Transport of the Ruthenium Complex [Ru(GA)(dppe)₂]PF₆ into Triple-Negative Breast Cancer Cells Is Facilitated by Transferrin Receptors.

Article CiteThis:Mol.Pharmaceutics2019,16,1167−1183 pubs.acs.org/molecularpharmaceutics Transport of the Ruthenium Complex [Ru(GA)(dppe) ]PF into 2 6 Triple-Negative Breast Cancer Cells Is Facilitated by Transferrin Receptors † † ‡ † ̃ ‡ Marina A. Naves, Angelica E. Graminha, Legna C. Vegas, Liany Luna-Dulcey, Joao Honorato, Anto ̂ nio C. S. Menezes, § Alzir A. Batista, ‡ and Marcia R. Cominetti *,† † ‡ Laboratory of Biology of Aging (LABEN), Department of Gerontology and Laboratory of Structure and Reactivity of Inorganic Compounds (LERCI), Department of Chemistry, Federal University of Saõ Carlos, CEP 13565-905 Saõ Carlos, SP, Brazil § ́ ́ Campus of Exact Sciences and Technology (CCET), State University of Goias, CEP 75132-903 Anapolis, GO, Brazil * S Supporting Information ABSTRACT: Thetriple-negativebreastcancersubtype(TNBC)is highly aggressive and metastatic and corresponds to 15−20% of diagnosed cases.TNBC treatmentis hampered,because thesecells usually do not respond to hormonal therapy, and they develop resistancetochemotherapeuticdrugs.Ontheotherhand,thesevere side effects of cisplatin represent an obstacle for its clinical use. Ruthenium (Ru)-based complexes have emerged as promising antitumor and antimetastatic substitutes for cisplatin. In this study, we demonstrated the effects of a Ru/biphosphine complex, containing gallic acid (GA) as a ligand, [Ru(GA)(dppe) ]PF , 2 6 hereaftercalledRu(GA),onaTNBCcellline,andcomparedthem to the effects in a nontumor breast cell line. Ru(GA) complex presentedselectivecytotoxicityagainstTNBCovernontumorcells, inhibited its migration and invasion, and induced apoptosis. These effects were associated with the increased amount of transferrinreceptors(TfR)ontumorcells,comparedtonontumorones.SilencingofTfRdecreasedRu(GA)effectsonTNBC cells,demonstratingthatthesereceptorswereatleastpartiallyresponsibleforRu(GA)deliveryintotumorcells.TheRu(GA) compoundmustbefurtherstudiedindifferentinvivoassaysinordertoinvestigateitsantitumorpropertiesanditstoxicityin complex biological systems. KEYWORDS: antitumor effects, apoptosis, ruthenium complex, transferrin, transferrin receptors, triple-negative breast cancer ■ INTRODUCTION biological activity, such as interaction with transporter blood proteins(albumin andtransferrin) anddifferentDNA-binding Triple-negative breast cancer (TNBC) is the most aggressive types, which may turn them into an encouraging and among breast cancers and represents 15 to 20% of all cases. promising class of metallodrugs. This complex/protein TNBC is associated with high resistance to treatment, lower interaction facilitates the complex transportation, increasing survivalrate,andincreasedmetastasistothebrain,bones,lung, and liver.1 TNBC cells lack the expression of the estrogen its solubility in the blood plasma, providing protection to the complexes against premature degradation or elimination and receptor (ER), progesterone receptor (PR), and human leading to the initiation of cell death mechanisms. Also, this epidermal growth factor receptor 2 (HER 2), which impairs ability allows the ruthenium/complexes to be internalized by hormone therapy, leaving chemotherapy as one of the only options for advanced cases.2 transferrin receptors by (TfR)-mediated endocytosis.4 There- fore, these proteins could be responsible for the transport, Cisplatin-resistant neoplastic cells are able to avoid intra- delivery, and uptake of Ru into the tumor cells, decreasing its cellular drug accumulation, prevent its binding to cellular toxicity by reducing unspecific bindings.5−8 It was already targets, and decrease DNA damage recognition as well as signaling to cell death.3 Therefore, the search for new reported that tumor cells express higher amounts of TfR antitumor drugs that could be more specific to neoplastic compared to nontumor cells, and its expression can be cells with fewer side effects is of great importance and correlated with tumor stage or cancer progression.9,10 represents a challenge to chemotherapeutic drug research aiming for a better quality of life for cancer patients. Received: November2, 2018 Ruthenium(Ru)-basedcomplexesemergedinrecentstudies Revised: January10,2019 as an alternative to cisplatin. One perspective of ruthenium Accepted: January11,2019 complexes is that they exhibit properties that can help their Published: January11,2019 ©2019AmericanChemicalSociety 1167 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 .)CTU( 55:43:21 ta 9102 ,32 rebotcO no GRUBNEHTOG FO VINU aiv dedaolnwoD .selcitra dehsilbup erahs yletamitigel ot woh no snoitpo rof senilediuggnirahs/gro.sca.sbup//:sptth eeS Molecular Pharmaceutics Article Furthermore,theaccelerated metabolismof tumorcellsfavors electrochemical system and redox potential. Fc is oxidized at theoxidationreactionofRucompoundsandleadstotheneed 0.43 V (Fc+/Fc). of extra nutrients, including iron. These characteristics may Synthesis and Characterization of the [Ru(GA)- contribute to the increased uptake of Ru complexes by tumor (dppe) ]PF . For synthesis, 0.052 mmol of gallic acid (GA) 2 6 cells, compared to nontumor cells.11,12 and 0.052 mmol of Et N were dissolved in a solution 3 SeveralstudiesreporttheantitumoreffectsofRucomplexes containing 5 mL of methanol and 25 mL of CH Cl under 2 2 both in in vitro8,13−18 and in vivo.8,19−21 The most notable an argon atmosphere. Subsequently, 0.05 g (0.052 mmol) of actionsaretheirhighcytotoxicityagainstarangeoftumorcell cis-[RuCl (dppe) ]28 and 0.0095 g (0.052 mmol) of KPF 2 2 6 lines and induction of apoptosis by DNA damage, cell cycle were added to thesolution. After 8 h of reaction, CH Cl was 2 2 arrest, and ROS-mediated mitochondrial dysfunction.22 Gallic evaporated,and1mLofwaterwasadded.Theprecipitatewas acid (GA, 3,4,5-trihydroxybenzoic acid) is a triphenol widely then filtered, washed with water and hexane, and dried in distributedinplantkingdom,relatedtoantimicrobial,antiviral, vacuum. The final efficiency of reaction was 92% (0.058 g/ anti-inflammatory, antioxidant, and antitumor activities, molecularweight:1211.97g/mol).Thecomplexwasdissolved through the activation of different signaling pathways.23,24 in DMSO (Sigma-Aldrich) and applied to cells in a final Inserting ligands with a previous recognized biological activity concentration of 1% DMSO. [Ru(GA)(dppe) ]PF ·1/ 2 6 to Ru complexes can lead to synergistic effects with the metal 2CH Cl : Elemental analysis: found (%): C, 56.78; H, 4.32; 2 2 center or with the complex being an interesting strategy for calc(%):C,56.97;H,4.34.Absorptionspectrum:UV−vis[(n- metallodrug development. Recently, a work published by our octanol) λ (ε)]: 256 nm (36702 M −1 cm −1, intraligand group showed that the natural products lawsone (2-hydroxy- π−π*), 346 nm (2958 M −1 cm −1, (MLCT) dπ → Ru 1,4-naphthoquinone) and lapachol (1-hydroxy-3-(3-methyl- 3pσ*dπ (dppe) and dπ Ru → π* (GA) orbitals) (Figure S2). Molar but-2-enyl)naohthalene-1,4-dione), when coordinated to conductivity in CH 2 Cl 2 : 54.9 S cm2 mol −1. IR (KBr) cm −1: ruthenium/phosphinecomplexesincreasedtheantileishmanial, ν(OH) 3510, ν(CH) 3057, ν as (COO − ) 1615, ν s (COO − ) antiplasmodial,25 and antitumoral26 activities of the new 1499, ν s (C=C arom ) 1485, ν(C=C phosp ) 1434, δ (OH phenol ) complexes, when compared with the free ligands. Taking this 1311, ν (C−O phenol ) 1280, ν (P−CH) 1096, ν(P−F) 844, hypothesisintoconsideration,theaimofthepresentstudywas γ(arom) 697, δ (P−F) 558, ν(Ru−P) 512, ν(Ru−O) 481 P t a o g F a 6 i e ( n v [ s R a t l u u M ( a G t D e A A ) t - h ] M ) e , B e - w 2 ff 3 i e t 1 h cts T G N o A B f C t a h s c e e l l i l c g s o a . n m d p , le a x s [ a Ru c ( y G to A to )( x d ic pp a e g ) e 2 n ] t - ( 3 ( J G F P i − A g P u ) r − 1 e H 6 S .9 ] 3 + ) ( a . H n N d z M ) 1 . 0 R 6 M 7 3 A . 1 0 P L 8 { D 9 1H I [ - M T } O − (p F P p F m ( ] m ) +. / ( 3 z 1 C ) P : H (1 8 2 H C 98 ) l 2 . ) 0 N : 9 M 3 δ R [ 5 M ( 9 1 .9 − 6 , 1 P . 5 F 9 8 6 7 − . 6 4 ■ MHz, CDCl -d, 298 K): δ(ppm); 6 58.7 (t) and 58.0 (t); 2J = 3 17.2Hz.1HNMR(400.132MHz,CDCl -d,298K):δ(ppm): EXPERIMENTAL SECTION 3 9.93 (m, OH of GA); 7.78 (4H, m, dppe-ring); 7.54 (5H, m, All chemicals used in the synthesis of the complex were of dppe-ring);7.45(4H,m,dppe-ring);7.37(3H,m,dppe-ring); reagent grade or comparable purity. Solvents were purified by 7.31 (3H, m, dppe-ring); 7.26 (CDCl ); 7.12 (4H, m, dppe- 3 standard methods, and the chemicals and reactions were ring);7.06(4H,m,dppe-ring);6.98(6H,m,dppe-ring);6.90 handled under an argon atmosphere. RuCl 3 ·3H 2 O, 1,2- (3H, m, dppe-ring); 6,34 (2H, s, CH of GA); 2.81 (2H, m, bis(diphenylphosphino)ethane (dppe), and GA were pur- CH dppe); 2.26 (2H, m, CH dppe); 2.08 (2H, m, CH 2 2 2 chased from Sigma-Aldrich. The spectra in theinfrared region dppe); 1.44 (2H, m, CH dppe) (Figures S4 and S5). 13C 2 (IR) were recorded on an FT-IR Bomem-Michelson 102 NMR (100.62 MHz, CDCl -d) δ(ppm): 181.7 (COO of spectrometer in the 4000−250 cm −1 region, using cesium GA); 143.4 (CO-meta of 3 GA); 136.2 (CO-para of coo G rd A); iodide (CsI) pellets. The UV−vis spectra of the complex in 136.1−128.2 (CH dppe-ring); 124.5 (C-ipso of GA); 107.7 dichloromethane (CH 2 Cl 2 ) were recorded on a Hewlett- (CH-ortho of GA); 77.0 (CDCl 3 ); 33.2 (CH 2 of dppe); 16.6 Packard diode array-8452A spectrometer. Molar conductivity (CH of dppe) (Figure S6). 2 valueswereobtainedat293Kusingthecomplexinasolution CellLinesandCulture.NontumorbreastcellsMCF-10A, at 10 −3 mol L −1 in CH 2 Cl 2 and in a Meter Lab CDM2300 kindly donated by Prof. Dr. Robin Anderson from the Olivia instrument. The microanalyses were performed using a Newton John Cancer Research Institute (Australia), were FISIONS CHNS, mod. EA 1108 microanalyzer. Nuclear culturedinDulbecco’sModifiedEagleMediumF12(DMEM/ magneticresonance(NMR)of1H, 31P{1H},and13C{1H}was F12) supplemented with 5% of horse serum, 20 ng/mL of recorded on a Bruker DRX 400 MHz spectrometer, using epidermalgrowthfactor(EGF),0.5μg/mLofhydrocortisone, tetramethylsilane as the reference and CDCl as the solvent. and 10 μg/mL of insulin. TNBC MDA-MB-231 and MDA- 3 The 31P{1H} chemical shifts are reported in relation to MB-468cellswereobtainedfromBancodeCel ́ ulasdoRiode phosphoricacid(H PO ),85%.Massspectrawereobtainedby Janeiro (BCRJ) (Rio de Janeiro, RJ, Brazil) and cultured in 3 4 direct infusion in a MALDI-TOF Autoflex Speed (Bruker) DMEM supplemented with 10% fetal bovine serum (FBS). massspectrometer,andsampleswerepreparedusingsinapinic TNBC MDA-MB-468 cells were a gift from Dr. Alexandre acid as the matrix,27 employing acetonitrile as solvent. Bruni Cardoso (Department of Biochemistry, University of Electrochemical Characterization. Cyclic voltammetry Saõ Paulo, Brazil) and were maintained in DMEM high experiments were carried out at room temperature in a glucose medium supplemented with 10% fetal bovine serum solution of CH Cl containing 0.10 M tetrabutylammonium (FBS).Allcelllinesweremaintainedat37°Cinanincubator 2 2 perchlorate, Bu NClO (TBAP - Fluka Purum), using a BAS- with95%O and5%CO ,and1%penicillin/streptomycinwas 4 4 2 2 100B/Welectrochemicalanalyzer(BioanalyticalSystemsInc.). added to each culture medium. During these experiments, working and auxiliary electrodes of Cytotoxicity. For evaluation of Ru(GA) effects of on cell platin(Pt)wereusedaswellasaLuggincapillaryprobeanda viability,theMTTassaywasperformed.29Cells(1×104cells/ reference electrode of silver (Ag) and silver chloride (AgCl). well) were seeded into a 96-well plate for 24 h. The complex Ferrocene (Fc) was employed as a calibrator for the Ru(GA) and its ligand GA were dissolved in sterile DMSO. 1168 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Each complex sample (1 μL) was added to each 100 μL of (Nikon,T5100)withmagnificationof40×inordertoobserve medium (1% DMSO final concentration). Cells were exposed the wound closure. Photos were taken at 0, 6, 12, and 24 h. to Ru(GA) (0.0122 to 75 μM) for 24 and 48 h. Cisplatin The closure area of the migrating cells was measured using dissolved in DMF was used as a positive control for cell ImageJ software, and the percentage of wound closure was cytotoxicityatthesameconcentrations.GAwasaddedtocells calculated, comparing time zero and 24 h, using eq 1 as at concentrations from 3.12 to 200 μM. After treatment, the previously described.33,34 medium was removed, and 100 μL of MTT (1 mg/mL) was addedtoeachwell.Cellswereincubatedfor4h,andformazan %wound closure = [(A t=0h − A t=Δh )/A t=0h ]100 (1) crystalsweresolubilizedin100μLofDMSO.Theabsorbance Invasion.TheeffectofRu(GA)ontumorcellinvasiveness was measured in a spectrophotometer at 540 nm. The was determined by the ability to transmigrate through a layer reference of 100% of viability was the control cells (with no of Matrigel in transwell chambers (BD Biosciences) as complex). IC 50 was calculated using Hill’s equation in previously described.35 Briefly, inserts containing Matrigel GraphPad Prism software [y = y min + (y max − y min )/(1 + (gelatinous protein mixture, which simulates the extracellular 10log(IC50 −x)), where y is the value on the Y axis; y max is the matrix)werepreviouslyhydratedwithmediumfreeofFBSfor maximum value for y; y min is the lower value for y; x is the 2hinanincubatorat37°C.Then,MDA-MB-231cells(0.5× corresponding value on the X axis, and IC 50 is the 105cells/insert)wereplacedontheupperchambers(BioCoat concentration of compound that gives a response half way MatrigelInvasionChambersKitfromBDBiosciences)ina24- between y max and y min ]. The selectivity index (SI) was wellplate,alongwithincreasingconcentrations(0.195to3.12 calculated as the ratio of the IC 50 of MCF-10A/IC 50 of μM)ofRu(GA)inafinalvolumeof350μLofmediumfreeof MDA-MB-231 and of IC 50 of MCF-10A/IC 50 of MDA-MB- FBS. In the lower chambers, 750 μL of medium containing 468.30 10% FBS was added to act as a chemoattractant. Cells were Partition Coefficient (P). The Ru(GA) water−octanol allowedtomigratefor22hat37°C,andthosethatremained partition coefficient was determined using the shake flash intheupperchamberswerethenremovedusingacottonswab. method.31 The complex was added into an equal volume Cellsthatinvadedtotheothersideof thechamberwerefixed mixture of water (750 μL) and octanol (750 μL), under withmethanolfor5minandstainedwithtoluidineblue(1%) shaking for 24 h at 1000 rpm and 27 °C. Samples were and sodium borate (1%) in water for 5 min. Membranes of centrifugedfor5minat500rpm,andtheorganicandaqueous each insert were cut and placed in histological slides. Inserts phases were separated. The concentration of drug in each were photographed under microscope in 5−10 random areas, phasewasmeasuredbyUV−visinordertodeterminevaluesof and cells were counted in ImageJ software. Treatments were P using the formula log P = log(C o /C w ) where C o and C w are compared to the positive control for invasion (cells with no the molar concentrations of the complex in octanol and in compound, with 10% FBS in the lower chamber) and to the water, respectively. negative control (cells with no compound and no FBS in the Morphology. MDA-MB-231 and MCF-10A cells (1.0 × lower chamber). 105/well) were seeded in a 12-well plate and after 24 h were Phalloidin Staining. To evaluate the damage on the exposedtoincreasingconcentrationsofthecomplex(0.049to cytoskeletal organization caused by the complex, cells were 25 μM) for an additional 48 h. Cells were examined under an stained with Alexa Fluor 488 phalloidin (ThermoFisher inverted optical microscope (Nikon, T5100) with amplifica- Scientific). MDA-MB-231 and MCF-10A (1 × 104 cells/ tion of 100×. Photos were taken at 8, 24, and 48 h. The well) wereseededinablack 96-wellplatewith aclearbottom efficiencyoftreatmentwascomparedtocontrolcells(withno (Corning Incorporated)andafter 24hofincubationat37°C treatment). were exposed to increasing concentrations of the compound Colony Formation. MDA-MB-231 (3 × 102/plate) were (12.5 to 50 μM) for 6 h. After treatment, medium was seededinto6cmPetridishesandafter24hofincubationat37 removed,andcellswerefixedwithparaformaldehyde(3.7%)in °C were exposed to increasing concentrations of the complex PBSfor30minandpermeabilizedwithtritonX-100(0.1%)in (0.0122 to 0.78 μM) for 48 h. After treatment, the medium PBSfor5minatroomtemperature.Thereactionwasblocked was replaced by fresh medium with no compound, and cells bytheadditionofbovineserumalbumin(BSA)2%inPBSfor were incubated for another 10 days. After incubation, cells 30min.CellsweremarkedwithAlexaFluor488phalloidinfor were fixed with methanol and acetic acid (3:1) for 5 min and 20 min and stained with DAPI 1 μg/mL (Life Technologies, stained with methanol with 0.5% crystal violet for 15 min. Carlsbad, CA) for nuclear labeling for 10 min. Images were Coloniesformedwereanalyzedinbothnumberandsize,with obtained using an automated microscope ImageXpress Micro the assistance of ImageJ software, using the tool “particle XLS System (Molecular Devices) with magnification of 400× analysis” that measures the area and average size of each and digital confocal adjust. colony and counts the number of colonies in each selected Induction of Nuclei Damage. To evaluate whether image.32 Ru(GA)is capable of causing nuclei damage, which can result Migration.TheeffectsofRu(GA)ontumorcellmigration in apoptosis, MDA-MB-231 cells (1.0 × 104 cells/well) were were evaluated by a wound healing assay. MDA-MB-231 cells seeded in a black 96-well plate with a clear bottom (Corning (2.0×105/well)wereseededintoa12-wellplate,andafter24 Incorporated). After 24 h of incubation at 37 °C, cells were h, a scratch was made in the central portion of every well, treated with different concentrations (12.5 to 100 μM) of removing the cells from this region with the help of a Ru(GA) for8 hand then fixed with methanol for 10 min and micropipette tip. Wells were washed with PBS to remove stained with 1 μg/mL DAPI (Life Technologies, Carlsbad, unbound cells. Next, cells were exposed to increasing CA) for nuclear labeling for 10 min. Images were obtained concentrations of the complex (0.195 to 3.12 μM) and using an automated microscope ImageXpress Micro XLS incubatedfor24h.Duringtheincubationtime,theareaofthe System (Molecular Devices) with magnification of 400× and wound was examined using an inverted optical microscope digital confocal adjust. Additionally, a quantification of cells 1169 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article with condensed chromatin and fragmented nuclei after ATT GCT TTC C3′), caspase-9 − NM_001229.3 − treatment was performed. (Forward: 5′GGAAGAGCT GCAGGTGGAC3′;Reverse: Apoptosis. Apoptosis induced by Ru(GA) was also 5′CCT GCC CGC TGG ATG TC3′), caspase-8 − evaluatedbyflowcytometry,usingaPE−Annexin-VApoptosis NM_001080125 − (Forward: 5′GGA TGA GGC TGA Detectionkit(BDBiosciences).MDA-MB-231andMCF-10A CTT TCT G3′; Reverse: 5′CTG GCA AAG TGA CTG cells (0.7 × 105 cells/well) were seeded into a 24-well plate GAT G3′), GAPDH − NM_001256799.1 − (Forward: and after 24 h of incubation at 37 °C were exposed to 5′GAC TTC AAC AGC GAC ACC CAC3′; Reverse: increasing concentrations of Ru(GA) (3.12 to 50 μM) for 24 5′CAC CAC CCT GTT GCT GTA G3′). h. Cells were incubated in the dark with 2.5 μL of PE− Western Blotting. MDA-MB-231 cells (1.0 × 106/plate) annexin-V and 2.5 μL of 7AAD for 15 min. Analyses were were incubated for 24 h with increasing concentrations of the performed in an Accuri C6 (BD Biosciences) flow cytometer, complex(6.25to25μM)inPetridishes(6cm)at37°C.After recording15.000eventsforeachcondition,afterscrapingcells incubation, cells were lysed using CelLytic M reagent (Sigma- fromwells.Emittedfluorescencebyeachdyewasquantifiedin Aldrich),accordingtofabricantinstructions,inordertoobtain CellQuestsoftware(BDBiosciences),anditisproportionalto the protein content. Protein concentrations of supernatants the percentage of cells in apoptosis. The apoptotic rate of were determined using a Pierce BCA Protein Assay kit treatment was compared to control (cells with no treatment). (Thermo-Scientific). Protein samples (15 μg) were applied Camptothecin at 32 μM was used as the positive control. onto 4−20% Tris-glycine gels for 1 h at 100 V, transferred to CellCycle.MDA-MB-231cells(1.0×106cells/well)were nitrocellulosemembranes(BioRadLaboratories),andblocked placedintoa6-wellplateandafter24hofincubationat37°C with 1% casein (BioRad Laboratories) for 1 h, with all were treated with increasing concentrations of the compound processes following fabricant instructions. Subsequently, (0.195 to 3.12 μM) for 24 h. Then, cells were washed with membranes were incubated overnight with anti-caspase-3 PBS and fixed with ethanol (70%) for 24 h at −20 °C. After (Abcam 1:1000), anti-caspase-9 (Novusbio 1:1000), anti-Bax fixation, cells were washed with PBS, incubated with RNase (Abcam 1:1000), and anti-Bcl-2 primary antibodies (Abcam (0.2mg/mL) for30 min at37 °C,and incubated with PI(20 1:500), followed by incubation with HRP-conjugated secon- μg/mL) for 1 h on ice and in the dark. DNA contents were dary antibody (1:5000) (Abcam) for 2 h. β-Actin was used as analyzed by flow cytometer Accuri C6 (BD Biosciences) an endogenous control. Proteins were analyzed by chem- recording 20.000 events for each condition and cells in each iluminescence using Clarity Western ECL Substrate (BioRad phase of the cell cycle were quantified in CellQuest software Laboratories).SpecificbandswerevisualizedwithaChemiDoc (BD Biosciences). Camptothecin at 32 μM was used as the MP imager (BioRad Laboratories), quantified with ImageJ positive control. software, and normalized to β-actin. Reverse Transcriptase Quantitative Real-Time PCR Interaction with DNA. Two experiments to evaluate the (RT-qPCR). MDA-MB-231 and MCF-10A cells (1.0 × 106 interaction of Ru(GA) with DNA were conducted: spectro- cells/dish)wereseededinto6cmPetridishesandafter24hof scopic titrations and fluorescence suppression of thiazole incubationweretreatedwiththecomplex(6.25to25μM)for orange (TO). All measurements with DNA (calf thymus (ct) 20 h. Total RNA was extracted using TRIzol reagent from Sigma-Aldrich) were performed in a Tris-HCl buffer (5 (Invitrogen), according to fabricant instructions. Next, RNA mMTris-HCland50mMNaCl,pH7.4).DNAconcentration was quantified in Nanodrop (Thermo Scientific), and 1.5 μg per nucleotide was determined by absorption spectrophoto- was reversely transcribed to cDNA using the Enhanced Avian metric analysis using a molar absorption coefficient of 6.600 RTFirstStrandSynthesiskit(Sigma-Aldrich).Reactionswere mol −1 L cm −1 at 260 nm.37 The spectroscopic titrations were adjusted to each primer, depending upon its melting carriedoutbyaddingincreasingamountsofDNAtoasolution temperature, using Rotor-Gene 6 software in a Rotor-Gene of the complex, at a fixed concentration (20 μM), in a quartz RG 3000 (Corbett Life Science). In brief, 12.5 μL of SYBR cell and recording the UV−vis spectrum of the complex after Green JumpStart Taq ReadyMix (Sigma-Aldrich), 10 μL of each addition. The intrinsic binding constant K b was pure water, 2 μL of the primer pair, and 0.5 μL of cDNA at a determined by eq 2, where [DNA] is the concentration of finalconcentrationof500nmolL −1madeafinal25μLaliquot DNA in a determined measurement, and the apparent of thereactionsystem.After amplification, analyses of melting absorptioncoefficients(ε a ,ε f ,andε b )correspondtospectrum curves were performed by monitoring the fluorescence of complex recorded after addiction of specific [DNA], associated with the temperature increase. Samples were spectrum of complex in the totally bound form, and spectrum amplified in triplicate. The internal calibrator, used as a basis of free complex, respectively.38 tostandardizetheresultsofexpression,wasthecontrolgroup, [DNA] [DNA] 1 ΔCts average. Calibration was determined by ΔΔCt = ΔCt = + (gene of interest) − ΔCt (internal control gene). Gene ε a − ε f ε b − ε f [K b (ε b − ε f )] (2) expression was assessed by relative quantification, using the For fluorescence suppression of TO, measurements were formula 2 −ΔΔCt and GAPDH expression as the internal performed by applying a solution of ct-DNA (100 μM) with control.36 The expression of genes related to apoptosis in TO (10 μM). The quenching of intensity of TO emission treated cells was compared to the control (cells with no between540to680nm(excitationat520nm)wasmonitored treatment). Primers used: Bax − NM_001291428.1 − using Ru(GA) as a suppressor upon increasing concentrations (Forward: 5′CAT CCA GGA TCG AGC AGG3′; Reverse: (0 to 10 μM), dissolved in DMSO. Experiments were carried 5′GCA TGC GCT TGA GAC ACT C3′), Bcl-2 − out in triplicate in an opaque 96-well plate. Measurements NM_000633.2 − (Forward: 5′GGT GGG AGG GAG GAA wereperformedinaSynergyH1BioTekfluorimeterat298K. GAAT3′;Reverse:5′GCAGAGGCATCACATCGAC3′), Interaction with Human Albumin and Transferrin. caspase-3 − NM_004346.3 − (Forward: 5′GTG CTA CAA Evaluation of Ru(GA) interaction with human serum albumin TGC CCC TGG AT3′; Reverse: 5′GCC CAT TCA TTT (HSA) and apo-transferrin (apo-Tf) was performed by 1170 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article measuring the decrease of tryptophan fluorescence. Proteins Statistical Analysis. All experiments were carried out in wereadded,separately,at2.0μMinTris-HClbufferatpH7.4. triplicateandindependentlytoensurethereliabilityofresults. Tryptophanquenchingintensityemissionbetween300to500 Data were expressed as mean ± SD. Determination of IC , 50 nm(excitationat280nmforHSAand270nmforapo-Tf)was statisticalmeasurements,andgraphicsweremadeinGraphPad monitored using Ru(GA) a as suppressor at increasing Prism software (Intuitive Software for Science, San Diego, concentrations (5 to 50 μM) and dissolved in DMSO, while CA). Results were compared through one-way ANOVA the HSA and apo-Tf concentrations were kept constant in all (acceptable significance p < 0.05), followed by Dunnett’s samples. Experiments were carried out in triplicate in an test, which compares all groups with the control group. opaque 96-well plate. Measurements were performed in a Data Availability. All data generated or analyzed during SynergyH1BioTekfluorimeterat298and310K.Thebinding thisstudyareincludedinthispublishedarticleaswellasinthe constant(K ),whichindicatesamoderatetostronginteraction S■upporting Information. b between protein and complex, and the number of binding RESULTS sites/moleculeof protein (n) were determinedby plottingthe double-loggraphofthefluorescence datausinglog[(F −F)/ Synthesis and Characterization of Ru(GA). Ru(GA) 0 F] = log K + nlog[Q]; where F and F are the fluorescence was prepared in the presence of the complex cis- b 0 intensities in the absence and presence of complex, [RuCl (dppe) ], GA ligand, and KPF at room temperature. 2 2 6 respectively, and [Q] is the concentration of the complex. The complex with the general formula [Ru(GA)(dppe) )]PF 2 6 Results were also analyzed by the classical Stern−Volmer was formed by substitution of chloride ligands and equation in order to calculate the quenching constant.39 The coordination of the monoanionic chelated GA ligand as thermodynamicparameter(ΔH°)wascalculatedfromln(K / shown in Figure 1. Good yields of syntheses and analytically b1 K ) = [(1/T ) − (1/T )]ΔH/R, where K and K are the b2 1 2 b1 b2 bindingconstantsattemperaturesT andT inKelvin,andRis 1 2 the gas constant. Furthermore, the changes in free energy (ΔG)andentropy(ΔS)werecalculatedfromΔG=−RTlnK = ΔH − TΔS.40 Transport of Ru(GA) by Albumin and Transferrin. To evaluate the interference of albumin and transferrin in the transportandinteractionofRu(GA)withcells,MDA-MB-231 andMCF-10A(1.0×104cells/well)wereseededina96-well plate and after 24 h were exposed for 48 h to a concentration neartheIC calculatedintheMTTassay(0.5μMforMDA- 50 MB-231 and 3.12 μM for MCF-10A). Additionally, increasing concentrationsofbovineserumalbumin(BSA,Sigma-Aldrich, 20 to 300 μg/mL) and human transferrin (apo-Tf, Sigma- Aldrich, 0 to 100 μg/mL) were added to the cells. The percentage of viable cells in each treatment was evaluated by Figure1.Structuresofruthenium(II)complexeswithGAanddppe, an MTT assay. The reference of 100% of viability was the asproposed by NMR. control cells (with no treatment), and cells that received only protein were treated with the highest concentration shown pure complexes were obtained as determined by elemental (300 μg/mL for BSA and 100 μg/mL for apo-Tf). analyses. Values calculated for elemental analyses are in Anti-TfRStaining.MDA-MB-231andMCF-10Acells(1.0 agreementwiththetheoreticalvalues,suggestingtheformation ×106) wereseparately fixedwithparaformaldehyde (3.7%) in of mononuclear species of Ru(II). Molar conductivity PBS for 30 min and with 0.3 M glycine in PBS for 10 min. measurements of the complex in dichloromethane at 298 K CellswereblockedwithBSA(10%)inPBSfor1h.Afterthese indicate1:1electrolyte,asacation[Ru(GA)(dppe) )]+species 2 processes, cells were stained with anti-transferrin receptor and an anion PF − (counterion), corroborating with the 6 antibody (Abcam #ab1086) at 2 μg/mL for 1 h and with proposed formulations for the complex [Ru(GA)(dppe) )]- 2 secondaryantibodyanti-MsFITC(Abcam)at1μg/mLfor40 PF . The proposed composition of the complex was also 6 min. Analyses were performed in an Accuri C6 (BD confirmed by mass spectrometry analysis, where ESI-MS(+) Biosciences) flow cytometer recording 20.000 events. Emitted spectra showed a signal at 1067.089 Da (calc.: 1067.190 Da) fluorescence was quantified in CellQuest software (BD corresponding to the loss of free hexafluorophosphate ions Biosciences) proportionally to the amount of cells expressing [M−PF ]+, followed by the loss of the free GA, forming [M− 6 TfR. Cells without antibody incubation were used as negative PF −(O−O)−H]+ with a peak at 898.093 Da and also peaks 6 control. of low intensity, which correspond to degradation of the free SmallInterferingRNATransfection.MDA-MB-231(7.0 ligandgroup.ThesedataarelistedintheExperimentalSection × 105 cells) was seeded into T25 culture flasks. The two and in Figures S1−S5. sequences of TfR small interfering RNA (siRNA) and the The infrared spectra of the GA ligand show bands control sequence (NT) were purchased from ThermoFisher characteristic of the vibrations ν(OH), ν (C=O), ν(C=O), as s Scientific (Ambion #4390824 TFRC s725 and s727). siRNA and δ(OH)in the regions of 3495, 1666, and 1448, and 1317 sequences were transfected into cells using DharmaFECT4 cm −1,respectively.Inthespectraofthecomplex,thereisashift from Dharmacon following the manufacture’s protocol. After of bands ν and ν(C=O) (Figure S3) in lower frequencies as s 48 h, cells were used for viability assays (with or without indicating the coordination. To confirm the mode of transferrin), and after 72 h, cells were used for anti-TfR coordination of the GA ligand, the difference among the incubation. symmetric and asymmetric stretch of νCOO − (Δcm −1) was 1171 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article calculated.41 According to Δcm −1 values calculated for the of the precursor cis-[RuCl (dppe) ] (E = 0.94; E = 0.87 2 2 pa 1/2 ligandandcomplex,thedisplacementofν COO − (Δ = V), indicating stabilization for the ruthenium metal center. as ligandfree 198 cm −1 and Δ = 116 cm −1) showed a smaller This stabilization is due to the substitution of two chlorine complex wavelength to the complex, confirming the bidentate donors by a monoanionic carboxylic group chelating the GA coordination of the carboxylate ion to the ruthenium metal ligand. The complex has also an irreversible process (III), at center (Table 1). approximately 1.2 V, which is assigned to the oxidation of phenol groups from the GA ring.42 The stability of Ru(GA) Table 1. Vibrational Frequencies of the Carboxylate Ion of was analyzed by 31P{1H} NMR in DMSO/DMEM and by the Free and Coordinated Ligand as Well as the NMR UV−vis in DMSO for 72 h, without structural modification 31P{1H} and Electrochemical Potential of Ru(GA) (Figures S8 and S9). Cytotoxicity of Ru(GA) and Morphological Observa- compounds ν as COO− ν s COO− (cm Δ −1) 31P{ δ 1H} ( 3J H P− z P ) E (V 1/ ) 2 t b i a o s n al s A .C ), y M to D to A xi - c M it B y - o 4 f 6 t 8 he (T R N u( b G a A sa ) l a B g ) a , i a n n st d M M D C A F - - M 10 B A -2 c 3 e 1 lls ( w T a N s GA 1666 1468 198 - - - evaluated, results expressed as IC values are described in 50 Ru(GA) 1615 1499 116 58.7; 17.2 1.59 Table 2, and concentration−response curves are shown in 58.1 Figure 3A. Ru(GA) exhibited higher cytotoxicity against TN cells as seen by the lower IC value when compared to The 31P{1H} NMR spectrum of the complex in CDCl 50 3 nontumor MCF-10A, with selectivity indexes (SI) of 7.19 (Figure 2A and Table 1) indicated two triplets with chemical (MDA-MB-231) and 5.82 (MDA-MB-468) at 48 h of treatment. Cisplatin, tested for comparison, also showed higher cytotoxicity on TN cells compared to MCF-10A. An MTT assay wasalso performedin24 h of Ru(GA) treatment, andtheIC resultscalculatedwere4.50±0.29μMforMDA- 50 MB-231,1.62±0.20μMforMDA-MB-468,and12.04±2.33 μM for MCF-10A (SI 2.68 for MDA-MB-231 and 7.43 for MDA-MB-468, data not shown), demonstrating that the selectivity of the complex for tumor cells is maintained over time. A viability assay with the precursor cis-[RuCl (dppe) ] 2 2 was not possible, since this compound, under the same solubilization conditions of Ru(GA), is poorly soluble in aqueousmediumaswellasinorganicsolventsallowedforcell culture, such as DMSO. Beyond the advantage of increasing the solubility of the new complex formed, the coordination between the ruthenium precursor and GA resulted in a more cytotoxic compound, when compared to free GA (Table 2). Figure 2. 31P{1H} NMR spectra (CDCl) of the complex Ru(GA). In order to investigate the effects on cell morphology, 3 145× 99mm (300 ×300 DPI). Ru(GA) was incubated with MDA-MB-231 and MCF-10A at different concentrations for 48 h. MDA-MB-231 cells treated shiftsat58.7(t)and58.1(t)ppm,with3J P−P =17.2Hz,which with Ru(GA) at 12.5 μM for 8 h presented a decrease in cell is a typical standard for bisphosphines in the cis position (two density compared to untreated cells (Figure S10). In 24 h of identical molecules completing the coordination sphere). One treatment at 6.25 μM, the density of cells decreased, and at groupofbisphosphinehastwotypesofphosphorusatoms(P 12.5 μM, morphology was profoundly altered, promoting the A and P ), each one with two chemically equivalent but appearance of round cells, which is indicative of cell B magnetically different neighbors, thus generating triplets. The detachment and probably cell death. In 48 h, these alterations redox behavior of the metal complexes was investigated by areevidentatallconcentrationstested.Ontheotherhand,the cyclic and differential pulse voltammograms (Figure S7 and mostevidentalterationpromotedbyRu(GA)onMFC-10Ais Table 1). Electrochemical experiments were carried out in the decrease in cell density at 6.25 μM in 24 h, with the CH Cl and 0.1 M TBAP, 100 mV s −1, Ag/AgCl. The cyclic appearanceofsomeroundcellsat3.12μMin48h.Increasing 2 2 voltammogramofthecomplexshowsaquasireversibleprocess detachment of these cells can be seen at 25 μM in 24 h and (I/II) corresponding to a one-electron redox of the redox 6.25 μM at 48 h (Figure S10). These results corroborate with coupleRuIII/RuII(E =1.64V;E =1.59V).TheE value the selective cytotoxicity of Ru(GA) for the breast tumor cell pa 1/2 1/2 foundforthecomplexwasconsiderablymoreanodicthanthat line when compared to the nontumor cell line. Table 2. Cytotoxic Effects of Ru(GA) and Cisplatin on Breast Cell Lines after 48 h of Treatment IC ±SD(μM)a 50 compound MDA-MB-231 MDA-MB-468 MCF-10A SIb logPb GA >200 163.3±7.2 >200 n.a.c n.a.c Ru(GA) 0.81±0.08 1.00±0.10 5.82±0.33 7.19;5.82d 0.277±0.004 cisplatin 2.43±0.20 0.43±0.05 29.45±0.85 12.12;68.5d −2.53e aIC values are expressed as mean ± SD of three independent assays in triplicate. bSelectivity index (SI) was calculated as described in the 50 ExperimentalSection.cn.a.=notapplicable.dSIforMDA-MB-231andMDA-MB-486,respectively.eLogPofcisplatinwasbasedonBussetal.70 1172 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Figure3.CytotoxicityofRu(GA).(A)Concentration−responsecurvesofMDA-MB-231andMCF-10AcellsafterRu(GA)treatment(0.022to75 μM)for48h.(B)ClonogenicassayofMDA-MB-231cellstreatedwithindicatedconcentrationsofRu(GA)for48h.AphotographofPetridishes in a representative experiment is shown along with graph quantifications of colony number and size. Data represents mean ± SD from three independentassaysintriplicate. Significanceatthe *p < 0.05,**p <0.001, ***p < 0.0001levelsusing ANOVAand Dunnet’s test. Ru(GA)concentrationsstartingfrom0.049μMsignificantly cytoskeleton, reducing its density and promoting F-actin inhibited colony number and size in MDA-MB-231 after 48 h filament condensation, even with less time of treatment. For of treatment, indicating a cytotoxic and cytostatic behavior, MCF-10A, these changes only appear with 50 μM Ru(GA) respectively.Atthehighestconcentration(0.78μM),Ru(GA) (Figure5).Likecisplatin,Ru(GA),asshowninthisstudy,was completelyinhibitedtheTNBCcolonyformation(Figure3B). able to selectively cause damage to the cytoskeleton in tumor Inhibition of Cell Migration and Invasion. To evaluate cells, which may indicate an initiator factor for apoptosis the effects of Ru(GA) on other cellular aspects, such as signaling,withconsequentalterationintheexpressionofgenes migration and invasion, wound healing and Boyden chamber and proteins related to the process.43 assays were performed, respectively. Results from wound Induction of Apoptosis. The capacity of Ru(GA) in healing show that in 24 h of treatment, the complex inducing apoptosis in MDA-MB-231 cells was evaluated by significantly inhibited wound closure in approximately 10% PE−annexin-VandDAPIstaining.Resultsshowthatapoptosis at 0.195 μM, 20% at 0.78 μM, and 50% at 1.56 and 3.12 μM in tumor cells was induced in a concentration-dependent when compared to untreated cells (Figure 4A). These effects manner(Figure6A),withearlysignificantinductionstartingat are related to cell migration rather than cell death, since all 6.25μM,reachingapproximately60%ofapoptoticcellsatthe Ru(GA) concentrations tested are not cytotoxic at this highestconcentration.However,onlyconcentrationsof25and incubation time. Results from the invasion assay are shown 50 μM were needed to induce significant apoptosis in MCF- in Figure 4B as well as evidence that treatment significantly 10A, with approximately 40 and 50% of the apoptotic rate, inhibited invasion of TNBC cells at 1.56 and 3.12 μM. respectively (Figure 6A). DAPI staining showed that 8 h of Phalloidin Staining. Alterations in cytoskeleton proteins treatment with the complex resulted in more chromatin can interfere in mechanisms of migration and invasion in condensation,nucleidamage,andpresenceofapoptoticbodies tumorcells.ToevaluatewhetherRu(GA)iscapableofaltering compared to untreated MDA-MB-231 cells (Figure 6B). The a tumor cell’s cytoskeleton and therefore responsible for the triggeringoftheapoptosisprocesswasconfirmedbyadditional decrease in cell motility and invasiveness, MDA-MB-231 was assays described as follows. stainedwithphalloidinafter6hofRu(GA)treatment.Results Evaluationoftheexpressionlevelofapoptosis-relatedgenes show that 12.5 μMRu(GA) strongly altersthe MDA-MB-231 was performed trough RT-qPCR. Four proapoptotic genes 1173 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Figure4.EffectsofRu(GA)onMDA-MB-231cellmigrationandinvasion.(A)Woundhealingassayat6,12,and24hoftreatmentwithindicated Ru(GA)concentrationsunderamplificationof40×.Graphrepresentsthepercentageofwoundclosure,comparingtreatedcellswithcontrolgroup. Datarepresentsmean±SDofthreeindependentassaysintriplicate.Significancein24hatthe**p<0.0001levelsusingANOVAandDunnet’s test.(B)EffectsofRu(GA)oncellinvasionthroughMatrigel.CellswereseededininsertscontainingMatrigelwithindicatedconcentrationsof complexfor22hwithFBSaschemoattractant.Invasivecellswerefixedandcountedmanually.Insertswerephotographedunder40×amplification. Graph representsinhibition of invasive cells.Positive control(C+) representsinvasive cellswithno treatmentand FBSin lowerchambers, and negativecontrol(C−)representsinvasivecellswithnotreatmentandnoFBSinlowerchambers.Representativeimagesoftheinsertsareexhibited. Data represents mean ± SD of three independent assays in triplicate. Significance at the *p < 0.05, **p < 0.001, ***p < 0.0001 levels using ANOVAand Dunnet’stest. Figure5.EffectsofRu(GA)onMDA-MB-231andMCF-10Acytoskeleton.Cellsweretreatedwithcomplexatincreasingconcentrationsfor6h. CellswerefixedandstainedwithAlexaFluor488phalloidinandDAPI.Imageswereobtainedwithamplificationof400×.Whitearrowsindicate lossofcellular prolongations,actin filament condensation,and reductionof cytoskeletondensity when comparedtountreatedcells. (Bax, caspase-8, caspase-9, and caspase-3) and one anti- to untreated MDA-MB-231 cells. Moreover, at6.25, 12.5, and apoptoticgene(Bcl-2)wereanalyzed.Ru(GA)at12.5and25 25 μM, Ru(GA) also upregulated caspase-3, a final effector of μM significantly upregulated Bax expression, when compared the apoptosis signaling pathway. However, the expression of 1174 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Figure6.EffectsofRu(GA)onapoptosisofMDA-MB-231andMCF-10A.Cellswereexposedtoindicatedconcentrationsofthecompoundfor24 h.Camptothecin(32μM)wasusedasapositivecontrolofapoptosis.Aftertreatment,cellswereincubatedwithPE−annexin-Vand7AADfor15 mininthedarkatroomtemperature,scraped,andthenanalyzedbyflowcytometry.(A)CytometryanalysisofMDA-MB-231andMCF-10A.The percentage of apoptotic cells for each cell line is represented in a graph. Data represents mean ± SD of three independent assays in triplicate. Significanceatthe**p<0.0001levelusingANOVAandDunnet’stest.(B)MDA-MB-231nucleardamagecausedbyRu(GA).Cellsweretreated withcomplexatincreasingindicatedconcentrationsfor8h.CellswerefixedandstainedwithDAPI.Imageswereobtainedwithamplificationof 400×.Whitearrowsindicateincreasingobservationofchromatincondensationandnuclearfragmentationappearancewhencomparedtountreated cells. Bcl-2andcaspase-8,aproteaserelatedtotheextrinsicsignaling Bcl-2 ratio is still favorable to the maintenance of pathwayforapoptosis,didnotchangeafterRu(GA)treatment. mitochondrial integrity (Figure 7A). On the other hand, caspase-9 levels were significantly Taking into consideration that the initiation of apoptosis in upregulated, suggesting an induction of apoptosis via an MDA-MB-231 cells occurs through intrinsic activation, the expression of proteins related to this signaling in tumor cells intrinsic pathway (Figure 7A). For MCF-10A cells, no significant alterations for Bax and caspase-3 expression but was evaluated by Western blotting (Figure 7B). The results an upregulation of Bcl-2 at 6.25, 12.5, and 25 μM was found, corroborate to the those found in RT-qPCR, suggesting a signaling to apoptosis due to mitochondrial dysfunction, since compared to untreated MCF-10A cells. Both caspase-8 and -9 there was an increase in the Bax/Bcl-2 ratio as well as the genes were upregulated in MCF-10A cells, with higher levels active caspase-9 and caspase-3 levels. A cell cycle distribution for caspase-9 (Figure 7A). These results confirm the higher assayshowedanincreaseofthesub-G1cellpopulationat0.78 sensitivityoftumorcellstoRu(GA)foundinthePE−annexin- and 3.12 μM after 24 h of incubation (Figure 7C and V assay, since at the same time of treatment, nontumor cells cytometry analysis on Figure S12), which is also indicative of havenotyetinitiatedtheexpressionofcaspase-3,andtheBax/ apoptosis.44 1175 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Figure7.InductionofapoptosisbyRu(GA).(A)EffectsofRu(GA)treatmentintheexpressionofgenesrelatedtoapoptosis.MDA-MB-231and MCF-10Acellswereincubatedwithindicatedconcentrationsofcomplexfor20h.Aftertreatment,RNAwasextractedwithTRIzolreagent.cDNAs were synthesized, and amplification of endogen control and target genes was analyzed as previously described. (B) Representative images of apoptosis-relatedproteinexpressionbyWesternblottinganalysis.(C)CellcycledistributionofMDA-MB-231cellsaftertreatmentwithRu(GA). Cells were exposed to indicated concentrations of complex for 24 h. Camptothecin (32 μM) was used as a positive control of apoptosis. After treatment, cells were permeabilized, incubated with PI for 1 h in ice, and analyzed by flow cytometry. The percentage of cells in each phase is representedinagraph.Analysisofthesub-G1phaseishighlightedinorange.Datarepresentsmean±SDofthreeindependentassaysintriplicate. Significanceatthe *p <0.05and **p <0.0001levels usingANOVA andDunnet’s test. Figure8.(A)ElectronicabsorptionspectraofRu(GA)(20μM)andtheeffectcausedbyincreasedct-DNAconcentration.(B)Effectofdifferent Ru(GA)concentrationsonthe fluorescence spectraofthe DNA-TO adduct (excitationat520 nm,298K). Interaction with DNA. To explore whether Ru(GA) absorptionspectralbandat260nm,observedforthecomplex, would induce apoptosis by targeting DNA in TNBC cells, wasusedtoevaluatethepossibleinteractionwithct-DNA.The spectroscopic titrations and fluorescence suppression of experimentswerecarriedoutkeepingtheconcentrationofthe thiazole orange (TO) studies were performed. The intense Ru(GA) constant and varying the concentration of the ct- 1176 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Table 3. Stern−Volmer Quenching Constanta, Biomolecular Quenching Rate Constantb, Binding Constantc, the Number of Binding Sitesd, ΔG°e, ΔH°f, and ΔS°g Values for the Complex−HSA/Apo-Tf System at Different Temperatures protein T(K) K (×104) K (×1013) K n ΔG° ΔH° ΔS° SV q b HSA 298 7.65±0.06 6.33 4.31×105(±1.10) 1.18 −3.21 25.1 192 310 7.62±0.12 6.24 6.38×105(±1.06) 1.22 −3.44 apo-Tf 298 7.25±0.02 4.83 1.05×107(±1.16) 1.65 −4.00 36.4 257 310 7.21±0.04 4.60 1.85×107(±1.15) 1.57 −4.31 aK ,M−1.bK, M−1s−1.cK,M−1.dn.eKJ·mol−1.fKJ·mol−1.gJ·mol−1K. SV q b Figure9.Fluorescencequenchingspectraof(A)HSAand(B)apo-TfwithdifferentRu(GA)concentrationsat310KinaTrizmabuffer,pH7.4. Insetgraphics representStern−Volmerplots showingtryptophan quenchinginHSA and apo-Tfat298and 310K. DNA.Whenaddingct-DNAsolutiontothecomplex,astrong and ΔS° values reveal the predominance of hydrophobic decrease in absorption intensity was observed, indicating interactions of the complex with HSA. Furthermore, the changes in the electronic spectra (Figure 8A). The intrinsic negative ΔG° values reveal that the interaction process is bindingconstant K found was9.13 ± 1.10 × 104M −1, which spontaneous; it occurs with no external intervention, and no b indicates a moderate interaction when compared to K of energy must be applied for the reaction to occur.40,46 b ethidium bromide (order of 106 to 107 M −1), known to be a The results found for apo-Tf also show a decrease in DNAintercalative.45However,interactiontestsofthecomplex tryptophan fluorescence in the presence of the complex withtheorangethiazoledye(TO),aDNAintercalatingagent, (Figure9B).Thetemperatureincreasealsodidnotchangethe showed a decrease in TO fluorescence after adding increasing scale of fluorescence suppression; an unaltered K and high SV concentrations of Ru(GA) (Figure 8B). valueforK indicatesastaticinteraction,asobservedforHSA. q Interaction with Human Albumin and Apo-Trans- Thennumberevidencedthatthereisonlyonebindingsitefor ferrin. To investigate whether Ru(GA) could interact with the complex in the apo-Tf molecule. The K value found b albumin and transferrin, fluorescence quenching experiments suggests a strong interaction between apo-Tf and Ru(GA). werecarried out. Fluorescencequenching datawere measured Parameters ΔH° and ΔS° point to hydrophobic interaction atdifferent temperatures(298 and310 K),and theresultsare between molecules, and negative ΔG° values indicate a shown in Table 3. Regarding HSA, results show a decrease in spontaneous process. tryptophan fluorescence in the presence of Ru(GA) (Figure Transport of Ru(GA) by Albumin and Apo-Trans- 9A) with no influence of the temperature. The quenching ferrin.Toevaluatetheinterferenceofalbuminandtransferrin constant K was not altered with increasing temperature, in the transport and interaction of Ru(GA) with cells, MDA- SV indicating that the interaction is through the formation of a MB-231 and MCF-10A were exposed to a treatment with the compound−proteincomplexandstaticinteraction,insteadofa complex in a constant concentration (near IC , 0.5 μM for 50 dynamiccollisioninteraction.Moreover,thevaluesofK were MDA-MB-231 and 3.12 μM for MCF-10A), along with q 6.33 and 6.24 × 1013 M −1 s −1 for 298 and 310 K respectively, increasing concentrations of BSA and human apo-transferrin, far higher than 2.0 × 1010 M −1 s −1, the maximum possible separately, for 48 h. The percentage of viable cells in each valuefordynamicinteraction,reinforcingthepresenceofstatic treatment was evaluated by an MTT assay. The reference of interactionbetweenthecomplexandprotein.40Thenumberof 100% of viability was the negative control cells (without binding sites is approximately equal to 1, indicating that there treatment).Inthisexperiment,itwasexpectedthatadecrease isonlyonebindingsiteforthecomplexintheHSAmolecule. in cell viability with increasing protein concentrations would Thermodynamic parameters, enthalpy change (ΔH°), occur, comparing all treatment groups to the positive control entropy change (ΔS°), and free energy change (ΔG°), are (cells receiving only complex and no protein), which would themainmeansusedtoindicatetypesofinteractionoperating indicate that the protein can enhance the interaction between between the complex and proteins. From the thermodynamic cells and complex. standpoint, ΔH > 0 and ΔS > 0 imply a hydrophobic Results show that even though the complex exhibited a interaction,ΔH<0andΔS<0reflectthevanderWaalsforce strong interaction with albumin, as demonstrated in the orhydrogenbondformation,andΔH<0andΔS>0suggest previous experiments, this protein may be responsible for the anelectrostaticforce.AsobservedinTable3,thepositiveΔH° transport of Ru(GA); however, it does not interfere in the 1177 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Figure 10. Transport of Ru(GA) by albumin and transferrin. MDA-MB-231 and MCF-10A cells were incubated with 0.5 and 3.12 μM concentrations of compound, respectively, and with increasing indicated concentrations of bovine serum albumin (A) and apo-Tf (B) for 48 h. ViablecellswereestimatedbyMTTassay,andresultswerecomparedtocellstreatedwithcompoundbutabsenceofprotein.Datarepresentsmean ±SDfrom three independentassays intriplicate.Significanceatthe *p< 0.001 and**p <0.0001 levelsusingANOVA and Dunnet’stest. Figure11.TfRsilencingattenuatestheeffectsofRuGA.(A)ContentofTfRonMCF-10AandMDA-MB-231cells(leftpanel)anddifferencesin theexpressionofTfRbeforeandafter72hofsilencingwithsiRNATFRCsequencess725ands727.(B)MDA-MDA-231cellstransfectedwith siRNATFRCsequencess725ands727wereincubatedwith0.5μMcompoundandwithincreasingindicatedconcentrationsofapo-Tffor48h. ViablecellswereestimatedbyMTTassay,andresultswerecomparedtocellstreatedwithcompoundbutinabsenceofprotein.(C)IC50valuesof nontransfectedandtransfectedcellswithTFRCsequencess725ands727.Datarepresentmean±SDfromthreeindependentassaysintriplicate. Significanceatthe *p <0.001 and **p < 0.0001levelsusing ANOVAand Dunnet’s test. 1178 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Ru(GA) activity on MDA-MB-231 cells, since increasing DoHuRu/DOTAP ([trans-RuCl (DMSO) ]−Na+ complexed 4 2 concentrations of BSA with a constant concentration of with nucleolipids) had an IC of 12.1 ± 0.3 μM for MDA- 50 Ru(GA) had no significant effects on MDA-MB-231 viability MB-231and10.3±0.2μMforMCF-7andalsowascapableof in this experiment, when compared to positive control. Even inducing morphological alterations on these cell lines, upon increasing the BSA concentration to 300 μg/mL (3-fold including cell shrinkage and loss of cell−cell contact under a apo-transferrin concentration), the result did not change 48 h treatment.51 (Figure 10A). On the other hand, when apo-transferrin was The Ru(GA) complex was investigated for its ability to increased, while maintaining a constant Ru(GA) concen- interfere with other cellular processes, such as migration and tration, cell viability significantly decreased compared to cells invasion.Tumorcellmigrationandinvasionofadjacenttissues treatedonlywithRu(GA),indicatingthat,possibly,transferrin is a process that leads to the extravasation of tumor cells to may be responsible for, or can enhance, the interaction and distant organs to form secondary tumors or metastasis.52 If a transportofRu(GA)intotumorcells(Figure10B).Regarding compoundiscapableofinhibitingcellmigrationandinvasion, MCF-10A, cell viability did not significantly change. These it can interrupt important processes of metastasis. Previous results suggest that Ru complexes might be more selective for studies demonstrated that picolinate arene Ru(II) complex tumor cells because of the higher amounts of TfR on their inhibitedcellmigrationandinvasionofHeLacells53andthata membranes. ruthenium polypyridyl complex effectively inhibited the TfR Silencing Attenuates Ru(GA) Effects on TNBC migration and invasion of MDA-MB-231 cells in a cytotox- Cells.InordertotestthehypothesisofhigheramountsofTfR icity-independent manner.17 In addition, a previous study on tumor cells, we investigate their levels in both nontumor demonstrated the capacity of an NAMI-A Ru complex in and TNBC cells and showed an overexpression of these altering the cytoskeleton structure in HeLa cells.54 receptors on tumor cells by cytometry analysis (Figure 11A, Cytoplasmic alterations, such as cytoskeleton damages, as leftpanel).TofurtherdemonstratetheinfluenceofTfRonthe those induced by the Ru(GA) complex, can be a source of Ru(GA) transport into TNBC cells, MDA-MB-231 was some signals leading to apoptosis.43 Prior studies described silenced for TfR before Ru(GA) treatment. Results with that cisplatin can directly interact with F-actin, preventing its TFRC s725 and s727 transfected cells demonstrated that polymerization and depolymerization, altering cytoskeleton silenced cells have less amounts of TfR (Figure 11A, right functions.43 Members of the Bcl-2 family contribute to both panel). An MTT assay after 48 h of treatment with Ru(GA) cytoskeletal organization and dysfunction processes. At the showed a significantly increased IC value for s725 TfR- sign of cytoskeleton damage and disorganization, these 50 silenced MDA-MB-231 (1.21 ± 0.16 μM) when compared to proteins are responsible for initiating proapoptotic signaling, s727 (1.14 ± 0.19 μM) and nonsilenced cells (0.81 ± 0.08 such as the activation of caspases, culminating in cell death. μM)(Figure11C),indicatingthatTfRsilencinginMDA-MB- Thecytoskeletonisthetargetofactivecaspases,whichcleaves 231 cells decreases Ru(GA) cytotoxic effects. After 48 h of proteins from this cellular structure, facilitating the formation Ru(GA) (0.5 μM) treatment with increasing apo-transferrin ofapoptoticbodies.43Rucomplexesthatcausecellcyclearrest concentrations, there was an improvement of the cell viability andtriggerapoptosisintumorcellscanbeapromisingstrategy rate, especially at higher concentrations, when compared to for cancer treatment. Several studies have demonstrated that nontransfected cells. While the cell viability of untransfected rutheniumcomplexesareabletoinduceapoptosis.TheRu(II) MDA-MB-231 reaches approximately 27.5% (Figure 10A, polypyridyl complexes increased apoptotic features such as bottomgraph),theviabilityofs725TfR-silencedMDAcellsis nuclear shrinkage, chromatin condensation, and cytoplasmic about 53% (Figure 11B, upper panel), indicating that these blebbing in BEL-7402 cells. Additionally, the complex receptors are at least partially responsible for the Ru(GA) promoted cell cycle arrest at the G0/G1 phase, raising the t■ransport to tumor cells. apoptotic cells rate in a concentration-dependent manner.22,55 The methylimidazole complex [Ru(MeIm)4(p-cpip)]2+ had DISCUSSION the same effects on the lung carcinoma cell line A549.18 Liu Chemotherapy treatments currently used are not able to and colleagues demonstrated that the apoptotic effects of Ru eradicate all tumor cells, which may lead to recurrences and complexes on osteosarcoma cell line MG-63, in which the drug resistance.47 Therefore, the search for more selective numberofcellswithDNAdamageincreased,culminateincell cytotoxic compounds is fundamental. Here we demonstrated cycle arrest at G0/G1 and apoptosis.13 The same effects were thatRu(GA)presentstime-dependentselectivetoxicityagainst described on HeLa56 and MDA-MB-231 cells.17 In addition, TNBC cells over nontumor breast cells. Similar results were nanoaggregates DoHuRu/POPC and DoHuRu/DOTAP described for the arene ruthenium(II) complex [(η6-C H )- formulations were able to induce proapoptotic effects on 6 6 Ru(H iip)Cl]Cl, named RAWQ11, that also inhibits most MCF-7 and MDA-MB-231 cells.51 2 efficiently the proliferation of MDA-MB-231 cells (IC 20.80 We have demonstrated here that the complex Ru(GA) 50 ± 0.40 μM) when compared to MCF-10A (IC > 300 μM) increases the number of cells at sub-G1 phase, which is a 50 and human normal kidney HK 2 (IC 110.30 ± 0.61 μM).48 parameter that indicates apoptosis induction in TNBC cells. 50 In addition, a Ru(II)-based complex with thiourea ligands, Thiscouldberesult,atleastinpart,oftheinteractionbetween [Ru(PPh ) (BzBn Th)(bipy)]PF ,presentedanIC of0.22± Ru(GA) and nuclear DNA. After oxidative stress, cancer cells 3 2 2 6 50 0.15 μM against the human prostate cancer cell line DU-145 often overexpress TfR on the nuclear surface,57,58 thus andanIC of0.68±0.47μMagainstthehumanlungcancer potentially enabling enhanced internalization of the Ru(GA) 50 cell line A549.49 The other Ru(II) complex with chloroquine complex into the nucleus of the cell, where it can exert its andchelatingligands,[RuCl(CTZ)(dppe)(bipy)]PF ,showed effects. However, upon comparison of Ru(GA) structural 6 IC values of 2.30 ± 1.64 and 0.80 ± 0.10 μM for human features with those of other complexes from the literature, it 50 breast cancer cell lines MDA-MB-231 and MCF-7, respec- can be determined that the most likely interaction for this tively.50 Finally, a Ru(III) nucleolipidic complex denominated complexwithDNAwouldbeelectrostatic.K valuesfoundfor b 1179 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article Ru(GA) interaction with DNA are comparable with those indicates a hydrophobic behavior and thus possible facility to foundfor[Ru(tpy)(PHBI)]2+and[Ru(tpy)(PHNI)]2+(PHBI cross cell membranes, might have contributed to its biological = 2-(2-benzimidazole)-1,10-phenanthroline; PHNI = 2-(2- properties. naphthoimidazole)-1,10-phenanthroline; tpy = 2,2′:6′,2″-ter- Despite these findings, some questions remain to be pyridyl)59 and other Ru(II) complexes that bind to DNA answered,suchasbioavailabilityoftheRu(GA),itsinteraction mainly through electrostatic interaction.60,61 Further, the with the reticuloendothelial or immunity systems, and its Ru(GA) K value is higher than that found for the complexes ability to reach the target in vivo in a proper amount to exert b with the lapachol ligand, cis-[Ru(PPh ) (lap) ] and [Ru(lap)- anticancer effects. Ru(GA) might be sequestered by 3 2 2 (PPh ) (phen)]PF ,forwhichtheK valueswerefound8.0× reticuloendothelial system, and only a small part might reach 3 2 6 b 102 and 3.0 × 103 M −1, respectively.25 Regarding the TO cancer cells, thus interfering with its properties found in interactionassay,eventhoughitindicatedthattheintercalative vitro.67,68 Possible toxicities of Ru(GA) on off-target organs dye was being replaced by the complex, we propose that the such as the liver or spleen can also be a matter of concern, electrostatic interaction between DNA and the complex is sinceTfR-1isexpressedinseveralhealthytissues,andTfR-2is capableofmomentarilycausingadisruptionintheDNA−TO- highly expressed by hepatocytes.69 Further in vivo studies binding system via a subtle change in the DNA electronic evaluating bioavailability andoxidative stress induction aswell density, thus inducing a decrease in the fluorescence process. asantitumor,antimetastatic,andtoxicitypropertiesofRu(GA) Nevertheless, electrostatic interactions are considered to be in complex biological systems will help to elucidate these weaker, when compared to intercalations, for example. Even questions. though it is shown that Ru(GA) causes nuclei fragmentation Taken together, our results show that Ru(GA) presented a by the DAPI assay, the DNA is not the primordial target for selectivecytotoxicagainstcancercells,comparedtonontumor Ru(GA), since electrostatic interactions are not capable of cells. Moreover, it was able to interfere with some cellular inducing strong damage to the DNA molecule and, hence, processes in TNBC cells, such as migration and invasion, triggering apoptosis. Thus, it is suggested that the leading leading to nuclei damage, cell cycle arrest, and apoptosis. The events to apoptotic bodies are a combination of these effects were more pronounced in tumor cells because of the interactions with mitochondrial dysfunction and cytoskeleton higher amounts of TfR compared to nontumor cells. The damage, as previously described. silencing of these receptors decreased Ru(GA) effects on Ontheotherhand,thepresenceofGAinthecomplexcould TNBC cells, demonstrating that the interaction mechanism havecontributedtothisproapoptoticeffect.TriphenolGAhas between Ru complexes and TfR should be deeply further been related to various biological properties, including investigated and can serve as a strategy for drug delivery antitumor capacity. Studies have demonstrated that the s■pecifically into tumor cells. antitumor effects of GA are related to its ability of ROS generation and intracellular Ca2+ increase, which triggers ASSOCIATED CONTENT * cytochromecreleaseintothecytoplasmandcaspaseactivation S Supporting Information through a mitochondrial membrane dysfunction, leading to The Supporting Information is available free of charge on the apoptosis.62 GA also seems to induce apoptosis by activating ACS Publications website at DOI: 10.1021/acs.molpharma- Fas receptors and promoting an overexpression of p53 and ceut.8b01154. p21,resultingincellcyclechangessuchasincreasingtheG1/S Table S1. Data from the calculation of the molar phaseandcelldeath.62Youandcolleagues23demonstratedthe absorptivity coefficient (E) of Ru(GA); Figure S1. ROSlevelsandapoptoticratesonHeLacellsincreasedwitha Experimental spectrum of MALDI-TOF mass and 50μMGAtreatment.GAinducedcytotoxicityinbreastcancer theoretical spectra of Ru(GA); Figure S2. Absorption MCF-7cellswithanIC 50 of80.5μMin24h,withanincreased spectrum in the UV−vis region of Ru(GA); Figure S3. apoptoticrate.63 On theother hand, Yen andco-workers,64 in InfraredspectrumofthecomplexofRu(GA);FigureS4. a study with human lymphocytes, demonstrated that, depend- 1H NMR spectrum of Ru(GA); Figure S5. 1H COSY ing on concentration, GA has the ability of decreasing spectrum of Ru(GA); Figure S6. 13C {1H} NMR deoxyribose,DPPHradical(byDPPHmethod),andhydrogen spectrum of Ru(GA); Figure S7. Ru(GA) stability in peroxides, indicating also an antioxidant activity. Moreover, DMSO/DMEM by 31P {1H} NMR spectra; Figure S8. GA is capable of inhibiting cyclooxygenase-2 (COX-2). This Ru(GA) stability in DMSO by UV−vis spectrum at enzyme is overexpressed in tumor cells and represents a key different times; Figure S9. Cyclic and differential pulse enzyme in the synthesis of prostaglandins, which are voltammograms of Ru(GA); Figure S10. Morphological responsible for local inflammatory reactions, such as capillary observations caused by Ru(GA) cytotoxicity on MDA- permeability and macrophage attraction, that will provide MB-231 and MCF-10A cells; Figure S11. Full-length survival signals for tumor cells.65 blots of proteins related to apoptosis; Figure S12. Flow We verified in this work a selective cytotoxic effect of the cytometryhistogramofMDA-MB-231cellstreatedwith Ru(GA) complex on TNBC cells over nontumor cells. This Ru(GA) (PDF) wasexplainedbythehigheramountsofTfRontumorcells,as ■ transferrin was demonstrated to serve as a mediator to deliver Ru complexes into the cells.66 Our results are in agreement AUTHOR INFORMATION with those of Zhao and colleagues,16 who demonstrated that Corresponding Author when TfR was blocked by specific antibodies in breast cancer *Mailing Address: Marcia R. Cominetti, Department of MCF-7 cells, the cellular uptake of Ru(II) polypyridyl Gerontology, Federal University of São Carlos, Rod. complexes decreased by 53.6%. These findings suggest that Washington Luis ́ , km 235, CEP 13565-905 Saõ Carlos, SP, TfRisresponsibleforpartoftheRucomplexes’transportinto Brazil; E-mail: mcominetti@ufscar.br; Phone: +55 (16) 3306- cancer cells. In addition, the Ru(GA) log P value, which 6663; Fax: +55 (16) 3351-9628. 1180 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article ORCID (6) Jakupec, M. A.; Galanski, M.; Arion, V. B.; Hartinger, C. G.; Legna C. Vegas: 0000-0003-3557-5544 Keppler, B. K. Antitumour metal compounds: more thantheme and ̃ variations. Dalton transactions (Cambridge, England: 2003) 2008, Joao Honorato: 0000-0002-1127-6083 No. 2, 183−94. Marcia R. Cominetti: 0000-0001-6385-7392 (7) Nesǐc,́ M.; Popovic,́ I.; Leskovac, A.; Petkovic,́ M. Biological Author Contributions activityandbindingpropertiesof[Ru(II)(dcbpy)2Cl2]complexto The manuscript was written through contributions of all bovine serum albumin, phospholipase A2 and glutathione. BioMetals authors. All authors discussed the results, commented on the 2016,29(5), 921−933. manuscript, and gave approval to the final version of the (8) Popolin, C. P.; Cominetti, M. R. A review of ruthenium complexes activities on different steps of the metastatic process in manuscript. M.A.N., A.E.G., L.C.V., L.L.D., J.H., and M.R.C. breast cancercells. Mini-Rev.Med. Chem. 2017, 17,1435. conceived and designed the experiments and analyzed data. (9) Habeshaw, J. A.; Lister, T. A.; Stansfeld, A. G.; Greaves, M. F. M.A.N. and A.E.G. prepared the manuscript and assembled Correlation of transferrin receptor expression with histological class figures. A.E.G., J.H., A.C.S.M., and A.A.B., prepared and and outcome in non-Hodgkin lymphoma. Lancet 1983, 321, 498− characterized the Ru(GA) ruthenium complex and conducted 501. all chemical experiments. M.A.N and L.C.V. performed the (10) Daniels, T. R.; Bernabeu, E.; Rodriguez, J. A.; Patel, S.; interaction studies with DNA, HSA, and apo-transferrin. Kozman,M.;Chiappetta,D.A.;Holler,E.;Ljubimova,J.Y.;Helguera, M.A.N. and L.L.D. conducted in vitro experiments of MTT, G.;Penichet,M.L.Thetransferrinreceptorandthetargeteddelivery morphology, colony, migration, and invasion. M.A.N. oftherapeuticagentsagainstcancer.Biochim.Biophys.Acta,Gen.Subj. performed in vitro assays of TfR staining, phalloidin staining, 2012,1820(3), 291−317. PE−annexin,DAPIstaining,PCR,Westernblotting,cellcycle, (11) Allardyce, C. S.; Dyson, P. J. Ruthenium in medicine: current and siRNA transfection. A.A.B and M.R.C. revised the clinical uses and future prospects. Platinum Metals Review 2001, 45 manuscript. (2), 62−69. (12) Bergamo, A.; Sava, G. Ruthenium anticancer compounds: Notes myths and realities of the emerging metal-based drugs. Dalton T■he authors declare no competing financial interest. transactions (Cambridge,England:2003) 2011,40(31),7817−23. (13) Liu, S.-H.; Zhao, J.-H.; Deng, K.-K.; Wu, Y.; Zhu, J.-W.; Liu, ACKNOWLEDGMENTS Q.-H.; Xu, H.-H.; Wu, H.-F.; Li, X.-Y.; Wang, J.-W.; et al. Effect of radiation on cytotoxicity, apoptosis and cell cycle arrest of human The Laboratory of Structure and Reactivity of Inorganic osteosarcoma MG-63 induced by a ruthenium (II) complex. Compounds (LERCI) from UFSCar is acknowledged for Spectrochim. Acta,Part A 2015,140,202−209. providing the ruthenium complex, and the Laboratory of (14)Tan,C.;Wu,S.;Lai,S.;Wang,M.;Chen,Y.;Zhou,L.;Zhu,Y.; Biology and Biochemistry (LBBM) from UFSCar is acknowl- Lian,W.;Peng,W.;Ji,L.;Xu,A.Synthesis,structures,cellularuptake edged for additional equipment and support. The National and apoptosis-inducing properties of highly cytotoxic ruthenium- Counsel of Technological and Scientific Development − Norharman complexes. Dalton transactions (Cambridge, England: CNPqandtheSaõ PauloResearchFoundation−FAPESPare 2003) 2011,40(34),8611−21. acknowledged for the financial support (grants #2013/00798- (15)Pierroz,V.;Joshi,T.;Leonidova,A.;Mari,C.;Schur,J.;Ott,I.; 2■; #2015/24940-8, and #2016/16312−0). Spiccia, L.; Ferrari, S.; Gasser, G. Molecular and cellular character- ization of the biological effects of ruthenium(II) complexes incorporating 2-pyridyl-2-pyrimidine-4-carboxylic acid. J. Am. Chem. ABBREVIATIONS Soc. 2012, 134(50),20376−87. apo-Tf, transferrin; ct-DNA, calf thymus DNA; COX2, (16) Zhao, Z.; Luo, Z.; Wu, Q.; Zheng, W.; Feng, Y.; Chen, T. cyclooxygenase-2; ESI-MS, electrospray ionization mass Mixed-ligandrutheniumpolypyridylcomplexesasapoptosisinducers spectrometry; ER, estrogen receptor; GA, gallic acid; HER-2, in cancer cells, the cellular translocation and the important role of human epidermal growth factor receptor 2 (HER 2); MLCT, ROS-mediated signaling. Dalton transactions (Cambridge, England: metal to ligand charge transfer; MTT, 3-(4,5-dimethylthiazol- 2003) 2014,43(45),17017−28. (17)Cao,W.;Zheng,W.;Chen,T.Rutheniumpolypyridylcomplex 2-Yl)-2,5-diphenyltetrazolium bromide; PR, progesterone inhibits growth and metastasis of breast cancer cells by suppressing receptor; Tf-R, transferrin receptors; TBAP, tetrabutylammo- FAK signaling with enhancement of TRAIL-induced apoptosis. Sci. nium perchlorate; TNBC, triple-negative breast cancer; TO, Rep. 2015, 5,9157. t■hiazole orange (18)Chen,T.;Liu,Y.;Zheng,W.J.;Liu,J.;Wong,Y.S.Ruthenium polypyridyl complexes that induce mitochondria-mediated apoptosis REFERENCES incancer cells.Inorg. Chem. 2010, 49(14),6366−8. (19)Carnizello,A.P.;Barbosa,M.I.;Martins,M.;Ferreira,N.H.; (1) Toss, A.; Cristofanilli, M. Molecular characterization and Oliveira,P.F.;Magalhaẽ s,G.M.;Batista,A.A.;Tavares,D.C.Invitro targeted therapeutic approaches in breast cancer. Breast Cancer Res. 2015, 17,60. and in vivo antitumor activity of a novel carbonyl ruthenium (2)Blok,E.J.;Derks,M.G.;vanderHoeven,J.J.;vandeVelde,C. compound, the ct-[RuCl (CO)(dppb)(bipy)] PF6 [dppb= 1, 4-bis J.; Kroep, J. R. Extended adjuvant endocrine therapy in hormone- (diphenylphosphine) butane and bipy= 2, 2′-bipyridine]. J. Inorg. receptor positive early breast cancer: current and future evidence. Biochem.2016,164,42−48. CancerTreat.Rev. 2015,41(3), 271−6. (20)Benadiba,M.;deM.Costa,I.;Santos,R.L.;Serachi,F.O.;de (3) Kelland, L. The resurgence of platinum-based cancer chemo- Oliveira Silva, D.; Colquhoun, A. Growth inhibitory effects of the therapy.Nat. Rev.Cancer2007,7 (8),573−84. Diruthenium-Ibuprofen compound,[Ru2Cl (Ibp) 4], in human (4)Guo,W.;Zheng,W.;Luo,Q.;Li,X.C.;Zhao,Y.;Xiong,S.X.; glioma cells in vitro and in the rat C6 orthotopic glioma in vivo. Wang, F. Y. Transferrin Serves As a Mediator to Deliver Organo- JBIC, J. Biol.Inorg. Chem. 2014, 19(6),1025−1035. metallic Ruthenium(II) Anticancer Complexes into Cells. Inorg. (21)Clavel,C.M.;Paunescu,E.;Nowak-Sliwinska,P.;Griffioen,A. Chem. 2013,52(9), 5328−5338. W.; Scopelliti, R.; Dyson, P. J. Discovery of a highly tumor-selective (5) Clarke, M. J. Ruthenium metallopharmaceuticals. Coord. Chem. organometallicruthenium(II)-arenecomplex.J.Med.Chem.2014,57 Rev.2003, 236(1), 209−233. (8), 3546−58. 1181 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article (22) Xie, Y.-Y.; Li, Z.-Z.; Lin, G.-J.; Huang, H.-L.; Wang, X.-Z.; (39) Lakowicz, J. R.; Weber, G. Quenching of fluorescence by Liang, Z.-H.; Jiang, G.-B.; Liu, Y.-J. DNA interaction, cytotoxicity, oxygen. A probe for structural fluctuations in macromolecules. apoptotic activity, cell cycle arrest, reactive oxygen species and Biochemistry 1973, 12(21),4161−70. mitochondrial membrane potential assay induced by ruthenium (II) (40)Ganeshpandian,M.;Loganathan,R.;Suresh,E.;Riyasdeen,A.; polypyridylcomplexes.Inorg. Chim.Acta 2013,405,228−234. Akbarsha, M. A.; Palaniandavar, M. New ruthenium(II) arene (23) You, B. R.; Moon, H. J.; Han, Y. H.; Park, W. H. Gallic acid complexesofanthracenyl-appendeddiazacycloalkanes:effectofligand inhibitsthegrowthofHeLacervicalcancercellsviaapoptosisand/or intercalation and hydrophobicity on DNA and protein binding and necrosis.Food Chem.Toxicol. 2010,48(5), 1334−1340. cleavage and cytotoxicity. Dalton transactions (Cambridge, England: (24) Lu, Y.; Jiang, F.; Jiang, H.; Wu, K.; Zheng, X.; Cai, Y.; 2003) 2014,43(3), 1203−19. Katakowski, M.; Chopp, M.; To, S.-S. T. Gallic acid suppresses cell (41) Nakamoto, K. Infrared and Raman spectra of inorganic and viability, proliferation, invasion and angiogenesis in human glioma coordination compounds,4th ed.;Wiley, 1986. cells.Eur. J. Pharmacol.2010,641(2), 102−107. (42) Enache, T. A.; Oliveira-Brett, A. M. Phenol and para- (25)Barbosa,M.I.;Correâ ,R.S.;deOliveira,K.M.;Rodrigues,C.; substitutedphenolselectrochemicaloxidationpathways.J.Electroanal. Ellena,J.;Nascimento,O.R.;Rocha,V.P.;Nonato,F.R.;Macedo,T. Chem. 2011,655 (1),9−16. S.; Barbosa-Filho, J. M.; et al. Antiparasitic activities of novel (43) Sancho-Martínez, S. M.; Prieto-García, L.; Prieto, M.; Loṕ ez- ruthenium/lapachol complexes.J. Inorg.Biochem.2014,136,33−39. Novoa, J. M.; Loṕ ez-Hernań dez, F. J. Subcellular targets of cisplatin (26) Oliveira, K. M.; Correâ , R. S.; Barbosa, M. I.; Ellena, J.; cytotoxicity:anintegratedview.Pharmacol.Ther.2012,136(1),35− Cominetti, M. R.; Batista, A. A. Ruthenium (II)/triphenylphosphine 55. (44) Kajstura, M.; Halicka, H. D.; Pryjma, J.; Darzynkiewicz, Z. complexes: An effective way to improve the cytotoxicity of lapachol. Polyhedron2017,130,108−114. Discontinuous fragmentation of nuclear DNA during apoptosis revealed by discrete “sub G1” peaks on DNA content histograms. (27) Hunsucker, S. W.; Watson, R. C.; Tissue, B. M. Character- Cytometry,Part A 2007,71A(3), 125−131. ization of inorganic coordination complexes by matrix-assisted laser (45)Heng,M.P.;Sinniah,S.K.;Teoh,W.Y.;Sim,K.S.;Ng,S.W.; desorption/ionization mass spectrometry. Rapid Commun. Mass Spectrom.2001, 15(15),1334−40. Cheah, Y. K.; Tan, K. W. Synthesis of a DNA-targeting nickel (II) complexwithtestosteronethiosemicarbazonewhichexhibitsselective (28)Bautista,M.T.;Cappellani,E.P.;Drouin,S.D.;Morris,R.H.; cytotoxicity towards human prostate cancer cells (LNCaP). Schweitzer, C. T.; Sella, A.; Zubkowski, J. PREPARATION AND Spectrochim. Acta,Part A 2015,150,360−72. SPECTROSCOPIC PROPERTIES OF THE ETA-2-DIHYDRO- (46) Ross, P. D.; Subramanian, S. Thermodynamics of protein GENCOMPLEXES[MH(ETA-2-H2)(PR2CH2CH2PR2)2]+(M= association reactions: forces contributing to stability. Biochemistry FE, RU, R= PH, ET) AND TRENDS IN PROPERTIES DOWN 1981,20(11),3096−3102. THE IRON GROUP TRIAD. J. Am. Chem. Soc. 1991, 113 (13), (47) Schwartsmann, G.; Ratain, M. J.; Cragg, G. M.; Wong, J. E.; 4876−4887. Saijo,N.;Parkinson,D.R.;Fujiwara,Y.;Pazdur,R.;Newman,D.J.; (29)Mosmann,T.Rapidcolorimetricassayforcellulargrowthand Dagher,R.;DiLeone,L.Anticancerdrugdiscoveryanddevelopment survival: application to proliferation and cytotoxicity assays. J. throughout theworld. J. Clin. Oncol. 2002,20(18 Suppl), 47s−59s. Immunol.Methods 1983,65(1−2),55−63. (48) Wu, Q.; He, J.; Mei, W.; Zhang, Z.; Wu, X.; Sun, F. Arene (30) Badisa, R. B.; Darling-Reed, S. F.; Joseph, P.; Cooperwood, J. ruthenium (II) complex, a potent inhibitor against proliferation, S.; Latinwo, L. M.; Goodman, C. B. Selective cytotoxic activities of migration and invasion of breast cancer cells, reduces stress fibers, two novel synthetic drugs on human breast carcinoma MCF-7 cells. focal adhesions and invadopodia. Metallomics: integrated biometal AnticancerRes. 2009,29(8), 2993−2996. science 2014,6(12),2204−2212. (31)Baka,E.;Comer,J.E.;Takacs-Novak,K.Studyofequilibrium (49) Correa, R. S.; de Oliveira, K. M.; Delolo, F. G.; Alvarez, A.; solubility measurement by saturation shake-flask method using Mocelo,R.;Plutin,A.M.;Cominetti,M.R.;Castellano,E.E.;Batista, hydrochlorothiazide as model compound. J. Pharm. Biomed. Anal. A. A. Ru (II)-based complexes with N-(acyl)-N′, N′-(disubstituted) 2008, 46(2),335−41. thiourea ligands: Synthesis, characterization, BSA-and DNA-binding (32)Franken,N.A.;Rodermond,H.M.;Stap,J.;Haveman,J.;van studiesofnewcytotoxicagentsagainstlungandprostatetumourcells. Bree, C. Clonogenic assay of cells in vitro. Nat. Protoc 2006, 1 (5), J. Inorg. Biochem.2015,150,63−71. 2315−9. (50)Colina-Vegas,L.;Villarreal,W.;Navarro,M.;deOliveira,C.R.; (33)Yarrow,J.C.;Perlman,Z.E.;Westwood,N.J.;Mitchison,T.J. Graminha, A. E.; Maia, P. I. d. S.; Deflon, V. M.; Ferreira, A. G.; Ahigh-throughputcellmigrationassayusingscratchwoundhealing,a Cominetti, M. R.; Batista,A. A. Cytotoxicity of Ru (II) piano−stool comparisonofimage-basedreadoutmethods.BMCBiotechnol.2004, complexes with chloroquineand chelating ligands against breast and 4, 21. lung tumor cells: Interactions with DNAand BSA. J. Inorg. Biochem. (34) Yue, P. Y.; Leung, E. P.; Mak, N.; Wong, R. N. A simplified 2015,153,150−161. method for quantifying cell migration/wound healing in 96-well (51) Irace, C.; Misso, G.; Capuozzo, A.; Piccolo, M.; Riccardi, C.; plates.J. Biomol. Screening2010, 15(4),427−433. Luchini,A.;Caraglia,M.;Paduano,L.;Montesarchio,D.;Santamaria, (35)Fuzer,A.M.;Filho,J.C.;Becceneri,A.B.;DosSantos,D.A.; R. Antiproliferative effects of ruthenium-based nucleolipidic nano- daSilva, M. F.; Vieira, P. C.; Fernandes, J. B.; Selistre-de-Araujo, H. aggregates in human models of breast cancer in vitro: insights into S.; Cazal, C. M.; Cominetti, M. R. Effects of limonoid cedrelone on theirmode ofaction. Sci. Rep.2017,7, 45236. MDA-MB-231 breast tumor cells in vitro. Anti-Cancer Agents Med. (52) Quail, D. F.; Joyce, J. A. Microenvironmental regulation of Chem. 2013,13(10),1645−1653. tumorprogressionandmetastasis.Nat.Med.2013,19(11),1423−37. (36) Livak, K. J.; Schmittgen, T. D. Analysis of relative gene (53) Gligorijevic, N.; Arandelovic, S.; Filipovic, L.; Jakovljevic, K.; expression data using real-time quantitative PCR and the 2(-Delta Jankovic, R.; Grguric-Sipka, S.; Ivanovic, I.; Radulovic, S.; Tesic, Z. Delta C(T)) Method. Methods (Amsterdam, Neth.) 2001, 25 (4), Picolinateruthenium(II)-arenecomplexwithinvitroantiproliferative 402−8. and antimetastatic properties: comparison to a series of ruthenium- (37)Fredericq,E.;Oth,A.;Fontaine,F.Theultravioletspectrumof (II)-arene complexes with similar structure. J. Inorg. Biochem. 2012, deoxyribonucleic acids and their constituents. J. Mol. Biol. 1961, 3, 108,53−61. 11−7. (54) Sava, G.; Frausin, F.; Cocchietto, M.; Vita, F.; Podda, E.; (38)Veal,J.M.;Rill,R.L.NoncovalentDNAbindingofbis(1,10- Spessotto, P.; Furlani, A.; Scarcia, V.; Zabucchi, G. Actin-dependent phenanthroline) copper (I) and related compounds. Biochemistry tumour cell adhesion after short-term exposure to the antimetastasis 1991, 30(4),1132−1140. rutheniumcomplexNAMI-A.Eur.J.Cancer2004,40(9),1383−96. 1182 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183 Molecular Pharmaceutics Article (55)Han,B.-J.;Jiang,G.-B.;Yao,J.-H.;Li,W.;Wang,J.;Huang,H.- L.; Liu, Y.-J. DNA interaction, antioxidant activity, and bioactivity studies of two ruthenium (II) complexes. Spectrochim. Acta, Part A 2015, 135, 840−849. (56)Tan,C.;Wu,S.;Lai,S.;Wang,M.;Chen,Y.;Zhou,L.;Zhu,Y.; Lian,W.;Peng,W.;Ji, L.;etal. Synthesis,structures, cellularuptake and apoptosis-inducing properties of highly cytotoxic ruthenium- Norharmancomplexes. DaltonTrans.2011,40(34),8611−8621. (57) Mazzucchelli, S.; Truffi, M.; Baccarini, F.; Beretta, M.; Sorrentino, L.; Bellini, M.; Rizzuto, M. A.; Ottria, R.; Ravelli, A.; Ciuffreda,P.;Prosperi,D.;Corsi,F.H-Ferritin-nanocagedolaparib:a promisingchoiceforbothBRCA-mutatedandsporadictriplenegative breastcancer. Sci. Rep.2017, 7, 7505. (58)Yang,D.C.;Wang,F.;Elliott,R.L.;Head,J.F.Expressionof transferrin receptor and ferritin H-chain mRNA are associated with clinical and histopathological prognostic indicators in breast cancer. AnticancerRes. 2001,21(1B), 541−549. (59) Jiang, C. W.; Chao, H.; Li, H.; Ji, L. N. Syntheses, characterization and DNA-binding studies of ruthenium(II) terpyr- idine complexes: [Ru(tpy)(PHBI)]2+ and [Ru(tpy)(PHNI)]2+. J. Inorg.Biochem.2003,93(3−4),247−55. (60)deCamargo,M.S.;daSilva,M.M.;Correa,R.S.;Vieira,S.D.; Castelli,S.;D’Anessa,I.;DeGrandis,R.;Varanda,E.;Deflon,V.M.; Desideri, A.; et al. Inhibition of human DNA topoisomerase IB by nonmutagenic ruthenium (II)-based compounds with antitumoral activity.Metallomics 2016,8 (2),179−192. (61) Correâ , R. S.; da Silva, M. M.; Graminha, A. E.; Meira, C. S.; dosSantos,J.A.F.;Moreira,D.R.;Soares,M.B.;VonPoelhsitz,G.; Castellano,E.E.;Bloch,C.;etal.Ruthenium(II)complexesof1,3- thiazolidine-2-thione: Cytotoxicity against tumor cells and anti- Trypanosoma cruzi activity enhanced upon combination with benznidazole.J. Inorg. Biochem.2016,156,153−163. (62) Locatelli, C.; Filippin-Monteiro, F. B.; Creczynski-Pasa, T. B. Alkyl esters of gallic acid as anticancer agents: a review. Eur. J. Med. Chem. 2013,60,233−9. (63)Wang,K.;Zhu,X.;Zhang,K.;Zhu,L.;Zhou,F.Investigationof Gallic Acid Induced Anticancer Effect in Human Breast Carcinoma MCF7 Cells. J. Biochem.Mol.Toxicol. 2014,28 (9),387−393. (64)Yen,G.-C.;Duh,P.-D.;Tsai,H.-L.Antioxidantandpro-oxidant propertiesofascorbicacidandgallicacid.FoodChem.2002,79(3), 307−313. (65)Verma,S.;Singh,A.;Mishra,A.Gallicacid:molecularrivalof cancer.Environ. Toxicol. Pharmacol.2013,35(3),473−85. (66)Guo,W.;Zheng,W.;Luo,Q.;Li,X.;Zhao,Y.;Xiong,S.;Wang, F. Transferrin serves as a mediator to deliver organometallic ruthenium(II) anticancer complexes into cells. Inorg. Chem. 2013, 52(9),5328−38. (67)Liu,J.;Zheng,X.P.;Gu,Z.J.;Chen,C.Y.;Zhao,Y.L.Bismuth sulfidenanorodsasaprecisionnanomedicineforinvivomultimodal imaging-guided photothermal therapy of tumor. Nanomedicine 2016, 12(2),486−487. (68)Lu,Z.X.;Huang,F.Y.;Cao,R.;Zhang,L.M.;Tan,G.H.;He, N.Y.;Huang,J.;Wang,G.Z.;Zhang,Z.J.LongBloodResidenceand Large Tumor Uptake of Ruthenium Sulfide Nanoclusters for Highly EfficientCancer PhotothermalTherapy.Sci. Rep.2017,7,41571. (69)Herbison,C.E.;Thorstensen,K.;Chua,A.C.;Graham,R.M.; Leedman, P.; Olynyk, J. K.; Trinder, D. The role of transferrin receptor1and2intransferrin-boundironuptakeinhumanhepatoma cells. American journal of physiology. Cell physiology 2009, 297 (6), C1567−75. (70)Buss,I.;Garmann,D.;Galanski,M.;Weber,G.;Kalayda,G.V.; Keppler, B. K.; Jaehde, U. Enhancing lipophilicity as a strategy to overcome resistance against platinum complexes? J. Inorg. Biochem. 2011, 105(5), 709−17. 1183 DOI:10.1021/acs.molpharmaceut.8b01154 Mol.Pharmaceutics2019,16,1167−1183