Synthesis, chemical characterization, PARP inhibition, DNA binding and cellular uptake of novel ruthenium(II)-arene complexes bearing benzamide derivatives in human breast cancer cells
Abstract
Inhibitors of poly(ADP-ribose) polymerase-1 have demonstrated significant clinical effectiveness in tumors with BRCA mutations. Building upon rational drug design principles, derivatives of the PARP inhibitor 3-aminobenzamide, specifically 2-amino-4-methylbenzamide and 3-amino-N-methylbenzamide, were coordinated with the ruthenium(II) ion to create potential therapeutic agents capable of affecting DNA and inhibiting the PARP enzyme. Four conjugated complexes, denoted as C1, C2, C3, and C4, with defined chemical formulas, were synthesized and characterized. The MTT assay revealed that C1 exhibited the highest antiproliferative activity across HCC1937, MDA-MB-231, and MCF-7 breast cancer cell lines. The efficiency of in vitro PARP-1 enzymatic activity inhibition followed a decreasing order: C2, C4, 3-AB, C1, and C3. Inductively coupled plasma mass spectrometry analysis of intracellular accumulation and distribution in BRCA1-mutated HCC1937 cells indicated that all four complexes entered the cells within a 24-hour period. Complex C1 displayed the highest intracellular accumulation, exhibited nuclear-targeting properties, and demonstrated the strongest DNA binding, leading to cell cycle arrest in the S phase.
Keywords: antitumor agents; breast cancer; PARP inhibitor; ruthenium(II)
Introduction
The escalating global incidence of cancer-related deaths underscores the critical importance of drug development research. It is now well-established that the development of novel antitumor drugs intended to supersede cisplatin must overcome the severe adverse effects and the accompanying resistance that cisplatin induces. Ruthenium-based compounds represent a highly promising class of metal-based anticancer drug candidates. The synthetic chemistry of ruthenium complexes is well-developed, allowing for the creation of a wide array of compounds. Among the ruthenium complexes synthesized to date, ruthenium(III) complexes such as NAMI-A, which targets metastases in solid tumors, and KP1339, effective against platinum-resistant tumors, have completed phase II clinical trials. Organometallic ruthenium(II)-arene compounds, featuring a distinct metallodrug structure where three coordination sites are occupied by an η6-coordinated arene that stabilizes the Ru+2 oxidation state, are under extensive investigation. Compounds like RAPTA-C, RAED-C, and RAPTA-T have shown efficacy in reducing the number and weight of lung metastases. To date, numerous structurally diverse ruthenium(II)-arene complexes have been synthesized, with some demonstrating notable anticancer activities in laboratory settings. Ruthenium(II) exhibits ligand exchange kinetics similar to platinum(II), and the octahedral geometry of ruthenium complexes offers unique possibilities for binding to nucleic acids.
Ongoing research into the mechanisms of action of cisplatin and ruthenium-based drugs highlights the presence of diverse cellular targets beyond nuclear DNA for these agents. There is particular interest in investigating enzyme inhibition by metal complexes. In this context, Kilpin and Dyson proposed a classification of metal-based enzyme inhibitors based on whether the activity of the metal complexes primarily depends on the chemical properties of the metal center, the bioactive ligand, or a combination of both.
Poly(ADP-ribose) polymerases, a particularly interesting family of enzymes found in eukaryotes, are involved in various cellular processes, including DNA repair, chromatin remodeling, transcriptional regulation, and cell death. PARP-1 is one of the most abundant chromatin-bound proteins, accounting for over 90% of the total PARP activity. It is considered a key player in base excision repair and the repair of single-strand DNA breaks in response to ionizing radiation, oxidative stress, or DNA-binding agents. Structurally, PARP-1 consists of three distinct functional domains: an N-terminal DNA-binding domain containing three zinc fingers crucial for PARP-1 binding to DNA breaks, a central automodification domain that serves as an acceptor of ADP-ribose moieties, and a C-terminal catalytic domain forming the active site of PARP-1, which transfers ADP-ribose subunits from NAD+ to protein acceptors.
PARP-1 utilizes beta nicotinamide adenine dinucleotide as a substrate to covalently add poly(ADP-ribose) chains onto itself and other nuclear acceptor proteins, a process known as PARylation. The role of PARP-1 in DNA damage response and cell death regulation spurred the development of potent small molecules called PARP inhibitors. Generally, PARP inhibitors are classified as competitive inhibitors because they impede the catalytic activity of PARP-1 by interacting with the NAD+ binding site. Nicotinamide, benzamide, and 3-aminobenzamide were identified as the first generation of classical PARP inhibitors. Despite their limited potency and specificity for clinical use, these compounds are significant for research purposes. Structure-activity relationship analysis has indicated that potent PARP inhibitors should possess an electron-rich aromatic or polyaromatic heterocyclic system, featuring a pharmacophore with a cis-configured carboxamide, imide, or carbamoyl group. To date, four third-generation PARP inhibitors, namely olaparib, rucaparib, niraparib, and talazoparib, have been approved by the United States Food and Drug Administration as single-agent therapies for targeting breast cancer susceptibility gene-mutated breast, ovarian, prostate, and pancreatic cancers. PARP inhibitors are the first clinically approved drugs designed to exploit the concept of synthetic lethality in tumors harboring mutations in BRCA1 or BRCA2 genes, which are responsible for the repair of double-strand DNA breaks by homologous recombination. Synthetic lethality is a genetic concept based on the principle that a defect in either one of two genes has minimal impact on the cell or organism, but a combination of defects in both genes leads to cell death. Cells with impaired homologous recombination pathways rely on alternative DNA repair mechanisms for survival, thus making PARP inhibitors a promising therapeutic approach for BRCA-mutated cancers.
Although recent studies have suggested that the application of PARP inhibitors could be extended to tumors exhibiting BRCAness, characterized by defects in homologous recombination repair genes other than BRCA1 and BRCA2, the acquired resistance resulting from the development of secondary mutations in homologous recombination repair genes that restore the function of the homologous recombination repair pathway limits the clinical efficacy of PARP inhibitors when used as single agents. Therefore, combination therapy represents a rational strategy to enhance the utility of these inhibitors as chemopotentiators of commonly used cytotoxic chemotherapeutic agents, such as cisplatin. Unfortunately, multicomponent drug combinations can exhibit adverse effects due to complex pharmacokinetics and unpredictable drug-drug interactions. Consequently, multi-targeting single drugs offer potential advantages in terms of pharmacokinetic simplicity and improved outcomes. A growing trend in drug design involves linking metals with pharmacophoric moieties of bioactive ligands to achieve diverse spectra of biological activity and enhance the properties of both components. Existing research has shown that various ruthenium(II)-arene complexes efficiently enter tumor cells and bind to DNA.
Therefore, these complexes may provide a suitable framework for delivering bioactive ligands with PARP-1 inhibitory potential more directly to their cellular targets. Furthermore, a study by Mendes and colleagues demonstrated that compounds based on platinum, ruthenium, or gold can exhibit high PARP-1 inhibitory activity, supporting the hypothesis that the displacement of zinc from the zinc finger motif of PARP-1 by metal ions leads to reduced PARP-1 activity.
In the present study, novel ruthenium(II)-arene complexes bearing 3-AB derivatives as ligands were synthesized and evaluated for their inhibitory potential against the catalytic activity of PARP-1. Additionally, their growth inhibitory effects were investigated in a panel of human breast cancer cell lines, including BRCA1-mutant, triple-negative, and hormone-responsive types. To elucidate the mechanism of action and identify intracellular targets, we analyzed cell cycle progression, DNA binding, cellular uptake, and distribution across cellular compartments in treated cells. The influence of specific structural modifications in the molecules on various aspects of their anticancer potential is discussed.
Experimental Section
Material and Methods
Ruthenium(III) chloride trihydrate was obtained from Johnson Matthey. The compounds 2-amino-4-methylbenzamide and 3-amino-N-methylbenzamide were purchased from Sigma Aldrich. The compound with the formula [Ru(η6-p-cymene)Cl2]2 was synthesized following a previously reported procedure. Similarly, the compound with the formula [Ru(η6-toluene)Cl2]2 was prepared according to a published method. All solvents used were obtained commercially and used directly without further purification. Infrared spectra were recorded using a Nicolet 6700 FTIR spectrometer with the ATR technique. Proton and carbon-13 nuclear magnetic resonance spectra were acquired on a Bruker Avance III 500 spectrometer. Chemical shifts for proton and carbon-13 spectra were referenced to the residual proton and carbon-13 signals present in deuterated dimethyl sulfoxide and deuterated methanol. Conductivity measurements were performed using a CrisonMultimeter MM 41 instrument. The conductivities of the synthesized complexes were determined using 1 mM solutions in dimethyl sulfoxide. Electrospray ionization mass spectra were obtained using a mass spectrometer (Bruker, Model Esquire 3000), employing a mixture of acetonitrile and methanol as the solvent. Elemental analyses were carried out at the Microanalytical Service of the Faculty of Chemistry of the University of Vienna.
Synthesis of Complexes
Synthesis of [(η6-toluene)Ru(L1)Cl]PF6 (C1). A suspension of 2-amino-4-methylbenzamide (0.028 g, 0.190 mmol) in 3 mL of methanol was added to a suspension of [Ru(η6-toluene)Cl2]2 (0.050 g, 0.095 mmol) in 3 mL of methanol. The resulting reaction mixture was stirred at room temperature for 1 hour. The yellow solution was concentrated under reduced pressure to a volume of 2 mL, and solid ammonium hexafluorophosphate (0.03 g) was added. The mixture was then stirred at room temperature for an additional 2 hours. The resulting yellow precipitate was collected by filtration and dried under reduced pressure. The yield of the product was 55%.
Infrared spectroscopy data (ATR, νmax/cm-1): 844.1 and 666.3 (aromatic C-H out-of-plane bending), 1436.9 (N-CO stretching valence), 1534 (N-CO stretching symmetric), 1592 (N-H bending of NH2 amide and amine), 1646 (C=O stretching), 3126.0 (aromatic C-H stretching), 3253.7 to 3309.9 (NH2 stretching).
Proton nuclear magnetic resonance spectroscopy data (500 MHz, [D6]DMSO, δ in ppm): 2.14 and 2.20 (singlets, 3H, CH3 of toluene), 2.16 (singlet, 3H, CH3 of L1), 5.70 to 6.16 (multiplet, 4H, C2 to C6 of toluene), 6.31 (multiplet, 1H, C3 of L1), 6.47 (singlet, 1H, C5 of L1), 6.60 (multiplet, 2H, NH2 of L1), 6.93 and 7.60 (singlets, 2H, NH2 of the amide group of L1), 7.41 to 7.43 (doublet, 1H, C6 of L1).
Carbon-13 nuclear magnetic resonance spectroscopy data (50 MHz, [D6]DMSO, δ in ppm): 18.54 (CH3 of toluene), 21.06 (CH3 of L1), 82.19, 84.30, 89.45, 93.02, 96.45, 105.38 (6 carbons, CH of toluene), 111.16, 115.70, 116.50, 128.79, 141.61, and 150.25 (6 carbons, CH of L1), 171.21 (C=O).
Electrospray ionization mass spectrometry data (m/z): negative mode 377 [M–PF6–H]− (calculated for RuC15H17N2OCl 377.6) and 413 [M–PF6+Cl]− (calculated for RuC15H17N2OCl2 413.1); positive mode 343 [M–PF6–Cl]+ (calculated for RuC15H17N2O 343.1).
Elemental analysis: calculated for RuC15H18N2OClPF6·H2O (%): C, 33.23; H, 3.69; N, 5.17; found (%): C, 33.21; H, 3.60; N, 5.60.
Synthesis of [(η6-p-cymene)Ru(L1)Cl]PF6 (C2). A suspension of 2-amino-4-methylbenzamide (0.0245 g, 0.164 mmol) in 3 mL of methanol was added to a suspension of [Ru(η6-p-cymene)Cl2]2 (0.0500 g, 0.082 mmol) in 3 mL of methanol, and the mixture was stirred at room temperature. After 24 hours, solid ammonium hexafluorophosphate (0.05 g) was added to the yellow solution. The reaction mixture was concentrated under reduced pressure to a volume of 2 mL, and diethyl ether was added. The resulting yellow precipitate was collected by filtration and dried under reduced pressure. The yield of the product was 63.9%.
Infrared spectroscopy data (ATR, νmax/cm-1): 837.8 (aromatic C-H out-of-plane bending), 1408.7 (N-CO stretching valence), 1549.6 (N-CO stretching symmetric), 1565 (N-H bending of NH2 amide and amine), 1647.9 (C=O stretching), 2973.0 to 3213.3 (aromatic C-H stretching), 3310.7 and 3455.2 (NH2 stretching of amide and amine N-H).
Proton nuclear magnetic resonance spectroscopy data (500 MHz, [D4]CD3OD, δ in ppm): 1.19 to 1.24 (multiplet, 6H, (CH3)2CH of cymene), 1.90 (singlet, 3H, CH3 of cymene), 2.50 (singlet, 3H, CH3 of L1), 2.60 (multiplet, 1H, (CH3)2CH of cymene), 5.30 to 5.29 (doublet, 1H, C2 of cymene), 5.39 to 5.45 (doublet of doublets, 2H, C3 and C5 of cymene), 5.58 to 5.57 (doublet, 1H, C2 of cymene), 7.22 to 7.23 (doublet, 1H, C3 of L1), 7.47 (doublet, 1H, C5 of L1), 7.65 to 7.66 (doublet, 1H, C6 of L1).
Carbon-13 nuclear magnetic resonance spectroscopy data (50 MHz, [D4]CD3OD, δ in ppm): 17.86 (CH3 of cymene), 22.03 and 22.55 (2 carbons, CH(CH3)2), 32.09 (CH3 of L1), 79.50, 81.94, 82.98, 82.94, and 96.15 (6 carbons, CH of cymene), 101.8, 122.78, 127.70, 133.08, 143.23, and 146.97 (6 carbons, CH of L1), 174.12 (C=O of L1).
Proton nuclear magnetic resonance spectroscopy data (500 MHz, [D6]DMSO, δ in ppm): 1.18 to 1.20 (multiplet, 6H, (CH3)2CH of cymene), 2.09 (singlet, 3H, CH3 of cymene), 2.14 (singlet, 3H, CH3 of C2-L1), 2.83 (multiplet, 1H, (CH3)2CH of cymene), 5.73 to 5.80 (doublet, 4H, C2, C3, C5, and C6 of cymene), 6.30 to 6.29 (doublet, 1H, C3 of L1), 6.47 (doublet, 1H, C5 of L1), 6.52 (multiplet, 2H, NH2 of C2-L1), 7.61 and 6.92 (singlets, 2H, NH2 of the amide group of L1), 7.44 to 7.42 (doublet, 1H, C6 of L1).
Carbon-13 nuclear magnetic resonance spectroscopy data (50 MHz, [D6]DMSO, δ in ppm): 17.73 (CH3 of cymene), 21.48 (2 carbons, CH(CH3)2), 30.33 (CH3 of L1), 85.67, 85.42, 99, 97, and 106.21 (6 carbons, CH of cymene), 111.52, 115.75, 128.78, 141.30 (6 carbons, CH of L1), 171.37 (C=O of L1).
Electrospray ionization mass spectrometry data (m/z): negative mode 419 [M–PF6–H]− (calculated for RuC18H23N2OCl 419.6) and 455 [M–PF6+Cl]− (RuC18H23N2OCl2 455.1); positive mode 385 [M–PF6–Cl]+ (calculated for RuC18H23N2O 385.6).
Elemental analysis: calculated for RuC18H24N2OClPF6 (%): C, 38.19; H, 4.24; N, 4.95; found (%): C, 38.58; H, 3.89; N, 5.18.
Synthesis of [(η6-toluene)Ru(L2)Cl2] (C3). Solid 3-amino-N-methylbenzamide (0.028 g, 0.190 mmol) and [Ru(η6-toluene)Cl2]2 (0.0500 g, 0.095 mmol) were suspended in 20 mL of methanol and stirred at room temperature for 24 hours. The resulting orange precipitate was collected by filtration and dried under reduced pressure. The yield of the product was 40%.
Infrared spectroscopy data (ATR, νmax/cm-1): 856 to 695 (aromatic C-H out-of-plane bending), 1544 (N-CO stretching symmetric, NH2 bending), 1592 (N-H bending of NH2 amide and amine), 1634 (C=O stretching), 3028 (aromatic C-H stretching), 3317 to 3137 (NH2 stretching and NH valence stretching).
Proton nuclear magnetic resonance spectroscopy data (500 MHz, [D6]DMSO, δ in ppm): 2.02 (singlet, 3H, CH3 of toluene), 2.61 (doublet, 3H, -NH-CH3), 5.22 (singlet, 2H, -NH2), 5.59 and 5.87 (multiplet and triplet, 5H, CH of toluene), 6.54, 6.79, 6.89, and 6.93 (doublet of doublets, doublet of doublets, triplet, and triplet, 4H, CH of arene), and 8.04 (doublet, 1H, NH).
Carbon-13 nuclear magnetic resonance spectroscopy data (50 MHz, [D6]DMSO, δ in ppm): 20.54 (CH3), 28.20 (-NH-CH3), 84.20, 86.81, 91.46, and 107.59 (6 carbons, CH of toluene), 114.77, 116.08, 118.24, 130.60, 137.51, and 150.66 (6 carbons, CH of arene), 169.40 (C=O).
Electrospray ionization mass spectrometry data (m/z): negative mode 451 [M+Cl]− (calculated for RuC15H18N2OCl3 449.6) and 413 [M–H]− (calculated for RuC15H17N2OCl2 413.1); positive mode 343 [M–2Cl]+ (calculated for RuC15H17N2O 343.1).
Elemental analysis: calculated for RuC15H18N2OCl2 (%): C, 43.47; H, 4.35; N, 6.76; found (%): C, 43.19; H, 4.51; N, 6.75.
Synthesis of [(η6-p-cymene)Ru(L2)Cl2] (C4). A suspension of [Ru(η6-p-cymene)Cl2]2 (0.0500 g, 0.082 mmol) and 3-amino-N-methylbenzamide (0.0245 g, 0.164 mmol) in 10 mL of methanol was stirred under an argon atmosphere at room temperature for 48 hours. The resulting orange precipitate was collected by filtration and dried under reduced pressure. The yield of the product was 42%.
Infrared spectroscopy data (ATR, νmax/cm-1): 882 to 698 (aromatic C-H out-of-plane bending), 1560 (N-CO stretching symmetric, NH2 bending), 1592 (N-H bending of NH2 amide and amine), 1632 (C=O stretching), 3082.6 (aromatic C-H stretching), 3314 to 3135 (NH2 stretching and NH valence stretching).
Proton nuclear magnetic resonance spectroscopy data (500 MHz, [D6]DMSO, δ in ppm): 1.17 (doublet, 6H, -CH(CH3)2), 2.07 (singlet, 3H, CH3 of cymene), 2.74 (doublet, 3H, -NH-CH3), 2.82 (multiplet, 1H, -CH(CH3)2), 5.20 (singlet, 2H, -NH2), 5.75 and 5.80 (multiplets, 4H, CH of cymene), 6.63, 6.89, 7.00, and 7.03 (doublet of doublets, doublet of doublets, triplet, and triplet, 4H, CH of arene), and 8.13 (doublet, 1H, NH).
Carbon-13 nuclear magnetic resonance spectroscopy data (50 MHz, [D6]DMSO, δ in ppm): 18.26 (CH3 of cymene), 21.90 (2 carbons, -CH(CH3)2), 26.59 (-NH-CH3), 30.37 (-CH(CH3)2), 85.91, 86.76, 100.49, and 106.79 (6 carbons, CH of cymene), 113.20, 114.53, 116.68, 129.00, 135.89, and 148.96 (6 carbons, CH of arene), and 167.78 (C=O).
Electrospray ionization mass spectrometry data (m/z): negative mode 493 [M+Cl]− (calculated for RuC18H25N2OCl3 491.6); positive mode 385 [M–2Cl]+ (calculated for RuC18H24N2O 385.1).
Biological Studies
Cell Culture
MDA-MB-231 and MCF-7 cell lines were grown in Dulbecco’s modified Eagle’s medium, while HCC1937, MDA-MB-453, MDA-MB-361, and BEAS-2B cells were maintained in Roswell Park Memorial Institute 1640 nutrient medium. All media were supplemented with 10% fetal calf serum, adjusted to pH 7.2, 2 mM L-glutamine, 25 mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid, and a Penicillin-Streptomycin solution at final concentrations of 100 U/mL penicillin and 100 µg/mL streptomycin. The media for HCC1937, MDA-MB-231, and MCF-7 cells were additionally supplemented with D-glucose to a final concentration of 4.5 g/L. Cells were cultured as a flat monolayer in tissue culture flasks at a constant temperature of 37 degrees Celsius in an incubator with a humidified atmosphere of 5% carbon dioxide.
MTT Assay
The impact of the synthesized ruthenium(II)-arene complexes (C1-C4), the starting ruthenium(II)-arenes (C5, C6), the ligands (L1, L2), and the reference compound cisplatin on the viability of cultured cells was determined using the colorimetric 3-(4.5-dimethylthiazol-2-yl)-2.5-diphenyltetrazolium bromide assay. Cells were seeded in 96-well culture plates at specific cell densities for each cell line in a total volume of 100 µL of nutrient medium and cultured overnight at 37 degrees Celsius. Stock solutions of the tested compounds were prepared in dimethyl sulfoxide at a concentration of 80 mM prior to use. The final concentration of dimethyl sulfoxide in the wells did not exceed 1% by volume. The cells were subsequently treated with serial dilutions of freshly prepared ruthenium compounds and ligands (50 µM, 100 µM, 200 µM, 400 µM, and 800 µM) or cisplatin (6.25 µM, 12.5 µM, 25 µM, 50 µM, and 100 µM) in the nutrient medium. Following continuous drug exposure for 72 hours, 20 µL of 3-(4.5-dimethylthiazol-2-yl)-2.5-diphenyltetrazolium bromide solution (5 mg/mL) was added to each well and incubated with the cells for 4 hours at 37 degrees Celsius. Finally, to dissolve the formed formazan crystals, 100 µL of 10% sodium dodecyl sulfate was added to each well. Absorbance readings were taken after 24 hours using a microplate reader at a wavelength of 570 nm. The percentage of cell viability was calculated by comparing the absorbance values of treated cells to those of untreated cells, which were taken as 100%. The IC50 values, representing the concentration of the tested compound that reduced the number of viable cells in a treated population to 50% compared to an untreated control, were determined by analyzing the correlation between the percentage of cell viability and the drug concentration using GraphPad Prism 6.0 software. The results are presented as the mean ± standard deviation and reported in µM. The selectivity index for each tested compound was calculated as the ratio of the IC50 value in BEAS-2B cells to the IC50 value in each of the cancer cell lines: HCC1937, MDA-MB-231, MDA-MB-453, MCF-7, and MDA-MB-361. Changes in the morphology of HCC1937, MCF-7, and MDA-MB-231 cells induced by the investigated ruthenium(II)-arene complexes C1-C4 were evaluated using inverted microscopy with an 80x/0.2 objective after 72 hours of continuous drug treatment, prior to the addition of the 3-(4.5-dimethylthiazol-2-yl)-2.5-diphenyltetrazolium bromide solution.
Flow Cytometric Analysis of Cell Cycle Phase Distribution
HCC1937 cells were seeded in 6-well culture plates in the nutrient medium at a density of 2.5 x 105 cells per well. After 24 hours of growth, the cells were continuously exposed to the ruthenium(II)-arene complexes (C1-C4), the starting ruthenium(II)-arenes (C5, C6), and the ligands (L1, L2) at concentrations of 200 µM and 500 µM, or to cisplatin at concentrations of 5 µM and 10 µM. Following a 72-hour incubation period, the cells were harvested by trypsinization, centrifuged, resuspended in phosphate-buffered saline, and fixed with 70% ice-cold ethanol at -20 degrees Celsius overnight. Flow cytometric analysis of the cell cycle phase distribution was performed on the fixed HCC1937 cells after staining with propidium iodide. Briefly, after fixation, the cells were washed with cold phosphate-buffered saline, incubated with 100 µg/mL of ribonuclease A (1 mg/mL in phosphate-buffered saline) for 30 minutes, and stained with 50 µg/mL of propidium iodide (400 µg/mL in phosphate-buffered saline) in the dark immediately before analysis. For each sample, 10000 cells were collected, and the cell cycle phase distribution was analyzed using a fluorescence-activated cell sorting BD Calibur flow cytometer and Cell Quest computer software.
Cellular Uptake, Subcellular Distribution and DNA Binding Studies
For cellular uptake, subcellular distribution, and DNA binding studies, 5 x 105 HCC1937 cells were grown in 25 cm2 culture flasks in triplicate. After 72 hours, the nutrient medium was replaced with fresh medium containing 200 µM of the ruthenium(II)-arene complexes (C1-C4), and the cells were incubated for an additional 24 hours. The cells were then harvested by trypsinization, and the cell pellet was collected by centrifugation at 1000 rpm for 7 minutes. Cell viability was determined using the Trypan blue exclusion assay. For subcellular distribution studies, the cell pellet was further lysed using a Subcellular Protein Fractionation Kit for Cultured Cells, following the manufacturer’s instructions, which yielded five protein fractions: cytoplasmic, membrane, nuclear soluble, chromatin-bound, and cytoskeletal. For DNA binding studies, total DNA from the HCC1937 cell pellet was isolated using a salting-out procedure. The DNA concentration was determined by measuring the absorbance at wavelengths of 260 nm and 280 nm using a BioSpec-nanospectrophotometer. Measurements of total intracellular uptake, DNA binding, and accumulation in subcellular compartments of the ruthenium(II)-arene complexes were performed in HCC1937 cells using inductively coupled plasma mass spectrometry with a Thermo Scientific iCAP Qc inductively coupled plasma mass spectrometry spectrometer and Qtegra operational software. The levels of ruthenium found in the total cells and in distinct cell compartments after treatment were expressed as the amount of ruthenium in nanograms taken up per 106 cells. The amount of ruthenium bound to the cellular DNA was expressed as picograms of ruthenium per microgram of DNA.
PARP-1 Inhibitory Assay
The potential of the examined ruthenium(II)-arene complexes (C1-C4) with benzamide ligands (L1, L2), as well as the starting ruthenium(II)-arenes (C5, C6), for inhibiting PARP-1 was investigated using the HT Universal Colorimetric PARP Assay, following the manufacturer’s instructions and previously reported procedures. This assay quantifies the incorporation of biotinylated poly(ADP-ribose) onto histone proteins in a 96-well strip well format. 3-aminobenzamide, provided in the kit, was used as a control inhibitor. Briefly, stock solutions of the investigated ruthenium compounds and ligands were prepared in dimethyl sulfoxide before use and then serially diluted with 1 x PARP buffer to the required concentrations (1 µM, 10 µM, 100 µM, and 1000 µM). 3-aminobenzamide was also serially diluted with 1 x PARP buffer to final concentrations of 1 µM, 10 µM, 100 µM, and 1000 µM. The tested compounds were incubated with PARP-1 enzyme (0.5 Units/well), biotinylated NAD+, activated DNA, and histones that were coated on the well. After a 60-minute incubation at room temperature, the solution was removed, and streptavidin-horseradish peroxidase was added for a further 1-hour incubation. The final reaction was detected after incubation with a pre-warmed colorimetric substrate in the dark for 15 minutes at room temperature. The colorimetric reaction was stopped by the addition of 0.2 M hydrochloric acid, resulting in a yellow color that remained stable for up to 1 hour at room temperature. The absorbance was read at 450 nm using a microplate reader. Differences in color intensities reflect the PARP-1 inhibition efficiencies of the tested compounds. By considering activity controls (without inhibitor) and negative controls (without PARP-1), the monitored mean optical density values were converted into relative values corresponding to the respective percentage of PARP-1 inhibition. From a plot of compound concentration versus the percentage of PARP-1 inhibition, generated using GraphPad Prism 6.0 software, IC50 values for 50% PARP-1 inhibition were determined.
Agarose Gel Electrophoresis Assay
The interaction of the ruthenium(II)-arene complexes (C1-C4), the starting ruthenium(II)-arenes (C5, C6), the ligands (L1, L2), and cisplatin with pHOT-1 plasmid DNA was studied using agarose gel electrophoresis. 300 ng of pHOT-1 plasmid DNA was incubated with the ruthenium compounds and ligands (200 µM and 400 µM) or cisplatin (12.5 µM, 25 µM, 50 µM, and 100 µM) for 24 hours at 37 degrees Celsius in a 5 mM Tris-HCl/50 mM NaCl buffer at pH 7.2. The experiments were also conducted in the presence of 100 µM DAPI or methyl green, which were added to the pHOT-1 plasmid DNA 1 hour prior to the addition of the investigated ruthenium complexes C1-C4. The final solutions were incubated for 24 hours. The samples were then analyzed by 0.8% agarose gel electrophoresis in 1 x TBE buffer at pH 8.0 for 2 hours at 30 V. After electrophoresis, the gel was stained for 1 hour by soaking it in an aqueous GelRed solution and subsequently visualized using a UV transilluminator.
Results and Discussion
Synthesis and Characterization
Four new ruthenium(II)-arene complexes (C1-C4) were synthesized through the reaction of 3-aminobenzamide analogues, 2-amino-4-methylbenzamide and 3-amino-N-methylbenzamide, with the aim of obtaining potentially effective PARP inhibitors. The starting ruthenium(II) complexes were dimeric species containing either a toluene or a p-cymene moiety. The synthesis of the ruthenium(II)-p-cymene-based complexes with the ligands had to be performed under an oxygen-free atmosphere, as previously reported for similar compounds. All syntheses were carried out in methanol solution. The yellow-orange complexes with 3-amino-N-methylbenzamide precipitated during the reaction, while the complexes with 2-amino-4-methylbenzamide precipitated only after the addition of an excess of ammonium hexafluorophosphate. Spectral and conductivity measurement data confirmed that 2-amino-4-methylbenzamide bound in a bidentate manner, coordinating through both nitrogen atoms, whereas 3-amino-N-methylbenzamide coordinated to the metal ion in a monodentate fashion.
Infrared Spectroscopy
A broad, intensive band observed between 3406 and 3168 cm-1 in the spectrum of 2-amino-4-methylbenzamide, corresponding to the standard vibrations of NH2 amine and amide groups, became sharper and shifted to lower wavenumber values in the spectra of complexes C1 and C2. This change is attributed to coordination to the metal ion via these two nitrogen atoms. The band associated with the N-C amide symmetric stretching vibration at 1615 cm-1 shifted to 1534 and 1549 cm-1 for complexes C1 and C2, respectively. NH2 deformation vibration bands appeared at 1565 cm-1 in both complexes, compared to 1548 cm-1 in the ligand, due to coordination through the amine nitrogen atoms.
Changes in shape and slight shifts in wavenumbers were observed in the infrared spectra of complexes C3 and C4 in the region from 3300 to 3100 cm-1, which could indicate coordination via the amide nitrogen atom. Sharp, well-defined, and intensive bands corresponding to standard symmetric N-C and NH amide deformation vibrations appeared at 1563 and 1598 cm-1, respectively, compared to less intensive and broader bands at 1559 and 1592 cm-1 in the ligand’s infrared spectrum. These alterations in wavenumbers and band shape suggest coordination of the amide nitrogen atom to the metal ion.
Nuclear Magnetic Resonance Spectroscopy
Proton Nuclear Magnetic Resonance Spectroscopy
For complexes C1 and C2, slight upfield shifts were observed in the proton nuclear magnetic resonance spectra for the signals corresponding to the amide protons. Well-defined, separated signals for the amine protons and the proton from the C3 carbon in the aromatic ring were noticeable in the proton nuclear magnetic resonance spectrum of 2-amino-4-methylbenzamide. These signals became closer and formed a multiplet signal in complex C1 and a broad singlet in the proton nuclear magnetic resonance spectrum of complex C2. The changes in the chemical shift values observed in the proton nuclear magnetic resonance spectra of the ligand and the complexes indicate bidentate coordination via both nitrogen atoms in the ligand moiety. Noticeable changes in the chemical shift values for the protons in the arene cap of the synthesized ruthenium(II)-arene complexes were also observed compared to the starting ruthenium(II) arene dimers. For the protons in the toluene moiety, changes in the shape of the signals and the splitting of one signal into two separated signals were observed. In the cymene-based ruthenium(II) complex, C2, a broad doublet was observed for the four aromatic protons. In the proton nuclear magnetic resonance spectra of C1 and 2-amino-4-methylbenzamide recorded in deuterated methanol, significant changes in chemical shift values for all protons were observed, but due to proton exchange between the ligand and the solvent, the amide and amine protons were not visible. Signals for all protons from the ligand moiety were shifted downfield in the proton nuclear magnetic resonance spectrum of C2 compared to the ligand. The signal for protons from the cymene moiety was observed in a specific region for aromatic protons and at different chemical shifts for methyl and isopropyl groups.
In the proton nuclear magnetic resonance spectra of complexes C3 and C4, upfield shifts were observed for the signal corresponding to the amide proton compared to the ligand. Additionally, signals for all other protons in complex C3 had slightly lower values, suggesting ligand coordination to the ruthenium(II) ion. As expected, signals for toluene or cymene protons were at lower chemical shifts than in the corresponding ruthenium(II) dimers. In these complexes, coordination to the metal ion is also proposed to occur via the amide proton, based on the observed changes in chemical shifts compared to the ligand itself.
Carbon-13 Nuclear Magnetic Resonance Spectroscopy
No significant changes were observed in the carbon-13 nuclear magnetic resonance spectra of the synthesized complexes compared to the ligand spectrum when recorded in deuterated dimethyl sulfoxide. Characteristic signals for the carbon atom from the carbonyl group were around 171 ppm, signals for arene carbon atoms from the ligand moiety were in the range of 150 to 111 ppm, while toluene or cymene arene carbon atoms gave signals from 105 to 82 ppm or 96 to 85 ppm, respectively. The signal for the carbon atom from the ligand methyl group appeared around 21 ppm in the carbon-13 nuclear magnetic resonance spectrum of the ligand, as well as in both spectra of the complexes. Below 30 ppm, signals appeared for the carbon atom from the toluene methyl group and carbon atoms from the cymene isopropyl group.
All signals for aromatic carbon atoms from the ligand were slightly shifted upfield in the deuterated methanol spectrum of complex C2. Signals for cymene aromatic carbon atoms also appeared in the region from 96 to 82 ppm, and the methyl group carbon atoms signals were located at specific chemical shift values for both the ligand and cymene, while the isopropyl carbon atom signal appeared at a specific chemical shift value.
In the carbon-13 nuclear magnetic resonance spectra of the ligand and the corresponding complexes C3 and C4, changes in chemical shift values were observed, with slightly higher values in the complexes compared to the ligand. A change in the chemical shift value for the carbonyl group was also observed, further supporting the coordination of the amide proton to the ruthenium(II) ion.
Chemical behaviour of the Ru(II)-arenes in DMSO
The stability of ruthenium(II)-arene complexes in commonly used solvents for biological assays, such as water or dimethyl sulfoxide, is an important factor in understanding the biological activity of potential new metallodrugs. Some studies have indicated that nitrogen or oxygen donor ligands, or even the arene cap, can be replaced by dimethyl sulfoxide. To investigate the behavior of the synthesized ruthenium(II)-arene compounds (C1-C4), changes in their proton nuclear magnetic resonance spectra in deuterated dimethyl sulfoxide solutions were monitored over 72 hours at specific time points after dissolution. These spectra showed traces of free arene (cymene or toluene) after 24 hours, with a slow increase in its concentration in solution over time. Given that the solutions of complexes used in biological tests were prepared immediately before use and that the final concentration of dimethyl sulfoxide did not exceed 1% in the total media, it can be assumed that significant changes in the complex structures did not occur under the experimental conditions of the biological assays.
Mass Spectrometry
The mass spectra of the synthesized complexes displayed characteristic bands in both negative and positive ion modes. For complexes C1 and C2, the molecular ion was not detected; however, two types of peaks were observed in the negative mode, corresponding to the [M-PF6-H]- ion and the [M-PF6+Cl]- ion, while in the positive recording mode, peaks for [M-PF6-Cl]+ ions were detected. In the mass spectra of complexes C3 and C4, peaks for [M+Cl]- and [M-H]- ions were detected in the negative mode, and peaks for [M-2Cl]+ ions were observed in the positive recording mode.
Conductivity Measurements
Molar conductivity measurements were conducted to further confirm the structures of the synthesized complexes. All complexes exhibited good solubility in dimethyl sulfoxide, so 1 mM solutions in this solvent were used for conductivity measurements. For complexes C1 and C2, the conductivity values were 22.7 Ω-1 cm2 mol-1 and 24.3 Ω-1 cm2 mol-1, respectively, while for complexes C3 and C4, the values were 1.65 Ω-1 cm2 mol-1 and 3.46 Ω-1 cm2 mol-1, respectively. Based on these observed values, it was confirmed that complexes C1 and C2 behave as 1:1 electrolytes, and the molar conductivity measurement values for C3 and C4 are consistent with neutral complexes.
Hydrolysis of synthesized complexes (Cl/H2O exchange process)
The generally accepted mechanism of cytotoxic action for ruthenium-arene complexes of the type [Ru(η6-arene)(XY)(Z)]n+, where XY represents a chelating ligand and Z is a labile halide, involves the hydrolysis of the Ru−Z bond, leading to the formation of an active Ru−OH2 species. The relative amounts of Ru−OH2 and Ru−OH species present depend on the pH values. While hydrolysis can be suppressed extracellularly due to high chloride ion concentrations, it becomes feasible intracellularly where chloride ion concentrations are lower. The rates of hydrolysis for such complexes can vary significantly. In this study, the time dependence of the water/chloride exchange process was monitored using ultraviolet-visible spectroscopy. An initial 2 mM solution of complex C3 in water at pH 6.8 was prepared, where the Ru−OH2 species is expected to be the most abundant at this pH. Subsequently, potassium chloride solutions were added to achieve final concentrations of 100 mM, 24 mM, and 4 mM, mimicking blood plasma, cytoplasm, and the cell nucleus, respectively. In all cases, changes in the ultraviolet-visible spectra were observed before and immediately after the addition of potassium chloride. No further changes in the spectra were observed over time, indicating that this reaction proceeded to completion immediately and exhibited no time dependence, suggesting a fast reaction.
The work of Sarah J. Dougan and Peter J. Sadler, which reviewed the comparative aqueous solution chemistry, cancer cell cytotoxicity, and reactivity towards different biomolecules for this class of complexes containing various arenes, chelating ligands, and monodentate leaving groups, clearly demonstrated that variations in the identity of all three ligand types can profoundly influence the pharmacological properties of these complexes. This observation has been corroborated in several studies. Previous work with similar organoruthenium(II) complexes containing the p-cymene ligand and a pyridine derivative coordinated in a monodentate or bidentate manner suggests that there is no simple correlation between fast hydrolysis of complexes, high-yield reaction with 9-methyladenine, and the inhibition of tumor cell growth. The unique combination of ligands dictates the specific mechanism of activation and cytotoxic action, which must be an integral part of the investigation of novel ruthenium complexes. For the complexes examined in this work, further thorough investigation is required to elucidate their coordination to enzymes, DNA, or other cellular components and different biomolecules.
PARP-1 Inhibition
To investigate the inhibitory potential of four novel ruthenium(II)-arene complexes (C1-C4) with benzamide ligands (L1, L2) on PARP-1 enzymatic activity, the HT Universal Colorimetric PARP Assay Kit was used. This kit measures the in vitro incorporation of biotinylated poly(ADP-ribose) onto histone proteins. The inhibitory effects of the novel complexes, their starting complexes (C5, C6), and the derived ligands (L1, L2) were compared to determine if the incorporated benzamide ligands or the arene ligands (p-cymene versus toluene) contribute to the PARP-1 inhibitory potency of the resulting ruthenium(II) complex. 3-aminobenzamide was used as a positive reference compound. The results, expressed as IC50 values determined after a 60-minute in vitro incubation, are presented. The literature value of 3-aminobenzamide was reproduced, while the benzamide derivatives, L1 and L2, exhibited significantly lower PARP-1 inhibitory activity, providing only a limited percentage of inhibition at the highest concentration tested. The four novel ruthenium(II)-arene complexes displayed inhibitory effects with IC50 values within a specific range. All complexes, except C3, showed similar concentration-dependent PARP-1 inhibitory activity and reached maximum inhibition at the highest concentration tested. Complexes with a p-cymene moiety, C2 and C4, were superior PARP-1 inhibitors compared to complexes with a toluene moiety, C1 and C3. Notably, C2 and C4 exhibited lower IC50 values against PARP-1 than 3-aminobenzamide. Furthermore, the starting ruthenium(II)-p-cymene complex, C6, was approximately eight times more potent in inhibiting PARP-1 than the ruthenium(II)-toluene complex, C5.
From a general structure-activity perspective, complexes with a bidentately coordinated ligand (C1 and C2) demonstrated higher efficacy in inhibiting PARP-1 compared to the other two complexes (C3 and C4). Additionally, complexes carrying p-cymene (C2 and C4) were better PARP-1 inhibitors than those carrying toluene (C1 and C3). This suggests that structural variations influence the specificity and potency of the complexes to inhibit PARP-1 activity. The obtained data were consistent with previous reports on the PARP-1 inhibitory potential of ruthenium(II)-arenes. This study, to the best of our knowledge, included the starting binuclear ruthenium(II)-arene complexes C5-C6 for the first time. The ruthenium(II)-p-cymene scaffold appears promising for further structural improvements and investigation of PARP-1 inhibitory potential. The stronger PARP-1 inhibitory activity of the bidentate conjugate molecules C1 and C2 also correlated with their generally higher antiproliferative effects.
Cell Growth Inhibition
The cell growth inhibition of four new ruthenium(II)-arene complexes, as well as their corresponding starting binuclear ruthenium(II) complexes and ligands, was assessed on a panel of five human breast cancer cell lines and one human non-cancer bronchial epithelium cell line using an MTT assay, with cisplatin used as a reference compound. The genetic and phenotypic characteristics of the breast cancer cell lines used in this study are summarized. HCC1937, MDA-MB-231, and MDA-MB-453 cell lines are considered triple-negative breast cancer cells, while MCF-7 and MDA-MB-361 are positive for estrogen and progesterone receptors and are therefore described as hormone-sensitive. In this breast cancer cell line panel, only HCC1937 cells are homozygous for a specific BRCA1 mutation and experience a complete loss of BRCA1 function, while the other investigated breast cancer cells have wild-type BRCA1 gene expression.
The two ligands, L1 and L2, showed very low cell growth inhibition potential against all tested cells and were non-toxic even at a high concentration. However, when coordinated to the ruthenium(II)-arene moieties, the resulting complexes, C1-C4, displayed cell growth inhibition potential with IC50 values in the micromolar range. The precursor ruthenium(II)-arenes, C5 and C6, exhibited IC50 values varying within a specific range.
Differences observed in the sensitivity pattern among breast cancer cell lines treated with the investigated complexes C1-C4 most probably arose due to the genetic and phenotypic variations between the cells. Generally, higher sensitivity to the action of C1-C4 was found in HCC1937, MDA-MB-231, and MCF-7 cells, which displayed some similarities in response. Hormone-responsive MDA-MB-361 cells showed approximately equal low sensitivity to C1-C6. Structure-activity comparison indicated that complexes with a bidentately bound benzamide ligand, C1 and C2, exhibited lower IC50 values than complexes C3 and C4, with a monodentate ligand, except in HCC1937 cells, where C2 was the least active.
In summary, the antiproliferative efficiency of C1-C4 in HCC1937 cells decreased in a specific order, and in MDA-MB-231 and MCF-7 cells, it followed a different order. The results indicated the highest activity of C1 in HCC1937, MDA-MB-231, and MCF-7 cells, with corresponding selectivity index values determined according to the response in non-malignant BEAS-2B cells. Complex C2 also displayed higher cytoselectivity toward MDA-MB-231 and MCF-7 cells.
Cisplatin showed IC50 values in the low micromolar range, with the highest activity toward MCF-7 cells and the lowest toward MDA-MB-361 cells. The response of triple-negative breast cancer cells from this panel to cisplatin treatment was approximately within the same concentration range.
It is important to note that the antiproliferative effects of C1-C4 in the tested panel of cell lines cannot be attributed solely to their PARP-1 inhibitory effects. Despite the observed potency for PARP-1 inhibition, complexes C2 and C4 displayed weak cell growth inhibition in HCC1937, MDA-MB-231, and MCF-7 cells. The cytotoxicity range of the C1-C4 complexes is within the known range for ruthenium complexes, such as RAPTA compounds and NAMI-A, which showed weak in vitro cytotoxicity even at a high concentration. In the case of NAMI-A, high cytotoxicity is not a prerequisite for further development as an anticancer drug candidate, as it exhibits antimetastatic activity despite being essentially non-toxic to primary cancer cells. Additional structure-activity studies were performed to reveal possible multi-targeting properties of complexes C1-C4 and their mechanism of action in HCC1937 breast cancer cells. This cell line was of interest due to its homozygous mutation in the BRCA1 gene and complete loss of BRCA1 function, supporting its potential response to PARP inhibitors via synthetic lethality. However, it must be noted that additional genetic alterations, such as the loss of PTEN, a gene frequently mutated in BRCA1-deficient breast cancers, may accumulate and restore homologous recombination repair efficiency. All of this complexity must be considered when attempting to explain the observed selectivity of the investigated complexes C1-C4. Moreover, MDA-MB-231 and MCF-7 breast cancer cells, despite being BRCA1/2 wild-type, may harbor a homologous recombination deficient phenotype, rendering them sensitive to PARP inhibitors. Together, these data support the findings of cell line-dependent biological activity of the tested ruthenium complexes in the breast cancer cell line panel.
Impact on Cell Morphology
Since the MTT assay primarily measures mitochondrial dehydrogenase activity, the IC50 values obtained cannot differentiate between cytotoxic and antiproliferative effects of the tested drugs. Examining the morphological appearance of cells provides valuable information about concentration- and time-dependent changes in cellular response. Observation of morphological changes in HCC1937, MDA-MB-231, and MCF-7 cell lines following treatment with C1-C4 for 72 hours revealed a concentration-dependent reduction in cell number. Treatment of MCF-7 cells with 400 µM of complexes C1-C4 resulted in a mixed cell population consisting of rounded cells and cells with preserved morphology. More pronounced changes in cell density and morphology were observed at higher concentrations (800 µM) of C1 and C2. MDA-MB-231 cells showed a similar response, accompanied by cell elongation and the appearance of pseudopods.
In HCC1937 cells, C1-C4 caused a dose-dependent decrease in cell number, along with the appearance of small rounded cells and individual enlarged single- or multi-vacuolated cells, already at a treatment concentration of 200 µM. At 800 µM, the complexes profoundly altered HCC1937 cell morphology, promoting a larger proportion of rounded cells, indicative of cell death. The majority of HCC1937 cells treated with C1 became rounded and formed clusters. Taken together, these findings support the results of the MTT assay, which indicated a higher cell growth inhibitory potential of complex C1 compared to its structurally related C2-C4 complexes.
Cell Cycle Arrest
The effect of the investigated compounds on cell cycle progression in HCC1937 cells was analyzed by flow cytometry. After 72 hours of treatment with ruthenium complexes and ligands, or cisplatin, the DNA content in cells was estimated by propidium iodide staining. Treatment with a low concentration of cisplatin induced significant arrest of the cell cycle in the G2-M phase. Increasing the cisplatin concentration triggered accumulation in the S phase of the HCC1937 cell cycle. Treatment with the ligands L1 and L2 did not induce any significant changes in the cell cycle phase distribution at the tested concentrations. Among the investigated ruthenium complexes, only C5 and novel C1, applied at a higher concentration, induced evident changes in the cell cycle of HCC1937 cells. C5 caused arrest of the cell cycle in the S phase and an accompanying decrease in the percentage of cells in the G1 phase. A negligible increase was detected in the Sub-G1 region, characteristic of fragmented DNA and apoptotic cells, after C5 and C6 treatment. C1-induced S phase arrest in the BRCA1-mutated HCC1937 cells suggested the capability of this compound to block DNA replication. This effect may be related to a similar ability of its constituent ruthenium(II)-toluene complex, C5. The present results correlate well with the data showing the DNA binding potential of ruthenium(II) obtained by inductively coupled plasma mass spectrometry. It is likely that the highest activity of C1 in HCC1937 cells is due to the more efficient cell cycle arrest.
Differential Cellular Accumulation
To further investigate the mechanism underlying the activity of C1-C4 complexes in HCC1937 cells and to locate their possible targets within the cell, total cellular uptake and subcellular distribution studies were performed using inductively coupled plasma mass spectrometry. The affinity of C1-C4 for DNA binding was determined based on the ruthenium(II) content found in the genomic DNA isolated from treated HCC1937 cells. The results revealed that after 24 hours of continuous treatment with equimolar concentrations, all four complexes entered the HCC1937 cells and accumulated in different amounts. The intracellular content of C1-C4 followed a decreasing order. C1 entered the HCC1937 cells most efficiently. C2 showed the lowest ability to enter the HCC1937 cells. A similar trend was observed in the DNA binding affinity of the complexes, with C1 showing a significantly higher amount bound to DNA compared to C2. The inductively coupled plasma mass spectrometry analysis of compound distribution across cellular compartments disclosed a similar localization pattern of C1, C3, and C4 in HCC1937 cells, with the highest accumulation in the cytosol and a lower amount in the membrane/organelle fraction. In contrast, C2 was retained in the membrane/organelle fraction as the largest portion and to a lesser extent in the cytosol. Furthermore, significant ruthenium content was found in the nuclear fraction of HCC1937 cells treated with C1-C4. Only a small amount of ruthenium atoms was found in the cytoskeletal fraction.
Drug localization in specific cellular compartments and its interaction with biomolecules is not only structure-dependent but also time- and concentration-dependent, as reported for different ruthenium complexes. Based on the obtained data and structure-activity correlations, it is notable that complex C1 exhibits an advantage over the other tested complexes in its ability to enter the cell and bind to nuclear DNA. The highest intracellular uptake of C1 and its accompanying accumulation in the DNA fraction of HCC1937 cells are consistent with the higher growth inhibitory activity of this compound. Additionally, the notable accumulation of C1 in the nuclear fraction of cells may indicate a tendency of this type of “combi-molecules” to also interact with DNA-associated proteins, such as PARP-1. This is consistent with the PARP-1 inhibitory potential of C1 observed in the present study, as well as with previous reports of cellular uptake of some ruthenium(II)-arene complexes with phenanthridine-based ligands.
The lower intracellular content of C2, C3, and C4 could be attributed to either reduced uptake or enhanced efflux of these compounds, which contributes to their lower growth inhibitory activity. The preferential uptake of C1 by HCC1937 cells highlights that structural differences among the investigated ruthenium complexes significantly influence the affinity and stability of their interactions with cellular proteins and nuclear DNA.
Interaction with Plasmid DNA
The ability of ruthenium(II)-arene complexes and their starting binuclear complexes and ligands to alter the tertiary structure of DNA in a cell-free system was determined by testing the electrophoretic mobility of pHOT-1 plasmid DNA. Cisplatin was used as a reference compound. Small changes in the tertiary structure of DNA can be easily monitored on the plasmid DNA, as its superhelical nature closely mimics certain forms of intracellular DNA, such as chromatin. The supercoiled form of DNA experiences less resistance from the gel due to its compact nature and migrates faster, whereas the open (nicked) circular form moves slower through the gel. The pHOT-1 plasmid consisted mainly of the supercoiled Form I and the open circular Form II. The addition of cisplatin to the plasmid resulted in a gradual disappearance of the two distinct bands representing Form I and Form II, and the subsequent formation of one smeared band. Cisplatin caused a concentration-dependent decrease in the rate of migration of Form I and, at the same time, accelerated the mobility of Form II. At a specific concentration, cisplatin caused a significant decrease in the rate of migration of Form I, likely due to relaxation (unwinding) of the supercoiled DNA. At higher concentrations, cisplatin caused the coalescence of Form I and Form II bands. This observation indicated local untwisting at the sites of cisplatin-DNA adducts and was consistent with previous literature data.
Incubation of pHOT-1 plasmid DNA with the ligands L1 and L2 did not induce any significant alterations in the tertiary structure of the plasmid DNA that could be visualized. Interestingly, the two investigated ruthenium(II)-arenes, C5 and C6, altered the electrophoretic mobility of pHOT-1 plasmid DNA. C5 displayed the capability to interact with plasmid DNA, and treatment at a specific concentration induced untwisting of the supercoiled Form I and the appearance of a smeared band. This effect can be explained by the capacity of the complex to form adducts with plasmid DNA, which, in turn, induce unwinding of the negatively supercoiled helix and a reduction in the number of negative supercoils, leading to slower mobility of the plasmid bands. Interestingly, C5 at a higher concentration and C6 at a lower concentration caused the disappearance of Form I and Form II bands and the formation of one band whose mobility exceeded that of the supercoiled Form I. This result indicated enhanced coiling and the formation of a more compact DNA structure upon C5 or C6 binding, implicating the role of the ruthenium(II) metal center in DNA binding and damage.
Although treatment with C1-C4 at a lower concentration did not alter the electrophoretic mobility of pHOT-1 plasmid DNA, increasing the concentration caused various changes in the mobility of the plasmid DNA. C1 showed the capability to interact with plasmid DNA, as seen from the appearance of a low-intensity smeared band. C2 and C3 induced relaxation and unwinding of the supercoiled Form I and a mobility shift to some extent. In contrast, C4 caused the appearance of a low-intensity band with higher mobility than the respective control plasmid DNA, indicating profound coiling and DNA condensation, as observed also with its parental ruthenium(II)-p-cymene complex, C6. The degree of superhelicity of the plasmid DNA was altered to a certain extent by all tested complexes, which correlates with previous reports for ruthenium(II)-arenes. Since unwinding of the double strand of the DNA helix can be caused by the formation of monofunctional or intercalating adducts, these findings again suggest that the tested ruthenium complexes manifest their activity, at least partly, through interaction with DNA. To further investigate possible DNA binding preferences of the four complexes, gel electrophoretic analysis of treated pHOT-1 plasmid DNA was performed in the presence of a minor groove binding agent and a major groove binding agent. The supercoiled pHOT-1 plasmid was treated separately with the groove binders prior to the addition of C1-C4. When applying the complexes at a higher concentration, there were no significant alterations in the mobility pattern of plasmid forms, suggesting that C1-C4 complexes do not compete with the groove binders for DNA binding sites and therefore do not act as minor or major groove binders. Treatment of the groove binder-bound plasmid DNA with C1-C4 displayed lower intensity bands than treatment with the complexes only. It is well known that the gel staining dye used acts by intercalation and groove binding, preferably to the minor groove of DNA. The disappearance of the plasmid DNA bands implies significant saturation of the dye binding sites and a strong DNA intercalating potential of the investigated ruthenium complexes, especially seen for C1 and C4, resulting in less efficient staining of the gel for DNA visualization.
Conclusions
In the present study, four new ruthenium(II)-arene complexes, C1-C4, carrying derivatives of 3-aminobenzamide with anticipated multifunctional potential, were synthesized and tested for their anticancer activity in vitro. The complexes exhibited efficiency in inhibiting PARP-1 activity at micromolar concentrations, following a specific order of potency. Complexes bearing the η6-p-cymene moiety demonstrated better PARP-1 inhibition than their starting complex and the benchmark PARP inhibitor, suggesting that their ability to impair PARP-1 action may derive through interaction with the catalytic domain or zinc fingers of the PARP-1 enzyme. The capability of the precursor ruthenium(II)-p-cymene binuclear complex to inhibit PARP-1 indicates a noteworthy affinity of this part of the complex (and probably especially the metal center) alone to interfere with PARP-1 activity.
The investigated C1-C4 complexes displayed growth inhibitory effects toward breast cancer cells, including BRCA1-mutant, BRCA1-wild-type triple-negative, and hormone-responsive cell lines, in the micromolar range of concentrations. Interestingly, three cell lines displayed approximately two-fold higher sensitivity to the action of the complexes than the other two cell lines. C1 exhibited the highest antiproliferative potential, with similar IC50 values across these sensitive cell lines. Structure-activity comparison revealed that bidentately coordinated ligands in the ruthenium(II)-arene complexes potentiated biological action, as C1 and C2 exhibited higher activity than the monodentately bound C3 and C4. Further investigation of the mechanism of action of C1-C4 in BRCA1-mutant cells disclosed a marked relation between the growth inhibitory potential of the complexes and their cellular uptake. The results of the inductively coupled plasma mass spectrometry analysis revealed the efficiency of C1-C4 uptake by the cells following a specific decreasing order. Complexes C1, C3, and C4 mostly accumulated in the cytoplasm of the cells, allowing them to interact with various intracellular molecules and manifest their complex mechanism of action. C1 displayed notable nuclear-targeting properties, as it was evidently found in the chromatin-bound nuclear fraction, showed the highest DNA binding affinity, and the ability to interfere with DNA replication (arrest of the cell cycle in the S phase). Furthermore, inductively coupled plasma mass spectrometry analysis in the cells and electrophoretic mobility study in a cell-free system using plasmid DNA showed the affinity of C1 to bind DNA and form DNA adducts. In vitro PARP-1 testing revealed its potential for PARP-1 inhibition. Based on these results, it can be concluded that C1 demonstrated the potential to obstruct the structural and functional properties of DNA (DNA replication), either through binding to DNA or interfering with PARP-1 activity, or by a combination of both. Additionally, C1-C4 complexes may have other cellular targets that were not revealed in this study but could be anticipated from the significant amount of complexes found in the membrane and cytoplasmic fractions of cells. Complex C2 showed the highest accumulation in the membrane fraction and also selectivity for highly invasive cells, thus may present a candidate for further investigation of antimetastatic potential.
Despite the low PARP-1 inhibitory potential of the deriving benzamide-based ligands, the results pointed out that the ruthenium(II)-arene complexes C1, C2, and C4 exhibited an enhanced ability to compromise PARP-1 enzymatic action. In recent years, it has become clear that many of the available PARP-1 inhibitors have a polypharmacology profile and may act at distant unrelated targets. Taken together, these results support the hypothesis of the multi-targeting affinity of ruthenium as a single agent and justify the strategy based on the rational design of hybrid molecules with metals coordinated to bioactive ligands. This strategy may help in obtaining a new category of more potent anticancer drug candidates with a broader spectrum of pharmacological activities in tumors harboring BRCA1 mutations.
Highlights
* Ruthenium(II) arene complexes exhibit good efficiency in inhibiting PARP-1 activity.
* Ruthenium complexes exhibit good antiproliferative activity against breast cancer cells.
* Ruthenium(II) arene complexes display notable nuclear-targeting properties.
* Ruthenium(II) arene complexes interfere with DNA replication.