Anticancer Cyclometalated Iridium(III) Complexes with Planar Ligands

Feb 28, 2019 - Design of N-Benzoxaborole Benzofuran GSK8175—Optimization of Human Pharmacokinetics Inspired by Metabolites of a Failed Clinical HCV ...
0 downloads 0 Views 2MB Size
Subscriber access provided by WEBSTER UNIV

Article

Anticancer Cyclometalated Iridium(III) Complexes with Planar Ligands: Mitochondrial DNA Damage and Metabolism Disturbance Jianjun Cao, Yue Zheng, Xiao-Wen Wu, Cai-Ping Tan, Mu-He Chen, Na Wu, Liang-Nian Ji, and Zongwan Mao J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01704 • Publication Date (Web): 28 Feb 2019 Downloaded from http://pubs.acs.org on February 28, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Anticancer Cyclometalated Iridium(III) Complexes with Planar Ligands: Mitochondrial DNA Damage and Metabolism Disturbance Jian-Jun Cao,† Yue Zheng,† Xiao-Wen Wu, Cai-Ping Tan,* Mu-He Chen, Na Wu, Liang-Nian Ji and Zong-Wan Mao* MOE Key Laboratory of Bioinorganic and Synthetic Chemistry, School of Chemistry, Sun Yat-Sen University, Guangzhou 510275, China. Email: [email protected]; [email protected] † These authors contributed equally to this work.

1

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 43

ABSTRACT: Emerging studies have shown that mitochondrial DNA (mtDNA) is a potentially target for cancer therapy. Herein, six cyclometalated Ir(III) complexes Ir1‒Ir6 containing a series of extended planar diimine ligands have been designed and assessed for their efficacy as anticancer agents. Ir1‒Ir6 show much higher cytotoxicity than cisplatin, and they can effectively localize to mitochondria. Among them, complexes Ir3 and Ir4 with dipyrido[3,2-a:2',3'-c]phenazine (dppz) ligands can bind to DNA tightly in vitro, intercalate to mtDNA in situ, and induce mtDNA damage. Ir3- and Ir4-impaired mitochondria undergo mitochondrial membrane potential decline, disability of ATP generation, disruption of mitochondrial energetic and metabolic status, which subsequently causes protective mitophagy, G0/G1 phase cell cycle arrest and apoptosis. In vivo antitumor evaluations also show that Ir4 can inhibit tumor xenograft growth effectively. Overall, our work proves that targeting mitochondrial genome may present an effective

strategy

to

develop

metal-based

anticancer

agents

to

overcome

cisplatin-resistance.

2

ACS Paragon Plus Environment

Page 3 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

1. INTRODUCTION The serendipitous discovery and successful clinical applications of platinum-based anticancer agents have inspired the search for other mental-based biochemical agents.1,2 In recent years, non-platinum anticancer drugs have received widespread attention because they can act through different mechanisms, such as targeting subcellular organelles,3,4 proteins,5,6 mitochondrial DNA and DNA mismatches.7 Cyclometalated Ir(III) complexes with polypyridyl ligands and proper cationic d6 metal center, have recently been rapidly extended to yield a wide variety of bioimaging probes or therapeutics.8,9 By structural modifications, cyclometalated iridium complexes can easily permeate the cell membrane and accumulate in different subcellular organelles, e.g., mitochondria9 and lysosomes.10 Under dark and light conditions, these complexes exhibit interesting anticancer activities.11,12 However, the anticancer mechanisms and the molecular targets of these compounds are largely unknown. Mitochondria play crucial roles in various processes of carcinogenesis, including energy source, cytosolic biosynthetic precursors production and cell death regulation.13,14 The proximal exposures of mitochondrial DNA (mtDNA) to reactive oxygen species (ROS) lead to relatively high mutation rate in mtDNA that have been identified exist in multiform cancer.15 These mitochondrial genomic aberrations result in mitochondrial function alternation to maintain cancer metabolism.16 The traditional DNA-targeted antineoplastic drug, cisplatin is normally confronted with serious resistance due to accelerated DNA repair in nuclei.17 In contrast, mtDNA doesn’t possess the nucleic 3

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 43

excision repair (NER) pathway and lacks the protection of histones, so mtDNA is more vulnerable to pharmacological actions as compared with nuclear DNA.18 Thus, targeting mtDNA presents promising therapeutic approaches for the treatment of aggressive and cisplatin-resistant cancers.19 To prove this hypothesis, recently, researchers have modified several clinical drugs targeting nuclear DNA with mitochondrial targeting moieties to interact with mtDNA.19-21 Besides mitochondria-targeted nanocarriers,22 the modification of anticancer agents with mitochondrial-oriented groups, such as mitochondria-targeting peptide23 and triphenylphosphonium group, is also proved to be very effective.24 Mitochondrial metabolism is emerging as a possible target for cancer therapy.25 In the early observations, researchers find that cancer cells take up glucose and produce large quantities of lactate, which is well-known as the Warburg effect.26 Cancer cells primarily utilize glycolysis as the major metabolic pathway for proliferation, which is one main aspect of mitochondrial abnormalities in tumors.26 Furthermore, copious amounts of ROS are produced in mitochondria, which can promote mtDNA damage and genetic instability.27 Moreover, mitochondrial dysfunction is gradually recognized as the metabolic hallmark of cancer cells, and genetic evidence also show that mitochondrial metabolism is essential for tumorigenesis.28 Tumor cells rely on glycolysis and mitochondrial metabolism to provide the necessary building blocks for macromolecule (e.g., nucleotides, lipids and amino acids and ATP), which is essential for tumor cell survival, growth, and proliferation.29

4

ACS Paragon Plus Environment

Page 5 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

It has been reported that lipophilic cationic metal complexes, e.g., cyclometalated Ir(III) complexes, can preferentially accumulate in mitochondria to perform their imaging or anticancer functions.30,31 However, their specific binding molecules in mitochondria are rarely elucidated. In this work, we synthesized a series of mitochondrial-located Ir(III)-based metallo-intercalators Ir1‒Ir6 containing extended aromatic ligands (Figure 1a). The DNA binding properties in vtiro, and mtDNA damage in situ were explored. Their impact on mitochondrial energetic/metabolic status and the mechanisms through which they induce cell death were also investigated. Finally, the in vivo antitumor potency was evaluated in a xenograft mice model. We demonstrate that mtDNA is an important molecular target of these complexes and targeting mitochondrial metabolism may represent an effective strategy for metallo-drugs to overcome cisplatin resistance.

Figure 1. a) Chemical structures of Ir1‒Ir6. b) The X-ray crystal structure and atom-numbering scheme for Ir4 at a 30% thermal ellipsoids probability level. The 5

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 43

hydrogen atoms, counter ions and solvents are omitted for clarity. c) UV/Vis spectra of Ir1‒Ir6 in CH3CN measured at 298 K. 2. RESULTS AND DISCUSSION 2.1.

Synthesis

and

Characterization.

The

ligands

(dpq:

pyrazino[2,3-f][1,10]phenanthroline;32 dppz: dipyrido[3,2-a:2',3'-c]phenazine;33 dppn: benzo[i]dipyrido[3,2-a:2',3'-c]phenazine34) were synthesized by literature methods. The Ir(III) dimers ([(dfppy)2Ir(μ-Cl)]2, dfppy: 2-(2,4-difluorophenyl)pyridine; [(ptz)2Ir(μ-Cl)]2, ptz: 2-phenylthiazole) were synthesized using the method which was previously reported.35 The iridium complexes were obtained by refluxing two equivalents of ligands and the corresponding cyclometalated Ir(III) dimers in CH2Cl2/CH3OH (1/1, v/v) followed by anion metathesis with NH4PF6 and purification by column chromatography on silica gel. Ir1‒Ir6 were confirmed by 1H NMR spectroscopy (Figure S1–S6, Supporting Information (SI)), ESI-MS, high performance liquid chromatography (HPLC; SI, Figure S7–S12) and elemental analysis. Ir4 was also characterized by X-ray crystallography (Figure 1b; SI, Table S1). The selected bond lengths and angles of Ir4 are listed in Table S2 (SI). The Ir(III) atom in Ir4 adopts a distorted octahedral geometry and the trans angles at the iridium center range from 171.9(4)º to 173.2(3)º. The two Ir‒C bonds are in a mutual cis arrangement. The trans influence of carbon donors renders slightly shorter Ir‒N bond lengths of dppz ligand (approximately 2.135(8)‒2.142(9) Å) than those of the cyclometalated ligands (approximately 2.041(8) ‒2.074(9) Å). The absorption spectra of all complexes were measured in phosphate buffer saline (PBS), CH3CN and CH2Cl2 (Figure 1c; SI, Figure S13). The intense absorbance bands at 6

ACS Paragon Plus Environment

Page 7 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

approximately 250‒350 nm are assigned to spin-allowed ligand-centered (1LC) π–π* transitions arising from cyclometalated (C–N) and diimine (N–N) ligands. When the π-conjugation on the diimine ligand is expanded, these most intense bands exhibit a systematic bathochromic-shift, possibly due to the decrease in the energy of the diimine ligand localized π* orbital.36 The less intense absorbance shoulders at around 350‒450 nm are mainly attributable to charge-transfers (metal to ligand charge transfer, ligand-to-ligand charge transfer and/or intraligand charge transfer) transitions, which are likely to be mixed with the 1LC π–π* character of ligands dppz and dppn.37 All complexes display very weak absorbance tails beyond 450 nm that are more likely contributed by the spin-forbidden π–π*/3CT transitions.36 The emission spectra of Ir1‒Ir6 were studied in PBS, CH3CN and CH2Cl2 (SI, Table S3). The complexes Ir1 and Ir2 containing the dpq ligand emit strongly in CH2Cl2 and show solvent-dependent emission quantum yields. By contrast, other complexes containing dppz or dppn show very weak emission.

2.2. Lipophilicity, Cellular Uptake and Localization. Cellular uptake behaviors of metal complexes have been reported to be affected by many factors, e.g., lipophilicity, molecular size and substitute group.12 The lipophilicity is commonly referred to as the partition coefficient of the compound in n-octanol/water (Po/w). The log Po/w values of Ir(III) complexes in this work have been determined by a classical shake-flask method. The complexes containing dppn (Ir5 and Ir6) ligands show substantially higher lipophilicity than dppz (Ir3 and Ir4) and dpq (Ir1 and Ir2), which is in accordance with 7

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 43

the hydrophobic nature of the diimine ligands and the enlargement of the planar area. Similar trend is also observed for the cellular uptake efficacy of iridium, which is quantitatively determined by inductively coupled plasma-mass spectrometry (ICP-MS) (SI, Table S4). To further investigate the subcellular distribution of Ir1‒Ir6, the iridium contents in different cellular compartments isolated from Ir(III)-treated A549 cells were measured by ICP-MS (Figure 2). As expected, the contents of iridium in mitochondria are much higher than those obtained in cytosol and nuclei, especially for Ir4. Utilizing their intrinsic emission, the localization of Ir1 and Ir2 in A549 cells was further investigated by laser scanning confocal microscopy (SI, Figure S14). Ir1 and Ir2 can be visualized in mitochondria of A549 cells specifically, which is confirmed by colocalization

experiments

with

the

mitochondrion-specific

fluorescent

probe

MitoTracker Deep Red (MTDR) and the lysosome-specific LysoTracker Deep Red (LTDR). In conformity with our expectations, the results indicate that these complexes can target mitochondria in A549 cells, which offers the possibility for complexes to interact with mtDNA.

8

ACS Paragon Plus Environment

Page 9 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Figure 2. Distribution of complexes Ir1‒Ir6 in cellular compartments of A549 cells measured by ICP-MS. Cell were treated with Ir(III) at a concentration of 5 μM for 1 h. 2.3. In Vitro Cytotoxicity. The in vitro cytotoxicity of Ir1‒Ir6 and cisplatin was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay on A549 (human lung cancer), HepG2 (human liver cancer), HeLa (human cervical cancer), PC3 (human prostate cancer), A549R (cisplatin-resistant A549), HeLa-ρ0 (mtDNA-less HeLa) and HLF (human lung fibroblasts) cells. After 48 h treatment, all complexes exhibit high cytotoxicity with IC50 values ranging from 0.1 to 8.4 μM, which is higher than those of cisplatin against all human cancer cell lines tested (Table 1). Generally, the order of the cytotoxicity of Ir1‒Ir6 is as follows: Ir3 ≈ Ir4 > Ir5 ≈ Ir6 > Ir1 ≈ Ir2. The highest cytotoxicity is observed in HeLa cells. For example, the IC50 value of Ir4 against HeLa cells is 0.1, which is about 100-fold higher that of cisplatin. Notably, Ir1‒Ir6 show much lower cytotoxicity against human nomal HLF cells. Ir1‒Ir6 are highly active in PC3 cell line that is androgen-independent and inherently resistant to cisplatin therapy.22 The IC50 values of Ir3 (0.76 μM) and Ir4 (0.72 μM) are about 30- and 32-fold lower than that of cisplatin, respectively. The high cytotoxicity of Ir1‒Ir6 is also retained in the cisplatin-resistant A549 cell line. The inhibitory activities of Ir3 and Ir4 are almost 200 times higher than that of cisplatin. These results indicate that Ir1‒Ir6 are likely to overcome both the intrinsic and acquired drug resistance of cisplatin. As we propose that mtDNA might present an important target of these complexes, we use a mtDNA-deficient cell line HeLa-ρ0 to investigate the content of mtDNA on 9

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 43

cytotoxicity of Ir1‒Ir6. We calculated the mtDNA-targeting factors (MTFs, the ratios of IC50 values of mtDNA-free ρ0 cell line to the wild-type cell line) to investigate whether the toxicity is due to mtDNA damage.38 The MTFs calculated for Ir1‒Ir6 range between 3.0‒9.0. Remarkably, the MTFs measured for Ir3 and Ir4 are 9.0 and 8.9, respectively. In contrast, no obvious difference in cytotoxicity is observed for cisplatin in HeLa and HeLa-ρ0 cells. These results indicate that mtDNA damage caused by Ir1‒Ir6 contributes to their anticancer activities. The soft agar colony formation assay is a well-established method to semi-quantitatively evaluate the suppressive effects of the anticancer agents on the anchorage-independent growth of tumor cells.39 The colony formation capability of Ir3and Ir4-treated A549 cells is greatly suppressed (Figure 3). MTT assay also shows that the cell proliferation is not influenced under the concentrations tested (SI, Figure S15).

Figure 3. Inhibition of colony formation by Ir3 and Ir4 at the indicated concentrations. A549 cells were incubated with the Ir (III) complexes for 1 week.

10

ACS Paragon Plus Environment

Page 11 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Table 1 Cytotoxicity of the tested compounds against different cell lines a IC50 (μM)

A549

HepG2

HeLa

PC3

A549R

HeLa-ρ0 (MTFs)

HLF

Ir1

1.11±0.1

8.41±0.4

0.72±0.06

2.16±0.07

2.51±0.2

2.88±0.10 (4.0)

10.62±0.9

Ir2

0.72±0.1

3.80±0.2

0.79±0.08

4.22±0.01

2.46±0.1

3.76±0.2 (4.7)

5.47±0.7

Ir3

0.35±0.02

1.88±0.2

0.15±0.01

0.76±0.03

0.43±0.3

1.41±0.10 (9.0)

5.34±1.2

Ir4

0.25±0.09

1.11±0.1

0.10±0.01

0.72±0.03

0.39±0.01

0.89±0.06 (8.9)

5.58±1.1

Ir5

0.33±0.02

2.19±0.2

0.43±0.02

1.03±0.03

0.71±0.5

1.36±0.09 (3.0)

7.25±1.5

Ir6

0.25±0.07

2.29±0.2

0.33±0.06

1.23±0.02

0.63±0.04

2.00±0.2 (6.0)

4.88±1.2

cisplatin

25.9±2.1

10.47±1.5

10.1±1.6

23.4±1.7

89.1±6.2

10.9±1.2 (1.0)

12.1±2.2

a

Cells were incubated with the indicated compounds for 48 h. Data are presented as the

means ± standard deviations (SD), and cell viability was assessed after 48 h of incubation.

2.4. In Vitro DNA Binding Properties. As Ir3 and Ir4 show prominent antitumor activity and relatively high MTF values, we choose them as the model compounds in the following studies. First, the DNA binding affinities of Ir3 and Ir4 were determined by UV/Vis spectra titrations and ethidium bromide (EB) competitive binding assay. The UV/Vis spectra of Ir3 (Figure 4a) and Ir4 (Figure 4b) exhibit obvious hypochromism 11

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 43

and bathochromic shifts upon addition of calf thymus DNA (CT-DNA). Meanwhile, an obvious decline in the fluorescence of EB bound to CT-DNA was detected upon the addition of Ir3 (Figure 4c) and Ir4 (Figure 4d). Both of these experiments suggest that Ir3 and Ir4 can bind to DNA tightly in vitro and Ir4 possesses higher binding affinity than Ir3.

Figure 4. UV/Vis spectra of a) Ir3 (10 μM) and b) Ir4 (10 μM) titrated with CT-DNA in Tris-HCl buffer (50 mM, pH 7.4). Fluorescence emission spectra of EB-DNA quenched by c) Ir3 and d) Ir4. The solutions containing EB (3 μM) and CT-DNA (30 μM) were incubated with Ir3 or Ir4 (0, 1, 2, 5, 8, 11, 14 μM) at 25 oC for 30 min.

12

ACS Paragon Plus Environment

Page 13 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

2.5. MtDNA Binding and Damage. The classical mtDNA intercalating dye PicoGreen was used to determine whether Ir3 and Ir4 can bind to mtDNA in live cells (Figure 5a). The fluorescence of PicoGreen can be quenched by the addition of a second intercalator.40 PicoGreen stained cells show bright nuclei surrounded by numerous punctate cytoplasmic mtDNA nucleoid staining. The addition of Ir3 and Ir4 (about 4 × IC50) results in the quenching of fluorescent signal of PicoGreen in mtDNA nucleoid to varying degrees. At the same time, the fluorescence in the nuclei is barely influenced. The results show that Ir3 and Ir4 can intercalate into mtDNA with no obvious nuclear accumulation. Interestingly, Ir4 causes the entire quenching of mtDNA nucleoid staining, which is similar to that observed for the well-established mtDNA intercalator, EB. Similar phenomena are observed in HeLa cells (Figure S16). For HeLa-ρ0 cells, due to the deficiency in mtDNA, the fluorescence intensity of mtDNA nucleoid staining is very low in the control cells. Then, the degradation of mtDNA in Ir3- and Ir4-treated A549 cells was detected by gel electrophoresis. Significantly weakened bands of mtDNA isolated from Ir(III)-treated A549 cells are detected (Figure 5b). The amounts of residual mtDNA treated by Ir3 and Ir4 decrease to about 63% and 59% compared with the control, respectively. However, Ir3 and Ir4 do not cause obvious mtDNA degradation in the non-redox environment (SI, Figure S17). The oxidation of guanine residues can produce 8-Oxoguanine (8-OG) that is the most common product of DNA oxidative damage.41 An immunofluorescence assay is applied to detect the presence of 8-OG in mtDNA caused by treatment of Ir3 and Ir4.42 13

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 43

An enhancement in the immunofluorescent signals of 8-OG in the cytoplasm is observed in A549 cells upon Ir(III) treatment (SI, Figure S18). The result confirms our hypothesis that Ir3 and Ir4 can cause mtDNA oxidative damage. Each human cell contains a unique nuclear genome and plentiful copies of the mitochondrial genome. It has been reported that damaging mtDNA can cause replication blockage and depletion of mtDNA.43 The copy number of mtDNA relative to nuclear DNA in Ir3- and Ir4-treated A549 cells was measured by the real time-polymerase chain reaction (RT-PCR).44 It can be seen from Figure 5c, the copy number of mtDNA is reduced by about 49% and 56% in A549 cells treated with Ir3 (0.5 μM, 24 h) and Ir4 (0.5 μM, 24 h), respectively. Next, we amplified a mitochondrial genomic segment (8.9 kb) and a nuclear genomic segment (12.2 kb) in Ir(III)-treated A549 cells by PCR (Figure 5d). Ir3 and Ir4 can reduce amplification efficiency of mtDNA segment significantly, while their impact on the amplification of nuclear DNA segment is not so obvious. The results confirm the selectivity of Ir3 and Ir4 for mtDNA over nuclear DNA. Among 37 genes encoded in human mitochondrial genome, 13 genes encoding essential polypeptides belong to the mitochondrial electron transport chain and constitutes the vital parts of the OXPHOS system.45 The impacts of Ir3 and Ir4 on the transcription levels of these genes in A549 cells are quantitatively assessed by RT-PCR.46 After 6 h treatment, the expression of all of the 13 genes shows a downward trend. On a standard of a fold change of more than two times, the expression of ND1, ND2, and ND3 is commonly decreased following treatment with Ir3 and Ir4 (Figure 5e and 5f). ND1-3 are three of seven subunits encoded by the mitochondrial genome of NADH 14

ACS Paragon Plus Environment

Page 15 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

(Nicotinamide adenine dinucleotide) dehydrogenase that belongs to the largest complex of the respiratory chain, namely Complex I.47 NADH dehydrogenase can catalyze NADH dehydrogenation and the transfer of electrons from NADH to Coenzyme Q10.48 Meanwhile, the expression of ND5, ND6 and Cytb also descends upon Ir3 treatment, and an expression decline of COX3, ATP8 is observed for Ir4. Similar results were obtained by using the other housekeeping gene, gapdh (SI, Figure S19). These results suggest that Ir3 and Ir4 can suppress the transcription of mtDNA and may further affect mitochondrial functions.

Figure 5. a) The quenching of PicoGreen fluorescence by Ir3 (1 μM), Ir4 (1 μM) and EB (1.0 μg/mL) for 1.5 h in A549 cells. Scale bar: 10 μm. b) Degradation of mtDNA by Ir3 (5 μM) and Ir4 (5 μM) in the presence of ascorbate (50 μM) /H2O2 (50 μM) at 37 °C for 1 h. c) Impact of Ir3 or Ir4 on mtDNA copy number in A549 cells at the indicated concentrations. d) Impact of Ir3 and Ir4 on the amplification the mtDNA and nuclear 15

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 43

DNA segments. e) and f) The impact of Ir3 (1 μM, e) and Ir4 (1 μM, f) on the transcriptional levels of 13 genes encoded in mitochondrial genome determined by RT-PCR. The relative fold changes in genes expression were normalized according to the average expression of the housekeeping gene β-actin.

2.6. Induction of Mitochondrial Dysfunction. We then examine the impact of Ir3 and Ir4 on mitochondrial functions. Mitochondrial membrane potential (MMP, ∆Ψm) is the basic characteristic to reflect mitochondrial integrity.49 Mitochondrial depolarization induced by Ir3 and Ir4 was detected by rhodamine 123 (Rh123) staining monitored by flow cytometry. Rh123 is a MMP-sensitive mitochondrial dye, which leaks out from mitochondria along with the decline of MMP.49 Compared with the control cells, the fluorescence intensity of Rh123 in Ir3 (1 μM, 6 h) and Ir4 (1 μM, 6 h)-treated cells decrease to about 50% and 31%, respectively (Figure 6a). In addition, treatment of Ir3 and Ir4 results in a significant dose-dependent decrease in ATP production (SI, Figure S20). Compared with the control cells, the ATP content decreases to 20.6 ± 4.2% and 18.8 ± 4.4% in cells treated with Ir3 (1 μM, 6 h) and Ir4 (1 μM, 6 h), respectively. Next, the impact of Ir3 and Ir4 on mitochondrial respiration was investigated by quantitative detection of the oxygen consumption rate (OCR) using a Seahorse XF24 extracellular flux analyzer. In order to calculate the key parameters, several respiratory regulators that can target the components of electron transport chain are used to modulate the mitochondrial OXPHOS (Figure 6b). The decreases in basal OCR, ATP production,

16

ACS Paragon Plus Environment

Page 17 of 43 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

the maximum respiration, and non-mitochondrial respiration are detected in Ir3- and Ir4-treated A549 cells (SI, Figure S21).

Altered mitochondrial metabolism is closely related to tumorigenesis and cancer progression.50 The impact of Ir3 and Ir4 on mitochondrial metabolism was analyzed by measuring the metabolites of A549 cells using gas chromatography-time-of-flight mass spectrometry (GC-TOF-MS). The data were first analyzed using principal component analysis (PCA), a clear separation of the control group and the treated groups can be detected

(SI,

Figure

S22).

Following

analysis

of

orthogonal

partial

least

square-discriminant analysis (OPLS-DA), candidate biomarkers were picked out from the S-plots. The R2Y (the goodness of fit) and Q2Y (predictability) values obtained from the OPLS-DA score plots for the Ir3-treated samples and controls are 1.000 and 0.898, respectively (SI, Figure S23a). The location of the cluster of the Ir3-treated group is well separated from and that of the controls, indicating the difference in the metabolic profiles of Ir3-treated samples and the controls. Also, a clear discrimination is also detected in Ir4-treated samples and controls (SI, Figure S23b, R2Y = 0.909, Q2Y = 0.813). The S-plots are shown in Figure S24 and S25 (SI). 32 potential biomarkers and 28 potential biomarkers with the variable importance in the projection (VIP) >1 and p-value