Photoactivated Osmium Arene Anticancer Complexes.

{"full_text": " pubs.acs.org/IC Article\n\n\n\n Photoactivated Osmium Arene Anticancer Complexes\n Xuling Xue, Ying Fu, Liang He, Luca Salassa, Ling-Feng He, Yuan-Yuan Hao, Madeleine J. Koh,\n Cl\u00e9ment Souli\u00e9, Russell J. Needham, Abraha Habtemariam, Claudio Garino, Kirill A. Lomachenko,\n Zhi Su,* Yong Qian, Martin J. Paterson,* Zong-Wan Mao,* Hong-Ke Liu,* and Peter J. Sadler*\n Cite This: Inorg. Chem. 2021, 60, 17450\u221217461 Read Online\n\n\n ACCESS Metrics & More Article Recommendations *\n s\u0131 Supporting Information\nSee https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.\n\n\n\n\n ABSTRACT: Half-sandwich Os-arene complexes exhibit promising\n anticancer activity, but their photochemistry has hardly been\n Downloaded via MOSCOW STATE UNIV on May 12, 2026 at 13:38:45 (UTC).\n\n\n\n\n explored. To exploit the photocytotoxicity and photochemistry of\n Os-arenes, O,O-chelated complexes [Os(\u03b76-p-cymene)(Curc)Cl]\n (OsCUR-1, Curc = curcumin) and [Os(\u03b76-biphenyl)(Curc)Cl]\n (OsCUR-2), and N,N-chelated complexes [Os(\u03b76-biphenyl)(dpq)-\n I]PF6 (OsDPQ-2, dpq = pyrazino[2,3-f ][1,10]phenanthroline) and\n [Os(\u03b76-biphenyl)(bpy)I]PF6 (OsBPY-2, bpy = 2,2\u2032-bipyridine), have\n been investigated. The Os-arene curcumin complexes showed\n remarkable photocytotoxicity toward a range of cancer cell lines\n (blue light IC50: 2.6\u22125.8 \u03bcM, photocytotoxicity index PI = 23\u221234),\n especially toward cisplatin-resistant cancer cells, but were nontoxic to normal cells. They localized mainly in mitochondria in the\n dark but translocated to the nucleus upon photoirradiation, generating DNA and mitochondrial damage, which might contribute\n toward overcoming cisplatin resistance. Mitochondrial damage, apoptosis, ROS generation, DNA damage, angiogenesis inhibition,\n and colony formation were observed when A549 lung cancer cells were treated with OsCUR-2. The photochemistry of these Os-\n arene complexes was investigated by a combination of NMR, HPLC-MS, high energy resolution \ufb02uorescence detected (HERFD), X-\n ray adsorption near edge structure (XANES) spectroscopy, total \ufb02uorescence yield (TFY) XANES spectra, and theoretical\n computation. Selective photodissociation of the arene ligand and oxidation of Os(II) to Os(III) occurred under blue light or UVA\n excitation. This new approach to the design of novel Os-arene complexes as phototherapeutic agents suggests that the novel\n curcumin complex OsCUR-2, in particular, is a potential candidate for further development as a photosensitizer for anticancer\n photoactivated chemotherapy (PACT).\n\n\n \u25a0 INTRODUCTION\n Encouraged by the \u201cfrom bench to clinic\u201d story of platinum\n important biomolecules such as nucleobases, and (ii) photo-\n redox reactions at the Os center to induce oxidative damage to\n anticancer drugs, the study of other transition metals as cancer cells. Such Os-arenes might then have potential in\n anticancer agents has become a rapidly expanding \ufb01eld.1\u22123 For photoactivated chemotherapy (PACT) or photodynamic\n example, complexes of Group 8 metals iron, ruthenium, and therapy (PDT).8,9\n osmium have shown promising anticancer activity in vitro and Osmium(II) arene complexes with N,N-chelating ligands,\n in vivo.4\u22126 Photoactivatable complexes are attractive because of such as [Os(arene)(N\u2212N)I]+, can exhibit potent anticancer\n their potential ability to minimize e\ufb00ects on normal tissue activity in vitro and in vivo,10\u221220 and O,O-chelated curcumin\n through the use of light directed to the tumor. These complexes also have good anticancer potency. To date, 12\n complexes only exert high antitumor activity under light metal-arene based curcumin complexes have been reported\n irradiation while being inactive in the dark, which can with moderate anticancer cytotoxicity (Table S1), but their\n minimize side e\ufb00ects on normal cells and tissues.6 Ru- and phototoxicity and anticancer mechanisms have not been\n Ir-based complexes have been well studied especially as explored.13\u221220 Curcumin (1,7-bis(4-hydroxy-3-methoxyphen-\n polypyridine-type complexes, some of which have high yl)-1,6-heptadiene-3,5-dione, CurcH) has attracted much\n anticancer activity. The photochemistry of Os-arene complexes\n and their potential for use in phototherapy have hardly been\n investigated. Received: January 25, 2021\n Very few Os-arene complexes are known to photodissociate Published: September 9, 2021\n the arene ligand in preference to their mono- or bidentate\n ligands.6,7 Based on this observation, two strategies for the\n design of Os-based photoactivated anticancer agents can be\n envisaged: (i) photosubstitution of Os-bound ligands by\n\n \u00a9 2021 American Chemical Society https://doi.org/10.1021/acs.inorgchem.1c00241\n 17450 Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 1. (A,B) Molecular structures of OsCUR-1, OsCUR-2, RuCUR-2, OsDPQ-2, and OsBPY-2. (C) X-ray crystal structure of OsCUR-1,\nwhere the hydrogen atoms and PF6\u2212 are omitted for clarity. (D) IC50 values (\u03bcM) of complexes OsCUR-2 and cisplatin (cis-Pt) toward various cell\nlines for 48 h treatment.\n\nattention for therapy, due to its multifunctionalities in bound arene ligand and oxidation of the Os(II) center were\ninhibiting carcinogenesis and limiting tumor growth.19,20 investigated for excitation by UVA and blue light.\nMoreover, curcumin is a photosensitizer which absorbs visible\nlight (400\u2212500 nm), and itself is a candidate photodynamic\ntherapy (PDT) agent.19 Despite exhibiting interesting\n \u25a0 RESULTS\n Synthesis. All Os complexes were prepared by a general\nanticancer activity, curcumin (CurcH) still has limited utility one-step reaction of the Os-arene dimers [Os(arene)X2]2\nbecause of its low bioavailability and hydrolytic instability (arene = biphenyl or p-cymene, X = Cl or I) with a chelating\nunder physiological conditions.19,20 As a result, metal curcumin ligand (details in Supporting Information, Experimental\ncomplexes have been explored as a means of avoiding the Section, Figure 1, Scheme S1). The Ru complex RuCUR-2\nhydrolysis of the diketone through formation of coordination was synthesized for comparison. The purity of all complexes\nbonds, including complexes of Pt(II), Zn(II), Cu(II), and was determined to be \u226595% by NMR and elemental analysis.\nGa(III).19\u221223 Single-crystal di\ufb00raction of OsCUR-1 revealed the classical\n Herein, we report the synthesis and characterization of O,O- piano-stool structure with the O,O-donor curcumin chelated to\nchelated Os(II) complexes [Os(\u03b76-p-cymene)(Curc)Cl] Os in the {Os(\u03b76-p-cymene)Cl} unit, and the monomers were\n(OsCUR-1) and [Os(\u03b76-biphenyl)(Curc)Cl] (OsCUR-2) further connected through hydrogen bonds and \u03c0\u2212\u03c0\nand for comparison Ru(II) complex [Ru(\u03b76-biphenyl)(Curc)- interactions (Tables S2\u2212S4, Figures 1, S1, and S2). The\nCl] (RuCUR-2) and N,N-chelated [Os(\u03b76-biphenyl)(dpq)I]- structure of OsCUR-1 is very similar to that of [Ru(\u03b76-p-\nPF6 (OsDPQ-2, dpq = pyrazino[2,3-f ][1,10]phenanthroline) cymene)(Curc)Cl]15 and bis((1,7-bis(3,4-dimethoxyphenyl)-\nand [Os(\u03b76-biphenyl)(bpy)I]PF6 (OsBPY-2, bpy = 2,2\u2032- hept-1,6-diene-3,5-dione)-(\u03b7 6-p-cymene)chlororuthenium-\n (II),16 with bond lengths in similar ranges. An isostructural\nbipyridine). The structure of OsCUR-1 was determined by\n complex [Os(\u03b76-p-cymene)(Curc)Cl] was reported recently by\nsingle-crystal X-ray di\ufb00raction. The photocytotoxicity and\n Dyson et al., crystallized in the orthorhombic space group\nanticancer mechanism of these Os-arene complexes were Pbca, in contrast to the monoclinic C2/c space group observed\nstudied by a range of chemical and physical techniques and cell here.17\nbiological assays, including the MTT colorimetric cytotoxicity Cytotoxicity and Photocytotoxicity. The in vitro\nassays, confocal microscopy, ICP-MS analysis, \ufb02ow cytometry, anticancer activity of complexes OsDPQ-2 and OsBPY-2\nWestern blotting, comet assays, colony formation, and was investigated. OsDPQ-2 showed good anticancer activity in\nHUVECs tube formation assays. The photochemistry of the the dark, with an IC50 value of 2 \u03bcM, similar to cisplatin under\ncomplexes was investigated by various techniques including the same experimental conditions (Chart S1). However,\nhigh-energy resolution \ufb02uorescence detected (HERFD) X-ray OsBPY-2 is inactive, IC50 > 50 \u03bcM. The distribution of Os\nadsorption near-edge structure (XANES)24 and total \ufb02uo- in A2780 human ovarian cancer cells was studied after\nrescence yield (TFY) XANES spectroscopy,25 UV\u2212vis, incubation with 4 \u03bcM OsDPQ-2. The Os content in four\n\ufb02uorescence, and NMR spectroscopy, HPLC-MS, and fractions, cytosol, membrane plus particulate fraction, nucleus,\ntheoretical computation. Selective photodissociation of the \u03c0- and cytoskeleton, was determined by ICP-MS. Most of the Os\n 17451 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 2. (A) Intracellular Os and Ru distribution in A549 cells incubated with 10 \u03bcM OsCUR-1, OsCUR-2, or RuCUR-2 for 12 h in the dark or\nlight (photoirradiation for 15 min and then dark incubation for 11 h 45 min) determined by ICP-MS (ppb/106 cells). (B,C) Confocal \ufb02uorescence\nmicroscopy images of A549 cells treated with complexes OsCUR-1, OsCUR-2, and RuCUR-2, before (B) and after photoirradiation (C), showing\nthat the complexes are accumulated in the mitochondria and nucleus, respectively. The excitation wavelength was 460 nm, with emission at 500\u2212\n600 nm.\n\n(80%) was in the membrane plus particulate fraction, and 10% Subcellular Localization in the Dark or upon\nwas in the nucleus (Figure S3). To examine whether OsDPQ- Irradiation. The subcellular localization of OsCUR-1,\n2 in the membranes a\ufb00ected the cell cycle progression of OsCUR-2, RuCUR-2, and curcumin in A549 cells was\nA2780 cells, a further cell cycle analysis was performed after investigated by confocal microscopy (Figures 2 and S5)\nincubation with varying concentrations of OsDPQ-2 for 24 h without photoirradiation. Green luminescence of OsCUR-1,\n(1, 5, and 20 \u03bcM). These experiments revealed that S-phase OsCUR-2, RuCUR-2, or curcumin was observed within 1 h,\ncell cycle arrest was induced by OsDPQ-2 (Figure S4). suggesting signi\ufb01cant cellular uptake of OsCUR-1, OsCUR-2,\n To test the potential for using Os arene compounds as RuCUR-2, and curcumin in the cytoplasm (Figure S5). High\nphototherapeutic agents, OsCUR-1 and OsCUR-2 (controls: Pearson\u2019s colocalization coe\ufb03cients were obtained by confocal\nRuCUR-2, curcumin, and cisplatin) were assessed against a microscopy for OsCUR-1, OsCUR-2, RuCUR-2, and\npanel of human cell lines: lung carcinoma A549, cisplatin- curcumin with the mitochondria-speci\ufb01c probe MitoTracker\nresistant A549R, breast cancer MCF-7, cervical cancer HeLa, Deep Red (MTDR), implying that nearly 90% of these\nliver cancer HepG2, normal lung HLF, and liver LO2 cell lines, complexes were observed to be in the mitochondria (Figure\nas shown in Figure 1D and Table S27. Signi\ufb01cant phototoxicity 2B). The ICP-MS data also suggested that in the dark most of\nupon irradiation was observed for these osmium compounds, the Os-arene or Ru-arene curcumin complexes were located in\nwith higher potency toward cisplatin-resistant A549R cancer the mitochondria and cytoplasm (Figure 2A).\ncells and high selectivity for cancer cells over normal cells. The Redistribution of the metal-arene complexes after photo-\nbiphenyl Os(II) complex OsCUR-2 exhibited the highest irradiation was also revealed by confocal microscopy and ICP-\nphototoxicity, the highest PI values, and selectivity toward MS. In the confocal images of pretreated A549 cells after\nboth the cisplatin-resistant A549R cells and A549 cells. For photoirradiation and incubation for another 12 h, the intense\nexample, OsCUR-2 is relatively noncytotoxic in the dark (IC50: green luminescence of the metal-arene complexes emerged in\n73\u2212135 \u03bcM) toward all tested cells. Upon irradiation, the A549 cells and exhibited a partially overlapping pro\ufb01le with the\nphototoxicity IC50 decreased to 5.8 and 2.6 \u03bcM with PI values \ufb02uorescence of nucleus dye 2\u2032-(4-ethoxyphenyl)-5-(4-methyl-\nof 23.3 and 34.4 against A549 and A549R cancer cells, 1-piperazinyl)-2,5\u2032-bi-1H-benzimidazole trihydrochloride\nrespectively, while the IC50 value was 44.8 \u03bcM for normal HLF (Hoechst 33342, Figure 2C). Since the metal centers remained\ncells (relatively nontoxic). The selectivity factor for A549R coordinated to curcumin after the irradiation, the observation\ncells was up to 17.2-fold compared with normal HLF cells of luminescence in the nucleus suggested more Os/Ru-Cur\nupon irradiation. Curcumin itself displayed moderate cytotox- fragments had accumulated in the nucleus after the irradiation.\nicity against A549 cancer cells and HLF normal cells in the The ICP-MS data also indicated that the content of Os/Ru\ndark, showing no selectivity for lung cancer cells over normal dramatically increased in the nucleus, accompanied by a\nlung cells. As expected, cisplatin exhibited little photo- decrease in the mitochondria after the irradiation (Figure 2A).\ncytotoxicity toward the tested cells under these conditions. For example, the Os content of OsCUR-2 in A549 cells\n 17452 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 3. Characterization of apoptosis for A549 cancer cells treated with OsCUR-2 under photoirradiation (425 nm LED light for 15 min, 36 J\ncm\u22122): (A) morphological observations with Hoechst 33342 staining. (B) MMP (\u0394\u03a8m) analysis stained with Rh-123. (C) Observation of ROS\ngeneration with the probe DCFH-DA by confocal microscopy during OsCUR-2-mediated photocytotoxicity (425 nm LED light for 15 min, 36 J\ncm\u22122). (D) Western blotting analysis of the indicated proteins from A549 cells in the dark and after treatment with various concentrations of\nOsCUR-2 and photoirradiation. (E) Concentration-dependent inhibition of z-VAD-fmk on cell death induced by OsCUR-2-mediated\nphotocytotoxicity toward A549 human lung cancer cells. Data are represented as means \u00b1 SD of three independent experiments. **, P < 0.01, as\ncompared with the group treated with OsCUR-2 (10 \u03bcM) in the absence of z-VAD-fmk. Scale bar = 40 \u03bcm.\n\ndropped from 32.7 to 18.7 ppb/106 cells in the mitochondria Caspase-3 plays a decisive role in the execution of apoptosis\nbut increased from 5.2 to 9.5 ppb/106 cells in the nucleus after and is responsible for the cleavage of PARP during cell death.29\nthe irradiation, in agreement with the confocal images. As compared with control cells in the dark, dose-independent\n Apoptosis. Upon irradiation, OsCUR-2 exhibited the increases in cleaved-caspase-3 and cleaved-PARP (clPARP)\nhighest photocytotoxicity toward A549 human lung cancer activities were detected in A549 cells treated with OsCUR-2\ncells and cisplatin-resistant A549R lung cancer cells; thus, upon light irradiation (Figure 3D). Moreover, z-VAD-fmk, a\nA549 cells were chosen for exploration of the mechanism of pan-caspase inhibitor, e\ufb03ciently attenuated cell death caused\nthe phototoxicity. Apoptosis is one of the most common cell by photoactivated OsCUR-2 (Figure 3E). In the presence of z-\ndeath pathways, accompanied by cell shrinkage, nuclear VAD-fmk (100 \u03bcM), the cell viability increased from 40.7% to\nfragmentation, membrane blebbing, and apoptotic body 64.7% when cells were treated with OsCUR-2 (10 \u03bcM) and\nformation.26 To verify whether OsCUR-2 induced apoptosis irradiated by blue light, 425 nm. These data suggest that\nin A549 cells upon irradiation, the morphological changes in photoactivated OsCUR-2 kills A549 cells mainly via apoptosis,\nA549 cells were \ufb01rst examined by staining with Hoechst through caspase-dependent mechanisms.\n33342.27 As shown in Figure 3A and Figure S14A, upon ROS Generation. Generation of reactive oxygen species\nirradiation with OsCUR-2, apoptosis-induced morphological (ROS) is the main mechanism responsible for photosensitizer-\nfeatures were observed with an increasing dependence on induced cell death.30 The ability of OsCUR-2 to generate ROS\nconcentration, including plasma membrane blebbing, frag- within A549 cells was investigated using the ROS probe, 2\u2032,7\u2032-\nmented nuclei, and apoptotic bodies. In contrast, the control dichloro\ufb02uorescein diacetate (DCF-DA).31 Upon light irradi-\ncells (both in the dark with or without OsCUR-2 and in the ation, the intensity of green intracellular \ufb02uorescence from\nlight without OsCUR-2) displayed normal morphology with OsCUR-2 in A549 cells increased signi\ufb01cantly in a\nround and homogeneous nuclei. Apoptosis also results in the concentration-dependent manner (Figure 3C), indicating the\ndepolarization of the mitochondria and a decrease in the generation of ROS. Flow cytometric analysis further con\ufb01rmed\nmitochondrial membrane potential (MMP, \u0394\u03a8m) of the that a 9-fold increase in the ROS probe signal intensity was\ncells.28 Rhodamine 123 (Rh-123) was used to investigate the observed in A549 cells treated with OsCUR-2 (25 \u03bcM) under\nMMP by confocal microscopy, since it is readily washed out light irradiation as compared with the control cells treated with\nonce the MMP is lost. Green \ufb02uorescence of Rh-123 was OsCUR-2 in the dark alone (Figure S6).\ndetected in A549 cells after incubation with OsCUR-2 upon DNA Damage. The expression level of \u03b3-H2AX, an\nirradiation (Figure 3B and Figure S14B), suggesting that the established molecular marker of DNA damage, and one of\ncancer cells maintained their viability. The \u0394\u03a8m showed a the major and early cellular responses to the induction of\nmarked decreasing trend as revealed by the \ufb02uorescence nuclear foci, was determined for A549 cells after treatment\nintensity with increasing concentrations of OsCUR-2 upon with OsCUR-2 upon irradiation and analyzed by immuno-\nirradiation. \ufb02uorescence analysis and Western blotting.32 \u03b3-H2AX\n 17453 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 4. (A) Cell immunostaining assay, (B) statistical quanti\ufb01cation, and (C) Western blotting of the expression level of \u03b3-H2AX of A549 cells\nafter the treatment of OsCUR-2 under irradiation. *, P < 0.05, **, P < 0.01, ***, P < 0.001. Scale bar = 40 \u03bcm. (D) and (E) \ufb02uorescence images\nand statistical quanti\ufb01cation obtained from an alkaline comet assay for A549 cells on preincubation for 4 h with complex OsCUR-2 with or without\nirradiation. Scale bar = 10 \u03bcm. (F) Western blotting of the expression level of \u03b3-H2AX of cisplatin-resistant A549R cells after the treatment of\nOsCUR-2, cis-Pt, or CurcH under irradiation. The irradiation was 425 nm LED light for 15 min, 36 J cm\u22122. (G\u2212K) In vitro antiangiogenic activity\nof listed complexes (5 \u03bcM) and the control on HUVECs. (L,M) Images of colony formation and statistic quanti\ufb01cation of A549 cells after\ntreatment with OsCUR-2 with or without photoirradiation (425 nm LED light for 15 min, 36 J cm\u22122). ns, not signi\ufb01cant. **, P < 0.01. ***, P <\n0.001.\n\n 17454 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 5. 1H NMR spectra of RuCUR-2 (A\u2212D) and OsCUR-2 (G\u2212J) (100 mM, DMSO-d6) before and after photoirradiation (425 nm, 100 mW)\ntogether with 1H NMR spectra of biphenyl (E) and curcumin (F). (A and G) initial complexes, (B,C and H,I) after irradiation for speci\ufb01c times,\nand (D or J) following a rest in the dark for another 180 min for RuCUR-2 or 1390 min for OsCUR-2. Triangles, circles, and squares denote the\nsignals of released biphenyl.\n\nexpression was signi\ufb01cantly elevated in a dose-independent formation and might have the potential to suppress tumor\nmanner in A549 cells after the treatment (Figure 4A\u2212C), metastasis.\nsuggesting the photoactivity of OsCUR-2 enhances the Optical Properties. Both OsCUR-1 and OsCUR-2 show\nfrequency of DNA double-strand breaks (DSBs). The intense absorption at 425 nm, while RuCUR-2 has an intense\nexpression levels of \u03b3-H2AX in cisplatin-resistant A549R cells band at 408 nm with a shoulder at 445 nm (Figure S7).\nafter the treatment with OsCUR-2, cis-Pt, or CurcH under Furthermore, both RuCUR-2 and OsCUR-2 exhibit maximum\nirradiation are shown in Figure 4F. Signi\ufb01cantly elevated \u03b3- emissions at 545 or 560 nm, close to that of curcumin itself\nH2AX expression was observed only for OsCUR-2, while (548 nm) with excitation in the range of 450\u2212461 nm. The\nthose for cisplatin and CurcH remained unchanged. This maximum emission of OsCUR-1 is signi\ufb01cantly blue-shifted to\nindicated that e\ufb00ective DNA damage in cisplatin-resistant cells 507 nm with excitation at 461 nm (Figure S8 and Table S5).\nwas caused by OsCUR-2, which may contribute to overcoming The dramatic di\ufb00erence between OsCUR-1 and OsCUR-2 can\ncisplatin resistance. be ascribed to the di\ufb00erent arene groups, p-cymene vs\n A single cell gel electrophoresis assay (SCGC, comet assay), biphenyl. The UV\u2212vis spectra of OsDPQ-2 and OsBPY-2\ncommonly used to detect cellular DNA damage as single- are similar (Figure S9). OsDPQ-2 has absorption maxima at\nstrand breaks, was also used.33 As seen in Figure 4D,E, the 280 nm, 300 nm (shoulder), 420 nm, and 500 nm (shoulder).\nlength of the comet dramatically increased compared to the After irradiation of OsDPQ-2 and OsBPY-2 by UVA (365\ncontrol samples (\u223c2.5-fold longer), and di\ufb00erently sized and nm), UV\u2212vis spectra were recorded at various time intervals\nfragmented dead cells were observed after treatment with over the course of 4 h. The absorption intensity at 280 and 420\nOsCUR-2 upon irradiation. In contrast, the control samples nm decreased, while that at 500 nm increased (Figure S9).\nexhibited almost condensed nuclei. The comet length for Biphenyl Photodissociation Studied by 1H NMR and\nOsCUR-2-treated A549 cells in the dark elongated only LC-MS. Time-dependent 1H NMR was used to monitor the\nslightly, which indicated much less DNA damage in the dark photoirradiation reactions of OsCUR-2, RuCUR-2, OsDPQ-\nthan that after photoirradiation. The results of the comet assay 2, and OsBPY-2 with blue light (425 nm) or UVA (365 nm)\nillustrate that the DNA double helix has been at least partially at ambient temperature (Figures 5 and S12). After irradiation\ndenatured and that some nuclear DNA becomes single for 50 or 45 min, new peaks at 7.66, 7.47 and 7.37 ppm were\nstranded upon photoirradiation with OsCUR-2 treatment. observed for OsCUR-2 or RuCUR-2, which can be assigned to\n Angiogenesis and Colony Formation. The e\ufb00ects of the free biphenyl ligand. The intensities of these peaks\nOsCUR-1, OsCUR-2, and RuCUR-2 on angiogenesis were increased when the samples were kept in the dark after\nstudied by using human umbilical vein endothelial cells irradiation. These results implied that the biphenyl arene was\n(HUVECs) (Figure 4G\u2212K).34 These complexes, as well as released after photoirradiation of OsCUR-2 and RuCUR-2.\ncurcumin itself, exhibited antiangiogenic activity. In addition, However, it was notable that no new set of signals for other\nthe colony formation assay con\ufb01rmed the antiproliferative photoproducts from OsCUR-2 and RuCUR-2 such as metal-\nactivity, invasiveness, sensitivity, and long-term toxic e\ufb00ect of curcumin fragments was observed.\nOsCUR-2 upon irradiation with A549 cells (Figure 4L,M). Similar results were observed for 1H NMR experiments\nColony formation by A549 cells was signi\ufb01cantly inhibited when OsDPQ-2 and OsBPY-2 were photoirradiated by either\nafter a two-week incubation with OsCUR-2 upon irradiation in blue light (425 nm) or UVA (365 nm). However, 1H NMR\na dose-independent manner. There was less than 1% of survival signals disappeared after irradiation for 1 or 2 h. Then, the\nof A549 cells after OsCUR-2 treatment and photoirradiation at biphenyl photodissociation reaction was further studied by LC-\na concentration of 24 \u03bcM. In contrast, more than 95% of A549 MS experiments. Only one LC peak was observed at 17.2 min\ncancer cells survived after treatment without OsCUR-2 both in with m/z of 705.0276 before irradiation, attributable to\nthe dark and after photoirradiation. These results suggest that [Os(\u03b76-bip)(dpq)I]+, but two new LC peaks appeared at\nOsCUR-2 can e\ufb03ciently inhibit angiogenesis and colony 18.6 and 19.5 min with m/z of 582.9697 and 552.9572 with\n 17455 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n\n\n\nFigure 6. DFT optimized structures of (A) OsBPY-2 before light irradiation and (B) its putative photoproduct [Os(bpy)(acetone)3I]2+. (C)\nExperimental and calculated X-ray absorption spectra of the OsBPY-2 in acetone solution before (black line) and after (blue line) irradiation with\nblue light. Shift along the vertical axis is for the sake of clarity.\n\ncharacteristic Os isotopic peaks after irradiation for 4 h, To analyze the spectra (see SI Section Experimental\nassignable to biphenyl-depleted OsDPQ-2 ([Os(dpq)I + 2H]+ Procedures for details), we modeled the structure of a putative\nand [Os(dpq)I+ + MeOH]), respectively, further con\ufb01rming photoproduct taking into account the loss of the biphenyl\nthe loss of biphenyl from OsDPQ-2 after the irradiation ligand and the formation of an octahedral Os center\n(Figure S13). coordinated to three solvent molecules (acetone, as [Os(bpy)-\n It was remarkable that during the photoreaction, 1H NMR (acetone)3I]2+), Figure 6B. The good agreement between the\npeaks corresponding to the initial OsBPY-2 disappeared over calculated and experimental XANES spectra overall con\ufb01rms\n\u223c2 h, and the only peaks visible in the spectrum were for free the generation of an Os(III) center upon light irradiation of\nbiphenyl (Figure S12). The photodecomposition reaction of OsBPY-2.\nOsBPY-2 was further investigated by X-ray absorption\nspectroscopy, a powerful technique to investigate local\ngeometry and electronic structure of metal ions in organo-\n \u25a0 DISCUSSION\n We have studied the photochemistry, phototoxicity, and\nmetallic compounds.35 We acquired the Os L3-edge HERFD- anticancer mechanism of action of half-sandwich Os(II)\n(High Energy Resolution Fluorescence Detected) and TFY arene complexes containing the O,O-chelating and N,N-\n(Total Fluorescence Yield) XANES (X-ray Absorption Near chelating ligands, together with chloride or iodide as a\nEdge Structure) spectra of initial and light-irradiated OsBPY-2 monodentate ligand. Very few studies of photoactivatable\nat the ID26 beamline of the ESRF synchrotron, following osmium complexes have been reported so far. Brewer et al.\npreviously described procedures.36 Experimental Os L3-edge reported the inhibition of growth of African green monkey\nHERFD and TFY XANES of the OsBPY-2 and light-irradiated kidney epithelial (Vero) cells by an osmium bipyridine\nOsBPY-2 together with the theoretical spectrum are shown in complex [{(bpy)2Os(dpp)}2RhCl2]Cl5 [bpy = 2,2\u2032-bipyridine;\nFigure 6. The resolution gain of the HERFD spectrum dpp = 2,3-bis(2-pyridyl)pyrazine].6 The current work appears\ncompared to the TFY is clearly visible. Calculated XANES to be the \ufb01rst study of the photochemistry and photobiology of\nspectra were in good agreement with the experiment, half-sandwich Os(II) complexes.\ncon\ufb01rming the reliability of the structural model obtained. Phototoxicity, Cellular Distribution, and Anticancer\nThe changes occurring in the experimental XANES spectra Mechanism. The Os/Ru-arene curcumin complexes OsCUR-\nupon irradiation of the sample with blue light are qualitatively 1, OsCUR-2, and RuCUR-2 exhibited low dark toxicity\nreproduced by the theory in the pre-edge (a), white line (b), toward both cancer and normal cell lines, but showed potent\nand postedge (c) spectral regions. Notably, changes in the phototoxicity toward cancer cell lines after irradiation,\nconventional TFY XANES are much smaller, which would especially toward the cisplatin-resistant A549R human lung\nhave made the analysis much more di\ufb03cult without the high- cancer cells. This high phototoxicity suggests that Os-arene\nresolution data. curcumin complexes might be promising candidates as\n 17456 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\nphotochemotherapeutic agents and overcome cisplatin resist- estimation of the oscillator strength relative to the lowest\nance.37,38 The phototoxicity of the biphenyl complex OsCUR- energy transition at 467 nm. This transition corresponds to the\n2 is the highest among the three metal-arene curcumin HOMO \u2192 LUMO transition.\ncomplexes, perhaps re\ufb02ecting the higher inertness of Os(II) Photoactivated Dissociation of the Arene Group and\ncompared to Ru(II).39 OsCUR-1, OsCUR-2, and RuCUR-2 Oxidation of Os(II) upon Photoirradiation. Taube and co-\nlocalized mainly in the mitochondria of A549 cells without workers reported that the Os-benzene dimer [Os(\u03b76-C6H6)-\nirradiation; the same cellular localization was also observed for Cl2]2 decomposes upon UVA (365 nm) irradiation, free\npreviously reported curcumin metal complexes, such as V(IV) benzene is released as a reaction product,7 and arene\nand Nb(III),19\u221221,28 indicating that after coordination with photosubstitution reactions for Os(II) arene 1,4,7-trimethyl-\nmetal centers, curcumin retains its ability to target mitochon- 1,4,7-triazacyclononane and 1,4,7-triazacyclononane complexes\ndria.22 The apoptosis-induced morphological features, a can be achieved after prolonged (>24 h) photolysis using UV\ndecrease in the mitochondrial membrane potential (MMP, irradiation (mercury arc lamp).7 In our case, the photo-\n\u0394\u03a8m) together with increases in cleaved-caspase-3 and activated dissociation of the arene ligand was observed by 1H\ncleaved-PARP (clPARP) activities, suggested that these NMR, LC-MS, HERFD, and TFY XANES techniques when\nphotoactivated Os-arene complexes kill cancer cells via these half-sandwich Os(II) biphenyl complexes were irradiated\napoptosis through caspase-dependent mechanisms (Figure with UVA or blue light (Figures 6, S12, and S13) and also\n3). A 9-fold increase in the ROS signal intensity, antiangio- con\ufb01rmed by DFT calculations. The released free biphenyl\ngenic activity, and signi\ufb01cant inhibition of colony formation might further interact with DNA by intercalation and result in\nwere also observed, indicating a multitargeting anticancer DNA damage. RuCUR-2 exhibited a higher extent of biphenyl\nmechanism of action. dissociation compared to OsCUR-2 under similar irradiation\n Activity in Cisplatin-Resistant Cells. Mitochondria were conditions.4\ndamaged in A549R cells after photoirradiation and Os partially On the other hand, the photodissociation may also result\nmigrated to the nuclei (Figure 2). Both mitochondrial damage from the low-lying dd excited state, as all absorbing transitions\nand translocation of Os from the mitochondria to nuclei after include at least minor d-to-d character (Tables S9, S14, and\nphotoirradiation could lead to the DNA damage. A S19). This is also supported by the fact that the dd component\nsigni\ufb01cantly elevated expression level of \u03b3-H2AX upon in RuCUR-2 transitions is stronger than in OsCUR-1 and\nirradiation of A549R cells treated with OsCUR-2 was OsCUR-2 transitions, explaining a higher activity with\nobserved, while for cisplatin or curcumin, the level remained RuCUR-2.44\nunchanged, indicative of DNA damage in A549R cells caused The photoirradiation reactions of Os/Ru-arene curcumin\nby OsCUR-2. Furthermore, the comet assay illustrated that the complexes are di\ufb00erent from that of the Pt(II)-curcumin\nDNA double helix in A549R cells was partially denatured, and complex [Pt(Curc)(NH3)2](NO3), in which the curcumin is\nthe nuclear DNA became single stranded upon photo- released from the coordination complex.22 The quantum yields\nirradiation with OsCUR-2 (Figure 4). These results indicated for the photodissociation of biphenyl ligand from RuCUR-2\nthat irradiated OsCUR-2 overcomes cisplatin resistance via and OsCUR-2 were also determined, based on 1H NMR data,\nmitochondrial and DNA damage. which indicated that \u223c50% and \u223c9% biphenyl ligands had\n Computational Studies of Photophysical Properties. been released under the similar irradiation conditions from\nDFT calculations were carried out for the Os-arenes (Figure RuCUR-2 and OsCUR-2, respectively. This result suggested\nS11, Tables S6\u2212S23).40 Singlet state TDDFT calculations that interaction between Ru(II) and the biphenyl was weaker\nwere run in aqueous solutions (cpcm method) to assign the than that for Os(II).\nabsorption bands in the electronic spectrum.41,42 The The release of the arene ligand from OsDPQ-2 can be\ncomputed electronic transitions for OsCUR-1, OsCUR-2, induced by photoirradiation with either blue light and UVA,\nand RuCUR-2 indicate maxima at 444, 445, and 456 nm and the iodido osmium complexes with bidentate ligands\n(Tables S7, S12, and S17), respectively, consistent with reported in this work appear to react much faster under\nexperimental results. The orbital energies for the HOMO irradiation. LC-MS data con\ufb01rmed the release of the biphenyl\nand LUMO are listed in Table S6. The UV absorption for ligand from OsDPQ-2 after irradiation since two biphenyl-\nOsCUR-1 is mainly due to an S0 \u2192 S3 electronic transition, depleted species were detected. OsDPQ-2 and OsBPY-2 were\nand analysis of the Natural Transition Orbitals (NTOs) clearly studied to elucidate their photodecomposition pathways using\nshows the MLCT character in this transition (Table S9). The DFT calculations (Tables S23 and S24). Geometry optimiza-\nphotoluminescence emission for OsCUR-1 is due to the tions of OsBPY-2 in the ground state (S0) and lowest-lying\nintraligand charge transfer (ILCT) according to the NTO triplet state (T1) were performed in the gas phase, and the\nanalysis43a (Tables S8\u2212S11). Similarly, the UV\u2212vis absorp- nature of all stationary points was con\ufb01rmed by normal-mode\ntions of OsCUR-2 and RuCUR-2 are ascribed to the MLCT analysis. The X-ray coordination bond lengths and angles\nS0 \u2192 S1/S2 electronic transitions for OsCUR-2 and S0 \u2192 S2 around the Os center are listed in Tables S24 and S25.41,42 The\nelectronic transition for RuCUR-2 (Tables S12\u2212S19), with most dramatic di\ufb00erence is the weakened coordination\nabsorption maxima at 445 and 483 nm (shoulder), respectively between the Os center and the biphenyl arene group, where\n(Figure S7). For OsDPQ-2 (Tables S20\u2212S23), the most the Os\u2212centroid bond length has elongated by 0.275 \u00c5 (from\nintense band at 280 nm is due to mixed states of 1MLCT and 1.767 to 2.042 \u00c5) after the light irradiation, suggesting the\n1\n \u03c0\u2212\u03c0 character according to the calculation,43b and the dissociative character of the biphenyl arene group in OsDPQ-\nshoulder in the 300\u2212350 nm region can be ascribed to 2. Several of the lowest energy transitions have contributions\n1\n MLCT (Os, I \u2192 dpq) transitions; the lowest-energy bands from the LUMO+3 and LUMO+4 orbitals, which have \u03c3*\nare again of 1MLCT type, involving di\ufb00erent Os d orbitals. The antibonding character toward the Os\u2212I bond and partially also\nagreement between the experimental spectra and the toward the Os\u2212arene (Table S26). Such transitions therefore\ncomputed transitions is satisfactory, despite the under- have partial dissociative character. The lowest-lying triplet and\n 17457 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\ntwo other triplet states of higher energy have dissociative Os(II)-arene complexes. OsCUR-2 particularly appears to be\ncharacter (Tables S15\u2212S18), which were 3MC45,46 with the a candidate for further development as a phototherapeutic\ncontributions from the same dissociative orbitals according to agent with the potential to overcome cisplatin resistance. This\nthe spin density surface (Table S19). The nature and energy of work provides a new strategy to generate novel anticancer\nthe low-energy triplets is consistent with the lack of drugs via release of arene ligands and oxidation of metal\n\ufb02uorescence from OsDPQ-2 as well as its photochemical centers upon light irradiation.\n\n \u25a0\nbehavior.43\n The photo-oxidation of Os(II) in OsDPQ-2 and OsBPY-2 EXPERIMENTAL SECTION\nwas \ufb01rst revealed by 1H NMR experiments. The 1H NMR\nsignals for OsDPQ-2 disappeared when OsDPQ-2 was Materials and Synthesis. The ligand curcumin was purchased\n from HEOWNS. The starting metal-arene dimeric materials ([(\u03b76-p-\nphotoactivated by blue light or UVA. Low-spin Os(II) (5d6) cymene)OsCl2]2, [(\u03b76-biphenyl)OsCl2]2, [(\u03b76-biphenyl)OsI2]2, and\nis diamagnetic, and its complexes give sharp NMR peaks. In [(\u03b76-biphenyl)RuCl2]2) were prepared according to previously\ncontrast, Os(III) is paramagnetic and expected to give very reported procedures.39 All other chemicals and reagents were\nbroad (even unobservable) peaks.47 commercially available and were used as-received. MTDR was\n The reactivity of OsBPY-2 was investigated by HERFD and purchased from Invitrogen, and DCFH-DA was purchased from\nTFY XANES, and the structure of a putative photoproduct via Sigma-Aldrich. Antibodies were purchased as follows: anticaspase-3\nloss of the biphenyl ligand and the formation of an octahedral (SC-7148; Santa Cruz Biotechnology, Inc.), anti-GAPDH (AP0063;\nOs center coordinating three solvent molecules (acetone) was Abgent, Suzhou, China), anti-\u03b3-H2AX (ab2893; Abcam, Cambridge,\nmodeled based on the experimental data (Figure 6). The MA, USA), and antivinculin antibody (MAB3574, Millipore).\nemerging XANES pre-edge feature probably corresponds to a [Os(p-cymene)(Curc)Cl] (OsCUR-1). This compound was\n obtained by a two-step reaction. The curcumin was dissolved in\n2p \u2192 t2g transition, where the t2g vacancy results from Os(II) methanol, and NaOMe was added. After 1 h stirring at room\noxidation. An analogous photoproduct [(CH3CN)3Os(tacn)]- temperature, [(\u03b76-p-cymene)OsCl2]2 was added. The resulting dark-\n(PF6)2 (where tacn = 1,4,7-triazacyclononane) has been red solution was stirred and re\ufb02uxed for 24 h. The solvent was\nisolated from the photolysis of [(\u03b76-p-cymene)Os(tacn)]- removed under reduced pressure, and the residual was redissolved in\n(PF6)2 in CH3CN.7 Experimental HERFD and TFY XANES of dichloromethane. Sodium chloride was \ufb01ltered from the mixture. An\nthe nonirradiated OsBPY-2 together with the theoretical orange-red precipitate was obtained after rotary evaporation, and this\nspectrum are shown in Figure 6. The good agreement between precipitate was puri\ufb01ed by TLC with dichloromethane/methanol as\nthe calculated and experimental XANES spectra overall the eluent; further puri\ufb01cation was performed by recrystallization\ncon\ufb01rmed the generation of an Os(III) center upon light from CHCl3. Reaction yield 68%. 1H NMR (400 MHz, DMSO-d6) \u03b4\nirradiation of OsBPY-2. The formation of the photoproduct is 9.56 (s, 2H), 7.41 (d, 2H), 7.24 (d, 2H), 7.06 (d, 2H), 6.81 (d, 2H),\n 6.56 (d, 2H), 6.14 (d, 2H), 5.91 (d, 2H), 5.68 (s, 1H), 3.83 (s, 6H),\ndue to the loss of an arene ligand and leads to an octahedral Os 2.80\u22122.59 (m, 1H), 2.22 (s, 3H), 1.28 (d, 6H). ESI-MS (in CH3CN):\ncomplex coordinated by solvent molecules, yet retaining the calcd for [(\u03b76-p-cym)Os(curcuminato)]+ m/z 691.82, found m/z\niodido and bpy ligands.\n\n\u25a0\n 693.42. Anal. Calcd for OsCUR-1 (C31H33O6ClOs, %): C, 51.20; H,\n 4.57. Found (%): C, 50.88; H, 4.58.\n CONCLUSIONS [Os(biphenyl)(Curc)Cl] (OsCUR-2). This compound was synthe-\nThere are few reports of the photochemistry of Os(II) arene sized by using a procedure similar to OsCUR-1 except replacing [(\u03b76-\n p-cymene)OsCl2]2 with [(\u03b76-biphenyl)OsCl2]2. Reaction yield 73%.\ncomplexes. The Os(II) arene curcumin complexes OsCUR-1 1\n H NMR (400 MHz, DMSO-d6) \u03b4 9.54 (s, 2H), 7.78 (d, 3.1 Hz, 2H),\nand OsCUR-2, studied here, exhibit high potent phototoxicity\n 7.49 (m, 3H), 7.15 (t, 4H), 7.00 (d, 2H), 6.80 (d, 2H), 6.66 (d, 2H),\nupon blue light irradiation, especially toward cisplatin-resistant 6.48 (d, 2H), 6.40 (t, 2H), 6.30 (t, 1H), 5.63 (s, 1H), 3.83 (s, 6H).\ncancer cell lines, and with high selectivity toward cancer cells. ESI-MS (in CH3CN): calcd for [(\u03b76-biphenyl)Os(curcuminato)]+ m/\nThese complexes were translocated from mitochondria to the z 711.81, found m/z 713.33. Anal. Calcd for OsCUR-2\nnucleus after irradiation, resulting in mitochondrial damage, (C33H29O6ClOs, %): C, 53.04; H, 3.91. Found (%): C, 52.16; H,\napoptosis, ROS generation, DNA damage, angiogenesis 4.21.\ninhibition, and colony formation. Even complex RuCUR-2 [Ru(biphenyl)(Curc)Cl] (RuCUR-2). A similar method was\nwith similar optical properties to those of OsCUR-1 and adopted for the synthesis of RuCUR-2 as for OsCUR-1 in which\nOsCUR-2 exhibited much lower cytotoxicity toward cancer precursor [(\u03b76-p-cymene)OsCl2]2 was replaced by [(\u03b76-biphenyl)-\ncells compared to the Os complexes, due to the stronger heavy RuCl2]2. Reaction yield 80%. 1H NMR (500 MHz, CDCl3) \u03b4 7.81 (d,\n 2H), 7.50 (m, 3H), 7.04\u22126.92 (m, 4H), 6.89 (d, 2H), 6.35 (d, 2H),\natom e\ufb00ect of Os compared to Ru. The photochemical\n 5.98 (t, 2H), 5.87 (d, 2H), 5.80 (d, 3H), 5.41 (s, 1H), 3.94 (s, 6H).\nreactions of these Os(II)-arene complexes were elucidated by ESI-MS (in CH3CN): calcd for [(\u03b76-biphenyl)Ru(curcuminato)]+ m/\nvarious techniques. This is the \ufb01rst report that at least two z 622.25, found m/z 623.25. Anal. Calcd for RuCUR-2\nreactions occur simultaneously during the photoreaction of an (C33H29O6ClRu(H2O)0.66(CH2Cl2)0.66, %): C, 55.64; H, 4.39.\nOs(II) arene complex, including the selective photodissocia- Found (%): C, 55.87; H, 4.73.\ntion of the biphenyl arene ligand and oxidation of Os(II) to [Os(biphenyl)(dpq)I]PF6 (OsDPQ-2). [(\u03b76-biphenyl)OsI2]2 was\nOs(III) induced by irradiation with blue light or UVA. The mixed with dpq and dissolved in ethanol. The solution was stirred\nreleased biphenyl ligand, the translocation of the Os-arene or under re\ufb02ux at 353 K for 2 h. The solution color changed from brown\narene-depleted species from mitochondria to the nucleus, to orange. The solution was hot-\ufb01ltered, and ammonium hexa\ufb02uor-\nmight be responsible for the DNA damage and inhibition of ophosphate (68.1 mg, 0.418 mmol) was added. The yellow-orange\n precipitate was collected by vacuum \ufb01ltration and washed with\ncolony formation, as well as contribute to overcoming cisplatin\n ethanol and diethyl ether. It was dried under vacuum overnight. Yield:\nresistance. Furthermore, mitochondrial damage, apoptosis, and 47.1 mg (65.9%). ESI-MS calcd for C26H18IN4Os m/z 705.2, found\nROS generation might arise from oxidation of Os(II) to m/z 704.9. 1H NMR (600 MHz, acetone-d6) \u03b4 9.74 (d, 2H, J = 6 Hz),\nOs(III) in the photoreaction. The present studies have 9.67 (d, 2H, J = 6 Hz), 9.38 (s, 2H), 8.15 (dd, 2H, J = 6 Hz), 7.55 (d,\ninvestigated the photocytotoxicity, anticancer mechanism of 2H, J = 6 Hz), 7.45 (t, 1H, J = 6 Hz), 7.33 (t, 2H, J = 6 Hz), 7.02 (d,\naction, and photochemical decomposition pathways for 2H, J = 6 Hz), 6.92 (t, 2H, J = 6 Hz), 6.71 (t, 1H, J = 6 Hz). Anal.\n\n 17458 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\nCalcd for OsDPQ-2 (C26H18F6IN4OsP, %): C, 36.80 H, 2.14, N, Authors\n6.60. Found (%): C, 36.55; H, 2.11; N, 6.39. Xuling Xue \u2212 Jiangsu Collaborative Innovation Center of\n [Os(biphenyl)(bpy)I]PF6 (OsBPY-2). [(\u03b76-biphenyl)OsI2]2 was Biomedical Functional Materials, School of Chemistry and\nmixed with bpy and dissolved in ethanol. The solution was stirred Materials Science, Nanjing Normal University, Nanjing\nunder re\ufb02ux at 353 K for 2 h. There solution color changed from dark\n 210023, China\nbrown to orange. The solution was hot-\ufb01ltered, and ammonium\nhexa\ufb02uorophosphate (137.6 mg, 0.844 mmol) was added. The orange Ying Fu \u2212 Department of Chemistry, University of Warwick,\nprecipitate was collected by vacuum \ufb01ltration and washed with Coventry CV4 7AL, U.K.; National Center for Advancing\nethanol and diethyl ether. It was dried under vacuum overnight. Yield: Translational Sciences (NCATS/NIH), Rockville, Maryland\n104.9 mg (80.5%). ESI-MS calcd for C22H18IN2Os: m/z 629.0, found 20850, United States\nm/z 629.0. 1H NMR (600 MHz, acetone-d6) \u03b4 9.21 (d, 2H, J = 6.0 Liang He \u2212 MOE Key Laboratory of Bioinorganic and\nHz), 8.65 (d, 2H, J = 6.0 Hz), 8.14 (t, 2H, J = 6.0 Hz), 7.53 (t, 2H, J = Synthetic Chemistry, School of Chemistry, Sun Yat-Sen\n6.0 Hz), 7.48 (d, 2H, J = 6.0 Hz), 7.40 (d, 1H, J = 6.0 Hz), 7.33 (t, University, Guangzhou 510275, China\n2H, J = 6.0 Hz), 6.86 (d, 2H, J = 6.0 Hz), 6.56 (t, 2H, J = 6.0 Hz), Luca Salassa \u2212 Donostia International Physics Center,\n6.50 (t, 1H, J = 6.0 Hz). Anal. Calcd for OsBPY-2 (C22H18F6IN2OsP, Donostia 20018, Spain; Ikerbasque, Basque Foundation for\n%): C, 34.21; H, 2.35; N, 3.63. Found (%): C, 34.45; H, 2.29; N, Science, Bilbao 48011, Spain; Kimika Fakultatea, Euskal\n3.61.\n Herriko Unibertsitatea, UPV/EHU, Donostia 20080, Spain\n\n\u25a0\n*\n ASSOCIATED CONTENT\ns\u0131 Supporting Information\n Ling-Feng He \u2212 Jiangsu Key Laboratory for Molecular and\n Medical Biotechnology, College of Life Science, Nanjing\n Normal University, Nanjing 210023, China\n Yuan-Yuan Hao \u2212 Jiangsu Collaborative Innovation Center of\nThe Supporting Information is available free of charge at\n Biomedical Functional Materials, School of Chemistry and\nhttps://pubs.acs.org/doi/10.1021/acs.inorgchem.1c00241.\n Materials Science, Nanjing Normal University, Nanjing\n Characterization of complexes OsCUR-1, OsCUR-2, 210023, China\n RuCUR-2, OsDPQ-2, and OsBPY-2 including 1H Madeleine J. Koh \u2212 Department of Chemistry, University of\n NMR spectra, ESI-MS, elemental analyses, single-crystal Warwick, Coventry CV4 7AL, U.K.\n X-ray structure analysis, and UV\u2212vis and \ufb02uorescence Cl\u00e9ment Souli\u00e9 \u2212 Institute of Chemical Sciences, School of\n spectroscopic properties; computational details; bio- Engineering & Physical Sciences, Heriot-Watt University,\n logical procedures such as ICP-MS measurement, MTT Edinburgh, Scotland EH14 4AS, U.K.\n detection, confocal microscopy imaging, and Western Russell J. Needham \u2212 Department of Chemistry, University of\n blotting experiments (PDF) Warwick, Coventry CV4 7AL, U.K.\n Abraha Habtemariam \u2212 Department of Chemistry, University\nAccession Codes of Warwick, Coventry CV4 7AL, U.K.\nCCDC 1571503 contains the supplementary crystallographic Claudio Garino \u2212 Department of Chemistry and NIS\ndata for this paper. These data can be obtained free of charge Interdepartmental Center, University of Turin, Turin I-\nvia www.ccdc.cam.ac.uk/data_request/cif, or by emailing 10135, Italy; orcid.org/0000-0002-7854-6076\ndata_request@ccdc.cam.ac.uk, or by contacting The Cam- Kirill A. Lomachenko \u2212 Department of Chemistry and NIS\nbridge Crystallographic Data Centre, 12 Union Road, Interdepartmental Center, University of Turin, Turin I-\nCambridge CB2 1EZ, UK; fax: +44 1223 336033. 10135, Italy; European Synchrotron Radiation Facility,\n\n\n\u25a0\n 38043 Grenoble, France; The Smart Materials Research\n Institute, Southern Federal University, Rostov-on-Don\n AUTHOR INFORMATION 344090, Russia; orcid.org/0000-0003-0238-1719\nCorresponding Authors Yong Qian \u2212 Jiangsu Collaborative Innovation Center of\n Zhi Su \u2212 Jiangsu Collaborative Innovation Center of Biomedical Functional Materials, School of Chemistry and\n Biomedical Functional Materials, School of Chemistry and Materials Science, Nanjing Normal University, Nanjing\n Materials Science, Nanjing Normal University, Nanjing 210023, China\n 210023, China; orcid.org/0000-0003-1690-527X; Complete contact information is available at:\n Email: zhisu@njnu.edu.cn https://pubs.acs.org/10.1021/acs.inorgchem.1c00241\n Martin J. Paterson \u2212 Institute of Chemical Sciences, School of\n Engineering & Physical Sciences, Heriot-Watt University, Author Contributions\n Edinburgh, Scotland EH14 4AS, U.K.; orcid.org/0000- X.X., Y.F., L.H., L.S., and L.-F.H. contributed to this\n 0002-0012-974X; Email: M.J.Paterson@hw.ac.uk manuscript equally. X.X., Y.F., L.H., L.S., L.-F.H., Y.-Y.H.,\n Zong-Wan Mao \u2212 MOE Key Laboratory of Bioinorganic and M.J.K., C.S., R.J.N., A.H., C.G., and K.A.L. designed and\n Synthetic Chemistry, School of Chemistry, Sun Yat-Sen performed the experiments. Z.S., M.J.P., Z.-W.M., H.-K.L., and\n University, Guangzhou 510275, China; orcid.org/0000- P.J.S. proposed and supervised the project. All the authors\n 0001-7131-1154; Email: cesmzw@mail.sysu.edu.cn were involved in discussing the results and writing the\n Hong-Ke Liu \u2212 Jiangsu Collaborative Innovation Center of manuscript.\n Biomedical Functional Materials, School of Chemistry and\n Notes\n Materials Science, Nanjing Normal University, Nanjing\n The authors declare no competing \ufb01nancial interest.\n\n \u25a0\n 210023, China; orcid.org/0000-0002-0968-6518;\n Email: liuhongke@njnu.edu.cn\n Peter J. Sadler \u2212 Department of Chemistry, University of ACKNOWLEDGMENTS\n Warwick, Coventry CV4 7AL, U.K.; orcid.org/0000- We thank Dr. Zhe Liu, Dr. Yao Zhao, and Dr. Lihong Li for\n 0001-9160-1941; Email: P.J.Sadler@warwick.ac.uk their help with NMR and MS experiments and Dr. P. Glatzel,\n 17459 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\nDr. E. Gallo, Dr. D. Gianolio, and Prof. C. Lamberti for the bisdemethoxycurcumin co-ligands. Dalton Trans. 2015, 44 (47),\nassistance during the XAS experiment at ID26 beamline of the 20523\u221220531.\nESRF. We thank NSFC (nos. 22077066, 21771109, 21778033, (14) Pettinari, R.; Condello, F.; Marchetti, F.; Pettinari, C.;\n21701195, 21837006, and 21977052), NSF of Jiangsu Smolenski, P.; Riedel, T.; Scopelliti, R.; Dyson, P. J. Dicationic\nProvince (no. BK20171472), ERC (grant 247450), EPSRC ruthenium(II)-arene-curcumin complexes containing methylated\n 1,3,5-triaza-7-phosphaadamantane:synthesis, structure, and cytotox-\n(grants EP/F034210/1, EP/P030572/1, EP/P001459, and\n icity. Eur. J. Inorg. Chem. 2017, 2017 (22), 2905\u22122910.\nEP/T021675), Anglo American Platinum, and the Severo (15) Caruso, F.; Rossi, M.; Benson, A.; Opazo, C.; Freedman, D.;\nOchoa Centres of Excellence Programme for L.S. by the Monti, E.; Gariboldi, M. B.; Shaulky, J.; Marchetti, F.; Pettinari, R.;\nSpanish State Research Agency (ref CEX2018-000867-S) for Pettinari, C. Ruthenium-arene complexes of curcumin: X-ray and\ntheir \ufb01nancial support.\n\n\u25a0\n density functional theory structure, synthesis, and spectroscopic\n characterization, in vitro antitumor activity, and DNA docking studies\n REFERENCES of (p-cymene)Ru(curcuminato)chloro. J. Med. Chem. 2012, 55 (3),\n (1) Shi, H. Y.; Sadler, P. J. How promising is phototherapy for 1072\u22121081.\ncancer? Br. J. Cancer 2020, 123 (6), 871\u2212873. (16) Pettinari, R.; Marchetti, F.; Condello, F.; Pettinari, C.; Lupidi,\n (2) Boros, E.; Dyson, P. J.; Gasser, G. Classification of metal-based G.; Scopelliti, R.; Mukhopadhyay, S.; Riedel, T.; Dyson, P. J.\ndrugs according to their mechanisms of action. Chem. 2020, 6 (1), Ruthenium(II)-arene RAPTA type complexes containing curcumin\n41\u221260. and bisdemethoxycurcumin display potent and selective anticancer\n (3) Huang, H. Y.; Banerjee, S.; Qiu, K. Q.; Zhang, P. Y.; Blacque, O.; activity. Organometallics 2014, 33 (14), 3709\u22123715.\nMalcomson, T.; Paterson, M. J.; Clarkson, G. J.; Staniforth, M.; (17) Pettinari, R.; Marchetti, F.; Nicola, C. D.; Pettinari, C.;\nStavros, V. G.; Gasser, G.; Chao, H.; Sadler, P. J. Targeted photoredox Cuccioloni, M.; Bonfili, L.; Eleuteri, A. M.; Therrien, B.; Batchelore,\ncatalysis in cancer cells. Nat. Chem. 2019, 11 (11), 1041\u22121048. L. K.; Dyson, P. J. Novel osmium(II)-cymene complexes containing\n (4) Liu, H. K.; Sadler, P. J. Metal complexes as DNA intercalators. curcumin and bisdemethoxycurcumin ligands. Inorg. Chem. Front.\nAcc. Chem. Res. 2011, 44 (5), 349\u2212359. 2019, 6 (9), 2448\u22122457.\n (5) Xue, X. L.; Qian, C. G.; Tao, Q.; Dai, Y. X.; Lv, M. D.; Dong, J. (18) Wilson, C. S.; Prior, T. J.; Sandland, J.; Savoie, H.; Boyle, R. W.;\nW.; Su, Z.; Qian, Y.; Zhao, J.; Liu, H. K.; Guo, Z. J. Using Murray, B. S. Homo- and hetero-dinuclear arene-linked osmium(II)\nbioorthogonally catalyzed lethality strategy to generate mitochondria- and ruthenium(II) organometallics: probing the impact of metal\ntargeting antitumor metallodrugs in vitro and in vivo. Natl. Sci. Rev. variation on reactivity and biological activity. Chem. - Eur. J. 2020, 26\n2020, DOI: 10.1093/nsr/nwaa286. (50), 11593\u221211603.\n (6) (a) Holder, A. A.; Zigler, D. F.; Tarrago-Trani, M. T.; Storrie, B.; (19) Banerjee, S.; Chakravarty, A. R. Metal complexes of curcumin\nBrewer, K. J. Photobiological impact of [{(bpy)2Ru(dpp)}2RhCl2]Cl5 for cellular imaging, targeting, and photoinduced anticancer activity.\nand [{(bpy)2Os(dpp)}2RhCl2]Cl5 [bpy = 2,2\u2032-bipyridine; dpp = 2,3- Acc. Chem. Res. 2015, 48 (7), 2075\u22122083.\nbis(2-pyridyl)pyrazine] on Vero cells. Inorg. Chem. 2007, 46 (12), (20) Wanninger, S.; Lorenz, V.; Subhan, A.; Edelmann, F. T. Metal\n4760\u22124762. (b) Smith, N. A.; Sadler, P. J. Photoactivatable metal complexes of curcumin-synthetic strategies, structures and medicinal\ncomplexes: from theory to applications in biotechnology and applications. Chem. Soc. Rev. 2015, 44 (15), 4986\u22125002.\nmedicine. Philos. Trans. R. Soc., A 2013, 371 (1995), 20120519. (21) Bhattacharyya, U.; Kumar, B.; Garai, A.; Bhattacharyya, A.;\n(c) Schatzschneider, U. Photoactivated biological activity of Kumar, A.; Banerjee, S.; Kondaiah, P.; Chakravarty, A. R. Curcumin\ntransition-metal complexes. Eur. J. Inorg. Chem. 2010, 2010, 1451\u2212 \u201cdrug\u201d stabilized in oxidovanadium(iv)-bodipy conjugates for\n1467. (d) Imberti, C.; Zhang, P. Y.; Huang, H. Y.; Sadler, P. J. New mitochondria-targeted photocytotoxicity. Inorg. Chem. 2017, 56\ndesigns for phototherapeutic transition metal complexes. Angew. (20), 12457\u221212468.\nChem., Int. Ed. 2020, 59, 61\u221273. (22) Mitra, K.; Gautam, S.; Kondaiah, P.; Chakravarty, A. R. The cis-\n (7) Wang, R.; Eberspacher, T. A.; Hasegawa, T.; Day, V.; Ware, D. diammineplatinum(ii) complex of curcumin: a dual action dna\nC.; Taube, H. Tmtacn, tacn, and triammine complexes of (eta 6- crosslinking and photochemotherapeutic agent. Angew. Chem. 2015,\narene) OsII: syntheses, characterizations, and photosubstitution 127 (47), 14195\u221214199.\nreactions (tmtacn = 1,4,7-trimethyl-1,4,7-triazacyclononane; tacn = (23) Pucci, D.; Crispini, A.; Mendiguchia, B. S.; Pirillo, S.; Ghedini,\n1,4,7-triazacyclononane). Inorg. Chem. 2001, 40 (4), 593\u2212600.\n M.; Morelli, S.; De Bartolo, L. Improving the bioactivity of Zn(II)-\n (8) Farrer, N. J.; Salassa, L.; Sadler, P. J. Photoactivated\n curcumin based complexes. Dalton Trans. 2013, 42 (26), 9679\u22129687.\nchemotherapy (PACT): the potential of excited-state d-block metals\n (24) Singh, J.; Alayon, E. M. C.; Tromp, M.; Safonova, O. V.;\nin medicine. Dalton Trans. 2009, 48, 10690\u221216701.\n Glatzel, P.; Nachtegaal, M.; Frahm, R.; van Bokhoven, J. A.\n (9) Zhang, P. Y.; Huang, H. Y. Future potential of osmium\ncomplexes as anticancer drug candidates, photosensitizers and Generating highly active partially oxidized platinum during oxidation\norganelle-targeted probes. Dalton Trans. 2018, 47 (42), 14841\u2212 of carbon monoxide over Pt/Al2O3: in situ, time-resolved, and high-\n14854. energy-resolution x-ray absorption spectroscopy. Angew. Chem., Int.\n (10) Needham, R. J.; Sanchez-Cano, C.; Zhang, X.; Romero- Ed. 2008, 47 (48), 9260\u22129264.\nCanelon, I.; Habtemariam, A.; Cooper, M. S.; Meszaros, L.; Clarkson, (25) Rosenberg, R. A.; Simons, J. K.; Frigo, S. P.; Tan, K.; Chen, J.\nG. J.; Blower, P. J.; Sadler, P. J. In-cell activation of organo- M. X-ray fluorescence detection of low-Z elements using a\nOsmium(II) anticancer complexes. Angew. Chem., Int. Ed. 2017, 56 microchannel plate detector. Rev. Sci. Instrum. 1992, 63 (4), 2193\u2212\n(4), 1017\u22121020. 2194.\n (11) Peacock, A. F. A.; Parsons, S.; Sadler, P. J. Tuning the (26) Taylor, R. C.; Cullen, S. P.; Martin, S. J. Apoptosis: controlled\nhydrolytic aqueous chemistry of osmium arene complexes with N,O- demolition at the cellular level. Nat. Rev. Mol. Cell Biol. 2008, 9 (3),\nchelating ligands to achieve cancer cell cytotoxicity. J. Am. Chem. Soc. 231\u2212241.\n2007, 129 (11), 3348\u22123357. (27) He, L.; Tan, C. P.; Ye, R. R.; Zhao, Y. Z.; Liu, Y. H.; Zhao, Q.;\n (12) Coverdale, J. P. C.; Romero-Canelon, I.; Sanchez-Cano, C.; Ji, L. N.; Mao, Z. W. Theranostic iridium(III) complexes as one- and\nClarkson, G. J.; Habtemariam, A.; Wills, M.; Sadler, P. J. Asymmetric two- photon phosphorescent trackers to monitor autophagic\ntransfer hydrogenation by synthetic catalysts in cancer cells. Nat. lysosomes. Angew. Chem., Int. Ed. 2014, 53 (45), 12137\u221212141.\nChem. 2018, 10 (3), 347\u2212354. (28) Prasad, P.; Pant, I.; Khan, I.; Kondaiah, P.; Chakravarty, A. R.\n (13) Pettinari, R.; Marchetti, F.; Pettinari, C.; Condello, F.; Petrini, Mitochondria-targeted photoinduced anticancer activity of\nA.; Scopelliti, R.; Riedel, T.; Dyson, P. J. Organometallic rhodium(III) oxidovanadium(IV) complexes of curcumin in visible light. Eur. J.\nand iridium(III) cyclopentadienyl complexes with curcumin and Inorg. Chem. 2014, 2014 (14), 2420\u22122431.\n\n 17460 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\fInorganic Chemistry pubs.acs.org/IC Article\n\n (29) Li, J.; Yuan, J. Caspases in apoptosis and beyond. Oncogene C2Deltag and X2Pi g-D2Deltag transitions. J. Chem. Phys. 2005, 122\n2008, 27 (48), 6194\u22126206. (1), 014315.\n (30) Li, Y. C.; Gu, Z. Y.; Zhang, C.; Li, S. H.; Zhang, L.; Zhou, G. (47) Das, D.; Scherer, T. M.; Das, A.; Mondal, T. K.; Mobin, S. M.;\nQ.; Wang, S. X.; Zhang, J. C. Synthesis, characterization and ROS- Fiedler, J.; Priego, J. L.; Jim\u00e9nez-Aparicio, R.; Kaim, W.; Lahiri, G. K.\nmediated antitumor effects of palladium(II) complexes of curcumi- The intricate paramagnetic state of [Os(Q)2(bpy)]+, Q = 4,6-di-tert-\nnoids. Eur. J. Med. Chem. 2018, 144, 662\u2212671. butyl-o- iminobenzoquinone. Dalton Trans. 2012, 41 (38), 11675\u2212\n (31) LeBel, C. P.; Ischiropoulos, H.; Bondy, S. C. Evaluation of the 11683.\nprobe 2\u2032,7\u2032-dichloroflurescin as an indicator of reactive oxygen species\nformation and oxidative stress. Chem. Res. Toxicol. 1992, 5 (2), 227\u2212\n231.\n (32) Xue, X. L.; Qian, C. G.; Fang, H. B.; Liu, H. K.; Yuan, H.; Guo,\nZ. J.; Bai, Y.; He, W. J. Photoactivated lysosomal escape of a\nmonofunctional PtII complex Pt-BDPA for nucleus access. Angew.\nChem., Int. Ed. 2019, 58 (36), 12661\u221212666.\n (33) Zhang, L.; Cheng, X.; Gao, Y. Y.; Bao, J. D.; Guan, H. X.; Lu, R.\nR.; Yu, H. X.; Xu, Q.; Sun, Y. Induction of ROS-independent DNA\ndamage by curcumin leads to G2/M cell cycle arrest and apoptosis in\nhuman papillary thyroid carcinoma BCPAP cells. Food Funct. 2016, 7\n(1), 315\u2212325.\n (34) Friedl, P.; Alexander, S. Cancer invasion and the microenviron-\nment: plasticity and reciprocity. Cell 2011, 147 (5), 992\u22121009.\n (35) Bordiga, S.; Groppo, E.; Agostini, G.; van Bokhoven, J. A.;\nLamberti, C. Reactivity of surface species in heterogeneous catalysts\nprobed by in situ X-ray absorption techniques. Chem. Rev. 2013, 113\n(3), 1736\u22121850.\n (36) Lomachenko, K. A.; Garino, C.; Gallo, E.; Gianolio, D.;\nGobetto, R.; Glatzel, P.; Smolentsev, N.; Smolentsev, G.; Soldatov, A.\nV.; Lamberti, C.; Salassa, L. High energy resolution core-level X-ray\nspectroscopy for electronic and structural characterization of osmium\ncompounds. Phys. Chem. Chem. Phys. 2013, 15 (38), 16152\u221216519.\n (37) Garai, A.; Pant, I.; Banerjee, S.; Banik, B.; Kondaiah, P.;\nChakravarty, A. R. Photorelease and cellular delivery of mitocurcumin\nfrom its cytotoxic cobalt(III) complex in visible light. Inorg. Chem.\n2016, 55 (12), 6027\u22126035.\n (38) Garipler, G.; Mutlu, N.; Lack, N. A.; Dunn, C. D. Deletion of\nconserved protein phosphatases reverses defects associated with\nmitochondrial DNA damage in Saccharomyces cerevisiae. Proc. Natl.\nAcad. Sci. U. S. A. 2014, 111 (4), 1473\u22121478.\n (39) Peacock, A. F. A.; Habtemariam, A.; Fernandez, R.; Walland,\nV.; Fabbiani, F. P. A.; Parsons, S.; Aird, R. E.; Jodrell, D. I.; Sadler, P.\nJ. Tuning the reactivity of osmium(II) and ruthenium(II) arene\ncomplexes under physiological conditions. J. Am. Chem. Soc. 2006,\n128 (5), 1739\u22121748.\n (40) Becke, A. D. Density-functional thermochemistry. III. The role\nof exact exchange. J. Chem. Phys. 1993, 98 (7), 5648\u22125652.\n (41) Hay, P. J.; Wadt, W. R. Ab initio Effective core potentials for\nmolecular calculations-Potentials for the transition-metal atoms Sc to\nHg. J. Chem. Phys. 1985, 82 (1), 270\u2212283.\n (42) McLean, A. D.; Chandler, G. S. Contracted Gaussian basis sets\nfor molecular calculations. I. Second row atoms, Z = 11\u221218. J. Chem.\nPhys. 1980, 72 (10), 5639\u22125648.\n \u030c\n (43) (a) Vlcek, A., Jr.; Z\u0301 alis,\u030c S. Modeling of charge-transfer\ntransitions and excited states in d6 transition metal complexes by\nDFT techniques. Coord. Chem. Rev. 2007, 251 (3\u22124), 258\u2212287.\n(b) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R.\nMolecular excitation energies to high-lying bound states from time\ndependent density-functional response theory: Characterization and\ncorrection of the time-dependent local density approximation\nionization threshold. J. Chem. Phys. 1998, 108 (11), 4439\u22124449.\n (44) Krauklis, V. I.; Reshetova, I. K.; Podkopaeva, Y. O.; Chizhov, V.\nY. Electronically exited states in triphenylphosphine complexes of\nRuthenium (II): TD DFT study. J. Photochem. Photobiol., A 2018,\n354, 112\u2212118.\n (45) Browne, W. R.; O\u2019Boyle, N. M.; McGarvey, J. J.; Vos, J. G.\nElucidating excited state electronic structure and intercomponent\ninteractions in multicomponent and supramolecular systems. Chem.\nSoc. Rev. 2005, 34 (8), 641\u2212663.\n (46) Ram\u00edrez-Sol\u00eds, A.; Daudey, J. P. Ab initio study on the\nspectroscopy of CuCl2. II. Benchmark calculations on the X2Pi g-\n\n 17461 https://doi.org/10.1021/acs.inorgchem.1c00241\n Inorg. Chem. 2021, 60, 17450\u221217461\n\f", "pages_extracted": 12, "text_length": 84633}