalton Trans 39: 3290–3301. 18. Pierce LT, Cahill MM, Winfield HJ, McCarthy FO (2012) Synthesis and identification of novel indolo[2,3-a]pyrimido[5,4-c]carbazoles as a new class of anti-cancer agents. Eur J Med Chem 56: 292–300. 19. Hong S, Kim J, Seo JH, Jung KH, Hong SS, et al. (2012) Design, Synthesis, and Evaluation of 3,5-Disubstituted 7-Azaindoles as Trk Inhibitors with Anticancer and Antiangiogenic Activities. J Med Chem 55: 5337–5349. 20. Cheve´ G, Bories C, Fauvel B, Picot F, Tible A, et al. (2012) De novo design, synthesis and pharmacological evaluation of new azaindole derivatives as dual inhibitors of Abl and Src kinases. Med Chem Commun 3: 788–800. 21. Dyke HJ, Price S, Williams K (2010) US Patent 20100216768. 22. Sˇtarha P, Marek J, Tra´vnı´cˇek Z (2012) Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations. Polyhedron 33: 404–409. 23. Sˇtarha P, Tra´vnı´cˇek Z, Popa A, Popa I, Muchova´ T, et al. (2012) How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole. J Inorg Biochem 115: 57–63. 24. Muchova´ T, Pracharˇova´ J, Sˇtarha P, Olivova´ R, Vra´na O, et al. (2013) Insight into the toxic effects of cis-dichloridoplatinum(II) complexes containing 7azaindole halogeno derivatives in tumor cells. J Biol Inorg Chem 18: 579–589. 25. Salemi G, Gueli MC, D’Amelio M, Saia V, Mangiapane P, et al. (2009) Blood levels of homocysteine, cysteine, glutathione, folic acid, and vitamin B12 in the acute phase of atherothrombotic stroke. Neurol Sci 30: 361–364. 26. Tra´vnı´cˇek Z, Matikova´-Malˇarova´ M, Novotna´ R, Vancˇo J, Sˇteˇpa´nkova´ K, et al. (2011) In vitro and in vivo biological activity screening of Ru(III) complexes involving 6-benzylaminopurine derivates with higher pro-apoptotic activity than NAMI-A. J Inorg Biochem 105: 937–948.
PLOS ONE | www.plosone.org
27. Talhouk R, Chin J, Unemori E, Werb Z, Bissell M (1991) Proteinases of the mammary gland: developmental regulation in vivo and vectorial secretion in culture. Development 112: 439–49. 28. Reedijk J (1999) Why does cisplatin reach guanine-N7 with competing S-donor ligands available in the cell? Chem Rev 99: 2499–2510. 29. Nagle PS, Rodriguez F, Nguyen B, Wilson WD, Rozas I (2012) High DNA Affinity of a Series of Peptide Linked Diaromatic Guanidinium-like Derivatives. J Med Chem 55: 4397–4406. 30. Spuches AM, Kruszyna HG, Rich AM, Wilcox DE (2006) Thermodynamics of the As(III)-Thiol Interaction: Arsenite and Monomethylarsenite Complexes with Glutathione, Dihydrolipoic Acid, and Other Thiol Ligands. Inorg Chem 44: 2964–2972. 31. Timerbaev AR, Hartinger CG, Aleksenko SS, Keppler BK (2006) Interactions of Antitumour Metallodrugs with Serum Proteins: Advances in Characterization Using Modern Analytical Methodology. Chem Rev 106: 2224–2248. 32. Esposito BA, Najjar R (2002) Interactions of antitumoural platinum-group metallodrugs with albumin. Coord Chem Rev 232: 137–149. 33. Brown SD, Trotter KD, Sutcliffe OB, Plumb JA, Waddell B, et al. (2012) Combining aspects of the platinum anticancer drugs picoplatin and BBR3464 to synthesize a new family of sterically hindered dinuclear complexes; their synthesis, binding kinetics and cytotoxicity. Dalton Trans 41: 11330–11339. 34. Carreira M, Calvo-Sanjua´n R, Sanau´ M, Zhao X, Magliozzo RS, et al. (2012) Cytotoxic hydrophilic iminophosphorane coordination compounds of d8 metals. Studies of their interactions with DNA and HSA. J Inorg Biochem 116: 204– 214. 35. Quirante J, Ruiz D, Gonzalez A, Lo´pez C, Cascante M, et al. (2011) Platinum(II) and palladium(II) complexes with (N,N9) and (C,N,N9)2 ligands derived from pyrazole as anticancer and antimalarial agents: Synthesis, characterization and in vitro activities. J Inorg Biochem 105: 1720–1728. 36. Carland M, Grannas MJ, Cairns MJ, Roknic VJ, Denny WA, et al. (2010) Substituted 9-aminoacridine-4-carboxamides tethered to platinum(II)diamine complexes: Chemistry, cytotoxicity and DNA sequence selectivity. J Inorg Biochem 104: 815–819. 37. Vrzal R, Sˇtarha P, Dvorˇa´k Z, Tra´vnı´cˇek Z (2010) Evaluation of in vitro cytotoxicity and hepatotoxicity of platinum(II) and palladium(II) oxalato complexes with adenine derivatives as carrier ligands. J Inorg Biochem 104: 1130–1132. 38. de Mier-Vinue´ J, Gay M, Montan˜a AM, Sa´ez RI, Moreno V, et al. (2008) Synthesis, biophysical studies, and antiproliferative activity of platinum(II) complexes having 1,2-bis(aminomethyl)carbobicyclic ligands. J Med Chem 51: 424–431. 39. Yamori T (2003) Panel of human cancer cell lines provides valuable database for drug discovery and bioinformatics. Cancer Chemother Pharmacol 52: S74–S79. 40. Margiotta N, Denora N, Ostuni R, Laquintana V, Anderson A, et al. (2010) Platinum(II) complexes with bioactive carrier ligands having high affinity for the translocator protein. J Med Chem 53: 5144–5154. 41. Xu G, Yan Z, Wamg N, Liu Z (2011) Synthesis and cytotoxicity of cisdichloroplatinum (II) complexes of (1S,3S)-1,2,3,4-tetrahydroisoquinolines. Eur J Med Chem 46: 356–363. 42. Komeda S, Takayama H, Suzuki T, Odani A, Yamori T, et al. (2013) Synthesis of antitumour azolato-bridged dinuclear platinum(II) complexes within vivo antitumour efficacy and uniquein vitro cytotoxicity profiles. Metallomics 5: 461– 468. 43. World Health Organization (2009) World Health Statistics 2009. Available: http://www.who.int/whosis/whostat/2009/en/. Accessed 2014 Feb 13. 44. Hua H, Li M, Luo T, Yin Y, Jiang Y (2011) Matrix metalloproteinases in tumourigenesis: an evolving paradigm. Cell Mol Life Sci 68: 3853–3868. 45. Zhang B, Ramesh G, Uematsu S, Akira S, Reevers WB (2008) TLR4 Signaling Mediates Inflammation and Tissue Injury in Nephrotoxicity. J Am Soc Nephrol 19: 923–932. 46. Reedijk J (1987) The mechanism of action of platinum anti-tumor drugs. Pure Appl Chem 59: 181–192. 47. Siddik ZH (2003) Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 22: 7265–7279. 48. Agarwal ML, Taylor WR, Chernov MV, Chernova OB, Stark GR (1998) The p53 Network. J Biol Chem 273: 1–4. 49. Gong JG, Costanzo A, Yang HQ, Melino G, Kaelin WG, et al. (1999) The tyrosine kinase c-Abl regulates p73 in apoptotic response to cisplatin-induced DNA damage. Nature 399: 806–809. 50. Hengartner MO (2000) The biochemistry of apoptosis. Nature 407: 770–776. 51. Chauhan P, Sodhi A, Shrivastava A (2009) Cisplatin primes murine peritoneal macrophages for enhanced expression of nitric oxide, proinflammatory cytokines, TLRs, transcription factors and activation of MAP kinases upon coincubation with L929cells. Immunobiology 214: 197–209.
11
March 2014 | Volume 9 | Issue 3 | e90341
PŘÍLOHA 5 Štarha, P.; Trávníček, Z.; Popa, I.; Dvořák, Z. Synthesis, characterization and in vitro antitumor activity of platinum(II) oxalato complexes involving 7-azaindole derivatives as coligands Molecules 19 (2014) 10832–10844
Molecules 2014, 19, 10832-10844; doi:10.3390/molecules190810832 OPEN ACCESS
molecules ISSN 1420-3049 www.mdpi.com/journal/molecules Article
Synthesis, Characterization and in Vitro Antitumor Activity of Platinum(II) Oxalato Complexes Involving 7-Azaindole Derivatives as Coligands Pavel Štarha 1, Zdeněk Trávníček 1,*, Igor Popa 1 and Zdeněk Dvořák 2 1
2
Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacký University, 17. listopadu 12, CZ 77146 Olomouc, Czech Republic; E-Mails:
[email protected] (P.S.);
[email protected] (I.P.) Regional Centre of Advanced Technologies and Materials, Department of Cell Biology and Genetics, Faculty of Science, Palacký University, Šlechtitelů 11, CZ 78371 Olomouc, Czech Republic; E-Mail:
[email protected]
* Author to whom correspondence should be addressed; E-Mail:
[email protected]; Tel.: +420-585-634-352; Fax: +420-585-634-954. Received: 7 July 2014; in revised form: 16 July 2014 / Accepted: 17 July 2014 / Published: 25 July 2014 Abstract: The platinum(II) oxalato complexes [Pt(ox)(naza)2] (1–3) were synthesized and characterized by elemental analysis (C, H, N), multinuclear NMR spectroscopy (1H, 13C, 15 N, 195Pt) and electrospray ionization mass spectrometry (ESI-MS); naza = 4-chloro-7azaindole (4Claza; 1), 3-bromo-7-azaindole (3Braza; 2) or 4-bromo-7-azaindole (4Braza; 3). The prepared substances were screened for their in vitro antitumor activity on the osteosarcoma (HOS) and breast adenocarcinoma (MCF7) human cancer cell lines, where 2 showed moderate antitumor effect (IC50 = 27.5 μM, and 18.3 μM, respectively). The complex 2 was further tested on a panel of six others human cancer cell lines, including the malignant melanoma (G361), cervix carcinoma (HeLa), ovarian carcinoma (A2780), cisplatin-resistant ovarian carcinoma (A2780R), lung carcinoma (A549) and prostate adenocarcinoma (LNCaP). This substance was found to be moderate antitumor effective against G361 (IC50 = 17.3 μM), HeLa (IC50 = 31.8 μM) and A2780 (IC50 = 19.2 μM) cell lines. The complex 2 was also studied by NMR for its solution stability and by ESI-MS experiments for its ability to interact with biomolecules, such as cysteine, glutathione or guanosine 5'-monophosphate.
Molecules 2014, 19
10833
Keywords: platinum(II) complexes; oxalato complexes; 7-azaindole derivatives; multinuclear NMR; antitumor activity
1. Introduction Platinum carboxylates represent a notable group of transition metal complexes, which have been used for the treatment of various types of cancer for many years [1–3]. Various carboxylate anions involved in the structures of the clinically used or studied platinum-based metallotherapeutics can be mentioned, particularly the cyclobutane-1,1-dicarboxylate dianion (involved in carboplatin), glycolate dianion (involved in nedaplatin), lactate dianion (involved in lobaplatin), malonate dianion (involved in heptaplatin), acetate anion (involved in satraplatin) or neodecanoate anion (involved in aroplatin) [4–9]. One of the carboxylate-based leaving groups is the oxalate dianion involved in the well-known substance oxaliplatin clinically used mainly for the treatment of colorectal tumours [10]. In other words, the platinum(II) oxalato complexes are, thanks to the mentioned oxaliplatin, biologically relevant and worth studying group of compounds. Search for novel antitumor active platinum complexes in terms of novel carrier (i.e., N-donor ligands) and leaving (i.e., carboxylates) ligands is one of the crucial challenge of modern bioinorganic chemistry [11–13], although those involved in clinically used drugs are still substantial part of this research, as exemplified by the NH3 carrier ligands involved in original platinum-based drug cisplatin, as well as in currently studied picoplatin [14]. As for the biologically effective platinum oxalato complexes, it is quite interesting that although these complexes can be considered, thanks to clinically used anticancer drug oxaliplatin [15], as biologically perspective group of compounds [16], not many papers dealing with such compounds in connection with their biological effect have been reported in last five years [17–24]. Our research group reported the platinum(II) oxalato complexes involving variously substituted N6-benzyladenine derivatives [17–19], whose the most effective representatives showed IC50(HOS) = 3.6 µM ([Pt(ox)(L1)]; L1=2-chloro-N6-(2-methoxybenzyl)-9-isopropyladenine) and IC50(MCF7) = 3.6 µM ([Pt(ox)(L2)]; L2=2-chloro-N6-(2,4-dimethoxybenzyl)-9-isopropyladenine). Utku et al. described the platinum(II) oxalato complexes with 2-phenylbenzimidazole ligand and their antibacterial and antifungal activity [20]. The work of Silva et al. dealt with the [Pt(ox)(L3)] (L3 symbolizes a long-chain aliphatic diamine) and their cytotoxic effect against A549, B16-F1, B16-F10 and MDA-MB-231 cancer cells and BHK-21 and CHO non-cancer cells. The obtained results proved four reported oxalato complexes as less active against the named cancer cells (IC50(A549) = 12.6–60.6 µM, IC50(B16-F1) = 14.5–21.9 µM, IC50(B16-F10) = 16.6–38.0 µM, IC50(MDA-MB-231) = 16.6–25.1 µM) as compared with cisplatin (IC50(A549) = 2.7 µM, IC50(B16-F1) = 3.5 µM, IC50(B16-F10) = 4.2 µM, IC50(MDA-MB-231) = 1.4 µM) [21]. Other platinum(II) oxalato complexes involved 1,2diaminocyclohexane with variously monoalkyl-substituted nitrogen atom (L4) [22]. These substances were studied for their in vitro cytotoxicity against HepG-2, MCF7, A549 and HCT-116 human cancer cell lines. The antitumor effect of the most effective complex, expressed as IC50 = 3.7, 8.8, 18.4, and 2.1 µM, respectively, against the mentioned cells, exceed both cisplatin and oxaliplatin used as standards in this study. The biological perspective of the oxalate anion, used as O-donor ligand in the
Molecules 2014, 19
10834
biologically active transition metal complexes, can be also demonstrated on the palladium(II) oxalato complexes with N6-benzyladenin-based N-donor ligands, which in many cases showed noticeable antitumor activity against human cancer cell lines in many cases [19,25,26]. The herein described platinum(II) oxalato complexes with variously substituted 7-azaindole halogeno-derivatives (1–3; Scheme 1), namely 4-chloro-7-azaindole (4Claza; the complex 1), 3-bromo-7-azaindole (3Braza; the complex 2) and 4-bromo-7-azaindole (4Braza; the complex 3), follow recently reported analogues involving differently substituted 7-azaindoles (3-chloro-7-azaindole, 3Claza; 3-iodo-7-azaindole, 3Iaza; 5-bromo-7-azaindole, 5Braza) [24]. Although the mentioned recently reported platinum(II) oxalato complexes with 3Claza, 3Iaza and 5Braza N-donor ligands did not show any biological effect on osteosarcoma HOS, breast carcinoma MCF7 and prostate carcinoma LNCaP human cancer cell lines (these complexes were poorly soluble in the medium used), we decided to study, whether substitution variability of the 7-azaindole moiety within the platinum(II) oxalato complexes may led to the formation of antitumor active compounds. Scheme 1. The synthesis and schematic representation of the prepared [Pt(ox)(naza)2] complexes (1–3) given with the atom numbering scheme of the used 7-azaindole derivatives.
R3=H and R4=Cl for 4Claza and [Pt(ox)(4Claza)2] (1); R3=Br and R4=H for 3Braza and [Pt(ox)(3Braza)2] (2) and R3=H and R4=Br for 4Braza and [Pt(ox)(4Braza)2] (3).
2. Results and Discussion 2.1. General Properties The platinum(II) oxalato complexes [Pt(ox)(naza)2] (1–3; Scheme 1) were prepared by a one-step synthetic procedure using the bis(oxalato)platinate(II) salt as a starting compound [17]. The composition of the products was proved by the results of elemental analysis. The electrospray ionization mass spectra obtained in the positive mode (ESI+) of the studied complexes contained the molecular peaks detected at 589.2 m/z (calc. 589.0; 5%; {[Pt(ox)(4Claza)2]+H}+ for 1), 678.9 m/z (calc. 678.9; 40%; {[Pt(ox)(3Braza)2]+H}+ for 2) and 679.0 m/z (calc. 678.9; 25%; {[Pt(ox)(4Braza)2]+H}+ for 3), their adducts with different cations, namely 627.1 m/z (calc. 626.9; 30%; {[Pt(ox)(4Claza)2]+K}+ for 1), 611.1 m/z (calc. 611.0; 50%; {[Pt(ox)(4Claza)2]+Na}+ for 1), 716.9 m/z (calc. 716.8; 100%; {[Pt(ox)(3Braza)2]+K}+ for 2), 701.0 m/z (calc. 700.9; 15%; {[Pt(ox)(3Braza)2]+Na}+ for 2), 716.9 m/z (calc. 716.8; 100%; {[Pt(ox)(4Braza)2]+K}+ for 3) and 701.1 m/z (calc. 700.9; 55%; {[Pt(ox)(4Braza)2]+Na}+ for 3). The peak of the released N-donor ligand was found at 153.1 m/z (calc. 153.0; 5%; {4Claza+H}+
Molecules 2014, 19
10835
for 1), 197.0 m/z (calc. 197.0; 20%; {3Braza+H}+ for 2) and 197.1 m/z (calc. 197.0; 15%; {4Braza+H}+ for 3). The pseudomolecular peaks of the {[Pt(ox)(naza)2]−H}– species were detected by means of ESI− (electrospray ionization in the negative mode) mass spectrometry at 585.9 m/z (calc. 586.0; 60%; {[Pt(ox)(4Claza)2]−H}− for 1) and 676.0 m/z (isomeric complexes 2 and 3; calc. 675.9; 100%; {[Pt(ox)(3Braza)2]−H}– for 2, and: calc. 675.9; 85%; {[Pt(ox)(4Braza)2]–H}− for 3). The ESImass spectra of the studied compounds also contain the peaks whose mass correspond to the {[Pt(ox)(naza)]−H}− fragment (434.0 m/z, calc. 434.0; 30%; {[Pt(ox)(4Claza)]−H}− for 1; 478.1 m/z, calc. 477.9; 30%; {[Pt(ox)(3Braza)]−H}− for 2; 478.1 m/z, calc. 477.9; 65%; {[Pt(ox)(4Braza)]−H}− for 3) and {naza−H}− ligand (151.0 m/z, calc. 151.0; 10%; {4Claza−H}− for 1; 195.1 m/z, calc. 195.0; 15%; {3Braza−H}− for 3; 195.1 m/z, calc. 195.0; 20%; {4Braza−H}− for 3). The complexes 1–3, as well as the starting compounds naza and K2[Pt(ox)2]∙2H2O, were studied by means of multinuclear and 2D NMR spectroscopy. All the 1H, 13C and 15N signals of free naza molecules were unambiguously detected in the spectra of corresponding platinum(II) complexes and assigned by means of the below-mentioned 2D NMR experiments. The significantly different 15 N-NMR coordination shift values of the N1 (|Δδ| = 2.5–3.4 ppm) and N7 (|Δδ| = 114.0–115.1 ppm) atoms clearly proved the coordination of naza molecules through the N7 atoms. The 13C-NMR chemical shifts of the C11 and C12 atoms of the oxalate dianion (detected at 165.9–166.0 ppm) and the 195 Pt chemical shifts, which equal −1770.1 (1), −1783.5 (2) and −1772.5 (3), correlate well with the values of the formerly reported platinum(II) oxalato complexes with 7-azaindole [23] or its derivatives [24]. 2.2. NMR and ESI-MS Stability and Interaction Studies 1
H and 195Pt NMR spectroscopy (solution of 2 in DMF-d7 and DMF-d7/H2O mixture, 9:1 v/v) and electrospray ionization mass spectrometry (ESI-MS; solution of 2 in methanol and water/methanol mixture, 1:1 v/v) (the presence of the organic solvents ensured the solubility of the studied complex, because carrying out of the experiments in water was prevented by limited solubility of the mentioned complex in water) were used to investigate the behaviour of the representative complex 2 in the mentioned organic or water-containing solvents. As it is generally accepted for the antitumor active platinum(II) complexes, hydrolysis is a crucial step within the mechanism of action, which lead to the replacement of leaving groups (i.e., the oxalato ligands) and formation of the activated and more reactive aqua- and/or hydroxidoplatinum(II) species, probably with formulas cis-[Pt(H2O)2(3Braza)2]2+ and/or cis-[Pt(OH)2(3Braza)2] in our case [1,2,27,28]. The hydrolysis of the platinum(II) oxalate complexes should be connected with opening of the PtO2C2 ring and/or substitution of the oxalate dianion by two H2O or OH− species as ligands, both resulting in the change of inner coordination sphere (resulting in new peaks in mass spectra) and electron density within the initial complex, which is known to provide different 1H and 195Pt NMR chemical shifts [27,28]. Since we did not observe any new signals in both the 1H and 195Pt NMR spectra (Figure 1), it can be concluded that the complex 2 is stable and do not undergo any changes within the structure during 5 days in DMF-d7 as well as in the DMF-d7/H2O mixture. Similarly it was found that the complex is stable and did not show any change in the composition from the mass spectrometry point of view, because its mass spectra (methanol solutions) recorded after 12 h did not contain any novel peaks as compared with the spectra obtained on the fresh solution of 2.
Molecules 2014, 19
10836
Figure 1. Time dependent 195Pt NMR spectra on the representative complex 2 dissolved in DMF-d7/H2O mixture (9:1 v/v) showing on the stability of the complex.
In the case of water/methanol mixture solution of 2, we detected several new peaks in the mass spectra in comparison with the spectra of 2 dissolved in pure methanol, but their isotopic distribution did not correspond to that of platinum-containing species (Figure 2). In other words, no new platinumcontaining species was found in the mass spectra recorded on the water/methanol solution of 2, which, as in the case of NMR, proved that no hydrolytic processes proceeded under the experimental condition used. Thus it can be said that we did not get any evidence of the hydrolysis usually involved within the mechanism of action of the cytotoxic active platinum(II) complexes including the clinically used platinum(II) oxalate complex oxaliplatin [1,2,29]. On the other hand, it is known that the oxaliplatin hydrolysis in water (leading to diaqua-species) and under in vivo conditions has different course, since the latter one provides the carbonato or phosphato adducts (instead of the above mentioned diaqua ones), which consequently enhance the reactivity of such species towards nucleophiles including nucleobases [30–32]. It means that although we did not observed any processes usually associated with the action of cytotoxic platinum(II) complexes (see Section 2.3. for the in vitro cytotoxicity of 1–3) under experimental conditions, the cytotoxic action itself is not excluded with respect to different conditions in the cells (cytosol involving various ions) as compared with those used in the herein discussed NMR and ESI-MS experiments. ESI-MS was also used to study the ability of 2 to interact with sulphur-containing biomolecules (cysteine (cys) and reduced glutathione (GSH)) or guanosine 5'-monophosphate (GMP) in water/methanol mixture (1:1 v/v) (again, the presence of the organic solvent ensured the solubility of the studied complex). It is well-known that ability of the cytotoxic platinum(II) complexes to interact with the intracellular sulphur-containing compounds correlates with their activity as well as with the resistance of the respective tumours in terms of inactivation of the platinum(II) species and their removing from the cell [1,29]. With respect to this phenomena, ability of the studied platinum(II) complexes to interact with the sulphur-containing biomolecules should be investigated by relevant techniques. In the case of this work, we studied an interaction of 2 with the mixture of cys and GSH in water/methanol mixture. We did not observe any adducts of 2 (or its fragments formed during ionization) with cysteine or reduced glutathione assignable to the species formed by their interaction (Figure 2), which corresponds to the above-described reluctance of 2 to undergo hydrolysis. The only exception from this statement is very weak peak of the {[Pt(cys)(ox)(3Braza)2]+H}+ species (Figure 2, inset) detected in the ESI+ spectra of the studied complex 2 at 799.5 m/z (calcd. 799.9 m/z), which most probably contains a ring-opened product of the interaction with a monodentate bound oxalate dianion.
Molecules 2014, 19
10837
Figure 2. The ESI+ mass spectra (recorded on fresh solutions and after 12 h) for the stability in methanol/water mixture, and for the interaction of the representative complex 2 with the mixture of cysteine and reduced glutathione (cys+GSH), and guanosine 5'-monophosphate (GMP), given together with the spectra of the fresh solutions of the individual reactants, i.e., the complex 2, cys+GSH mixture and GMP.
* stands for {cys+H}+, ♯ for {GSH+H}+,♦ for {H2GMP+H}+, {HNaGMP+H}+ and {Na2GMP+H}+, ○ for {3Braza+H}+ and ● for {[Pt(ox)(3Braza)2]+H}+ and {[Pt(ox)(3Braza)2]+Na}+; Inset: Part of the ESI+ mass spectrum of the mixture of the complex 2 with cys+GSH (recorded 12 h after the preparation) showing the peak of the {[Pt(cys)(ox)(3Braza)2]+H}+ species, together with the simulated isotopic distribution labeled as the green triangles.
The mechanism of action itself of the clinically used antitumor active platinum(II) complexes is based on the covalent binding of activated platinum(II) species to the nuclear DNA molecule of the tumour cells [33], which is also expected for most of the platinum(II) complexes having cytotoxic effect. A simple model to study the ability of the platinum(II) complexes to bind DNA molecule is based on binding reactions with various nucleobase-based compounds such as GMP employed in this work. However, we have to state that we did not detect any species whose mass and isotopic distribution would correspond with those of adduct of the studied complex, its fragments or its hydrolysis products with GMP (Figure 2). 2.3. In Vitro Cytotoxicity The complexes 1–3 were screened for their in vitro antitumor activity against two types of the human cancer cell lines - osteosarcoma (HOS) and breast adenocarcinoma (MCF7) (Table 1). In the
Molecules 2014, 19
10838
case of 1 and 3, the testing was limited by low solubility of the compounds, which can be expressed as <1.0 μM. Interestingly, the solubility of 2 was much higher (>50.0 μM) in the medium used. This compound showed the moderate in vitro anticancer activity, concretely 27.5 ± 3.4 μM against HOS and 18.3 ± 3.6 μM against MCF7 cells, which is comparable effect with that of cisplatin (IC50(HOS) = 25.4 ± 8.5 μM, IC50(MCF7) = 18.1 ± 5.1 μM) (Table 1, Figure 3). It has to be mentioned, that we also tried to compare the results of 2 with another platinum-based drug, oxaliplatin, which involve the same living group in its structure as the studied complexes 1–3. However, such comparison is limited by the fact that oxaliplatin did not show any cytotoxic effect on both cell lines up to the 50.0 μM concentration (IC50(HOS) > 50.0 μM, IC50(MCF7) > 50.0 μM). Nevertheless, it can be stated that the complex 2 exceeded the in vitro antitumor activity of oxaliplatin on HOS and MCF7 human cancer cell lines. Table 1. The results of the in vitro antitumor activity of the studied platinum(II) oxalato complexes (1–3), cisplatin (CDDP) and oxaliplatin (OXA) against osteosarcoma (HOS), breast adenocarcinoma (MCF7), malignant melanoma (G361), cervix carcinoma (HeLa), ovarian carcinoma (A2780), cisplatin-resistant ovarian carcinoma (A2780R), lung carcinoma (A549) and prostate carcinoma (LNCaP) human cancer cell lines, as obtained by an MTT assay on the cells exposed to the compounds for 24 h. HOS MCF7 1 >1.0 >1.0 2 27.5 ± 3.4 18.3 ± 3.6 3 >1.0 >1.0 CDDP 25.4 ± 8.5 18.1 ± 5.1 OXA >50.0 >50.0
G361 nt 17.3 ± 3.1 nt 5.8 ± 2.4 >50.0
HeLa nt 31.8 ± 6.2 nt 39.9 ± 4.6 >50.0
A2780 nt 19.2 ± 3.7 nt 21.8 ± 3.9 >50.0
A2780R nt >50.0 nt 32.0 ± 9.6 >50.0
A549 nt >50.0 nt >50.0 >50.0
LNCaP nt >50.0 nt 3.8 ± 1.5 >50.0
With respect to the results obtained on HOS and MCF7, the complex 2 was tested against next six human cancer cell lines, namely malignant melanoma (G361), cervix carcinoma (HeLa), ovarian carcinoma (A2780), cisplatin-resistant ovarian carcinoma (A2780R), lung carcinoma (A549) and prostate adenocarcinoma (LNCaP). It has been observed that its in vitro antitumor activity equals IC50 = 17.3 ± 3.1 μM (G361; 5.8 ± 2.4 μM for cisplatin, >50.0 μM for oxaliplatin), IC50 = 31.8 ± 6.2 μM (HeLa; 39.9 ± 4.6 μM for cisplatin, >50.0 μM for oxaliplatin), IC50 = 19.2 ± 3.7 μM (A2780; 21.8 ± 3.9 μM for cisplatin, >50.0 μM for oxaliplatin) and IC50 > 50.0 μM (A2780R, A549 and LNCaP; 32.0 ± 9.6, >50.0 and 3.8 ± 1.5 μM for cisplatin, respectively, >50.0 μM for oxaliplatin) cell lines (Table 1, Figure 3). The antitumor activity of 2 against G361, HeLa, A2780, A2780R, A549 and LNCaP can be evaluated as moderate and slightly higher on A2780 and HeLa in comparison with clinically used platinum-based therapeutic cisplatin and on G361, A2780 and HeLa as compared with another platinum-based drug oxaliplatin. As it is mentioned above, 1–3 follow recently reported [24] analogous oxalato complexes involving different types of 7-azaindole halogeno-derivatives, concretely 3-chloro-7-azaindole (3Claza), 3-iodo7-azaindole (3Iaza) and 5-bromo-7-azaindole (5Braza), which, similarly to 1 and 3, did not show any effect against both HOS and MCF7 cell lines up to the concentration of 10.0, 25.0, and 0.5 μM, respectively. Concerning all six platinum(II) oxalato complexes with different 7-azaindole derivatives together, it can be said that the biological activity, in terms of bioavailability, is strongly affected by
Molecules 2014, 19
10839
the position of halogeno-substituent of the 7-azaindole moiety, because the solubility of the complexes with 3Claza, 3Braza and 3Iaza (10.0–50.0 μM) is significantly higher than that of the complexes involving the 7-azaindole derivatives substituted in the position 4 (4Claza and 4Braza) or 5 (5Braza), whose solubility did not exceed 1.0 μM in the medium used. Figure 3. The in vitro antitumor activity of the complex 2 and cisplatin (CDDP) on osteosarcoma (HOS), breast adenocarcinoma (MCF7), malignant melanoma (G361), cervix carcinoma (HeLa) and ovarian carcinoma (A2780) human cancer cell lines.
Asterisk (*) symbolizes significant difference (p < 0.05) in in vitro antitumour activity of 2 as compared to cisplatin; nt = not tested.
3. Experimental Section 3.1. Materials and Methods Potassium tetrachloridoplatinate(II) (K2[PtCl4]), potassium oxalate monohydrate (K2(ox)∙H2O), 4-chloro-7-azaindole (4Claza), 3-bromo-7-azaindole (3Braza), 4-bromo-7-azaindole (4Braza), cisplatin, oxaliplatin, cysteine (cys), reduced glutathione (GSH), guanosine 5'-monophosphate disodium salt (GMP) and solvents were purchased from Sigma-Aldrich Co. (Prague, Czech Republic) and Acros Organics Co. (Pardubice, Czech Republic) and used as received. General methods: Elemental analysis was performed on a Flash 2000 CHNS Elemental Analyzer (Thermo Scientific, Waltham, MA, USA). Electrospray ionization mass spectrometry (ESI-MS): Mass spectra were obtained on fresh methanol solutions and after 2 h and 12 h by an LCQ Fleet ion trap mass spectrometer using both the positive (ESI+) negative (ESI−) mode electrospray ionization technique (Thermo Scientific, QualBrowser software, version 2.0.7, Thermo Fischer Scientific, Waltham, MA, USA). The 10 µM (final concentration) solution of 2 in methanol was mixed together with the same volume of water (hydrolysis studies; presence of methanol ensured the solubility of the studied complex, because carrying out of the experiments in water was prevented by limited solubility of the mentioned complex in water), water solutions of the mixture of GSH (6 µM) and cys (260 µM) or water solution of GMP. 20 µL of the mixtures was analysed by means of flow injection analysis/mass spectrometry (FIA/ESI-MS) in both the positive and negative ionization modes 0 h, 2 h and 12 h after the preparation.
Molecules 2014, 19
10840
NMR spectroscopy: 1H, 13C and 195Pt NMR spectra and two dimensional correlation 1H-1H gs-COSY, 1 H-13C gs-HMQC, 1H-13C gs-HMBC and 1H-15N gs-HMBC experiments (DMF-d7 solutions) were performed at 300 K on a Varian 400 device (Santa Clara, CA, USA) at 400.00 MHz (1H), 100.58 MHz (13C), 86.00 MHz (195Pt) and 40.53 MHz (15N); gs = gradient selected, COSY = correlation spectroscopy, HMQC = heteronuclear multiple quantum coherence, HMBC = heteronuclear multiple bond coherence. 1 H and 13C spectra were adjusted against SiMe4, while 195Pt spectra were calibrated against K2[PtCl6] in D2O found at 0 ppm. 1H-15N gs-HMBC experiments were obtained at natural abundance and calibrated against the residual signals of DMF (8.03 ppm for 1H, 104.7 ppm for 15N). The splitting of proton resonances in the reported 1H-NMR spectra is defined as s = singlet, d = doublet, t = triplet, br = broad band, m = multiplet. The coordination shift (Δδ; ppm) is calculated as Δδ = δcomplex − δligand. The stability studies of the DMF-d7 and DMF-d7/H2O (9:1 v/v) solutions of 2 was carried out by means of 1H and 195Pt NMR after 1, 2, 3, 4 and 5 days of standing at laboratory temperature. 3.2. Synthesis of Complexes 1–3 A solution of 1.0 mmol of 4Claza (for 1), 3Braza (for 2) or 4Braza (for 3) in 10 mL of hot (50 °C) ethanol was slowly poured into the solution of K2[Pt(ox)2]∙2H2O (0.5 mmol) in 10 mL of hot (50 °C) distilled water. The reaction mixtures were stirred at 50 °C for two days. The products, which formed, were filtered off, washed (5 mL of distilled water and 5 mL of ethanol) and dried at 40 °C (Figure 1). The described syntheses followed a procedure reported in our recent works for analogous complexes with different 7-azaindoles [23,24]. Bis(4-chloro-7-azaindole)-κN7}(oxalato-κ2O,O’)platinum(II) (1, C16H10N4Cl2O4Pt) Light grey solid; yield 80%; 1H-NMR (400.0 MHz, DMF-d7): δ/Δδ = 13.41/1.34 (br, NH-1), 8.71/0.49 (d, J = 6.4, CH-6), 7.98/0.30 (d, J = 3.6, CH-2), 7.37/0.17 (d, J = 6.3, CH-5), 6.79/0.21 (d, J = 3.6, CH-3) ppm. 13 C-NMR (100.6 MHz, DMF-d7): δ/Δδ = 165.9 (C-11,12), 148.2/−1.6 (C-7a), 146.6/3.0 (CH-6), 138.4/3.8 (C-4), 129.3/1.9 (CH-2), 122.2/3.1 (C-3a), 117.5/1.9 (CH-5), 100.5/2.2 (CH-3) ppm. 15 N-NMR (40.5 MHZ, DMF-d7): δ/Δδ = 145.6/3.4 (NH-1), 154.6/−114.6 (N-7) ppm. 195Pt NMR (86.0 MHz, DMF-d7): δ = −1770.1 ppm. ESI MS (30 ev): m/z = 611.1 (M+Na), 585.9 (M−H), 434.0 (M−4Claza−H), 153.1 (4Claza+H), 151.0 (4Claza−H). Anal. Calc.: C, 32.7%; H, 1.7%; N, 9.5%. Found: C, 32.8%; H, 1.6%; N, 9.6%. Bis(3-bromo-7-azaindole)-κN7}(oxalato-κ2O,O’)platinum(II) (2, C16H10N4Br2O4Pt) Light grey solid; yield 75%; 1H-NMR (400.0 MHz, DMF-d7): δ/Δδ = 13.49/1.32 (br, NH-1), 8.82/0.48 (d, J = 5.7, CH-6), 8.11/0.23 (d, J = 8.0, CH-4), 8.09/0.31 (s, CH-2), 7.31/0.10 (m, CH-5) ppm. 13C-NMR (100.6 MHz, DMF-d7): δ/Δδ = 166.0 (C-11,12), 147.5/3.2 (CH-6), 147.0/−0.9 (C-7a), 130.6/4.0 (CH-4), 128.0/2.1 (CH-2), 122.6/3.3 (C-3a), 118.0/1.4 (CH-5), 89.3/1.6 (C-3) ppm. 15N-NMR (40.5 MHZ, DMF-d7): δ/Δδ = 143.5/2.5 (NH-1), 159.4/−115.1 (N-7). 195Pt NMR (86.0 MHz, DMF-d7): δ = −1783.5 ppm. ESI MS (30 ev): m/z = 678.9 (M+H), 676.0 (M−H), 478.1 (M−3Braza−H), 197.1 (3Braza+H), 195.1 (3Braza–H). Anal. Calc.: C, 28.4%; H, 1.5%; N, 8.3%. Found: C, 28.4%; H, 1.5%; N, 8.3%. Bis(4-bromo-7-azaindole)-κN7}(oxalato-κ2O,O’)platinum(II) (3, C16H10N4Br2O4Pt) Light grey solid; yield 80%; 1H-NMR (400.0 MHz, DMF-d7): δ/Δδ = 13.40/1.31 (br, NH-1), 8.60/0.47 (d, J = 6.3, CH-6),
Molecules 2014, 19
10841
8.00/0.30 (d, J = 3.5, CH-2), 7.51/0.16 (d, J = 6.2, CH-5), 6.72/0.21 (d, J = 3.5, CH-3) ppm. 13C-NMR (100.6 MHz, DMF-d7): δ/Δδ = 165.9 (C-11,12), 147.3/−1.6 (C-7a), 146.3/2.9 (CH-6), 129.4/1.9 (CH-2), 128.0/4.1 (C-3a), 124.6/3.2 (C-4), 120.6/1.8 (CH-5), 102.1/2.2 (CH-3) ppm. 15N-NMR (40.5 MHZ, DMF-d7): δ/Δδ = 145.5/3.1 (NH-1), 155.4/−114.0 (N-7). 195Pt NMR (86.0 MHz, DMF-d7): δ = −1772.5 ppm. ESI MS (30 ev): m/z = 679.0 (M+H), 676.0 (M−H), 478.1 (M−4Braza−H), 197.1 (4Braza+H), 195.1 (4Braza−H). Anal. Calc.: C, 28.4%; H, 1.5%; N, 8.3%. Found: C, 28.3%; H, 1.5%; N, 8.4%. 3.3. In Vitro Cytotoxicity Testing Breast adenocarcinoma (MCF7; ECACC No. 86012803), osteosarcoma (HOS; ECACC No. 87070202), malignant melanoma (G361; ECACC No. 88030401), cervix epitheloid carcinoma (HeLa; ECACC No. 93021013), A2780 ovarian carcinoma (ECACC No. 93112519), A2780R cisplatin-resistant ovarian carcinoma (ECACC No.93112517), lung carcinoma (A549; ECACC No.86012804) and prostate adenocarcinoma (LNCaP; ECACC No. 89110211) cancer cell lines were purchased from European Collection of Cell Cultures (ECACC; Prague, Czech Republic). In vitro cytotoxicity was determined by an MTT assay against MCF7 (the complexes 1–3), HOS (1–3), G361 (2), HeLa (2), A2780 (2), A2780R (2), A549 (2) and LNCaP (2) human cancer cell lines. The cells were maintained in a humidified incubator (37 °C, 5% CO2). The cells were treated with 1–3 or standards (cisplatin, oxaliplatin) at the 0.01, 0.1, 1.0, 5.0, 25.0 and 50 μM concentrations for 24 h, using multi-well culture plates of 96 wells. In parallel, the cells were treated with vehicle (DMF; 0.1%, v/v) and Triton X-100 (1%, v/v) to assess the minimal (i.e., positive control) and maximal (i.e., negative control) cell damage, respectively. The MTT assay was measured spectrophotometrically at 540 nm (TECAN, Schoeller Instruments LLC). The data were expressed as the percentage of viability, when 100% and 0% represent the treatments with DMF and Triton X-100, respectively. The cytotoxicity data from the cancer cell lines were acquired from three independent experiments (conducted in triplicate) using cells from different passages. The IC50 values (µM) were calculated from viability curves. The results are presented as arithmetic mean ± SD. The significance of the differences between the results was assessed by the ANOVA analysis, followed by Tukey’s post-hoc test for multiple comparisons, with p < 0.05 considered to be significant (QC Expert 3.2, Statistical software, TriloByte Ltd., Pardubice, Czech Republic). 4. Conclusions This work describes three new platinum(II) oxalato complexes [Pt(ox)(naza)2] (1–3) and evaluates their in vitro cytotoxicity on the selected human cancer cell lines. The testing revealed the complex 2 (involving 3Braza) as in vitro antitumor active against HOS, MCF7, G361, HeLa and A2780 with IC50 ≈ 17–32 µM. Because the complex 2 differs from 1 (involving 3Claza) and 3 (involving 4Braza) in the nature or position of the 7-azaindole moiety substituent, it can be concluded that 3Braza represents a very perspective N-donor ligand, which could be used as a carrier ligand involved into platinum(II) complexes with different leaving group than oxalato one.
Molecules 2014, 19
10842
Acknowledgments The authors thank the Czech Science Foundation (GAČR P207/11/0841), Operational Program Research and Development for Innovations—European Regional Development Fund (CZ.1.05/2.1.00/03.0058) of the Ministry of Education, Youth and Sports of the Czech Republic, and Palacký University in Olomouc (PrF_2013_015 and PrF_2014_009). The authors also thank Bohuslav Drahoš and Ján Vančo for help with ESI-MS experiments, Radka Křikavová for assistance with NMR experiments, and Alexandr Popa for collaboration on the syntheses. Author Contributions Conceived and designed the experiments: PŠ, ZT, IP, ZD. Performed the experiments: PŠ, ZT, IP, ZD. Analyzed the data: PŠ, ZT, IP, ZD. Wrote the paper: PŠ, ZT. Conflicts of Interest The authors declare no conflict of interest. References 1. 2. 3. 4. 5.
6. 7.
8.
9.
Kelland, L. The resurgence of platinum-based cancer chemotherapy. Nat. Rev. Cancer 2007, 7, 573–584. Kelland, L.R.; Farrell, N.P. Platinum Based Drugs in Cancer Therapy; Humana Press: Totowa, NJ, USA, 2000. Gielen, M.; Tiekink, E.R.T. Metallotherapeutic Drugs and Metal-Based Diagnostic Agents; John Wiley & Sons, Ltd.: Chichester, UK, 2005. Harrap, K.R. Preclinical studies identifying carboplatin as a viable cisplatin alternative. Cancer Treat. Rev. 1985, 12, 21–33. Akaza, H.; Togashi, M.; Nishio, Y.; Miki, T.; Kotake, T.; Matsumura, Y. Yoshida, O.; Aso, Y. Phase II study of cis-diammine(glycolato)platinum, 254-S, in patients with advanced germ-cell testicular cancer, prostatic cancer, and transitional-cell carcinoma of the urinary tract. Cancer Chemoth. Pharm. 1992, 31, 187–192. McKeage, M.J. Lobaplatin: A new antitumour platinum drug. Expert Opin. Inv. Drugs 2001, 10, 119–128. Kim, D.K.; Kim, G.; Gam, J.; Cho, Y.B.; Kim, H.T.; Tai, J.H.; Kim, K.H.; Hong, W.S.; Park, J.G. Synthesis and antitumor activity of a series of [2-substituted-4,5-bis(aminomethyl)-1,3-dioxolane] platinum(II) complexes. J. Med. Chem. 1994, 37, 1471–1485. Kelland, L.R.; Abel, G.; McKeage, M.J.; Jones, M.; Goddard, P.M.; Valenti, M.; Murrer, B.A.; Harrap, K.R. Preclinical antitumor evaluation of bis-acetato-ammine-dichloro-cyclohexylamine platinum(IV): An orally active platinum drug. Cancer Res. 1993, 53, 2581–2583. Dragovich, T.; Mendelson, D.; Kurtin, S.; Richardson, K.; Von Hoff, D.; Hoos, A. A Phase 2 trial of the liposomal DACH platinum L-NDDP in patients with therapy-refractory advanced colorectal cancer. Cancer Chemoth. Pharm. 2006, 58, 759–764.
Molecules 2014, 19
10843
10. Kidani, Y.; Inagaki, K.; Iigo, M.; Hoshi, A.; Kuretani, K. Antitumor activity of 1,2diaminocyclohexaneplatinum complexes against Sarcoma-180 ascites form. J. Med. Chem. 1978, 21, 1315–1318. 11. Zhang, J.C.; Liu, D.D.; Li, Y.P.; Sun, J.; Wang, L.W.; Zang, A.M. Status of Non-classical mononuclear platinum anticancer drug development. Mini-Rev. Med. Chem. 2009, 9, 1357–1366. 12. Butler, J.S.; Sadler, P.J. Targeted delivery of platinum-based anticancer complexes. Curr. Opin. Chem. Biol. 2013, 17, 175–188. 13. Harper, B.W.; Krause-Heuer, A.M.; Grant, M.P.; Manohar, M.; Garbutcheon-Singh, K.B.; Aldrich-Wright, J.R. Advances in platinum chemotherapeutics. Chem. Eur. J. 2010, 16, 7064–7077. 14. Holford, J.; Sharp, S.Y.; Murrer, B.A.; Abrams, M.; Kelland, L.R. In vitro circumvention of cisplatin resistance by the novel sterically hindered platinum complex AMD473. Br. J. Cancer 1998, 77, 366–373. 15. Stein, A.; Arnold, D. Oxaliplatin: A review of approved uses. Expert Opin. Pharmacother. 2012, 13, 125–137. 16. Cleare, M.J. Transition metal complexes in cancer chemotherapy. Coord. Chem. Rev. 1974, 12, 349–405. 17. Štarha, P.; Trávníček, Z.; Popa, I. Platinum(II) oxalato complexes with adenine-based carrier ligands showing significant in vitro antitumor activity. J. Inorg. Biochem. 2010, 104, 639–647. 18. Trávníček, Z.; Štarha, P.; Popa, I.; Vrzal, R.; Dvořák, Z. Roscovitine-based CDK inhibitors acting as N-donor ligands in the platinum(II) oxalato complexes: Preparation, characterization and in vitro cytotoxicity. Eur. J. Med. Chem. 2010, 45, 4609–4614. 19. Vrzal, R.; Štarha, P.; Dvořák, Z.; Trávníček, Z. Evaluation of in vitro cytotoxicity and hepatotoxicity of platinum(II) and palladium(II) oxalato complexes with adenine derivatives as carrier ligands. J. Inorg. Biochem. 2010, 104, 1130–1132. 20. Utku, S.; Topal, M.; Dögen, A.; Serin, M.S. Synthesis, characterization, antibacterial and antifungal evaluation of some new platinum(II) complexes of 2-phenylbenzimidazole ligands. Tur. J. Chem. 2010, 34, 427–436. 21. Silva, H.; Barra, C.V.; Rocha, F.V.; Frézard, F.; Lopes, M.T.P.; Fontes, A.P.S. Novel platinum(II) complexes of long chain aliphatic diamine ligands with oxalato as the leaving group. Comparative cytotoxic activity relative to chloride precursors. J. Braz. Chem. Soc. 2010, 21, 1961–1967. 22. Sun, Y.; Yin, R.; Gou, S.; Zhao, J. Antitumor platinum(II) complexes of N-monoalkyl-1R, 2R-diaminocyclohexane derivatives with alkyl groups as hindrance. J. Inorg. Biochem. 2012, 112, 68–76. 23. Štarha, P.; Marek, J.; Trávníček, Z. Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations. Polyhedron 2012, 33, 404–409. 24. Štarha, P.; Trávníček, Z.; Popa, A.; Popa, I.; Muchová, T.; Brabec, V. How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole. J. Inorg. Biochem. 2012, 105, 57–63. 25. Štarha, P.; Trávníček, Z.; Popa, I. Synthesis, characterization and in vitro cytotoxicity of the first palladium(II) oxalato complexes involving adenine-based ligands. J. Inorg. Biochem. 2009, 103, 978–988.
Molecules 2014, 19
10844
26. Štarha, P.; Popa, I.; Trávníček, Z. Palladium(II) oxalato complexes involving N6-(benzyl)-9isopropyladenine-based N-donor carrier ligands: Synthesis, general properties, 1H, 13C and 15 N{1H} NMR characterization and in vitro cytotoxicity. Inorg. Chim. Acta 2010, 363, 1469–1478. 27. Siddik, Z.H. Cisplatin: Mode of cytotoxic action and molecular basis of resistance. Oncogene 2003, 22, 7265–7279. 28. Berners-Price, S.J.; Ronconi, L.; Sadler, P.J. Insights into the mechanism of action of platinum anticancer drugs from multinuclear NMR spectroscopy. Prog. Nucl. Mag. Res. Sp. 2006, 49, 65–98. 29. Mistry, P.; Kelland, L.R.; Abel, G.; Sidhar, S.; Harrap, K.R. The relationships between glutathione, glutathione-S-transferase and cytotoxicity of platinum drugs and melphalan in eight human ovarian carcinoma cell lines. Br. J. Cancer 1991, 64, 215–220. 30. Luo, F.R.; Yen, T.Y.; Wyrick, S.D.; Chaney, S.G. High-performance liquid chromatographic separation of the biotransformation products of oxaliplatin. J. Chromatogr. B 1999, 724, 345–356. 31. Ravera, M.; Bagni, G.; Mascini, M.; Dabrowiak, J.C.; Osella, D. The activation of platinum(II) antiproliferative drugs in carbonate medium evaluated by means of a DNA-biosensor. J. Inorg. Biochem. 2007, 101, 1023–1027. 32. Kim, Y.S.; Shin, S.; Cheong, M.; Hah, S.S. Mechanistic Insights into in vitro DNA Adduction of Oxaliplatin. Bull. Korean Chem. Soc. 2010, 31, 2043–2046. 33. Reedijk, J. Increased understanding of platinum anticancer chemistry. Pure Appl. Chem. 2011, 83, 1709–1719. Sample Availability: Samples of the compounds 1–3 are available from the authors. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).
PŘÍLOHA 6 Štarha, P.; Trávníček, Z.; Dvořák, Z.; Radošová-Muchová, T.; Prachařová, J.; Vančo, J.; Kašpárková, J. Potentiating effect of UVA irradiation on anticancer activity of carboplatin derivatives involving 7-azaindoles. PLoS ONE (2015) akceptovaný manuskript
1
Potentiating effect of UVA irradiation on anticancer
2
activity of carboplatin derivatives involving
3
7-azaindoles
4 5
Pa el Šta ha1, )de ěk T á íček1,*, )de ěk D ořák2, Te eza Radošo á-Mu ho á3,
6
Jitka P a hařo á4, Já Va čo1, Ja a Kašpá ko á4
7 8
1
9
Inorganic Chemistry, Fa ulty of S ie e, Pala ký U iversity, Olo ou , Cze h Repu li
Regional Centre of Advanced Technologies and Materials & Department of
10
2
11
Biology and Genetics, Faculty of S ie e, Pala ký U iversity, Olomouc, Czech Republic
12
3
13
Department of Biophysics, Faculty of Science, Pala ký U iversity, Olomouc, Czech
14
Republic
15
4
16
Republic
Regional Centre of Advanced Technologies and Materials & Department of Cell
Ce tre of the Regio
Ha á for Biote h ologi al a d Agri ultural Resear h &
Department of Biophysics, Faculty of S ie e, Pala ký U iversity, Olo ou , Cze h
17 18
* Corresponding author
19
E-mail:
[email protected] (ZT)
20
-1-
21
A stra t
22
The moderate-to-high in vitro cytotoxicity against ovarian A2780 (IC50 = 4.7–14.4
23
µM , p ostate LNCaP IC50 = 18.7–
24
human cancer cell lines of the platinum(II) cyclobutane-1,1'-dicarboxylato complexes
25
Pt(cbdc)(naza)2] (1–6; cbdc = cyclobutane-1,1'-dicarboxylate(2-); naza = halogeno-
26
substituted 7-azaindoles), derived from the anticancer metallodrug carboplatin, are
27
reported. The complexes containing the chloro- and bromo-substituted 7-azaindoles
28
(1, 2, and 4–6) showed a significantly higher (p < 0.05) cytotoxicity against A2780 cell
29
line as compared to cisplatin used as a reference drug. Addition of the non-toxic
. µM a d p ostate PC-3 (IC50 = 17.6–
. µM
30
o e t atio
. µM of L-buthionine sulfoximine (L-BSO, an effective inhibitor of
31
-glutamylcysteine synthase) markedly increases the in vitro cytotoxicity of the
32
selected complex 3 against A2780 cancer cell line by a factor of about 4.4. The
33
cytotoxicity against A2780 and LNCaP cells, as well as the DNA platination, were
34
effe ti el e ha ed
35
the highest phototoxicity determined for compound 3, resulting in a 4-fold decline in
36
the A2780 cells viability from 25.1% to 6.1%. The 1H NMR and ESI-MS experiments
37
suggested that the complexes did not interact with glutathione as well as their ability
38
to interact with guanosine monophosphate. The studies also confirmed UVA light
39
induced the formation of the cis-[Pt(H2O)2(cbdc`)(naza)] intermediate, where cbdc`
40
represents monodentate-coordinated cbdc ligand, which is thought to be
41
responsible for the enhanced cytotoxicity. This is further supported by the results of
42
transcription
UVA light i adiatio
max
= 365 nm) of the complexes, with
mapping experiments showing that
-2-
the
studied complexes
43
preferentially form the bifunctional adducts with DNA under UVA irradiation, in
44
contrast to the formation of the less effective monofunctional adducts in dark.
45
46
Keywords: Carboplatin; Platinum(II) complexes; Cyclobutane-1,1'-dicarboxylato; 7-
47
Azaindole; Cytotoxicity; Phototoxicity.
48
-3-
49
I trodu tio
50
Although the well-known story of platinum-based anticancer metallotherapeutics
51
have slowly reached their second half-century, the application, development and
52
research is still one of the leading branches of bioinorganic chemistry [1–3].
53
However, there is still room for improvement with regard to the therapeutic effects
54
of platinum-based antitumor active complexes, combined with the suppression of
55
negative side-effects (e.g. nephrotoxicity, neurotoxicity or myelosuppression) and/or
56
ability to overcome both the intrinsic or acquired resistance of various tumor cells
57
against chemotherapeutics [2–5]. One of the current approaches for reaching the
58
aforementioned objectives is based on the irradiation at selected wavelengths of
59
light and converting the initially inactive drugs into significantly enhanced cytotoxics
60
[6-8]. Recently, the considerably increased ability of the 2nd generation platinum-
61
based anticancer drug carboplatin to bind to the DNA upon UVA irradiation, resulting
62
in increased cytotoxicity, was reported [8].
63
Carboplatin represents a complex, whose composition offers the possibility of
64
facile derivatization. The approach is based on the derivatization of the cyclobutane-
65
1,1'-dicarboxylate(2-) (cbdc) (e.g. the diammineplatinum(II) complexes with the
66
furoxan-substituted cyclobutane moiety [9]), while the second involves the
67
replacement of the NH3 carrier-ligands (e.g. with adenine-based N-donor ligands
68
[10]). In this work, the second approach was applied to yield a series of cyclobutane-
69
1,1'-dicarboxylatoplatinum(II) complexes where both the NH3 ligands are substituted
70
by various 7-azaindoles (naza). 7-Azaindole was recently used as a suitable N-donor
71
carrier ligand of various types of antitumor active platinum(II) complexes, and a -4-
72
number of dichlorido [11–13], mixed-ligand [14,15] and oxalato [11] complexes have
73
been reported to date. The herein presented complexes represent a logical step
74
towards the extension of the group of dichlorido and oxalato platinum(II) complexes,
75
involving the analogical halogeno-derivatives of 7-azaindole, recently developed by
76
our research group [11–13]. In the case of dichlorido complexes, considerably high in
77
vitro cytotoxicity (with IC50 values up to 0.6 µM) was found against various human
78
cancer cell lines (ovarian A2780, breast MCF7, osteosarcoma HOS, lung A549,
79
cervical HeLa, malignant melanoma G361 and prostate LNCaP). These cisplatin
80
analogues complexes also successfully overcame an acquired resistance to cancer
81
cells (ovarian carcinoma model) and effectively reduced the tumor tissues volume
82
during the in vivo experiments on mice (L1210 lymphocytic leukemia model), while
83
showing less serious negative side-effects on the healthy tissues as compared with
84
cisplatin [13]. The above-mentioned positive findings, regarding the in vitro and in
85
vivo anticancer activities of platinum(II) complexes bearing 7-azaindole monodentate
86
ligands, motivated us to study the carboplatin analogues involving the mentioned N-
87
donor ligands (Fig. 1), their cytotoxicity on selected human cancer cell lines and
88
mechanisms of their action under normal conditions and upon UVA light irradiation,
89
using the set of advanced analytical and biological methods.
90 91
Fig. 1. Structural formula of the studied complexes [Pt(cbdc)(naza)2] (1–6). The
92
formula is given together with their atom numbering scheme; R3, R4, R5 = Cl, H, H (3-
93
chloro-7-azaindole (3Claza), involved in complex 1); Br, H, H (3-bromo-7-azaindole
94
(3Braza), 2); I, H, H (3-iodo-7-azaidole (3Iaza), 3); H, Cl, H (4-chloro-7-azaindole
-5-
95
(4Claza), 4); H, Br, H (4-bromo-7-azaindole (4Braza), 5); and H, H, Br (5-bromo-7-
96
azaindole (5Braza), 6).
97
98
Materials a d Methods
99
Chemicals and Biochemicals
100
The reagents (K2[PtCl4], 3Claza, 3Braza, 3Iaza, 4Claza, 4Braza and 5Braza,
101
cyclobutane-1,1'-dicarboxylic
102
dimethylformamide (DMF), acetone, methanol, diethyl ether) and other chemical
103
(reduced glutathione (GSH), guanosine 5'-monophosphate disodium salt hydrate
104
(GMP)) were supplied by Sigma-Aldrich (Prague, Czech Republic) and Acros Organics
105
(Pardubice, Czech Republic). Calf thymus DNA (CT DNA; 42% G:C, mean molecular
106
mass of approximately 20,000 kDa) was isolated as previously described [16].
107
Sephadex G-50 (coarse) was from Sigma-Aldrich (Prague, Czech Republic). MTT, 3-
108
(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, was from Calbiochem
109
(Darmstadt, Germany). RPMI 1640 medium, fetal bovine serum, trypsin/EDTA, and
110
Dul e o’s
111
was from Serva (Heidelberg, Germany).
odified Eagle’s
acid,
ediu
NaOH,
eefo
AgNO3),
solvents
(N,N’-
PAA Pas hi g, Aust ia . Ge ta i i
112
113
General Method for the Synthesis of 1–6
114
The platinum(II) cyclobutane-1,1'-dicarboxylato complexes, [Pt(cbdc)(naza)2] (1–6;
115
naza = 3Claza for 1, 3Braza for 2, 3Iaza for 3, 4Claza for 4, 4Braza for 5 and 5Braza for
-6-
116
6), were prepared by well-established Dhara`s method [17]. Briefly, 1.0 mmol (415
117
mg) of K2[PtCl4] was dissolved in 15 mL of deionized water at room temperature and
118
KI (830 mg; 5.0 mmol) was added. The solution turned black during 1 h of stirring at
119
room temperature and then 2.0 mmol of naza dissolved in 15 mL of methanol were
120
poured in. The mixture was stirred overnight at room temperature and the obtained
121
yellow solid, i.e. cis-[PtI2(naza)2] (yields 90%), was filtered off and washed with
122
deio ized
123
over silica gel. The obtained platinum(II) diiodido complexes (0.5 mmol) were
124
dissolved in DMF (5 mL) and silver(I) cyclobutane-1,1'-dicarboxylate (0.5 mmol) was
125
added into the solution. The mixtures were stirred at room temperature and in the
126
dark for 48 h. The formed AgI precipitate was collected a d ashed ith DMF
127
mL). Deionized water (25 mL) was poured into the filtrate and the obtained white
128
precipitate was removed by filtration and washed successively with deionized water
129
×
L,
ate
×
etha ol
L a d
×
etha ol
×
L , d ied a d sto ed i desi ato
L a d dieth l ethe
×
×
L . The p odu ts of
130
[Pt(cbdc)(naza)2] (1–6; Fig. 1) were dried in desiccator over silica gel and stored
131
without any further purification. Characterization data for 1–6 are given in
132
Supporting Information (S1 Text).
133
134
Physical Measurements
135
A combustion analysis (C, H, N) was performed using a Flash 2000 CHNS Elemental
136
Analyzer (Thermo Scientific). Electrospray ionization mass spectroscopy (ESI-MS) of
137
the methanol solutions was performed on an LCQ Fleet Ion Trap mass spectrometer
138
(Thermo Scientific; QualBrowser software, version 2.0.7) in both the positive (ESI+) -7-
139
and negative (ESI–) ionization modes. Infrared spectra were recorded on a Nexus 670
140
FT-IR (Thermo Nicolet) using the ATR technique in the 400–4000 cm–1 region. The
141
NMR spectra (1H,
142
gradient selected, COSY = correlation spectroscopy, HMQC = heteronuclear multiple
143
quantum coherence, HMBC = heteronuclear multiple bond coherence) were
144
acquired on the DMF-d7 solutions at 300 K on either a JEOL JNM-ECA600II
145
spectrometer at 600.00 MHz (1H) and 150.86 MHz (13C) (complexes 1, 4 and 6) or a
146
Varian 400 spectrometer at 400.00 MHz (1H) and 100.58 MHz (13C) (complexes 2, 3
147
and 5). Proton and carbon spectra were calibrated against the residual DMF-d6 1H
148
NMR (8.03, 2.92 and 2.75 ppm) and 13C NMR (163.15, 34.89 and 29.76 ppm) signals.
149
The symbols s (singlet), d (doublet), t (triplet), qui (quintet), br (broad signal) and m
150
(multiplet) is used for the splitting of the proton resonances.
13
C, 1H–1H gs-COSY, 1H–13C gs-HMQC and 1H–13C gs-HMBC; gs =
151 152
Solution Stability Studies
153
Stability of the studied complexes in DMF-d7 (1–6) and DMF-d7/H2O mixture (1:1,
154
v/v; 5) was monitored by 1H NMR spectroscopy (Varian 400 MHz device) after 24 h
155
(both solutions) and 14 days (only DMF-d7 solutions) of standing at room
156
temperature under ambient light.
157
158
Methods of Biological Testing
159
Cell Culture and In Vitro Cytotoxicity Testing
160
In vitro cytotoxicity of the complexes 1–6, cisplatin and carboplatin was tested by
161
an MTT assay against ovarian carcinoma A2780 (ECACC No. 93112519), prostate -8-
162
carcinoma LNCaP (ECACC No. 89110211) and prostate carcinoma PC-3 (ECACC No.
163
90112714) human cancer cells obtained from European Collection of Cell Cultures
164
(ECACC) , as described in our previous works [11,13]. The cell lines were maintained
165
in a humidified incubator (
°C, 5% CO2). The cells were treated with 1–6, cisplatin
166
and carboplatin at the 0.01–
.
167
cells were treated in parallel with vehicle (DMF; 0.1%, v/v), and Triton X-100 (1%,
168
v/v) to assess the minimal (100% of cell viability), and maximal (0% of cell viability)
169
cell damage, respectively. The exposure time was 24 h. The MTT assay was used to
170
determine the cell viability by the spectrophotometric measurements of the
171
solubilized dye at 540 nm (TECAN, Schoeller Instruments LLC).
M o e t atio s usi g 96-well culture plates. The
172
Analogical in vitro cytotoxicity experiments were performed in the case of complex
173
3 on the A2780 cells with the addition of L-buthionine sulfoximine (L-BSO), which
174
as i depe de tl added to ea h
ell to gi e the . µM fi al o e t atio of L-
175
BSO
176
experiments focusing on the modulation of anticancer active transition metal
177
complexes) [18]. These experiments were performed with two negative controls
178
(DMF and
179
between the controls.
180
. µM o e t atio of L-BSO is known to be non-toxic and optimal for the
.
µM L-BSO), and no statistically different results were obtained
The IC50 values (compound concentrations that produce 50% of cell growth
181
inhibition; µM±SD
e e a ui ed
182
conducted in triplicate) performed on the cells from different passages. The
183
statistical evaluation (p < 0.05 were considered as significant) of the obtained data
184
was carried out by ANOVA using QC Expert 3.2 statistical software (TriloByte Ltd.).
three independent experiments (each
185 -9-
186
UVA Light Irradiation
187
A LZC-4V photoreactor (Luzchem, Ottawa, ON, Canada) employed with a
188
temperature controller was used for irradiation (4.3 mW cm-2; max = 365 nm) of the
189
DNA samples in cell-free media and using the UVA tubes, as described previously [8].
190 191
Platination of DNA in Cell-free Media
192
CT DNA (0.2 mg mL–1) was mixed with 1–6 or, for comparative purposes, with
193
carboplatin in NaClO4 (10 mM) and immediately irradiated (UVA, max = 365 nm) for
194
3 h at
195
conditions, similarly as described in [8]. . The ri value was 0.08 (ri = the molar ratio of
196
free platinum complex to nucleotide phosphates at the onset of incubation with
197
DNA). After the incubation, the samples were quickly filtered using a Sephadex G-50
198
column to remove free (unbound) platinum. The platinum content in these DNA
199
samples (rb, defined as the number of the molecules of platinum complex
200
coordinated per nucleotide residue) was determined by flameless atomic absorption
201
spectrometry (FAAS).
°C i the da k a d the kept fo additio al
h u de the
e tio ed
202 203
In Vitro Phototoxicity
204
A2780 and LNCaP cells were seeded in 96-well tissue culture plates in 100 L
205
medium in the absence of antibiotics at a density of 5,000 cells per well and placed
206
in the incubator for 24 h, as previously reported [8]. The solutions of 2, 3, 4 or 5 in
207
the medium (100 L) were added. The cells were incubated for 24 h. After that the
208
cells were washed out, and the medium containing the platinum(II) complex was
- 10 -
209
replaced by a drug-free medium in the absence of antibiotics, followed by 20 min
210
irradiation with UVA or sham irradiation. After additional 24 h, cell viability was
211
evaluated by an MTT (vide supra).
212 213
Stability and Interaction Studies after UVA Light Irradiation
214
The 1H NMR spectra (Varian 400 MHz device) of the selected representative
215
complex 5, its mixture with two molar equivalents of GSH (5+GSH), and its mixture
216
with two molar equivalents of guanosine monophosphate (5+GMP) in DMF-d7/H2O
217
mixture (1:1, v/v) were recorded right after the UVA irradiation (20 min) after 24 h of
218
standing at room temperature under ambient light. In the case of irradiated 5 itself
219
(i.e. without GSH or GMP) the 1H NMR spectrum was also recorded after 96 h of
220
standing at room temperature under ambient light. The ESI mass spectra in both the
221
positive and negative ionization modes were recorded using all the mentioned
222
solutions (i.e. 5, 5+GSH and 5+GMP) 24 h after UVA irradiation by the
223
ThermoFinnigan LCQ Fleet Ion Trap mass spectrometer (Thermo Scientific).
224 225
Transcription Mapping of DNA Adducts In Vitro
226
Linear pSP73KB/HpaI DNA was incubated with the selected platinum complexes 3
227
or 5 so that the DNA samples with the platinum(II) complex were irradiated with
228
UVA fo
229
or, alternatively, the DNA samples were incubated with the platinum(II) complexes in
230
the da k at
231
ethanol to remove unbound complex and the obtained solid (pellet) was dissolved in
i
ith su se ue t i u atio fo additio al . h i the da k at
°C fo
°C,
h. Afte i u ation, the samples were precipitated with
- 11 -
232
0.01 M NaClO4. The aliquots of the samples were used to determine level of Pt
233
bound to DNA (rb, defined as the number of molecules of the platinum(II) complex
234
bound per nucleotide residue) by using FAAS and spectrophotometric determination
235
of DNA at 260 nm.
236
Transcription of the linear pSP73KB/HpaI DNA treated with the complexes with
237
DNA-dependent T7 RNA polymerase, followed by the electrophoretic analysis of
238
transcripts were carried out according to the manufacturer (Promega Protocols and
239
Applications, 43–46, 1989/90) recommended protocols, as described previously [19].
240
The DNA concentration used in this assay (relative to the monomeric nucleotide
241
content) as
µM.
242 243
Interstrand DNA Cross-linking in a Cell-free Medium
244
Linear pSP73KB DNA/EcoRI (2455 bp) was mixed with 3 or 5 and immediately
245
irradiated with UVA for 30 min with subsequent incubation for additional 4.5 h in the
246
da k at
247
o ple es i
°C [8]. Alternatively, the DNA was incubated with the platinum(II) the da k at
°C fo
h. Afte i u atio , the sa ples
ee
248
precipitated to remove free, unbound platinum complex, dissolved in 0.01 M NaClO 4
249
and the rb in the aliquots of these samples was estimated by FAAS and
250
spectrophotometric determination of DNA at 260 nm. DNA in the remaining part of
251
sa ples as ´-end-labeled by means of the Klenow fragment of DNA polymerase I
252
in the presence of [-32P]dATP. The labeled samples were evaluated for DNA
253
interstrand cross-links according to the previously published procedures by
254
electrophoresis under denaturing conditions on alkaline agarose gel (1%) [19,20].
255
After the electrophoresis had been completed, the intensities of the bands - 12 -
256
corresponding to single strands of DNA and interstrand cross-linked duplex were
257
quantified. The frequency of interstrand cross-links was calculated as ICL/Pt (%) =
258
XL/4918 rb (the DNA fragment contained 4918 nucleotide residues), where ICL/Pt (%)
259
is the number of interstrand cross-links per adduct multiplied by 100, and XL is the
260
number of interstrand cross-links per molecule of the linearized DNA duplex, and
261
was calculated assuming a Poisson distribution of the interstrand cross-links as XL = -
262
ln A, where A is the fraction of molecules running as a band corresponding to the
263
non-cross-linked DNA.
264 265
Fluorescence Quenching Experiments
266
Fluorescence measurements of systems consisting of ethidium bromide (EtBr) and
267
CT DNA with addition of platinum complexes 3 or 5 were carried out at a 546 nm
268
excitation wavelength,, and the emitted fluorescence was analyzed at 590 nm. These
269
measurements were performed on a Varian Cary fluorescence spectrophotometer
270
usi g a
271
NaCl to avoid secondary binding of EtBr to DNA [21,22]. The concentrations were
272
0.01 mg mL− for DNA and 0.04 mg mL− for EtBr, which corresponded to the
273
saturation of all sites of EtBr in DNA [21].
ua tz ell. The fluo es e e i te sit
274
- 13 -
as
easu ed at
°C in 0.4 M
275
Results
276
Chemistry
277
A series of six complexes of the general formula Pt(cbdc)(naza)2] (1–6; Fig. 1) was
278
prepared in ca. 40% yields (relating to K2[PtCl4]) and their chemical purity (>95%) was
279
checked by combustion analysis (see S1 Text) and by 1H NMR spectroscopy (S1 Fig).
280
All the corresponding 1H and 13C signals (with the appropriate integral intensities)
281
of coordinated naza and cbdc ligands were detected in the spectra (S1 Fig). An N7
282
coordination mode of the naza ligands was clearly proved for the complexes 1–6
283
from the calculated 1H and 13C NMR coordination shifts (S1 Table). All the complexes
284
were found to be stable in DMF-d7 over 14 days (no changes were detected in the 1H
285
NMR spectra). In the case of the selected complex 5 dissolved in the DMF-d7/H2O
286
mixture (1:1, v/v), a new set of signals corresponding to 4Braza ligand (e.g. N1–H
287
signal at 11.94 ppm or C6–H signal 8.19 ppm) was detected after 24 h of standing at
288
room temperature under ambient light (Fig. 2B). The chemical shifts detected were
289
different from those of free 4Braza molecule in the same solvent (e.g. N1–H signal at
290
11.82 ppm).
291 292
Fig. 2. 1H NMR stability studies. Time-dependent (A and C - fresh solutions; B and D -
293
after 24 h of standing at room temperature under ambient light) 400 MHz 1H NMR
294
spectra (N1–H region of 4Braza) as observed before (A and B) and after (C and D)
295
UVA i adiatio
296
solution (1:1, v/v).
i ,
max
= 365 nm) of complex 5 dissolved in the DMF-d7/H2O
- 14 -
297 298
The ESI-MS spectra, measured in both the positive (ESI+) and negative (ESI–)
299
ionization mode, of all the studied complexes contained the molecular peaks, i.e.
300
{Pt(cbdc)(naza)2]+H}+, and {Pt(cbdc)(naza)2]–H}–, respectively (S3 Fig.). Moreover,
301
the adducts with sodium ions, {Pt(cbdc)(naza)2]+Na}+, as well as the
302
{Pt(cbdc)(naza)]–H}– and {naza+H}+ species were identified in the appropriate
303
spectra as well.
304
305
Biological activities testing
306
In Vitro Cytotoxicity
307
The in vitro cytotoxicity of the prepared complexes 1–6 (applied within the .
M, depe di g o
the solu ilit
i
ate -
308
concentration range of 0.01–
309
containing medium) can be generalized as high against the A2780 ovarian carcinoma
310
cancer cell line (IC50 = 4.7–
311
. µM a d
. µM a d PC-3 (IC50 = 17.6–
ode ate i the ase of LNCaP IC50 = 18.7–
. µM p ostate a i o a a e
ell li es Ta le
312
1). All the complexes exceeded the activity of the reference drug cisplatin against
313
A2780, with the most effective complex 6 being ca. 4.6-fold more cytotoxic. Except
314
for compound 3, the in vitro cytotoxicity of the studied complexes against the A2780
315
cells was significantly higher (ANOVA, p < 0.05) as compared to cisplatin. In the case
316
of LNCaP and PC-3, the evaluation was limited by the fact that IC50 of cisplatin could
317
not be obtained, because it is higher than the highest applied concentration, above
318
which the compounds are generally considered to be ineffective (i.e. IC50 >
319
Complex 4 was found to be the most cytotoxic on LNCaP and PC-3 cell lines (Table 1). - 15 -
. µM .
320
The studied complexes also exceeded the in vitro cytotoxicity of carboplatin, which
321
was found to be non-to i up to the
322
cell lines used (Table 1).
. µM o e t atio agai st all th ee a e
323 324
Table 1. In vitro antitumor activity of 1–6, cisplatin and carboplatin against A2780,
325
LNCaP and PC-3 cancer cell lines. The results of the in vitro antitumor activity testing
326
of 1–6, cisplatin and carboplatin against human ovarian (A2780) and prostate (LNCaP
327
and PC-3) cancer cell lines. Cells were treated with the tested compounds for 24 h,
328
measurements were performed in triplicate, and cytotoxicity experiment was
329
repeated in three different cell passages. Data are expressed as IC50 ± SD Complex
A2780
LNCaP
PC-3
1
. ± . *
2
. ± . *
>20.0a
. ± .
3
. ± .
>50.0a
. ± .
4
. ± . *
. ± .
. ± .
5
. ± . *
. ± .
. ± .
6
. ± . *
. ± .
. ± .
Carboplatin >50.0a Cisplatin
. ± .
M.
. ± .
. ± .
>50.0a
>50.0a
>50.0a
>50.0a
asterisk (*), significantly different values (p < 0.05) between 1–6 and cisplatin; a) IC50 were not reached up to the given concentration 330 331
The in vitro cytotoxicity of the selected representative complex 3 was found to be
332
sig ifi a tl highe i
the p ese e of
.
333
glutathione synthesis, as the IC50 value determined for the A2780 cell line equalled - 16 -
µM L-BSO, an effective inhibitor of
334
to . ± . µM. This
335
activity as compared to the same complex without added L-BSO.
eans about 4.4-times enhancement in the antiproliferative
336 337
Phototoxicity in Cell Cultures
338
The effect on the cell viability determined for each tested compound was
339
significantly (p 0.05) higher when the complex was applied in combination with
340
UVA irradiation, as compared to the reference sample (in the dark) (Fig. 3).
341
Importantly, the control cells (with and without UVA exposure) grew at the same
342
rate.
343 344
Fig. 3. Cytotoxic activity of the complexes 2–5. Cytotoxic activity of 2–5 at their 10
345
µM o e t atio s agai st the A
346
lines. Viability of the untreated, sham-irradiated cells was taken as 100%. The
347
asterisk (*) denotes significant difference (p < 0.05) between the irradiated and
348
sham – irradiated cells.
top pa el a d LNCaP
otto
pa el
ell
349 350 351
DNA Binding in Cell-free Media Samples of double-helical CT DNA were incubated with the complex at ri value of
352
0.08 in 0.01 M NaClO4 at
°C a d su se ue tl di ided i to t o pa ts. O e pa t
353
was irradiated with UVA light (max = 365 nm, 4.3 mW cm-2) immediately after
354
addition of the complex; the other (control) sample was kept in the dark. After 5 h of
355
incubation, the samples were assayed for platinum content bound to DNA, as
356
described above, by FAAS. The amount of platinum bound to DNA in the samples,
- 17 -
357
which were kept in the dark, ranged from 22 to 42% (Table 2). In contrast, after 5 h
358
of continuous UVA irradiation, the platination of DNA increased ca. 2–3-fold, as
359
compared to the samples incubated in the dark (Table 2).
360 361
Table 2. DNA binding of 1–6 in cell free media. Data are expressed as percentage of
362
plati u
363
three independent experiments.
ou d to DNA to total plati u
Complex
dark
UVA
1
±
±
2
±
±
3
±
±
4
±
±
5
±
±
6
±
±
Carboplatin
±
i o e. Data ep ese t the
ea ± SD of
±
364 365
Stability Studies after the UVA Irradiation
366
The 1H NMR spectrum of the complex [Pt(cbdc)(4Braza)2] (5) dissolved in DMF-
367
d7/H2O (1:1, v/v) was recorded before and after 20 min of UVA irradiation (max =
368
365 nm; 4.3 mW cm-2). Before the irradiation, 5 showed one signal at 13.02 ppm,
369
corresponding to the N1–H atom of the coordinated 4Braza ligand, while two new
370
N1–H signals were detected at 11.82 and 12.35 ppm after the irradiation (Figs. 2 and
371
4). The 1H NMR spectrum of the starting complex 5 (before the irradiation) showed,
372
as assumed, one quintet (C13–H2) and one triplet (C11–H2, C12–H2) at 1.89, and 2.49
- 18 -
373
ppm, respectively (S4 Fig.). One new C13–H2 quintet (2.08 ppm) was detected in the
374
1
375
the irradiated sample did not show any new signals nor any change in the integral
376
intensities, as compared with the fresh irradiated solution, after 24 h (Fig. 2), but at
377
96 h lower intensity of the signal at 12.35 ppm and one new signal at 12.18 ppm
378
were detected. Additional 1H NMR experiments were performed on the irradiated
379
solution of 5 (96 h after the irradiation) spiked with free 4Braza dissolved in the
380
same solution (DMF-d7/H2O, 1:1, v/v), leading to marked intensity increase of the
381
signal at 11.82 ppm, which suggested that the signal at 11.82 ppm belongs to the
382
4Braza molecule released from the studied complex (Fig. 4).
H NMR spectrum of the irradiated sample. The 1H NMR spectroscopy performed on
383 384
Fig. 4. UVA irradiation effect on the composition of the complex 5. Time-dependent
385
(after 24 (A) or 96 h (B and C) of standing at room temperature under ambient light)
386
400 MHz 1H NMR spectra (N1–H region of 4Braza) as observed in the DMF-d7/H2O
387
(1:1, v/v) solutions of complex 5 (A and B), and complex 5 spiked with free 4Braza
388
(C), with UVA irradiation (20 min, max = 365 nm), respectively.
389 390
The ratio of the integral intensities, showing on the portion of the
391
rearranged/decomposed complex, of the N1–H signal of the parent compound and
392
both the newly emerged signals is 1.00 (13.02 ppm) to 0.65 (12.35 ppm) to 1.06
393
(11.82 ppm), which indicates that the amount of the decomposed complex is ca 63%
394
after 20 min UVA irradiation.
395
The ESI+ and ESI– mass spectra of complex 5 before the irradiation contained the
396
{[Pt(cbdc)(4Braza)2]+H}+ (733.2 m/z), {[Pt(cbdc)(4Braza)2]–H}– (731.2 m/z) and - 19 -
397
{[Pt(cbdc)(4Braza)2]+Na}+ (755.2 m/z) peaks of the starting complex or their adducts
398
with the Na+ ion, as well as the peaks of the {[Pt(cbdc)(4Braza)]–H}– (533.6 m/z) and
399
{4Braza+H}+ (197.1 m/z) species (S5 Fig.). All these peaks were detected also in the
400
spectra of the irradiated sample, but with significant changes in intensities.
401
Concretely, the peak of the {4Braza+H}+ fragment showed about 5-fold higher
402
relative abundance in the ESI+ mass spectrum after the irradiation, which
403
corresponded with markedly higher intensity of the peak of the {[Pt(cbdc)(4Braza)]–
404
H}– species with one 4Braza molecule released (S5 Fig.).
405 406
Photoreaction with Biomolecules (GSH and GMP)
407
Analogous 1H NMR and ESI-MS experiments, as described above for complex 5,
408
were performed also for the mixtures of 5 with GSH (symbolized as 5+GSH) or GMP
409
(5+GMP) in DMF-d7/H2O (1:1, v/v).
410
The 1H NMR spectra of 5+GSH, both before and after the irradiation, contained
411
the characteristic GSH signals, such as triplet at 8.68 ppm and doublet at 8.56 ppm of
412
the N–H hydrogen atoms of glycine, and cysteine part of the GSH molecule,
413
respectively. As for 4Braza, the N1–H region of the 1H NMR spectrum (before
414
irradiation) does not contain any new signals, as discussed above. As for the 1H NMR
415
spectrum of the irradiated 5+GSH mixture, only the same two new signals (i.e. at
416
12.35 and 11.82 ppm) in the N1–H region of 4Braza as in the case of the irradiated
417
starting material itself, and no new signals in the region of the N–H hydrogen atoms
418
of glycine and cysteine were observed. Furthermore, any peaks whose mass
419
corresponded to the adducts of GSH and 5 or fragments were detected in the ESI+
420
and ESI– spectra of 5+GSH, both before and after irradiation (S5 Fig.). - 20 -
421
Regarding the 5+GMP mixture, the signals of the coordinated 4Braza as well as
422
C8–H signal of free GMP (at 8.37 ppm) were clearly detected in the 1H NMR spectra
423
before the irradiation. The UVA irradiation caused several changes: the signals (e.g.
424
8.19 ppm for C6–H; N1–H signals were not detected) of the released 4Braza ligand
425
were detected and one new signal was observed very close to the C8–H signal of free
426
GMP (at 8.31 ppm) with the integral intensity about three times lower compared to
427
the C6–H signal of 4Braza of the starting material in the mixture (S6 Fig.). The mass
428
spectra of the mixtures containing the GMP also showed the new peaks (in addition
429
to
430
{[Pt(cbdc)(4Braza)2]+H}+,
431
{[Pt(cbdc)(4Braza)]–H}– species) of the {GMP–Na+2H}+, {GMP+H}+, {GMP–2Na+H}–
432
and
433
{[Pt(cbdc)(4Braza)(GMP)]+H}+, and {[Pt(cbdc)(4Braza)(GMP)]–2Na+H}– species at
434
368.3, 408.3, 362.4, 920.2, 942.3, and 896.3 m/z, respectively (S5 Fig. and S6 Fig.).
those
also
discussed
those
above,
corresponding
{[Pt(cbdc)(4Braza)2]+Na}+,
corresponding
to
the
to
the
{4Braza+H}+,
{[Pt(cbdc)(4Braza)2]–H}–,
{[Pt(cbdc)(4Braza)(GMP)]–Na+2H}+,
435 436
Characterizations of DNA Adducts Formed in Dark and under the UVA
437
Irradiation
438
Besides the DNA binding capacity, the important factor which modulates the
439
cytotoxicity of platinum compounds is the nature of the conformational changes
440
induced in DNA. In order to determine the nature of DNA adducts formed by
441
Pt(cbdc)(naza)2] complexes in the dark and under the UVA irradiation, several
442
biochemical and biophysical methods have been applied. As model compounds,
- 21 -
443
complexes 3 and 5 have been selected since they exhibited the highest phototoxic
444
effects.
445 446
Transcription Mapping of Platinum–DNA Adducts
447
Experiments on in vitro RNA synthesis by T7 RNA polymerase were carried out
448
using a linear pSP73KB/HpaI DNA fragment, treated with complex 3 or 5 in dark or
449
under the UVA irradiation conditions (see Materials and Methods). The major stop
450
sites produced by the templates treated with 5 either in dark or under the irradiation
451
are shown in Fig. 5 (lanes 5-dark and 5-UV). These stop sites were similar to those
452
produced by cisplatin (Fig. 5, lane cisplatin), i.e. appeared mainly at GG or AG sites -
453
the preferential DNA binding sites for this metallodrug [23]. The stop sites produced
454
by transplatin (shown for comparative purposes) were less regular and appeared
455
mainly at single G and C sites - the preferential DNA binding sites for this platinum
456
complex [24]. Importantly, the efficiency to block the RNA polymerases differed
457
significantly for the adducts formed by 5 in the dark and the adducts formed upon
458
UVA irradiation. The adducts formed by 5 (at rb = 0.003) under irradiation conditions
459
were much more effective in inhibiting the RNA synthesis compared to the adducts
460
formed by 5 in the dark at the same or even higher level of platination (r b = 0.003
461
and 0.01) [cf. lanes 5-UV (0.003), 5-dark (0.003) and 5-dark (0.01) in Fig. 5. Similar
462
results were obtained also for complex 3 (not shown).
463 464
Fig. 5. Inhibition of RNA synthesis by 5, cisplatin and transplatin. Inhibition of RNA
465
synthesis by T7 RNA polymerase on the pSP73KB/HpaI fragment modified by 5 under
466
the irradiation or in the dark, cisplatin or transplatin. Autoradiogram of 6% - 22 -
467
polyacrylamide/8 M urea gel. Lanes: Control UVA, unmodified template irradiated
468
with UVA; Control dark, non-irradated unmodified template; A, U, G, C, chain
469
terminated marker RNAs; 5-UV (0.003), the template modified at rb = 0.003 by
470
irradiated 5; 5-dark (0.01), the template modified at rb = 0.01 by 5 in the dark; 5-dark
471
(0.003), the template modified by 5 at rb = 0.003 in the dark; cisplatin (0.01), the
472
template modified at rb = 0.01 by cisplatin; transplatin (0.01), the template modified
473
at rb = 0.01 by transplatin.
474 475
Interstrand DNA Cross-links
476
Bifunctional platinum compounds, which coordinate to the base residues in DNA,
477
form various types of interstrand and intrastrand cross-links. Such cross-links in the
478
target DNA are important factors involved in the DNA damaging action of the
479
genotoxic agents. Therefore, we have quantified the interstrand cross-linking
480
efficiency of 3 or 5 when photoactivated or in the dark using linear pSP73KB/EcoRI
481
DNA. The DNA samples were treated with complex 3 or 5 in dark or under the UVA
482
irradiation conditions as described above. Samples were analyzed by agarose gel
483
electrophoresis under denaturing conditions. The interstrand cross-linked DNA
484
appears in the autoradiogram as the top bands (Fig. 6A), as it migrates more slowly
485
than the single-strand DNA (the bottom bands). The frequencies of interstrand cross-
486
links formed by photoactivated 3 and 5
487
Interestingly, the modification of DNA in dark resulted in the absence of the slowly
488
migrating bands, indicating that these samples contained no detectable interstrand
489
cross-linking, although the levels of platination (r b) were similar to those in the
490
irradiated samples. - 23 -
ee
± , a d
± %, espe ti el .
491 492
Fig. 6. Formation of interstrand cross-links and dependence of ethidium bromide
493
(EtBr) fluorescence. A. The formation of interstrand cross-links by complex 5 under
494
the irradiation with UVA (lanes 1–5) or in the dark (lanes 6–10). Lanes: 1, control,
495
untreated DNA (incubated under irradiation conditions); 2–5, rb = 0.0027, 0.0013,
496
0.0007 and 0.0004, respectively; 6, control, untreated DNA (incubated in the dark);
497
7–10, rb = 0.0033, 0.0016, 0.0009 and 0.0007, respectively. B. Dependence of
498
ethidium bromide (EtBr) fluorescence on rb for DNA modified by irradiated 5
499
(squares) or by 5 i the da k t ia gles . Data a e a e age ± SD fo th ee i depe de t
500
experiments. Data for cisplatin (dashed line) and monofunctional dienplatin (dotted
501
line) recorded under identical experimental conditions are taken from the literature
502
[25].
503 504
Characterization of DNA Adducts by Fluorescence Experiments
505
Ethidium bromide (EtBr), as a fluorescent probe, can be used to characterize the
506
DNA binding of small molecules, such as platinum antitumor drugs and to distinguish
507
the bifunctional from monofuntional DNA–adducts of platinum complexes [21,22].
508
Binding of EtBr to DNA by intercalation is blocked in a stoichiometric manner by the
509
formation of bifunctional adducts of a series of platinum complexes, including
510
cisplatin, which results in a loss of fluorescence intensity. However, DNA binding of
511
monofunctional
512
diethylenetriamineplatinum(II) chloride) results only in small decrease of the
513
fluorescence intensity [25]. Modification of DNA by 3 or 5 under irradiation
514
conditions resulted in a decrease of EtBr fluorescence (shown in Fig. 6B for 5) similar
complexes
such
- 24 -
as
dienplatin
(chlorido-
515
to that caused by cisplatin at equivalent rb values. On the contrary, the decrease
516
caused by the adducts of 5 formed in the dark was only slightly higher at equivalent
517
rb values than that induced by monodentately DNA-binding dienplatin (having only
518
one leaving ligand). The analogous results were obtained also for the complex 3.
519
520
521
Dis ussio The studied Pt(cbdc)(naza)2] complexes (1–6; Fig.
, p epa ed
Dha a´s
522
method [17], represent the derivatives of the clinically used platinum-based drug
523
carboplatin involving 7-azaindole (naza) derivatives as N-donor carrier ligands. The
524
coordination mode of the naza ligands in the studied complexes was determined to
525
be through the N7 atom as in the previous works reported in the X-ray structures of
526
the platinum(II) dichlorido [11,26] and oxalato [26] complexes with the analogous
527
naza ligands. The same coordination mode was clearly proven also for the herein
528
reported complexes 1–6 by the 1H and 13C NMR coordination shifts (S1 Table).
529
The prepared complexes showed high (against ovarian carcinoma cells) or
530
moderate (against both the prostate carcinoma cancer cell lines) in vitro cytotoxicity
531
(Table 1). In comparison to the recently studied dichlorido complexes (IC50 = 1.8–2.6
532
µM agai st A
533
ligands, the complexes studied in this work (1–6) are less effective against the
534
mentioned human cancer cell lines.
a d . – . µM agai st LNCaP ells [11]) with analogous N-donor
535
One of the hot-topics in the field of platinum bioinorganic and medicinal
536
chemistry is the preparation of agents having good stability and no or very low
537
cytotoxic effect (so called prodrugs) and study their activation towards biologically - 25 -
538
active species [27]. Obviously, the studied complexes were not found to be inactive,
539
but their cytotoxicity was still markedly lower as compared to their dichlorido
540
analogues [11,13], which meant that there is still room to improve the biological
541
activity of the studied carboplatin-analogues. To reach this goal, we used two of
542
several possible strategies which can be applied to increase the biological activity of
543
cytotoxic transition metal complexes, the first one was based on the addition of L-
544
BSO (a selective inhibitor of -glutamylcysteine synthase), effectively blocking the
545
inactivation of the complexes by GSH conjugation, and the second one using the
546
photoactivation of the studied compounds by UVA light.
547 548
The complex 3 showed ca. 4.4-fold enhancement of in vitro cytotoxicity when it as ad i ist ated to the A
ells togethe
ith . µM L-BSO. As L-BSO is a well-
549
k o
i hi ito of -glutamylcysteine synthase, it has a profound effect on the
550
mechanism of action of the cytotoxic transition metal complexes having either
551
cisplatin-like or redox processes modulating mechanism of action [18,28,29]. In
552
other words, decreasing of the GSH cellular levels, caused by the L-BSO addition,
553
affects the cytotoxicity of the complexes inactivated by a GSH-mediated cellular
554
detoxification (e.g. cisplatin in the cisplatin-resistant cancer cell lines) as well as the
555
cytotoxicity of the complexes, whose biological effect is mediated through the
556
cellular redox processes. Since it has been proved by the 1H NMR and ESI-MS
557
experiments that the studied complexes do not interact with GSH, the increase of
558
the biological effect of 3 on the A2780 cancer cells could come from the modulation
559
of other cellular redox pathways.
560
Iit is known for carboplatin that its in vitro cytotoxicity could be enhanced by the
561
UVA irradiation [8]. That is why we decided to study the UVA irradiation effect on - 26 -
562
the biological profile of the herein studied carboplatin derivatives. We chose A2780
563
and LNCaP cells, which were exposed to 2–5 for 24 h, followed by 20 min irradiation
564
with UVA or sham irradiation. We observed that upon UVA irradiation, the in vitro
565
cytotoxicity of the tested substances increased markedly, as compared with the
566
experiments performed in dark (Fig. 3). These promising results motivated us to
567
further perform more detailed molecular biological and biophysical studies to
568
uncover the mechanistic aspects responsible for the considerable biological activity
569
of the studied complexes.
570
The studied compounds were stable in DMF-d7 over 14 days, as judged by 1H NMR
571
spectra. On the other hand, analogical experiments performed in the DMF-d7/H2O
572
mixture (1:1, v/v) provided one new N1–H signal of the 4Braza N-donor ligand (11.94
573
ppm) after 24 h (Fig. 2). The chemical shift value differs from that of free 4Braza
574
(11.82 ppm), which suggested (together with the fact that only one new signal was
575
detected - release of the 4Braza molecule from complex 5 would have to led to at
576
least two new signals) that the mentioned N1–H signal belongs to the 4Braza ligand
577
coordinated in the species having different composition from that of the starting
578
material. Speculatively, the mentioned species could contain an open six-membered
579
PtO2C3 ring (formed by the central atom and bidentate-coordinated cbdc ligand
580
within the starting complex 5) and could correspond to the composition cis-
581
[Pt(4Braza)2(cbdc`)(H2O)] (cbdc` = monodentate-coordinated cbdc ligand), which was
582
both experimentally [30,31] and theoretically [32] proven for carboplatin, but we did
583
not get any evidence for this process (e.g. from ESI-MS performed on this sample) in
584
the case of herein reported carboplatin derivatives.
- 27 -
585
With an intention to better understand the composition and behavior of the
586
studied complexes (represented by the selected complex 5) before and after the
587
UVA irradiation, as well as in the presence of GSH, GMP or genomic DNA, we decided
588
to perform the stability and interaction studies (1H NMR, ESI-MS) of 5 and its
589
mixtures with GSH (one of the major reducing sulfur-containing agents of human
590
plasma with known coordination affinity towards Pt(II) atom, representing both the
591
transport opportunities and some ways of inactivation of Pt(II) anticancer drugs) or
592
GMP (well-known model system of the target binding site on DNA molecule attacked
593
by the cytotoxic platinum(II) complexes). In the case where the studied complex was
594
irradiated by UVA light (max = 365 nm) for 20 min, two new N1–H signals of 4Braza
595
were detected in the 1H NMR spectra at 12.35 and 11.82 ppm together with the
596
signal of the starting material at 13.02 ppm, while the signal at 11.94 ppm (as
597
detected for the unirradiated complex 5) was not found (Fig. 2). One of these signals
598
(11.82 ppm) belongs to the free 4Braza molecule, as proved by the 1H NMR
599
experiments with the irradiated solution of 5 spiked with the solution of free 4Braza
600
ligand (Fig. 4). This is also consistent with the results of ESI-MS, where the intensity
601
of the peaks of the {[Pt(cbdc)(4Braza)]–H}– and {4Braza+H}+ fragments, whose
602
formation is directly associated with a release of the N-donor ligand from the parent
603
complex, markedly increased after the UVA irradiation (S5 Fig.). The second new
604
peak (12.35 ppm) split into two (12.35 and 12.18 ppm) after longer than 24 h
605
standing at room temperature under ambient light (Fig. 4). We believe that this
606
change
607
transformation) of the species formed after the irradiation, and not by the formation
608
of the new species with different composition, because the overall integral intensity
is
caused
by
rearrangement
- 28 -
or
isomerization
(e.g.
cis-to-trans
609
of these two signals is in the same ratio to the other two N1–H peaks at 13.02 and
610
11.82 ppm, as in the case of the spectrum recorded 24 h after the irradiation.
611
Further, this observation indirectly proved the opening of the six-membered PtO2C3
612
chelate ring, without which the above mentioned isomerization would not be
613
possible. Opening of this chelate ring was also indicated by the 1H NMR results,
614
because the chemical shift of one new C13–H2 quintet has been detected in the 1H
615
NMR spectrum of the irradiated sample at 2.08 ppm (S4 Fig.) differing from both the
616
complex 5 and free H2cbdc molecule. With respect to the described findings, it can
617
be assumed that the composition of the platinum-containing species formed from
618
complex 5 after the UVA irradiation corresponds to cis-[Pt(H2O)2(cbdc`)(4Braza)]
619
(N1–H signal of 4Braza at 12.35 ppm), which partially rearranges with time to trans-
620
[Pt(H2O)2(cbdc`)(4Braza)] (12.18 ppm). Unfortunately, no direct evidence for these
621
statements from the mass spectra of the irradiated sample was found, probably due
622
to decomposition of the mentioned aqua species connected with the Pt–OH2 bond
623
cleavage under electrospray ionization conditions.
624
Photoreaction with GSH does not lead to the formation of adducts of GSH with
625
complex 5 or fragments, as judged by the 1H NMR and ESI-MS experiments (S5 Fig.).
626
Although the changes in the NMR and mass spectra before and after the irradiation
627
were detected, they correspond only to irradiation (as discussed above for the
628
original complex) and not to interaction with this sulfur-containing biomolecule. The
629
statement that GSH does not interact with complex 5 or with the species formed
630
from complex 5 after UVA irradiation can be proved also by the fact that no new N1–
631
H signals of 4Braza were observed in the 1H NMR spectra, as compared to the 1H
- 29 -
632
NMR spectra of 5 alone (as discussed above). Further, the platinum-containing
633
adducts with GSH were not found by the ESI-MS before and after irradiation (S5 Fig.).
634
On the other hand, GMP showed the ability to interact with Pt(II) atom in the
635
representative complex 5, since the 1H NMR spectra of the irradiated 5+GMP
636
mixture contained one new signal of the C8–H hydrogen atom of GMP at 8.31 ppm,
637
as well as the signals of free 4Braza released from the parent complex (S6 Fig.).
638
Although one could expect the coordination of GMP to the Pt(II) atom (as known for
639
carboplatin [30]), it could not be unambiguously judged based on the proton NMR
640
spectra, because no new set of signals of 4Braza, coordinated within the GMP-
641
containing species, was detected. Still, even if we could think about the substitution
642
of both the 4Braza molecules involved in the starting complex by two GMP
643
molecules to be possible, this process would be firstly not very probable, and
644
secondly it was directly disproved by ESI-MS results showing the peaks assignable to
645
the
646
{[Pt(cbdc)(4Braza)(GMP)]–2Na+H}– species (S5 Fig..). It has to be noted, that these
647
peaks were identified only in the spectra of the irradiated samples, contrary to the
648
peaks at 1118.2, 1140.2 and 1094.0 m/z of the species whose mass corresponds to
649
{[Pt(cbdc)(4Braza)2(GMP)]–Na+2H}+,
650
{[Pt(cbdc)(4Braza)(GMP)]–2Na+H}–, respectively, which were detected in the
651
appropriate spectra even before the UVA irradiation of the studied 5+GMP mixture,
652
which indicates that these adducts formed only as a consequence of the electrospray
653
ionization process.
{[Pt(cbdc)(4Braza)(GMP)]–Na+2H}+,
{[Pt(cbdc)(4Braza)(GMP)]+H}+
{[Pt(cbdc)(4Braza)2(GMP)]+H}+
and
and
654
Since DNA is the major pharmacological target of the antitumor platinum drugs
655
[2,33,34] it was also of great interest to examine whether the enhanced cytotoxicity - 30 -
656
of the complexes correlates with DNA binding of the photoactivated derivatives,
657
similarly as in the case of carboplatin [8]. The initial experiments were aimed to
658
quantify the binding of 1–6 and carboplatin to mammalian DNA in cell-free media.
659
The results proved that the amount of platinum bound to DNA was markedly
660
enhanced due to the UVA irradiation (after 5 h) as compared to the samples
661
incubated in dark (Table 2). The results of DNA binding of carboplatin in dark and
662
under continuous irradiation with UVA are in good agreement with the previously
663
published data [8], confirming that less than 5% of carboplatin is bound to DNA in
664
the sample which was kept in the dark. This is in contrast to the increased level of
665
DNA-platination in the sample irradiated with UVA, so that more than 50% of
666
platinum from carboplatin was bound after 5 h (Table 2). Notably, under comparable
667
conditions, the amount of molecules of carboplatin bound to DNA was lower than
668
that of molecules of 1–6. Recent works have shown that the transcription on DNA
669
templates modified by bidentate adducts of platinum complexes can be prematurely
670
terminated at the level or in the proximity of such adducts, while the
671
monofunctional DNA adducts of platinum complexes were unable to terminate the
672
RNA synthesis [19,35,36]. So, the considerably different efficacy of DNA adducts
673
formed by 3 or 5 in dark and under the irradiation conditions (at the same level of
674
DNA platination) to inhibit RNA polymerase is consistent with different frequency of
675
mono- and bifunctional adduct formed by these complexes in dark and under the
676
irradiation conditions. Thus, the results of transcription mapping experiments (Fig. 5)
677
support the hypothesis that under irradiation conditions, complexes 3 and 5
678
preferentially form the bifunctional adducts with DNA, capable of effective
- 31 -
679
termination of RNA synthesis by RNA polymerases. On the other hand, in dark, the
680
formation of less effective monofunctional adduct prevails.
681
The results of transcription mapping experiments are in good agreement with the
682
characterization of DNA adducts formed by 3 or 5 in the dark and under irradiation
683
conditions estimated by the EtBr fluorescence quenching (Fig. 6B). These results
684
show that 3 and 5 form in the dark the DNA adducts which resemble, from the
685
viewpoint of their capability to inhibit EtBr fluorescence, those formed by
686
monofunctional platinum complexes. Notably, the DNA adducts formed by 3 and 5
687
under irradiation conditions inhibited EtBr fluorescence to the same extent as
688
bifunctional cisplatin. Hence, the fluorescent analysis is consistent with the idea and
689
supports the postulate that the major DNA adducts formed by 3 or 5 in dark are
690
mainly monofunctional lesions. In contrast, under comparable conditions (at the
691
same level of DNA platination), but under the irradiation conditions, 3 and 5 form on
692
DNA mainly bifunctional adducts similar to those formed by cisplatin.
693
The latter conclusion is also reinforced by the observation (Fig. 6A) that 3 or 5
694
formed a significant amount of bifunctional interstrand cross-links under the
695
irradiation of DNA (even slightly higher than cisplatin at the same level of platination
696
[19]) whereas 3 or 5 formed under comparable conditions no such bifunctional
697
lesions in the dark in DNA.
698
In addition, the results characterizing the monofunctional binding of 3 or 5 to
699
highly polymeric double-helical DNA are consistent with the formation of a ring-
700
opened species [Pt(cbdc`)(naza)2(H2O)], containing monodentate cbdc`, in the dark.
701
The bifunctional cross-links are probably formed as a consequence of the
702
photoactivation by UVA and very likely occur as a consequence of the reaction of - 32 -
703
DNA with bifunctional cis-[Pt(H2O)2(cbdc`)(naza)] active intermediate (containing two
704
easily exchangeable H2O ligands). The structure of the such bifunctional product was
705
suggested on the basis of 1H NMR and ESI-MS characterization of the UVA-irradiated
706
solutions of complex 5 (vide supra).
707
To conclude the presented work, we prepared and characterized a series of
708
carboplatin derivatives, involving the halogeno-substituted 7-azaindoles as the N-
709
donor carrier ligands. The in vitro cytotoxicity of the prepared complexes against
710
A2780 human ovarian carcinoma cell line was markedly (ca. 4.4-times) increased by
711
the addition L-BSO. Based on the results of the detailed 1H NMR and ESI-MS studies
712
carried out on the starting complex 5 and its mixtures with biomolecules GSH or
713
GMP, as well as on the results of DNA-platination, it is obvious that UVA irradiation
714
(20 min, max = 365 nm) led to the release of one 4Braza ligand and hydrolysis of one
715
of the Pt–O bonds between central Pt(II) atom and the chelating cbdc dianion.
716
Additionally, the UVA irradiation led to subsequent formation of the activated
717
species (most probably cis-[Pt(H2O)2(cbdc`)(naza)]) and resulted in markedly higher
718
cytotoxicity of 5 against A2780 ovarian carcinoma and LNCaP prostate
719
adenocarcinoma human cancer cell lines, as compared with sham-irradiated
720
samples. Moreover, DNA binding of the studied complexes is markedly enhanced by
721
the irradiation, as was proven on both chemical (ability to interact with GMP) and
722
biological (higher CT DNA platination) experimental levels. Thus, in connection with
723
the acquired results we have reason to believe that the complexes 1–6 could
724
represent suitable candidates for use in photoactivated cancer chemotherapy.
725
- 33 -
726
A k owledge e ts
727
The authors thank to Ms. Kateři a Ku ešo á fo help
728
D . Radka Křika o á fo pe fo
729
performing ESI-MS experiments.
ith the
toto i it testi g,
i g NMR e pe i e ts a d D . Bohusla D ahoš fo
730
731
Refere es
732
1. Rosenberg B, VanCamp L, Krigas T. Inhibition of cell division in Escherichia coli by
733
electrolysis products from a platinum electrode. Nature. 1965; 205: 698–699.
734
2. Kelland L. The resurgence of platinum-based cancer chemotherapy. Nature Rev
735
Cancer. 2007; 7: 573–584.
736
3. Barry NPE, Sadler PJ. Exploration of the medical periodic table: towards new
737
targets. Chem Commun. 2013; 49: 5106–5131.
738
4. Kelland LR, Farrell NP. Platinum-Based Drugs in Cancer Therapy. Humana: Totowa;
739
2000.
740
5. Siddik ZH. Cisplatin: mode of cytotoxic action and molecular basis of resistance.
741
Oncogene. 2003; 22: 7265–7279.
742
6. Smith NA, Sadler PJ. Photoactivatable metal complexes: from theory to
743
applications in biotechnology and medicine. Phil Trans R Soc. 2013; A 371: 20120519.
744
7. Mackay FS, Woods JA, Heringova P, Kasparkova J, Pizarro AM, Moggach SA, et al.
745
A potent cytotoxic photoactivated platinum complex. Proc Natl Acad Sci USA. 2007;
746
104: 20743–20748.
- 34 -
747
8. Mlcouskova J, Stepankova J, Brabec V. Antitumor carboplatin is more toxic in
748
tumor cells when photoactivated: enhanced DNA binding. J Biol Inorg Chem. 2012;
749
17: 891–898.
750
9. Zhao J, Gou S, Sun Y, Fang L, Wang Z. Antitumor platinum(II) complexes containing
751
platinum-based moieties of present platinum drugs and furoxan groups as nitric
752
oxide donors: synthesis, DNA interaction, and cytotoxicity. Inorg Chem. 2012; 51:
753
−
.
754
10. D ořák L, Popa I, Šta ha P, T á íček ). In vitro cytotoxic-active platinum(II)
755
complexes derived from carboplatin and involving purine derivatives. Eur J Inorg
756
Chem. 2010; 3441–3448.
757
11. Šta ha P, T á íček ), Popa A, Popa I, Mu ho á T, B a e V. How to modify 7-
758
azaindole to form cytotoxic Pt(II) complexes: highly in vitro anticancer effective
759
cisplatin derivatives involving halogeno-substituted 7-azaindole. J Inorg Biochem.
760
2012; 115: 57–63.
761
12. Mu ho á T, P a hařo á J, Šta ha P, Oli o á R, V á a O, Benesova B, et al. Insight
762
into the toxic effects of cis-dichloridoplatinum(II) complexes containing 7-azaindole
763
halogeno derivatives in tumor cells. J Biol Inorg Chem. 2013; 18: 579–589.
764
13. Šta ha P, Hošek J, Va čo J, D ořák ), Su hý P, Popa I, et al. Pharmacological and
765
molecular effects of platinum(II) complexes involving 7-azaindole derivatives. PLoS
766
ONE. 2014; 9: e90341.
767
14. New EJ, Roche C, Madawala R, Zhang JZ, Hambley TW. Fluorescent analogues of
768
quinoline reveal amine ligand loss from cis and trans platinum(II) complexes in
769
cancer cells. J Inorg Biochem. 2009; 103: 1120–1125.
- 35 -
770
15. )a o a A, Rod íguez V, Cutillas N, Yellol GS, Espi osa A, Samper KG, et al. New
771
steroidal 7-azaindole platinum(II) antitumor complexes. J Inorg Biochem. 2013; 128:
772
48–56.
773
16. B a e V, Paleček E
774
surfaces: II. Conformational changes in double-helical polynucleotides. Biophys
775
Chem. 1976; 4: 79–92.
776
17. Dhara SC. A rapid method for the synthesis of cis-[Pt(NH3)2Cl2]. Indian J Chem.
777
1970; 8: 193–194.
778
18. Ro e o-Ca eĺ
779
multitargeting via redox modulation. Inorg Chem. 2013; 52: 12276–12291.
780
19. Brabec
781
diamminedichloroplatinum(II) are preferentially formed between guanine and
782
complementary cytosine residues. Proc Natl Acad Sci USA. 1993; 90: 5345–5349.
783
20. Farrell N, Qu Y, Feng L, Van Houten B. Comparison of chemical reactivity,
784
cytotoxicity, interstrand cross-linking and DNA sequence specificity of bis(platinum)
785
complexes containing monodentate or bidentate coordination spheres with their
786
monomeric analogues. Biochemistry. 1990; 29: 9522–9531.
787
21. Butour JL, Macquet JP. Differentiation of DNA - platinum complexes by
788
fluorescence. The use of an intercalating dye as a probe. Eur J Biochem. 1977; 78:
789
455–463.
790
22. Butour JL, Alvinerie P, Souchard JP, Colson P, Houssier C, Johnson NP. Effect of
791
the amine nonleaving group on the structure and stability of DNA complexes with
792
cis-[Pt(R-NH2)2(NO3)2]. Eur J Biochem. 1991; 202: 975–980.
V,
I te a tio of u lei a ids
ith ele trically charged
I, Sadle PJ. Next-generation metal anticancer complexes:
Leng
M.
DNA
interstrand
- 36 -
cross-links
of
trans-
793
23. Fichtinger-Schepman AMJ, Van der Veer JL, Den Hartog JHJ, Lohman PHM,
794
Reedijk J. Adducts of the antitumor drug cis-diamminedichloroplatinum(II) with DNA:
795
formation, identification, and quantitation. Biochemistry. 1985; 24: 707–713.
796
24. Eastman A. The formation, isolation and characterization of DNA adducts
797
produced by anticancer platinum complexes. Pharm Ther. 1987; 34: 155–166.
798
25. Ramos-Lima FJ, Vrana O, Quiroga AG, Navarro-Ranninger C, Halamikova A,
799
Rybnickova H, et al. Structural characterization, DNA interactions, and cytotoxicity of
800
new transplatin analogues containing one aliphatic and one planar heterocyclic
801
amine ligand. J Med Chem. 2006; 49: 2640–2651.
802
26. Šta ha P, Ma ek J, T á íček ). Cisplatin and oxaliplatin derivatives involving 7-
803
azaindole: Structural characterisations. Polyhedron. 2012; 33: 404–409.
804
27. Zhao Y, Woods JA, Farrer NJ, Robinson KS, Pracharova J, Kasparkova J, et al.
805
Diazido mixed-amine platinum(IV) anticancer complexes activatable by visible-light
806
form novel DNA adducts. Chem-Eur J. 2013; 19: 9578–9591.
807
28. Jansen BAJ, Brouwer J, Reedijk J. Glutathione induces cellular resistance against
808
cationic dinuclear platinum anticancer drugs. J Inorg Biochem. 2002; 89: 197–202.
809
29. Romero-Ca eĺ I, Salassa L, Sadle PJ. The contrasting activity of iodido versus
810
chlorido ruthenium and osmium arene azo- and imino-pyridine anticancer
811
complexes: control of cell selectivity, cross-resistance, p53 dependence, and
812
apoptosis pathway. J Med Chem. 2013; 56: 1291–1300.
813
30. Frey U, Ranford JD, Sadler PJ. Ring-opening reactions of the anticancer drug
814
carboplatin: NMR characterization of cis-[Pt(NH3)2(CBDCA-O)(5'-GMP-N7)] in
815
solution. Inorg Chem. 1993; 32: 1333–1340.
- 37 -
816
31. Hay RW, Miller S. Reactions of platinum(ll) anticancer drugs. Kinetics of acid
817
hydrolysis of cis-diammine(cyclobutane-l,1-dicarboxylato)platinum(II) "Carboplatin".
818
Polyhedron. 1998; 17: 2337–2343.
819
32. Pavelka M, Fatima M, Lucas A, Russo N. On the hydrolysis mechanism of the
820
second-generation anticancer drug Carboplatin. Chem-Eur J. 2007; 13: 10108–10116.
821
33. Jamieson ER, Lippard SJ. Structure, recognition, and processing of cisplatin-DNA
822
adducts. Chem Rev. 1999; 99: 2467–2498.
823
34. Brabec V. DNA modifications by antitumor platinum and ruthenium compounds:
824
their recognition and repair. Prog Nucleic Acid Res Mol Biol. 2002; 71: 1–68.
825
35. Lemaire MA, Schwartz A, Rahmouni AR, Leng M. Interstrand cross-links are
826
preferentially formed at the d(GC) sites in the reaction between cis-
827
diamminedichloroplatinum(II) and DNA. Proc Natl Acad Sci USA. 1991; 88: 1982–
828
1985.
829
36. Brabec V, Boudny V, Balcarova Z. Monofunctional adducts of platinum(II)
830
produce in DNA a sequence-dependent local denaturation. Biochemistry. 1994; 33:
831
1316–1322.
832
- 38 -
833
Supporti g I for atio
834
S1 Text. The characterization data (1H and 13C NMR, elemental analysis, FTIR and
835
ESI-MS) for 1–6.
836
S1 Fig. 1H NMR,
837
spectra of 5. The 1H-NMR (up left), 13C-NMR (up right), 1H–1H gs-COSY (middle left),
838
1
839
solution of 5 in DMF-d7; the chemical shift values are given in Experimental section in
840
the main text.
841
S1 Table. The 1H and
842
δligand; ppm) of the prepared complexes.
843
The 1H and 13C NMR oo di atio shifts
844
the prepared complexes.
845
S2 Fig. ESI+ mass spectrum of 5. ESI+ mass spectrum (0–800 m/z range) of the
846
methanolic solution of the complex 5 (A) and its part between 720 and 765 m/z
847
showing the molecular peak (together with isotopic distribution) and its adduct with
848
sodium ion observed experimentally (B) and calculated (C).
849
S3 Fig. Time-dependent 1H NMR spectra before and after UVA irradiation of 5.
850
Time-dependent (fresh solution and after 24 h) 400 MHz 1H NMR spectra as
851
observed before and after UVA irradiation (20 min, 365 nm) of the complex 5
852
dissolved in the DMF-d7/H2O solution (1:1, v/v).
853
S4 Fig. ESI+ and ESI– mass spectra of 5 and its mixtures with GSH or GMP with or
854
without UVA irradiation. ESI+ (left) and ESI– (right) mass spectra (100–1200 m/z
855
range) of the complex 5 and its mixtures with GSH or Na2GMP (dissolved in the DMF-
13
C NMR, 1H–1H gs-COSY, 1H–13C gs-HMQC and 1H–13C gs-HMBC
H–13C gs-HMQC (middle right) and 1H–13C gs-HMBC (down) spectra obtained on the
13
C NMR coordination shifts ( al ulated as ∆δ = δcomplex –
al ulated as ∆ =
- 39 -
complex
–
ligand;
ppm) of
856
d7/H2O, 1:1, v/v) as detected on the samples with or without irradiation (20 min, 365
857
nm) 24 h after preparation. ♦ stands for {4Braza+H}+, × fo {GMP–Na+2H}+, {GMP+H}+
858
or {GMP–2Na–H}–, ● fo {GS–SG+H}+ or {GS–SG+Na}+, ○ fo {[Pt
859
{[Pt(cbdc)(4Braza)2]+Na}+ or {[Pt(cbdc)(4Braza)2]–H}–, ♯ for {[Pt(cbdc)(4Braza)]–H}–,
860
a d ◊ fo
861
{[Pt(cbdc)(4Braza)(GMP)]–2Na–H}–.
862
S5 Fig. 1H NMR spectrum after UVA irradiation of the mixture of 5 and GMP. 400
863
MHz 1H NMR spectrum as observed after UVA irradiation (20 min, 365 nm) of the
864
mixture of the complex 5 and GMP dissolved in the DMF-d7/H2O solution (1:1, v/v).
865
S6 Fig. ESI– mass spectrum of the {[Pt(cbdc)(4Braza)(GMP)]–2Na+H}– species.
866
Experimental (up) and simulated (down) mass spectrum isotope distribution of the
867
{[Pt(cbdc)(4Braza)(GMP)]–2Na+H}– species detected in the ESI– mass spectrum of
868
the complex 5 and GMP mixture dissolved in the DMF-d7/H2O solution (1:1, v/v). The
869
fresh mixture was irradiated (20 min, 365 nm) and the spectrum was recorded 24 h
870
after preparation.
871
S7 Fig. Impact of UVA irradiation of platinum(II) carboxylato complexes with 7-
872
azaindoles as carrier ligands on their cytotoxicity.
{[Pt
d
d
4Braza)2]+H}+,
4Braza)(GMP)]–Na+2H}+, {[Pt(cbdc)(4Braza)(GMP)]+H}+ or
873
874 875
- 40 -
PŘÍLOHA 7 Štarha, P.; Stavárek, M.; Tuček, J.; Trávníček, Z. 4-Aminobenzoic acid-coated maghemite nanoparticles as potential anticancer drug magnetic carriers: a case study on highly cytotoxic cisplatin-like complexes involving 7-azaindoles Molecules 19 (2014) 1622–1634
Molecules 2014, 19, 1622-1634; doi:10.3390/molecules19021622 OPEN ACCESS
molecules ISSN 1420-3049 www.mdpi.com/journal/molecules Article
4-Aminobenzoic Acid-Coated Maghemite Nanoparticles as Potential Anticancer Drug Magnetic Carriers: A Case Study on Highly Cytotoxic Cisplatin-Like Complexes Involving 7-Azaindoles Pavel Štarha 1, Martin Stavárek 1, Jiří Tuček 2 and Zdeněk Trávníček 1,* 1
2
Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacký University, 17. listopadu 12, Olomouc CZ 77146, Czech Republic; E-Mails:
[email protected] (P.Š.);
[email protected] (M.S.) Regional Centre of Advanced Technologies and Materials, Department of Experimental Physics, Faculty of Science, Palacký University, 17. listopadu 12, Olomouc CZ 77146, Czech Republic; E-Mail:
[email protected]
* Author to whom correspondence should be addressed; E-Mail:
[email protected] Tel.: +420-585-634-352; Fax: +420-585-634-954. Received: 23 December 2013; in revised form: 17 January 2014 / Accepted: 23 January 2014 / Published: 28 January 2014
Abstract: This study describes a one-pot synthesis of superparamagnetic maghemite-based 4-aminobenzoic acid-coated spherical core-shell nanoparticles (PABA@FeNPs) as suitable nanocomposites potentially usable as magnetic carriers for drug delivery. The PABA@FeNPs system was subsequently functionalized by the activated species (1* and 2*) of highly in vitro cytotoxic cis-[PtCl2(3Claza)2] (1; 3Claza stands for 3-chloro7-azaindole) or cis-[PtCl2(5Braza)2] (2; 5Braza stands for 5-bromo-7-azaindole), which were prepared by a silver(I) ion assisted dechlorination of the parent dichlorido complexes. The products 1*@PABA@FeNPs and 2*@PABA@FeNPs, as well as an intermediate PABA@FeNPs, were characterized by a combination of various techniques, such as Mössbauer, FTIR and EDS spectroscopy, thermal analysis, SEM and TEM. The results showed that the products consist of well-dispersed maghemite-based nanoparticles of 13 nm average size that represent an easily obtainable system for delivery of highly cytotoxic cisplatin-like complexes in oncological practice.
Molecules 2014, 19
1623
Keywords: maghemite; nanoparticles; magnetic; platinum complexes; 7-azaindole derivatives; drug delivery
1. Introduction The platinum(II) complexes, such as cisplatin, oxaliplatin or carboplatin, are well-established anticancer chemotherapeutics used worldwide for the treatment of various types of cancer [1]. Application of these drugs causes several negative side-effects, such as nephrotoxicity, neurotoxicity or myelosuppression, which represent a permanent incentive for bioinorganic chemists to find novel non-platinum drugs (e.g., ruthenium complexes) or platinum complexes with diminished side-effects or to study the possibilities of the targeted drug delivery of the known as well as novel cytotoxic platinum complexes to the tumour tissues. The latter option, i.e., targeted drug delivery, offers various approaches, as reviewed e.g., in [2–4]. One of them is based on the use of magnetic nanoparticles (NPs) coated with a suitable shell, comprising e.g., organic molecules of the same entity, that is able to interact with the drug [5,6]. A distribution of such systems within the organism suffering with cancer could be affected by an external magnetic field, whose application concentrates the drug into the tumour tissue. Among all the magnetic nanoparticles of transition metals and their oxides, iron oxide-based nanosystems hold a paramount position in various medical fields due to their promising properties including magnetic (e.g., superparamagnetism, strong magnetic response and saturation under small applied magnetic fields, excellent heating performance in the frequencies of the alternating magnetic field safe for humans) and biochemical (e.g., very low toxicity, biocompatibility, biodegradability) features [7,8]. To date, they have been successfully employed as negative contrast agents in magnetic resonance imaging, for cell labelling and separation, drug delivery, and as functional components in magnetically-assisted hyperthermia for cancer treatment. Once iron oxide nanoparticles are functionalized with suitable bioactive substances, the resulting system then performs both the diagnostic and therapeutic actions, giving birth to a novel branch of medicine known as theranostics [9]. Many iron oxide-based (both maghemite or magnetite) nanocarriers designed and fabricated for magnetic drug delivery purposes have been reported in the literature to date, involving various synthetic approaches and the use of different organic-coating layers [5,6,10–19]. Regarding the nanocomposites with maghemite-based core-shell NPs functionalized with platinum-based drugs (i.e., structurally similar to those reported in this work), to the best of our knowledge only two such works have been reported to date. Both of them report maghemite-based NPs with dechlorinated cisplatin bound to 4-oxo-4-(3(triethoxysilyl)propylamino)butanoic acid (OTPBA) [20,21]. These systems showed remarkable, timedependent in vitro cytotoxicity against MCF7, HeLa, A549 and A549R human cancer cell lines, which is comparable with that of cisplatin after 72 h, and moreover, they can be simultaneously used as MRI contrast agents. Several other works have dealt with similar magnetite-based NPs suitable for magnetic drug delivery and studied cisplatin as the model drug as follows: Deng and Lei reported the Fe3O4/SiO2 cores with PEG–PLA shell (PEG = polyethylene glycol, PLA = polylactic acid) and loaded cisplatin [22]. A similar system, but with a PLA shell, was studied by Devi et al., who focused on
Molecules 2014, 19
1624
loading and release properties of cisplatin [23]. Ashjari et al. published the preparation of cisplatinfunctionalized magnetite NPs with biodegradable poly(lactic-co-glycolic acid) (PLGA) with different morphological properties of the resulting composite [24]. Several other studies focused on the analogical systems (magnetite core and cisplatin as the functionalizing agent) differing in organic shells, such as folate acid- [25], squalene- [26], carboxymethylcellulose- [27], dextrane- [28] or poly(ethyl-2-cyanoacrylate)- [29] based layer. Finally, Li et al. studied the effect of cisplatin-loaded magnetite-based NPs on multidrug resistance and its mechanism [30]. Although all of the mentioned nanosystems have to be considered as universal in terms of functionalization by the platinum complexes, it has to be noted that to the best of our knowledge only one work [31] has reported iron oxide-based NPs functionalized with platinum complexes other than the mentioned cisplatin. In particular, the magnetite-silica composite nanoparticles were investigated as carriers of a photoactive platinum diimine complex. In an effort to research the possibilities of targeted delivery of the recently reported highly in vitro cytotoxic cisplatin-like complexes involving 7-azaindole derivatives investigated by our team [32–34], we developed a novel system based on easily obtainable magnetic nanoparticles. They consist of maghemite-based 4-aminobenzoic acid (PABA)-coated core-shell nanoparticles (PABA@FeNPs) functionalized by the activated (i.e., dechlorinated) platinum(II) complexes bearing various 7-azaindoles (the complexes 1 and 2, whose activated form is symbolized as 1* and 2*; Figure 1a). The obtained systems 1*@PABA@FeNPs and 2*@PABA@FeNPs (Figure 1b) were thoroughly characterized by relevant techniques including Mössbauer and FTIR spectroscopy, simultaneous TG/DTA thermal analysis, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) equipped with energy-dispersive X-ray spectroscopy (EDS). The resulting nanocomposites, which were found to be 13.0 ± 2.1 nm of size with acceptable dispersibility, are of high potential from the magnetic drug delivery point of view. Figure 1. (a) The structural formulas of highly cytotoxic cis-[PtCl2(3Claza)2] (1; 3Claza = 3-chloro-7-azaindole) and cis-[PtCl2(5Braza)2] (2; 5Braza = 5-bromo-7azaindole), and the corresponding activated diaqua species symbolized as 1* and 2* used for the interaction with PABA@FeNPs; (b) the proposed composition of the studied 1*@PABA@FeNPs and 2*@PABA@FeNPs based on the maghemite core coated with 4-aminobenzoic acid and functionalized with 1* or 2*; and (c) photos of the obtained aqueous suspensions of PABA@FeNPs (up) and 2*@PABA@FeNPs (down) without (left) and with (right) an external magnetic field.
Molecules 2014, 19
1625
2. Results and Discussion 2.1. Preparation and Properties of PABA@FeNPs The maghemite-based nanoparticles (FeNPs) coated with 4-aminobenzoic acid (PABA@FeNPs) were prepared be a method using a mixture of the Fe(III) and Fe(II) salts (FeCl3·6H2O and FeCl2·4H2O in this work), which was mixed together with PABA in deionized water under atmospheric conditions. NH4OH was finally added to the mixture which resulted in the formation of the maghemite-based PABA@FeNPs nanoparticles. It has to be noted that the usual preparation of maghemite-based NPs involves magnetite NPs (prepared from the mixture of Fe(III) and Fe(II) salts under nitrogen [35,36]) and subsequently oxidized by e.g., diluted nitric acid [20,21]. Since we aimed to prepare maghemitebased NPs, we used a different approach—we did not perform the syntheses under nitrogen but under atmospheric conditions, which was found to be sufficient for the magnetite oxidation to occur without addition of any other oxidizing agent. The presence of the organic layer coating the maghemite NPs within the obtained PABA@FeNPs was proved by the FTIR spectra recorded in the 400–4,000 cm−1 region (Figure 2 and Figure S1) as well as by the thermal analysis (Figure S2) [35,37]. The FTIR spectrum of PABA@FeNPs contained the characteristic peak of maghemite at ca. 550 cm−1 clearly assignable to the (Fe–O) vibration, as well as a series of peaks, whose positions in the spectrum correlates with those of free PABA molecule (see Figure S1 and Experimental Section). The results of the FTIR spectroscopy also indirectly proved that PABA is bonded to γ-Fe2O3 within the PABA@FeNPs nanocomposite through the deprotonated carboxyl group, because there is only one major peak in the region characteristic for this functional group with a band centred at 1,603 cm−1, as compared with four in total peaks found in the spectrum of free PABA at 1,571, 1,597, 1,623 and 1,660 cm−1 (Figure S1). Figure 2. FTIR spectra of the maghemite nanoparticles coated with 4-aminobenzoic acid (PABA@FeNPs; red line), the parent complex 2 involving 5-bromo-7-azaindole (blue line), and the resulting system with the activated complex (2*) bound on the maghemite nanoparticles coated with 4-aminobenzoic acid (2*@PABA@FeNPs).
Molecules 2014, 19
1626
The PABA@FeNPs hybrid systems exhibited, as expected, a strong attraction to the magnetic field (Figure 1c) and good stability in solution (no macroscopic changes, e.g., colour or magnetism, after one weak of standing in the fridge) and in solid state (usable with no changes of their properties after more than two months of standing in the fridge). 2.2. Functionalization of PABA@FeNPs by Platinum(II) Complexes We assumed a bonding of the platinum(II) complexes through Pt–N bonds formed between the Pt(II) atom of the activated species and the amino group of PABA. As previously reported, PABA with a substituted carboxyl group (which simulates the bonding of PABA to the maghemite NPs through the mentioned functional group) binds to platinum through the amino group [38–40]. Further, the X-ray structures of several platinum(II) complexes involving aniline or its variously substituted derivatives, such as 4-alkylanilines (e.g., [41,42]) are described in the literature. These facts proved PABA is a suitable coating agent for magnetic therapeutic/theranostic systems functionalized with platinum-based agents. The initial cis-dichloridoplatinum(II) complexes of the composition cis-[PtCl2(naza)2] (1 and 2), which were recently described in the literature by our team as having significant antitumor properties [33,34], were activated by their reactions with a stoichiometric amount of silver(I) nitrate, resulting in dechlorination and formation of the diaquaplatinum(II) species cis-[Pt(H2O)2(3Claza)2]2+ (1*) and cis-[Pt(H2O)2(5Braza)2]2+ (2*). The activated species, involving the labile Pt–aqua bonds which represent suitable sites for consequent formation of the mentioned Pt–N bonds with PABA, were allowed to interact in acetone with the PABA@FeNPs nanoparticles for 48 h. The final systems (1*@PABA@FeNPs and 2*@PABA@FeNPs), which showed strong attraction to the external magnetic field (Figure 1c), were magnetically isolated, purified and stored in the fridge. The presence of the platinum(II) species within the obtained maghemite-based nanocomposite was proved by FTIR spectroscopy (Figure 2) as well as by EDS spectroscopy (Figure 3f). FTIR spectroscopic experiments were performed with the aim to better characterize the studied systems. The detailed analysis of FTIR spectra revealed that bands observed in the spectrum of 1*@PABA@FeNPs at 785, 1,018, 1,099, 1,277, 1,338, 1,516, 1,602, 2,906 and 3,107 cm−1 and in the spectrum of 2*@PABA@FeNPs at 700, 883, 946, 1,267, 1,341, 1,473, 1,603 and 3,082 cm−1 were not detected in the spectrum of the PABA@FeNPs system. Moreover, the positions of these bands correlate well with those detected in the spectra of the complexes 1 and 2 (Figure 2 for the complex 2 and 2*@PABA@FeNPs), thus showing on the presence of the platinum(II) 7-azaindole species within the resulting nanocomposites. An interpretation of the far-FTIR spectra recorded at 150–600 cm−1 provides indirect proof of the covalent bonding between the platinum(II) species and PABA@FeNPs within the studied nanosystems 1*@PABA@FeNPs and 2*@PABA@FeNPs. In particular, the performed far-FTIR experiments regarding 2, 2* and 2*@PABA@FeNPs indicated changes in inner coordination spheres in the vicinity of the central platinum(II) atom, i.e., the changes going from a PtCl2N2 donor set (the starting complex 2; two (Pt–Cl) maxima at 336 and 345 cm−1), through a PtN2O2 donor set (the dechlorinated complex 2*; two (Pt–O) maxima at 322 and 332 cm−1) to a PtN4 one (the resulting system 2*@PABA@FeNPs; no bands detected in the region mentioned for both the (Pt–Cl) or (Pt–O) vibrations), as depicted in Figure S3. In other words, although we did not detect the vibrations connected with the anticipated
Molecules 2014, 19
1627
Pt–N bonds between the platinum(II) species and amino group of PABA, we assume that these vibrations are overlapped (analogically to those at ca 520 cm−1 assignable to Pt–N bonds between the central Pt(II) atom and 7-azaindole rings) by a wide and intensive band belonging to (Fe–O). The (Fe–O) vibration was, together with a band centred at 579 cm−1 assignable to the vibrations connected with the deformation of the 7-azaindole moiety [43], the only band detected in the discussed far-FTIR spectrum of 2*@
[email protected] images of the prepared 1*@PABA@FeNPs and 2*@PABA@FeNPs systems provided the relevant information regarding the shape, size, and uniformity of the resulting NPs (Figures 3a,b). The systems were found to be spherical, core-shell well-dispersed composites with an average size of 13.0 ± 2.1 nm. SEM was used to investigate the surface morphology of the prepared maghemite-based NPs (Figure 3c,e). A comparison of the SEM images depicted for PABA@FeNPs (Figure 3c) and 2*@PABA@FeNPs (Figure 3e) did not show any noticeable difference between their properties, since both systems were detected by SEM (as well as by the above-mentioned TEM) as having a spherical shape of individual NPs, which agglomerated together to the structure without any specific shape. Figure 3. TEM images of 2*@PABA@FeNPs {(a) with 100 nm resolution and (b) with 50 nm resolution}, SEM images of PABA@FeNPs (c) and 2*@PABA@FeNPs (e) given with their EDS patterns (d, f).
Simultaneous TG/DTA thermal analysis revealed a considerable difference between the weight losses of PABA@FeNPs and those involving functionalized platinum(II) complexes (represented by 2*@PABA@FeNPs; Figure S2). PABA@FeNPs did not show any weight increase in the 100–150 °C range (after the loss of water physically adsorbed on the the prepared NPs), which is known to be
Molecules 2014, 19
1628
connected with an oxidation of Fe2+ ions, which indirectly proved the chemical composition of the magnetic core (maghemite). The PABA@FeNPs system was thermally stable between 114 and 148 °C, when its thermal degradation (connected with an oxidation of the organic layer coating the maghemite core) started and continued to 460 °C (total weight loss equals 8.0%). The product of the thermal decomposition, most probably γ-Fe2O3 as previosly reported for maghemite NPs [44], did not show any weight change up to 530 °C, however, an exothermic effect unambiguously assignable to the γ-Fe2O3 to α-Fe2O3 conversion was detected on the DTA curve with the maximum at 488 °C (Figure S2). A considerably different weight loss (19.8%) as well as TG-curve shape (a continual decomposition) was found for 2*@PABA@FeNPs indirectly proving the presence of the platinum(II) species within the resulting system (Figure S2). 2.3. Mössbauer Spectroscopy The 57Fe Mössbauer spectra of the samples studied are depicted in Figure 4, while the values of the Mössbauer hyperfine parameters, derived from the fitting of the recorded Mössbauer spectra, are listed in Table 1. Figure 4. Room-temperature Mössbauer spectra of (a) PABA@FeNPs and (b) 2*@PABA@FeNPs; doublet - assigned to the Fe3+ relaxation component, singlet - asigned to the Fe3+ superparamagnetic component.
Table 1. Values of the Mössbauer hyperfine parameters, derived from the fitting of the room-temperature Mössbauer spectra of PABA@FeNPs and 2*@PABA@FeNPs, where δ is the isomer shift, ΔEQ is the quadrupole splitting and RA is the spectral area of the individual spectral components. Sample
Component
PABA@FeNPs
Doublet
δ ± 0.01 (mm/s) 0.34
Singlet
0.35
Doublet
0.35
Singlet
0.35
2*PABA@FeNPs
ΔEQ ± 0.01 (mm/s) 0.66
RA ± 1 (%) 18 82
0.69
88 78
Assignment Fe3+ relaxation component Fe3+ superparamagnetic component Fe3+ relaxation component Fe3+ superparamagnetic component
Molecules 2014, 19
1629
The room-temperature Mössbauer spectrum of both the PABA@FeNPs and 2*@PABA@FeNPs sample shows only relaxation components (i.e., singlet and doublet, see Figures 4a,b, respectively) with the values of the Mössbauer hyperfine parameters (see Table 1) typical of high-spin Fe(III) atom in iron(III) oxides [45]; there is no indication of presence of the Fe2+ valence state. Thus, the nanoparticles in both samples are solely of -Fe2O3 origin. This is expected in connection with their small size with a large surface area securing their complete oxidation. On the timescale of the Mössbauer technique, all the nanoparticles in both assemblies behave in a superparamagnetic manner at room temperature. The doublet component belongs to the nanoparticles with thermally fluctuating superspins having relaxation times much smaller than the characteristic measurement time (τm) of the Mössbauer spectroscopy, while the presence of a singlet corresponds to those nanoparticles the superspin of which thermally fluctuates between the energetically favored orientations with a relaxation time close to τm. The PABA@FeNPs and 2*@PABA@FeNPs systems would show superparamagnetic features in the magnetization measurements at room temperature, however, their magnetic characteristics will be significantly driven by finite-size and surface effects (manifested, for example, by a smaller saturation magnetization or lack of saturation and reduced magnetic response under small applied magnetic fields). 3. Experimental 3.1. Materials and Methods The starting materials K2[PtCl4], 3-chloro-7-azaindole (3Claza), 5-bromo-7-azaindole (5Braza), iron(III) chloride hexahydrate (FeCl3·6H2O), iron(II) chloride tetrahydrate (FeCl2·4H2O), p-aminobenzoic acid (PABA), 25% NH4OH, silver nitrate (AgNO3) and solvents were supplied by Sigma-Aldrich Co. (Prague, CzechRepublic) and Acros Organics Co. (Pardubice, CzechRepublic), and used as received. The platinum(II) complexes cis-[PtCl2(3Claza)2] (1) and cis-[PtCl2(5Braza)2] (2) (Figure 1a) were prepared as described in our recent paper [33]. Transmission electron microscopy (TEM) was carried out on a JEOL 2010 microscope (200 kV, 1.9 Å point-to-point resolution). A drop of high-purity water with the ultrasonically dispersed samples was placed onto a holey-carbon film supported by a copper-mesh TEM grid and dried in air at room temperature. Scanning electron microscopy (SEM) was performed, together with energy-dispersive X-ray (EDS) spectroscopy, by a Hitachi 6600 FEG microscope (5–15 keV accelerating voltage; the dried samples were placed on an aluminum holder equipped with double-sided adhesive carbon tape). The 57Fe Mössbauer spectra of the studied samples were recorded at room temperature employing a Mössbauer spectrometer operating at the constant acceleration mode and equipped with a 50 mCi 57 Co(Rh) source. The isomer shift values are related to α-Fe at room temperature. The Mössbauer spectra were fitted with the MossWinn software program; prior to fitting, the signal-to-noise ratio was enhanced by a statistically based algorithm [46]. Infrared spectra (400–4000 cm−1 and 150–600 cm−1 regions) were recorded on a Nexus 670 FT-IR (Thermo Nicolet, Waltham, MA, USA) using the ATR technique. Simultaneous thermogravimetric (TG) and differential thermal (DTA) analyses were performed using an Exstar TG/DTA 6200 thermal analyzer (Seiko Instruments Inc., Chiba, Japan) from room temperature to 650 °C (5.0 °C min−1) in dynamic air atmosphere (50 mL min−1).
Molecules 2014, 19
1630
3.2. PABA@FeNPs Nanoparticles 4-Aminobenzoic acid (PABA; 0.62 g, 4.5 mmol) was, due to its poor solubility at laboratory temperature, dissolved in hot (80 °C) deionized water (75 mL) and then, FeCl3·6H2O (1.17 g; 4.3 mmol) dissolved in deionized water (5 mL) was added. The mixture was stirred at 80 °C for 30 min and then FeCl2·4H2O (0.86 g; 4.3 mmol) dissolved in deionized water (2 mL) was added to the solution. After 30 min of stirring at 80 °C the last reagent (10 mL of 25% NH4OH) was slowly added. The solution turned dark black as the PABA@FeNPs formed. The suspension was intensively stirred at 80 °C for 60 min. Finally, nanoparticles were magnetically isolated and washed with deionized water (3 × 20 mL) and acetone (3 × 20 mL). Part of the final ferrofluid suspension of PABA@FeNPs (Figure 1c) was stored under degassed acetone in the fridge, while the rest of the product was dried with nitrogen gas and in desiccator over silica gel, and stored in the fridge. PABA@FeNPs: FTIR (νATR/cm−1): 552vs, 783w, 1,179m, 1,395vs, 1,493m, 1,603s, 3,219s, 3,334s. 3.3. Synthesis of 1*@PABA@FeNPs and 2*@PABA@FeNPs The complexes 1 (110 mg; 0.2 mmol) and 2 (130 mg; 0.2 mmol) were dissolved in acetone (10 mL) and two molar equivalents of AgNO3 dissolved in a minimum volume of deionized water were added. The mixture was stirred at laboratory temperature in the dark for 24 h. After that, AgCl was removed by filtration and washed with acetone (3 × 1 mL) to produce the filtrate containing the solution of the dechlorinated complexes of the composition cis-[Pt(H2O)2(3Claza)2]2+ (1*), and cis-[Pt(H2O)2(5Braza)2]2+ (2*) (Figure 1a). PABA@FeNPs (0.5 mL of the acetone suspension involving 100 mg of PABA@FeNPs) was poured in and the mixture was stirred for 48 h to produce the final systems 1*@PABA@FeNPs and 2*@PABA@FeNPs (Figures 1b,c). These products were magnetically isolated, washed with acetone (3 × 20 mL), dried (nitrogen and then in desiccator over silica gel) and stored in the fridge. 1*@PABA@FeNPs: FTIR (νATR/cm−1): 550vs, 785w, 1,018m, 1,099w, 1,179m, 1,208m, 1,277s, 1,338s, 1,398vs, 1,516m, 1,602s, 1,694w, 2,906s, 3,107s. 2*@PABA@FeNPs: FTIR (νATR/cm−1): 547vs, 700w, 883w, 946w, 1,017w, 1,101w, 1,177m, 1,267vs, 1,341s, 1,400vs, 1,473s, 1,500s, 1,603m, 1,646s, 1,694w, 3,082s. For far-FTIR spectra of 2, 2*, PABA@FeNPs and 2*@PABA@FeNPs see Supplementary Materials. 4. Conclusions A simple approach was applied to obtain magnetic 4-aminobenzoic acid-coated maghemite nanoparticles with good stability in solution, with high magnetic response to the external magnetic field and showing superparamagnetic behaviour as proved by the Mössbauer spectroscopy experiments at room temperature. The systems were designed to be able to bind highly cytotoxic platinum(II) complexes involving 7-azaindole derivatives represented by the diaquaplatinum(II) species 1* and 2* prepared by a Ag(I) ion-assisted activation from the initial highly cytotoxic dichlorido complexes. The incorporation of the platinum(II) species was proved by relevant techniques (FTIR, EDS), while a combination of the microscopic techniques (SEM, TEM) showed the obtained core-shell nanocomposites as having the spherical shape and an average size of 13 nm in diameter. Although we are aware of the fact that other important properties of the reported systems (e.g., release of the
Molecules 2014, 19
1631
complex, in vitro cytotoxicity, in vitro toxicity or MRI experiments), should by also elucidated (the results will be a subject of our forthcoming studies), we have reason to believe that the prepared systems fulfil the basic requirements (magnetism, size, dispersibility or functionalization with therapeutically active substance) for the nanoparticles used in the field of theranostic nanomedicine. Supplementary Materials Supplementary materials can be accessed at: http://www.mdpi.com/19/2/1622/s1. Acknowledgments The authors gratefully thank the Czech Science Foundation (GAČRP207/11/0841), Operational Program Research and Development for Innovations-European Regional Development Fund (CZ.1.05/2.1.00/03.0058), Operational Program Education for Competitiveness-European Social Fund (CZ.1.07/2.3.00/20.0017) of the Ministry of Education, Youth and Sports of the Czech Republic and Palacký University in Olomouc (PrF_2013_015). The authors would like to thank Jana Gáliková and Klára Šafářová for performing FTIR, and TEM and SEM experiments, respectively. Author Contributions Conceived and designed the experiments: PŠ, ZT. Performed the experiments: MS, PŠ, JT. Analyzed the data: PŠ, JT, ZT. Wrote the paper: PŠ, JT, ZT. Conflicts of Interest The authors declare no conflict of interest. References 1. 2. 3. 4. 5. 6. 7.
Kelland, L. The resurgence of platinum-based cancer chemotherapy. Nat. Rev. Cancer 2007, 7, 573–584. Butler, J.S.; Sadler, P.J. Targeted delivery of platinum-based anticancer complexes. Curr. Opin. Chem. Biol. 2013, 17, 175–188. Zhang, Y.; Chan, H.F.; Leong, K.W. Advanced materials and processing for drug delivery: The past and the future. Adv. Drug Deliv. Rev. 2013, 65, 104–120. Goncalves, A.S.; Macedo, A.S.; Souto, E.B. Therapeutic nanosystems for oncology nanomedicine. Clin. Transl. Oncol. 2012, 14, 883–890. Chomoucka, J.; Drbohlavova, J.; Huska, D.; Adam, V.; Kizek, R.; Hubalek, J. Magnetic nanoparticles and targeted drug delivering. Pharmacol. Res. 2010, 62, 144–149. Kievit, F.M.; Zhang, M. Surface Engineering of iron oxide nanoparticles for targeted cancer therapy. Acc. Chem. Res. 2011, 44, 853–862. Gupta, A.K.; Gupta, M. Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications. Biomaterials 2005, 26, 3995–4021.
Molecules 2014, 19 8.
9. 10.
11.
12. 13.
14.
15. 16.
17.
18.
19. 20.
21.
22. 23.
1632
Laurent, S.; Forge, D.; Port, M.; Roch, A.; Robic, C.; Elst, L.V.; Muller, R.N. Magnetic iron oxide nanoparticles: Synthesis, Stabilization, Vectorization, Physicochemical characterizations, and biological applications. Chem. Rev. 2008, 108, 2064–2110. Kelkar, S.S.; Reineke, T.M. Theranostics: Combining Imaging and Therapy. Bioconjugate. Chem. 2011, 22, 1879–1903. Wang, N.; Guan, Y.; Yang, L.; Jia, L.; Wei, X.; Liu, H.; Guo, G. Magnetic nanoparticles (MNPs) covalently coated by PEO–PPO–PEO block copolymer for drug delivery. J. Colloid Interface Sci. 2013, 395, 50–57. Nowicka, A.M.; Kowalczyk, A.; Jarzebinska, A.; Donten, M.; Krysinski, P.; Stojek, Z. Progress in targeting tumor cells by using drug-magnetic nanoparticles conjugate. Biomacromolecules 2013, 14, 828–833. Karsten, S.; Nan, A.; Turcu, R.; Liebscher, J. A new access to polypyrrole-based functionalized magnetic core-shell nanoparticles. J. Polym. Sci. Part. A Polym. Chem. 2012, 50, 3986–3995. Fan, C.H.; Ting, C.Y.; Lin, H.J.; Wang, C.H.; Liu, H.L.; Yen, T.C.; Yeh, C.K. SPIO-conjugated, doxorubicin-loaded microbubbles for concurrent MRI and focused-ultrasound enhanced brain-tumor drug delivery. Biomaterials 2013, 34, 3706–3715. Gautiera, J.; Munniera, E.; Paillard, A.; Hervéa, K.; Douziech-Eyrollesa, L.; Soucé, M.; Duboisa, P.; Chourpa, I. A pharmaceutical study of doxorubicin-loaded PEGylated nanoparticles for magnetic drug targeting. Int. J. Pharm. 2012, 423, 16–25. Kim, J.E.; Shin, J.Y.; Cho, M.H. Magnetic nanoparticles: An update of application for drug delivery and possible toxic effects. Arch. Toxicol. 2012, 86, 685–700. Qi, L.; Wu, L.; Zheng, S.; Wang, Y.; Fu, H.; Cui, D. Cell-Penetrating Magnetic Nanoparticles for Highly Efficient Delivery and Intracellular Imaging of siRNA. Biomacromolecules 2012, 13, 2723–2730. Huang, P.; Li, Z.; Lin, J.; Yang, D.; Gao, G.; Xu, C.; Bao, L.; Zhang, C.; Wang, K.; Song, H.; et al. Photosensitizer-conjugated magnetic nanoparticles for in vivo simultaneous magnetofluorescent imaging and targeting therapy. Biomaterials 2011, 32, 3447–3458. Pan, B.; Cui, D.; Sheng, Y.; Ozkan C.; Gao, F.; He, R.; Li, Q.; Xu, P.; Huang, T. Dendrimer-modified magnetic nanoparticles enhance efficiency of gene delivery system. Cancer Res. 2007, 67, 8156–8163. Knezevic, N.Z.; Ruiz-Hernandez, E.; Hennink, W.E.; Vallet-Regi, M. Magnetic mesoporous silica-based core/shell nanoparticles for biomedical applications. RSC Adv. 2013, 3, 9584–9593. Wang, J.; Wang, X.; Song, Y.; Wang, J.; Zhang, C.; Chang, C.; Yan, J.; Qiu, L.; Wua, M.; Guo, Z. A platinum anticancer theranostic agent with magnetic targeting potential derived from maghemite nanoparticles. Chem. Sci. 2013, 4, 2605–2612. Wang, J.; Wang, X.; Song, Y.; Zhu, C.; Wang, J.; Wang, K.; Guo, Z. Detecting and delivering platinum anticancer drugs using fluorescent maghemite nanoparticles. Chem. Commun. 2013, 49, 2786–2788. Deng, H.; Lei, Z. Preparation and characterization of hollow Fe3O4/SiO2@PEG–PLA nanoparticles for drug delivery. Composites Part. B 2013, 54, 194–199. Devi, S.V.; Prakash, T. Kinetics of cisplatin release by in-vitro using poly(D,L-Lactide) coated Fe3O4 nanocarriers. IEEE Trans. Nanobiosci. 2013, 12, 60–63.
Molecules 2014, 19
1633
24. Ashjari, M.; Khoee, S.; Mahdavian, A.R. Controlling the morphology and surface property of magnetic/cisplatin-loaded nanocapsules via W/O/W double emulsion method. Colloids Surf. A Physicochem. Eng. Asp. 2012, 408, 87–96. 25. Xie, M.; Xu, Y.; Liu, J.; Zhang, T.; Zhang, H. Preparation and characterization of Folate targeting magnetic nanomedicine loaded with cisplatin. J. Nanomater. 2012, 2012, 921034. 26. Arias, J.L.; Reddy, L.H.; Othman, M.; Gillet, B.; Desmaele, D.; Zouhiri, F.; Dosio, F.; Gref, R.; Couvreur, P. Squalene based nanocomposites: A new platform for the design of multifunctional pharmaceutical theragnostics. ACS Nano. 2011, 5, 1513–1521. 27. Xing, R.; Wang, X.; Zhang, C.; Wang, J.; Zhang, Y.; Song, Y.; Guo, Z. Superparamagnetic magnetite nanocrystal clusters as potential magnetic carriers for the delivery of platinum anticancer drugs. J. Mater. Chem. 2011, 21, 11142–11149. 28. Sonoda, A.; Nitta, N.; Nitta-Seko, A.; Ohta, S.; Takamatsu, S.; Ikehata, Y.; Nagano, I.; Jo, J.; Tabata, Y.; Takahashi, M.; et al. Complex comprised of dextran magnetite and conjugated cisplatin exhibiting selective hyperthermic and controlled-release potential. Int. J. Nanomed. 2010, 5, 499–504. 29. Yang, J.; Lee, H.; Hyung, W.; Park, S.B.; Haam, S. Magnetic PECA nanoparticles as drug carriers for targeted delivery: Synthesis and release characteristics. J. Microencapsul. 2006, 23, 203–212. 30. Li, K.; Chen, B.; Xu, L.; Feng, J.; Xia, G.; Cheng, J.; Wang, J.; Gao, F.; Wang, X. Reversal of multidrug resistance by cisplatin-loaded magnetic Fe3O4 nanoparticles in A549/DDP lung cancer cells in vitro and in vivo. Int. J. Nanomed. 2013, 8, 1867–1877. 31. Zhang, Z.; Chai, A. Core-shell magnetite-silica composite nanoparticles enhancing DNA damage induced by a photoactive platinum-diimine complex in red light. J. Inor. Biochem. 2012, 117, 71–76. 32. Štarha, P.; Marek, J.; Trávníček, Z. Cisplatin and oxaliplatin derivatives involving 7-azaindole: Structural characterisations. Polyhedron 2012, 33, 104–409. 33. Štarha, P.; Trávníček, Z.; Popa, A.; Popa, I.; Muchová, T.; Brabec, V. How to modify 7-azaindole to form cytotoxic Pt(II) complexes: Highly in vitro anticancer effective cisplatin derivatives involving halogeno-substituted 7-azaindole J. Inorg. Biochem. 2012, 105, 57–63. 34. Muchová, T.; Prachařová, J.; Štarha, P.; Olivová, R.; Vrána, O.; Benešová, B.; Kašpárková, J.; Trávníček, Z.; Brabec, V. Insight into the toxic effects ofcis-dichloridoplatinum(II) complexes containing 7-azaindole halogeno derivatives in tumor cells. J. Biol. Inorg. Chem. 2013, 18, 579–589. 35. Maity, D.; Zoppellaro, G.; Sedenkova, V.; Tucek, J.; Safarova, K.; Polakova, K.; Tomankova, K.; Diwoky, C.; Stollberger, R.; Machala, L.; et al. Surface design of core–shell superparamagnetic iron oxide nanoparticles drives record relaxivity values in functional MRI contrast agents. Chem. Commun. 2012, 48, 11398–11400. 36. Hamoudeh, M.; Al Faraj, A.; Canet-Soulas, E.; Bessueille, F.; Leonard, D.; Fessi, H. Pharmaceutical nanotechnology elaboration of PLLA-based superparamagnetic nanoparticles: Characterization, magnetic behaviour study and in vitro relaxivity evaluation. Int. J. Pharm. 2007, 338, 248–257. 37. Dorniani, D.; Bin Hussein, M.Z.; Kura, A.U.; Fakurazi, S.; Shaari, A.H.; Ahmad, Z. Preparation of Fe3O4 magnetic nanoparticles coated with gallic acid for drug delivery. Int. J. Nanomed. 2012, 7, 5745–5756.
Molecules 2014, 19
1634
38. Cafaggi, S.; Esposito, M.; Parodi, B.; Viale, M. A water-soluble 1,2-diaminocyclohexaneplatinum(II) complex containing procaine hydrochloride: Synthesis and antiproliferative activity in vitro. Pharmazie 1994, 49, 617–618. 39. Kukushkin, Y.N.; Blyumental, T.O.; Konovalov, L.V. Complex compounds of platinum(II) with p- and m-aminobenzoic acids. Zh. Obshch. Khim. 1979, 49, 1376–1379. 40. Rudyi, R.I.; Solomentseva, A.I.; Cherkashina, N.V.; Evstafeva, O.N.; Salyn, Y.V.; Moiseev, I.I. Complexes of platinum(II) and platinum(IV) with aromatic amines. Koord. Khim. 1976, 2, 499–506. 41. Davies, G.R.; Hewertson, W.; Mais, R.H.B.; Owston, P.G.; Patel, C.G. π-Complexes of platinum (II) with unsaturated hydrocarbons. Part II. Crystal and molecular structure of trans-dichloro-(πdi-t-butylacetyl-ene)-p-toluidineplatinum (II). J. Chem. Soc. A 1970, 1873–1877. 42. Rochon, F.D.; Bonnier, C. Study of Pt(II)-aromatic amines complexes of the types cis- and trans-Pt(amine)2I2, [Pt(amine)4]I2 and I(amine)Pt( -I)2Pt(amine)I. Inorg. Chim. Acta 2007, 360, 461–472. 43. Karthikeyan, B. Density functional calculations on the structure, vibrational frequencies and normal modes of 7-Azaindole. Spectrochim. Acta 2006, A64, 1083–1087. 44. Ye, X.; Lin, D.; Jiao, Z.; Zhang, L. The thermal stability of nanocrystalline maghemite Fe2O3. J. Phys. D-Appl. Phys. 1998, 31, 2739–2744. 45. Cornell, R.M.; Schwertmann, U. The Iron Oxides: Structure, Properties, Reactions, Occurrence and Uses; Wiley-VCH Publishers: Weinheim, Germany, 2003. 46. Prochazka, R.; Tucek, P.; Tucek, J.; Marek, J.; Mashlan, M.; Pechousek, J. Statistical analysis and digital processing of the Mössbauer spectra. Meas. Sci. Technol. 2010, 21, 025107. Sample Availability: Samples of the compounds 1, 2, 1*, 2*, PABA@FeNPs, 1*PABA@FeNPs and 2*PABA@FeNPs are available from the authors. © 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).
PŘÍLOHA 8 Štarha, P.; Smola, D.; Tuček, J.; Trávníček, Z. Efficient synthesis of a maghemite/gold hybrid nanoparticle system as a magnetic carrier for the transport of platinum-based metallotherapeutics Int. J. Mol. Sci. 16 (2015) 2034–2051
Int. J. Mol. Sci. 2015, 16, 2034-2051; doi:10.3390/ijms16012034 OPEN ACCESS
International Journal of
Molecular Sciences ISSN 1422-0067 www.mdpi.com/journal/ijms Article
Efficient Synthesis of a Maghemite/Gold Hybrid Nanoparticle System as a Magnetic Carrier for the Transport of Platinum-Based Metallotherapeutics Pavel Štarha 1, David Smola 1, Jiří Tuček 2 and Zdeněk Trávníček 1,* 1
2
Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacký University, 17. listopadu 12, Olomouc CZ-77146, Czech Republic; E-Mails:
[email protected] (P.S.);
[email protected] (D.S.) Regional Centre of Advanced Technologies and Materials, Department of Experimental Physics, Faculty of Science, Palacký University, 17. listopadu 12, Olomouc CZ-77146, Czech Republic; E-Mail:
[email protected]
* Author to whom correspondence should be addressed; E-Mail:
[email protected]; Tel.: +420-585-634-352; Fax: +420-585-634-954. Academic Editor: Bing Yan Received: 11 December 2014 / Accepted: 13 January 2015 / Published: 16 January 2015
Abstract: The preparation and thorough characterization of a hybrid magnetic carrier system for the possible transport of activated platinum-based anticancer drugs, as demonstrated for cisplatin (cis-[Pt(NH3)2Cl2], CDDP), are described. The final functionalized mag/Au–LA–CDDP* system consists of maghemite/gold nanoparticles (mag/Au) coated by lipoic acid (HLA; LA stands for deprotonated form of lipoic acid) and functionalized by activated cisplatin in the form of cis-[Pt(NH3)2(H2O)2]2+ (CDDP*). The relevant techniques (XPS, EDS, ICP-MS) proved the incorporation of the platinum-containing species on the surface of the studied hybrid system. HRTEM, TEM and SEM images showed the nanoparticles as spherical with an average size of 12 nm, while their superparamagnetic feature was proven by 57Fe Mössbauer spectroscopy. In the case of mag/Au, mag/Au–HLA and mag/Au–LA–CDDP*, weaker magnetic interactions among the Fe3+ centers of maghemite, as compared to maghemite nanoparticles (mag), were detected, which can be associated with the non-covalent coating of the maghemite surface by gold. The pH and time-dependent stability of the mag/Au–LA–CDDP* system in different media, represented by acetate (pH 5.0), phosphate (pH 7.0) and carbonate (pH 9.0) buffers and connected with
Int. J. Mol. Sci. 2015, 16
2035
the release of the platinum-containing species, showed the ability of CDDP* to be released from the functionalized nanosystem. Keywords: nanoparticles; maghemite; magnetic; cisplatin; drug delivery
1. Introduction Targeted drug delivery involving nanoparticles (e.g., iron oxides or gold) is currently generally accepted as a suitable and prospective alternative pathway in conventional anticancer chemotherapy using platinum-based metallotherapeutics [1–3]. The basic advantage is that the targeted delivery could reduce the amount of drug administrated to the body, and as a consequence of this, the negative side effects (e.g., nephrotoxicity, neurotoxicity or myelosuppression) can be partially eliminated [4]. Additionally to tumor therapy, the magnetic iron oxide-based nanoparticles also offer diagnostic (e.g., MRI) potential [5–7]. On the other hand, the gold nanoparticles are known to show a surface plasmon resonance usable in photothermal therapy or diagnostics [8–11]. Hybrid systems, consisting of gold and iron oxide nanoparticles, offer a combination of these properties within one nanosystem [12], which has been of a great interest in the field of drug delivery agents [13]. In particular, the coating of the magnetic iron oxide nanoparticles with the biocompatible precious metal, gold, prevents not only the chemical and enzymatic degradation of the iron oxide core, but also seems to be suitable for further binding of various types of compounds (especially sulfur-containing ones) and functionalization [14,15]. There are plenty of works, as shown below, reporting gold/iron oxide-based nanoparticles available for further derivatization and/or functionalization of gold. Surprisingly, only one of them describes functionalization with platinum-based species, namely cisplatin [13]. In this case, the activated cisplatin was bound to the thiolated polyethylene glycol (PEG) linker (the thiol part of the linker was lipoic acid (HLA), also used in this work). In vitro cytotoxicity of these nanoparticles was found to be more than 100-times higher against both the cisplatin-sensitive and resistant human ovarian carcinoma cancer cell lines (A2780, A2780/cp70) as compared to cisplatin. It can be highlighted that the herein reported mag/Au–LA–CDDP* nanoparticles represent a synthetically more easily and quickly obtainable system (there is no need for the preparation of the thiolated PEG linker), which is (very important from the pharmacological point of view) able to bind even more platinum(II) species (as discussed below). Regarding other maghemite/gold nanoparticle systems, those layered by differently-thiolated PEG were reported as a promising MRI contrast diagnostic agent for malignant tumors [16]. Similar systems, but with anti-podoplanin antibody bound to PEG on the surface of the studied nanoparticles, were shown as suitable MR imaging agents in evaluating lymphangiogenesis in breast cancer cells in vivo [17]. Fan et al. prepared gold-coated maghemite core-shell nanoparticles that highly specifically targeted SK-BR-3 human breast cancer cells through the S6 aptamer bound on the surface, which exhibit SK-BR-3 cells targeting themselves [18]. Modification of the surface by fibrinogen resulted in interesting cell targeting properties, representing another possible application of gold-coated iron oxide core-shell nanoparticles [19]. Similar nanoparticles functionalized with different agents (PEG, glucose or fluorochrome) did not show any in vitro cytotoxicity against the cervix carcinoma HeLa cancer cell line [20]. Interesting possibilities within this field of study are hybrid nanosystems with an iron oxide
Int. J. Mol. Sci. 2015, 16
2036
core and a gold shell separated by an intermediate layer (e.g., polyethyleneimine) [21] or dumbbell-like nanocomposites consisting of a gold nanoparticle bound together with an iron oxide one, where the gold part can be functionalized by the sulfur-containing molecules or linkers [22]. In this study, we deal with the preparation and characterization of potential theranostic maghemite/gold nanoparticles layered with a simple sulfur-containing carboxylic acid (lipoic acid) and directly functionalized by the platinum metallotherapeutic-based species, such as the herein used activated cisplatin. The reported data represent an optimized and easily realized pathway leading to gaining of the discussed biocompatible nanosystem, which may serve as a necessary base for further biological studies. 2. Results and Discussion 2.1. Preparation and Functionalization The synthesis of the magnetite nanoparticles, maghemite nanoparticles, iron oxide/gold nanoparticles system, as well as the iron-oxide/gold nanoparticle hybrid system layered by organic sulfur-containing carboxylic acid is well-established in the literature [13,23–25]. In the case of this work, the freshly prepared magnetite nanoparticles were gently oxidized by diluted nitric acid to produce the maghemite nanoparticles (mag), which are macroscopically pronounced by the color change from black to brown. The coating of maghemite by gold was carried out by the slow addition of a 1% HAuCl4 water solution, again connected with a color change of the nanoparticles (mag/Au) to dark purple (further addition of the HAuCl4 solution led to the formation of gold nanoparticles, observable as a purple/pink color above the mag/Au nanoparticles). The prepared mag/Au nanoparticles kept their magnetic properties (being attracted by the external magnet) and are stable in the tetramethylammonium hydroxide (TMAOH) solution for at least two weeks (there are no color changes of the nanoparticles themselves nor the solution above the nanoparticles). As is frequently described in the literature [13,25], gold nanoparticles can easily interact with sulfur-containing molecules. We let a slight excess of lipoic acid (HLA) react with mag/Au, resulting in mag/Au–HLA nanoparticles. Within the final step of the synthetic procedure, including the crucial functionalization, the mag/Au–HLA nanoparticles interacted with the activated platinum-based drug, cisplatin, i.e., with cis-[Pt(NH3)2(H2O)2]2+ species (CDDP*) prepared by the reaction of cisplatin (cis-[Pt(NH3)2Cl2], CDDP) with two molar equivalents of AgNO3 in the dark [26], to give the final product of the mag/Au–LA–CDDP* system (Figure 1). Regarding the yield of the synthesis, we got 155 mg of mag/Au–LA–CDDP*, but this cannot be expressed as a percent, because every synthetic intermediate was partly used for synthesis and partly stored or used for characterization. Regarding the lipoic acid used, it has to be stated that it was previously used, for example, for the modification of the gold seeds of the iron oxide magnetic core, where lipoic acid was subsequently modified by the interaction with iminodiacetic acid, which finally bound copper(II) ions [27], or lipoic acid was used as the thiol part of the thiolated polyethylene glycol (PEG) linker layering the gold-coated maghemite nanoparticles and binding the activated cisplatin as functionalizing platinum-based species [13]. In comparison to the latter work, here, we applied a different approach resulting in the magnetic gold-coated maghemite nanoparticles layered by lipoic acid itself, with its carboxylic group ready for the covalent binding of the biologically-active platinum-based species, such as activated
Int. J. Mol. Sci. 2015, 16
2037
cisplatin. Interestingly, the use of the lipoic acid itself (this work), instead of the LA–PEG thiolated polyethylene glycol linker [13], not only simplified and accelerated the synthesis, but also provided the nanosystems with a higher Pt content (as discussed below).
Figure 1. The reaction pathway leading to the preparation of the mag/Au–LA–CDDP* nanoparticle system. 2.2. Characterization 2.2.1. HRTEM, TEM and SEM Microscopy The prepared mag/Au–LA–CDDP* nanosystems are of a uniform spherical shape (average size of 12.2 ± 1.9 nm), as proven by the HRTEM, TEM and SEM techniques applied to characterize their morphology and dispersibility (Figure 2). Any impurity (e.g., crystalline lipoic acid or CDDP*) was not microscopically detected within the studied product. The intermediates, mag, mag/Au and mag/Au–HLA, showed similar morphological properties compared with the discussed mag/Au–LA–CDDP*. The dispersibility of all of the intermediates and the final functionalized nanoparticles was comparable, as well, with no agglomeration observed after the individual synthetic steps.
Figure 2. HRTEM (left; 10 nm size bar), TEM (top right; 50 nm size bar) and SEM (bottom right; 500 nm size bar) images, as obtained for the studied functionalized mag/Au–LA–CDDP* nanoparticles.
Int. J. Mol. Sci. 2015, 16
2038
Although we applied the synthetic procedure reported for the preparation of the gold-coated maghemite core-shell nanoparticles, we did not detect gold as a compact layer coating the surface of the maghemite, but more or less randomly distributed on the magnetic iron oxide support (Figure 3).
Figure 3. High-angle annular dark-field detector (HAADF)-STEM images of mag/Au nanoparticles (a–d; 9 nm size bars) showing iron (b), gold (c) and both iron and gold present in one nanoparticle (d). 2.2.2. XPS Spectroscopy The XPS spectra were recorded for the final mag/Au–LA–CDDP* nanosystem and its synthetic precursor, mag/Au–HLA (Figure 4). As can be anticipated, the XPS spectra of both systems were similar (the data given below belong to mag/Au–LA–CDDP*). The photoelectron peaks assignable to the maghemite part of the composite were at 711.0 and 724.2 eV (Fe2p3/2 and Fe2p1/2) and 530.2 eV (O1s) [24,28]. The spectra also contained several gold photoelectron peaks, namely Au4p3/2 (353.8 eV), Au4p5/2 (333.4 eV), Au4f5/2 (87.0 eV) and Au4f7/2 (83.8 eV) [29–31]. Similar data were also reported for the gold nanoparticles [32] or gold seeds on the surface of the magnetite nanoparticles [27], which were covalently layered with the herein used lipoic acid. As for the lipoic acid itself, the peaks of C1s at 284.6 eV (aliphatic chain carbon atoms) and 288.4 eV (carboxylic group carbon atoms), as well as the S2p peak (163.0 eV) have to be considered as evidence of the lipoic acid incorporation into the studied nanoparticles [27,33]. The differences between the XPS spectra of mag/Au–HLA and mag/Au–LA–CDDP*, clearly showing for the CDDP* the presence within mag/Au–LA–CDDP*, can be found at the 399.8 eV region
Int. J. Mol. Sci. 2015, 16
2039
characteristic for N1s peaks and mainly in the regions connected with platinum, whose photoelectron peaks were detected at binding energies of 315.8 (Pt4d5/2), 75.7 (Pt4f5/2) and 72.3 (Pt4f7/2) eV (Figure 4) [24,34,35]. The position of the Pt4f photoelectron peaks within the XPS spectrum of mag/Au–LA–CDDP* indicated that platinum keeps the +II oxidation state, because it falls between Pt(0) (74.5 and 71.0 eV) and Pt(IV) (78.0 and 74.5 eV) [34]. Further, the position of the Pt4f and N1s peaks correlated well with those reported for cisplatin, cis-[Pt(NH3)2(H2O)Cl]+ or activated cisplatin bound to mercaptosuccinic acid on the surface of the functionalized hydroxyapatite nanoparticles [34,35].
Figure 4. The results of the XPS spectroscopy of the mag/Au–LA–CDDP* nanosystems (red lines) and their comparison with mag/Au–HLA (black lines), given for the 0–750 eV region (top) with the details of the Pt4f/Au4f region at 67.5–90 eV (bottom left) and C1s region in the 280–295 eV range (bottom right). 2.2.3. EDS Spectroscopy The EDS spectra were recorded for the products of all of the synthetic steps (mag, mag/Au, mag/Au–HLA, mag/Au–LA–CDDP*) (Figure 5). The discussed XPS results correlated very well with those obtained by the EDS spectroscopy and nicely reflected the synthetic procedure. Particularly, coating of mag nanoparticles, whose EDS contained only the Fe (0.70, 6.43 and 7.08 keV) and O (0.53 keV) peaks, with gold resulted in the Au peaks detected in the EDS spectrum of mag/Au nanocomposites at 1.67, 2.15 and 9.75 keV. Interaction of the mag/Au system with lipoic acid was shown as a new peak at 0.31 keV, a characteristic region for carbon. All of the mentioned peaks were also found in the EDS spectrum of the functionalized mag/Au–LA–CDDP* nanoparticles together with one new peak at 9.49 keV, unambiguously assignable to platinum. The other platinum peaks were not detected in the spectrum of the functionalized nanoparticles, because they are overlapped by the mentioned Au peaks at 1.67 and 2.15 keV.
Int. J. Mol. Sci. 2015, 16
2040
Figure 5. EDS spectra of the final functionalized mag/Au–LA–CDDP* nanoparticles and their synthetic intermediates, mag/Au–HLA, mag/Au and mag nanoparticles, with the assigned peaks of carbon, oxygen, iron and gold/platinum. Insets: detail of the gold/platinum region (depicted in the 9.25–10.00 keV range). 2.2.4. Quantification by ICP-MS The platinum and gold contents within the mag/Au–LA–CDDP* nanoparticles were determined by ICP-MS. The results showed that the mentioned functionalized nanoparticles contain platinum and gold in a molar ratio of 1:2.6. Related to the weight of mag/Au–LA–CDDP*, the weight content of platinum is 4.0% (10.6% for gold). Expressed in mol per 1 g of Au, the Pt content within the studied mag/Au–LA–CDDP* nanoparticles equals 1.97 × 10−3 mol per 1 g of Au, which is a ca. 2.5-times higher Pt-to-Au ratio compared with the similar previously reported maghemite/gold nanoparticles layered by PEG thiolated by lipoic acid and functionalized with the activated cisplatin (7.9 × 10−4 mol per 1 g of Au) [13]. 2.2.5. Simultaneous TG/DTA Thermal Analyses A comparison of the results of the simultaneous TG/DTA thermal analyses showed the considerable differences between mag/Au–LA–CDDP* nanoparticles and their synthetic intermediates mag/Au–HLA and mag/Au (Figure 6). As for mag/Au, we detected the usual weight loss (30–145 °C) connected with the loss of physically adsorbed water [28,36], which fluently continued up to 508 °C by the next weight loss, which is, as supported by a massive exothermic effect (maximum at 243 °C), most probably connected with an oxidation of the remaining citrate on the surface of the mentioned
Int. J. Mol. Sci. 2015, 16
2041
nanoparticles (the presence of the citrate within mag/Au can be supported by the EDS results, which showed the Na peak at 1.05 keV; Figure 5) [37]. In the case of mag/Au–HLA, there are two steps following the above-mentioned initial water loss, detected at 101–413 °C and above 496 °C, connected with the decomposition of the organic layer (lipoic acid) of these nanoparticles. However, a sharp exothermic peak (maximum at 597 °C) is not connected with this process, but it has to be assigned to the γ-Fe2O3 to α-Fe2O3 conversion [36].
Figure 6. Thermogravimetry (TG) and differential thermal analysis (DTA) results obtained for mag/Au–LA–CDDP* (red curves), mag/Au–HLA (dashed green curves) and mag/Au (dotted purple curves). The final functionalized mag/Au–LA–CDDP* nanoparticles showed a similar TG curve, but with a marked weight loss at 367–465 °C accompanied by an exothermic effect with a maximum at 416 °C, which most probably relates to the platinum-containing species decomposition. Again, the exothermic effect of the γ-Fe2O3 to α-Fe2O3 conversion was found on the DTA curve at 609 °C. Another important finding comes from the sharp exothermic effect detected on the DTA curve of mag/Au–LA–CDDP* at 263 °C, which can be assigned to the cis-to-trans rearrangement of the platinum-containing species, which showed that a Pt(NH3)2 motif binds to one LA through one Pt–O band; in other words, only one H2O ligand of the interacting cis-[Pt(NH3)2(H2O)2]2+ species was replaced by the O-donor, lipoic acid. 2.2.6. IR Spectroscopy The IR spectrum of mag/Au–HLA is depicted, together with the spectra of mag nanoparticles and free HLA, in Figure 7. Both the spectra of mag/Au–HLA and mag nanoparticles contained peak at ca. 540 cm−1, clearly assignable to the ν (Fe–O) vibration of maghemite [24,38]. In the spectrum of mag/Au–HLA, we also detected a series of peaks revealed at 1604 and 1387 cm−1, which may be associated with the C=O and C–O stretching vibrations, respectively, and the peaks at 2931 and 2974 cm−1 corresponding to the stretching vibrations of aliphatic C–H bonds. The peaks observed at ca. 920 and 629 cm−1 could be connected with the deformation O–H vibrations and stretching C–S vibrations, respectively. These results can support the presence of HLA in the discussed mag/Au–HLA nanoparticles, since their positions correlate well with the positions of the peaks in the spectrum of free lipoic acid [39,40].
Int. J. Mol. Sci. 2015, 16
2042
Figure 7. The IR spectra of the mag/Au–HLA nanoparticles (green; the detail of the aliphatic C–H vibration region is given as an inset) and its comparison with the mag nanoparticles (mag; black) and free lipoic acid (HLA; red). 2.2.7. 57Fe Mössbauer Spectroscopy In order to gain deeper insight into the structural and magnetic properties of the studied samples, Fe Mössbauer spectroscopy was employed. In Mössbauer spectroscopy, the Fe nucleus acts as a probe monitoring the physicochemical characteristics of the local surroundings through the hyperfine interactions of an electromagnetic nature. The measured 57Fe Mössbauer spectra of the studied samples are shown in Figure 8, and the values of the Mössbauer hyperfine parameters, derived from the fitting of the respective Mössbauer spectrum, are listed in Table 1. 57
Figure 8. 57Fe Mössbauer spectrum of the mag, mag/Au, mag/Au–HLA and mag/Au–LA–CDDP* nanoparticles, measured at 300 K and without an external magnetic field.
Int. J. Mol. Sci. 2015, 16
2043
Table 1. Values of the Mössbauer hyperfine parameters for maghemite (mag), maghemite/gold (mag/Au), maghemite/gold hybrid nanoparticle system layered by lipoic acid (mag/Au–HLA) and maghemite/gold nanoparticles layered by lipoic acid and functionalized by activated cisplatin (mag/Au–LA–CDDP*) derived from the fitting procedure of the respective 300-K Mössbauer spectrum. δ is the isomer shift; ΔEQ is the quadrupole splitting; Bhf is the hyperfine magnetic field; RA is the relative spectral area of individual spectral components. Sample mag
mag/Au
mag/Au–HLA
mag/Au–LA–CDDP*
Component Sextet Sextet Singlet Sextet Sextet Sextet Singlet Sextet Sextet Sextet Singlet Sextet Sextet Sextet Singlet
δ ± 0.01 (mm/s) 0.26 0.35 0.31 0.27 0.37 0.31 0.31 0.25 0.37 0.31 0.31 0.26 0.36 0.32 0.31
ΔEQ ± 0.01 (mm/s) 0.01 0.00 0.00 0.01 0.03 0.01 0.01 0.04 0.00 0.00 0.04 -
Bhf ± 0.3 (T) 48.7 48.2 48.7 48.3 44.5 48.6 48.2 44.4 48.8 48.3 44.3 -
RA ± 1 (%) 13 25 62 11 20 9 60 11 20 9 60 11 20 9 60
Assignment γ-Fe2O3–T-sites γ-Fe2O3–O-sites Fe3+ relaxation γ-Fe2O3–T-sites γ-Fe2O3–O-sites γ-Fe2O3/Au surface Fe3+ relaxation γ-Fe2O3–T-sites γ-Fe2O3–O-sites γ-Fe2O3/Au surface Fe3+ relaxation γ-Fe2O3–T-sites γ-Fe2O3–O-sites γ-Fe2O3/Au surface Fe3+ relaxation
Int. J. Mol. Sci. 2015, 16
2044
The slightly asymmetric profile of the room-temperature 57Fe Mössbauer spectrum of the maghemite (mag) nanoparticles implies the presence of more than one sextet component with similar values of the Mössbauer hyperfine parameters. Thus, to correctly fit the spectrum, the mag nanoparticles were placed in an external magnetic field applied in a parallel direction with respect to the propagation of γ-rays. The in-field room-temperature Mössbauer spectrum of the mag nanoparticles clearly showed two resolved sextets (not shown) with values of the Mössbauer hyperfine parameters typical of γ-Fe2O3; one sextet corresponding to the tetrahedral Fe3+ cation sites (i.e., T-sites) of the γ-Fe2O3 spinel structure and the other sextet reflecting octahedral Fe3+ cation sites (i.e., O-sites) of the γ-Fe2O3 spinel structure [41]. It is known that in an external magnetic field, the T- and O-related sextets are well separable, due to the addition and subtraction of the external magnetic field for the γ-Fe2O3 T-sites and O-sites, respectively [41]. In addition, the spectral area of the O-sextet to T-sextet (~1.90) is far from the ideal ratio (~1.66), indicating that γ-Fe2O3 is not purely stoichiometric. For γ-Fe2O3 with sizes of less than 20 nm, the non-stoichiometry results from either distribution of vacancies on the T-sites or the presence of some non-oxidized Fe2+ ions [41]. However, in the case of mag, both factors may play an equal role, as their signatures cannot be clearly identified in the zero-field and in-field room-temperature Mössbauer spectra. Nevertheless, the knowledge of the values of the Mössbauer hyperfine parameters of the two sextets and their spectral areas, derived from the in-field Mössbauer spectrum, assisted in helping to fit the zero-field Mössbauer spectrum of the mag nanoparticles. Besides the two sextets ascribed to cation T-sites and O-sites in the γ-Fe2O3 spinel structure, an extra component, a broad singlet, emerges (see Figure 8). It has features of the Fe3+ relaxation component and indicates an onset of the passage of nanoparticle superspin to the superparamagnetic regime at the timescale of the Mössbauer technique [41]. On the other hand, the clearly evolved sextet components imply that superspins of a fraction of γ-Fe2O3 nanoparticles (i.e., the largest one) still remain in the magnetically blocked state at 300 K with respect to the characteristic measuring time of the Mössbauer technique. The coexistence of sextets and singlet reflects the particle size distribution in the system, as the passage to the superparamagnetic state is, at a given temperature and measuring time, governed by the nanoparticle size [41]. As the spectral area of a singlet is dominant, most of the magnetic nanoparticles in the assembly likely behave in the superparamagnetic manner. The room-temperature Mössbauer spectrum of the maghemite/gold nanoparticles (mag/Au) shows four spectral components (see Figure 8), i.e., three sextets and one singlet. The sextets with the higher values of the hyperfine magnetic fields can be again ascribed to the cation T-sites and O-sites in the γ-Fe2O3 spinel structure, and the singlet corresponds to those γ-Fe2O3 nanoparticles entering the superparamagnetic regime. In addition, one more sextet is clearly identified; its values of the Mössbauer hyperfine parameters reflect the local surrounding of Fe3+ ions, however, affected by the presence of non-magnetic and/or diamagnetic atoms. As pure gold is weakly diamagnetic, it acts as a shield, decreasing the magnetic field in which it is placed. Since a much lower value of the hyperfine magnetic field is observed for the third sextet (see Table 1), it can be explained by partially covering the surface of γ-Fe2O3 nanoparticles with gold, inducing further weakening of the magnetic interactions among the Fe3+ magnetic moments in the surface layers of γ-Fe2O3 nanoparticles. The interaction of gold with the surface of γ-Fe2O3 nanoparticles is purely non-covalent, as no change in the isomer shift value of the interaction sextet is observed.
Int. J. Mol. Sci. 2015, 16
2045
The room-temperature Mössbauer spectrum of the mag/Au–HLA and final functionalized mag/Au–LA–CDDP* nanoparticles resembles that of the mag/Au sample (see Figure 8 and Table 1), implying that both lipoic acid and activated cisplatin (CDDP*) do not influence the surface state of the γ-Fe2O3 nanoparticles. The spectra show only a smaller signal-to-noise ratio, most probably due to firmer embedding of γ-Fe2O3 nanoparticles in the organic medium. 2.2.8. Stability Studies The stability of the mag/Au–LA–CDDP* nanoparticles was investigated in three buffers at different pH values and time points (Table 2). Table 2. Pt content (µg) at different times and pH values for the functionalized mag/Au–LA–CDDP* nanoparticles dispersed in phosphate, carbonate and acetate buffers, incubated at 37 °C for 24, 48 and 72 h and checked by ICP-MS from the supernatant collected over the remnant nanoparticles attracted to the magnet. Solvent (pH) Acetate buffer (5.0) Phosphate buffer (7.0) Carbonate buffer (9.0)
Incubation Time 24 h 48 h 72 h 2.5 ± 0.1 1.7 ± 0.3 1.6 ± 0.2 5.9 ± 0.2 5.2 ± 0.2 7.5 ± 0.1 27.5 ± 0.2 31.5 ± 0.2 33.7 ± 0.3
It has been found that the stability of the prepared nanocomposites is quite pH dependent, because the content of the released platinum-containing species increased with the pH value of the appropriate buffer. In the case of the acidic acetate buffer, the release of the platinum-containing species is negligible, while at pH 7 (phosphate buffer), it reaches the values acceptable for further biological studies (e.g., in vitro cytotoxicity) [24]. However, even a several times higher release of the platinum-containing species was found with the carbonate buffer under basic pH values. 3. Experimental Section 3.1. Materials and Methods The starting materials, FeCl3·6H2O, FeCl2·4H2O, lipoic acid (HLA), 25% NH4OH, tetramethylammonium hydroxide (TMAOH), sodium citrate tribasic dihydrate, HAuCl4, cisplatin, AgNO3 and solvents, were supplied by Sigma-Aldrich Co. (Prague, Czech Republic) and Acros Organics Co. (Pardubice, Czech Republic) and used as received. Simultaneous thermogravimetric (TG) and differential thermal (DTA) analyses were performed using an Exstar TG/DTA 6200 thermal analyzer (Seiko Instruments Inc., Chiba, Japan); dynamic air atmosphere (100 mL·min−1), 25–700 °C (5.0 °C·min−1). Scanning electron microscopy (SEM) was performed by a Hitachi 6600 FEG microscope (5 keV accelerating voltage), together with energy-dispersive X-ray spectroscopy (EDS) (Hitachi, Tokyo, Japan). HRTEM microscopic images were recorded by an FEI Titan 60–300 kV (0.06 nm point resolution) with an X-FEG emission gun, a Cs image corrector and an STEM high-angle annular dark-field detector (HAADF) (FEI, Hillsboro, OR, USA). The elemental mapping was carried out by STEM-energy dispersive X-ray spectroscopy
Int. J. Mol. Sci. 2015, 16
2046
(EDS). Transmission electron microscopy (TEM) images were taken on a JEOL 2010 microscope (200 kV, 1.9 Å point-to-point resolutions; JEOL, Peabody, MA, USA). A drop of high-purity water with the ultrasonically dispersed samples was placed onto a holey-carbon film supported by a copper-mesh TEM grid and dried in air at room temperature. The diameter of the nanoparticles was measured by ImageJ software (ImageJ, Bethesda, MD, USA). X-ray photoelectron spectroscopy (XPS) results were obtained using a PHI 5000 VersaProbe II device (Physical Electronics, Chanhassen, MN, USA). Infrared spectra (400–4000 cm−1 region) were recorded by the ATR technique on a Nexus 670 FT-IR (Thermo Nicolet, Waltham, MA, USA). 57Fe Mössbauer spectra were recorded in a transmission geometry employing a Mössbauer spectrometer operating in a constant velocity regime and equipped with a 50 mCi 57Co(Rh) source. For in-field 57Fe Mössbauer measurement, the maghemite nanoparticles (mag) were placed inside the chamber of the cryomagnetic system (Oxford Instruments, Abingdon, UK); an external magnetic field of 5 T was applied in parallel direction with respect to the propagation of γ-rays. For fitting the Mössbauer spectra, the MossWinn software program was used [42]; prior to fitting, the signal-to-noise ratio was adjusted by the filtering procedures built in the MossWinn software program and by the statistically-based approach developed by Prochazka et al. [43]. The isomer shift values are referred to α-Fe at room temperature. 3.2. The Platinum and Gold Contents The mag/Au–LA–CDDP* nanosystems (2.0 mg) were dissolved in 0.1 mL of aqua regia and then 5000× diluted by distilled water. The Pt and Au contents were determined by inductively-coupled plasma mass spectrometry (ICP-MS) with the obtained values corrected for the adsorption effects (ICP-MS spectrometer 7700×, Agilent, Santa Clara, CA, USA). 3.3. Stability Studies The functionalized mag/Au–LA–CDDP* nanoparticles (2.0 mg) were dispersed in 5 mL of phosphate (pH 7.0), carbonate (pH 9.0) and acetate (pH 5.0) buffers and incubated at 37 °C for 24, 48 and 72 h. The content of platinum, showing the ability of the system to release the drug under different conditions, was checked by ICP-MS from the supernatant collected over the remnant nanoparticles attracted to the magnet. 3.4. Synthetic Procedures 3.4.1. Maghemite Nanoparticles (mag) The water (50 mL) solution of the mixture of FeCl3·6H2O (2.16 g; 4.0 mmol) and FeCl2·4H2O (0.80 g; 2.0 mmol) was stirred under nitrogen gas until the temperature reached 80 °C. After that, 5 mL of 25% NH4OH were added to the solution, and the resulting black mixture was intensively stirred for the next 30 min. Thereafter, the obtained magnetite (Fe3O4) nanoparticles were attracted by the external magnet, washed with water (3 × 10 mL) and 0.01 M HNO3 (3 × 10 mL) and stirred in 25 mL of 0.01 M HNO3 at 90 °C for 1 h. The prepared maghemite (γ-Fe2O3) nanoparticles were magnetically isolated and washed with an adequate amount of deionized water (3 × 10 mL). Part of the product was washed with acetone (3 × 10 mL) and dried under N2 atmosphere to be stored in the solid state (in the
Int. J. Mol. Sci. 2015, 16
2047
fridge); another part was dispersed in 0.1 M TMAOH to be stored as a suspension; and the rest was used for the upcoming synthesis. 3.4.2. Maghemite/Gold Nanoparticle System (mag/Au) Sodium citrate (200 mL of 0.1 M water solution) was added to the water suspension of maghemite nanoparticles (200 mg), and the suspension was stirred at 90 °C for 10 min. One-milliliter aliquots of HAuCl4 (1% water solution; 30 mL) were added to this suspension every 5 min with continuous stirring. The prepared mag/Au nanoparticles were, as in the case of maghemite nanoparticles, isolated by the external magnet, washed with deionized water (3 × 10 mL) and either washed with 3 × 10 mL of acetone and dried under a N2 atmosphere to be stored in the solid state, dispersed in 0.1 M TMAOH to be stored as a suspension or used for the synthesis. 3.4.3. Maghemite/Gold Nanoparticles Layered by Lipoic Acid (mag/Au–HLA) The mag/Au nanoparticles (20 mL of the TMAOH suspension) were attracted by the external magnet, the TMAOH poured out and distilled water (20 mL) poured in. The lipoic acid (HLA; 200 mg) was poured in about a 1.25 molar excess (related to gold) into the suspension, which was intensively stirred overnight. The obtained mag/Au–HLA nanoparticle suspension was split into two portions, i.e., one was isolated as the solid state product and the second one as the TMAOH suspension, as described above for both the maghemite and mag/Au nanoparticles. 3.4.4. Maghemite/Gold Nanoparticles Layered by Lipoic Acid and Functionalized by Activated Cisplatin (mag/Au–LA–CDDP*) The final step of the synthesis involves the reaction of the mag/Au–HLA nanoparticles (those from the TMAOH suspension with TMAOH replaced by distilled water) with the cis-[Pt(NH3)2(H2O)2]2+ species (CDDP*), which formed from cisplatin (cis-[Pt(NH3)2Cl2], CDDP; 132 mg to keep the 1:1 molar ratio with HLA) by its activation (24 h, in the dark, water/DMF 1:1 v/v) with two molar equivalents of AgNO3 (149 mg). The suspension of mag/Au–HLA with the activated cisplatin (CDDP*) was stirred for 24 h followed by magnetic isolation, washing (3 × 5 mL of DMF, 3 × 5 mL of water and 3 × 5 mL of acetone) and drying under a nitrogen gas flow. The obtained product, mag/Au–LA–CDDP* nanoparticles, was stored in the solid state in the fridge. 4. Conclusions Efficient synthesis of a maghemite/gold hybrid nanoparticle system, as a possible magnetic carrier for the transport of platinum-based metallotherapeutics, is described. Easily obtainable superparamagnetic maghemite/gold-lipoic acid (mag/Au–HLA) nanoparticles, consisting of the maghemite/gold particles layered by a sulfur-containing carboxylic acid (lipoic acid), were found (by HRTEM and TEM) to be of a spherical shape and to have an average size of about 12 nm in diameter. This system is suitable for the binding of the platinum-containing species, particularly activated cisplatin of the composition cis-[Pt(NH3)2(H2O)2]2+. The arrangement of gold on the surface of the maghemite
Int. J. Mol. Sci. 2015, 16
2048
nanoparticles can be indirectly supported by the weakening of the magnetic interactions among the Fe3+ magnetic moments, which is probably caused by non-covalent interactions of the maghemite core with the gold situated on its surface. The organic layer, represented by lipoic acid, as well as the incorporation of the functionalizing platinum-containing species did not cause an additional alteration of the magnetic interactions. The platinum-containing species binds monofunctionally (i.e., through one Pt–O bond) through the carboxylic groups of mag/Au–HLA within the final mag/Au–LA–CDDP* nanosystem. The incorporation of the activated cisplatin within the final functionalized mag/Au–LA–CDDP* nanoparticles was proven by XPS and EDS spectroscopy and quantified by ICP-MS as being 4.0% in terms of the platinum content related to the total weight of mag/Au–LA–CDDP*. The stability of the mag/Au–LA–CDDP* nanoparticles in solution was found to be more pH dependent than time dependent, i.e., the highest platinum content released from the nanoparticles was determined under basic conditions (pH 9), but acceptable platinum content was detected also in the phosphate buffer of the physiologically more relevant pH (7.0). Acknowledgments The authors gratefully thank the Czech Science Foundation (GAČRP207/11/0841), the Ministry of Education, Youth and Sports of the Czech Republic (Project LO1305) and Palacký University in Olomouc (PrF_2014_009). The authors would like to thank Klára Čépe, Jana Stráská and Ondřej Tomanec for performing the HRTEM, TEM and SEM/EDS experiments, Jana Gáliková for recording the FTIR spectra, David Milde for performing ICP-MS and Martin Petr for carrying out the XPS measurements. Author Contributions Conceived of and designed the experiments: Pavel Štarha, Jiří Tuček, Zdeněk Trávníček; Performed the experiments: Pavel Štarha, David Smola, Jiří Tuček and Zdeněk Trávníček; Analyzed the data: Pavel Štarha, David Smola, Jiří Tuček and Zdeněk Trávníček; Wrote the paper: Pavel Štarha, David Smola, Jiří Tuček and Zdeněk Trávníček. Conflicts of Interest The authors declare no conflict of interest. References 1. 2. 3. 4.
Oberoi, H.S.; Nukolova, N.V.; Kabanov, A.V.; Bronich, T.K. Nanocarriers for delivery of platinum anticancer drugs. Adv. Drug Deliv. Rev. 2013, 35, 1667–1685. Butler. J.S.; Sadler, P.J. Targeted delivery of platinum-based anticancer complexes. Curr. Opin. Chem. Biol. 2013, 17, 175–188. Barry, N.P.E.; Sadler, P.J. Challenges for metals in medicine: How nanotechnology may help to shape the future. ACS Nano 2013, 7, 5654–5659. Kelland, L. The resurgence of platinum-based cancer chemotherapy. Nat. Rev. Cancer 2007, 7, 573–584.
Int. J. Mol. Sci. 2015, 16 5.
6. 7. 8. 9.
10. 11. 12. 13.
14.
15. 16.
17.
18.
19.
2049
Tassa, C.; Shaw, S.Y.; Weissleder, R. Dextran-coated iron oxide nanoparticles: A versatile platform for targeted molecular imaging, molecular diagnostics and therapy. Acc. Chem. Res. 2011, 44, 842–852. Xie, J.; Chen, K.; Huang, J.; Lee, S.; Wang, J.; Gao, J.; Li, X.; Chen, X. PET/NIRF/MRI triple functional iron oxide nanoparticles. Biomaterials 2010, 31, 3016–3022. Kievit, F.M.; Zhang, M. Surface engineering of iron oxide nanoparticles for targeted cancer therapy. Acc. Chem. Res. 2011, 44, 853–862. Huff, T.B.; Tong, L.; Zhao, Y.; Hansen, M.N.; Cheng, J.X.; Wei, A. Hyperthermic effects of gold nanorods on tumor cells. Nanomedicine 2007, 2, 125–132. Wagner, D.S.; Delk, N.A.; Lukianova-Hleb, E.Y.; Hafner, J.H.; Farach-Carson, M.C.; Lapotko, D.O. The in vivo performance of plasmonic nanobubbles as cell theranostic agents in zebrafish hosting prostate cancer xenografts. Biomaterials 2010, 31, 7567–7574. Bardhan, R.; Lal, S.; Joshi, A.; Halas, N.J. Theranostic nanoshells: From probe design to imaging and treatment of cancer. Acc. Chem. Res. 2011, 44, 936–946. Lim, J.; Majetich, S.A. Composite magnetic–plasmonic nanoparticles for biomedicine: Manipulation and imaging. Nano Today 2013, 8, 98–113. Huang, C.; Jiang, J.; Muangphat, C.; Sun, X.; Hao, Y. Trapping iron oxide into hollow gold nanoparticles. Nanoscale Res. Lett. 2011, 6, doi:10.1007/s11671-010-9792-x. Wagstaff, A.J.; Brown, S.D.; Holden, M.R.; Craig, G.E.; Plumb, J.A.; Brown, R.E.; Schreiter, N.; Chrzanowski, W.; Wheate, N.J. Cisplatin drug delivery using gold-coated iron oxide nanoparticles for enhanced tumour targeting with external magnetic. Inorg. Chim. Acta 2012, 393, 328–333. Wang, L.; Luo, J.; Fan, Q.; Suzuki, M.; Suzuki, I.S.; Engelhard, M.H.; Lin, Y.; Kim, N.; Wang, J.Q.; Zhong, C.J. Monodispersed core−shell Fe3O4@Au nanoparticles. J. Phys. Chem. B 2005, 109, 21593–21601. Zhang, L.; Zhu, X.; Jiao, D.; Sun, Y.; Sun, H. Efficient purification of His-tagged protein by superparamagnetic Fe3O4/Au–ANTA–Co2+ nanoparticles. Mater. Sci. Eng. C 2013, 33, 1989–1992. Kumagai, M.; Sarma, T.K.; Cabral, H.; Kaida, S.; Sekino, M.; Herlambang, N.; Osada, K.; Kano, M.R.; Nishiyama, N.; Kataoka, K. Enhanced in vivo magnetic resonance imaging of tumors by PEGylated iron-oxide–gold core–shell nanoparticles with prolonged blood circulation properties. Macromol. Rapid Commun. 2010, 31, 1521–1528. Yang, H.; Zou, L.G.; Zhang, S.; Gong, M.F.; Zhang, D.; Qi, Y.Y.; Zhou, S.W.; Diao, X.W. Feasibility of MR imaging in evaluating breast cancer lymphangiogenesis using Polyethylene glycol-GoldMag nanoparticles. Clin. Radiol. 2013, 68, 1233–1240. Fan, Z.; Shelton, M.; Singh, A.K.; Senapati, D.; Khan, S.A.; Ray, P.C. Multifunctional plasmonic shell–magnetic core nanoparticles for targeted diagnostics, isolation, and photothermal destruction of tumor cells. ACS Nano 2012, 6, 1065–1073. Krystofiak, E.S.; Mattson, E.C.; Voyles, P.M.; Hirschmugl, C.J.; Albrecht, R.M.; Gajdardziska-Josifovska, M.; Oliver, J.A. Multiple morphologies of gold–magnetite heterostructure nanoparticles are effectively functionalized with protein for cell targeting. Microsc. Microanal. 2013, 19, 821–834.
Int. J. Mol. Sci. 2015, 16
2050
20. Salado, J.; Insausti, M.; Lezama, L.; Gil de Muro, I.; Moros, M.; Pelaz, B.; Grazu, V.; de la Fuente, J.M.; Rojo, T. Functionalized Fe3O4@Au superparamagnetic nanoparticles: In vitro bioactivity. Nanotechnology 2012, 23, doi:10.1088/0957-4484/23/31/315102. 21. Hoskins, C.; Min, Y.; Gueorguieva, M.; McDougall, C.; Volovick, A.; Prentice, P.; Wang, Z.; Melzer, A.; Cuschieri, A.; Wang, L. Hybrid gold-iron oxide nanoparticles as a multifunctional platform for biomedical application. J. Nanobiotech. 2012, 10, doi:10.1186/1477-3155-10-27. 22. Xu, C.; Wang, B.; Sun, S. Dumbbell-like Au−Fe3O4 nanoparticles for target-specific platin delivery. J. Am. Chem. Soc. 2009, 131, 4216–4217. 23. Maity, D.; Zoppellaro, G.; Sedenkova, V.; Tucek, J.; Safarova, K.; Polakova, K.; Tomankova, K.; Diwoky, C.; Stollberger, R.; Machala, L.; et al. Surface design of core–shell superparamagnetic iron oxide nanoparticles drives record relaxivity values in functional MRI contrast agents. Chem. Commun. 2012, 48, 11398–11400. 24. Wang, J.; Wang, X.; Song, Y.; Wang, J.; Zhang, C.; Chang, C.; Yan, J.; Qiu, L.; Wu, M.; Guo, Z. A platinum anticancer theranostic agent with magnetic targeting potential derived from maghemite nanoparticles. Chem. Sci. 2013, 4, 2605–2612. 25. Lo, C.K.; Xiao, D.; Choi, M.M.F. Homocysteine-protected gold-coated magnetic nanoparticles: Synthesis and characterisation. J. Mater. Chem. 2007, 17, 2418–2427. 26. Guo, S.; Miao, L.; Wang, Y.; Huang, L. Unmodified drug used as a material to construct nanoparticles: Delivery of cisplatin for enhanced anti-cancer therapy. J. Control. Release 2014, 174, 137–142. 27. Reddy, A.N.; Anjaneyulu, K.; Basak, P.; Rao, N.M.; Manorama, S.V. A simple approach to the design and functionalization of Fe3O4–Au nanoparticles for biomedical applications. ChemPlusChem 2012, 77, 284–292. 28. Wang, L.; Luo, J.; Maye, M.M.; Fan, Q.; Rendeng, Q.; Engelhard, M.H.; Wang, C.; Lin, Y.; Zhong, C.J. Iron oxide–gold core–shell nanoparticles and thin film assembly. J. Mater. Chem. 2005, 15, 1821–1832. 29. Pérez-Mirabet, L.; Surinyach, S.; Ros, J.; Suades, J.; Yáñez, R. Gold and silver nanoparticles surface functionalized with rhenium carbonyl complexes. Mater. Chem. Phys. 2012, 137, 439–447. 30. Yu, Y.; Cao, C.Y.; Chen, Z.; Liu, H.; Li, P.; Dou, Z.F.; Song, W.G. Au nanoparticles embedded into the inner wall of TiO2 hollow spheres as a nanoreactor with superb thermal stability. Chem. Commun. 2013, 49, 3116–3118. 31. Szytula, A.; Fus, D.; Penc, B.; Jezierski, A. Electronic structure of RTX (R = Pr, Nd; T = Cu, Ag, Au; X = Ge, Sn) compounds. J. Alloy. Compd. 2001, 317, 340–346. 32. Anand, N.; Ramudu, P.; Reddy, K.H.R.; Seetha, K.; Rao, R.; Jagadeesh, B.; Babu, V.S.P.; Burri, D.R. Gold nanoparticles immobilized on lipoic acid functionalized SBA-15: Synthesis, characterization and catalytic applications. Appl. Catal. A 2013, 454, 119–126. 33. Roux, S.; Garcia, B.; Bridot, J.L.; Salome, M.; Marquette, C.; Lemelle, L.; Gillet, P.; Blum, L.; Perriat, P.; Tillement, O. Synthesis, characterization of dihydrolipoic acid capped gold nanoparticles, and functionalization by the electroluminescent luminol. Langmuir 2005, 21, 2526–2536. 34. Guven, A.; Rusakova, I.A.; Lewis, M.T.; Wilson, L.J. Cisplatin@US-tube carbon nanocapsules for enhanced chemotherapeutic delivery. Biomaterials 2012, 33, 1455–1461.
Int. J. Mol. Sci. 2015, 16
2051
35. Bach, L.G.; Islam, R.; Vo, T.S.; Kim, S.K.; Lim, K.T. Poly(allyl methacrylate) functionalized hydroxyapatite nanocrystals via the combination of surface-initiated RAFT polymerization and thiol–ene protocol: A potential anticancer drug nanocarrier. J. Colloid Interface Sci. 2013, 394, 132–140. 36. Ye, X.; Lin, D.; Jiao, Z.; Zhang, L. The thermal stability of nanocrystalline maghemite Fe2O3. J. Phys. D: Appl. Phys. 1998, 31, doi:10.1088/0022-3727/31/20/006. 37. Corr, S.J.; Raoof, M.; Mackeyev, Y.; Phounsavath, S.; Cheney, M.A.; Cisneros, B.T.; Shur, M.; Gozin, M.; McNally, P.J.; Wilson, L.J.; et al. Citrate-capped gold nanoparticle electrophoretic heat production in response to a time-varying radiofrequency electric-field. J. Phys. Chem. C 2012, 116, 24380–24389. 38. Dorniani, D.; Bin Hussein, M.Z.; Kura, A.U.; Fakurazi, S. Shaari, A.H.; Ahmad, Z. Preparation of Fe3O4 magnetic nanoparticles coated with gallic acid for drug delivery. Int. J. Nanomed. 2012, 7, 5745–5756. 39. Ikuta, N.; Tanaka, A.; Otsubo, A.; Ogawa, N.; Yamamoto, H.; Mizukami, T.; Arai, S.; Okuno, M.; Terao, K.; Matsugo, S. Spectroscopic studies of R(+)-α-lipoic acid–cyclodextrin complexes. Int. J. Mol. Sci. 2014, 15, 20469–20485. 40. Young, A.G.; Green, D.P.; McQuillan, A.J. IR spectroscopic studies of adsorption of dithiol-containing ligands on CdS Nanocrystal films in aqueous solutions. Langmuir 2007, 23, 12923–12931. 41. Tucek, J.; Zboril, R.; Petridis, D. Maghemite nanoparticles by view of Mössbauer spectroscopy. J. Nanosci. Nanotechnol. 2006, 6, 926–947. 42. Klencsar, Z.; Kuzmann. E.; Vertes, A. User-friendly software for Mössbauer spectrum analysis. J. Radioanal. Nucl. Chem. 1996, 210, 105–118. 43. Prochazka, R.; Tucek, P.; Tucek, J.; Marek, J.; Mashlan, M.; Pechousek, J. Statistical analysis and digital processing of the Mössbauer spectra. Meas. Sci. Technol. 2010, 21, doi:10.1088/09570233/21/2/025107. © 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).
PŘÍLOHA 9 Dvořák, Z.; Štarha, P.; Trávníček, Z. Evaluation of in vitro cytotoxicity of 6-benzylaminopurine carboplatin derivatives against human cancer cell lines and primary human hepatocytes Toxicol. in Vitro 25 (2011) 652–656
Toxicology in Vitro 25 (2011) 652–656
Contents lists available at ScienceDirect
Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit
Evaluation of in vitro cytotoxicity of 6-benzylaminopurine carboplatin derivatives against human cancer cell lines and primary human hepatocytes Zdeneˇk Dvorˇák a,⇑, Pavel Štarha b, Zdeneˇk Trávnícˇek b a ´ University, 17. listopadu 12, 771 46 Regional Centre of Advanced Technologies and Materials, Department of Cell Biology and Genetics, Faculty of Science, Palacky Olomouc, Czech Republic b ´ University, 17. listopadu 12, 771 46 Olomouc, Czech Republic Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacky
a r t i c l e
i n f o
Article history: Received 4 October 2010 Accepted 3 January 2011 Available online 11 January 2011 Keywords: Human hepatocytes Cytotoxicity 6-Benzylaminopurine derivative Carboplatin derivative CDK inhibitor
a b s t r a c t A series of seven platinum(II) cyclobutane-1,1-dicarboxylato (cbdc) complexes {[Pt(cbdc)(Ln)2], 1–7}, derived from carboplatin by a substitution of two NH3 molecules for two 2,6,9-trisubstituted 6-benzylaminopurine-based N-donor ligands (Ln), was studied by the MTT assay for their in vitro cytotoxic activity against seven human cancer cell lines, i.e. lung carcinoma (A549), cervix epithelioid carcinoma (HeLa), osteosarcoma (HOS), malignant melanoma (G361), breast adenocarcinoma (MCF7), ovarian carcinoma (A2780) and its cisplatin-resistant analogue (A2780cis), and against two primary cultures of human hepatocytes (LH31 and LH32). The prepared complexes were cytotoxic against several cancer cells, in some cases even more than cisplatin. The best results were achieved for complexes 1 (IC50 = 17.4 ± 2.0 lM) and 2 (IC50 = 14.8 ± 2.1 lV) against HOS cells, 1 (IC50 = 15.1 ± 6.8 lM), 2 (IC50 = 13.6 ± 5.2 lM) and 6 (IC50 = 19.0 ± 6.6 lM) against MCF7, 6 (IC50 = 6.4 ± 0.1 lM) against A2780, and 1–6 (IC50 = 15.6 ± 4.0, 12.9 ± 3.7, 15.8 ± 3.8, 16.6 ± 5.5, 22.1 ± 2.5, and 5.6 ± 1.7 lM, respectively) against A2780cis. Viability of human hepatocytes was not declined by the tested complexes up to the concentration of 50 lM (for 1, 3–7) and 20 lM (for 2; caused by lower solubility of this complex). Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction Carboplatin, diamminecyclobutane-1,1-dicarboxylatoplatinum(II) complex, [Pt(cbdc)(NH3)2], followed cisplatin, cis-[PtCl2 (NH3)2], as a representative of the second generation of platinum-based anticancer drugs (Kelland and Farrell, 2000). Although the carrier ligand, i.e. NH3, is identical in the case of both metallotherapeutic drugs, carboplatin is known to be less toxic as compared with cisplatin. In other
Abbreviations: A2780, human ovarian carcinoma cell line; A2780cis, human ovarian carcinoma cisplatin-resistant cell line; A549, human Caucasian lung carcinoma; A9opy, E-2-[1-(9-anthryl)-3-oxo-3-prop-2-enylpyridine; AM, acetoxymethyl; cbdc, dianion of cyclobutane-1,1-dicarboxylic acid; CDK, cyclin-dependent kinase; dach, trans-1,2-diaminocyclohexane; DMF, N,N0 -dimethylformamide; dmso, coordinated dimethyl sulfoxide molecule; G361, human Caucasian malignant melanoma; HeLa, human negroid cervix epithelioid carcinoma; HOS, human Caucasian osteosarcoma; ipram, isopropylamine; K-562, chronic myelogenous leukaemia; Ln, variously substituted 6-benzylaminopurine derivatives; L8, 2chloro-6-(3-methoxybenzyl)amino-9-isopropylpurine; L9, 2-(1-ethyl-2-hydroxyethylamino)-6-(4-methoxybenzyl)amino-9-isopropylpurine; L10, 2-chloro-6-(2,42-chloro-6-(2-methoxybendimethoxybenzyl)amino-9-isopropylpurine; L11, zyl)amino-9-isopropylpurine; MCF7, human Caucasian breast adenocarcinoma; meim, 1-methylimidazole; mepz, 1-methylpyrazole; MTT, 3-(4,5-dimethylthiazol2-yl)-2,5-diphenyltetrazolium bromide; ros, 2-(1-ethyl-2-hydroxyethylamino)-6(benzyl)amino-9-isopropylpurine. ⇑ Corresponding author. Tel.: +420 58 5634903; fax: +420 58 5634905. E-mail address:
[email protected] (Z. Dvorˇák). 0887-2333/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.tiv.2011.01.002
words, mitigation of negative side-effects compensates for lower activity of carboplatin, which is very important for the application of this substance in cancer therapy. One of our recent papers describes a series of the platinum(II) cyclobutane-1,1-dicarboxylato complexes of the [Pt(cbdc)(Ln)2] general composition, where n = 1–6 and L1 = 2-chloro-6-(2-fluoro5-bromobenzyl)amino-9-isopropylpurine (coordinated in the complex 1), L2 = 2-chloro-6-(3,4-dichlorobenzyl)amino-9-isopropylpurine (2), L3 = 2-chloro-6-(3-bromobenzyl)amino-9-isopropylpurine (3), L4 = 2-chloro-6-(2-trifluoromethylbenzyl)amino-9isopropylpurine (4), L5 = 2-chloro-6-(3-trifluoromethylbenzyl) amino-9-isopropylpurine (5) and L6 = 2-chloro-6-(4-trifluoromethylbenzyl) amino-9-isopropylpurine (6) (Dvorˇák et al., 2010). These complexes were tested by the AM assay for their in vitro cytotoxicity against the K-562 and MCF7 human cancer cells. The complex 3 (IC50 = 4.5 ± 1.0 lM) was evaluated as more active than cisplatin (IC50 = 4.7 lM) against the K-562 cells, while the complexes 1–4 (IC50 = 9.0 ± 2.3, 4.3 ± 0.2, 5.0 ± 0.3, and 7.9 ± 2.3 lM, respectively) exceeded the in vitro cytotoxic activity of cisplatin (IC50 = 10.9 lM) against the MCF7 cell line. Moreover, all the compounds were several times more cytotoxic as compared with carboplatin (Dvorˇák et al., 2010). The above-mentioned statements encouraged us to prepare one more carboplatin derivative with 6-benzylaminopurine-based Ndonor ligand, [Pt(cbdc)(L7)2] (7). L7 symbolizes highly effective
Z. Dvorˇák et al. / Toxicology in Vitro 25 (2011) 652–656
and selective CDK inhibitor 2-(3-hydroxypropylamino)-6-(benzyl)amino-9-isopropylpurine (bohemine), which was formerly, together with similar 6-benzylaminopurine-based CDK inhibitor 2-(1-ethyl-2-hydroxyethylamino)-6-(benzyl)amino-9-isopropylpurine (roscovitine), reported as a suitable N-donor ligand of the cytotoxic active platinum(II)-dichlorido and oxalato complexes ˇ et al., 2005 and Trávnícˇek et al., 2010). Moreover, we also (Malon decided to study deeply the biological activity of all the complexes 1–7 and to broaden the number of human cancer cell lines to assess in vitro cytotoxicity of the complexes. As it is discussed below, the tested platinum(II) complexes were in several cases more in vitro cytotoxic than cisplatin. This positive finding led us to investigate how the prepared substances 1–7 affect healthy noncancer cells in the hepatotoxicity test against primary cultures of human hepatocytes.
2. Materials and methods 2.1. Materials Collagen-coated culture dishes were purchased from BD Biosciences (Le Pont de Claix, France). All the chemicals and solvents were purchased from commercial sources, namely Sigma–Aldrich Co., Acros Organics Co., Lachema Co. and Fluka Co., and they were used as received. [Pt(cbdc)(dmso)2] and 2-(hydroxypropylamino)6-(benzyl)amino-9-isopropylpurine (bohemine, L7) were prepared by slightly modified synthetic procedures described by Bitha et al. (1990), and Oh et al. (1999), respectively.
2.1.1. Platinum(II) complexes 1–7 The synthesis and characterization of the complexes [Pt(cbdc)(Ln)2] (n = 1–6 for the complexes 1–6, see also Fig. 1) were reported by Dvorˇák et al. (2010). The complex [Pt(cbdc)(L7)2] (7; Fig. 1) was prepared according to the procedure described in the same literature source. Briefly, [Pt(cbdc)(dmso)2] reacted in the distilled water/isopropyl alcohol mixture (1:1, v/v) with two molar equivalents of bohemine (L7). The light grey product, which formed in two days of stirring at 90 °C, was filtered off and washed by distilled water and isopropyl alcohol.
653
[Pt(cbdc)(L7)2] (7): Anal. Calc. for C42H54N12O6Pt: C, 49.6; H, 5.4; N, 16.5. Found: C, 49.9; H, 5.3; N, 16.1%. IR (Nujol; cm 1): 526m (PtN), and 563vs (PtO). IR (KBr; cm 1): 3122m, 3068m (CHar), 2970m, 2932m, 2875m (CHal), 1661s (COox), 1612vs (CN), and 1551s (CC). Raman (cm 1): 3057s (CHar), 2978s, 2938vs, 2914vs (CHal), 1661w (COox), and 1611vs (CN). 1H NMR (400 MHz, DMFd7, ppm): d 8.74 (s, 1H, C8H), 8.66 (br, 1H, N6H), 7.51 (dd, 2H, J = 7.3 Hz, 1.5 Hz, C11H, C15H), 7.27 (tt, 2H, J = 7.3 Hz, 1.5 Hz, C12H, C14H), 7.20 (tt, 1H, J = 7.3 Hz, 1.5 Hz, C13H), 6.72 (t, 1H, J = 6.2 Hz, N2H), 4.74 (d, 2H, J = 6.0 Hz, C9H), 4.72 (sp, 1H, J = 6.8 Hz, C16H), 4.49 (br, 1H, C21H), 3.59 (t, 2H, J = 6.2 Hz, C21H), 3.42 (q, 2H, J = 6.2 Hz, C19H), 2.89 (t, 4H, J = 8.2 Hz, C25H, C27H), 1.77 (t, 2H, J = 8.2 Hz, C26H), 1.75 (t, 2H, C20H), 1.54 (d, 6H, J = 6.8 Hz, C17H, C18H). 13C NMR (400 MHz, DMF-d7, ppm): d 177.70 (C22, C23), 160.58 (C2), 153.53 (C6), 151.01 (C4), 140.62 (C10), 139.17 (C8), 128.82 (C12, C14), 128.30 (C11, C15), 127.21 (C13), 111.45 (C5), 60.71 (C21), 56.73 (C24), 48.30 (C16), 44.62 (C9), 39.45 (C19), 33.45 (C20), 31.38 (C25, C27), 22.06 (C17, C18), 15.85 (C26). 15N NMR (400 MHz, DMF-d7, ppm): d 179.9 (N9), 127.1 (N7), 89.8 (N6), 89.6 (N2). 195Pt NMR (400 MHz, DMF-d7, ppm): d-1611. 2.2. Characterization of [Pt(cbdc)(L7)2] (7) Elemental analyses were performed on a Fisons EA-1108 CHNSO Elemental Analyzer (Thermo Scientific). IR spectra were recorded on a Nexus 670 FT-IR spectrometer (Thermo Nicolet) at 400– 4000 cm 1 (KBr pellets) and 150–600 cm 1 (Nujol technique). Raman spectra were recorded using an NXR FT-Raman Module (Thermo Nicolet) between 150 and 3750 cm 1. 1H, 13C and 195Pt NMR spectra and 1H–1H gs-COSY, 1H–13C gs-HMQC, 1H–13C gs-HMBC two dimensional correlation experiments of the DMF-d7 solutions were measured at 300 K on a Bruker 300 device. 1H spectra were also, together with 1H–15N gs-HMBC experiments, recorded at 340 K. 1H and 13C spectra were adjusted against the signals of tetramethylsilane (Me4Si). 195Pt spectra were calibrated against K2PtCl6 in D2O found at 0 ppm. 1H–15N gs-HMBC experiments were obtained at natural abundance and calibrated against the residual signals of DMF adjusted to 8.03 ppm (1H) and 104.7 ppm (15N). The splitting of proton resonances in the reported 1H spectra is defined as s = singlet, d = doublet, t = triplet, q = quadruplet, sp = septuplet, br = broad band, dd = doublet of doublets, tt = triplet of triplets. 2.3. Human cancer cell lines Human cancer cell lines were purchased from European Collection of Cell Cultures (ECACC). The following cell lines were employed in the current study: human ovarian carcinoma cells (A2780; ECACC No. 93112519), human ovarian carcinoma cisplatin-resistant cells (A2780cis; ECACC No. 93112517), human Caucasian malignant melanoma (G361; ECACC No. 88030401), human Caucasian breast adenocarcinoma (MCF7; ECACC No. 86012803), human Caucasian lung carcinoma (A549; ECACC No. 86012804), human Caucasian osteosarcoma (HOS; ECACC No. 87070202) and human negroid cervix epithelioid carcinoma (HeLa; ECACC No. 93021013). The cells were cultured according to the ECACC instructions. Briefly, culture medium was DMEM (for cell lines HOS, MCF7, HeLa, A549), RPMI1640 (for cell lines A2780 and A2780cis) and McCoy´s (for cell line G361). The medium was supplemented with penicillin, streptomycin and 10% of foetal bovine serum. The cells were maintained at 37 °C and 5% CO2 in a humidified incubator. 2.4. Primary cultures of human hepatocytes
Fig. 1. The proposed structure of the complexes [Pt(cbdc)(Ln)2] (1–7), given with specification of the R1 and R2 substituents of the 6-benzylamino-9-isopropylpurine moiety.
Human hepatocytes were isolated from liver tissue, resected from multiorgan donors. A tissue acquisition protocol was in
654
Z. Dvorˇák et al. / Toxicology in Vitro 25 (2011) 652–656
accordance with the requirements issued by a local ethical commission in the Czech Republic. Human liver tissues used in this study were obtained from two donors: LH31 (male, 28 years) and LH32 (male, 70 years). Hepatocytes were isolated by two-step collagenase perfusion and the cells were plated on collagen-coated culture dishes using cell density of 14 104 cells/cm2 (Pichard´s Garcia et al., 2002). The culture medium was Williams and HAM F-12 (1:1) supplemented with penicillin, streptomycin, ascorbic acid, linoleic acid, holo-transferin, ethanolamine, glucagon, insulin, dexamethasone, pyruvate, glucose, glutamine, amphotericin. The medium was enriched for plating with 2% foetal calf serum (v/v). The medium was exchanged for a serum-free medium the day after and the culture was stabilized for additional 24 h. Thereafter, the cells were ready for treatments. The cultures were maintained at 37 °C and 5% CO2 in a humidified incubator. 2.5. Cytotoxicity assays Human cancer cell lines and primary cultures of human hepatocytes were treated with the tested compounds for 24 h, using multi-well culture plates of 96 wells (Vrzal et al., 2010). The following compounds – cisplatin, oxaplatin, carboplatin and 1–7 were applied to the cells up to the concentration of 50 lM. In parallel, the cells were treated with vehicle (DMF; 0.1%, v/v) and Triton X-100 (1%, v/v) to assess the minimal (i.e. positive control) and maximal (i.e. negative control) cell damage, respectively. Cells were incubated with MTT for 3–4 h, and after removal of the medium and washing the cells with PBS, formazan dye was dissolved in DMSO containing 1% of ammonia. Absorbance was measured spectrophotometrically at 540 nm (TECAN, Schoeller Instruments LLC). The data were expressed as the percentage of viability, when 100% and 0% represent the treatments with DMF and Triton X-100, respectively. The data on human hepatocytes were obtained from two independent cultures (obtained from two different donors). The data from cancer cell lines were acquired from three independent cell passages. 3. Results 3.1. Synthesis and characterization of the complex [Pt(cbdc)(L7)2] (7) The complex 7 was prepared according to the formerly described synthetic procedure employed for the synthesis of the complexes 1–6 (Dvorˇák et al., 2010), as described in Section 2.1.1 (Fig. 1). The characterization of the compound, by the methods summarized in Section 2.2., proved a composition corresponding to the [Pt(cbdc)(L7)2] formula. The results of the detailed NMR spectroscopic study indicated that the cbdc anion is bidentatecoordinated to the metal centre through two oxygen atoms and both L7 molecules are bound to the Pt(II) atom through their N7 atoms. 3.2. Cytotoxicity in human cancer cell lines We examined cytotoxicity of the prepared platinum(II) complexes and three standard platinum-derived cytotoxic compounds (i.e. cisplatin, oxaliplatin, carboplatin) in seven different commercial human cancer cell lines, including A2780, A2780cis, G361, MCF7, A549, HOS and HeLa. The obtained results (IC50 ± SD values given in lM) are summarized in Figs. 2–4, as well as in Table S1 of Supplementary material. No cytotoxic effects of the tested compounds were found in human A549 cells. Due to the limited solubility of the compounds, the estimate IC50 values were >50 lM for the complexes 1, 3–7, and cisplatin. The IC50 value for complex 2 was >20 lM (Fig. 2A). The cytotoxicity of the tested compounds was weak in HeLa cancer cells. With the exception of cisplatin
Fig. 2. Cytotoxicity of the tested compounds against the A549, HeLa and HOS cell lines. The cells were plated at 96-well dishes and cultured according to the manufacturer instructions. The tested compounds were applied to the cells for 24 h in concentrations ranging from 0.01 to 50 lM. As the positive and negative control, Triton-X100 (1% v/v) and vehicle (DMF; 0.1% v/v) were used, respectively. The cytotoxicity was assessed by the MTT test, and the values of IC50 were calculated. The bar graphs show the IC50 values against (Panel A) A549 cells, (Panel B) HeLa cells and (Panel C) HOS cells. The data are expressed as a mean ± SD from three independent experiments.
(IC50 = 39.9 ± 4.6 lM), 1 (IC50 = 47.0 ± 1.5 lM) and 4 (IC50 = 40.2 ± 4.2 lM), the estimated IC50 values for the tested compounds
Z. Dvorˇák et al. / Toxicology in Vitro 25 (2011) 652–656
655
Fig. 4. Cytotoxicity of the tested compounds against the A2780 and A2780cis cell lines. The cells were plated at 96-well dishes and cultured according to the manufacturer instructions. The tested compounds were applied to the cells for 24 h in concentrations ranging from 0.01 to 50 lM. As the positive and negative control, Triton-X100 (1% v/v), and vehicle (DMF; 0.1% v/v) were used, respectively. The cytotoxicity was assessed by the MTT test, and the values of IC50 were calculated. The bar graphs show the IC50 values in A2780 and A2780cis cells. The data are expressed as a mean ± SD from three independent experiments.
3.3. Cytotoxicity in primary cultures of human hepatocytes In none of the primary human hepatocyte cultures, the IC50 values were reached by the tested compounds. This was due to the limited solubility of these compounds. Therefore, the estimated IC50 values for all the tested complexes were IC50 > 50 lM, with the exception of carboplatin (IC50 > 1 lM) and 2 (IC50 > 20 lM). 4. Discussion
Fig. 3. Cytotoxicity of the tested compounds against the G361 and MCF7 cell lines. The cells were plated at 96-well dishes and cultured according to the manufacturer instructions. The tested compounds were applied to the cells for 24 h in concentrations ranging from 0.01 to 50 lM. As the positive and negative control, TritonX100 (1% v/v) and vehicle (DMF; 0.1% v/v) were used, respectively. The cytotoxicity was assessed by the MTT test, and the values of IC50 were calculated. The bar graphs show the IC50 values against (Panel A) G361 cells and (Panel B) MCF7 cells. The data are expressed as a mean ± SD from three independent experiments.
exceeded their solubility (Fig. 2B). Similarly, in osteosarcoma cells, the significant cytotoxic effects were detected only for cisplatin (IC50 = 34.2 ± 6.4 lM), 1 (IC50 = 17.4 ± 2.0 lM) and 2 (IC50 = 14.8 ± 2.1 lM), whereas the other compounds were found to be non-cytotoxic within the tested concentration range (Fig. 2C). The best cytotoxic potency of the tested compounds was attained in the case of G361 (Fig. 3A) and MCF7 (Fig. 3B) cells. The exception was the complex 7, which was not cytotoxic in either cell line in the concentration up to 50 lM. We also examined comparative effects of the tested compounds in cisplatin-sensitive (A2780) and cisplatin-resistant (A2780cis) human ovarian carcinoma cell line. Cisplatin exerted significantly higher cytotoxicity in A2780 cells (IC50 = 11.5 ± 1.6 lM) as compared to A2780cis cells (IC50 = 30.3 ± 6.1 lM), as expected. Beside the complex 7, which was not cytotoxic, all the complexes displayed significant cytotoxicity in both A2780 and A2780cis cells (Fig. 4). We have also found that none of the employed cancer cell lines were sensitive to oxaliplatin (IC50 > 50 lM) and carboplatin (IC50 > 1 lM) up to the concentration given by the compound solubility.
In the present work, we tested in vitro cytotoxicity of the [Pt(cbdc)(Ln)2] complexes (1–7) and compared the obtained results with those of cisplatin, oxaliplatin and carboplatin. The cytotoxicity was evaluated in seven commercial human cancer cell lines (A2780, A2780cis, G361, MCF7, A549, HOS and HeLa) derived from various types of cancer (vide supra) and in two primary cultures of human hepatocytes. The cells were challenged for 24 h with the tested compounds and the MTT test was used to assess the resulting cytotoxicity. We have found out that oxaliplatin (IC50 > 50 lM) and carboplatin (IC50 > 1 lM) did not cause significant cell damage in any cell line used, because the cytotoxic potential of oxaliplatin and carboplatin was restricted by their limited solubility. Therefore, it makes sense to test platinum-based derivatives with increased solubility and clinically reasonable cytotoxicity in vitro. Indeed, in the current paper we describe different in vitro cytotoxicity of the tested platinum-derivatives. The tested complexes were not cytotoxic against human Caucasian lung carcinoma cells A549, and very weak cytotoxicity was observed against human negroid cervix epithelioid carcinoma cells (HeLa) and human Caucasian osteosarcoma cells (HOS). In contrast, significant dose-dependent cytotoxicity of the tested compounds was observed against human Caucasian malignant melanoma cells (G361), human Caucasian breast adenocarcinoma cells (MCF7) and human ovarian carcinoma cell line (A2780) and its cisplatin-resistant variant (A2780cis). The cytotoxic effect of the studied compounds 1–7 in primary cultures of human hepatocytes, an in vitro model considered as the most suitable for studies of xenobiotic cytotoxicity and metabolism, was studied as well. We used primary human hepatocytes obtained from two human liver donors. The obtained results of in vitro hepatotoxicity revealed that the studied platinum(II)
656
Z. Dvorˇák et al. / Toxicology in Vitro 25 (2011) 652–656
complexes 1–7 did not affect healthy human hepatocytes up to the concentration of 50 lM. The presented platinum(II) cyclobutane-1,1-dicarboxylato complexes follow the recently reported highly in vitro cytotoxic ˇ et al., 2005 and oxalato dichlorido {cis-[PtCl2(Ln)2]; Malon {[Pt(ox)(Ln)2]; Štarha et al., 2010; Trávnícˇek et al., 2010; Vrzal et al., 2010} complexes prepared in our laboratory. The cis[PtCl2(Ln)2] complexes were tested against the G361, HOS, MCF7 and K-562 (not used in this work) cancer cells. We performed a comparison of the calculated IC50(cisplatin)/IC50(complex) ratio of the formerly tested dichlorido and oxalato complexes with the cyclobutan-1,1-dicarboxylato ones reported in this work to better describe and compare in vitro cytotoxicity of these types of complexes. The ratio of the most active cis-[PtCl2(ros)2] complex with the coordinated CDK inhibitor roscovitine equals 3.0 (it means that the tested complex is three times more active than cisplatin) ˇ et al., 2005). against G361 and HOS and 5.0 against MCF7 (Malon The oxalato complexes were tested against the same cancer cell lines as herein described complexes 1–7 (Trávnícˇek et al., 2010; Vrzal et al., 2010). Again we can calculate the ratio of the in vitro cytotoxicity against respective cancer cell line of cisplatin versus the appropriate oxalato complex, which equals 1.7 against G361 for [Pt(ox)(L8)2], 1.7 against HeLa for [Pt(ox)(L9)2], 5.4 against MCF7 for [Pt(ox)(L10)2], 9.5 against HOS for [Pt(ox)(L11)2], 3.6 against A2780 for [Pt(ox)(L8)2] and 9.5 against A2780cis for [Pt(ox)(L8)2]. It has to be mentioned that all of these ratios are higher (it means more in vitro cytotoxic active) as compared with 1–7, in particular with 0.6 against G361 for 6, 1.0 against HeLa for 4, 1.4 against MCF7 for 2, 2.3 against HOS for 2, 1.8 against A2780 for 6 and 5.4 against A2780cis for 6. Another important finding is that the studied platinum(II) complexes 1–7 overcome cisplatin resistance. It was demonstrated by the IC50(A2780cis)/IC50(A2780) ratio, which is equal to 2.6 for cisplatin and 0.9 (1), 1.1 (2), 1.4 (3), 1.2 (4), 1.2 (5) and 0.9 (6). We can compare these results with those reported in the literature sources for different in vitro cytotoxic mononuclear platinum(II) complexes. The above-mentioned [Pt(ox)(L8)2] complex has the IC50(A2780cis)/IC50(A2780) ratio of 1.0, those of cis-[PtCl2(ipram)(meim)] and cis-[PtCl2(ipram)(mepz)] complexes are equal to 1.9, and 2.3, respectively (Pantoja et al., 2006), while that of the pyrophosphato (pyro) complex [Pt(pyro)(dach)] was determined to be 2.4 (Bose et al., 2008). As for six platinum(II) dichlorido complexes with 2-aminomethylpyrrolidine-based ligands, their IC50(A2780cis)/IC50(A2780) ratios equal 1.1–4.9 (Diakos et al., 2009). The last example, which could be mentioned concerning in vitro cytotoxicity against A2780 and A2780cis cells, is cis-[PtCl2(A9opy)] with the IC50(A2780cis)/IC50(A2780) ratio of 1.4 (Marqués-Gallego et al., 2009). On the other hand, the effectiveness of tested compounds against cisplatin-resistant cells is not, in principle, general, since cisplatin resistance comprises multiple mechanisms, including decreased drug uptake, increased efflux, increased inactivation by glutathione, increased excision of DNA adducts etc. In conclusion, we demonstrated that the tested carboplatin derivatives 1–7 may be classified as tentatively promising anticancer drugs, because they were found to be toxic against human cancer cells but not against healthy human hepatic cells. Further, the platinum(II) complexes of the general formula [Pt(cbdc)(Ln)2] significantly more effectively inhibit growth of both A2780 and A2780cis cancer cells as compared with several recently reported mononuclear platinum(II) complexes. Conflict of interest We declare no conflict of interest.
Acknowledgements This work was financially supported by the Grant Agency of the Czech Republic (Grant Nos.: GACR503/10/0579 and GACR304/10/ 0149), the Ministry of Education, Youth and Sports of the Czech Republic (a Grant No.: MSM6198959218), Faculty of Science of Palacky University in Olomouc (a Grant No.: PrF_2010_018), and by the Operational Program Research and Development for Innovations – European Social Fund (a Grant No.: CZ.1.05/2.1.00/ 03.0058). We thank Mr. Lukáš Dvorˇák for his help with the synthesis of the complexes, Ms. Radka Novotná for IR and Raman spectra measurements and Dr. Igor Popa for NMR spectra measurements and interpretation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.tiv.2011.01.002. References Bitha, P., Morton, G.O., Dunne, T.S., Delos Santos, E.F., Lin, Y., Boone, S.R., Haltiwanger, R.C., Pierpoint, C.G., 1990. (Malonato)bis[sulfinylbis[methane]S]platinum(II) compounds: versatile synthons for a new general synthesis of antitumor symmetrical and dissymmetrical (Malonato)platinum(II) complexes. Inorganic Chemistry 29, 645–652. Bose, R.N., Maurmann, L., Mishur, R.J., Yasui, L., Gupta, S., Grayburn, W.S., Hofstetter, H., Salley, T., 2008. Non-DNA-binding platinum anticancer agents: cytotoxic activities of platinum-phosphato complexes towards human ovarian cancer cells. PNAS 105, 18314–18319. Diakos, C.I., Zhang, M., Beale, P.J., Fenton, R.R., Hambley, T.W., 2009. Synthesis, characterisation and in vitro cytotoxicity studies of a series of chiral platinum(II) complexes based on the 2-aminomethylpyrrolidine ligand: X-ray crystal structure of [PtCl2(R-dimepyrr)] (R-dimepyrr = N-dimethyl-2(R)aminomethylpyrrolidine). European Journal of Medicinal Chemistry 44, 2807– 2814. Dvorˇák, L., Popa, I., Štarha, P., Trávnícˇek, Z., 2010. In vitro cytotoxic active platinum(ii) complexes derived from carboplatin and involving purine derivatives. European Journal of Inorganic Chemistry 3441, 3448. Kelland, L.-R., Farrell, N.-P., 2000. Platinum-based Drugs in Cancer Therapy. Humana Press, Totowa, New Jersey. ˇ , M., Trávnícˇek, Z., Marek, R., Strnad, M., 2005. Synthesis, spectral study and Malon cytotoxicity of platinum(II) complexes with 2, 9-disubstituted-6benzylaminopurines. Journal of Inorganic Biochemistry 99, 2127–2138. Marqués-Gallego, P., Kalayda, G.V., Jaehde, U., den Dulk, H., Brouwer, J., Reedijk, J., 2009. Cellular accumulation and DNA platination of two new platinum(II) anticancer compounds based on anthracene derivatives as carrier ligands. Journal of Inorganic Chemistry 103, 791–796. Oh, C.H., Lee, S.C., Lee, K.S., Woo, E.R., Hong, C.Y., Yang, B.S., Baek, D.J., Cho, J.H., 1999. Synthesis and biological activities of C-2, N-9 substituted 6-benzylaminopurine derivatives as cyclin-dependent kinase inhibitor. Archiv der Pharmazie 332, 187–190. Pantoja, E., Gallipoli, A., van Zutphen, S., Tooke, D.M., Spek, A.L., Navarro-Ranninger, C., Reedijk, J., 2006. In vitro antitumor activity and interaction with DNA model bases of cis-[PtCl2(iPram)(azole)] complexes and comparison with their trans analogues. Inorganica Chimica Acta 359, 4335–4342. Pichard-Garcia, L., Gerbal-Chaloin, S., Ferrini, J.B., Fabre, J.M., Maurel, P., 2002. Use of long-term cultures of human hepatocytes to study cytochrome P450 gene expression. Methods in Enzymology 357, 311–321. Štarha, P., Trávnícˇek, Z., Popa, I., 2010. Platinum(II) oxalato complexes with adeninebased carrier ligands showing significant in vitro antitumor activity. Journal of Inorganic Biochemistry 104, 639–647. Trávnícˇek, Z., Štarha, P., Popa, I., Vrzal, R., Dvorˇák, Z., 2010. Roscovitine-based CDK inhibitors acting as N-donor ligands in the platinum(II) oxalato complexes: preparation, characterization and in vitro cytotoxicity. European Journal of Medicinal Chemistry 45, 4609–4614. Vrzal, R., Štarha, P., Dvorˇák, Z., Trávnícˇek, 2010. Evaluation of in vitro cytotoxicity and hepatotoxicity of platinum(ii) and palladium(ii) oxalato complexes with adenine derivatives as carrier ligands. Journal of Inorganic Biochemistry 104, 1130–1132.
PŘÍLOHA 10 Trávníček, Z.; Štarha, P.; Vančo, J.; Šilha, T.; Hošek, J.; Suchý, P.; Pražanová, G. Anti-inflammatory active gold(I) complexes involving 6-substituted-purine derivatives J. Med. Chem. 55 (2012) 4568–4579
Article pubs.acs.org/jmc
Anti-inflammatory Active Gold(I) Complexes Involving 6-SubstitutedPurine Derivatives Zdeněk Trávníček,*,† Pavel Štarha,† Ján Vančo,† Tomás ̌ Šilha,† Jan Hošek,†,‡ Pavel Suchý, Jr.,§ and Gabriela Pražanov᧠†
Regional Centre of Advanced Technologies and Materials, Department of Inorganic Chemistry, Faculty of Science, Palacký University, 17 listopadu 1192/12, 771 46 Olomouc, Czech Republic ‡ Department of Natural Drugs, and §Department of Human Pharmacology and Toxicology, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences Brno, Palackého 1/3, 612 42 Brno, Czech Republic S Supporting Information *
ABSTRACT: The gold(I) complexes of the general formula [Au(Ln)(PPh3)]·xH2O (1−8; n = 1−8 and x = 0−1.5), where Ln stands for a deprotonated form of the benzyl-substituted derivatives of 6-benzylaminopurine, were prepared, thoroughly characterized (elemental analyses, FT-IR, Raman and multinuclear NMR spectroscopy, ESI+ mass spectrometry, conductivity, DFT calculations), and studied for their in vitro cytotoxicity and in vitro and in vivo anti-inflammatory effects on LPS-activated macrophages (derived from THP-1 cell line) and using the carrageenan-induced hind paw edema model on rats. The obtained results indicate that the representative complexes (1, 3, 6) exhibit a strong ability to reduce the production of pro-inflammatory cytokines TNF-α, IL-1β and HMGB1 without influence on the secretion of anti-inflammatory cytokine IL-1RA in the LPSactivated macrophages. The complexes also significantly influence the formation of edema, caused by the intraplantar application of polysaccharide λ-carrageenan to rats in vivo. All the tested complexes showed similar or better biological effects as compared with Auranofin, but contrary to Auranofin they were found to be less cytotoxic in vitro. The obtained results clearly indicate that the gold(I) complexes behave as very effective anti-inflammatory agents and could prove to be useful for the treatment of difficult to treat inflammatory diseases such as rheumatoid arthritis. (PPh3)] were tested and found to be in vitro antimicrobial active agents [HL = 4-amino-N-(5-methylisoxazole-3-yl)benzenesulfonamide, mercaptopropionic acid, mercaptonicotinic acid and mercaptobenzoic acid, D-penicillamine, pyrazole, imidazole, 1,2,3-triazole, 1,2,4-triazole, tetrazole].11−13 In vitro antitumor activity against the cervival carcinoma (HeLa-229), ovarian carcinoma (A2780), and cisplatin-resistant ovarian carcinoma (A2780cis) human cancer cell lines was investigated for 10 gold(I) complexes of the above-mentioned general composition with the anions of 3-(aryl)-2-sulfanylpropenic acid. 14 Recently, it has been reported that Auranofin [triethylphosphine-(2,3,4,6-tetra-O-acetyl-β-D-thiopyranosato)gold(I)] and other gold(I) and gold(III) complexes have a positive therapeutic effect on HIV-infection15,16 and their application could be beneficial for AIDS suffering patients.17 Nevertheless, the best known application of gold(I) compounds concerns the treatment of rheumatoid arthritis.
INTRODUCTION Adenine derivatives substituted at the N6 position, such as 6benzylaminopurine (Bap) and its benzyl-substituted analogues, belong to the group of both natural and artificial growth regulators called cytokinins.1 Furthermore, the Bap derivatives substituted besides the benzyl aromatic ring also at a purine moiety may be integrated into a family of cyclin dependent kinase inhibitors, such as 2-[(R)-(1-ethyl-2-hydroxyethylamino)]-6-benzylamino-9-isopropylpurine (R-roscovitine),2 which have received attention because of their biological, particularly antitumor, activity.3 From the coordination or bioinorganic chemistry point of view, variously substituted Bap derivatives represent a group of polydentate N-donor ligands biologically active themselves, and moreover, the transition metal complexes involving such ligands were found to be highly biologically active as well.4−7 Gold(I) coordination compounds are known to play several biochemical roles.8−10 Focusing on the gold(I) triphenylphosphine (PPh3) complexes involving N-donor heterocyclic ligands, some compounds of the general composition [Au(L)-
■
© 2012 American Chemical Society
Received: October 20, 2011 Published: April 30, 2012 4568
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
from the human acute monocytic leukemia (THP-1) cell line. The influence of the tested compounds on carrageenan-induced paw edema on rats was also evaluated. The obtained results are discussed within the framework of the following text.
Nowadays, their representatives Auranofin, Solganal [{(2S,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)-oxane2-thiolato}gold(I)], and Myochrysin [sodium ((2-carboxy-1carboxylatoethyl)thiolato)gold(I), GST] are the most important clinically used gold(I) complexes.18,19 An administration of metallic gold or its complexes for the treatment of different diseases, known as chrysotherapy, came through various phases during its several thousand years history. The last one started in 1929,20 when the first parenteral drug preparations, containing gold or its compounds, were introduced into the therapy of inflammation-related indications. A significant progress in chrysotherapy occurred in mid-1980s, when the first oral formulations, containing drug Auranofin,21 were approved. At the time of the introduction of parenteral chrysotherapy, only a little was known about the pharmacodynamics and mechanism of action of gold-containing compounds in biological systems. Since then, major advances have been achieved to elucidate the mechanisms of action, although this topic is not completely closed even nowadays.16,22−25 Several mechanisms of action were proposed to explain the anti-inflammatory activity of goldcontaining drugs. A lot of discussions were dedicated to the identification of the biologically active end-products of gold drugs metabolism, and most of investigators agreed that Au(I)26 as well as Au(I) metabolites (such as [Au(CN)2]−,27,28 Au(III),29 or Au nanoparticles30,31) play a major role. Because of the composition and structure diversities of active species, a number of various mechanisms of action were proposed. Specifically: (a) the mechanism induced by Au(I) compounds and based on the cellular enzymes inhibition,32 especially the serine esterases, such as elastase, cathepsin G, B, K, S.33,34 The next identified cellular processes, also directly influenced by Au(I) compounds, may be as follows: the inhibition of lysosome enzymes release (i.e., β-glucuronidase, lysosyme) in polymorphonuclears,32,35 the inhibition of T-lymphocytes, macrophages, immunoglobulin levels, rheumatoid factor titers, tumor necrosis factor,35 monocyte capacity to produce superoxide anions,36,37 bone resorption, and vascular cell adhesion molecules on endothelial cells, the release of stress proteins from macrophages and surface adhesion of polymorphonuclear neutrophils,38,39 the inhibition of selenium enzymes, especially activity of thioredoxin reductase,40,41 and expression of different oxidative-stress proteins such as ezrin or peroxiredoxins;42 (b) the conversion of Au(I) compounds to the [Au(CN)2]− species, which suppress oxidant production of inflammatory cells, including neutrophils, monocytes, and lymphocytes;28 (c) the oxidation of Au(I) to Au(III) involved into a reversible redox process, in which Au(III) can be considered to be a better oxidant than Au(I) and dominates both the anti-inflammatory and toxic effects of gold salts;29 (d) the reduction of Au(I) or Au(III) to colloidal gold Au(0),43 which, similarly to Au(I), suppresses the activity of inflammatory-associated cytokines, tumor necrosis factor, immune complexes, and rheumatoid factors.44 On the basis of the clinical success of the last-mentioned gold(I) complexes, we decided to prepare, characterize, and study anti-inflammatory activity of the prepared complexes on in vitro and in vivo models. The ability to regulate the levels of inflammatory-related cytokines TNF-α (tumor necrosis factor‑α), IL-1β (interleukin-1β), and HMGB1 (high-mobility group protein B1), i.e., pro-inflammatory cytokines, and IL-10 (interleukin-10) and IL-1RA (interleukin-1 receptor antagonist), i.e., anti-inflammatory cytokines, were determined in vitro on lipopolysacharide-(LPS)-stimulated macrophages derived
CHEMISTRY Synthesis and General Properties. The gold(I) complexes of the composition [Au(L1)(PPh3)]·H2O (1), [Au(L2)(PPh3)] (2), [Au(L3)(PPh3)] (3), [Au(L4)(PPh3)] (4), [Au(L5)(PPh3)]·1.5H2O (5), [Au(L6)(PPh3)] (6), [Au(L7)(PPh3)] (7), [Au(L8)(PPh3)] (8) (Figure 1), where Ln stands
■
Figure 1. Synthetic pathway for the preparation of the gold(I) complexes (1−8) and schematic representations of Auranofin, indomethacin, and prednisone.
for the deprotonated form of the appropriate 6-benzylaminopurine derivative (see Supporting Information, Scheme S1), were prepared by the reaction of [AuCl(PPh3)] and HLn in acetone, as described in the Experimental Section, and obtained as white powder products (see Supporting Information). Complexes 1−8 are very well soluble in N,N′-dimethylformamide (DMF), methanol, ethanol, and butanol and partially soluble in water at room temperature. The complexes behave as nonelectrolytes in the 10−3 M methanolic solutions. The presence of the water molecules of crystallization in 1 and 5 was proved by simultaneous thermogravimetric (TG) and differential thermal (DTA) analyses, which also showed the dehydrated complexes 1 and 5, as well as the remaining gold(I) complexes, to be thermally stable up to 144−190 °C, when their thermal decay started (see Supporting Information, Figure S1). Electrospray ionization mass spectrometry in a positive mode (ESI+) mass spectra of 1−8 contain the [Au(Ln)(PPh3) 4569
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
+ H]+ molecular peaks (in the case of the hydrated complexes 1 and 5 without the water molecules of crystallization) as well as the [HLn + H]+ fragments corresponding to the respective 6benzylaminopurine derivative. The peaks of the [Au(Ln)(PPh3) + Na]+ adducts were observed in the mass spectra of 3−5 (Supporting Information, Figure S2). NMR, IR and Raman Spectroscopy. All the hydrogen and carbon atoms of the free HLn and PPh3 molecules were located in the 1H and 13C NMR spectra of 1−8, except for the N6H proton, which was detected in the case of 6−8 only, and the N9H one, which was not found, due to deprotonization of HLn, in any proton spectrum of 1−8 (see Supporting Information). The highest coordination shifts (Δδ = δcomplex − δligand; ppm) of the signals of the 6-benzylaminopurine derivatives were calculated for the C4 and C8 atoms, which were shifted by 4.83−10.27 ppm upfield and 8.33−9.08 ppm downfield, respectively, and C8H protons shifted by 0.17−0.32 ppm upfield (see Supporting Information, Table S1). The complex 1 was also studied by 31P NMR showing the P-atom signal shifted by 36 ppm upfield as compared with free PPh3. These results indicate that the 6-benzylaminopurine derivatives coordinate through the N9 atoms and PPh3 molecule coordinates to the central gold(I) atom through the phosphorus atom. The anticipated N9-coordination of the purine moiety to the Au(I) atom within the structure of 1−8 was reported for several gold(I) complexes, such as [Au(ade)PPh3]45 and [Au(ade)PEt3],46 where the deprotonated form of adenine (ade) coordinates to the central atom via the N9 atom of the purine moiety, as determined by the X-ray crystallographic study. The IR and Raman spectra of 1−8 contain the ν(CN)ar vibration of the purine ring of very strong intensity, whose maximum were detected at 1610−1619 cm−1 (IR) and 1612− 1617 cm−1 (Raman) and shifted by 3−30 cm−1 as compared with those observed in free 6-benzylaminopurine derivatives. The ν(CC)ar, ν(C−H)al, ν(C−H)ar, and ν(N−H) vibrations were, as well as ν(Car−Cl), ν(Car−F), and ν(Cmet−H) of the appropriate substituent, also detected in the spectra of the complexes (see Supporting Information). The maxima assignable to ν(Au−N) were detected at 543−547 cm−1 (IR) and 542−593 cm−1 (Raman), while the ν(Au−P) stretching vibrations were observed at 338−361 cm−1 (IR) and 358−368 cm−1 (Raman). Quantum Chemical Calculations. In an effort to predict the molecular structure of 1, the knowledge based on the results of spectroscopic and other analyses were summarized, together with the known structures of similar complexes, and the built model was optimized at the DFT-level using the hybrid B3LYP functional with the LACVP+** basis set using the Spartan (version 1.1.0) software.47−49 Optimized geometry of the complex 1 is shown in Supporting Information, Figure S3. The gold(I) atom is bidentate-coordinated by one N9coordinated 6-(2-fluorobenzylamino)purine anion (Au−N9 = 2.035 Å) and by one P-coordinated triphenylphoshine molecule (Au−P = 2.382 Å). More detailed discussion regarding interatomic parameters is given in Supporting Information.
Figure 2. THP-1 cells were treated with the decreasing concentration (50.00−0.00 μM) of 1, 3, 6, and Auranofin. After 24 h incubation, the number of metabolically active cells was determined by the WST-1 test. The viability was calculated as the percentage of vehicle-treated cells. The results are expressed as means ± SE for three independent experiments.
values of 1.91 μM for 1, 1.87 μM for 3, and 2.44 μM for 6. On the other hand, in the range of the concentrations between 1.00 and 0.02 μM, complexes 1, 3, and 6 caused significant accrual of numbers of metabolically active cells, which was manifested by higher amount of formazan, formed by the NAD(P)H-oxidase enzymatic system of viable cells from the WST-1 dye. It is widely known that kinetin (6-furfurylaminopurine) itself as well as some other cytokinins (including Bap and its derivatives) are able to positively affect the cell proliferation and metabolism, and may serve as antiaging agents (for more information, see recent review, Barciszewski et al., ref. 50). It has been found that the Bap derivates, which serve as N-donor ligands in compounds 1, 3, and 6, did not influence the number of metabolically active cells in the concentrations between 3.13 and 0.02 μM (data not shown). Therefore, the observed proproliferative effect is undoubtedly caused by the individual Au(I) complexes. Cytokine Expression. The progression of inflammation is connected with the expression of many genes, which trigger, maintain, and terminate it. The cooperation of individual types of immune cells during the inflammation is driven by the cytokines. The prepared gold(I) complexes have been supposed to exhibit an anti-inflammatory effect primarily because of their similarity to Auranofin, which was demonstrated by their ability to modulate cytokine expression.51 For this purpose, the expression of inflammatory-related cytokines was determined. To better understand the activity of the complexes (1−6), we tested their influence on the selected proand anti-inflammatory cytokines in the equitoxic concentration to Auranofin, i.e., 600 nM and further also in the 100 nM concentration. The biological effect of Au complexes was compared to Auranofin (tested in the 100 nM concentration) and prednisone (tested in the 1 μM concentration). The ability of the tested compounds to reduce gene transcription of a typical member of pro-inflammatory cytokines TNF-α is depicted in Figure 3a. The complex 1 was able to significantly decrease the level of mRNA of TNF-α at 2 and 4 h after the LPS treatment. On the other hand, all three selected complexes (1, 3, and 6) were able to reduce the TNF-α secretion by the factor of ∼2.5 (see Figure 3b). This effect is more similar to the effect of prednisone than to Auranofin, which reduced the TNF-α secretion only by the factor of 1.3. The mechanism of this regulation probably lies in the post-transcriptional phase because the level of TNF-α
IN VITRO ANTI-INFLAMMATORY ACTIVITY TESTING Cytotoxicity against THP-1 Cells. The cytotoxic effect of the selected representatives of the gold(I) complexes 1, 3, and 6 and Auranofin was evaluated against THP-1 cells before the testing of anti-inflammatory activity itself. Complexes 1, 3, and 6 showed strong cytotoxic action (Figure 2) with the LD50
■
4570
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
Figure 3. Effects of the gold(I) complexes and the reference drugs Auranofin and prednisone on the LPS-induced TNF-α gene expression (a) and TNF-α secretion (b). The cells were pretreated with complexes 1, 3, and 6 (600 nM) and Auranofin (100 nM), prednisone (1 μM), or the vehicle (DMSO) only. After 1 h of incubation, the inflammatory response was induced by LPS (except for the control cells). Gene expression was measured 2, 4, and 8 h after LPS treatment, and secretion was measured 24 h after LPS adding. The results are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), *** significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofin-treated cells (p < 0.05), ## significant difference in comparison to Auranofin-treated cells (p < 0.01).
Figure 4. Effects of the gold(I) complexes and the reference drugs Auranofin and prednisone on the LPS-induced IL-1β gene expression (a) and IL1β secretion (b). The cells were pretreated with complexes 1, 3, and 6 (600 nM) and Auranofin (100 nM), prednisone (1 μM), or the vehicle (DMSO) only. After 1 h of incubation, the inflammatory response was induced by LPS (except for the control cells). Gene expression was measured 2, 4, and 8 h after LPS treatment, and secretion was measured 24 h after LPS adding.The results are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01), *** significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofin-treated cells (p < 0.05), ## significant difference in comparison to Auranofin-treated cells (p < 0.01), ### significant difference in comparison to Auranofin-treated cells (p < 0.001).
mRNA was only slightly affected with the exception of 1. This correlated with the level of the secreted cytokine, where 1 showed the lowest level of the secreted TNF-α among all the tested compounds. The complexes (1, 3, and 6) were two times more efficient than Auranofin at the same effective concentration (i.e., at the same cytotoxicity levels) (Figure 3b). It has been found that the complexes (1, 3, and 6) showed the same minimal effect on the TNF-α secretion as compared to Auranofin in the equimolar concentration of 100 nM (Supporting Information, Figure S4). Auranofin was able to decrease the TNF-α secretion in concentrations higher than 5 μM.51,52 The IL-1β pro-inflammatory cytokine was tested as the second one. Complexes 1 and 3 significantly reduced its transcription (Figure 4a). Although complex 6 also decreased the IL-1β transcription 2 and 8 h after the LPS treatment, 4 h after adding LPS, when IL-1β transcription reached the maximum, this compound showed only a slight effect on this action. Compound 6 exhibits very similar quantitative effect on the transcription of this gene as prednisone but slower onset of its action. These observations corresponded to the secretion of
IL-1β (Figure 4b). The complexes 1 and 3 reduced the secreted amount of IL-1β by the factor of ∼7, while complex 6 was by only 3.4. Nevertheless, it was still twice lower than for Auranofin, which decreased the amount of the secreted IL-1β by the factor of 1.6. However, it has to be noted that Auranofin was used in the 6 times lower (equitoxic) concentration than the tested gold(I) complexes. In the 100 nM concentration, the tested Au(I) complexes showed the same or better ability to reduce the IL-1β secretion than Auranofin. The best effect on the secretion of this cytokine was observed for complex 6, which reduced its level 5.9 times in comparison with the vehicle-treated cells, while complexes 1 and 3 decreased the IL1β production by the factors of 3.6 and 2, respectively (Supporting Information, Figure S5). Different behavior of complex 6 at lower concentration could be caused by a different pattern of cytotoxic curve in the range of the concentrations of 3.13 and 0.02 μM (Figure 2). A previous study showed that Auranofin becomes more active at the concentrations over 10 μM.51 Similar effect was observed for another Au(I)-containing drug, Myochrysine (GST). Myochrisine (GST) decreased IL1β secretion at the concentration of 2.6 μM, but massive 4571
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
Figure 5. Effects of the gold(I) complexes and the reference drugs Auranofin and prednisone on the LPS-induced IL-1RA gene expression (a) and IL-1RA secretion (b). The cells were pretreated with complexes 1, 3, and 6 (600 nM) and Auranofin (100 nM), prednisone (1 μM), or the vehicle (DMSO) only. After 1 h of incubation, the inflammatory response was induced by LPS (except for the control cells). Gene expression was measured 2, 4, and 8 h after LPS treatment, and secretion was measured 24 h after LPS adding. The results are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01), *** significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofin-treated cells (p < 0.05), ## significant difference in comparison to Auranofin-treated cells (p < 0.01), ### significant difference in comparison to Auranofin-treated cells (p < 0.001).
decrease was achieved by the concentration of 25.6 μM.53 Interestingly, Auranofin decreased the secretion of IL-1β without affection of its transcription, and contrariwise, prednisone reduced transcription of this cytokine but did not change its secretion. IL-1RA acts as a physiological antagonist of IL-1β, and their mutual balance is important for the regulation of the progress of inflammation and for maintaining the homeostasis. The tested gold(I) complexes tended to decrease its transcription, similarly to prednisone (Figure 5a). Contrary to the transcription, complexes 1, 3, and 6 did not affect the secretion of IL-1RA significantly (Figure 5b). We were not able to explain this discrepancy yet, but it is possible that these compounds decreased the transcription, and on the other side, they stabilized mRNA or positively regulated translation and secretion, and thus they preserved the physiological conditions. Prednisone decreased the level of produced IL-1RA by the factor of 2.5 (Figure 5b). The metallodrug Auranofin slightly increased IL-1RA mRNA synthesis, and this trend correlated with the IL-1RA secretion, which was 1.3 times higher than in the vehicle-treated cells. With the aim to better describe the inflammatory process in the target cells, the ratio between the expression of IL-1β and IL-1RA was calculated. It is known that many of inflammatory diseases, like arthritis or inflammatory bowel disease, are characterized by either local overproduction of IL-1 (either IL-1α and/or IL-1β) and/or underproduction of IL-1RA.54 This situation leads into absolutely or relatively higher level of pro-inflammatory cytokine IL-1. As demonstrated in Figure 6, complexes 1 and 3 reached very low IL-1β/ IL-1RA ratios, comparable to the control cells (without the LPS treatment). On the other hand, complex 6, behaved similarly to Auranofin. The steroid anti-inflammatory drug prednisonetreated cells showed this ratio almost 3 times higher than the vehicle-treated cells. The extracellular HMGB1 also exhibits pro-inflammatory properties. As can be seen from Figure 7a, the tested compounds had only little effect on the HMGB1 transcription. However, they almost twice increased the production of mRNA for HMGB1 8 h after the LPS treatment. On the other hand, complexes 1, 3, and 6 reduced extracellular translocation of this dual-function protein, similarly to Auranofin (Figure 7b). Recently reported papers showed that the gold-containing drug
Figure 6. The calculated ratio of IL-1β/IL-1RA. The values were obtained from the ELISA experiments of individual cytokines as it is described in Figures 4 and 5. The results are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01),*** significant difference in comparison to vehicle-treated cells (p < 0.001), ### significant difference in comparison to Auranofintreated cells (p < 0.001).
Myochrysine (GST) decreased the HMGB1 secretion on different experimental models but concentrations used in these studies were higher than 50 μM.55,56 This work was also focused on the expression of antiinflammatory cytokine IL-10. However, the THP-1 cells are not able to secrete anti-inflammatory cytokine IL-10,57 thus only the transcription of this protein was measured. As it is shown in the Figure 8, the tested gold(I) complexes, Auranofin, and prednisone markedly reduced production of mRNA for IL-10 2 h after the LPS addition. In the other time intervals (4 and 8 h after the LPS treatment), the effect of the used compounds was minimal. The lower level of IL-10 mRNA was detected, but the study of Lampa et al.58 described a stimulatory effect of Myochrysine (GST) on the IL-10 secretion. The obtained results clearly showed that the transcription and expression of inflammatory-related cytokines TNF-α, IL1β, and HMGB1 (pro-inflammatory cytokines) and IL-10 and IL-1RA (anti-inflammatory cytokines) is influenced by the application of the tested gold(I) complexes. 4572
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
carrageenan-induced hind paw edema model, where the effect of the tested complexes on one of the symptoms of acute inflammation (i.e., the formation of edema) is investigated. Because of the structural similarity and intended purpose, we chose Auranofin as the primary standard of anti-inflammatory activity. According to the literature, it has been used in biological experiments in relatively wide range of dosage. In our experiments, we used it in the same dosages as the tested compounds, at 10 mg/kg in the form of the fine suspension in 25% DMSO (v/v in water for injections PhEur) applied intraperitonealy 30 min before the intraplantar injection of carrageenan. As a secondary standard, NSAID indomethacin was used in the dose of 5 mg/kg. The changes in the hind paw volume were continuously monitored plethysmometrically for 6 h after the carrageenan injection and assessed as percentual changes of the initial volume for every individual animal. The comprehensive overview of antiedematous activity profiles of the tested compounds are summarized in Figure 9.
Figure 7. Effects of the gold(I) complexes and the reference drugs Auranofin and prednisone on the LPS-induced HMGB1 gene expression (a) and HMGB1 secretion (b). The cells were pretreated with complexes 1, 3, and 6 (600 nM) and Auranofin (100 nM), prednisone (1 μM), or the vehicle (DMSO) only. After 1 h of incubation, the inflammatory response was induced by LPS (except for the control cells). Gene expression was measured 2, 4, and 8 h after LPS treatment. Secreted HMGB1 protein was detected in conditioned media by Western blot and immunodetection 24 h after LPS adding. The blot is a representative result of three independent experiments. The results of gene expression are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01), *** significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofin-treated cells (p < 0.05).
Figure 9. The time-dependent profile of antiedematous activity of the tested compounds.
Effect of the Au(I) Complexes on CarrageenanInduced Hind Paw Edema on Rats. On the basis of in vitro experiments, the in vivo tests of anti-inflammatory activity of the complexes 1, 3, and 6 were performed using the
The highest activity was found for 1, which was able to significantly decrease the volume of hind paw edema, even in the comparison with Auranofin (p < 0.001) and indomethacin (p < 0.01). The complex 6 was less effective than 1, and its activity profile correlated well with indomethacin, especially in the later stages of the experiment, but it showed significantly higher activity (p < 0.001) in comparison to Auranofin. Both complexes 1 and 6 showed antiedematous activity, comparable with the highly active derivatives of gold(I) complexes involving 3-(aryl)-2-sulfanylpropenoic acid59 and higher activity than other gold(I) and gold(III) complexes.60,61 The complex 3 showed only weak activity in this experiment. This could be due to its generally lower solubility, probably related to the presence of the chloro substituent at the benzyl moiety. With the intention to better understand and describe the effect of the tested compounds on the inflammation affected tissues, the sections from the animals were evaluated by cytological and histochemical methods. Cytological and Imunohistochemical Analyses of Animal Tissues. To assess the tissue consequences connected with the reduction of inflammation caused by the tested compounds after the intraplantar injection of carrageenan, the histopathological observations were made on the tissue sections obtained from the laboratory animals after the end of plethysmometric experiments. All animals were sacrificed by cervical dislocation, and immediately after that, the tissue samples were taken from the plantar area of hind paws.
Figure 8. Effects of the gold(I) complexes and the reference drugs Auranofin and prednisone on the LPS-induced IL-10 gene expression. The cells were pretreated with complexes 1, 3, and 6 (600 nM) and Auranofin (100 nM), prednisone (1 μM), or the vehicle (DMSO) only. After 1 h of incubation, the inflammatory response was induced by LPS (except for the control cells). The results are expressed as means ± SE for three independent experiments. AU = arbitrary unit. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01), *** significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofintreated cells (p < 0.05). 4573
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
Different histopathological changes in tissues, stained by the standard H&E-staining, were evaluated semiquantitatively using the five-point scale from 0 (no changes observed) up to 5 (maximum of changes). In addition to the classical histological investigation, the immunohistochemical detection of apoptosis (caspase 3), TNF-α, interleukin 6 (IL-6), and selectin E (CD62E) was performed. The results of the immunohistochemical IL-6 detection were due to the tissue nature being disputable because no differences were observed between the samples and control, and therefore they are not listed in the total evaluation. The results of histological and histoimmunochemical investigations are summarized in Table 1. Table 1. Semiquantitative Evaluations of Different Manifestations of Inflammation in Histological and Histochemical Samples Obtained from Animals Treated with Complexes 1, 3, and 6, and the reference drugs Auranofin and indomethacin 1 Inflammation PNM infiltration 0 into epidermis infiltration PMN 0 dermis infiltration PMN 2 hypodermis total infiltration 2 PMN extracellular edema 1 intracellular edema 0 vasodilatation 2 hemorrhage 1 disruption score 8 Immunohistochemistry apoptosis (caspase 0 3) TNF-α 1 CD 62 E 3
Auranofin indomethacin
25% DMSO
3
6
0
0
0
0
0
0
0
0
0
1
1
2
2
2
3
1
2
2
2
4
4 1 2 1 10
1 0 3 3 11
2 0 1 1 8
2 0 2 0 8
4 0 3 1 16
0
0
0
0
0
1 3
3 3
1 3
4 2
Figure 10. Histologic sections of the hind paw, stained with hematoxylin−eosin (200× magnification): tissue exposed to 25% DMSO (control; a) with the massive acute inflammatory reaction in the transition from dermis to hypodermis with a massive infiltrate of neurophils (PMN); tissue exposed to Auranofin (b) and 1 (c) with the acute inflammatory reaction in the hypodermis with a slight PMN infiltrate. 1, PMN infiltrate; 2, arteriola; 3, extracellular edema; 4, dilated veins with hemostasis; 5, vein with PMN presence in the bloodstream and hemostasis; 6, collagen connective tissue; 7, slight hemorrhage characterized by the presence of erythrocytes in interstitium; 8, muscle fiber.
The inflammation infiltrate contained mainly neutrophils (polymorphonuclear cells, PMNs), vasodilatation with the hemostasis and presence of leukocytes in bloodstream, and extracellular edema. These changes provided evidence of the acute inflammation. Table 1 shows significant presence of PMNs and edematous changes in hypodermis. Total histopathological changes are qualified by the disruption score. The highest disruption scores were found in tissues exposed to 25% DMSO (Figure 10a), which was used as a vehicle for the preparation of all the tested samples, with a high amount of PMNs in hypodermis and lower in dermis. The PMN distribution was mainly diffuse in dermis in the area near the veins, and in hypodermis, a massive infiltration (more than 1000 cells in microscope field of view, 200× magnification) was found in thin collagen connective tissue. Immunohistochemical TNF-α detection was significantly positive, which correlates with the histological results. The selectin E (CD62E) positiveness was less conclusive due to the nature of the tissues under study. The second highest disruption score was detected in tissues exposed to 6. The inflammatory infiltrate was significantly lower, which corresponds with the TNF-α positiveness. However, significant venous dilatation (characterized by the relaxation of venous epithelium, often accompanied by venostasis) and hemorrhage significantly
increased the score from the histopathological point of view. The tissue exposed to 3 can be characterized by the third highest disruption score. The inflammatory infiltrate was the lowest out of all the examined tissues; nevertheless, significant extracellular edema was present, together with a slight hydrophic degeneration of other cells (intracellular edema, characterized by small inclusions in cytoplasm of the cells). The disruption scores of the tissue exposed to Auranofin (Figure 10b), indomethacin, and 1 (Figure 10c) can be considered as identical. All the substances significantly decreased the inflammatory reaction. Apoptotic changes were not found in any of the tested samples. Interactions of 1−3 with a Mixture of Cysteine and Glutathione by Mass Spectrometry. Because of their 4574
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
properties (as soft Lewis acids), Au(I) ions prefer to form the strong coordination bonds with soft Lewis base ligands, such as thiolato or selenolato ions, or phosphine derivatives, while the latter ones form the most stable bonds. Therefore, in the biologically relevant environments (such as blood or serum), the Au(I) complexes tend to bind to sulfhydryl-containing substances, such as amino acid cysteine (Cys), or small proteins (such as glutathione, GSH) and in particular with high molecular proteins (such as albumin or globulins62) with the ligand exchange mechanism. In the case of Auranofin the exchange occurs for S-ligand of tetraacetylthioglucose,63 and in the next step the triethylphosphine (PEt3) ligand is substituted by another thiolato ligand while oxidized to triethylphosphineoxide (ref 64 and the references therein). The exchange of Nligands for S-ligands occurs relatively quickly (within 20 min when interacting with albumin and globulins in the blood),65 the P-ligand exchange occurs much more slowly, and it seems that in this mechanism the cooperative effects of adjacent thiolato or selenolato ligands in the neighborhood of interaction site play an important role, which is interpreted as one of the molecular mechanisms of incorporation of gold into the active site of selenium-containing flavoreductases such as thioredoxin reductase.66 In the scope of our work, we have verified the expected behavior of complexes 1−3 (applied in the concentration of 15 μM, corresponding to the highest therapeutic blood levels of gold during chrysotherapy67) in biologically relevant conditions68 using a mixture of cysteine (at 290 μM concentration) and reduced glutathione (at 6 μM concentration). On the basis of the results of ESI-MS experiments, we confirmed that the complexes 1−3 react with sulfhydryl-containing substances in a concentration dependent manner so that N-ligands of 6benzylaminopurine derivatives are substituted by cysteine, which was confirmed by the emergence of ion 580.1 m/z, corresponding to the [Cys-S-Au-PPh3]+ intermediate (see Figure 11 and Supporting Information, Figure S6). In biologically relevant concentrations of both low molecular sulfhydryl-containing compounds, the intermediate with glutathione, corresponding to [GS-Au-PPh3]+ and the mass of 766.0 m/z, reached the negligible intensity below the LOD (limit of detection, with signal-to-noise ratio S/N < 3).
Figure 11. ESI-MS spectra of (a; black line) the mixture of physiological levels of cysteine (290 μM) and glutathione (6 μM) in 50 mM ammonium acetate in water (b; green line) the interacting system containing cysteine (290 μM) + glutathione (6 μM) + complex 3 (15 μM) in the mixture of 50 mM ammonium acetate and methanol (1:1, v/v) immediately after preparation; (c; red line) the interacting system containing cysteine (290 μM) + glutathione (6 μM) + complex 3 (15 μM) in the mixture of 50 mM ammonium acetate and methanol (1:1, v/v) 20 min after preparation, both showing the appearance of peak at 580.08 m/z, corresponding to ion [Cys−S−Au-PPh3]+; (d; blue line) complex 3 solution in methanol.
managing difficult to treat inflammatory diseases. On the basis of such promising findings, the mechanism of their action and applicability as potential drugs will be further investigated.
■
EXPERIMENTAL SECTION
Chemistry. Materials and General Methods. The starting materials such as AuCl3, NaOH, and solvents were obtained from the commercial sources (Sigma−Sigma-Aldrich Co., Acros Organics Co., Lachema Co., and Fluka Co.) and used without further purification. The 6-benzylaminopurine derivatives (HL1−HL8) (see Supporting Information, Scheme S1) were synthesized according to the previously published procedure.69 The purity (≥95%) of the final products (complexes 1−8) was confirmed by elemental analysis (CHN), with the experimental values differing less than 0.4% from the calculated ones, and by 1H and 13C NMR spectroscopy. Elemental analyses (C, H, N) were performed on a Flash 2000 CHNO-S analyzer (Thermo Scientific). Conductivity measurements of 10−3 M methanol solutions of the prepared complexes were carried out using a conductometer 340i/SET (WTW) at 25 °C. IR spectra were recorded on a Nexus 670 FT-IR spectrometer (ThermoNicolet) using KBr pellets (400−4000 cm−1) and the Nujol technique (150− 600 cm−1). A Raman spectrometer Nicolet NXR 9650 equipped with the liquid nitrogen cooled NXE Genie germanium detector (ThermoNicolet) was used to record Raman spectra of all the complexes (150−4000 cm−1). 1H and 13C NMR spectra of the DMFd7 solutions were measured on a Varian 400 MHz NMR spectrometer at 300 K. Tetramethylsilane (TMS) was used as an internal reference standard during 1H and 13C NMR experiments. Mass spectra of the methanol solutions of the complexes were obtained by an LCQ Fleet ion trap mass spectrometer by the positive mode electrospray ionization (ESI+) technique (Thermo Scientific). The theoretical values were calculated by the QualBrowser software (version 2.0.7, Thermo Fisher Scientific). The electrospray-ionization mass spectrometry interaction experiments were performed on an Agilent HP1100 LC-MSD VL ion-trap mass spectrometer in positive ionization mode. Thermogravimetric (TG) and differential thermal analyses (DTA) were performed using a thermal analyzer Exstar TG/ DTA 6200 (Seiko Instruments Inc.) in dynamic air conditions (150
CONCLUSIONS The results indicated that the prepared gold(I) complexes of the type [Au(Ln)(PPh3)], involving 6-benzylaminopurine derivatives (HLn), exhibit a strong ability to reduce the production of pro-inflammatory cytokines TNF-α, IL-1β, and HMGB1 without influence on the secretion of anti-inflammatory cytokine IL-1RA in LPS-activated macrophages. Moreover, complexes 1 and 6 showed that they could significantly influence the formation of edema caused by the application of polysaccharide λ-carrageenan in vivo. All the tested gold(I) complexes exhibit similar or better effects as compared to Auranofin in equitoxic doses but lower cytotoxicity than Auranofin. The electrospray-ionization mass spectrometry experiments, under physiologically relevant conditions mimicking the interactions of the Au complexes with a mixture of cysteine and glutathione, confirmed that the complexes 1−3 react with sulfhydryl-containing substances in concentrationdependent manner, i.e., the 6-benzylaminopurine derivatives as ligands are substituted by cysteine, thus acting as pro-drugs, similarly to Auranofin. The obtained results entitle us to believe that the gold(I) complexes can become effective agents for
■
4575
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
mL min−1) between room temperature (∼25 °C) and 900 °C (gradient 2.5 °C min−1). Synthesis of [Au(L1)(PPh3)]·H2O (1), [Au(L2)(PPh3)] (2), [Au(L3)(PPh3)] (3), [Au(L4)(PPh3)] (4), [Au(L5)(PPh3)]·1.5H2O (5), [Au(L6)(PPh3)] (6), [Au(L7)(PPh3)] (7), [Au(L8)(PPh3)] (8). The amount of 1 mmol of [AuCl(PPh3)]70,71 dissolved in acetone (20 mL) was added to a solution of 1 mmol of the appropriate 6-benzylaminopurine derivative (HL1−HL8) in acetone (70 mL). Aqueous 1 M NaOH (1 mL) was added to this mixture. NaCl formed during 4 h of stirring, and it was filtered off. The obtained colorless clear filtrate was evaporated to dryness at 50 °C. The residue was dissolved in 30 mL of benzene. The clear solution was added dropwise into 250 mL of hexane. White precipitates of 1−8 were filtered off, washed with a small amount of ethanol and acetone, and dried at 40 °C under an infrared lamp. Complex 1: 1H NMR (400.00 MHz, DMF-d7, ppm): δ (SiMe4) 8.15 (bs, C2H, 1H), 7.96 (bs, C8H, 1H), 7.77 (m, C2H, C6H, 6H, PPh3), 7.70 (m, C3H, C4H, C5H, 9H, PPh3), 7.55 (bs, N6H, 1H), 7.42 (tt, 7.9, 1.6, C14H, 1H), 7.38 (qq, 7.5, 1.6, C15H, 1H), 7.13 (d, 9.2, C12H, 1H), 7.03 (t, 7.5, C13H, 1H), 5.09 (bs, C9H, 2H). 13C NMR (75.43 MHz, DMF-d7, ppm): δ (SiMe4) 162.05, 159.14 (C11), 156.39 (C6), 151.53 (C2), 148.52 (C8), 145.25 (C4), 135.06, 134.92 (C2, C6, PPh3), 131.09, 130.06 (C14), 130.67, 130.47 (C15), 130.46, 130.34 (C3, C5, PPh3), 129.21 (C10), 129.82, 129.28 (C1, PPh3), 129.12, 129.04 (C4, PPh3), 124.02, 124.00 (C13), 120.96 (C5), 114.64, 114.43 (C12), 39.15 (C9). ESI+ mass spectra (methanol, m/z) 244 (calcd 243), 702 (701). Yield: 55%. Anal. Calcd for C30H26N5OFPAu: C, 50.0; H, 3.6; N, 9.7. Found: C, 50.3; H, 3.8; N, 9.9%. Complex 3: 1H NMR (400.00 MHz, DMF-d7, ppm): δ (SiMe4) 8.13 (s, C2H, 1H), 7.93 (s, C8H, 1H), 7.77 (m, C2H, C6H, 6H, PPh3), 7.70 (m, C3H, C4H, C5H, 9H, PPh3), 7.41 (d, 8.5, C12H, C15H, 2H), 7.36 (d, 8.5, C13H, C14H, 2H), 5.14 (bs, C9H, 2H). 13C NMR (75.43 MHz, DMF-d7, ppm): δ (SiMe4) 156.38 (C6), 151.51 (C2), 148.44 (C8), 141.10 (C4), 139.10 (C10), 135.06, 134.92 (C2, C6, PPh3), 131.11 (C15), 131.07 (C12), 130.47, 130.35 (C3, C5, PPh3), 129.86 (C14), 129.78, 129.16 (C1, PPh3), 129.40, 129.27 (C4, PPh3), 120.92 (C5), 43.86 (C9). ESI+ mass spectra (methanol, m/z) 259 (calcd 258), 718 (717), 740 (740). Yield: 60%. Anal. Calcd for C30H24N5ClPAu: C, 50.1; H, 3.3; N, 9.7. Found: C, 50.0; H, 3.3; N, 9.9%. Complex 6: 1H NMR (400.00 MHz, DMF-d7, ppm): δ (SiMe4) 8.14 (s, C2H, 1H), 7.92 (bs, C8H, 1H), 7.78, (m, C2H, C6H, 6H, PPh3), 7.70 (m, C3H, C4H, C5H, 9H, PPh3), 7.50 (bs, N6H, 1H), 7.21 (s, C11H, 1H), 7.20 (m, C15H, 1H), 7.11 (t, 7.5, C14H, 1H), 7.01 (d, 7.5, C13H, 1H), 4.97 (bs, C9H, 2H). 13C NMR (75.43 MHz, DMF-d7, ppm): δ (SiMe4) 154.06 (C6), 151.57 (C2), 148.31 (C8), 145.15 (C4), 139.51 (C10), 139.25 (C12), 135.05, 134.92 (C2, C6, PPh3), 130.47, 130.35 (C3, C5, PPh3), 129.38, 129.26 (C4, PPh3), 128.14 (C14), 128.04 (C13), 128.02 (C11), 125.67 (C15), 121.16 (C5), 44.58 (C9). ESI+ mass spectra (methanol, m/z) 239 (calcd 238), 698 (697). Yield: 55%. Anal. Calcd for C31H27N5PAu: C, 53.3; H, 3.8; N, 10.0. Found: C, 53.6; H, 3.8; N, 10.1%. The results of NMR, mass spectrometry and elemental analyses (for 2, 4, 5, 7, and 8), IR and Raman spectroscopy, and conductivity measurements (for 1−8) are given in Supporting Information. Biological Activity Testing. Chemicals and Biochemicals. The RPMI 1640 medium and penicillin−streptomycin mixture were purchased from Lonza (Verviers, Belgium). Phosphate-buffered saline (PBS), fetal bovine serum (FBS), phorbol myristate acetate (PMA), prednisone (98%≤), Auranofin (98%≤), erythrosin B, and Escherichia coli 0111:B4 lipopolysaccharide (LPS) were purchased from SigmaAldrich (Steinheim, Germany). Cell proliferation reagent WST-1 was obtained from Roche (Mannheim, Germany). A RealTime ready cell lysis kit (Roche, Mannheim, Germany) served for isolation of RNA from cells and Transcriptor Universal cDNA Master (Roche, Mannheim, Germany) was used for reverse transcription of RNA to cDNA. Specific primers and probes (gene expression assays) for polymerase chain reactions (PCRs) were obtained from Applied Biosystems (Foster City, CA, USA). The following assays were chosen
for the quantification of gene expression: Hs00174128_m1 for TNF-α, Hs00961622_m1 for IL-10, Hs01555410_m1 for IL-1β, Hs00893626_m1 for IL-1RA, Hs01590761_g1 for HMGB1, and 4326315E for β-actin, which served as an internal control of gene expression. Quantitative PCR (qPCR) was performed with Fast Start Universal Probe Master (Roche, Mannheim, Germany). Instant ELISA kits (eBioscience, Vienna, Austria) were used to evaluate the production of TNF-α and IL-1β. Cytoscreen kit (BioSource Europe SA, Nivelles, Belgium) was used to detect the IL-1RA cytokine by the enzyme linked immunosorbent assay (ELISA) method. The supported nitrocellulose membrane 0.2 μm (Bio-Rad, Hercules, CA, USA) and albumin bovine fraction V (pH 7) (BSA) (Serva, Heidelberg, Germany) were used for Western blot. Rabbit polyclonal AntiHMGB1 and goat polyclonal antirabbit IgG (with conjugated peroxidase) antibodies (Sigma-Aldrich, Saint Louis, MO, USA) were applied for immunodetection. Conjugated peroxidase was detected by Opti-4CN substrate kit (Bio-Rad, Hercules, CA, USA). Maintenance and Preparation of Macrophages. For the measurement of biological activity, we used the human monocytic leukemia cell line THP-1 (ECACC, Salisbury, UK). The cells were cultivated at 37 °C in the RPMI 1640 medium supplemented with 2 mM L-glutamine, 10% FBS, 100 U/mL of penicillin, and 100 μg/mL of streptomycin in a humidified atmosphere containing 5% CO2. The stabilized cells (3rd−15th passage) were split into microtitration plates to get a concentration of 100000 cells/mL, and the differentiation to macrophages was induced by a phorbol myristate acetate (PMA) as we described previously.72 Cytotoxicity Testing. The THP-1 cells (floating monocytes, 500000 cells/mL) were incubated in 100 μL of the serum-free RPMI 1640 medium and seeded into 96-well plates in triplicate at 37 °C. Measurements were taken 24 h after the treatment with 0.024, 0.12, 0.6, 1.56, 3.13, 6.25, 12.5, 25, or 50 μM tested compounds dissolved at dimethylsulfoxide (DMSO). Viability was measured by WST-1 test according to manufacturer’s manual. The amount of created formazan (which correlates to the number of metabolically active cells in the culture) was calculated as a percentage of the control cells, which were treated only with DMSO and was set up as 100%. The results of the WST-1 test were confirmed by the erythrosine-exclusion test, as was described in a previous paper.73 The cytotoxic LD50 concentrations of the tested compounds were determined by the data from the equation generated by the KURV+ Version 4.4b software (Conrad Button Software, Arlington, WA, USA). Prednisone was used for in vitro experiments as a control, while indomethacin was used in in vivo studies. This experiment design was chosen because of the fact that indomethacin shows negligible effect on the cytokine expression (it is a typical cyclooxygenase inhibitor).73 On the other hand, steroid drugs, such as prednisone or dexamethasone, are able to affect gene expression.74 Drug Treatment and Induction of Inflammatory Response. Differentiated macrophages were pretreated for 1 h with 100 or 600 nM solutions of 1, 3, or 6, 100 nM solution of Auranofin, 1.0 μM solution of prednisone dissolved in DMSO (the final DMSO concentration was 0.1%), and with 0.1% DMSO solution itself (vehicle); the given concentrations of the tested compounds lack the cytotoxic effect (cell viability >94%). The inflammatory response was triggered by adding 1.0 μg/mL lipopolysaccharide (LPS) dissolved in water to pretreated macrophages, control cells were without the LPS treatment. Each experiment was repeated three times. RNA Isolation and Gene Expression Evaluation. To evaluate the expression of TNF-α, IL-1β, IL-1RA, IL-10, HMGB1, and β-actin mRNA, the total RNA was isolated directly from the cells in cultivation plates using a RealTime ready cell lysis kit according to the manufacturer’s instructions. The concentration and purity of the RNA were determined by UV spectrophotometry. The gene expression was quantified by two-step reverse-transcription quantitative (real-time) PCR (RT-qPCR). The reverse transcription step was performed by Transcriptor Universal cDNA Master using cell lysate as a template. The reaction consists of three steps: (1) primer annealing 29 °C for 10 min and (2) reverse transcription 55 °C for 10 min and (3) transcriptase inactivation 85 °C 4576
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
for 5 min. A Fast Start Universal Probe Master and gene expression assays were used for qPCR. These assays contain specific primers and TaqMan probes that bind to an exon−exon junction to avoid DNA contamination. The parameters for the qPCR work were adjusted according to the manufacturer’s recommendations: 50 °C for 2 min, then 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. The results were normalized to the amount of ROX reference dye, and the change in gene expression was determined by the ΔΔCT method.75 Transcription of the control cells was set as 1, and other experimental groups were multiples of this value. Evaluation of Cytokine Secretion by ELISA. Macrophages, which were pretreated with the tested compounds for 1 h, were incubated with LPS for next 24 h. After this period, the medium was collected and the concentration of TNF-α, IL-1β, and IL-1RA was measured by either Instant ELISA kit or Cytoscreen kit according to the manufactures’ manual. Determination of HMGB1 Release by Western Blot. Media from LPS-stimulated macrophages were obtained similarly as for ELISA. They were lyophilized, and the resting material was resuspended in the 1/10 of the original volume. The same volumes of samples were resolved in the 15% polyacrylamide gel. Then they were electophoretically transferred on the nitrocellulose membranes, which were subsequently blocked by 5% BSA dissolved in TBST buffer [150 mM NaCl, 10 mM Tris base, 0.1% (v/v) Tween-20]. The membranes were incubated with the primary anti-HMGB1 antibody at the concentration of 1:500 at 4 °C 16 h. After washing, the secondary antirabbit IgG antibody diluted 1:2000 was applied on the membranes and incubated for 1 h at laboratory temperature (∼23 °C). The amount of the bound secondary antibody was colorimetrically detected by Opti-4CN kit according to the manufacturer’s manual. Animals. Wistar-SPF (6−8 weeks male) rats were obtained from the AnLab, Ltd., Prague. The animals were kept in plexiglass cages at the constant temperature of 22 ± 1 °C and relative humidity of 55 ± 5% for at least 1 week before the experiment. They were given food and water ad libitum. All experimental procedures were performed according to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. In addition, all tests were conducted under the guidelines of the International Association for the Study of Pain.76 After a 1-week adaptation period, male Wistar-SPF rats (200−250 g) were randomly assigned to three groups (n = 10) of the animals in the study. The control group received 25% DMSO (v/v in water for injections PhEur, intraperitoneal; ip). The other three groups include a carrageenan-treated, an Auranofin positive control (Auranofin + carrageenan), and indomethacin positive control groups (indomethacin + carrageenan). Carrageenan-Induced Hind Paw Edema. The carrageenaninduced hind paw edema model was used for the determination of anti-inflammatory activity.77 Animals were ip pretreated with complexes 1, 3, and 6 (10 mg/kg), Auranofin (10 mg/kg), indomethacin (5 mg/kg), or 25% DMSO (v/v in water for injections PhEur) 30 min prior to the injection of 1% Carr (50 μL) into the plantar side of right hind paws of the rats. The paw volume was measured immediately after carrageenan injection and during the next 6 h after the administration of the edematogenic agent using a plethysmometer (model 7159, Ugo Basile, Varese, Italy). The degree of swelling induced was evaluated by the percentage of change of the volume of the right hind paw after carrageenan treatment from the volume of the right hind paw before carrageenan treatment. Auranofin and indomethacin were used as positive controls. After 6 h, the animals were sacrificed and the carrageenan-induced edema feet were dissected for cytological and histochemical evaluation. Histological Examination. For the histological examination, biopsies of paws were taken 6 h following the intraplantar injection of carrageenan. The tissue slices were fixed in 4% buffered formaldehyde for 1 week at room temperature, dehydrated by graded ethanol, and embedded in paraffin (Histowax). The sections (thickness 5 μm) were deparaffinized with xylene and stained with hematoxylin and eosin (H&E) stain and immunohistochemistry detection apoptosis: caspase 3 (AbCam), TNF-α (AbCam), and CD62E (AbCam). All samples were observed and photographed with
BX-40 Olympus microscopy. Every 3−5 tissue slices were randomly chosen from carrageenan, indomethacin, Auranofin, and studied gold(I) complexes groups. The histological examination of these tissue slices revealed an excessive inflammatory response with massive infiltration of neutrophils [polymorphonuclear leukocytes (PMNs)] by microscopy. Statistical Analysis. All experiments were performed in three independent experiments, and results are presented as mean values, with error bars representing the standard deviation (SE) of the mean. A one-way ANOVA test was used for statistical analysis, followed by a Tuckey’s posthoc test for multiple comparisons. A value of p < 0.05 was considered to be statistically significant. GraphPad Prism 5.02 (GraphPad Software Inc., San Diego, CA) was used to perform the analysis.
■
ASSOCIATED CONTENT
S Supporting Information *
The structural formulas of the 6-benzylaminopurine derivatives HL1−HL8, TG/DTA curves of the selected complexes, the figure showing the ESI+ mass spectra of 1 and 3, a geometry of the complex [Au(L1)(PPh3)] (1) optimized at the B3LYP/ LACVP+** level, a plot showing the effects of 1, 3, 6 (applied at the concentration of 100 nM), Auranofin and vehicle (DMSO) on the LPS-induced TNF-α and IL-1β secretion, the ESI+ mass spectra of the interactions of 1 and 6 with a mixture of cysteine and glutathione, selected 1H and 13C NMR coordination shifts, the discussion of quantum chemical calculations and selected experimental data (the results of NMR, mass spectrometry and elemental analyses (for 2, 4, 5, 7, and 8), IR and Raman spectroscopy and conductivity measurement (for 1−8)). This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +420 585 634 352. Fax: +420 585 634 954. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS This research was supported by the Ministry of Education, Youth and Sports of the Czech Republic (grant no. MSM6198959218), the Operational Program Research and Development for InnovationsEuropean Regional Development Fund (CZ.1.05/2.1.00/03.0058), the Operational Program Education for CompetitivenessEuropean Social Fund (CZ.1.07/2.3.00/20.0017) by Palacký University (student project PrF_2011_014 and PrF_2012_009). We thank Dr. Igor Popa for performing NMR experiments, Pavla Richterová for performing CHN elemental analyses, Radka Novotná for IR and Raman spectra measurements, and Alena Klanicová for mass spectra measurements.
■
■
REFERENCES
(1) Davies, P. J. Plant Hormones; Springer: Dordrecht, 1997. (2) Meijer, L.; Borgne, A.; Mulner, O.; Chong, J. P. J.; Blow, J. J.; Inagaki, N.; Inagaki, M.; Delcros, J. G.; Moulinoux, J. P. Biochemical and cellular effects of roscovitine, a potent and selective inhibitor of the cyclin-dependent kinases cdc2, cdk2 and cdk5. Eur. J. Biochem. 1997, 243, 527−536. (3) Benson, C.; Kaye, S.; Workman, P.; Garret, M.; Walton, M.; de Bono, J. Clinical anticancer drug development: targeting the cyclindependent kinases. Br. J. Cancer 2005, 92, 7−12.
4577
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
(4) Trávníček, Z.; Matiková-Mal'arová, M.; Novotná, R.; Vančo, J.; Štěpánková, K.; Suchý, P. In vitro and in vivo biological activity screening of Ru(III) complexes involving 6-benzylaminopurine derivatives with higher pro-apoptotic activity than NAMI-A. J. Inorg. Biochem. 2011, 105, 937−948. (5) Trávníček, Z.; Štarha, P.; Popa, I.; Vrzal, R.; Dvořaḱ , Z. Roscovitine-based CDK inhibitors acting as N-donor ligands in the platinum(II) oxalato complexes: preparation, characterization and in vitro cytotoxicity. Eur. J. Med. Chem. 2010, 45, 4609−4614. (6) Vrzal, R.; Štarha, P.; Dvořaḱ , Z; Trávníček, Z. Evaluation of in vitro cytotoxicity and hepatotoxicity of platinum(II) and palladium(II) oxalato complexes with adenine derivatives as carrier ligands. J. Inorg. Biochem. 2010, 104, 1130−1132. (7) Štarha, P.; Trávníček, Z.; Herchel, R.; Popa, I.; Suchý, P.; Vančo, J. Dinuclear copper(II) complexes containing 6-(benzylamino)purines as bridging ligands: synthesis, characterization, and in vitro and in vivo antioxidant activities. J. Inorg. Biochem. 2010, 103, 432−440. (8) Gielen, M.; Tiekink, E. R. T. Metallotherapeutic Drugs and MetalBased Diagnostic Agents; John Wiley & Sons, Ltd.: New York, 2005. (9) Milacic, V.; Ping Dou, Q. The tumor proteasome as a novel target for gold(III) complexes: Implications for breast cancer therapy. Coord. Chem. Rev. 2009, 253, 1649−1660. (10) Ott, I. On the medicinal chemistry of gold complexes as anticancer drugs. Coord. Chem. Rev. 2009, 253, 1670−1681. (11) Nomiya, K.; Noguchi, R.; Ohsawa, K.; Tsuda, K. Synthesis, crystal structure and antimicrobial activities of two isomeric gold(I) complexes with nitrogen-containing heterocycle andtriphenylphosphine ligands, [Au(L)(PPh3)] (HL = pyrazole and imidazole). J. Inorg. Biochem. 2000, 78, 363−370. (12) Marques, L. L.; Oliveira, G. M.; Lang, E. S. New gold(I) and silver(I) complexes of sulfamethoxazole: synthesis, X-ray structural characterization and microbiological activities of triphenylphosphine(sulfamethoxazolato-N2)gold(I) and (sulfamethoxazolato)silver(I). Inorg. Chem. Commun. 2007, 10, 1083−1087. (13) Nomiya, K.; Yamamoto, S.; Noguchi, R. Ligand-exchangeability of 2-coordinate phosphinegold(I) complexes with AuSP and AuNP cores showing selective antimicrobial activities against Gram-positive bacteria. Crystal structures of [Au(2-Hmpa)(PPh3)] and [Au(6Hmna)(PPh3)] (2-H2mpa = 2-mercaptopropionic acid, 6-H2mna = 6mercaptonicotinic acid). J. Inorg. Biochem. 2003, 95, 208−220. (14) Barreiro, E.; Casas, J. S.; Couce, M. D.; Sánchez, A.; SánchezGonzalez, A.; Sordo, J.; Varela, J. M.; Vázquez López, E. M. Synthesis, structure and cytotoxicity of triphenylphosphinegold(I) sulfanylpropenoates. J. Inorg. Biochem. 2008, 102, 184−192. (15) Lewis, M. G.; DaFonseca, S.; Chomont, N.; Palamara, A. T.; Tardugno, M.; Mai, A.; Collins, M.; Wagner, W. L.; Yalley-Ogunro, J.; Greenhouse, J.; Chirullo, B.; Norelli, S.; Garaci, E.; Savarino, A. Gold drug auranofin restricts the viral reservoir in the monkey AIDS model and induces containment of viral load following ART suspension. AIDS 2011, 25, 1347−1356. (16) Berners-Price, S. J.; Filipovska, A. Gold compounds as therapeutic agents for human diseases. Metallomics 2011, 3, 863− 873 (and the references within). (17) Fonteh, P. N.; Keter, F. K.; Meyer, D. HIV therapeutic possibilities of gold compounds. BioMetals 2010, 23, 185−196. (18) Williams, H. J.; Ward, J. R.; Reading, J. C.; Brooks, R. H.; Clegg, D. O.; Skosey, J. L.; Weisman, M. H.; Willkens, R. F.; Singer, J. Z.; Alarcon, G. S.; Field, E. H.; Clements, P. J.; Russell, I. J.; Hochman, R. F.; Boumpas, D. T.; Marble, D. A. Comparison of Auranofin, methotreaxate, and the combination of both in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum. 1992, 35, 259−269. (19) Sigler, J. W.; Bluhm, G. B.; Duncan, H.; Sharp, J. T.; Ensign, D. C.; McCrum, W. R. Gold Salts in the Treatment of Rheumatoid Arthritis: A Double-Blind Study. Ann. Intern. Med. 1974, 80, 21−26. (20) Pope, J. E.; Hong, P.; Koehler, B. E. Prescribing trends in disease modifying, antirheumatic drugs for rheumatoid arthritis: a survey of practicing Canadian rheumatologists. J. Rheumatol. 2002, 29, 255−260.
(21) Benedek, T. G. The history of gold therapy for tuberculosis. J. Hist. Med. Allied Sci. 2004, 59, 50−89. (22) Kean, W. F.; Kean, I. R. L. Clinical Pharmacology of gold. Inflammopharmacology 2008, 16, 112−125. (23) Kean, W. F.; Hart, L.; Buchanan, W. W. Auranofin. Br. J. Rheumatol. 1997, 36, 560−572. (24) Eisler, R. Chrysotherapy: a synoptic review. Inflamm. Res. 2003, 52, 487−501. (25) Whitehouse, M. W.; Graham, G. G. Is local biotransformation the key to understanding the pharmacological activity of salicylates and gold drugs? Inflamm. Res. 1996, 45, 579−592. (26) Canumalla, A. J.; Al-Zamil, N.; Phillips, M.; Isab, A. A.; Shaw, C. F., III. Redox and ligand exchange reactions of potential gold(I) and gold(III)-cyanide metabolites under biomimetic conditions. J. Inorg. Biochem. 2001, 85, 67−76. (27) Graham, G. G.; Kettle, A. J. The activation of gold complexes by cyanide produced by polymorphonuclear leukocytes. The formation of aurocyanide by myeloperoxidase. Biochem. Pharmacol. 1998, 56, 307− 312. (28) Goebel, C.; Kubicka-Muranyi, M.; Tonn, T.; Gonzalez, J.; Gleichmann, E. Phagocytes render chemicals immunogenic: oxidation of gold(I) to the T cell-sensitizing gold(III) metabolite generated by mononuclear phagocytes. Arch. Toxicol. 1995, 69, 450−459. (29) Zou, J.; Guo, Z.; Parkinson, J. A.; Chen, Y.; Sadler, P. J. Gold(III)-induced oxidation of glycine: relevance to the toxic sideeffects of gold drugs. J. Inorg. Biochem. 1999, 74, 352−352. (30) Abraham, G. E.; Himmel, P. B. Management of rheumatoid arthritis: rationale for the use of colloidal metallic gold. J. Nutr. Environ. Med. 1997, 7, 295−305. (31) Walz, E. T.; Dimartino, M. J.; Griswold, D. E.; Intoccia, A. P.; Flanagan, T. L. Biologic actions and pharmacokinetic studies of auranofin. Am. J. Med. 1983, 75, 90−108. (32) Kean, W. F.; Kassam, Y. B.; Lock, C. J. L.; Buchanan, W. W.; Rischke, J.; Nablo, L. Antithrombin activity of gold sodium thiomalate. Clin. Pharmacol. Ther. 1984, 35, 627−632. (33) Gunatilleke, S. S.; Barrios, A. M. Inhibition of lysosomal cysteine proteases by a series of Au(I) complexes: a detailed mechanistic investigation. J. Med. Chem. 2006, 49, 3933−3937. (34) Janoff, A. Inhibition of human granulocyte elastase by gold sodium thiomalate. Biochem. Pharmacol. 1970, 19, 626−628. (35) Asperger, S.; Cetina-Cizmek, B. Metal complexes in tumour therapy. Acta Pharm. (Zagreb, Croatia) 1999, 49, 225−236. (36) Brown, D. H.; Smith, W. E. The chemistry of the gold drugs used in the treatment of rheumatoid arthritis. Chem. Soc. Rev. (London) 1980, 9, 217−240. (37) Sato, H.; Yamaguchi, M.; Shibasaki, T.; Ishii, T.; Bannai, S. Induction of stress proteins in mouse peritoneal macrophages by the antirheumatic agents gold sodium thiomalate and auranofin. Biochem. Pharmacol. 1995, 49, 1453−1457. (38) Heimburger, M.; Lerner, R.; Palmblad, J. Effects of antirheumatic drugs on adhesiveness of endothelial cells and neutrophils. Biochem. Pharmacol. 1998, 56, 1661−1669. (39) Rubbiani, R.; Kitanovic, I.; Alborzinia, H.; Can, S.; Kitanovic, A.; Onambrle, L. A.; Stefanopoulou, M.; Geldmacher, Y.; Sheldrick, W. S.; Wolber, G.; Prokop, A.; Wölfl, S.; Ott, I. Benzimidazol-2-ylidene Gold(I) Complexes Are Thioredoxin Reductase Inhibitors with Multiple Antitumor Properties. J. Med. Chem. 2010, 53, 8608−8618. (40) Yan, K.; Lok, C.-N.; Bierla, K.; Che, C.-M. Gold(I) complex of N,N′-disubstituted cyclic thiourea with in vitro and in vivo anticancer propertiespotent tight-binding inhibition of thioredoxin reductase. Chem. Commun. 2010, 46, 7691−7693. (41) Magherini, F.; Modesti, A.; Bini, L.; Puglia, M.; Landini, I.; Nobili, S.; Mini, E.; Cinellu, M. A.; Gabbiani, C.; Messori, L. Exploring the biochemical mechanisms of cytotoxic gold compounds: a proteomic study. J. Biol. Inorg. Chem. 2010, 15, 573−582. (42) Merchant, B. Gold, the noble metal and the paradoxes of its toxicology. Biologicals 1998, 26, 49−59. (43) Gardea-Torresdey, J. L.; Tiemann, K. J.; Gamez, G.; Dokken, K.; Cano-Aguilera, I.; Furenlid, L. R.; Renner, M. W. Reduction and 4578
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
Journal of Medicinal Chemistry
Article
triethylphosphinegold(I) sulfanylpropenoates of the type [(AuPEt3)2xspa] [H2xspa = 3-(aryl)-2-sulfanylpropenoic acid]: an (H2O)6 cluster in the lattice of the complexes [(AuPEt3)2xspa]·3H2O. Inorg. Chem. 2008, 47, 6262−6272. (60) Coppi, G.; Borella, F.; Gatti, M. T.; Comini, A.; Dall’Asta, L. Synthesis, antiinflammatory and antiarthritic properties of a new tiopronine gold derivative. Boll. Chim. Farm. 1989, 128, 22−24. (61) Girard, G. R.; Hill, D. T.; Dimartino, M. J. Evaluation of some tetraalkylammonium gold(I) and gold(III) aurate salts for oral antiinflammatory and antiarthrithic activity. Inorg. Chim. Acta 1989, 166, 141−146. (62) Shaw, C. F., III; Coffer, M. T.; Klingbeil, J.; Mirabelli, C. K. Application of phosphorus-31 NMR chemical shift: gold affinity correlation to hemoglobin−gold binding and the first interprotein gold transfer reaction. J. Am. Chem. Soc. 1988, 110, 729−734. (63) Iqbal, M. S.; Saeed, M.; Taqi, S. G. Erythrocyte Membrane Gold Levels After Treatment with Auranofin and Sodium Aurothiomalate. Biol. Trace Elem. Res. 2008, 126, 56−64. (64) Isab, A. A.; Ahmad, S. Applications of NMR spectroscopy in understanding the gold biochemistry. Spectroscopy 2006, 20, 109−123. (65) Iqbal, M. S.; Taqi, S. G.; Arif, M.; Wasim, M.; Sher, M. In Vitro Distribution of Gold in Serum Proteins after Incubation of Sodium Aurothiomalate and Auranofin with Human Blood and its Pharmacological Significance. Biol. Trace Elem. Res. 2009, 130, 204− 209. (66) Saccoccia, F.; Angelucci, F.; Boumis, G.; Brunori, M.; Miele, A. E.; Williams, D. L.; Bellelli, A. On the mechanism and rate of gold incorporation into thiol-dependent flavoreductases. J. Inorg. Biochem. 2012, 108, 105−111. (67) Lewis, D.; Capell, H. A.; McNeil, C. J.; Iqbal, M. S.; Brown, D. H.; Smith, W. E. Gold levels produced by treatment with auranofin and sodium aurothiomalate. Ann. Rheum. Dis. 1983, 42, 566−570. (68) Salemi, G.; Gueli, M. C.; D’Amelio, M.; Saia, V.; Mangiapane, P.; Aridon, P.; Ragonese, P.; Lupo, I. Blood levels of homocysteine, cysteine, glutathione, folic acid, and vitamin B12 in the acute phase of atherothrombotic stroke. Neurol. Sci. 2009, 30, 361−364. (69) Kuhnle, J. A.; Fuller, G.; Corse, J.; Mackey, B. E. Anti-senescent activity of natural cytokinins. Physiol. Plant. 1977, 41, 14−21. (70) Mann, F. G.; Wells, A. F.; Purdie, D. The constitution of complex metalic salts: Part IV. The constitution of the phosphine and arsine derivatives of silver and aurous halides. The coordination of the coordinated argentous and aurous complex. J. Chem. Soc. 1937, 1828− 1836. (71) Bruce, M. I.; Nicholson, B. K.; Bin Shawkataly, O. Synthesis of gold-containing mixed-metal cluster complexes. Inorg. Synth. 1989, 26, 324−328. (72) Hošek, J.; Závalová, V.; Šmejkal, K.; Bartoš, M. Effect of Diplacone on LPS-Induced Inflammatory Gene Expression in Macrophages. Folia Biol. (Krakow, Pol.) 2010, 56, 124−130. (73) Hošek, J.; Bartoš, M.; Chudík, S.; Dall’acqua, S.; Innocenti, G.; Kartal, M.; Kokoška, L.; Kollar, P.; Kutil, Z.; Landa, P.; Marek, R.; Závalová, V.; Ž emlička, M.; Š mejkal, K. Natural Compound Cudraflavone B Shows Promising Anti-inflammatory Properties in Vitro. J. Nat. Prod. 2011, 74, 614−619. (74) Bahadir, O; Ç itoglu, G. S.; Ozbek, H; Dall’Acqua, S; Hošek, J; Smejkal, K. Hepatoprotective and TNF-α inhibitory activity of Zosima absinthifolia extracts and coumarins. Fitoterapia 2011, 82, 454−459. (75) Livak, K. J.; Schmittgen, T. D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(T)(−Delta Delta C) method. Methods 2001, 25, 402−408. (76) Zimmermann, M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983, 16, 109−110. (77) Chang, H. Y.; Sheu, M. J.; Yang, C. H.; Leu, Z. C.; Chang, Y. S.; Peng, W. H.; Huang, S. S.; Huang, G. J. Analgesic effects and the mechanisms of anti-inflammation of hispolon in mice. Evidence-Based Complementary Altern. Med. 2011, 2011, Article ID 478246 DOI: 10.1093/ecam/nep027.
Accumulation of Gold(III) by Medicago sativa Alfalfa Biomass: X-ray Absorption Spectroscopy, pH, and Temperature Dependence. Environ. Sci. Technol. 2000, 34, 4392−4396. (44) Abraham, G. E.; Himmel, P. B. Management of rheumatoid arthritis: rationale for the use of colloidal metallic gold. J. Nutr. Environ. Med. 1997, 7, 295−305. (45) Rosopulos, Y.; Nagel, U.; Beck, W. Metal complexes with biologically important ligands. 34. Allyl-palladium and triphenylphosphane gold(I) complexes with nucleobases and nucleosides. Chem. Ber. 1985, 118, 931−942. (46) Tiekink, E. R. T.; Kurucsev, T.; Hoskins, B.F. J. X-ray structure and UV spectroscopic studies of (adeninato-N9)triphenylphosphinegold(I). Crystallogr. Spectrosc. Res. 1989, 19, 823− 839. (47) Becke, A. D. A new mixing of Hartree−Fock and local density functional theories. J. Chem. Phys. 1993, 98, 1372−1377. (48) Rassolov, V.; Pople, J. A.; Ratner, M.; Redfern, P. C.; Curtiss, L. A. 6-31G* Basis Set for Third-Row Atoms. J. Comput. Chem. 2001, 22, 976−984. (49) Shao, V.; Molnar, L. F.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S. T.; Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.; O’Neill, D. P.; DiStasio, R. A., Jr.; Lochan, R. C.; Wang, T.; Beran, G. J. O.; Besley, N. A.; Herbert, J. M.; Lin, C. Y.; Van Voorhis, T.; Chien, S. H.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.; Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.; Dachsel, H.; Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney, S. R.; Heyden, A.; Hirata, S.; Hsu, C. P.; Kedziora, G.; Khalliulin, R. Z.; Klunzinger, P.; Lee, A. M.; Lee, M. S.; Liang, W. Z.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.; Pieniazek, P. A.; Rhee, Y. M.; Ritchie, J.; Rosta, E.; Sherrill, C. D.; Simmonett, A. C.; Subotnik, J. E.; Woodcock, H. L., III; Zhang, W.; Bell, A. T.; Chakraborty, A. K.; Chipman, D. M.; Keil, F. J.; Warshel, A.; Hehre, W. J.; Schaefer, H. F.; Kong, J.; Krylov, A. I.; Gill, P. M. W.; Head-Gordon, M. Advances in methods and algorithms in a modern quantum chemistry program package. Phys. Chem. Chem. Phys. 2006, 8, 3172−3191. (50) Barciszewski, J.; Massino, F.; Clark, B. F. Kinetina multiactive molecule. Int. J. Biol. Macromol. 2007, 40, 182−192. (51) Han, S.; Kim, K.; Kim, H.; Kwon, J.; Lee, Y. H.; Lee, C. K.; Song, Y.; Lee, S. J.; Ha, N.; Kim, K. Auranofin inhibits overproduction of pro-inflammatory cytokines, cyclooxygenase expression and PGE2 production in macrophages. Arch. Pharm. Res. 2008, 31, 67−74. (52) Jeon, K. I.; Jeong, J. Y.; Jue, D. M. Thiol-reactive metal compounds inhibit NF-kappa B activation by blocking I kappa B kinase. J. Immunol. 2000, 164, 5981−5989. (53) Seitz, M.; Valbracht, J.; Quach, J.; Lotz, M. Gold sodium thiomalate and chloroquine inhibit cytokine production in monocytic THP-1 cells through distinct transcriptional and posttranslational mechanisms. J. Clin. Immunol. 2003, 23, 477−484. (54) Arend, W. P. The balance between IL-1 and IL-1Ra in disease. Cytokine Growth Factor Rev. 2002, 13, 323−340. (55) Zetterstrom, C. K.; Jiang, W.; Wahamaa, H.; Ostberg, T.; Aveberger, A. C.; Schierbeck, H.; Lotze, M. T.; Andersson, U.; Pisetsky, D. S.; Erlandsson, H. H. Pivotal advance: inhibition of HMGB1 nuclear translocation as a mechanism for the anti-rheumatic effects of gold sodium thiomalate. J. Leukocyte Biol. 2008, 83, 31−38. (56) Schierbeck, H.; Wahamaa, H.; Andersson, U.; Harris, H. E. Immunomodulatory drugs regulate HMGB1 release from activated human monocytes. Mol. Med. 2010, 16, 343−351. (57) Lin, R. D.; Mao, Y. W.; Leu, S. J.; Huang, C. Y.; Lee, M. H. The Immuno-Regulatory Effects of Schisandra chinensis and Its Constituents on Human Monocytic Leukemia Cells. Molecules 2011, 16, 4836− 4849. (58) Lampa, J.; Klareskog, L.; Ronnelid, J. Effects of gold on cytokine production in vitro; increase of monocyte dependent interleukin 10 production and decrease of interferon-gamma levels. J. Rheumatol. 2002, 29, 21−28. (59) Barreiro, E.; Casas, J. S.; Couce, M. D.; Gato, Á .; Sánchez, A.; Sordo, J.; Varela, J. M.; López, E. M. V. Synthesis, structural characterization, and antiinflammatory activity of 4579
dx.doi.org/10.1021/jm201416p | J. Med. Chem. 2012, 55, 4568−4579
PŘÍLOHA 11 Hošek, J.; Vančo, J.; Štarha, P.; Paráková, L.; Trávníček, Z. Effect of 2-chloro-substitution of adenine moiety in mixed-ligand gold(I) triphenylphosphine complexes on anti-inflammatory activity: the discrepancy between the in vivo and in vitro models PLoS ONE 8 (2013) e82441
Effect of 2-Chloro-Substitution of Adenine Moiety in Mixed-Ligand Gold(I) Triphenylphosphine Complexes on Anti-Inflammatory Activity: The Discrepancy between the In Vivo and In Vitro Models Jan Hošek1, Ján Vančo1, Pavel Štarha1, Lenka Paráková2, Zdeněk Trávníček1* 1 Department of Inorganic Chemistry, Regional Centre of Advanced Technologies and Materials, Faculty of Science, Palacký University, Olomouc, Czech Republic, 2 Department of Human Pharmacology and Toxicology, Faculty of Pharmacy, University of Veterinary and Pharmaceutical Sciences, Brno, Brno, Czech Republic
Abstract A series of gold(I) triphenylphosphine (PPh3) complexes (1–λ) involving 2-chloro-N6-(substituted-benzyl)adenine derivatives as N-donor ligands was synthesized and thoroughly characterized by relevant methods, including electrospray-ionization (ESI) mass spectrometry and multinuclear NMR spectroscopy. The anti-inflammatory and antiedematous effects of three representatives 1, 5 and λ were evaluated by means of in vitro model based on the expression of pro- and anti-inflammatory cytokines and influence of the complexes on selected forms of matrix metalloproteinases secreted by LPS-activated THP-1 monocytes and in vivo model evaluating the antiedematous effect of the complexes in the carrageenan-induced rat hind-paw edema model. In addition to the pharmacological observations, the affected hind paws were post mortem subjected to histological and immunohistochemical evaluations. The results of both in vivo and ex vivo methods revealed low antiedematous and anti-inflammatory effects of the complexes, even though the in vitro model identified them as promising anti-inflammatory acting compounds. The reason for this discrepancy lies probably in low stability of the studied complexes in biological environment, as demonstrated by the solution interaction studies with sulfur-containing biomolecules (cysteine and reduced glutathione) using the ESI mass spectrometry. Citation: Hošek J, Vančo J, Štarha P, Paráková L, Trávníček Z (2013) Effect of 2-Chloro-Substitution of Adenine Moiety in Mixed-Ligand Gold(I) Triphenylphosphine Complexes on Anti-Inflammatory Activityμ The Discrepancy between the In Vivo and In Vitro Models. PLoS ONE κ(11)μ eκ2441. doiμ 10.1371/journal.pone.00κ2441 Editor: Dariush Hinderberger, Martin-Luther-Universität Halle-Wittenberg, Germany Received August κ, 2013ν Accepted October 23, 2013ν Published November 27, 2013 Copyright: © 2013 Hosek et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The authors would like to thank the Operational Program Research and Development for Innovations – European Regional Development Fund (CZ.1.05/2.1.00/03.005κ), the Operational Program Education for Competitiveness – European Social Fund (project CZ.1.07/2.3.00/20.0017) and Palacký University (Student project PrF_2013_015) for financial support. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. * E-mailμ
[email protected]
Introduction
behind the biological activities of gold(I) complexes were discovered. It has been shown that the gold(I) complexes, exhibiting the cytotoxic and antitumor activities, do not primarily target DNA [12] (as compared to platinum antitumor metallodrug cisplatin), but their main targets are the components of proteasome [13]. In addition, it was shown that the gold(I) species are able to take part in the redox cycling and interact with cellular redox processes by targeting mitochondria [14–16], leading to the decrease of the ATP concentration by uncoupling of oxidative phosphorylation, and thus inhibition of the oxidative ADP phosphorylation [17,1κ]. However, probably the major impact of gold(I) complexes on redox homeostasis of cancer cells is the inhibition of the cytosolic and mitochondrial thioredoxin
The mixed-ligand gold(I) complexes, involving the derivatives of phosphine, are in the scope of chemists for several reasons. One of these includes the ability of chiral gold(I)-phosphine complexes to involve in the catalytic asymmetric gold-catalyzed reactions providing versatile routes to enantiomerically enriched carbo- and hetero-cycles [1–3]. The other reasons embody the ability of gold(I)-phosphine complexes to interact with biological systems and act as biologically active agents, dominantly showing the cytotoxic [4–6], biocidal [7], or antiinflammatory activities [κ–10]. Over the years, since the oligodynamic effect of gold and its compounds was described by von Nägeli et al. [11], many diverse mechanisms standing
PLOS ONE | www.plosone.org
1
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
protocol was approved by the Expert committee on the protection of animals against cruelty at the University of Veterinary and Pharmaceutical Sciences (Permit Numberμ 113/2010, authorization for the use of animals No. 407/2011-30). To minimize the suffering of laboratory animals, all pharmacological interventions were done under anaesthesia. The animal tissues for ex vivo experiments were taken post mortem, immediately after all animals were sacrificed by cervical dislocation.
reductase (TrxR) system [1λ–21]. The ability of gold(I) species to interact with the active site of thioredoxin reductase can be clarified sufficiently by the application of the Pearson’s principle of hard and soft acids and bases, while the gold(I) species as a soft acid tend to bind with the soft base ligands [1λ]. Therefore it prefers the binding to selenolate groups of TrxR, subsequently leading to the inhibition of its activity both in cytosol and mitochondria, leaving the similar system of glutahione reductase, containing the thiolate groups in the active site, unaffected [1λ,22] up to high concentrations. The sum of all the above mentioned effects, the TrxR inhibition, disturbance of mitochondrial respiration, increased production of reactive oxygen species by redox cycling, mitochondrial swelling, and decreasing in the mitochondrial membrane potential, subsequently lead to apoptosis [4]. Additionally, it has been also found that Auranofin inhibits TrxR in a p53independent manner [23]. In addition to the alterations of the GSH and TrxR systems, the anti-inflammatory active compounds like Auranofin showed the ability to induce the HO-1 expression by activating Keap1/ Nrf2 signaling via Rac1/iNOS induction and MAPK activation [24]. It has been also shown that Auranofin can inhibit the activation of STAT3, NF- B, and the homodimerization of tolllike receptor 4 [25–27]. The biological perspective of gold(I) complexes as antiinflammatory and antiedematous agents [κ–10], represented by Auranofin® clinically utilized as a drug for the treatment of rheumatoid arthritis [2κ], led us previously to prepare a series of gold(I) triphenylphosphine complexes involving various N6(substituted-benzyl)adenine derivatives as adenine-based Ndonor ligands [10]. It has been shown that these compounds strongly reduced the production of pro-inflammatory cytokines TNF-α, IL-1β and HMGB1 without the influence on the secretion of anti-inflammatory cytokine IL-1RA in the LPSactivated macrophages. Several representatives of this group also reduced the formation of edema caused by the application of polysaccharide -carrageenan in vivo. All the previously studied gold(I) complexes exhibit similar or better effects as compared to Auranofin in equitoxic doses but lower cytotoxicity than Auranofin. It is generally known, that the substitution of organic ligands, such as the above mentioned N6-benzyladenines, involved in structures of transition metal complexes provide different rate or even type of biological activity of such complexes [2λ,30]. This fact motivated us to prepare a series of structurally analogical gold(I) complexes of the [Au(Ln)(PPh3)] type with a different type of the N6-benzyladenine derivatives. Concretely, we used 2-chloro-N6-(substituted-benzyl)adenine derivatives (HLn), which differ in the substitution at the C2 atom as compared to those involved in our previous work dealing with gold(I) triphenylphosphine complexes involving N6(substituted-benzyl)adenine-based.
Chemicals and Biochemicals H[AuCl4]·3H2O (Acros Organics, Pardubice, Czech Republic), triphenylphosphine (PPh3ν (Sigma-Aldrich Co., Prague, Czech Republic), NaOH (Sigma-Aldrich Co., Prague, Czech Republic) and all solvents (acetone, benzene, diethyl ether, dimethyl sulfoxide, N,N'-dimethylformamide, hexane, waterν Fisher-Scientific, Pardubice, Czech Republic) were obtained from the mentioned commercial sources and were used without further purification. The 2-chloro-N6benzyladenine derivatives HL1-HLλ were synthesized according to the previously published procedure [31] and characterized by elemental analysis, FT-IR, Raman and 1H and 13C NMR spectroscopy to prove their composition and purity. The [AuCl(PPh3)] complex was prepared as described previously [32,33]. RPMI 1640 medium and penicillin-streptomycin mixture were purchased from Lonza (Verviers, Belgium). Phosphate-buffered saline (PBS), fetal bovine serum (FBS), phorbol myristate acetate (PMA), Auranofin (λκ%≤), erythrosin B, and Escherichia coli 0111μB4 lipopolysaccharide (LPS) were purchased from Sigma-Aldrich (Steinheim, Germany). Cell Proliferation Reagent WST-1 was obtained from Roche (Mannheim, Germany). Instant ELISA Kits (eBioscience, Vienna, Austria) were used to evaluate the production of TNFα and IL-1β.
Chemistry The acetone solutions of [AuCl(PPh3)] (1 mmol in 10 mL) and the appropriate N6-(benzyl)adenine derivative (HL1-HLλν 1 mmol in 50 mL) were mixed together and consecutively, 1 mL of 1M NaOH was poured into the reaction mixture. The mixture was stirred for 4 h, after that NaCl was filtered off. The colourless filtrate was evaporated to dryness and the residue was dissolved in benzene (10 mL). The solution was added drop wise into hexane (200 mL). The white solid, which formed, was filtered off, washed with acetone (5 mL) and diethyl ether (10 mL) and dried at 40 °C under an infrared lamp. The results of elemental analysis, electrospray-ionization (ESI) mass spectrometry, 1H, 13C and 31P NMR, FT-IR and Raman spectroscopies, as well as thermogravimetric (TG/DTA) and molar conductivity measurements clearly confirmed the purity and composition of the obtained gold(I) complexes 1–9 (see Figures S1 and S2).
Materials and Methods
Physical Measurements
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The
PLOS ONE | www.plosone.org
Elemental analyses (C, H, N) were performed on a Flash 2000 CHNO-S Analyser (Thermo Scientific). The chlorine contents were determined using the Schöniger method [34].
2
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 1. Schematic representation of complexes 1–9. R symbolizes hydrogen for HL1 and 1, 3-fluoro for HL2 and 2, 2-chloro for HL3 and 3, 3-chloro for HL4 and 4, 2-methoxy for HL5 and 5, 3-methoxy for HL6 and 6, 4-methoxy for HL7 and 7, 4-hydroxy for HLκ and κ and 4-methyl for HLλ and λ. doiμ 10.1371/journal.pone.00κ2441.g001
PLOS ONE | www.plosone.org
3
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
(Thermo Scientific). The theoretic values were calculated by the QualBrowser software (version 2.0.7, Thermo Fisher Scientific). Thermogravimetric (TG) and differential thermal (DTA) analyses were performed using a thermal analyzer Exstar TG/DTA 6200 (Seiko Instruments Inc.) in dynamic air conditions (100 mL min-1) between room temperature (ca 25 °C) and λ00 °C (gradient 2.5 °C min-1).
Table 1. Selected 1H and 13C coordination shifts (calculated as Δδ = δcomplex – δligand) of the prepared gold(I) complexes 1–λ. Complex 1H NMR
13C NMR
31P NMR
N6H
C8H
N9H C2
C4
C5
C6
C8
P
1
-0.51
-0.23
nd
-1.κ2
-1.λ7
0.14
-0.40
7.14
3κ.40
2
-0.36
-0.23
nd
-1.κλ
-1.κ3
0.21
-0.36
7.κ0
noa
3
-0.56
-0.2λ
nd
-1.λ2
-1.κ1
0.44
-0.27
κ.λ7
31.41
4
-0.56
-0.2λ
nd
-1.λ0
-1.73
0.65
-0.12
κ.λ5
31.37
5
-0.30
-0.24
nd
-1.65
-5.73
0.40
-0.10
κ.κ1
37.κ1
6
-0.5κ
-0.2κ
nd
-2.06
-2.λκ
0.κ0
-0.07
κ.3κ
36.λ3
7
-0.55
-0.24
nd
-1.λκ
-1.λ0
0.45
-0.1λ
κ.4κ
noa
8
-0.37
-0.30
nd
-2.15
-2.2λ
2.00
0.3λ
κ.65
37.34
9
-0.54
-0.25
nd
-2.00
-1.λ1
0.60
-0.07
λ.06
37.4κ
Maintenance and Preparation of Macrophages For the determination of biological activities in vitro (cytotoxicity testing, induction of inflammatory response and the evaluation of cytokine secretion), we used the human monocytic leukemia cell line THP-1 (ECACC, Salisbury, UK). The cells were cultivated at 37 °C in RPMI 1640 medium supplemented with 2 mM of L-glutamine, 10% FBS, 100 U/mL of penicillin and 100 µg/mL of streptomycin in a humidified atmosphere containing 5% CO2. Stabilized cells (3rd–15th passage) were split into microtitration plates to get a concentration of 100.000 cells/mL and the differentiation to macrophages was induced by 50 ng/mL PMA dissolved in DMSO, as described previously [35].
nd = not detectedν noa = signal not observed even after 14 hrs of the experiment due to limited solubility of the complexes in DMF-d7. doiμ 10.1371/journal.pone.00κ2441.t001
Cytotoxicity Testing THP-1 cells (floating monocytes, 500.000 cells/mL) were incubated in 100 L of serum-free RPMI 1640 medium and seeded into λ6-well plates in triplicate at 37 °C. The measurements were done 24 h after the treatment with 0.03λ, 0.156, 0.625, 2.5, and 10 M solutions of the tested compounds 1, 5 and 9 dissolved in dimethyl sulfoxide (DMSO). Viability was determined by the WST-1 test according to the manufacturer’s manual. The amounts of created formazan (which correlate to the number of metabolically active cells in the culture) were calculated as a percentage of control cells, which were treated only with DMSO (100% viability). The cytotoxic LD50 concentrations of the tested compounds were determined by the data from the equation generated by the KURV+ Version 4.4b software (Conrad Button Software, Arlington, WA, USA).
Figure 2. Cytotoxicity of complexes 1, 5, and 9 on THP-1 cells. THP-1 cells were treated with the decreasing concentration (10–0.03λ M) of 1, 5, and λ, respectively. The number of metabolically active cells was determined by the WST-1 test after 24 h of incubation. The viability was calculated in comparison to the vehicle-treated cells. The results are expressed as means ± S.E. for three independent experiments.
Drug Treatment and Induction of Inflammatory Response
doiμ 10.1371/journal.pone.00κ2441.g002
Differentiated macrophages were pretreated with 100 nM or 600 nM solution of 1, 5 and λ in 0.1% DMSO, 100 nM solution of Auranofin in 0.1% DMSO, and with 0.1% DMSO solution itself (vehicle) for 1 hν the given concentrations of the tested compounds lack cytotoxic effect (cell viability >λ4%). The inflammatory response was triggered by adding of 1.0 µg/mL LPS dissolved in water to pre-treated macrophages. Control cells were pre-treated by DMSO only and subsequently incubated without the addition of LPS, thus serving as a source of the basal expression of pro-inflammatory cytokines. Each experiment was repeated three times.
Conductivity measurements of 10-3 M N,N'-dimethylformamide (DMF) and 10-3 M methanol solutions were carried out using a conductometer 340i/SET (WTW) at 25 °C. FT-IR spectra were recorded on a Nexus 670 FT-IR spectrometer (ThermoNicolet) using ATR technique at the 400–4000 cm-1 and 150–600 cm-1 regions. A Raman spectrometer Nicolet NXR λ650 equipped with the liquid nitrogen cooled NXE Genie germanium detector (ThermoNicolet) was used to record Raman spectra in the region of 150–3750 cm-1. 1H, 13C and 31P NMR spectra of DMFd7 solutions were measured on a Varian 400 MHz NMR spectrometer at 300 K with the tetramethylsilane (SiMe4) used as an internal reference standard (for 1H and 13C spectra) and κ5% H3PO4 (for 31P). Mass spectra of the methanol solutions were obtained by an LCQ Fleet ion trap mass spectrometer by the positive mode electrospray ionization (ESI+) technique
PLOS ONE | www.plosone.org
Evaluation of Cytokine Secretion by ELISA Macrophages, which were pretreated with the tested compounds for 1 h, were incubated with LPS for next 24 h. After this period, medium was collected and the concentration
4
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 3. Effects of the gold(I) complexes and the reference drug Auranofin on the LPS-induced TNF-α secretion. The cells were pretreated with complexes 1, 5 and λ (100 nM - depicted in dark blueν 600 nM - depicted in light blue), and Auranofin (100 nM), or the vehicle (DMSO) only. After 1 h of the incubation, the inflammatory response was induced by LPS [except for the control cells]. The secretion was measured 24 h after the LPS addition. The results are expressed as means for three independent experiments. * Significant difference in comparison to vehicle-treated cells (p < 0.05), *** significant difference in comparison to vehicle-treated cells (p < 0.001), ### significant difference in comparison to Auranofin-treated cells (p < 0.001). doiμ 10.1371/journal.pone.00κ2441.g003
of TNFα, and IL-1β was measured by Instant ELISA kit according to manufactures’ manual.
calculated densitometrically using the AlphaEasy FC 4.0.0 software (Alpha Innotech, USA).
Zymography
Animals
Conditioned media obtained by the same way as for cytokines evaluation were used for the measurement of matrixmetalloproteinases (MMP) activity by zymography, as described previously [36]. Briefly, 20 L of collected medium was loaded into a non-denaturating 12% polyacrylamide gel impregnated by 0.1% (w/v) gelatin. After the electrophoresis experiments, sodium dodecyl sulfate (SDS) from gels was washed out by 2.5% Triton X100 and the gels were incubated in developing buffer (50 mM Tris pH κ.κ, 5 mM CaCl2, 3 mM NaN3 and 0.5% Triton X100) at room temperature (ca 23 °C) for 30 min and at 37 °C overnight (16–20 h). Gels were stained by Coomassie blue. Intensities of digested regions were
Wistar - SPF (6–κ weeks male) rats were obtained from the AnLab, Ltd., Prague. The animals were kept in plexiglass cages at the constant temperature of 22±1 °C, and relative humidity of 55±5% for at least 1 week before the experiment. They were given food and water ad libitum. All experimental procedures were performed according to the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. In addition, all tests were conducted under the guidelines of the International Association for the Study of Pain [37]. After the one week adaptation period, male Wistar-SPF rats (200–250 g) were randomly assigned to the groups (n = 10). The control group received 25% DMSO (v/v in water for
PLOS ONE | www.plosone.org
5
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 4. Effects of the gold(I) complexes and the reference drug Auranofin on the LPS-induced IL-1β secretion. The cells were pre-treated with complexes 1, 5 and λ (100 nM - depicted in dark blueν 600 nM - depicted in light blue) and Auranofin (100 nM), or the vehicle (DMSO) only. After 1 h of the incubation, the inflammatory response was induced by the LPS [except for the control cells]. The secretion was measured 24 h after the LPS addition. The results are expressed as means for three independent experiments. *** Significant difference in comparison to vehicle-treated cells (p < 0.001), # significant difference in comparison to Auranofin-treated cells (p < 0.05), ## significant difference in comparison to Auranofin-treated cells (p < 0.01), ### significant difference in comparison to Auranofin-treated cells (p < 0.001). doiμ 10.1371/journal.pone.00κ2441.g004
injections PhEur, intraperitonealν i.p.). The other three groups include a carrageenan-treated, an Auranofin positive control (Auranofin + carrageenan) and indomethacin positive control (indomethacin + carrageenan) groups [10].
the rats. The paw volume was measured immediately after the carrageenan injection and during the next 6 h after the administration of the edematogenic agent using a plethysmometer (model 715λ, Ugo Basile, Varese, Italy). The degree of swelling induced was evaluated by the percentage of change of the volume of the right hind paw after the carrageenan treatment from the volume of the right hind paw before carrageenan treatment. The data were combined afterwards for all 10 animals within the each experimental group and subjected to statistical evaluation by the one-way ANOVA with post-hoc Tukey test of significance. All animals were sacrificed by cervical dislocation, and immediately after that, the affected hind paws were separated and underwent the process of dehydration and fixation, and
Carrageenan-induced Hind Paw Edema and Ex vivo Histological and Immunohistochemical Evaluation The carrageenan-induced hind paw edema model was used for the determination of the anti-inflammatory activity [10]. Animals were i.p. pretreated with the complexes 1, 5 and 9 (10 mg/kg), Auranofin (10 mg/kgν positive control), indomethacin (5 mg/kgν positive control) or 25% DMSO (v/v in water for injections PhEur), 30 min prior to the injection of 1% carrageenan (50 L) into the plantar side of right hind paws of
PLOS ONE | www.plosone.org
6
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 5. Effects of tested gold(I) complexes and the reference drug Auranofin on LPS-induced matrix metalloproteinases (MMP) activity. Cells were pretreated with the complexes 1, 5 and λ (100 nM - depicted in dark blueν 600 nM - depicted in light blue), Auranofin (100 nM), or the vehicle (DMSO) only. After 1 h of the incubation, the inflammatory response was induced by LPS (except for the control cells). Activity of (pro-)MMP-2 was detected by zymography, (a). Intensity of digested bands was analysed by densitometry analysis. The figure shows pro-MMP-2 / MMP-2 ratio, (b). Results are expressed as means for three independent experiments. * Significant difference in comparison to vehicle-treated cells (p < 0.05), ** significant difference in comparison to vehicle-treated cells (p < 0.01), # significant difference in comparison to Auranofin-treated cells (p < 0.05), ## significant difference in comparison to Auranofin-treated cells (p < 0.01). doiμ 10.1371/journal.pone.00κ2441.g005
Results and Discussion
were embedded into paraffin blocks by means of standard protocols [3κ]. The histopathological changes, like infiltration of different skin elements and deeper laying tissues by the white blood cells (dominantly neutrophils and lymphocytes) stained by the standard H&E-staining and Gömöri trichrome staining, were evaluated. In addition to the classical histological investigation, the immunohistochemical detection of apoptosis (caspase 3 and TUNEL), TNF-α, interleukin 6 (IL-6) and selectin E (CD62E) was performed using the Abcam® rat monoclonal abtibodies.
Characterization of the Complexes The composition of the prepared complexes 1–9 (Figure 1) correlates with the determined content of C, H, N and Cl, since the differences between the calculated and found values were up to 0.44%. The complexes 1−9 are very well soluble in DMF, methanol and ethanol and partially soluble in water at room temperature. The non-electrolytic character of the complexes was proved by the molar conductivity measurements, because the obtained values of the 10–3 M methanolic and DMF solutions (See Figures S1 and S2) fell to the intervals characteristic for nonelectrolytes [3λ]. ESI+ mass spectra of 1– λ contain the [Au(HLn)(PPh3)]+ molecular peak, as well as that of [HLn+H]+ belonging to the free adenine-based ligand. The [Au(Ln)(PPh3)+Na]+ adducts were detected in the spectra of 2, 5, 7 and 9. The simultaneous TG/DTA analysis proved the studied compounds as non-solvated ones (thermally stable up to 135 °C), except for 6 and 7 (monohydrates, see Figure S1). The bands of very strong intensity, whose maxima were detected in the 1606–1613 cm−1 (FT-IR) and 1604–161κ cm-1 (Raman) regions, may be assigned to the (C=N)ar vibration of the purine ring [40,41]. The bands found in both the FT-IR and Raman spectra between 2λ00 and 3000 cm-1, and at about 3060 cm -1 can be assigned to (C–H)aliph, including (C–H)met of 5–7 and 9, and (C–H)ar, respectively. The characteristic vibrations connected with the phenyl ring substitution, in particular (Car-F) for 2 at 1264 cm-1, and (Car-Cl) for 3 and 4 at ca 1160 cm-1 , were detected as well. The strong bands observed in the λλ6–1003 cm-1 (FT-IR) and 1000–1003 cm-1 (Raman) regions may be ascribed to the C2-Cl vibrations
Interactions with Cysteine and Reduced Glutathione Assessed by ESI Mass Spectrometry The interaction experiments between the selected representative complex 5 and the mixture of physiological levels of cysteine and glutathione were performed on a ThermoFinnigan LTQ Fleet Ion-Trap mass spectrometer, using the positive ionization mode. The FIA method was used to introduce the reaction system (100 L spikes) into the mass spectrometer, while the mixture of 10 mM ammonium acetate solution and methanol (10μλ0 v/v) was used as a mobile phase. The ESI- source was set up as followsμ source voltage was 4.4κ kV, the vaporizer temperature was 160°C, the capillary temperature was 275 °C, the sheath gas flow rate was 20 L/ min, and auxiliary gas flow rate was 5 L/min. The system was calibrated according to the manufacturer specifications and no further tuning was needed.
PLOS ONE | www.plosone.org
7
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 6. The time-dependent profile of antiedematous effect of tested compounds on carrageenan induced hind paw edema in rats. doiμ 10.1371/journal.pone.00κ2441.g006
coordinated through its Nλ atom and one PPh3 molecule coordinated through phosphorus atom.
observed in the spectra of all the complexes. The maxima at 53λ–545 cm-1 (FT-IR) and 531–555 cm-1 (Raman) may be assigned to the (Au–N) [42,43] stretching vibrations, while the maxima at 342–365 cm-1 (FT-IR) and 345–372 cm-1 (Raman) belong to the (Au–P) [44,45] vibrations of 1–λ. The complete results of the FT-IR and Raman spectroscopies are given in Figures S1 and S2. The obtained NMR spectroscopy results gave evidence of the composition of the gold(I) complexes by means of the presence of both types of ligands in the structure of 1–λ as well as their coordination mode. The highest 13C NMR coordination shifts (Δδ = δcomplex _ δligand, ppmν Table 1) were determined for the C4 and Cκ atoms shifted by 1.73–5.73 ppm upfield, and 7.14–λ.06 ppm downfield, respectively (see Table 1). This is caused by the fact that these atoms are adjacent to the Nλ coordination site. The signals observed in the 1H NMR spectra for Cκ-H were shifted by 0.23-0.30 ppm upfield. The N6–H signals were detected as slightly shifted in the case of 1–λ, while the Nλ–H signals were not found for all the studied complexes as expected for the deprotonated 2-chloro-N6benzyladenine-based ligands symbolized as Ln (see Table 1). The 31P NMR spectra of the complexes showed singlets at 30.λ7–32.44 ppm (see Figure S2) as compared to the signal of free PPh3 (-5.λ6 ppm). The obtained results indicated that the gold(I) atom is coordinated by one adenine derivative
PLOS ONE | www.plosone.org
Cytotoxicity to THP-1 Cells The cytotoxic effect of 1, 5 and λ was evaluated on the THP-1 cell line. The complexes 1, 5 and 9 exhibit similar LD50 values ranged from 1.45 to 1.5κ µM (Figure 2). The tested compounds demonstrated hormesis (in our case higher metabolic activity in the comparison with untreated cells) in concentrations lower than 1 µM. Similar effect was observed for other gold(I) complexes containing different N6benzyladenine derivatives and triphenylphosphine moiety described previously [10]. It could be caused by soft oxidative stress induced by Au(I) atom, which results in activation of repairing and/or pro-survival processes [46].
Cytokine Expression To evaluate the anti-inflammatory potential of the tested gold(I) complexes, the expression of two typical proinflammatory cytokines TNF-α and IL-1β was determined in vitro. To better understand the activity of 1, 5 and λ we tested their influence on the selected pro- and anti-inflammatory cytokines in both the equitoxic (600 nM) and equimolar (100 nM) concentrations as compared to Auranofin. Our previous results indicated that gold(I) complexes with N6-benzyladenine
κ
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
derivatives as N-donor ligands were able to attenuate the LPSinduced secretion of these cytokines [10]. A gentle diminution of the TNF-α secretion (by the factor of 1.1κ–1.43) was observed after the pretreatment of the cells by 100 nM solutions of 1, 5, 9 or Auranofin (Figure 3). On the other hand, when the higher concentration of the studied complexes was used (600 nM), the production of TNF-α decreased significantly by the factor of 10.6λ (for 1), κ.71 (for 5) and 20.07 (for 9). It has been previously described that gold sodium thiomalate, i.e. sodium ((2-carboxy-1carboxylatoethyl)thiolato)gold(I), GST, preferentially inhibits the IL-1β production rather than TNF-α [47]. This effect of goldcontaining compounds was proven by this experiment, where Auranofin 1.73-times inhibited the IL-1β secretion (Figure 4). In this parameter, the tested compounds 1, 5 and 9 were significantly more active than Auranofin and reduced the production of this cytokine by the factor of 2.25–3.3κ. When the higher concentration was used, the reduction reached the value of κ.12-times for 5. These results corresponded well with our previous findings on the similar gold(I) complexes with N6benzyladenine derivatives, where we noted only moderated effect of gold(I) complexes on the TNF-α secretion, but the production of IL-1β was strongly affected [10]. Newly synthesized complexes can be arranged according to their ability to reduce the production of pro-inflammatory cytokines TNF-α and IL-1β as followsμ Auranofin (the least potent) < 5 < 1 < λ (the most potent). The other hallmark of the inflammation is the production of matrix metalloproteinases (MMPs). The activity of MMP2 and pro-MMP2 is summarised in Figure 5a. The tested gold(I) complexes were able to significantly decrease the activity of pro-MMP2 by 15–40% and to change the pro-MMP2/MMP2 ratio in the favour of physiologically inactive pro-MMP2 (Figure 5b). The activity of MMPλ was not affected by the tested compounds (data not shown). The ability of the gold(I) complexes to attenuate activity of MMP2 sketches in their antiinflammatory potential.
In Vivo Pharmacological and Ex Vivo Histological Evaluations of Antiedematous Activity of 1, 5 and 9 Based on the in vitro experiments, the in vivo tests of antiinflammatory activity of the complexes 1, 5 and 9 were performed using the carrageenan-induced hind paw edema model, which investigate the effect of the tested complexes on one of the main symptoms of acute inflammation, i.e. the formation of edema. The clinically used gold-metallodrug Auranofin was used as the primary standard of the antiinflammatory activity. In the experiments, we used the same dosages of Auranofin as the tested compoundsν i.e. 10 mg/kg in the form of the fine suspension in 25% DMSO (v/v in water for injections PhEur) applied intraperitonealy 30 min before the intraplantar injection of carrageenan. As a secondary standard, acting dominantly by the different biochemical pathway (i.e. cyclooxygenase inhibition), the NSAID indomethacin was used in the dose of 5 mg/kg. The comprehensive overview of antiedematous activity profiles of the tested compounds is summarized in Figure 6.
Figure 7. Histologic sections of the hind paw, stained with Gömöri trichrome staining (a) and Hematoxylin-eosin (b-d) (photographed at 40x magnification). Tissue exposed to 25% DMSO (controlν a), and complex 5 (b) with the massive acute inflammatory reaction in the transition from dermis to hypodermis with the massive infiltrate of neurophils (PMN) in connective tissuesν tissue exposed to complex λ (c) with the massive acute inflammatory reaction reaching from the hypodermis up to the papillary layer of dermis with a massive infiltrate of neurophils (PMN)ν tissue exposed to Auranofin (d) with the inflammatory reaction in the hypodermis with a slight PMN infiltrate. 1 - PMN infiltrate, 2 - arteriola, 3 - dilated veins with haemostasis, 4 - vein with PMN presence in the blood stream and haemostasis, 5 - collagen connective tissue, 6 haemorrhage characterised by the presence of erythrocytes in interstitium, 7 - muscle fibres. doiμ 10.1371/journal.pone.00κ2441.g007
PLOS ONE | www.plosone.org
λ
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Figure 8. The ESI mass spectrometry analysis of interactions between the complex 5 and sulphur-containing biomolecules. The mass spectrum of the mixture of physiological concentrations of cysteine (2λ0 M) and reduced glutathione (6 M), (a)ν the mass spectrum of the solution of complex 5, (b)ν the mass spectrum of the interacting system containing the physiological concentrations of cysteine and reduced glutathione and 5, immediately after mixing, (c)ν the mass spectrum of the interacting system containing the physiological concentrations of cysteine and reduced glutathione and 5, 1 h after mixing, (d)ν the mass spectrum of the interacting system containing the physiological concentrations of cysteine and reduced glutathione and 5, 12 h after mixing, (e). doiμ 10.1371/journal.pone.00κ2441.g00κ
In contrast to the results of in vitro anti-inflammatory activity, all the tested complexes showed no anti-inflammatory activity in vivo. The complexes 1 and 5 did not influence the swelling induced by the carrageenan injection. On the other hand, the complex 9 augmented the hind paw edema significantly, as compared with the control (25% DMSO) by the one-way
PLOS ONE | www.plosone.org
ANOVA with post-hoc Tukey test on the level of significance p = 0.01. To assess the tissue consequences connected with the formation of inflammation and edema and the level of influence caused by the tested compounds after the intraplantar injection of carrageenan, the histopathological observations were made on the tissue sections. The histopathological changes in
10
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Interactions with a Mixture of Cysteine and Glutathione Assessed by ESI-MS
tissues, stained by the standard H&E-staining, were quite similar from the qualitative point of view in all samples. The affected tissues (dermis up to papillary layer and hypodermis) clearly show the signs of inflammation, e.g. the presence of inflammation infiltrate, vasodilatation with the haemostasis, presence of leukocytes in blood stream, and extracellular edema. The noticeable differences were found in the extent of infiltration of subcutaneous tissues and dermis by the polymorphonuclear cells (PMNs, mostly neutrophils) in the samples obtained from the Auranofin and Indomethacin treated groups (see Figure 7), where only focal mobilization of PMNs in dermis is evident and the number of PMNs decreases dramatically with the proximity of epidermis. On the other hand in all other samples, obtained from the groups of animals treated by the title complexes or the control group (only vehicle, 25% DMSO, was applied), the massive infiltration of PMNs in hypodermis and dermis was observed, accompanied by the presence of PMNs up to the papillary layer (for the representative examples see Figure 7). In addition to the classical histological investigation, the immunohistochemical detection of apoptosis (caspase 3, TUNEL assay), TNF-α, interleukin 6 (IL-6) and selectin E (CD62E) was performed. Unfortunately, the results of the immunohistochemical detection were disputable and inconclusive due to the tissue nature, because no differences were observed between the samples, standards and control, and therefore they are not discussed in detail. However, it is evident that the results of histological evaluation of tissue samples tightly correlate with the results obtained by plethysmometric measurements. The obtained results are rather surprising due to the fact that previously reported isostructural complexes, involving the ligands not bearing the chloro-substitution at the position C2 of the purine skeleton, showed remarkable antiedematous activity, comparable with metallodrug Auranofin as well as the highly active derivatives of gold(I) complexes involving 3(aryl)-2-sulfanylpropenoic acid [4κ] and higher activity than other gold(I) and gold(III) complexes [4λ,50]. The cause of this radical change in biological activity cannot be explained by trivial means, while the biological activity of these complexes itself is based on a multimodal effect. The main aspect influencing the quality of the action is undoubtedly the pharmacophore of Au(I)-trisubstituted phosphine derivative, however the quantity of the action is dependent from a variety of physical and chemical properties, like the stability in biological systems, influence of ligands on biodistribution and excretion of complexes, the intrinsic pro-inflammatory effect, etc. With the intention to better understand the mechanism of interaction of the tested complexes with biological systems, we decided to perform the solution experiments involving two main sulphur-containing constituents of human plasma, amino acid cysteine and reduced glutathione. Thus, the ESI mass spectrometry was used to analyse the specific species participating in the interactions with biological systems.
PLOS ONE | www.plosone.org
Due to chemical properties of Au(I) species they tend to form the strong coordination bonds with soft Lewis base ligands, such as thiolato or selenolato ions, or phosphine derivatives, while the latter ones form the most stable bonds. Therefore, Au(I) complexes tend to bind to sulfhydryl-containing substances in the biologically relevant environments (such as blood or serum), such as amino acid cysteine (Cys) or small proteins (such as glutathione - GSH) and in particular with high molecular proteins (such as albumin or globulins [51]), with the ligand exchange mechanism. The exchange of N-ligands for Sligands occurs relatively quickly (within 20 minutes when interacting with albumin and globulins in the blood [52]), the Pligand exchange occurs much more slowly, and it seems that in this mechanism the cooperative effects of adjacent thiolato or selenolato ligands in the neighborhood of interaction site play an important role, which is interpreted as one of the molecular mechanisms of incorporation of gold into the active site of selenium-containing flavoreductases, such as thioredoxin reductase [53]. In the scope of our work, we have verified the expected behaviour of selected complex 5 (applied in the concentration of 15 µM, corresponding to the highest therapeutic blood levels of gold during chrysotherapy [54]) in biologically relevant conditions [55] using a mixture of cysteine (at 2λ0 µM concentration) and reduced glutathione (at the 6 µM concentration). Based on the results of ESI-MS experiments, we confirmed that the complex 5 reacts with sulfhydrylcontaining substances in time-independent manner by the ligand-exchange mechanism based on the substitution of Nligand (N6-benzyladenine derivative) by the cysteine molecule. This mechanism was confirmed by the emergence of ion 57λ.λ3 m/z, corresponding to the [Cys-Au-PPh3]+ intermediate, and ion 103κ.03 m/z, corresponding to the ionic species [Cys(Au-PPh3)2]+ (see Figure κ). In biologically relevant concentrations of both low molecular sulfhydryl-containing compounds, we observed no intermediates involving the glutathione molecule. The interaction studies revealed also another important fact about the studied complexes. In contrast to the previously published structurally similar complexes, involving the ligands not bearing the chloro-substitution at the position C2 of the purine skeleton [10], the mass spectra of the complexes revealed a considerable instability of the complexes, as demonstrated by the appearance of the intensive ion 721.1λ m/z, corresponding to the [Au(PPh3)2]+ intermediate in all solutions (in the solution of complex 5 as well as in the mixtures with cysteine and glutathione), and other ionic species involving the residue Au-PPh3 (e.g. [M+Au-PPh3]+ν 1206.06 m/z), the free ligand ([HL5+H]+ν 2λ0.10 m/z), or the free triphenylphosphine residue (e.g. [PPh3+H]+ν 263.13 m/z or [PPh3+H+CH3OH]+ν 2λ5.13 m/z). The described instability in solutions might contribute significantly to the behaviour of title complexes in biological systems leading to the inactivity in context of antiedematous activity as presented above.
11
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
Conclusions
demonstrated by the interaction studies with sulfur-containing biomolecules (cysteine and GSH) using the electrosprayionization mass spectrometry.
The gold(I) complexes of the general formula [Au(Ln)(PPh3)] (1–λ) involving deprotonated 2-chloro-N6-(substitutedbenzyl)adenine derivatives (Ln) coordinated through the Nλ atom, and PPh3 molecule coordinated to the central atom through its phosphorus atom were prepared and characterized. The results obtained by the in vitro model of the LPS-activated THP-1 monocytes showed the ability of the representative complexes 1, 5 and 9 to interfere with the cell-cycle of THP-1 cells leading to hormetic effect and simultaneously decrease the secretion of pro-inflammatory cytokines TNF-α and IL-1β as well or even better as gold(I)-based metallodrug Auranofin at the same concentration level. The prepared gold(I) complexes also showed the ability to decrease the activity of matrix metalloproteinase 2, while leaving the MMPλ unaffected. In conjunction with the promising results of in vitro assays, the antiedematous effect was evaluated by carrageenan-induced rat hind-paw edema model. In addition to the pharmacological observations, the affected hind paws were post mortem subjected to histological and immunohistochemical evaluations. The results of both in vivo and ex vivo methods, however, revealed low antiedematous and anti-inflammatory effects of the gold(I) complexes. This discrepancy may be caused by low stability of the complexes in physiological conditions, as
Supporting Information Supporting Information S1. Figure S1. TG/DTA curves of the complexes 1 (left) and 6 (right) given together with the calculated and observed weight losses. Figure S2. 31P NMR spectrum of complex 6. (DOCX)
Acknowledgements The authors are grateful to Mr. Tomáš Šilha for the help with the preparation of the complexes, Dr. Igor Popa for performing NMR experiments, Assoc. Prof. Pavel Suchý and Ms. Gabriela Pražanová for the help with biological testing.
Author Contributions Conceived and designed the experimentsμ JH JV PS ZT. Performed the experimentsμ JH JV PS LP ZT. Analyzed the dataμ JH JV PS ZT. Contributed reagents/materials/analysis toolsμ JH JV PS ZT. Wrote the manuscriptμ JH JV PS ZT.
References 13. Buac D, Schmitt S, Ventro G, Kona FR, Dou QP (2012) Dithiocarbamate-based coordination compounds as potent proteasome inhibitors in human cancer cells. Mini Rev Med Chem 12μ 11λ3–1201. doiμ10.2174/13κλ55712κ02762040. PubMedμ 22λ315λ1. 14. Hoke GD, Rush GF, Bossard GF, McArdle JV, Jensen BD et al. (1λκκ) Mechanism of alterations in isolated rat liver mitochondrial function induced by gold complexes of bidentate phosphines. J Biol Chem 263μ 11203–11210. PubMedμ 245701κ. 15. Hoke GD, Rush GF, Mirabelli CK (1λκλ) The mechanism of acute cytotoxicity of triethylphosphine gold(I) complexes. III. Chlorotriethylphosphine gold(I)-induced alterations in isolated rat liver mitochondrial function. Toxicol Appl Pharmacol λλμ 50–60. doiμ 10.1016/0041-00κX(κλ)λ0110-5. PubMedμ 24712λ2. 16. Rackham O, Nichols SJ, Leedman PJ, Berners-Price SJ, Filipovska A (2007) A gold(I) phosphine complex selectively induces apoptosis in breast cancer cellsμ implications for anticancer therapeutics targeted to mitochondria. Biochem Pharmacol 74μ λλ2–1002. doiμ10.1016/j.bcp. 2007.07.022. PubMedμ 176λ7672. 17. Hoke GD, McCabe FL, Faucette LF, Bartus JO, Sung CM et al. (1λλ1) In vivo development and in vitro characterization of a subclone of murine P3κκ leukemia resistant to bis(diphenylphosphine)ethane. Mol Pharmacol 3λμ λ0–λ7. PubMedμ 1κλκλκ2. 1κ. Smith PF, Hoke GD, Alberts DW, Bugelski PJ, Lupo S et al. (1λκλ) Mechanism of toxicity of an experimental bidentate phosphine gold complexed antineoplastic agent in isolated rat hepatocytes. J Pharmacol Exp Ther 24λμ λ44–λ50. PubMedμ 2732λ55. 1λ. Gandin V, Fernandes AP, Rigobello MP, Dani B, Sorrentino F et al. (2010) Cancer cell death induced by phosphine gold(I) compounds targeting thioredoxin reductase. Biochem Pharmacol 7λμ λ0–101. doiμ 10.1016/j.bcp.200λ.07.023. PubMedμ 1λ665452. 20. Tonissen KF, Di Trapani G (200λ) Thioredoxin system inhibitors as mediators of apoptosis for cancer therapy. Mol Nutr Food Res 53μ κ7– 103. doiμ10.1002/mnfr.2007004λ2. PubMedμ 1κλ7λ503. 21. Vergara E, Casini A, Sorrentino F, Zava O, Cerrada E et al. (2010) Anticancer therapeutics that target selenoenzymesμ synthesis, characterization, in vitro cytotoxicity, and thioredoxin reductase inhibition of a series of gold(I) complexes containing hydrophilic phosphine ligands. Chemmedchem 5μ λ6–102. doiμ10.1002/cmdc. 200λ00370. PubMedμ 1λλ3766λ. 22. Lee MT, Ahmed T, Haddad R, Friedman ME (1λκλ) Inhibition of several enzymes by gold compounds. II. Beta-glucuronidase, acid phosphatase
1. Pradal A, Toullec PY, Michelet V (2011) Recent developments in asymmetric catalysis in the presence of chiral gold complexes. Synthesis 10μ 1501–1514. 2. Cao ZY, Wang XM, Tan C, Zhao XL, Zhou J et al. (2013) Highly stereoselective olefin cyclopropanation of diazooxindoles catalyzed by a C-2-symmetric spiroketal bisphosphine/Au(I) complex. J Am Chem Soc 135μ κ1λ7–κ200. doiμ10.1021/ja4040κλ5. 3. Liu B, Li KN, Luo SW, Huang JZ, Pang H et al. (2013) Chiral gold complex-catalyzed hetero-diels-alder reaction of diazenesμ highly enantioselective and general for dienes. J Am Chem Soc 135μ 3323– 3326. doiμ10.1021/ja3110472. PubMedμ 234214λ3. 4. Jungwirth U, Kowol CR, Keppler BK, Hartinger CG, Berger W et al. (2011) Anticancer activity of metal complexesμ involvement of redox processes. Antioxid Redox Signal 15μ 10κ5–1127. doiμ10.10κλ/ars. 2010.3663. PubMedμ 21275772. 5. Ott I, Qian X, Xu Y, Vlecken DHW, Marques IJ et al. (200λ) A gold(I) phosphine complex containing a naphthalimide ligand functions as a TrxR inhibiting antiproliferative agent and angiogenesis inhibitor. J Med Chem 52μ 763–770. doiμ10.1021/jmκ012135. PubMedμ 1λ123κ57. 6. Lima JC, Rodriguez L (2011) Phosphine-gold(I) compounds as anticancer agentsμ general description and mechanisms of action. Anticancer Agents Med Chem 11μ λ21–λ2κ. doiμ 10.2174/1κ71520117λ7λ27670. PubMedμ 21κ6423κ. 7. Navarro M (200λ) Gold complexes as potential anti-parasitic agents. Coord Chem Rev 253μ 161λ–1626. doiμ10.1016/j.ccr.200κ.12.003. κ. Gielen M, Tiekink ERT (2005) Metallotherapeutic drugs and metalbased diagnostic agents. Londonμ Willey. λ. Ott I (200λ) On the medicinal chemistry of gold complexes as anticancer drugs. Coord Chem Rev 253μ 1670–16κ1. doiμ10.1016/j.ccr. 200λ.02.01λ. 10. Trávníček Z, Starha P, Vančo J, Silha T, Hošek J et al. (2012) Antiinflammatory active gold(I) complexes involving 6-substituted purine derivatives. J Med Chem 55μ 456κ–457λ. doiμ10.1021/jm201416p. PubMedμ 22541000. 11. von Nägeli CW, Schwendener S, Cramer C (1κλ3) Ueber oligodynamische Erscheinungen in lebenden Zellen. Zürichμ Druck von Zürcher und Furrerμ 51. 12. Kostova I (2006) Gold coordination complexes as anticancer agents. Anticancer Agents Med Chem 6μ 1λ–32. doiμ 10.2174/1κ7152006774755500. PubMedμ 16475λ24.
PLOS ONE | www.plosone.org
12
November 2013 | Volume κ | Issue 11 | eκ2441
Diversity in Biological Responses to Au-Complexes
23.
24.
25. 26.
27.
2κ.
2λ. 30.
31.
32.
33. 34. 35.
36.
37. 3κ. 3λ. 40.
and L-malate dehydrogenase by sodium thiomalatoraurate(I), sodium thiosulfatoaurate(I) and thioglucosoaurate(I). J Enzyme Inhib 3μ 35–47 Hedström E, Eriksson S, Zawacka-Pankau J, Arnér ES, Selivanova G (200λ) p53-dependent inhibition of TrxR1 contributes to the tumorspecific induction of apoptosis by RITA. Cell Cycle κμ 3576–35κ3. PubMedμ 1λκ3κ062. Kim NH, Oh MK, Park HJ, Kim IS (2010) Auranofin, a gold(I)-containing antirheumatic compound, activates Keap1/Nrf2 signaling via Rac1/ iNOS signal and mitogen-activated protein kinase activation. J Pharmacol Sci 113μ 246–254. doiμ10.1254/jphs.0λ330FP. PubMedμ 206476κ6. Jeon KI, Byun MS, Jue DM (2003) Gold compound auranofin inhibits IkappaB kinase (IKK) by modifying Cys-17λ of IKKbeta subunit. Exp Mol Med 35μ 61–66. doiμ10.103κ/emm.2003.λ. PubMedμ 1275440κ. Kim NH, Lee MY, Park SJ, Choi JS, Oh MK et al. (2007) Auranofin blocks interleukin-6 signalling by inhibiting phosphorylation of JAK1 and STAT3. Immunology 122μ 607–614. doiμ10.1111/j. 1365-2567.2007.0267λ.x. PubMedμ 176454λ7. Youn HS, Lee JY, Saitoh SI, Miyake K, Hwang DH (2006) Auranofin, as an anti-rheumatic gold compound, suppresses LPS-induced homodimerization of TLR4. Biochem Biophys Res Commun 350μ κ66– κ71. doiμ10.1016/j.bbrc.2006.0λ.0λ7. PubMedμ 17034761. Williams HJ, Ward JR, Reading JC, Brooks RH, Clegg DO et al. (1λλ2) Comparison of auranofin, methotreaxate, and the combination of both in the treatment of rheumatoid arthritis. A controlled clinical trial. Arthritis Rheum 35μ 25λ–26λ. doiμ10.1002/art.17κ0350304. PubMedμ 1536666. Sorenson JRJ (1λ76) Copper chelates as possible active forms of the antiarthritic agents. J Med Chem 1λμ 135–14κ. doiμ10.1021/ jm00223a024. PubMedμ 1246036. Lemoine P, Viossata B, Dung NH, Tomas A, Morgant G (2004) Synthesis, crystal structures, and anti-convulsant activities of ternary [ZnII(3,5-diisopropylsalicylate)2], [ZnII(salicylate)2] and [ZnII(aspirinate)2] complexes. J Inorg Biochem λκμ 1734–174λ Oh CH, Lee SC, Lee KS, Woo ER, Hong CY et al. (1λλλ) Synthesis and biological activities of C-2, N-λ substituted 6-benzylaminopurine derivatives as cyclin-dependent kinase inhibitor. Arch Pharm (Weinheim) 332μ 1κ7–1λ0. doiμ10.1002/(SICI)1521-41κ4(1λλλ6)332μ6. PubMedμ 103λλ4κ6. Mann FG, Wells AF, Purdie D (1λ37) The constitution of complex metallic salts. Part VI. The constitution of the phosphine and arsine derivatives of silver and aurous halides. The configuration of the coordinated argentous and aurous complex. J Chem Socμ 1κ2κ–1κ36. doiμ10.103λ/jrλ370001κ2κ. Bruce MI, Nicholson BK, Shawkataly bin O, Shapley JR, Henly T (1λκλ) Synthesis of gold-containing mixed-metal cluster complexes. Inorg Syn 26μ 324–32κ Schöniger W (1λ56) Die mikroanalytische schnellbestimmung von halogen und schwefel in organischen verbindungen. Mikrochim Acta 44μ κ6λ–κ76. doiμ10.1007/BF01262130. Pěnčíková K, Kollár P, Müller Závalová V, Táborská E, Urbanová J et al. (2012) Investigation of sanguinarine and chelerythrine effects on LPS-induced inflammatory gene expression in THP-1 cell line. Phytomedicine 1λμ κλ0–κλ5. doiμ10.1016/j.phymed.2012.04.001. PubMedμ 225λ2163. Talhouk RS, Chin JR, Unemori EN, Werb Z, Bissell MJ (1λλ1) Proteinases of the mammary glandμ developmental regulation in vivo and vectorial secretion in culture. Development 112μ 43λ–44λ. PubMedμ 17λ4314. Zimmermann M (1λκ3) Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16μ 10λ–110. doiμ 10.1016/0304-3λ5λ(κ3)λ0201-4. PubMedμ 6κ77κ45. Kiernan JA (200κ) Histological and histochemical methodsμ theory and practice. 4th ed. Bloxham, UKμ Scion. Geary WJ (1λ71) The use of conductivity measurements in organic solvents for the characterisation of coordination compounds. Coord Chem Rev 7μ κ1–122. doiμ10.1016/S0010-κ545(00)κ000λ-0. Pouchert CJ (1λκ1) The Aldrich library of infrared spectra. Aldrich Chemical Company Press, Milwaukee.
PLOS ONE | www.plosone.org
41. Čajan M, Trávníček Z (2011) Structural (X-ray), spectral (FT-IR and Raman) and quantum chemical investigations of a series of 6benzylaminopurine derivatives. J Mol Struct λλ4μ 350–35λ. doiμ 10.1016/j.molstruc.2011.03.04λ. 42. Maćkowiak M, Kolasiewicz W, Markowicz-Kula K, Wedzony K (2005) Purvalanol A, inhibitor of cyclin-dependent kinases attenuates proliferation of cells in the dentate gyrus of the adult rat hippocampus. Pharmacol Rep 57μ κ45–κ4λ. PubMedμ 163κ2206. 43. Nomiya K, Noguchi R, Ohsawa K, Tsuda K, Oda M (2000) Synthesis, crystal structure and antimicrobial activities of two isomeric gold(I) complexes with nitrogen-containing heterocycle and triphenylphosphine ligands, [Au(L)(PPh3)] (HL = pyrazole and imidazole). J Inorg Biochem 7κμ 363–370. doiμ10.1016/S0162-0134(00)00065-λ. PubMedμ 10κ57λ1κ. 44. Bruce MI, Horn E, Humphrey PA, Tiekink ERT (1λλ6) A novel method of introducing the Au2(PR3)2 (R = Ph, OMe) unit into metal clusters Xray structures of three complexes containing Au2Ru3 cores and of Ru6C(µ-CO)2(CO)14{Au(PPh3)}2. J Organomet Chem 51κμ 121–13κ 45. Nomiya K, Yamamoto S, Noguchi R, Yokoyama H, Kasuga NC et al. (2003) Ligand-exchangeability of 2-coordinate phosphinegold(I) complexes with AuSP and AuNP cores showing selective antimicrobial activities against Gram-positive bacteria. Crystal structures of [Au(2Hmpa)(PPh3)] and [Au(6-Hmna)(PPh3)] (2-H2mpa = 2mercaptopropionic acid, 6-H2mna = 6-mercaptonicotinic acid). J Inorg Biochem λ5μ 20κ–220. doiμ10.1016/S0162-0134(03)00125-λ. PubMedμ 12763666. 46. Demirovic D, Rattan SI (2013) Establishing cellular stress response profiles as biomarkers of homeodynamics, health and hormesis. Exp Gerontol 4κμ λ4–λκ. doiμ10.1016/j.exger.2012.02.005. PubMedμ 225255λ1. 47. Seitz M, Valbracht J, Quach J, Lotz M (2003) Gold sodium thiomalate and chloroquine inhibit cytokine production in monocytic THP-1 cells through distinct transcriptional and posttranslational mechanisms. J Clin Immunol 23μ 477–4κ4. doiμ10.1023/BμJOCI.0000010424.41475.17. PubMedμ 15031635. 4κ. Barreiro E, Casas JS, Couce MD, Gato Á, Sánchez A et al. (200κ) Synthesis, structural characterization, and antiinflammatory activity of triethylphosphinegold(I) sulfanylpropenoates of the type [(AuPEt3)2xspa] [H2xspa = 3-(aryl)-2-sulfanylpropenoic acid]μ An (H2O)6 cluster in the lattice of the complexes [(AuPEt3)2xspa]·3H2O. Inorg Chem 47μ 6262– 6272. doiμ10.1021/icκ00314p. PubMedμ 1κ563κ77. 4λ. Coppi G, Borella F, Gatti MT, Comini A, Dall'Asta L (1λκλ) Synthesis, antiinflammatory and antiarthritic properties of a new tiopronine gold derivative. Boll Chim Farm 12κμ 22–24. PubMedμ 2775517. 50. Girard GR, Hill DT, Dimartino MJ (1λκλ) Evaluation of some tetraalkylammonium gold(I) and gold(III) aurate salts for oral antiinflammatory and antiarthrithic activity. Inorg Chim Acta 166μ 141– 146. doiμ10.1016/S0020-16λ3(00)κ07λλ-λ. 51. Shaw CF, Coffer MT, Klingbeil J, Mirabelli CK (1λκκ) Application of phosphorus-31 NMR chemical shiftμ gold affinity correlation to hemoglobin-gold binding and the first inter-protein gold transfer reaction. J Am Chem Soc 110μ 72λ–734. doiμ10.1021/ja00211a011. 52. Iqbal MS, Taqi SG, Arif M, Wasim M, Sher M (200λ) In vitro distribution of gold in serum proteins after incubation of sodium aurothiomalate and auranofin with human blood and its pharmacological significance. Biol Trace Elem Res 130μ 204–20λ. doiμ10.1007/s12011-00λ-κ330-0. PubMedμ 1λ1λ4667. 53. Saccoccia F, Angelucci F, Boumis G, Brunori M, Miele AE et al. (2012) On the mechanism and rate of gold incorporation into thiol-dependent flavoreductases. J Inorg Biochem 10κμ 105–111. doiμ10.1016/ j.jinorgbio.2011.11.005. PubMedμ 22166353. 54. Lewis D, Capell HA, McNeil CJ, Iqbal MS, Brown DH et al. (1λκ3) Gold levels produced by treatment with auranofin and sodium aurothiomalate. Ann Rheum Dis 42μ 566–570. doiμ10.1136/ard. 42.5.566. PubMedμ 64143κ7. 55. Salemi G, Gueli MC, D’Amelio M, Saia V, Mangiapane P et al. (200λ) Blood levels of homocysteine, cysteine, glutathione, folic acid, and vitamin B12 in the acute phase of atherothrombotic stroke. Neurol Sci 30μ 361–364. doiμ10.1007/s10072-00λ-00λ0-2. PubMedμ 1λ4κ41κ6.
13
November 2013 | Volume κ | Issue 11 | eκ2441
