The anti-breast cancer stem cell properties of gold(i)-non-steroidal anti-inflammatory drug complexes

Abstract

The anti-breast cancer stem cell (CSC) properties of a series of gold(i) complexes comprising various non-steroidal anti-inflammatory drugs (NSAIDs) and triphenylphosphine 1-8 are reported. The most effective gold(i)-NSAID complex 1, containing indomethacin, exhibits greater potency for breast CSCs than bulk breast cancer cells (up to 80-fold). Furthermore, 1 reduces mammosphere viability to a better extent than a panel of clinically used breast cancer drugs and salinomycin, an established anti-breast CSC agent. Mechanistic studies suggest 1-induced breast CSC death results from breast CSC entry, cytoplasm localisation, an increase in intracellular reactive oxygen species levels, cyclooxygenase-2 downregulation and inhibition, and apoptosis. Remarkably, 1 also significantly inhibits tumour growth in a murine metastatic triple-negative breast cancer model. To the best of our knowledge, 1 is the first gold complex of any geometry or oxidation state to demonstrate anti-breast CSC properties.

Scheme 1. Reaction scheme for the preparation of the gold(i) complexes containing triphenylphosphine and indomethacin, diflunisal, naproxen, diclofenac, salicylic acid, tolfenamic acid, mefenamic acid, or ibuprofen (1–8).

Fig. 1. X-ray structures of 1–8 comprising of triphenylphosphine and indomethacin, diflunisal, naproxen, diclofenac, salicylic acid, tolfenamic acid, mefenamic acid, or ibuprofen. Thermal ellipsoids are drawn at 50% probability. C atoms are shown in grey, N in dark blue, O in red, P in purple, Cl in dark green, F in light green, and Au in gold. The H atoms have been omitted and the Ph groups in PPh₃ are represented as wireframes for clarity.

Fig. 2. (A) Quantification of mammosphere formation with HMLER-shEcad cells untreated and treated with 1–3, 5-fluorouracil, capecitabine, cisplatin, carboplatin, and salinomycin at their respective IC₂₀ values after 5 days incubation. Error bars = SD and Student's t-test, * = p < 0.05. (B) Representative bright-field images (×10) of the mammospheres in the absence and presence of 1–3 at their respective IC₂₀ values after 5 days incubation.

Fig. 3. (A) ¹H NMR spectra for complex 1 (10 mM) in DMSO-d₆, in the absence and presence of N-acetylcysteine (NAC, 10 mM) over the course of 72 h at 37 °C. The ¹H NMR spectra of indomethacin and [Au^I(NAC)(PPh₃)] (both 10 mM) in DMSO-d₆ are also provided. (B) Representative scheme for the reaction of 1 with NAC.

Fig. 4. (A–C) FITC Annexin V-propidium iodide binding assay plots of untreated HMLER-shEcad cells, and HMLER-shEcad cells treated with 1 (IC₅₀ value ×2 for 48 h) or cisplatin (25 μM for 48 h). (D) Representative dose–response curves for the treatment of HMLER-shEcad cells with 1 in the absence and presence of z-VAD-FMK (5 μM) or PGE2 (20 μM), after 72 h incubation. (E) Representative histograms displaying the green fluorescence emitted by anti-COX-2 Alexa Fluor 488 nm antibody-stained HMLER-shEcad cells treated with LPS (2.5 μM) for 24 h followed by 48 h in fresh media (red) or media containing 1 (IC₅₀ value, blue) or indomethacin (20 μM, orange).

Fig. 5. (A) The change in tumour volume of 4T1-bearing mice (1 million cells inoculated, n = 5) over 16 days, following intraperitoneal administration of 1 or the vehicle three times a week. Unpaired t-test, *p < 0.05 and **p < 0.01. (B) Representative images of excised tumours from 1- or vehicle-treated mice. (C) The change in body weight of 1- or vehicle-treated mice during the efficacy study lasting 16 days. (D) Biodistribution of gold in 4T1-bearing mice following 1 treatment. (E) Hematoxylin and eosin (H&E) staining indicates reduced cellularity in tumour tissue obtained from 1-treated mice (compared to vehicle-treated mice), and unaltered cellularity in liver and kidney tissue obtained from 1-treated mice (compared to vehicle-treated mice).