← Back

Simultaneously Inducing and Tracking Cancer Cell Metabolism Repression by Mitochondria-Immobilized Rhenium(I) Complex.

PMID: 28368110
ResearchArticle www.acsami.org Simultaneously Inducing and Tracking Cancer Cell Metabolism Repression by Mitochondria-Immobilized Rhenium(I) Complex Jing Yang, †,§ Ji-Xian Zhao, †,§ Qian Cao, *,† Liang Hao, † Danxia Zhou, ‡ Zhenji Gan, ‡ Liang-Nian Ji, † and Zong-Wan Mao *,† † MOEKeyLaboratoryofBioinorganicandSyntheticChemistry,SchoolofChemistry,SunYat-SenUniversity,Guangzhou510275, China ‡ MOEKeyLaboratoryofModelAnimalsforDiseaseStudy,ModelAnimalResearchCenterofNanjingUniversity,Nanjing210061, China * S Supporting Information ABSTRACT: Mitochondrialmetabolismisessentialfortumorigenesis,andthe developmentofcancerisusuallyaccompaniedbyalternationsofmitochondrial function. Emerging studies have demonstrated that targeting mitochondria and mitochondrial metabolism is an effective strategy for cancer therapy. In this work, eight phosphorescent organometallic rhenium(I) complexes have been synthesized and explored as mitochondria-targeted theranostic agents, capable ofinducingandtracking thetherapeuticeffectsimultaneously.Complexes 1b− 4b can quickly and efficiently penetrate into A549 cells, specifically localizing within mitochondria, and their cytotoxicity is superior to cisplatin against the cancer cells screened. Notably, complex 3b [Re(CO) (DIP) (py-3-CH Cl)]+ 3 2 containingthiol-reactivechloromethylpyridylmoietyformitochondriaimmobi- lization showshighercytotoxicity andselectivityagainst cancer cellsthanother Re(I) complexes without mitochondria-immobilization properties. Mechanistic studies show that complexes 1b−4b induce a cascade of mitochondria-dependent events including mitochondrial damage, mitochondrial respiration inhibition, cellular ATP depletion, reactive oxygen species (ROS) elevation, and caspase-dependent apoptosis. By comparison, mitochondria-immobilized 3b causes more effective repression of mitochondrial metabolism than mitochondrial-nonimmobilized complexes. The excellent phosphorescence and O -sensitive lifetimes of mitochondria- 2 immobilized 3b can be utilized for real-time tracking of the morphological changes of mitochondria and mitochondrial respiration repression during therapy process, accordingly providing reliable information for understanding anticancer mechanisms. KEYWORDS: organometallic rhenium complex, mitochondria immobilization, metabolism repression, mitochondria dysfunction, PLIM, theranostic 1. INTRODUCTION Emergingclinicalevidenceandlaboratory-basedexperiments Mitochondria are well-known as the “cellular energy factory”, have demonstrated that targeting mitochondria and mitochon- drial metabolism is an effective therapeutic strategy against producing ATP and metabolites necessary for the cellular cancer,7−9 and some targeted molecules have already attended bioenergetic and biosynthetic demands.1 Recent studies clinical trials.10−16 These drugs exert their anticancer activities providegeneticandpharmacologicevidencethatmitochondrial by disrupting the mitochondrial energy producing system, the metabolismisessentialforoncogenerevolution,tumorigenesis, biosynthesis function, and the redox homeostasis, ultimately and tumor growth.2,3 The occurrence and development of leadingtotheactivationofmitochondrial-dependentcelldeath cancer are usually accompanied by alternations of mitochon- signaling pathways, usually the apoptotic pathways because of drial function, e.g., increased oxidative stress and decreased thevariousapoptosis-promotingproteinsexistinginmitochon- mitochondrial membrane potential.4 Moreover, the mitochon- dria.10−17 Thus, even when the endogenous apoptosis drialmetabolicpathwaysincancercellsaredifferentfromthose induction pathways were disrupted, mitochondria-targeted in normal cells. Cancer cells are primarily characterized by a drugs can still trigger cellular apoptosis efficiently, which is metabolic phenotype of aerobic glycolysis for ATP generation more advanced than other conventional apoptosis-inducing while normal cells generally use oxidative phosphorylation as the major metabolic pathway.5,6 In this perspective, mitochon- Received: February6, 2017 drial metabolism has emerged asa potentially fruitful arena for Accepted: April3, 2017 cancer therapy.1 Published: April3, 2017 ©XXXXAmericanChemicalSociety A DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle drugs and may overcome cancer resistance.17,18 On the other designedandsynthesized.ThebidentateN−Nligandtunesthe hand, a series of fluorescent mitochondria-targeted molecules lipophilicity of the compounds, and the monodentate ligand L have been synthesized for probing specific chemicals in is expected to contribute to the mitochondria affinity. It is mitochondria and tracking the dynamic changes of mitochon- worth mentioning that the reactive chloromethyl moiety dria.19,20 Inspired by these investigations, rational design of present in compounds 3a and 3b has the reactivity with emissive mitochondria-targeted multifunctional theranostic biologically relevant thiols, accordingly covalently binding with agents for cancer treatment are highly sought, which can thiol-containing proteins, such as bovine serum albumin (BSA, monitor the changes in mitochondria during therapeutic containsonefreecysteineresidue)andglutathione(GSH).22,38 process, giving insights into their anticancer mechanisms.21−23 Therefore, these two rhenium compounds are expected to be With the success of platinum-based anticancer drugs in the not only accumulating but also immobilized within mitochon- past few decades has emerged the development of other dria by covalent interactions with thiol-containing proteins. If nonplatinum metallodrugs. Recently, organometallic rhenium so, their metabolism repression efficacy and mitochondrial compounds have become a kind of new promising anticancer tracking capability will be enhanced because of the prolonged drug candidate, showing comparable or superior anticancer retention time within mitochondria even when the membrane activity compared with cisplatin.24−28 Most reported cytotoxic potential is already lost. Mitochondria-immobilized cyclo- andluminescentReorganometalliccomplexesarebasedonthe metalated iridium(III) complexes have been recently re- Re(CO) core, which makes the synthesis procedure and the ported.22,39 In this work, the anticancer properties of both 3 introduction of functional ligands easily accessible and mitochondria-immobilized and -nonimmobilized Re(I) com- chemically robust. These Re complexes can quickly and plexes were compared, including the in vitro cytotoxicity, efficiently penetrate into cells, and their cytotoxicity has been cancer cell selectivity, metabolism repression, mitochondria found highly proportional to their lipophilicity.24−26 Very damage, cellular ATP depletion, reactive oxygen species recently several rhenium complexes have also been reported elevation, and induction of apoptosis. Their capability for possessing photodynamic and photoactive therapeutic poten- mitochondrialimmobilization,insitutrackingofmitochondrial tials.29−32 On the other hand, phosphorescent Re complexes morphology, and cancer cell metabolism repression was also havebeenwidelyexploredasbioimagingandbiosensingagents explored depending on their excellent phosphorescence and exhibiting good photostability, long-lived emission states, high O 2 -sensitive lifetimes byusing confocal andPLIM microscopy. quantumyields,andlargeStokesshifts,andtheemissioncanbe tuned by varying the ligands.33−37 Therefore, phosphorescent 2. RESULTS AND DISCUSSION Re(I) complexes possess intrinsic advantages for the con- 2.1. Synthesis and Photophysical Properties. In the struction of novel multifunctional theranostic platforms. By presentwork,wesynthesizedeightphosphorescentrhenium(I) integrating the anticancer activities and the bioimaging complexes, [Re(CO) (N−N)L](PF ) (Scheme 1), in which 3 6 capabilities, Re complex can simultaneously induce and thebidentateN−Nligandwas1,10-phenanthroline(phen,1a− monitor the therapeutic effects. 4a) and 4,7-diphenyl-1,10-phenanthroline (DIP, 1b−4b), In this regard, eight organometallic Re(I) compounds respectively, and the monodentate ligand L was various [Re(CO) (N−N)L]PF (N−N = 1,10-phenanthroline (phen) pyridineanalogues.Thesecomplexeswereobtainedbyreacting 3 6 or 4,7-diphenyl-1,10-phenanthroline (DIP), Scheme 1) were 10 equiv of excess L with the dechlorination product of Re(CO) (N−N)Cl precursors in THF followed by anion 3 Scheme 1. (A) Chemical Structures of Re(I) Complexes in exchange with NH 4 PF 6 . Among them the synthesis route of ThisWork(CounterIon,PF );(B)X-rayCrystalStructures complexes3aand3bwasanexception,throughdirectreaction 6 o C f la 2 r a ity a ) nd 3a (H Atoms and Counter Ions PF 6 , Omitted for o o f f 2 C a H a 2 n C d l 2 2 . b A , ll re o s f p t e h c e tiv c e o l m y, p w le it x h es S w O e 2 r C e l 2 p i u n rifi a e m d in b i y m r u e m cry a s m ta o lli u z n a t - tion and characterized by ESI-MS (Figures S1−S8, Supporting Information (SI)), 1H NMR spectroscopy (Figures S9−S16, SI), and elemental analysis. Complexes 2a and 3a were characterized by X-ray crystallography, the crystallographic parameters and selected bond angles/length of which were described in Tables S1−S4 (SI), respectively. The perspective view (Scheme 1) showed that 2a and 3a adopted octahedral geometries,inwhichtheN−Nligandandtwocarbonylgroups were coplanar, and the L and another carbonyl group were distributed at the axial position. These findings are consistent with previously reported organometallic rhenium com- plexes.24−26 All of these Re(I) compounds were stable for at least 48 h in PBS (1% DMSO) at room temperature as monitored by UV−visible spectroscopy. 1H NMR spectra of compound 3b in DMSO-d and D O (v/v, 7/3) did not show 6 2 anychangesafter48hincubationatroomtemperature,further demonstrating that the reactive chloromethyl group did not undergo hydrolysis at least in 48 h (Figure S17, SI). The electronic absorption spectra of Re(I) complexes in CH Cl , PBS, and CH CN at 298 K exhibited intense 2 2 3 absorption bands at ca. 250−320 nm and relatively weak absorptions at ca. 320−450 nm, which were mainly attributed B DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle to the spin-allowed intraligand ππ* transitions and metal-to- smaller or even negative, indicating that 1b−4b were more ligand charge transfer (MLCT), respectively (Figure S18, SI). lipophilic than 1a−4a (Figure 1A). In combination with the Upon excitation at 405 nm, these complexes exhibited yellow cellular uptake results, it is found that, to some degree, the emission (maximum ca. 550 nm) from MLCT excited states greater the lipophilicity of these Re(I) complex, the larger the (Figure S19, SI). The photophysical data are summarized in cellular uptake efficacy. Table S5 (SI). Take 2b and 3b as model complexes; their We also studied the cellular uptake mechanism by using 2b emissionintensityandphosphorescence lifetimesexhibitedpH and 3b as examples. Confocal microscope revealed that insensitivity (Figure S20, SI) but excellent O sensitivity 2 incubation of A549 cells with 2b and 3b at lower temperature (Figure S21, SI). Stern−Volmer plots revealed a good linear (4 °C) or pretreatment of A549 cells with carbonyl cyanidem- relationship: the higher the oxygen concentration, the weaker chlorophenyl hydrazone (CCCP, a metabolic inhibitor) the phosphorescence andthe shorter the lifetime. (Figure S22, resulted inreduced cellularuptake efficiency, whereas pretreat- SI) 2.2. Cellular Uptake and Localization. Due to the ment of cells with chloroquine (an endocytosis modulator) excellentphosphorescencepropertiesofRe(I)complexes,their displayed negligible effect on the cellular uptake efficiency intracellular distribution can be easily monitored by fluo- (Figure S24, SI) These results indicate that the penetration of rescence microscopy. Laser scanning confocal microscopic these Re(I) complexes into A549 cells is dominated by an observation showed that complexes 1b−4b effectively pene- energy-dependent process rather than endocytic pathways. trated into A549 cells after 20 min incubation, as indicated by Then study was conducted to explore the location of the their intense emission in the cytoplasm (Figure S23B, SI). efficiently uptaken complexes 1b−4b in living A549 cells. In However, the cellular uptake of complexes 1a−4a was not confocal microscopic investigations, a commercially available efficient under the same conditions; even when the incubation mitochondria-specific stain, MitoTracker Deep Red (MTDR, timewaselongated,theintracellularemissionof1a−4awasnot 100 nM), was used. Figure 2A showed a high degree of clearly visualized (Figure S23A, SI). On the other hand, as colocalization between the visualized green phosphorescence rhenium is an exogenous element, the cellular uptake efficacies from Re(I) complexes 1b−4b (10 μM) and MTDR in A549 of Re(I) contents can be quantitatively determined by using cells, the Pearson’s correlation coefficients of which were inductively coupled plasma−mass spectrometry (ICP-MS; Figure 1B). After incubation of A549 cells with 20 μM determined to be 85%, 87%, 90%, and 86%, respectively. Meanwhile, negligible colocalization was observed between Figure1.(A)LipophilicityofRe(I)complexes.logP ismeasured O/W asthelogarithmicratiooftheconcentrationofcomplexesinn-octanol tothatintheaqueousphase.(B)CellularuptakeofRe(I)complexes inA549 cells measuredby ICP-MS. complexes for 1 h, the Re(I) contents in the whole cell were found remarkably higher for b series complexes rather than a series complexes, which was consistent with the confocal microscope. It should be noted that both 3a and 3b with the chloromethylpyridyl moiety exhibited the highest cellular uptake among each series. To find out the influencing factors of cellular uptake efficacies for these Re(I) complexes, their lipophilicity referred to as log P o/w (octanol/water partition coefficient) values was F im ig a u g r e e d b 2 y . C (A L ) SM In . t A er 5 c 4 e 9 llu c la e r lls c w ol e o r c e al i i n z c a u ti b o a n ted of w 1 it b h − 2 4 0 b μ w M ith and M a T t D 2 R 0 determined by the shake-flask method and UV−visible minandthenstainedwithMTDR(150nM,30min)at37°C(1b− 1 sp b e − ct 4 r b o , sc w o h p i y c . h I p t o i s s se c s l s e e a d rly mo se r e e n ex t t h e a n t de th d e de lo lo g ca P li o z / e w d v a a r l o u m es at o ic f 4 = b 6 , 5 λ 5 ex = ± 4 2 0 0 5 n n m m ) a . n ( d B) λ e D m i = str 5 ib 5 u 0 t ± ion 30 of n c m o ; m M p T le D xe R s , 1 λ b e − x = 4 6 b 3 ( 3 2 n 0 m μM an , d 1 λ h em ) N−Nligand,werepositivewhilethoseof1a−4awererelatively in various organelles of A549 cells measured by ICP-MS. C DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle 1b−4b and the lysosome-specific stain LysoTracker Deep Red extracted from complex 3b-treated cells displayed intensive (LTDR; Figure S25, SI). emissionwhereasemissivebandswerenotobservedinproteins ICP-MS was also utilized to further ascertain the isolatedfrom1b-,2b-,and4b-treatedA549cells.Theemission mitochondria-specific accumulation of 1b−4b in A549 cells. originated from the irradiation-induced phosphorescence of Afterincubationwith1b−4b(20μM)for1h,differentcellular Re(I) complex. This result indicates that complex 3b components such as mitochondria, cytoplasm, and nucleus containingchloromethylpyridylmoiety,ratherthanothertested wereisolatedfromA549cells,respectively, andRe(I)contents Re(I) compounds, could conjugate to intracellular proteins. in different cellular components were quantified. Figure 2B Then the immobilization ability of complexes 1b−4b in revealedthatRe(I)contentsinmitochondriaweremuchhigher mitochondria throughcovalent bindingwasfurther ascertained than those in cytoplasm and nucleus, indicating that 1b−4b by detecting the changes in the emission intensity of Re(I) couldtargetmitochondriawithhighspecificityinA549cells.It complexes upon cell fixation and elution. A commercially isworthmentioningthat,asmeasuredbyICP-MS,complex3b available mitochondrial-immobilized probe MTDR, also with chloromethylpyridyl moiety exerted the highest cellular containing chloromethylpyridyl moiety and possessing reac- uptake efficacies and the greatest mitochondrial accumulation tivity for thiol group, was used as a positive control, which over other Re(I) complexes explored in this work. could be retained during cell fixation.41 Another commercially 2.3. Mitochondrial Immobilization. Ithas been reported available mitochondrial dye, Rhodamine 123, was used as a that compounds containing a chloromethyl group are capable negative control, which could be easily washed away once the of reacting with the thiol groups of cysteine residues within loss of mitochondrial membrane potential occurred.42 As polypeptide and proteins to from stable covalent bonds.22,38 A shown in the confocal microscopy image (Figure 3B), after model experiment was conducted by treating chloromethyl- cell fixation and repeated elution the intracellular emissions of containing complex 3b with glutathione in a mixed solution of complexes 1b, 2b, 4b, and Rhodamine 123 were barely seen, alkaline water and methanol to demonstrate its reactivity with suggesting that they were easily washed away from fixed cells. biologically relevant thiols. The product was characterized by Meanwhile, only MTDR and complex 3b were still largely ESI-MS,andtheformationofarhenium−glutathioneconjugate retained in A549 cells under the same conditions, as indicated [Re(CO) (DIP) (py-3-CH −GSH)]+ was confirmed by the by the intensive intracellular emission. The combination of 3 2 molecular ion peak and a series of reasonable fragmentation these observations with colocalization results indicates that peaks (Figure S26, SI), indicating that complex 3b possesses complex3btargetsandselectivelyaccumulatesinmitochondria the reactivity for a biologically relevant thiol group. The and then conjugates with thiol-containing proteins in intracellular protein conjugation abilities of complexes 1b−4b mitochondria due to its thiol reactivity, thus being capable of were investigated using polyacrylamide gel electrophoresis acting asaneffectivemitochondria-immobilizedprobe forreal- (PAGE). As shown in Figure 3A, proteins isolated from lysed time tracking of mitochondrial morphologies. Moreover, the Re(I)-treated (30 μM, 1 h) A549 cells were gel-separated and immobilization ability may be beneficial for prolonging the clearlyvisualizedafterCBB(CoomassieBrilliantBlue)staining; retention time of 3b in mitochondria, accordingly enhancing however, upon irradiation at 365 nm only the proteins mitochondria-mediated cytotoxicity. 2.4. In Vitro Cytotoxicity and Selective Killing Cancer Cells. The in vitro cytotoxicity of the tested eight Re(I) compounds against several different cell lines (HeLa, human pulmonary carcinoma; A549, human lung carcinoma; A549R, cisplatin-resistant cell line; LO2, human normal liver cell line) was determined by MTT assay after 48 h of treatment in the dark. The resulting IC50 values were calculated and listed in Table1,andcisplatinwasadoptedasthepositivecontrol.Itcan be seen that complexes 1a−4a with phen ligand only showed Table 1. IC (μM) Values of Tested Compounds Toward Different Ce 50 ll Lines a IC (μM) 50 compd A549 A549R HeLa LO2 1a >100 >100 >100 >100 2a >100 >100 >100 >100 3a 75.8±2.3 37.3±1.1 64.6±2.2 >100 4a 39.8±0.7 36.5±1.8 52.5±3.0 47.9±1.5 1b 3.9±0.7 1.2±0.5 0.95±0.11 3.1±0.5 Figure 3. (A) Retaining of emission intensity of Rh123 (150 nM), 2b 5.5±0.6 2.7±0.5 1.7±0.4 7.6±0.9 MTDR (150 nM), and 1b−4b (20 μM) after fixation and washing. 3b 3.4±0.6 0.75±0.12 0.52±0.07 18.7±1.1 A549cellswereincubatedwithindicatedcompoundfor1h,fixedby 4b 22.4±1.2 8.5±1.1 5.9±0.9 6.4±0.7 paraformaldehyde, and washed twice with PBS/DMSO (9/1, v/v). cisplatin 21.5±2.5 65.6±1.6 8.9±1.0 29.9±2.1 Thecellswereimagedbytheconfocalinstrumentaftereachstep.Scale bar: 10 μm. (B) SDS-PAGE analysis of proteins purified from lysed aIC values are drug concentrations necessary for 50% inhibition of 50 1b−4b (20 μM, 1 h)-treated A549 cells. The gel was scanned with cellviability.Dataarepresentedasmeans±standarddeviations(SDs) transmissiveultraviolet(λ =365nm,top)andthenstainedwithCBB obtained in at least three independent experiments, and the drug ex (bottom). treatment period was 48 h. D DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle moderate cytotoxicity or even noncytotoxicity, whereas selectively localize in mitochondria, their capability to induce complexes 1b−4b with DIP ligand showed much higher mitochondrial dysfunction was investigated. Mitochondrial cytotoxicitythancisplatinagainstallthehumancancercelllines dysfunctionisusuallyaccompaniedbythelossofmitochondrial tested,withIC valuesrangingfrom0.52±0.08to22.4±0.5 membrane potential (MMP, ΔΨ ); thus the Re(I) complex- 50 m μM. This is in accordance with the cellular uptake level: the induced changes in MMP were initially detected by using flow higher the cellular uptake level, the greater the cytotoxicity of cytometry with 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimi- Re(I) complexes. dazolylcarbocyanine iodide (JC-1) staining, a mitochondria- On the other hand, selectively killing cancer cells with selective aggregate dye.43 The loss of MMP was characterized minimum side effect on normal cells is one of the major by an increase in green fluorescence (JC-1 monomers) and a obstacles for cancer treatment. According to the IC values decrease in red fluorescence (JC-1 aggregates). As shown in 50 listedinTable1,complexes1b−4balsopossesssomeextentof Figure 5A, in control cells red fluorescence was mainly selectivity toward cancer cell lines with less toxicity against observed indicating the high membrane potentials; after normal cells. The high cytotoxicity against the A549R cell line treating cells with 1b−4b for 6 h in the dark, a significant also indicates that they could overcome cisplatin resistance. dose-dependent red-to-green color shift is observed indicating Among them 3b exhibits the highest antiproliferative activity the remarkable decrease in MMP. Take 3b-treated cells as an against cancer cells, which is ca. 7-fold, 17-fold, and 87-fold example,thepercentageofcellswithdepolarizedmitochondrial morepotentthancisplatininkillingA549,HeLa,andcisplatin- membranes increased from 3.17 ± 0.5% (control) to 28.4 ± resistant A549R cells, respectively. Notably, 3b also possesses 1.1% (2.5 μM), 77.6 ± 0.5% (5 μM), and 91.7 ± 0.5% (10 thehighestselectivitytowardcancercellsA549(ca.6-fold)and μM), respectively. The representative JC-1 red/green ratio HeLa cells (ca. 36-fold) over normal cells LO2. signals in Re(I)-treated cells were also compared (Figure S27, A LO2 and A549 cell coculture model was constructed to SI), showing that the most prominent Re(I) complex-induced further elucidate the capability of 3b to selectively kill cancer MMP loss was observed in 3b-treated cells. The capability of cells(Figure4).Inthismodel,Hoechststainingwasperformed these Re(I) complexes to induce MMP loss is correlated with their cytotoxicity to some extent. Because mitochondrion is known as the bioenergetic center for cell metabolism, we further explored the impact of Re(I) complexesonmitochondrialmetabolicandbioenergeticstatus, mainly the intracellular ATP level and mitochondrial respiration. As shown in Figure 5B, Re(I)-treated A549 cells exhibit a dose-dependent decrease in the intracellular ATP levels compared with the untreated cells. Among all the tested compounds, 3b exhibits the strongest ability to reduce ATP levels, decreasing the ATP level from 120 ± 3.2 nM (control) to57.0±2.32nM(1.25μM),30.5±2.6nM(5μM),and15.3 ± 3.1 nM (10 μM) per million cells, respectively. Notably, the capability of 1b−4b to reduce the ATP levels in A549 cell is Figure4.Capabilityofcomplex2band3b(5μM,24h)toselectively correlated with their cytotoxicity. Further, we chose mitochondria-targeted complex 2b and induce cancer cell apoptosis in an A549/LO2 cocultured cell model measuredbyannexinV/PIdoublestaining.A549cellswereprelabeled mitochondrial-immobilized complex 3b as model compounds byHoechst (blue). Scale bar:20 μm. to explore their impact on mitochondrial respiration. Seahorse XF24 extracellular flux analyzer was utilized to quantify the onA549nucleiinadvance,andthenbothprestainedA549and mitochondrial respiration by measuring the oxygen consump- LO2 cells were treated with 3b and co-incubated followed by tion rate (OCR).44 Panels C and D of Figure 5 compare the annexinV/PIstaining.Theconfocalmicroscopicimagesclearly OCR changes of untreated and Re(I)-treated A549 cells upon showed that most of the A549 cells (blue nuclei) were also addition of respiration modulators (oligomycin, FCCP, and positively stained by annexin V (green color) and PI (red antimycin A/rotenone) at different time points, which can color), whereas LO2 cells were not successfully stained by target different components on the electron transport chain. annexin V/PI (Figure 4 lower panel). This indicates that Initially,treatmentwith3b(0.5and1.0μM)causesadramatic apoptoticcelldeathwasselectivelyinducedinA549cellswhile dose-dependent decrease in the basal OCR from 5.51 ± 0.31 LO2cells were still vigorous after the same treatment with 3b. (control), 2.29±0.30 (0.5μM),and1.01 ±0.12(1.0 μM)to Compound 2b was also conducted under the same conditions 0.45±0.14(2.0μM)(Figure5E).Afteradditionofoligomycin as a comparison, showing that the Hoechst staining (blue (an ATP synthase inhibitor, 1 μM), coupled respiration for nuclei) and annexin V/PI staining (green and red color) were ATPsynthesisanduncoupledrespirationfordrivingtheproton notperfectlylocalizedinthesamecells(Figure4upperpanel). pumping−leaking cycle across the inner mitochondrial These observations further confirmed that complex 3b could membrane can be affected.44,45 Compared with control cells, selectively kill cancer cells rather than normal cells while the 3b-treated cells display a prominent dose-dependent decrease selectivity of 2b for cancer cells was not good. The relatively inATPproductionfrom4.44±0.41(control)and0.61±0.11 highercytotoxicity andbettercancercellselectivityofcomplex (0.5 μM) to 0.17 ± 0.05 (1.0 μM) (Figure 5F). Proton leak 3b among all the tested Re(I) compounds may be partially alsochangesin3b-treatedcellsascomparedwithcontrolcells: attributed to its immobilization capability and prolonged lower dose treatment (0.5 μM) enhances proton leak, and retention time in mitochondria. higher dose treatment (1.0 μM) declines proton leak (Figure 2.5. Induction of Mitochondrial Dysfunction and 5G). Then after addition of FCCP (a potent mitochondrial Metabolic Repression. Because complexes 1b−4b could uncouplingagent,1μM),thecontrolabilityofATPsynthaseto E DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle Figure 5. Induction of mitochondrial dysfunction by rhenium complexes 1b−4b. (A) MMP of Re(I)-treated (6 h) A549 cells analyzed by flow cytometryatindicatedconcentrations(JC-1staining,λ =488nmandλ =530±30nm(green)and585±30nm(red)).(B)IntracellularATP ex em levels in Re(I)-treated (6 h) A549 cells. (C−D) Respiratory profiles of untreated and Re(I)-treated A549 cells at the indicated concentrations measuredbyaSeahorseXF24extracellularfluxanalyzer.Respirationmodulatorsoligomycin(1μM),FCCP(1μM),andthemixtureofrotenone(1 μM)andantimycinA(1μM)wereaddedatdifferenttimepoints.TheOCRvalueswerenormalizedto1μgproteindeterminationbytheBCA assay.(E)BasalrespirationcalculatedbysubtractingOCRvaluesaftertheadditionofthemixtureofrotenone(1μM)andantimycinA(1μM)from basalOCR.(F)ATPproductioncalculatedbysubtractingOCRvaluesaftertheadditionofoligomycinfrombasalOCR.(G)Protonleakcalculated by subtracting OCR values after the addition of the mixture of rotenone (1 μM) and antimycin A (1 μM) from OCR values obtained after the additionofFCCP.(H)NonmitochondrialrespirationwastheOCRvalueaftertheadditionofthemixtureofrotenone(1μM)andantimycinA(1 μM). respirationandtheprotongradientiseliminated,shownasthe also causes metabolic repression, shown as decreased basal rebound of the OCR peak curve in control cells. However, a OCR, reduced ATP production, enhanced proton leak, and dramatic decrease inthe OCRpeakrather thanthe rebound is inhibited nonmitochondrial respiration compared with control observedin3b-treatedcellsunderthesameconditions(Figure cells(Figure5E−H).However,itisworthnotingthat2batthe 5D).Thisindicatesthelossofsparerespiratorycapacityof3b- sameconcentrationsdisplaysamuchlessprominentimpacton treated cells.44,45 Finally, a mixture of antimycin A (a the OCR than 3b. mitochondrial complex III inhibitor, 1 μM) and rotenone (a All these results combined with the MMP and ATP assay mitochondrial complex I inhibitor, 1 μM) is added to shut indicate that the mitochondrial-immobilized complex 3b downthemitochondrialrespiratorychainthoroughly,resulting possesses stronger capability to induce mitochondrial dysfunc- in substantial decline of OCR peak both in control and 3b- tion and metabolic inhibition than the nonimmobilized treatedcells(Figure5D).TheremainingOCRvaluesrepresent complexes. This could be explained partially by the nonmitochondrial O consumption, including substrate oxida- mitochondrial immobilization capability of 3b and the 2 tion and cell surface O consumption.44,45 It is observed that prolonged retention time. 2 nonmitochondrial O consumption in 3b-treated cells is much 2.6.Real-TimeTrackingofMitochondriaDysfunction. 2 less (0.18 ± 0.05, 2.0 μM) than that in control cells (1.08 ± Because of the high quantum yields and the capability of 0.11) (Figure 5H). These results indicate that both the targeting and inducing mitochondrial dysfunction, Re(I) mitochondrialrespirationandnonmitochondrialrespirationare complexes may possess great potential acting as theranostic effectively inhibited, contributing to the relatively superior agents for real-time tracking of Re(I)-induced mitochondrial cytotoxicity of 3b. As a comparison, treatment with 2b at the dysfunction. As the lifetimes of2band3bare very sensitive to same concentrations under the same conditions (Figure 5C) oxygenconcentrations,theintracellularoxygenconsumptionin F DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle Re(I)-treated cells can be monitored by phosphorescence 3b-treatedcells.ThisisconsistentwiththeSeahorseresultsthat lifetime imaging (PLIM), accordingly reflecting the extent of 3b can inhibit mitochondrial metabolism and reduce oxygen Re(I)-induced inhibition of mitochondrial respiration and the consumption rate more effectively. PLIM results prove the mitochondrial metabolic status. The higher oxygen concen- promising application of 2b and 3b as theranostic agents tration in living cells represents less oxygen consumption, thus capable of inducing and monitoring the mitochondrial the slower oxygen consumption rate and the more inhibited metabolic inhibition simultaneously. mitochondrial metabolism. In addition, mitochondria are dynamic organelles and The upper panel of Figure 6A shows the PLIM imaging of mitochondrial dysfunction is usually accompanied by the 3b-treated fixed A549 cells under different oxygen concen- abnormal mitochondrial morphology.46 Complexes 2b and 3b were chosen as representatives of mitochondria- nonimmobi- lizedand-immobilizedRe(I)compound,respectively,andtheir emission can be monitored by confocal microscopy thus displaying the real-time changes in the mitochondrial morphology. As shown in Figure 7 and Figure S29A (SI), Figure 7. Real-time tracking of mitochondrial morphology in 3b- Figure6.(A)PLIMimagesof3b-treated(10μM,30min)fixedand treated A549 cells (20 μM)by CLSM. Scale bar: 5μm. livingA549cellsunderdifferentoxygenpartialpressuresat37°C(λ ex = 405 nm and f = 0.5 MHz). Scale bar: 20 μm. (B) Luminescence lifetime distributions of PLIM imaging of 3b-treated fixed cells. (C) both2band3b(10μM)couldeffectivelypenetrateintoA549 Fitted Stern−Volmer plots of 3b. Red dots and green triangles cells in 5 min, the intensive emission of which ascertained the represent the corresponding data collected from 3b-treated fixed and healthy mitochondria with normal filamentous network. living cells,respectively. Prolonged treatment of A549 cells with 2b and 3b caused gradualmitochondrialswelling.Aftertreatmentfor40min,the trations, and elongated lifetimes are observed upon decreasing hollow spheres and solid granules can be clearly seen in most O concentrationfrom21%to2%.Thecorresponding lifetime cells, representing the extremely swollen and damaged 2 distribution histogram is shown in Figure 6B. Because it is mitochondria. Notably, with the continuous extension of generally supposed that the O concentration in fixed cells is incubation time the capability of 2b and 3b to specifically 2 thesameastheextracellularenvironment,acalibrationcurveof localize in mitochondria may be affected by the loss of intracellular O concentrations as a function of phosphor- membranepotentialandmitochondrialdamage.InFigureS29B 2 escencelifetimesof3bisconstructed(Figure6C),andthered (SI), after treatment for 2 and 4.5 h, 3b was still colocalized dots are collected from the data of Figure 6B. Under the same excellently with MTDR (Pearson’s correlation coefficients of conditions, the calibration curve of 2b is also obtained using 85−91%),butthecolocalizationdegreeof2bwithMTDRwas PLIM imaging of 2b-treated fixed A549 cells (Figure S28, SI). reduced with lower Pearson’s correlation coefficients of 53− Then the PLIM imaging of 2b- and 3b-treated living cells is 56%. This can be explained by the fixation of 3b in collected (Figure 6A and Figure S28A, bottom panel, SI). mitochondria and the escape of nonfixable 2b from Underdifferentincubationconditions,thelifetimechangesand mitochondria when the membrane potential was lost. These the intracellular O concentrations in Re(I)-treated living cells findings indicate that mitochondrial-immobilized complex is 2 canbecalculatedaccordingtothecalibrationcurves,whichare more advantageous for real-time tracking of morphological representedasgreentrianglesinFigure6Candsummarizedin changes during mitochondrial damage. Tables S8 and S9 (SI). 2.7. Intracellular ROS Levels. Intracellular ROS is one of Forexample,underambientconditions,thelivingA549cells the most important mediators of cell death, and mitochondria treatedwith2band3bexhibitaphosphorescentlifetimeofca. are known as a major source of ROS. Hence the capability of 470 and 410 ns, respectively, indicating the mitochondrial mitochondria-targeted compounds 1b−4b to induce intra- oxygenconcentrationsof24%and29%.Whentheextracellular cellular ROS elevation was examined by using flow cytometry oxygen concentration is reduced to 2%, the phosphorescent with 2′,7′-dichlorofluorescein diacetate (H DCFDA) staining. 2 lifetimes of 2b and 3b were elongated to 650 and 620 ns and H DCFDA is naturally nonfluorescent and can be converted 2 themitochondrial oxygen concentrations werecalculated tobe into highly fluorescent 2′,7′-dichlorofluorescein (DCF) by 8% and 10%, respectively. The higher oxygen concentration in intracellular ROS.47 3b-treatedratherthan2b-treatedcellsislikelytheresultofless After 6 h treatment, 1b−4b caused a dramatically dose- oxygen consumption and thus a slower respiration rate, dependentincreaseinintracellularROSlevels(Figure8A).Ata indicating that mitochondrial metabolism is more inhibited in concentrationof10μM,themeanfluorescentintensityofDCF G DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle c, from the membrane between the space of mitochondria to cytosol,thusinitiatingthedeathsignalingcascade.48Complexes 2b and 3b were chosen as representative of mitochondrial- nonimmobilized and -immobilized Re(I) compound, respec- tively, to explore their capability to induce apoptosis. First, confocal microscopy was utilized to monitor the nuclear morphology of Re(I)-treated A549 cells stained with Hoechst 3342 (Figure S30, SI). Control cells show normal morphology and exhibit homogeneous nuclear staining, whereas treatment with2band3b(8μM,24h) bothincrease thepercentageofapoptoticcellswithmorphologicalcharacter- istics, e.g., membrane blebbing, cell shrinkage, and nuclear fragmentation.49 Next, phosphatidylserine externalization50 was detected as a hallmark of early apoptosis using Annexin V/PI labeling. As shown in flow cytometry analysis (Figure 9A), annexin V- positive/PI-negative cells are considered as early apoptotic Figure 8. (A) Intracellular ROS generation in Re(I)-treated (2.5−10 μM,6h)A549cellsmeasuredbyflowcytometry(λ =488nmand ex λ =525nm).(B)Dose-dependentinhibitionofcelldeathuponthe em incubation of A549 cells with ROS inhibitor NAC. Data are represented as means ±SDof three independent experiments. in 1b−4b-treated cells was approximate 4−7-fold higher than that in vehicle-treated cells. Upon the treatment of ROS inhibitor NAC, the cell viability was increased dramatically depending on the NAC concentration (Figure 8B). These findingsindicatethatcomplexes1b−4bcaninduceintracellular ROS elevation and ROS-dependent cell death. Because of the inefficient cellular uptake and low cytotoxicity of 1a−4a, they Figure 9. (A) Annexin V/PI double staining analyzed by flow were excluded from this and the following intracellular cytometry. A549 cells were incubated with 2b and 3b for 6 h. (B) investigations. Representative TEM images showing the morphological features of 2.8. Induction of Apoptosis. Mitochondria are essential A549cellstreatedwith3b(1and3μM)for24h.Panelsdandeare components of the intrinsic pathway of apoptosis, which enlargedviewsofpanelb,andpanelfistheenlargedviewofpanelc, regulatesthereleaseofpro-apoptoticproteins,e.g.,cytochrome respectively. N, nuclear. Scale bars: 2μm. H DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle whileannexinV/PIdoublepositivecellsareassignedtobelate consumption, thus reflecting the Re(I)-induced mitochondrial apoptotic and necrotic. It can be seen that treatment with 3b respiration repression. The excellent phosphorescence of 2b caused a dose-dependent increase in the percentage of cells in and 3b can also be utilized for real-time tracking of Re(I)- both early apoptotic and late/necrotic phases and mainly induced mitochondrialmorphological changes. Bycomparison, induced early apoptosis. After 24 h treatment with 3b, the complex 3b functions as a better theranostic agent for in situ percentages of early apoptotic cells increased from 5.2 ± 0.5% tracking oftherapeutic effect because itis still well-fixed within to 17.5 ± 1.12% (10 μM) and 63.3 ± 0.82% (20 μM). In mitochondria even when the mitochondria are already contrast, the capability of complex 2b to induce apoptosis was damaged. much weaker, and the percentages of cells in early apoptotic Overall, our work shows that targeting mitochondria and phase was only 6.7 ± 0.03% (10 μM) and 11.6 ± 0.01% (20 mitochondrial metabolism is an effective strategy for cancer μM) under the same conditions. The percentages of late therapy. Rational construction of multifunctional theranostic apoptotic cells among 3b-treated cells were also much higher anticancer agents with mitochondrial-immobilization property than those among untreated and 2b-treated cells. can enhance cytotoxicity and selectivity against cancer cells, as Transmission electron microscope (TEM) was also utilized well as the mitochondrial tracking function. to investigate the mitochondrial morphology of 3b-treated A549 cells (1 and 5 μM, 24h). Compared with vehicle-treated 4. MATERIALS AND METHODS cells, 3b induced substantial morphological changes including Re(CO)Cl (Sigma-Aldrich), 4,7-diphenyl-1,10-phenanthroline (DIP, swollen mitochondria with disrupted cristae at lower dose, a Sigma-Al 5 drich), NHPF (Alfa Aesar),pyridine (J&KScientificLtd.), remarkable increase in the numbers of damaged mitochondria, 4 6 3-(hydroxymethyl)pyridine (J&K), 3-(chloromethyl)pyridine (J&K), and cytoplasmic vacuolation at higher dose. These morpho- 3-pyridinylmethanamine(J&K),1-ethlimidazole(J&K),silvertrifluor- logical changes are the typical features of mitochondria omethanesulfonate(Sigma-Aldrich),cisplatin(Sigma-Aldrich),DMSO dysfunction-mediated apoptosis (Figure 9B). (dimethyl sulfoxide; Sigma-Aldrich), MTT (3-(4,5-dimethylthiazol-2- During apoptosis the release of pro-apoptotic proteins from yl)-2,5-diphenyltetrazolium bromide; Sigma-Aldrich), MTDR (Mito- mitochondriacanactivatethedeath-drivingproteolyticproteins Tracker Deep Red FM; Life Technologies, USA), LTDR (Lyso- caspases.51AhomogeneousluminescentCaspase-Gloassaywas Tracker Deep Red FM; Life Technologies), Hoechst 33342 (Sigma- operatedtoelucidatetheeffectsof2band3bonthecaspase-3/ Aldrich), JC-1(5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolyl- carbocyanine iodide; Sigma-Aldrich), and HDCFDA (2′,7′-dichlor- 7 activity. As shown in Figure S31, treatment of 2b and 3b odihydrofluorescein diacetate, Sigma-Aldrich) 2 were used as received. moderately induced the activation of caspase-3/7 in a dose- Caspase-3/7 activity assay kit, Cell Titer-Glo luminescent Cell dependent manner, indicating that both 2b and 3b can induce Viability Assay kit, and ribonucleotide triphosphates (rRTPs) kit caspase-dependentapoptoticpathways.Thiscanbeconsidered werepurchasedfromPromega(USA).Allthetestedcompoundswere as a consequence of Re(I)-induced mitochondrial dysfunction. dissolved in DMSO and diluted by PBS just before the cellular experiments,andtheconcentrationofDMSOwaslessthan1%(v/v). 3. CONCLUSIONS NMR spectra were recorded on a Bruker Avance 400 spectrometer (Germany). ESI-MS were recorded on a Thermo Finnigan LCQ Insummary,eightorganometallicRe(I)tricarbonylpolypyridyl DECA XP spectrometer (USA). Microanalysis (C, H, and N) was complexeswithdifferentlipophilicitieshavebeensynthesizedas carriedoutusinganElementalVarioELCHNSanalyzer(Germany). mitochondria-targeted anticancer agents. Among them, com- UV/vis spectra were recorded on a Varian Cary 300 spectropho- plexes 1b−4b can quickly and effectively penetrate into A549 tometer (USA). Emission measurements were conducted on an FLS cells and specifically localize in mitochondria while the cellular 920 combined fluorescence lifetime and steady state spectrometer uptake of 1a−4a is not efficient. The cytotoxicities of 1b−4b (Japan). Quantum yields of luminescence at room temperature were are superior to cisplatin against various cancer cells, including calculated using [Ru(bpy) 3 ](PF 6 ) 2 as the reference. Oxygen cisplatin-resistant A549 cells, whereas the cytotoxicities of 1a− concentration was controlled by flow counters (HORIBA STEC, SEC-E40JS,60cm3(STP)min−1)oroxygenconcentration-changeable 4a are much lower. This is highly related to their lipophilicity multigas incubator (Thermo Scientific, SERIES II WATER JACKET andcellularuptakeefficacy.Notably,complexes3bcontaininga CO Incubator,Model:3131,S/N:112620-1988)duringthespectra, 2 reactivechloromethylpyridylmoietycanbeimmobilizedwithin lifetimemeasurements,andcellimaging.ThePLIMsetupisintegrated mitochondria probably through nucleophilic substitution with with an Olympus FV1000 laser scanning confocal microscope reactive thiol groups within mitochondrial proteins. The equipped with a 40 immersion objective lens. The lifetime values mitochondrial immobilization of 3b results in relatively higher were calculated with professional software provided by PicoQuant cytotoxicity against cancer cells than other nonimmobilized Company. 4.1. Synthesis and Characterizations. General Synthetic Re(I) complexes. Moreover, 3b can selectively kill cancer cells Procedure of [Re(CO) (N−N)L](PF). A mixture of precursor [Re- among the co-incubated A549 and LO2 cells, whereas the (CO) (N−N)]Cl (0.2 3 0 mmol) a 6 nd AgCF SO (0.30 mmol) in cancer cell selectivity was not good for the nonimmobilized CHC 3 N (50 mL) was refluxed overnight unde 3 r N 3 protection. After complex 2b under the same condition. This may be attributed rem 3 oving off-white AgCl precipitate, the remain 2 ing solution was tothehighercellularuptakeefficacyaswellaslongerretention evaporated to obtain yellow solids of [Re(CO)(N−N) (CHCN)]- 3 3 time in mitochondria. (CFSO), which was used directly for further reaction without 3 3 Mechanism studies show that mitochondria-targeted 1b−4b purification. L (2 mmol) was added into 50 mL THF solution of mainly induce a series of mitochondria-dependent events, [Re(CO) 3 (N−N) (CH 3 CN)](CF 3 SO 3 ) and refluxed for 20 h under including mitochondrial damage, mitochondrial respiration N 2 protection. The solvent was evaporated, and the resulting solids were resolved in3 mL of CHCN, andthen aqueous solution of 10- inhibition, ATP production depletion, ROS level elevation, 3 foldexcessofNHPF wasaddedandstirredforanother1h.Finally, and caspase-dependent cellular apoptosis. By comparison, 4 6 the precipitates were washed with diethyl ether and further mitochondria-immobilized 3b causes more effective repression recrystallized from CHCN/diethyl ether. However, the synthesis of mitochondrial metabolism than nonimmobilized complexes. routeof3aand3bwasa 3 nexceptionandpresentedinSchemeS1(SI): Simultaneously,theO 2 -sensitivephosphorescentlifetimesof2b 2aor2bwasresolvedinminimumCH 2 Cl 2 and5mLofSO 2 Cl 2 was and3bcanbeutilizedforPLIMimagingofintracellularoxygen added, the mixture was refluxed for 5 h under N protection, and 2 I DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle NaOHsolutionwassimultaneouslyusedasthetailgasabsorber.The 7.69(s,10H;H3−H7,H10−H14ofPh-phen),7.40−7.44(m,J=7.8, 2 solventwasevaporated,andtheresultingsolidswereresolvedin3mL 5.8 Hz, 1H; H5 of pyridine), 4.70 (s, 2H; 2H; CH). ESI-MS: m/z 2 ofCHCN;aqueoussolutionof10-foldexcessofNHPF wasadded 729.9 [M −PF−]+. 3 4 6 6 and washed several times with diethyl ether and recrystallized from [Re(CO)(DIP)(L)](PF) (4b). Yield: 122.7 mg (70%). Elem. Anal. 3 4 6 CHCN/diethyl ether. (%).CalcdforC H FNOPRe·4HO(928.15):C,44.72;H,3.48; 3 33 24 6 4 3 2 [Re(CO) (phen)(L)](PF) (1a). Yield: 113 mg (82%). Elem. Anal N,6.04;C/N,7.40.Found:C,44.63;H,3.51;N,5.89;C/N,7.57.1H 3 1 6 (%). Calcd for C H FNOPRe·0.05ether (675.02): C, 37.05; H, NMR(400MHz,DMSO):δ9.48(d,J=5.4Hz,2H;H1andH16of 20 13 6 3 3 2.97; N, 6.11; C/N, 6.06. Found: C, 36.92; H, 2.94; N, 6.20; C/N, Ph2-phen),8.42−8.44(m,2H;H1,H6ofpyridine),8.15−8.20(m,J 5.95.1HNMR(400MHz,DMSO);δ9.78(dd,J=5.1,1.3Hz,2H; =5.4Hz,4H;H8,H9,H2,H15ofPh-phen),7.72(s,11H;H3−H7, 2 H1andH8ofphen),8.98(m,2H;H3andH6ofphen),8.46(dd,J= H10−H14 of Ph-phen, H4 of pyridine), 7.40−7.44 (dd, J = 7.8, 4.8 2 6.5,1.4Hz,2H;H1andH5ofpyridine),8.32(s,2H;H4andH5of Hz,1H;H5ofpyridine),4.68−4.71(m,2H,2H;CH).3.97−4.00(m, 2 phen), 8.30−8.20 (m, 2H; H2and H7 of phen), 7.87 (tt, J = 7.7, 1.5 2H,2H;NH).ESI-MS:m/z711.0[M−PF−]+;752.0[M−PF−]+ 2 6 6 Hz,1H;H3ofpyridine),7.33(dd,J=7.6,6.6Hz,2H;H2andH4of + CHCN. pyridine). ESI-MS: m/z 529.9 [M− PF−]+. 4.2 3 .CrystallographicStructureDetermination.Crystalsof2a 6 [Re(CO) (phen)(L)](PF)(2a).Yield: 109.7mg(76%).Elem.Anal. and3aqualifiedforX-rayanalysiswereobtainedbytheslowdiffusion 3 2 6 (%). Calcd for C H FNOPRe (705.03): C, 36.05; H, 2.35; N, of diethyl ether into the acetonitrile solution of complexes. X-ray 21 15 6 3 4 5.96; C/N, 6.04. Found: C, 35.65; H, 2.56; N, 5.94; C/N, 6.00. 1H diffraction measurements were performed on a Bruker Smart 1000 NMR(400MHz,DMSO):δ9.78(dd,J=5.1,1.2Hz,2H;H1andH8 CCDdiffractometerwithMoKαradiation(λ=0.71073Å)at173or ofphen),9.05(dd,J=8.3,1.2Hz,2H;H3andH6ofphen),8.36(t,J 150 K. The crystal structures were solved by direct methods with =7.8Hz,1H;H1ofpyridine),8.32(s,2H;H4,H5ofphen),8.30(d, program SHELXS and refined using the full matrix least-squares J = 5.5 Hz, 1H; H4 of pyridine), 8.26 (dd, J = 8.3, 5.1 Hz, 2H; H2, program SHELXL.40 The CCDC deposit numbers for 2a and 3a are H7ofphen),7.77(d,J=7.9Hz,1H;H2ofpyridine),7.28(dd,J=7.8, 1520808 and 1520690, respectively. Crystallographic data, details of 5.7Hz,1H,H3ofpyridine),5.34(t,J=5.7Hz,1H;OH),4.32(d,J= data collection, and structure refinements are listed in Tables S1 and 5.7 Hz, 2H; CH). ESI-MS: m/z 559.9 [M −PF−]+. S3.SelectedbonddistancesandanglesarelistedinTablesS2andS4. 2 6 [Re(CO) 3 (phen)(L 3 )](PF 6 )(3a).Yield: 103.6mg(70%).Elem.Anal. ThestructuralplotsweredrawnusingthexppackageinSHELXTLat (%). Calcd for C H ClFNOPRe (722.99): C, 34.26; H, 2.38; N, a 30% thermalellipsoids probability level. 20 14 6 3 3 5.43; C/N, 6.30. Found: C, 34.74; H, 2.36; N,5.79; C/N, 6.00. 1H 4.3.CellLinesandCultureConditions.A549,A549cisR,HeLa, NMR(400MHz,DMSO):δ9.79(dd,J=5.1,1.1Hz,2H;H1andH8 andLO2cellswereobtainedfromExperimentalAnimalCenterofSun ofphen),9.05(dd,J=8.3,1.1Hz,2H;H3andH6ofphen),8.56(s, Yat-sen University (Guangzhou, China). Cells were maintained in 1H;H1ofpyridine),8.40(d,J=5.2Hz,1H;H4ofpyridine),8.31(s, DMEM(Dulbecco’smodifiedEagle’smedium;GibcoBRL)orRPMI 2H;H4andH5ofphen),8.26(dd,J=8.3,5.1Hz,2H;H2andH7of 1640 (Roswell Park Memorial Institute 1640, Gibco BRL) medium, phen),7.92(d,J=8.0Hz,1H;H2ofpyridine),7.33(dd,J=7.9,5.7 whichcontained10%FBS(fetalbovineserum;GibcoBRL),100μg/ Hz,1H;H3ofpyridine),4.63(s,2H;CH).ESI-MS:m/z577.9[M− mL streptomycin (Gibco BRL), and 100 U/ml penicillin (Gibco 2 PF−]+. BRL). The cells were cultured in a humidified incubator, which 6 (% [ ) R . e C (C al O cd ) 3 ( f p o h r e C n)( H L 4 )] F (PF N 6 ) O (4 P a R ). e Y ·3 i H eld O : 9 ( 6 7 . 5 6 7 m .59 g ): (6 C 7 , % 3 ) 3 . .2 E 9 le ; m H . , A 2. n 9 a 3 l ; . p ex ro p v e i r d im ed en a t n ,c a e t l m ls o t s re p a h t e e r d e w o i f th 5% veh C ic O le 2 c a o n n d tr 9 o 5 l % (1% air D a M t 3 S 7 O ° ) C w . e I r n eu ea se ch d 21 16 6 4 3 2 N,7.40;C/N,4.49.Found:C,33.13;H,2.89;N,7.48;C/N,4.43.1H as the reference group. NMR(400MHz,DMSO):δ9.42(d,J=4.8Hz,2H;H1andH8of 4.4. Cellular Uptake. A549 cells were treated with 1a−4a (200 phen),8.97(d,J=8.1Hz,2H;H3andH6ofphen),8.45(d,J=47.1 μM)for2hand1b−4b(20μM)for30minat37°C,thenwashed Hz,2H;H1,H4ofpyridine),8.30(s,2H;H4,H5ofphen),8.18(d,J three timeswithice-cold PBS,andvisualizedbyconfocalmicroscopy =3.9Hz,1H;H2ofpyridine),8.09(dd,J=8.1,5.1Hz,2H;H2and (LSM 710, Carl Zeiss, Göttingen, Germany). Emission was collected H 6 3 0 .9 7 0 5 . o 0 f (d [ p M , h J en − = ), P 5 7 F . . 7 6 6 − 7 H ]+ ( z s , + ,1 2 C H H H ; ; 3 H C C 3 H N o 2 . ) f . p E y S ri I d -M ine S ) : , m 6. / 8 z 9− 5 6 5 . 9 6 . 6 0 ( [ m M ,2 − H; P N F 6 H −] 2 + ) ; , a ( t 20 C 55 o μ 0 l M o ± c ) a a l 2 i n z 0 a d n t M i m on T u D p A R o ss n a (2 e y x . 0 c 0 A it 5 n at 4 M i 9 o ) n c o e a r l t ls L 4 T 0 w 5 D er n R e m ( c . 1 o 5 -i 0 nc n u M ba ) te a d t3 w 7 it ° h C 1 f b o − r 4 3 b 0 [Re(CO) 3 (DIP)(L 1 )](PF 6 ) (1b). Yield: 132.1 mg (78%).; Elem. Anal. min. Cells were washed three times with PBS and visualized by (%).CalcdforC 32 H 21 F 6 N 3 O 3 PRe·11H 2 O(827.08):C,37.50;H,4.23; confocalmicroscopyimmediately.TheexcitationwavelengthsofRe(I) N,4.10;C/N,9.14.Found:C,37.27;H,4.27;N,4.02;C/N,9.27.1H complexes, MTDR and LTDR, are 405 and 633 nm, respectively. NMR(400MHz,DMSO):δ9.84(d,J=5.4Hz,2H;H1andH16of Emissionwascollectedat540±20nmfor1b−4band665±20nm Ph 2 -phen),8.62(d,J=6.5Hz,2H;H1andH5ofpyridine),8.23(d,J for MTDR andLTDR. =5.4Hz,2H;H8andH9ofPh2-phen),8.17(s,2H;H2,H15ofPh 2 - ICP-MS Measurement. A549 cells were seeded in 10 cm tissue phen),7.99−7.91(m,1H;H3ofpyridine),7.70(dd,J=10.4,4.9Hz, culturedishesatadensityof1×105cells/mLin5mLofRPMI1640 E 10 S H I-M ;C S 6 : H m 5 / a z tP 68 h 1 2 - .9 ph [ e M n), − 7. P 43 F 6 ( − t ] , + J . =7.9Hz,2H;H4,H2ofpyridine). μ m M ed ) iu fo m r ( 4 In h v a it n r d og 1 e b n − ). 4 C b e ( ll 2 s 0 w μ e M re ) in fo cu r b 1 at h e , d re a s t p 3 e 7 ct ° iv C ely w . it A h ft 1 e a r − d 4 ig a es ( t 1 io 0 n 0 [Re(CO) 3 (DIP)(L 2 )](PF 6 ) (2b). Yield: 142.1 mg (81%). Elem. Anal. in trypsin−EDTA solution, A549 cells were counted and digested in (%). Calcd for C 33 H 23 F 6 N 3 O 4 PRe·10H 2 O (1037.19): C, 46.26; H, 60% HNO 3 at room temperature overnight and then diluted with 2.71; N, 4.90; C/N, 9.44. Found: C, 46.17; H, 2.72; N, 4.78; C/N, Milli-Q HO to obtain 2% HNO solutions for ICP-MS (Thermo- 9.66.1HNMR(400MHz,DMSO):δ9.85(d,J=5.4Hz,2H;H1and Elemental, 2 USA) measurement of 3 the whole cell rhenium contents. H16ofPh 2 -phen),8.52(s,1H;H1ofpyridine),8.44(d,J=5.5Hz, For the measurements of rhenium contents in various organelles, 1H; H6 of pyridine), 8.22 (d, J = 5.4 Hz, 2H; H8 and H9 of Ph 2 - nuclear and cytoplasm fractions were separated by a nuclear and phen),8.16(s,2H;H2andH15ofPh 2 -phen),7.85(d,J=8.1Hz,1H; cytoplasmic protein extraction kit (Shanghai Sangon Biological H4ofpyridine),7.76−7.65(m,10H;C 6 H 5 atPh 2 -phen),7.33(m,1H; Engineering Technology & Services Co. Ltd.) and mitochondria H5ofpyridine),5.40(t,J=5.7Hz,1H;OH),4.41(d,2H;CH 2 ).ESI- fractions were separated by a cell Mitochondria Isolation Kit M (% S [ ) R : . e m C (C / a z O lc ) 7 d 3 1 (D 1 fo . I 9 P r ) [ ( C L M 3 3 ) 3 ] H − (P 2 F 2 P C 6 F ) l 6 F − (3 6 ] N b +. ) 3 . O Y 3 P ie R ld e : ·3 1 H 3 2 7 O .8 ( m 9 g 29 ( .2 7 2 6 ) % : ). C E , l 4 em 2.6 . 5 A ; n H al , . ( c d B a ig l e i e b y s r o t a i t o t i i m n on e a w B nd e io re te M c fr h i e l n l s i- h o Q l l y og p H y r ) e 2 , O p r a e r s e d p d i e lu c b t t y i i o v d e n l i . y lu , T t j i u h n s e g t a s b t e R a f n e o d N re a O rd t 3 h s e st f o o 6 c 0 r k % r s h o H e lu n N t iu i O o m n 3 3.04;N,4.52;C/N,9.44.Found:C,42.52;H,3.03;N,4.47C/N,9.51. with 2% HNO in Milli-QHO. 1HNMR(400MHz,DMSO):δ9.84(d,J=5.4Hz,2H;H1andH16 4.5. Mitoc 3 hondrial Im 2 mobilization. Confocal Microscopy ofPh2-phen),8.69(s,1H;H1ofpyridine),8.52(d,J=5.6Hz,1H; before and after Cell Fixation. A549 cells were cultured and H6ofpyridine),8.22(d,J=5.4Hz,2H;H8,H9ofPh-phen),8.14(s, incubatedwith1b−4b(10μM)andMTDRandRhodamine123for2 2 2H;H2,H15ofPh-phen),7.98(d,J=8.0Hz,1H;H4ofpyridine), h. The cells were then washed twice by PBS, fixed in 2 mL of 4% 2 J DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle paraformaldehyde(v/v),washedwithPBS(10%DMSO,v/v)atleast indicatedconcentrationsfor5hat37°Cundera5%CO atmosphere. 2 3 times, and visualized by confocal microscopy before and after Then cell culture medium was removed and replaced by assay fixation. medium, and the Microplates were placed into a 37 °C non-CO 2 Covalent Binding with Mitochondrial Proteins in SDS-PAGE. incubatortoequilibratefor1h.Afterthecompletionofcellularbasic A549 cells were incubated with complexes 1b−4b (30 μM) for 1 h. respiratory measurements, the repiration modulators at indicated Digestive cells were washed with PBS, lysed using RIPA buffer, and concentrations were loaded consecutively in sequence to affect the centrifuged to collect protein supernatant. After determination of electron transport chain, and the oxygen consumption in different protein concentration by BCA assay (Novagen Inc., USA), the same stages was measured directly. Wave 2.0.0 Software was utilized to amount of cellular total proteins was denatured by being boiled in calculate the key metabolic parameters of mitochondria. sampleloadingSDS-PAGE(polyacrylamidegelelectrophoresis)buffer Real-Time Tracking of Changes in Mitochondrial Morphology. for 10 min. The denatured proteins were loaded and separated by A549cellswereculturedin60mmdishes(Corning)andtreatedwith SDS-PAGE.Thenthegelwasscannedbytransmissionultraviolet(λ ex complexes2band3b(10μM).Cellimagingwascollectedevery5min = 365 nm) and analyzed by a FluorChem M imaging station by confocal microscopy (λ = 405 nm, λ = 540 ± 20nm). ex em AlphaView Software (ProteinSimple, CA, USA). The same gel was PLIM Imaging of Mitochondrial Respiration. A549 cells were thenstainedwithCommassieBrilliantBlue-250dye,andtheimaging culturedin60mmdishesandtreatedwithcomplexes2band3b(20 was alsocaptured and analyzed. μM)for30minat37°CfollowedbytwicePBSwash.Then1mLof 4.6. Cytotoxicity and Selectivity against Cancer Cells. MTT PBS was added for imaging of living cells. For imaging of fixed cells, Assay.Cellswereculturedin96-wellplatesandgrowntoconfluence. thecellsshouldbefixedby4%paraformaldehydeat4°Covernightin ThecompoundsweredissolvedinDMSO(1%,v/v),anddilutedwith advance.Thenthephosphorescentlifetimeimagingoflivingcellsand fresh media immediately. The cells were incubated with a series of fixed cells under different O concentrations was recorded by PLIM concentrations of the tested compounds for 44 h at 37 °C. A 20 μL 2 (setupisintegratedwithanOlympusFV1000laserscanningconfocal aliquotofMTTsolutionwasthenaddedtoeachwell,andtheplates microscope; λ = 405 nm, λ = 540 ± 20 nm). The O were incubated for an additional 4 h. The medium was carefully ex em 2 concentrations and O/N ratios were controlled by adjusting the removed,andDMSOwasadded(150μLperwell)andincubatedfor flowcountersintheliv 2 ece 2 llstation.Thecellimagesweretakenafter 10minwithshaking.Theabsorbanceat595nmwasmeasuredusinga thecellplateswereequilibratedunderacertainoxygenconcentration microplate reader (Infinite M200 Pro, Tecan, Man̈ nedorf, Switzer- for more than 30 min. The lifetime values were calculated with land). professional software provided by PicoQuant Co. Selective Killing Cancer Cells over Normal Cells. A549 cells were 4.8.MeasurementofIntracellularROS.Cellsweretreatedwith prestainedwithHoechst3342for10min,thenwashedwithPBSthree 1b−4battheindicatedconcentrationsfor6handthenincubatedwith times, suspended in fresh medium, then seeded into 35 mm dishes 10 μM HDCFDA inserum-free DMEM for15 minat 37°Cinthe with the same amount of LO2 cells, and incubated for 24 h for cell 2 dark. After the cells were washed twice with serum-free DMEM, the a 3 t b ta ( c 1 h 0 m μ en M t. ) T f h o e r 2 ce 4 ll h m , s ix ta tu in r e e d sw w e it r h ei 5 nc μ u L ba o t f ed An w n it e h xi c n o -V mp a l n e d xe 1 s 0 2b μL an o d f fluorescence intensity of the cells was measured immediately by flow cytometry with excitation at 488nmand emission at 530 nm.Green propidiumiodide,andimmediatelyvisualizedbyconfocalmicroscopy. MFI were analyzed usingFlow Jo7.6 software (Tree Star) 4.7. Mitochondrial Dysfunction and Real-Time Tracking. 4.9.InductionofApoptosis.TransmissionElectronMicroscopy. MMPAssay.A549cellswereculturedin60mmdishes(Corning)and treatedwithcomplexes1b−4bfor6h.Thecellswerethencollected, A549cellswereculturedin60mmdishes(Corning)andtreatedwith resuspendedat1×106/mLinprewarmedPBScontainingJC-15μg/ 2band3batindicatedconcentrationsfor24h.Thecellswerewashed mL,andincubatedfor30minat37°C.Subsequently,thecellswere by PBS and fixed overnight at 4 °C in phosphate buffer (pH 7.4) washedtwicewithPBSandimmediatelyanalyzedinaflowcytometer. containing 2.5% glutaraldehyde. After treatment with osmium tetroxide as postfixative, the cells were stained with uranyl acetate Fluorescence was monitored by measuring both the monomer (527 andleadcitrate,observedbyatransmissionelectronmicroscope(JEM nmemission;green)andtheaggregate(590nmemission;red)forms 100 CX, JEOL, Tokyo, Japan). Images were photographed using the of JC-1 following excitation at 488 nm. Red and green MFI were analyzedusingFlowJo7.6software(TreeStar,USA).Foreachsample, Eversmart Jazz program (Scitex). 10, 000 events were acquired. Hoechst Staining. A549 cells were seeded into 35 mm dishes AssayofATPConcentrationinCells.ATPconcentrationofA549 (Corning) and treated with 2b and 3b for 24 h. The cells were then was conducted by the CellTiter-Glo luminescent Cell Viability Assay washed once with PBS and fixed with 4% paraformaldehyde at room (Promega) according to the manufacturer’s instructions. Cells were temperature for 10 min. After that, cells were labeled with Hoechst culturedina96roundblackwellplatefor24htowelt.Asampleof 33342 (5 μg/mL in PBS) for 5 min. The cells were analyzed with a complexes 1b−4b at the indicated concentration was added into the confocal microscope immediately. cell to co-incubate for 6 h. The cell was washed by PBS once and Annexin V/PI Assay. The assay was carried out according to the balancedinPBSfor30minand100μLCellTiter-Gloluminescentcell manufacturer’s protocol. A549 cells were seeded in 6-well plates and viabilityreagentwasaddedintoeachwell.Themixturewaslysedfor2 treatedwithindicatedconcentrationsof2band3bfor24h.Thecells minbyashakenmachineandthenincubatedatroomtemperaturefor wereharvestedandstainedwithAnnexinVandPIasdescribedabove 10 min. The luminescence was measured using a TECAN Infinite atroomtemperaturefor15mininthedarkandanalyzedimmediately M200station.Onthesamecondition,astandardcurvewasobtained; byflowcytometry(λ ex =488nm),andthe absorbanceat 488nmof by the known concentration of standard ATP sample, ribonucleotide 1b−4b can be ignored. Data were analyzed by FlowJo Software triphosphates(10mM),wecanobtaintheATPconcentrationofthe (TreeStar). cell. Caspase-3/7ActivityAssay.Caspase-3/7activitywasconductedby Mitochondrial Bioenergetics Analysis. The Seahorse XF24 theCaspase-GloAssaykit(Promega)accordingtothemanufacturer’s Extracellular Flux analyzer (Seahorse Bioscience, Billerica, MA, instructions. Briefly, cells were cultured and treated with different USA) and XF Cell Mito Stress Test kit was utilized to quantify the concentrationsof2band3bfor6h,andthen50μLofcelllysatewas mitochondrial respiration by measuring the OCR (oxygen con- addedtoeachwell,followedbytheadditionof50μLofCaspase-Glo sumption rate). According to the manufacturer’s instruction, after a reagents.Themixturewasincubatedatroomtemperaturefor30min, series of preliminary experiments, the final concentrations of the and then the luminescence was measured using a TECAN Infinite respirationmodulatorsweredetermined:oligomycin,1μM;FCCP,1 M200 station. μM; a mixture of antimycin A and rotenone, 1 μM and 1 μM, 4.10. Statistical Analysis. All biological experiments were respectively. A549 cells were seeded in Seahorse 24-well XF Cell performed at least twice with triplicates in each experiment. CultureMicroplatesatadensityof3×104cellsperwell(0.275cm2), Representative results were depicted in this report, and data were cultured for 24 h, and then treated with complex 2b or 3b at the presented asmeans ±standard deviations (SDs). K DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ■ACS Applied Materials & Interfaces ResearchArticle ASSOCIATED CONTENT (10) Wang, F.; Ogasawara, M. A.; Huang, P. Small Mitochondria- * Targeting Molecules as Anti-Cancer Agents. Mol. Aspects Med. 2010, S Supporting Information 31 (1), 75−92. The Supporting Information is available free of charge on the (11) Armstrong, J. S. Mitochondrial Medicine: Pharmacological ACS Publications website at DOI: 10.1021/acsami.7b01764. Targeting of Mitochondria in Disease. Br. J. Pharmacol. 2007, 151, ESI-MS and 1H NMR spectra; stability of rhenium 1154−6115. compounds; UV−visible and emission spectra; lumines- (12)Zhang,X.;Fryknas,M.;Hernlund,E.;Fayad,W.;DeMilito,A.; cence intensities and lifetimes; cellular uptake and Olofsson, M. H.; Gogvadze, V.; Dang, L.; Pah̊ lman, S.; Schughart, L. subcellular localization; MMP assay; PLIM images; 2b- A.; Rickardson, L.; D’Arcy, P.; Gullbo, J.; Nygren, P.; Larsson, R.; induced mitochondrial morphological changes; Hoechst Linder, S. Induction of Mitochondrial Dysfunction as A Strategy for staining;Caspase-3/7assay;crystallographicdata;photo- TargetingTumorCellsinMetabolicallyCompromisedMicroenviron- ments. Nat. Commun. 2014,5, 3295. physical data; ICP-MS data; and lifetime data (PDF) (13)Wang,J.B.;Erickson,J.W.;Fuji,R.;Ramachandran,S.;Gao,P.; Crystallographic information for 2a (CIF) Dinavahi,R.;Wilson,K.F.;Ambrosio,A.L.;Dias,S.M.;Dang,C.V.; Crystallographic information for 3a (CIF) ■ Cerione,R.A.TargetingMitochondrialGlutaminase ActivityInhibits Oncogenic Transformation. Cancer Cell 2010,18,207−219. AUTHOR INFORMATION (14)Strohecker,A.M.;Guo,J.Y.;Karsli-Uzunbas, G.;Price,S.M.; Corresponding Authors Chen,G.J.;Mathew,R.;McMahon,M.;White,E.AutophagySustains *(Z.-W.M.) E-mail: cesmzw@mail.sysu.edu.cn. Mitochondrial Glutamine Metabolism and Growth of BrafV600E- *(Q.C.) E-mail: caoqian3@mail.sysu.edu.cn. Driven Lung Tumors. Cancer Discovery 2013,3,1272−1285. (15)Fendt,S.-M.;Bell,E.L.;Keibler,M.A.;Davidson,S.M.;Wirth, ORCID G. J.; Fiske, B.; Mayers, J. R.; Schwab, M.; Bellinger, G.; Csibi, A.; Zong-Wan Mao: 0000-0001-7131-1154 Patnaik, A.; Blouin, M. J.; Cantley, L. C.; Guarente, L.; Blenis, J.; Author Contributions Pollak,M.H.;Olumi,A.F.;VanderHeiden, M.G.;Stephanopoulos, § J.Y. and J.-X.Z. contributed equally to this work. G. Metformin Decreases Glucose Oxidation and Increases the Dependency of Prostate Cancer Cells on Reductive Glutamine Notes T■he authors declare no competing financial interest. Metabolism. Cancer Res. 2013,73,4429−4438. (16)Yuan,P.;Ito,K.;Perez-Lorenzo,R.;DelGuzzo,C.;Lee,J.H.; Shen,C.-H.;Bosenberg,M.W.;McMahon,M.;Cantley,L.C.;Zheng, ACKNOWLEDGMENTS B. Phenformin Enhances the Therapeutic Benefit of BRAFV600E Thanks very much to Prof. Qiang Zhao from Nanjing Inhibition in Melanoma. Proc. Natl. Acad. Sci. U. S. A. 2013, 110 University of Posts and Telecommunications (NUPT) for the (45), 18226−18231. utilization of PLIM equipment. We are grateful for financial (17) Smith, R. A.; Hartley, R. C.; Murphy, M. P. Mitochondria- support from the National Natural Science Foundation of Targeted Small Molecule Therapeutics and Probes. Antioxid. Redox China [Grants 21231007, 21572282, and 21401217], the 973 Signaling 2011,15, 3021−3038. Program [Grants 2014CB845604 and 2015CB856301], the (18) Hu, W.; Kavanagh, J. J. Anticancer Therapy Targeting the Science and Technology Program of Guangzhou [Grant Apoptotic Pathway. Lancet Oncol.2003,4,721−729. (19)Dickinson,B.C.;Chang,C.J.ATargetableFluorescentProbe 201607010379], the Natural Science Foundation of Jiangsu forImagingHydrogenPeroxideintheMitochondriaofLivingCells.J. Province [Grant BK20140600 (to Z.G.)], and Fundamental Am. Chem. Soc. 2008, 130,9638−9639. R■esearch Funds for the Central Universities. (20)Leung,C.W.;Hong,Y.;Chen,S.;Zhao,E.;Lam,J.W.;Tang,B. Z.APhotostableAIELuminogenforSpecificMitochondrialImaging REFERENCES and Tracking. J. Am. Chem. Soc. 2013,135, 62−65. (1) Weinberg, S. E.; Chandel, N. S. Targeting Mitochondria (21)Kelkar,S.S.;Reineke,T.M.Theranostics:CombiningImaging Metabolism for Cancer Therapy. Nat. Chem. Biol. 2014,11,9−15. and Therapy. Bioconjugate Chem. 2011,22,1879−1903. (2) Weinberg, F.; Hamanaka, R.; Wheaton, W. W.; Weinberg, S.; (22)Cao,J.-J.;Tan,C.-P.;Chen,M.-H.;Wu,N.;Yao,D.-Y.;Liu,X.- Joseph,J.;Lopez,M.;Kalyanaraman,B.;Mutlu,G.M.;Budinger,G.R. G.; Ji, L.-N.; Mao, Z.-W. Targeting Cancer Cell Metabolism with S.;Chandel,N.S.MitochondrialMetabolismandROSGenerationare Mitochondria-Immobilized Phosphorescent Cyclometalated Iridium- EssentialforKras-mediatedTumorigenicity.Proc.Natl.Acad.Sci.U.S. (iii) Complexes. Chem. Sci. 2017,8,631−640. A.2010,107 (19), 8788−8793. (23) Li, Y.; Tan, C. P.; Zhang, W.; He, L.; Ji, L. N.; Mao, Z. W. (3) Guo,J.Y.;Chen, H.-Y.; Mathew, R.;Fan,J.;Strohecker, A.M.; Phosphorescent Iridium(III)-Bis-N-heterocyclic Carbene Complexes Karsli-Uzunbas, G.; Kamphorst, J. J.; Chen, G.; Lemons, J. M. S.; asMitochondria-Targeted Theranostic andPhotodynamicAnticancer Karantza,V.;Coller,H.A.;DiPaola,R.S.;Gelinas,C.;Rabinowitz,J. agents. Biomaterials 2015, 39, 95−104. D.;White,E.ActivatedRasRequiresAutophagytoMaintainOxidative Metabolism and Tumorigenesis. Genes Dev.2011, 25, 460−470. (24) Leonidova, A.; Gasser, G. Underestimated Potential of Organometallic Rhenium Complexes as Anticancer Agents. ACS (4)Wallace,D.C.MitochondriaandCancer.Nat.Rev.Cancer2012, 12, 685−698. Chem. Biol.2014,9,2180−2193. (25) Balasingham, R. G.; Coogan, M. P.; Thorp-Greenwood, F. L. (5) Warburg, O. On the Origin of Cancer Cell. Science 1956, 123 (3191), 309−314. Complexes in Context: Attempting to Control the Cellular Uptake (6)Cairns,R.A.;Harris,I.S.;Mak,T.W.RegulationofCancerCell and Localisation of Rhenium fac-Tricarbonyl Polypyridyl Complexes. Metabolism. Nat. Rev. Cancer 2011,11, 85−95. Dalton Trans. 2011, 40,11663−11674. (7) Fulda, S.; Galluzzi, L.; Kroemer, G. Targeting Mitochondria for (26) Lo, K. K. Luminescent Rhenium(I) and Iridium(III) Cancer Therapy. Nat.Rev. Drug Discovery 2010,9,447−464. Polypyridine Complexes as Biological Probes, Imaging Reagents, and (8) Gogvadze, V.; Orrenius, S.; Zhivotovsky, B. Mitochondria as Photocytotoxic Agents. Acc. Chem. Res. 2015,48,2985−2995. Targets for Chemotherapy. Apoptosis 2009,14,624−640. (27)Liu,H.-K.;Sadler,P.J.MetalComplexesasDNAIntercalators. (9) Birsoy,K.;Sabatini, D.M.;Possemato, R.Untuningthe Tumor Acc. Chem. Res. 2011,44,349−359. MetabolicMachine:TargetingCancerMetabolism:ABedsideLesson. (28) Liu, Z.; Sadler, P. J. Organoiridium Complexes: Anticancer Nat. Med.2012,18,1022−1023. Agents and Catalysts. Acc. Chem. Res. 2014,47,1174−1185. L DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX ACS Applied Materials & Interfaces ResearchArticle (29)Abdel-Shafi,A.A.;Bourdelande,J.L.;Ali,S.S.Photosensitized (46) Tait, S. W. G.; Green, D. R. Mitochondria and Cell Death: Generation of Singlet Oxygen from Rhenium(I) and Iridium(III) Outer Membrane Permeabilization and Beyond. Nat. Rev. Mol. Cell Complexes. Dalton Trans. 2007, 251 (24), 2510−2516. Biol. 2010,11,621−632. (30)Leonidova,A.;Pierroz,V.;Rubbiani,R.;Lan,Y.;Schmitz,A.G.; (47) Chen, Q.; Vazquez, E. J.; Moghaddas, S.; Hoppel, C. L.; Kaech, A.; Sigel, R. K. O.; Ferrari, S.; Gasser, G. Photo-induced Lesnefsky, E. J. Production of Reactive Oxygen Species by UncagingofASpecificRe(I)OrganometallicComplexinLivingCells. Mitochondria: Central Role of Complex III. J. Biol. Chem. 2003, Chem. Sci. 2014,5,4044−4056. 278,36027−36031. (31) Wahler, K.; Ludewig, A.; Szabo, P.; Harms, K.; Meggers, E. (48)Hotchkiss,R.S.;Strasser,A.;McDunn,J.E.;Swanson,P.E.Cell Death in Disease: Mechanisms and Emerging Therapeutic Concepts. RheniumComplexeswithRed-Light-InducedAnticancerActivity.Eur. J. Inorg. Chem. 2014,2014,807−811. N. Engl.J. Med. 2009,361, 1570−1583. (49) Detmer, S. A.; Chan, D. C. Functions and Dysfunctions of (32) Yin Zhang, K.; Ka-Shun Tso, K.; Louie, M.-W.; Liu, H.-W.; Mitochondrial Dynamics.Nat. Rev. Mol. CellBiol. 2007,8,870−879. Kam-Wing Lo, K. A Phosphorescent Rhenium(I) Tricarbonyl (50)Riedl,S.J.;Shi,Y.MolecularMechanismsofCaspaseRegulation Polypyridine Complex Appended with a Fructose Pendant That During Apoptosis. Nat. Rev. Mol.Cell Biol. 2004,5, 897−907. Exhibits Photocytotoxicity and Enhanced Uptake by Breast Cancer Cells.Organometallics 2013,32,5098−5102. (33) Lo, K. K.; Hui, W. K.; Ng, D. C. Novel Rhenium(I) Polypyridine Biotin Complexes That Show Luminescence Enhance- ment and Lifetime Elongation upon Binding to Avidin. J. Am. Chem. Soc.2002, 124,9344−9345. (34) Lo, K. K.; Tsang, K. H.; Hui, W. K.; Zhu, N. Synthesis, Characterization, Crystal Structure, and Electrochemical, Photo- physical, and Protein-Binding Properties of Luminescent Rhenium(I) Diimine Indole Complexes. Inorg. Chem. 2005,44,6100−6110. (35) Amoroso, A. J.; Coogan, M. P.; Dunne, J. E.; Fernandez- Moreira,V.;Hess,J.B.;Hayes,A.J.;Lloyd,D.;Millet,C.;Pope,S.J. A.; Williams, C. Rhenium fac-Tricarbonylbisimine Complexes: Bio- logically Useful Fluorochromes for Cell Imaging Applications. Chem. Commun.2007, 3066−3068. (36) Gasser, G.; Pinto, A.; Neumann, S.; Sosniak, A. M.; Seitz, M.; Merz,K.;Heumann,R.;Metzler-Nolte,N.Synthesis,Characterization and Bioimaging of A Fluorescent Rhenium-Containing PNA Bioconjugate. Dalton Trans. 2012, 41, 2304−2313. (37) Louie, M.-W.; Ho-Chuen Lam, M.; Kam-Wing Lo, K. Luminescent Polypyridine Rhenium(I) Bis-Biotin Complexes as Crosslinkers for Avidin. Eur. J. Inorg. Chem. 2009, 2009 (28), 4265− 4273. (38) Amoroso, A. J.; Arthur, R. J.; Coogan, M. P.; Court, J. B.; Fernań dez-Moreira,V.;Hayes,A.J.;Lloyd,D.;Millet,C.;Pope,S.J. A. 3-Chloromethylpyridyl Bipyridine fac-Tricarbonyl Rhenium: A Thiol-Reactive Luminophore for Fluorescence Microscopy Accumu- lates inMitochondria. New J. Chem. 2008,32,1097−1102. (39)Wang,B.;Liang,Y.;Dong,H.;Tan,T.;Zhan,B.;Cheng,J.;Lo, K. K.; Lam, Y. W.; Cheng, S. H. A Luminescent Cyclometalated Iridium(III) Complex Accumulates in Mitochondria and Induces MitochondrialShorteningbyConjugationtoSpecificProteinTargets. ChemBioChem 2012, 13, 2729−2737. (40) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (41)Presley,A.D.;Fuller,K.M.;Arriaga,E.A.MitoTrackerGreen LabelingofMitochondrialProteinsandTheirSubsequentAnalysisby CapillaryElectrophoresiswithLaser-InducedFluorescenceDetection. J.Chromatogr.B:Anal.Technol.Biomed.LifeSci.2003,793,141−150. (42) Poot, M.; Zhang, Y. Z.; Kramer, J. A.; et al. Analysis of Mitochondrial Morphology and Function with Novel Fixable FluorescentStains.J.Histochem.Cytochem.1996,44(12),1363−1372. (43)Wang,H.;Joseph,J.A.QuantifyingCellularOxidativeStressby Dichlorofluorescein Assay using Microplate Reader. Free Radical Biol. Med.1999,27 (5−6),612−616. (44) Ferrick, D. A.; Neilson, A.; Beeson, C. Advances in Measuring Cellular Bioenergetics using Extracellular Flux. Drug Discovery Today 2008,13,268−274. (45) Wu, M.; Neilson, A.; Swift, A. L.; Moran, R.; Tamagnine, J.; Parslow, D.; Armistead, S.; Lemire, K.; Orrell, J.; Teich, J.; Chomicz, S.; Ferrick, D. A. Multiparameter Metabolic Analysis Reveals a Close Link Between Attenuated Mitochondrial Bioenergetic Function and Enhanced Glycolysis Dependency in Human Tumor Cells. Am. J. Physiol. Cell Ph. 2007,292,C125−C136. M DOI:10.1021/acsami.7b01764 ACSAppl.Mater.InterfacesXXXX,XXX,XXX−XXX