Identification of a Synergistic Multi-Drug Combination Active in Cancer Cells via the Prevention of Spindle Pole Clustering

A major limitation of clinically used cancer drugs is the lack of specificity resulting in toxicity. To address this, we performed a phenotypically-driven screen to identify optimal multidrug combinations acting with high efficacy and selectivity in clear cell renal cell carcinoma (ccRCC). The search was performed using the Therapeutically Guided Multidrug Optimization (TGMO) method in ccRCC cells (786-O) and nonmalignant renal cells and identified a synergistic low-dose four-drug combination (C2) with high efficacy and negligible toxicity. We discovered that C2 inhibits multipolar spindle pole clustering, a survival mechanism employed by cancer cells with spindle abnormalities. This phenotype was also observed in 786-O cells resistant to sunitinib, the first line ccRCC treatment, as well as in melanoma cells with distinct percentages of supernumerary centrosomes. We conclude that C2-treatment shows a high efficacy in cells prone to form multipolar spindles. Our data suggest a highly effective and selective C2 treatment strategy for malignant and drug-resistant cancers.

In the s-FSC approach, experimental data points are selected and tested based on the design of experiment approach (DoE) (Figure 1a), and more specifically using a series of orthogonal array composite design (OACD) matrices. These matrices are based on the combination of a two-level fractional factorial design and a three-level orthogonal array design generating a resolution IV design matrix. The two-level fraction factorial design allows for the estimation of linear and bi-linear effects (i.e., single drug and 2-drug interaction coefficients), while the three-level orthogonal design allows for the estimation of linear and quadratic effects. Thus, the design allows for internal cross-validation of linear effects. The resulting design is a resolution IV matrix and allows for accurate screening for the most influential factors in the system based on accurate estimations of each factor's main effect (i.e., main effects are not aliased by other main effects or two-factor effects) [19], in addition to estimates of interaction and quadratic effects. Such matrices have previously been published in the literature [20] or can be developed de novo using appropriate statistical software (SAS Institute Inc., Statware Inc., SPSS Inc., etc.). Input requirements include the number of compounds to be screened (referred to as factors) and number of compound doses to be considered (referred to as levels, generally three in levels are applied in this methodology corresponding to two compound doses and a dosage of zero).

Endothelial-Pericyte Coculture Network Formation Assay
The assay was performed as previously described [21]. Briefly, adherent HUVEC and pericytes were washed in serum-free M199 medium and then labelled 30 min at 37 °C with either CellTracker Green CMFDA dye (HUVEC, CMFDA488, Life Technology/Molecular Probes: cat no. C-2925 stock: 10 mM; diluted to final 1 μM in M199 without FCS) or CellTracker Orange CMRA dye (pericytes, CMRA548, Life Technology cat no. C-34551; stock 5 mM; diluted to final 1 μM in M199 without FCS). angiogenesis micro slides (Ibidi GmbH, Gräfelfing, Germany) were incubated on ice and 12 μL of icecold growth factor-reduced Matrigel (Corning) was added to each well. Subsequently, complete M199 medium (50 μL) was added to each well and incubated for 45 minutes at 37 °C. 5000 HUVEC and 2500 pericytes were added to the polymerized Matrigel in each well, and cultured in complete M199 for 7-10 hours. Live cell imaging with NikonA1R (time-lapse, Z stack imaging) and Fiji/ImageJ with Angiogenesis Analyzer toolset was used for analysis.

Chorioallantoic Membrane (CAM) of the Chicken Embryo
An in vivo model of developmental angiogenesis was used to validate the anti-angiogenic activity of C2 [22 , 23]. Fertilized chicken eggs were incubated in a hatching incubator at 37 °C with a relative humidity 65% [24]. On embryo development day (EDD) three, eggs were turned such that the narrow apex was facing up, a small hole was made in the top of the eggshell and closed with scotch tape. Eggs were returned to the incubated in a stationary position until EDD seven and the hole on the apex of the egg was expanded to approximately 3 cm in radius. A plastic ring was deposited on the CAM membrane and 20 μL treatments were administered within the ring. Treatment was performed twice, on EDD seven and eight, and the membranes were imaged via fluorescence angiograms on EDD nine using an epifluorescence microscope (Nikon AG, Eclipse FN1, Japan) coupled to pco.pixelfly. Fluorescein isothiocyanate dextran (FITC-dextran, 20 kDa, 20 μL, 25 mg/mL, Sigma-Aldrich) was injected intravascularly. To increase vascular contrast, 50 μL if black ink was injected (Pelikan, Witzikon, Switzerland) into the embryonic cavity. Image-based quantification using the number of branching points/mm 2 was performed using the ImageJ-based software [25].

Cell Cycle Analysis and Cell Death Induction by Flow Cytometry
Flow cytometry measurement of cellular DNA with the fluorochrome propidium iodide (PI) was used to identify cell cycle and apoptotic fractions. Cells were seeded in 6-well plates at a density of 20-40 × 10 3 cells/well and incubated for 24 hours. Medium or experimental condition was applied, and cells were incubated for an additional 72 hours. Both floating and attached cells were harvested, washed with PBS, resuspended and fixed with 70% ethanol, then incubated for 2 h at 4 °C. Cells were stained with PI/RNAse staining solution (ThermoFisher Scientific, 1825102) for 30 min, in the dark, at room temperature. Cells were analyzed with the Attune NxT acoustic focusing cytometer (Life Technologies, Carlsbad, CA, USA) in the BL2 channel. Maximum excitation of PI bound to DNA is at 536 nm, and emission is at 617 nm. Apoptotic cells were defined as having subG1 DNA staining and quantified with Attune TM NxT Software v. 2.5 (Life Technologies).

F-Actin and Nuclear DAPI Cell Staining
786-O cells were seeded on glass in a 24-well plate with a density of 6000 cells/well. After 72 hours of treatment with optimized drug combination or monotherapies, the cells were fixed with 4% formaldehyde for 10 minutes at RT, washed twice and permeabilized with 0.1% Triton-X for 15 minutes. Cells were incubated at RT with Alexa Fluor 488 Phalloidin (A12379, Thermo Fisher) diluted 1:200 for 20 minutes to stain f-actin, washed twice and incubated at RT with 1 ug/mL Dapi (D9542, Sigma) for 5 minutes. After a final wash-step the glasses were imaged on the Biotek Cytation3 Imaging reader using a 10× objective. Figure S1. 786-O and HEK-293T dose-response curves for all drugs included in the Therapeutically Guided Multidrug Optimization (TGMO)-based screen. Using a four-parameter nonlinear fit of the log-transformed doses for each compound (Graphpad Prism ® ), the ED20 concentration was calculated and selected as the high dose input in the TGMO-based screen.        cells chronically exposed to sunitinib (786-OsunR) treatment (1 µ M) using the fourparameter non-linear fit of the log-transformed doses (Graphpad Prism ® ). Error bars represent ± SEM.* p < 0.05 and ** p < 0.01 represent significance between the RCC naïve and 786-OsunR cells determined with a two-way ANOVA with Sidak's multiple comparisons test of 3-4 independent experiments. (b) Representative bright-field and fluorescent images were taken at 540 nm demonstrating sunitinib accumulation in chronically treated cells. Images were acquired with 40x magnification. The scale bar represents 20 µ m.