Phosphorogenic Dipyrrinato-Iridium(III) Complexes as Photosensitizers for Photodynamic Therapy.
Abstract
Financial support from MINECO (MAT2017-83856-C3-2-P and 3-P), and PID2020-11455 GB-C32 and C33) and Gobierno Vasco (IT912-16) is gratefully acknowledged. A. L-V. was supported by an i-PFIS grant from Instituto de Salud Carlos III (IFI17/00039). E. A.-Z. thanks MINECO for a postdoctoral contract. The authors also wish to thank the Spanish Government (project RTI2018-100910-B-C41 (MCUI/FEDER, EU)) and the Generalitat Valenciana (project PROMETEO 2018/024) for support.
Full text
Phosphorogenic Dipyrrinato-Iridium(III) Complexes as Photosensitizers for Photodynamic Therapy A. Prieto-Castañeda, A. Lérida-Viso, E. Avellanal-Zaballa, R. Sola-Llano, J. Bañuelos, A. R. Agarrabeitia, R. Martínez-Máñez and M. J. Ortiz https://doi.org/10.1016/j.dyepig.2021.109886 Dyes and Pigments 197 (2022) 109886 This is the accepted manuscript of the article that appeared in final form in Dyes and Pigments 197 : (2022) // Article ID 109886, which has been published in final form at https://doi.org/10.1016/ j.dyepig.2021.109886. © 2021 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/)
1 Phosphorogenic Dipyrrinato-Iridium(III) Complexes as Photosensitizers for 1 Photodynamic Therapy. 2 A. Prieto-Castañeda,a A. Lérida-Viso,b,c E. Avellanal-Zaballa,d R. Sola-Llano,d J. Bañuelos,d* A. R. 3 Agarrabeitia,a R. Martínez-Máñez,b,c,e,f* and M. J. Ortiza* 4 aDepartamento de Química Orgánica, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, Ciudad 5 Universitaria s/n, 28040, Madrid, Spain. 6 bUnidad Mixta de Investigación en Nanomedicina y Sensores, IIS La Fe, Universitat Politècnica de València, Avda de 7 Fernando Abril Martorell, n.106 46026, Valencia, Spain. 8 cUnidad Mixta UPV-CIPF de Investigación en Mecanismos de Enfermedades y Nanomedicina, Universidad Politécnica de 9 Valencia, Centro de Investigación Príncipe Felipe, Carrer d’Eduardo Primo Yúfera 3, 46012, Valencia, Spain. 10 dDepartamento de Química Física, Universidad del País Vasco-EHU, Apartado 644, 48080, Bilbao, Spain. 11 eInstituto Interuniversitario de Investigación de Reconocimiento Molecular y Desarrollo Tecnológico (IDM), Universitat 12 Politècnica de València, Universitat de València, Camino de Vera s/n, 46022 Valencia, Spain. 13 fCIBER de Bioingeniería, Biomateriales y Nanomedicina, Madrid, Spain. 14 15 ABSTRACT: We have designed and synthesized a family of Ir(III) metal complexes coordinated 16 with two ciclometallated bis-phenylpyridine ligands and an ancillary dipyrromethene which is 17 functionalized with a mesityl group ( Ir(dipy)-1 ), an α -chloroacetyl ester ( Ir(dipy)-2 ) or a chain 18 containing an ammonium cation ( Ir(dipy)-3 ). The Ir(III) complexes feature a high triplet state 19 population enabling red phosphorescence and efficient singlet oxygen generation. Ir(dipy)-2 20 and Ir(dipy)-3 are demonstrated to stain cells in both one-photon and two-photon confocal 21 imaging. Moreover, Ir(dipy)-2 and Ir(dipy)-3 produce ROS in cells upon irradiation, inducing 22 cell death by apoptosis. Colocalization studies in SK-Mel-103 cells show that Ir(dipy)-3 23 partially accumulated in mitochondria and induces upon irradiation a disruption in their 24 morphology. Overall our studies demonstrate that the prepared Ir(III) act as photosensitizers 25 able to kill cells under irradiation, being suitable candidates for photodynamic therapy 26 applications. 27 Keywords: Dipyrromethenes, Iridium(III) complexes, photosensitizers, singlet oxygen, photodynamic 28 therapy 29
30 1. Introduction 31 Albeit the therapeutic properties of light are known long ago [1], the concept of 32 phototherapy was not developed until the end of the 19th century [2], and it was not until 33 1960 that the modern era of Photodynamic Therapy (PDT) emerged [3-5], being necessary 34 thirty more years for its clinical approval. PDT is a minimally invasive treatment that requires 35 a simultaneous combination of i) a photosensitizer (PS), ii) light of a specific wavelength, and 36 iii) molecular oxygen. These three elements are not toxic by themselves, but their 37 combination triggers a toxic effect based on the generation of reactive oxygen species (ROS), 38 which promote oxidative cellular damage and destruction [6-9]. Actually, PDT is a suitable 39 medical tool, alternative to classical therapies based on surgery, chemotherapy or 40 radiotherapy, and it has shown effectiveness for the treatment of age-related macular 41 degeneration and actinic keratosis among other disseases [6]. However, PDT is especially 42 suited for cancer treatment, since PDT enables not only the elimination of cancerous cells but 43 also the destruction of the vasculature surrounding tumour cells, and the activation of an 44 immunological response [6,9]. Besides, PDT can display a dual selectivity, i.e., preferential 45 accumulation of the photosensitizer in diseased tissues, and toxicity trigerred by light that can 46 additionally be localized to confine the damage around a selected target region, lessening 47 side-effects. In addition, PDT-based treatments can be repeated without inducing resistance. 48 However, in spite of recent advances in this field, photodynamic therapy needs further 49 development to become a first-choice treatment. Actually, PDT is applied in a very limited 50 number of cases in clinic. In this scenario, intensive research is currently ongoing to improve 51 PDT and to enhance its application. In this regard, one of the challenges is the design of new 52 PSs with improved capabilities [7-10]. 53
3 One of the strategies for the synthesis of efficient PSs consists in the heightening of the 54 population of their triplet excited state by means of heavy-atom induced intersystem 55 crossing. In this approach, halogens (bromine or iodine) or transition metals are combined 56 with suitable chromophoric cores boosting the generation of cytotoxic ROS (like singlet 57 oxygen) mediated by the key triplet manifold [10,11]. Among transition-metal complexes, 58 those based on iridium(III) have attracted attention as therapeutically active compounds, 59 hence emerging as promising alternatives to design PSs. Certain Ir(III) complexes are inert, 60 biocompatible, stable, non-toxic, feature large Stokes shifts, display long-lived red 61 phosphorescence, have outstanding color-tuning capability, good resistance to 62 photobleaching, and can interact with specific cellular targets [11-12]. In addition Ir(III) 63 complexes can exhibit strong two-photon phosphorescence and have been applied 64 successfully in biosensing and bioimaging [12,13]. 65 Another noteworthy advantage of iridium(III) complexes is the chemical versatility of their 66 molecular structure. The octahedral coordination geometry offers a large number of 67 possibilities, allowing to readily modulate their photophysical properties by means of the 68 linkage of suitable ligands. Thus, homolectic and heteroleptic Ir(III) complexes with ligands 69 based on phenylpyridines, dipyridines or phenanthrolines have been widely tested and 70 applied in different areas, such as dye-sensitized solar cells, organic light-emitting diodes 71 (OLED), photosensitizers and/or bio-imaging [14-26]. 72 Dipyrrin (or dipyrromethene) is an attractive complexating ligand owing to their easy pre73 or post-functionalization [27]. Indeed, dipyrrin has been intensively exploited in the 74 construction of supramolecular or coordination polymeric architectures [28] for a number of 75 applications [29-31]. Nevertheless, it is noteworthy that there are few studies describing 76 heteroleptic systems formed by phenylpyridine and ancillary dipyrromethene ligands [32-34] 77
and, to the best of our knowledge, hitherto there is only one report in the literature dealing 78 with their application as PDT agents [35]. 79 In this scenario, and taking into account our interest in the development of new PSs for 80 PDT [21,36-38], herein we report a new strategy to access three new heteroleptic dipyrrinato81 iridium(III) complexes (i.e. Ir(dipy)-1 , Ir(dipy)-2 and Ir(dipy) - 3 ) through a straightforward and 82 cost-effective synthetic protocol (Figure 1). The photonic performance of these metal 83 complexes is studied paying special attention to characterize their phosphorescence emission 84 (detectable at room temperature and in aerated solutions) and their ability to generate 85 singlet oxygen. The in vitro photocytotoxic effect of the iridium complexes under irradiation 86 is investigated in the human melanoma cancer cell line SK-Mel-103. Ir(dipy)-2 and Ir(dipy)-3 87 under visible-light irradiation (λ > 475 nm) result in high ROS production leading to apoptotic 88 cell death at low concentration. Moreover, Ir(dipy)-2 and Ir(dipy)-3 also demonstrate two89 photon absorption under 900 nm irradiation. Studies of colocalization in organelles of 90 Ir(dipy)-2 and Ir(dipy)-3 and cell apoptosis/necrosis assays by flow cytometry are also carried 91 out. 92 O OCl ON R2R2 R 1 N N F F F F Ir N N Ir(dipy)-2: R 1 = Ir(dipy)-3: R 1 = Ir(dipy)-1: R1 = R2 = Me ; R2 = H ; R2 = H I 93 Fig. 1. Novel PSs based on dipyrrinate-iridium(III) complexes. 94 95 2. Experimental section 96 2.1. General methods 97
5 Anhydrous solvents were prepared by distillation over standard drying agents according to 98 common methods. All other solvents were of HPLC grade and were used as provided. Starting 99 chemical substrates and reagents were used as commercially provided unless otherwise 100 indicated. Flash chromatography was performed using silica gel (230-400 mesh). NMR spectra 101 were recorded using CDCl 3 at 20 o C. 1 H NMR and 13 C NMR chemical shifts ( δ ) were referenced 102 to internal solvent CDCl 3 ( δ = 7.260 and 77.03 ppm, respectively) or acetone-d6. DEPT 135 103 experiments were used to determine the type of carbon nucleus (C vs. CH vs. CH 2 vs. CH 3 ). 104 FTIR spectra were obtained from neat samples using the ATR technique. High-resolution mass 105 spectrometry (HRMS) was performed using MALDI-TOF. 106 2.2. Synthesis and characterization 107 2,4,6-Trimethylbenzaldehyde ( 1a ) and 4-hydroxybenzaldehyde ( 1b ) were purchased from 108 Sigma-Aldrich. Intermediates dipyrromethanes 2a [39] and 2b [40], dipyrromethenes 3a [41] 109 and 3b [40], and complex [Ir(dfppy) 2 Cl] 2 [42] were synthesized by the corresponding 110 described methods. 111 2.2.1. General procedure for the synthesis of dipyrrinato-iridium(III) complexes 112 To a solution of complex [Ir(dfppy) 2 Cl] 2 (1 equiv) and dipyrromethene (2 equiv) in CH 3 CN 113 (5 mL), sodium acetate trihydrate (3 equiv) was added under an inert argon atmosphere, and 114 the reaction was refluxed until the complete disappearance of the starting material. The 115 reaction mixture was then cooled to room temperature (rt) and the solvent was evaporated 116 under vacuum. The residue was purified by column chromatography on silica gel. 117 2.2.2. Synthesis of Ir(dipy)-1 118 According to the general procedure, [Ir(dfppy)2Cl]2 (50 mg, 0.04 mmol), dipyrromethene 3a (21.6 119 mg, 0.08 mmol) and sodium acetate trihydrate (16.8 mg, 0.12 mmol) in CH3CN (5 mL) were refluxed 120
for 18 h. Flash chromatography using hexane/CH2Cl2 (90:10) afforded Ir(dipy)-1 (28 mg, 41%) as an 121 orange solid. 1H NMR (700 MHz, CDCl3) δ 8.24 (d, J = 8.4 Hz, 2H, 2CH), 7.87 (dd, J = 5.6 and 2.1 Hz, 2H, 122 2CH), 7.67 (td, J = 8.4 and 2.1 Hz, 2H, 2CH), 6.92-6.90 (m, 4H, 4CH), 6.72 (s, 2H, 2CH), 6.44 (ddd, J = 123 12.6, 9.1 and 2.8 Hz, 2H, 2CH), 6.40 (dd, J = 4.2 and 1.4 Hz, 2H, 2CH), 6.20 (dd, J = 4.2 and 1.4 Hz, 2H, 124 2CH), 5.83 (dd, J = 8.4 and 2.1 Hz, 2H, 2CH), 2.35 (s, 3H, CH3), 2.02 (s, 6H, 2CH3) ppm. 13C NMR (176 125 MHz, CDCl3) δ 165.4 (d, JCF = 7.0 Hz, C), 163.5 (dd, JCF = 255.2 and 12.0 Hz, CF), 161.4 (dd, JCF = 260.0 126 and 12.3 Hz, CF), 161.0 (d, JCF = 4.9 Hz, C), 151.5 (CH), 149.6 (CH), 147.7 (C), 137.1 (CH), 137.0 (C), 127 136.1 (C), 135.8 (C), 133.4 (C), 130.1 (CH), 128.3 (C), 127.5 (CH), 123.0 (CH), 122.8 (CH), 121.9 (CH), 128 117.4 (CH), 114.0 (d, JCF = 16.0 Hz, CH), 97.3 (t, JCF = 26.8 Hz, CH), 21.1 (CH3), 19.7 (CH3) ppm. FTIR ν 129 2923, 2854, 1601, 1547, 1476, 1403, 1376, 1344, 1291, 1248, 1162, 1106, 1026, 987, 831 cm-1. HRMS-130 MALDI-TOF m/z 834.1946 (calcd. for C40H29F4IrN4: 834.1958). 131 2.2.3. Synthesis of complex 4 132 According to the general procedure, [Ir(dfppy)2Cl]2 (50 mg, 0.04 mmol), dipyrromethene 3b (19.4 133 mg, 0.08 mmol) and sodium acetate trihydrate (16.8 mg, 0.12 mmol) in CH3CN (5 mL) was refluxed 134 for 3 h. Flash chromatography using CH2Cl2/CH3OH (97:3) afforded 4 (16 mg, 24%) as a red solid. 1H 135 NMR (700 MHz, CDCl3) δ 8.23 (d, J = 8.4 Hz, 2H, 2CH), 7.77 (d, J = 5.6 Hz, 2H, 2CH), 7.67 (td, J = 7.7 136 and 0.7 Hz, 2H, 2CH), 7.31 (d, J = 8.4 Hz, 2H, 2CH), 6.94 (td, J = 6.7 and 0.7 Hz, 2H, 2CH), 6.86 (d, J = 137 8.4 Hz, 2H, 2CH), 6.78 (s, 2H, 2CH), 6.57 (dd, J = 4.4 and 0.7 Hz, 2H, 2CH), 6.43 (ddd, J = 9.8, 9.1 and 138 2.1 Hz, 2H, 2CH), 6.26 (dd, J = 4.2 and 1.4 Hz, 2H, 2CH), 5.78 (dd, J = 9.1 and 2.1 Hz, 2H, 2CH), 4.96 (s, 139 1H, OH) ppm. 13C NMR (176 MHz, CDCl3) δ 165.3 (d, JCF = 7.0 Hz, C), 163.5 (dd, JCF = 255.6 and 12.0 140 Hz, CF), 161.3 (dd, JCF = 260.1 and 12.0 Hz, CF), 160.1 (d, JCF = 5.8 Hz, C), 155.6 (C), 151.9 (CH), 149.6 141 (CH), 148.5 (C), 137.1 (CH), 134.4 (C), 132.0 (CH), 131.6 (CH), 128.3 (C), 122.9 (CH), 122.8 (CH), 122.1 142 (CH), 117.2 (CH), 114.0 (CH), 113.9 (d, JCF = 16.7 Hz, CH), 97.3 (t, JCF = 26.4 Hz, CH) ppm. FTIR ν 3358, 143
7 2925, 2852, 1601, 1543, 1476, 1405, 1378, 1344, 1290, 1246, 1200, 1106, 1031, 989, 819 cm-1. HRMS-144 MALDI-TOF m/z 808.1428 (calcd. for C37H23F4IrN4O: 808.1437). 145 2.2.4. Synthesis of Ir(dipy)-2 146 To a solution of 4 (27 mg, 0.03 mmol) and Et3N (5 drops) in CH2Cl2 (5 mL), chloroacetyl chloride 147 (0.004 mL, 0.05 mmol) was added under inert atmosphere of argon, and the reaction mixture was 148 stirred at 50 oC for 2 h. The reaction was then cooled to rt and the solvent was evaporated under 149 vacuum. The residue was purified by column chromatography on silica gel (hexane/CH2Cl2, 30:70) to 150 afford Ir(dipy)-2 (21 mg, 71%) as an orange-red solid. 1H NMR (700 MHz, CDCl3) δ 8.24 (d, J = 9.1 Hz, 151 2H, 2CH), 7.79 (d, J = 5.6 Hz, 2H, 2CH), 7.68 (td, J = 8.4 and 0.7 Hz, 2H, 2CH), 7.46 (d, J = 8.4 Hz, 2H, 152 2CH), 7.20 (d, J = 8.4 Hz, 2H, 2CH), 6.94 (td, J = 7.7 and 0.7 Hz, 2H, 2CH), 6.80 (s, 2H, 2CH), 6.52 (d, J = 153 4.2 Hz, 2H, 2CH), 6.46-6.44 (m, 2H, 2CH), 6.26 (dd, J = 4.2 and 0.7 Hz, 2H, 2CH), 5.78 (dd, J = 8.4 and 154 2.1 Hz, 2H, 2CH), 4.35 (s, 2H, CH2) ppm. 13C NMR (176 MHz, CDCl3) δ 165.7 (COO), 165.3 (d, JCF = 7.0 155 Hz, C), 163.5 (dd, JCF = 255.9 and 12.1 Hz, CF), 161.3 (dd, JCF = 259.4 and 12.5 Hz, CF), 160.6 (d, JCF = 156 6.2 Hz, C), 152.4 (CH), 150.2 (C), 149.5 (CH), 147.2 (C), 137.4 (C), 137.2 (CH), 134.0 (C), 131.6 (CH), 157 128.3 (C), 123.0 (CH), 122.8 (CH), 122.2 (CH), 119.9 (CH), 117.6 (CH), 114.0 (d, JCF = 15.8 Hz, CH), 97.4 158 (t, JCF = 26.9 Hz, CH), 40.9 (CH2) ppm. FTIR ν 2925, 2855, 1778, 1601, 1545, 1474, 1404, 1378, 1344, 159 1286, 1247, 1201, 1165, 1138, 1105, 1030, 989, 818 cm-1. HRMS-MALDI-TOF m/z 808.1138 (calcd. for 160 C39H24ClF4IrN4O2: 884.1153). 161 2.2.5. Synthesis of Ir(dipy)-3 162 A solution of 4 (16 mg, 0.02 mmol) and K2CO3 (8.2 mg, 0.06 mmol) in acetone (5 mL) was refluxed 163 under inert atmosphere of argon for 2 h. The reaction was cooled to rt, and then 2-chloro-N,N-164 dimethylethanamine hydrochloride (3.4 mg, 0.024 mmol) was added. The reaction mixture was 165 refluxed for 16 h, cooled to rt and filtered for remove inorganic salts. The solvent was evaporated 166
under vacuum and the residue was purified by column chromatography on silica gel (CH2Cl2/CH3OH, 167 98:2) to afford complex 5 (13 mg, 73%) as an orange-red solid. 1H NMR (300 MHz, CDCl3) δ 8.23 (d, J 168 = 8.4 Hz, 2H, 2CH), 7.77 (dd, J = 6.0 and 0.3 Hz, 2H, 2CH), 7.70-7.64 (m, 2H, 2CH), 7.35 (d, J = 8.7 Hz, 169 2H, 2CH), 6.97-6.93 (m, 4H, 4CH), 6.78 (t, J = 1.5 Hz, 2H, 2CH), 6.56 (dd, J = 4.2 and 1.5 Hz, 2H, 2CH), 170 6.47-6.39 (m, 2H, 2CH), 6.26 (dd, J = 4.2 and 1.5 Hz, 2H, 2CH), 5.77 (dd, J = 8.4 and 2.4 Hz, 2H, 2CH), 171 4.19 (t, J = 5.7 Hz, 2H, CH2O), 2.89 (t, J = 5.4 Hz, 2H, CH2N), 2.49 (s, 6H, 2CH3) ppm. 172 A solution of 5 (12.7 mg, 0.014 mmol) and iodomethane (1.0 mL) in CHCl3 (1.5 mL) was stirred 173 under inert atmosphere of argon at rt for 1 h. Et2O was added for precipitation. The precipitate was 174 filtered, then washed with Et2O and dried under vacuum to afford Ir(dipy)-3 (10 mg, 65%) as an 175 orange-red solid. 1H NMR (700 MHz, CDCl3) δ 8.31 (d, J = 8.4 Hz, 2H, 2CH), 7.97-7.93 (m, 4H, 4CH), 176 7.42 (d, J = 8.4 Hz, 2H, 2CH), 7.25 (td, J = 6.3 and 1.4 Hz, 2H, 2CH), 7.16 (d, J = 8.4 Hz, 2H, 2CH), 6.80 177 (t, J = 1.4 Hz, 2H, 2CH), 6.60 (td, J = 9.1 and 2.1 Hz, 2H, 2CH), 6.50 (dd, J = 4.9 and 1.4 Hz, 2H, 2CH), 178 6.29 (dd, J = 4.9 and 1.4 Hz, 2H, 2CH), 5.81 (dd, J = 8.4 Hz and 2.1 Hz, 2H, 2CH), 4.79-4.78 (m, 2H, 179 CH2O), 4.24 (t, J = 4.9 Hz, 2H, CH2N), 3.61 (s, 9H, 3CH3) ppm. 13C NMR (176 MHz, CDCl3) δ 164.8 (d, JCF 180 = 6.7 Hz, C), 163.4 (dd, JCF = 255.2 and 12.1 Hz, CF), 161.4 (d, JCF = 5.6 Hz, C), 161.3 (dd, JCF = 258.7 and 181 12.8 Hz, CF), 157.9 (C), 151.5 (CH), 149.8 (CH), 148.5 (C), 138.2 (CH), 134.3 (C), 132.6 (C), 131.8 (CH), 182 131.5 (CH), 128.5 (C), 123.0 (CH), 122.9 (CH), 122.8 (CH), 117.4 (CH), 113.6 (d, JCF = 15.8 Hz, CH), 113.4 183 (CH), 97.1 (t, JCF = 26.9 Hz, CH), 65.3 (CH2N), 62.3 (CH2O), 54.0 (CH3) ppm. FTIR ν 1603, 1542, 1477, 184 1404, 1378, 1345, 1247, 1201, 1030, 989 cm-1. HRMS-MALDI-TOF m/z 894.2403 (calcd. for 185 C42H35F4IrN5O+: 894.2401). 186 2.3. Photophysical properties and singlet oxygen generation 187 The photophysical properties were registered using quartz cuvettes with optical pathways 188 of 1 cm in diluted solutions (around 2·10 -6 M), prepared by adding the corresponding solvent 189
15 dipyrromethene derivatives) to complete the Ir(III) metal coordination sphere (Fig. 1). The 328 electronegativity of the fluorine atoms further stabilizes the chelation of the metallic cation 329 by phenylpyridine. This election of fluorinated ligands is backed up by our own previous 330 experience [21], as well as by the results reported by other research groups, which support 331 the viability of iridium-(III) complexes with flurophenylpyridine based ligands as suitable 332 agents for PDT and phototheragnosis [46-48]. On the other hand, dipyrromethene is a key 333 chelating ligand in our design since diferent functional groups can be appended at its meso 334 position to enhance the photonic performance of the metal complex and/or guide the 335 complex towards target cells or organelles [27,29,35]. Accordingly, we designed the Ir(III) 336 complexes Ir(dipy)-1 , Ir(dipy)2 and Ir(dipy)-3 (Fig.1) containing a mesityl group at the meso337 position of the dipyrrinato ligand ( Ir(dipy)-1 ), an α-chloroacetyl ester ( Ir(dipy)-2 ) or a chain 338 containing an ammonium cation ( Ir(dipy)-3 ). The mesityl group has been incorporated to 339 enhance the photostability [49,50], the α-chloroacetyl goup has been reported to react with thiol 340 groups on proteins [51], whereas ammonium derivatives can accumulate in mitochondria through 341 electrostatic interaction with the negative mitochondrial membrane [52]. 342 3.2. Synthesis 343 The synthetic routes followed to prepare the dipyrromethene-iridium(III) complexes are 344 outlined in Scheme 1. Following a usual synthetic procedure for dipyrromethenes, aldehydes 345 1a or 1b and pyrrole were reacted in acid medium to obtain dipyrromethanes 2a [39] and 2b 346 [40], which were further oxidized with DDQ to dipyrromethenes 3a [41] and 3b [40]. By a 347 complexation reaction following a representative synthetic procedure [33] between 3a or 3b 348 and [Ir(dfppy) 2 Cl] 2 [42], the complex Ir(dipy)-1 and the intermediate complex 4 were 349 obtained, respectively, in 41 and 24% yield. From this point, the binding of 4 with chloroacetyl 350
chloride in the presence of a base (Et 3 N) afforded Ir(dipy)-2 in 71% yield, while Ir(dipy)-3 was 351 obtained in two-steps from 4 that first was reacted with 2-chloro-N,N-dimethylethanamine in 352 a basic medium, to obtain 5 (73%) that was then treated with an excess iodomethane resulting 353 in Ir(dipy)-3 in 65% yield. All the complexes were fully charecterized by 1 H and 13 C NMR 354 spectroscopy (see ESI), FTIR and HRMS-MALDI-TOF. 355 R1 H O +N H NH HN R1 N HN R1 i ii 3a [Ir(dfppy)2Cl]2 iii R2R2 1a, 2a, 3a: R1 = R2 = Me 1b, 2b, 3b: R1 = OH; R2 = H 1 2 3 R2R2R2R2 iv N N Ir F F F F Cl Cl N N Ir F F F F N O N N F F F F Ir N N OH N N F F F F Ir N N iii 3b Ir(dipy)-1 4 Ir(dipy)-2 Ir(dipy)-3 5 vi v 356 Scheme 1. Synthesis of Ir(dipy)-1 , Ir(dipy)-2 and Ir(dipy)-3 . Reaction conditions: i) 1a , H 2 O/HCl (98.5:1.5), rt, 12 357 h; 1b , TFA, rt, 12 h. ii) DDQ, CH 2 Cl 2 , rt, 30 min. iii) sodium acetate trihydrate, CH 3 CN, reflux, 3-18 h. iv) α358 chloroacetyl chloride, Et 3 N, CH 2 Cl 2 , reflux, 2 h. v) 2-chloro-N,N-dimethylethanamine, K 2 CO 3 , acetone, reflux, 18 359 h. vi) iodomethane, CHCl 3 , rt, 1 h. 360 361 3.3. Photophysical Properties 362 The spectroscopic signatures of the three Ir(III)-based organometallic complexes are very 363 similar, being almost independent of the para-functionalization. Thus, the absorption profile 364 of Ir(dipy)-1 , Ir(dipy)-2 and Ir(dipy)-3 featured a band at ca. 475-485 nm with molar 365 absorptions up to 30000 M -1 cm -1 (Fig. 2). The profile and position of the absorption bands 366 resemble those recorded for the isolated dipyrromethene core. Besides, Ir(dipy)-1 , Ir(dipy)-2 367
17 and Ir(dipy)-3 shows a low negative solvatochromism; a trademark of dipyrromethene 368 delocalized π-systems (Fig. S1). Indeed, theoretical simulations predicted that the electronic 369 density of the molecular orbitals involved in such electronic transitions is exclusively located 370 at the dipyrromethene framework (Fig. 2 and S2). Optimized geometries resulted in an 371 expected iridium-centered octahedral conformation, where the dipyrrromethene ligand 372 adopts a planar geometry owing to the rigid coordination afforded by the iridium chelation 373 to the pyrrolic nitrogens. The meso-aryl moiety is twisted around 70 o with regard to the 374 dipyrromethene plane, reaching an orthogonal arrangement for the sterically hindered 375 mesityl in Ir(dipy)-1 . The fluorinated phenylpyridine ligands at the iridium center are axially 376 disposed far away from each other and do not have contact with the dipyrromethene. 377 The photoluminescence spectra of Ir(dipy)-1 , Ir(dipy)-2 and Ir(dipy)-3 displayed a residual 378 and almost negligible short wavelength emission (530-550 nm), followed by a broad and 379 dominant long-wavelength emission (670-675 nm), which endows a large Stokes shift (6000 380 cm -1 ) and falls within the biological window (Fig. 2 and S1). The first emission likely owes to 381 the strongly quenched fluorescence emission from the locally excited state of the 382 dipyrromethene group, while the second one is identified as a phosphorescence signal. The 383 low fluorescence from these compounds can be rationalized by the enhanced intersystem 384 crossing induced by the heavy atom effect promoted by the iridium chelating the ancillary 385 dipyrromethene, and mainly by the ability of these kind of cyclometalated compounds to 386 induce metal-ligand charge transfer (MLCT) and ligand-ligand charge transfer (LLCT) 387 processes. Considering such CT phenomena, the second and main long-wavelength emission 388 could be tentatively assigned to their own emission at a first sight. However, the recorded 389 emission is almost insensitive to the solvent polarity. Note that the wavelength and 390 fluorescence intensity of CT-like emissions are expected to markedly depend on the 391
environment, owing to the generated charge separation in such state, and this effect is not 392 observed for Ir(dipy)-1 , Ir(dipy)-2 nor Ir(dipy)-3 (see photoluminescence spectra in Fig. S1). 393 The lifetime associated to such long-wavelength emission was very long (hundreds of 394 nanoseconds, Table 1). Moreover, both the emission intensity and the lifetime (up to tens of 395 microseconds, Table 1) greatly enhanced in deaerated (oxygen-free) solutions (Fig. S3 and Fig. 396 S4, repectively) and when cooling the samples down to 77 K (Fig. S5 and Fig. 2, respectively). 397 Indeed, the time resolved emission spectra under an argon atmosphere recorded for Ir(dipy)- 398 1 , display such long-lived emission detectable even after delay times of tens of microseconds 399 (Fig. S6). In sumary, all these findings support that the long-lived triplet state is involved in 400 the recorded emission and hence pinpoint to phosphorescence rather than fluorescence as 401 the source of the red-emission (Fig. 2). A quick survey to the bibliography reveals that related 402 cyclometalated iridium complexes are able to display phosphorescence even under adverse 403 conditions, such as room temperature and aerated solutions [22,29]. Indeed, the emissive 404 triplet states populated from those MLCT and LLCT are suggested as the source of such weak 405 (efficiency lower than 0.3% in aerated solutions, increasing up to 8% upon removal of oxygen, 406 Table 1) red emissions [53]. In other words, if these compounds are able to stain cells (vide 407 infra), the tracking would be visualized thanks to a long-lived red phosphorescence. 408 Table 1. Photophysical properties of the cyclometalated iridium-based organometallic complexes in diluted 409 aerated solution of two representative solvents at room temperature; chloroform (CHCl 3 ) and acetonitrile (ACN). 410 See table S1 for photophysical data in more solvents. The photoluminescence quantum yield and lifetime in 411 argon-purged solutions of chloroform are also added in italics. 412 413 λ ab (nm) ε max (10 4 M -1 cm -1 ) λ pl (nm) ∆ν St (cm -1 ) φ τ (µs) φ ∆ Ir(dipy)-1 CHCl 3 484.0 3.4 675.0 5840 0.0025 0.085 0.475 22.3 0.92 ACN 480.0 3.1 676.0 6050 0.0014 0.327 0.92 Ir(dipy)-2 CHCl 3 482.5 3.8 675.0 5950 0.0023 0.070 0.375 10.4 0.75 ACN 478.5 3.3 668.0 5970 0.0015 0.215 1.00
19 Ir(dipy)-3 CHCl 3 483.0 2.8 676.0 5920 0.0025 0.045 0.376 15.3 0.84 ACN 479.0 2.6 673.0 6120 0.0013 0.286 0.94 414 Absorption ( λ ab ) and photoluminescence ( λ pl ) wavelength, molar absorption ( ε max ), Stokes shift ( ∆ν St ), 415 photoluminescence quantum yield ( φ ) and lifetime ( τ ), singlet oxygen generation quantum yield ( φ ∆ ). 416 417 418 Fig. 2. Absorption and photoluminescence spectra of the cyclometalated iridium-based complexes in diluted 419 solutions of chloroform. The theoretically simulated (wb97xd/lanl2dz) optimized ground state geometry and the 420 corresponding molecular orbitals involved in the main visible absorption transition are also depicted for the 421 representative organometallic compound Ir(dipy)-2 . See Fig. S1 for spectra in more solvents and Fig. S2 for the 422 optimized geometries and molecular orbitals of the rest of organometallic compounds. Inset: 423 Photoluminescence decay curve for representative Ir(dipy)-1 at room temperature and liquid nitrogen cooled 424 in ethanol. 425 426 This high triplet state population, allowing even the recording of phosphorescence under 427 room conditions, paves the way for an efficient singlet oxygen generation. Indeed, all 428 compounds display a strong phosphorescence emission from singlet oxygen in the near 429 infrared regardless of the solvent (Table 1 and S1). Thus, the three compounds in all the tested 430 media yield high efficiencies of singlet oxygen generation (higher than 70%), which reach 431 100% in many of them (Table 1 and S1). Therefore, these iridium complexes are expected to 432 behave as optimal photosensitizers to kill efficiently cells under aerobic conditions (vide infra). 433 Furthermore, the largely red-shifted emission of the compounds with respect to their light434
absorption region (Stokes shift around 6000 cm -1 , Table 1) would allow sensing and 435 bioimaging with strongly reduced background interferences (vide infra). 436 437 3.4. In vitro PDT experiments 438 PDT with complexes Ir(dipy)-1 , Ir(dipy)-2 and Ir(dipy)-3 were investigated against the 439 human melanoma cell line SK-Mel-103 using the WST-1 assay. SK-Mel-103 cells were treated 440 with increasing doses of the iridium complexes for 24 h and then irradiated for 30 min with a 441 36W LED source at 10 cm of distance through a filter that allowed passage of light of 442 wavelengths > 475 nm. The obtained results are depicted in Figure 3 that shows the 443 photoinduced cytotoxicities in SK-Mel-103 cancer cells treated with Ir(dipy)-2 and Ir(dipy)-3 . 444 Moreover, Ir(dipy)-1 up to 50 µM was not phototoxic in the tested conditions (Fig. S7) and 445 was not considered for further PDT studies. 446 447 Fig. 3. Photoinduced cytotoxicities in SK-Mel-103 cancer cells with iridium(III) complexes, Ir(dipy)-2 (A) and 448 Ir(dipy)-3 (B). Treatment was applied at increasing concentration up to 2 µM for 24 h before irradiation with 449 visible-light (λ>475) for 30 min. Cell viability was determined 24 h after irradiation step and in dark with WST-1 450 method. Values are expressed as mean ± SEM, and statistical significance was assessed by one-way ANOVA and 451 Tukey's post-test (n=18 data points). ** <0.010; **** <0.0001 indicate statistically significant changes. 452 453
21 As shown in Figure 3, after 24 h of treatment with PS Ir(dipy)-2 and Ir(dipy)-3 a fall in 454 cultures viability in a concentration-dependent way was found. These reductions in cell 455 viability are in agreement with the high singlet oxygen generation quantum yield observed 456 for these iridium complexes (Table 1). In contrast Ir(dipy)-1 , also displaying a high singlet 457 oxygen generation, was not able to induce cell death upon irradiation. The lack of 458 photocitotoxicity induced by Ir(dipy)-1 , along with the lack of efficiency to stain cells under 459 confocal imaging, suggest that the presence of the mesityl group at the meso-position of the 460 dipyrrinato ligand prevents the internalization of the compound in cells, and therefore no PDT 461 effect was observed. Several reasons can account for this unexpected finding. On the one 462 hand, the sterical hindrance of the mesityl leads to a bulkier molecule (see its orthogonal 463 disposition in Fig. S2) that might hamper its internalization. On the other hand, the absence 464 of polar/ionic targetable groups (as ester or ammonium in Ir(dipy)-2 or -3 , respectively) makes 465 Ir(dipy)-1 more hydrophobic and with less probability to interact with cells, hindering the 466 uptake. 467 Table S2 shows EC 50 values determined for SK-Mel-103 cells with Ir(dipy)-2 and Ir(dipy)- 468 3 . EC 50 values, calculated from the sigmoidal fitting of concentration-cell viability curves for 469 Ir(dipy)-2 and Ir(dipy)-3 (Fig. S8) were 0.11 µM and 0.39 µM, respectively, indicating that 470 compound Ir(dipy)-2 is more effective killing cells than Ir(dipy)-3 under irradiation conditions. 471 Besides, the intrinsic cytotoxicity of Ir(dipy)-2 and Ir(dipy)-3 was also evaluated in the absence 472 of light irradiation (dark control). Althought Ir(dipy)-2 and Ir(dipy)-3 show slightly dark 473 cytotoxicity, the phototoxicity index (PI) (PI = EC50dark/EC50light) for Ir(dipy)-2 and Ir(dipy)-3 is 45.45 474 and 23.7, respectively, which demonstrates their potential application for PDT as they effectively kill 475 cells by photoactivation. 476
477 3.5. Cellular uptake and organelles accumulation 478 The intracellular distribution of PSs has an important role in PDT outcome, as the 479 localization determines the mechanism of cell death and the cellular response to 480 photodamage [54]. Cellular uptake of compounds Ir(dipy)-2 and Ir(dipy)-3 in SK-Mel-103 cells 481 were demonstrated by one-photon confocal imaging (Fig. 4, A-D) for which a clear 482 cytoplasmic signal was observed after 24 h incubation. Ir(dipy)-2 and Ir(dipy)-3 also displayed 483 red phosphoresce under two-photon excitation (TPE) when irradiated at 900 nm (Fig. 4, E-H). 484 485 Fig. 4. Cellular uptake of compounds Ir(dipy)-2 and Ir(dipy)-3 in SK-Mel-103 cells. (i) (A-D) One-photon confocal 486 images of SK-Mel-103 in the absence or presence of 2.5 µM of Ir(dipy)-2 (A, B), respectively, and in the absence 487 or presence of 2.5 µM Ir(dipy)-3 (C, D), respectively. One-photon images were acquired by using one-photon 488 confocal microscope (Leica TCS SP8 HyVolution II). Excitation at 488 nm and emission collected at 640-780 nm. 489 (E-H) Two-photon confocal images of SK-Mel-103 in the absence or presence of 2.5 µM of Ir(dipy)-2 (E, F), 490 respectively, and in the absence or presence of 2.5 µM Ir(dipy)-3 (G, H), respectively. Two-photon images were 491 acquired by using multiphoton confocal microscope (Olympus FV1000MPE). Excitation at 900 nm and emission 492 was collected in the spectra of 650-740 nm. (ii) Quantification of the integrated fluorescence intensity relative 493 to cell number of SK-Mel-103 cells in the absence or presence of Ir(dipy)-2 and Ir(dipy)-3 with (A) one-photon 494 confocal imaging and (B) two-photon confocal imaging. Values are expressed as mean ± SEM, and statistical 495 significance was assessed by the two-tailed Student’s t-test. * <0.05; *** p <0.001. 496 497 Once demonstrated the internalization of the Ir(III) complexes, we studied more in detail 498 the preferential localization of Ir(dipy)-2 and Ir(dipy)-3 in the cells. The subcellular localization of 499
23 these iridium complexes was studied in SK-Mel-103 cells by confocal microscopy using commercially 500 available fluorescent markers of subcellular organelles (including nuclei, mitochondria, lysosomes 501 and endoplasmic reticulum) following the manufacturer recommendations. Studies were performed 502 with Ir(dipy)-3 as the results obtained in confocal for Ir(dipy)-2 were too weak to perform a 503 correct colocalization assay (Fig. S9). 504 Fig. 5 shows the subcellular colocalization of Ir(dipy)-3 (red) in SK-Mel-103 stained with 505 organelle markers (green). Colocalization analysis were performed on a pixel-by-pixel based analysis. 506 The red emission of Ir(dipy)-3 colocalized with the green fluorescence of the trackers for 507 mitochondria (Fig. 5B), for lysosome (Fig. 5C), and to a lesser extent for endoplasmic reticulum 508 (Fig. 5D). From these data, it can not be concluded a specific colocalization with any of the 509 organelles. Moreover, it is noteworthy that compared to the only Mitrotracker control (Fig 510 5A), a change in the morphology of the mitochondria after treatment with Ir(dipy)-3 was 511 clearly observed. A similar behavior was also found for Ir(dipy)-2 (Fig. S9). This effect was 512 effective even at low concentrations of Ir(dipy)-3 , although in this case the mitochondria was 513 less affected than at high concentration of the iridium complex (Fig. S10). All this data 514 suggested that there is in fact an accumulation of Ir(dipy)-3 in mitochondria which promotes 515 the destruction of their native morphology in a dose-dependent manner. This is in agreement 516 with previously reported data that suggested that fluorophores containing a chain-end 517 ammonium cation can have preference to target mitochondria [55]. Targeting mitochondria is 518 a desirable property to enhance PDT as this organelle is the cell energy supplier and a suitable 519 target to overcome resistance to apoptosis in anticancer therapies [52]. Moreover, 520 mitochondria play a central role in malignant tumor progression, and its biogenesis is many 521 times upregulated in cancers. 522
523 524 Fig. 5. Subcellular colocalization of Ir(dipy)-3 . SK-Mel-103 following 24 h incubation with Ir(dipy)-3 (red) at 2.5 525 µM and stained with the green trackers Mitotracker (B), Lysotracker (C), and ER-tracker (D). Images were 526 acquired by using one-photon confocal microscope (Leica TCS SP8 HyVolution II). Ir(dipy)-3 was excited at 488 527 nm and the emission was collected at 640-780 nm. Green trackers were excited at 488 and the emission was 528 collected at 490-530 nm. The 2D scatterplot diagram represents the degree of colocalization. Scale bar: 10 µm. 529 530 3.6. ROS generation in cells and cell death mechanism 531 It is well-known that PDT processes involve several steps in which there is an excited 532 triplet state (T 1 ) in the photosensitizer that is able to react with molecular oxygen resulting in 533 the formation of (ROS). This reaction happens through two mechanisms; one induces the 534 generation of superoxide anion, hydrogen peroxide and hydroxyl radicals (Type I mechanism), 535 whereas the other generates singlet oxygen ( 1 O 2 in Type II). It is mainly the generation of 536
31 [53] Langdon-Jones EE, Hallet, AJ, Routledge JD, Crole DA, Ward BD, Platts JA, Pope SJA. Using substituted cyclometalated quinoxaline 724 ligands to finely tune the luminescence properties of iridium(III) complexes. Inorg Chem 2013;52:448-456. 725 [54] Lv W, Zhang Z, Zhang KY, Yang H, Liu S, Xu A, Guo S, Zhao Q, Huang W. A Mitochondria-Targeted Photosensitizer Showing 726 Improved Photodynamic Therapy Effects Under Hypoxia. Angew Chem Int Ed Engl 2016;55: 9947-9951. 727 [55] Xu Z, Xu L. Fluorescent probes for the selective detection of chemical species inside mitochondria. Chem Commun 2016,52:1094-728 1119. 729 [56] Ye R-R, Tan C-P, He L, Chen M-H, Ji L-N, Mao Z-W. Cyclometalated Ir(III) complexes as targeted theranostic anticancer therapeutics: 730 combining HDAC inhibition with photodynamic therapy. Chem Commun (Camb). 2014;50:10945-10948. 731 [57] Nam JS, Kang M-G, Kang J, et al. Endoplasmic Reticulum-Localized Iridium(III) Complexes as Efficient Photodynamic Therapy 732 Agents via Protein Modifications. J Am Chem Soc. 2016;138:10968-10977. 733 [58] Wang C, Lystrom L, Yin H, et al. Increasing the triplet lifetime and extending the ground-state absorption of biscyclometalated 734 Ir(iii) complexes for reverse saturable absorption and photodynamic therapy applications. Dalton Trans. 2016;45:16366-16378. 735 [59] McKenzie LK, Sazanovich I V, Baggaley E, et al. Metal Complexes for Two-Photon Photodynamic Therapy: A Cyclometallated Iridium 736 Complex Induces Two-Photon Photosensitization of Cancer Cells under Near-IR Light. Chemistry. 2017;23:234-238. 737 738 796