Ouabain and chloroquine trigger senolysis of BRAF‐V600E‐induced senescent cells by targeting autophagy

Abstract The expression of BRAF‐V600E triggers oncogene‐induced senescence in normal cells and is implicated in the development of several cancers including melanoma. Here, we report that cardioglycosides such as ouabain are potent senolytics in BRAF senescence. Sensitization by ATP1A1 knockdown and protection by supplemental potassium showed that senolysis by ouabain was mediated by the Na,K‐ATPase pump. Both ion transport inhibition and signal transduction result from cardioglycosides binding to Na,K‐ATPase. An inhibitor of the pump that does not trigger signaling was not senolytic despite blocking ion transport, demonstrating that signal transduction is required for senolysis. Ouabain triggered the activation of Src, p38, Akt, and Erk in BRAF‐senescent cells, and signaling inhibitors prevented cell death. The expression of BRAF‐V600E increased ER stress and autophagy in BRAF‐senescent cells and sensitized the cell to senolysis by ouabain. Ouabain inhibited autophagy flux, which was restored by signaling inhibitors. Consequently, we identified autophagy inhibitor chloroquine as a novel senolytic in BRAF senescence based on the mode of action of cardioglycosides. Our work underlies the interest of characterizing the mechanisms of senolytics to discover novel compounds and identifies the endoplasmic reticulum stress‐autophagy tandem as a new vulnerability in BRAF senescence that can be exploited for the development of further senolytic strategies.

The IMR90/ER-RasVal12 cell line was a kind gift from Masashi Narita (Cambridge Research Institute, CRUK, Cambridge, UK), and the immortalized IMR90-hTERT/ER-RasVal12 cell line was generated as described in (Jeanblanc et al. 2012). Cells were cultured in high glucose DMEM (Gibco D6429, Thermo Fisher Scientific) with FBS 9% (Eurobio), at 37°C, 5% CO2 and 5% oxygen. Routine culture included splitting cells before confluence and regularly checking for mycoplasma infection. For oncogene-induced senescence (RasVSen), RasVal12 overexpression was triggered by incubating the cells with 100 ng/mL 4-hydroxytamoxifen for one week, renewing medium with 4-hydroxytamoxifen every 3 to 4 days. DNA damage senescence (EtoSen) was induced by incubating the cells with 50 µM etoposide for 48 h, then withdrawing the drug for 5 days before starting experiments.

Protein extraction and western blotting
We scraped cells on ice in culture media, collected and centrifuged them at 1000 g for 5 min at 4°C. We resuspended cells in ice-cold PBS and centrifuged them again at 1000 g for 5 min at 4°C. We carefully eliminated supernatant, and we resuspended cells in 2 volumes of PBS with protease and phosphatase inhibitors (B14001 and B15001A/B, Bimake) and 1 volume of Odyssey TBS Blocking Buffer (927-50000, Li-Cor) diluted ½ in TBS, for 1 h at room temperature with gentle shaking. We then incubated membranes overnight at 4°C with primary antibodies in blocking solution. We washed membranes in PBS or TBS with 0.1% Tween-20, and incubated them with NIR-secondary antibodies (Li-Cor) diluted in 0.1% Tween-20 PBS or TBS. We washed membranes in 0.1% Tween-20 PBS or TBS, and imaged them on a Li-Cor Odyssey CLx scanner.
For detection of NKAa1, cells were harvested as described above, then resuspended in an appropriate volume of RIPA buffer with protease and phosphatase inhibitors. After a 30-min incubation on ice, we centrifuged samples at 17000 g for 15 min at 4°C. The supernatant was collected and the required volume of 3X Sample Buffer was added. Samples were incubated at 37°C for 15 min, then loaded on Bis-tris acrylamide gels for electrophoresis. Western blotting was performed as above.
For signaling and autophagy studies, cells were flash-frozen after washing with PBS. PBS was removed, and plates plunged in a dry ice ethanol bath for 1 min (cells were not in contact with ethanol). Plates were incubated at -80°C overnight then proteins were extracted by scraping cells on ice in 1X Sample Buffer containing protease and phosphatase inhibitors. Samples were incubated at 70°C for 10 min and sonicated until viscosity was broken. We loaded proteins onto gels and carried out western blotting as above. See table S1 for the list of antibodies and their dilutions.

EdU incorporation
We seeded 4000 proliferating cells per well in 96-well plates and incubated them with 25 nM EdU (Molecular Probes, Invitrogen, Thermo Fisher Scientific) for 48 h. Cells were fixed with 1% formaldehyde (F1635, Sigma-Aldrich) for 15 min at room temperature. We detected incorporated EdU by click chemistry using the Click-iT EdU Imaging kit (Molecular Probes, Invitrogen, Thermo Fisher Scientific) following the manufacturer's instructions. Images were acquired on a CellInsight CX5 microscope. The percentage of EdU-positive cells in each well was determined after automated nucleus segmentation as described above.

Clonogenicity assay
We seeded 1000 cells per well in a 6-well plate (Corning 353046, Sigma Aldrich) and cultured for two weeks, renewing medium every 3 to 4 days. We used 20 µM etoposide as a negative control to block clone formation. Cells fixed with 2% formaldehyde for 10 min at room temperature, washed once with PBS and incubated with 0.1% crystal violet for 1 h at room temperature. Wells were washed 4x with PBS, and images acquired with a LiCor Odyssey CLx scanner at 700 nm.

Fluorescence microscopy
Cells were seeded in 96-well plates and treated with compounds of interest. Live cells were treated with 200 nM MitoTracker Red CMXRos (M7512, Invitrogen, ThermoFisher Scientific) for 15 min before fixation according to the manufacturer's protocol. Plates fixed with 1% formaldehyde for 15 min at room temperature. Cells permeabilized with 0.5% Triton X-100 for 10 min at room temperature. Blocking with 5% bovine serum albumin (BSA) (A7906, Sigma-Aldrich) and 0.1% Tween-20 (P1379, Sigma-Aldrich) in PBS, 1 h at room temperature. Nuclei stained with 500 ng/mL DAPI (D9542, Sigma-Aldrich) for 10 min at room temperature. Images acquired on a CellInsight CX5 microscope.

Relative plasma membrane potential measurement
We seeded 10000 cells per well in 96-well plates in serum-free MEM containing 500 nM DiSBAC2(3) (Thermo Fisher Scientific) and the various drugs to assess. The plates were incubated for 7 h. 10 µg/mL Hoechst 33342 was added to the cells 30 min before imaging. 10 min before imaging, maximum depolarization was achieved by adding 80 mM potassium Dgluconate (Sigma-Aldrich) in half the wells for each treatment. We imaged cells live with an Operetta screening microscope with a 10X objective. We calculated in each well the mean integrated DiSBAC fluorescence intensity per cell. Within each treatment, this value was normalized to the mean integrated DiSBAC fluorescence intensity in totally depolarized cells with potassium gluconate.
Target knock-down was assessed 48 hrours later by RT-qPCR and/or western blotting. Further treatments were applied as described. Surviving cells in 96-well plates were fixed and counted as described. ATP1A1 siRNAs were purchased from Ambion/Life Technologies (si-NKA1: s1719; si-NKA2: s1720).

RNA extraction, RT-qPCR, and RT-PCR
Cells were scraped on ice in culture media, collected, centrifuged at 1000 g for 5 min at 4°C, resuspended in ice-cold PBS and centrifuged again. Supernatant was eliminated and RNA extraction was done using the NucleoSpin RNA Plus kit (Macherey-Nagel, Hoerdt, France) following the manufacturer's protocol. RNA was eluted in ddH2O and quantified with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). and 50 s at 72°C; then 7 min at 72°C. PCR primers were synthesized by Sigma-Aldrich. PCR primer sequences were: XBP1 forward: 5'-GGAACAGCAAGTGGTAGA-3'; XBP1 reverse: 5'-CTGGAGGGGTGACAAC-3'. We loaded PCR products on a 3.5% agarose gel with 0.00005% ethidium bromide (Gen-Apex) for electrophoresis. We imaged the gel with a Fusion Solo S apparatus (Vilber).  Figure S1. Reversibility of ouabain-induced cytostaticity in Prolif cells. L'Hôte et al.  For all panels, colored overlapping dots represent independent replicates.

Supporting
Supporting Figure 5: siRNA-mediated knockdown of ATP1A1 in BJ cells.
A. Transcriptomics analysis of Illumina beadarray data reveals that a1 is the most expressed a subunit isoform of the Na,K-ATPase pump in BRafSen BJ cells. The NKAa1 protein subunit is encoded by the ATP1A1 gene.
B. ATP1A1 mRNA levels in Prolif, BRafSen, and EtoSen cells, as measured by RT-qPCR. Cells had been induced in senescence for 7 days before RNA extraction.
C. ATP1A1 mRNA levels in BJ cells following transfection with ATP1A1-targeting siRNAs (si-NKA1 and si-NKA2) for 48 h, as measured by RT-qPCR. Virtually 100% of ATP1A1 mRNA levels are effectively depleted by both siRNAs in all cell lines. NT = no target control siRNA. Data were aggregated from two independent biological replicates.
Supporting Figure 6: Total protein levels of Src, Erk, Akt, and p38, during ouabain and MB7 treatment, and ouabain and dasatinib treatment.

A.
As assessed by western blotting. Protein extracts are identical to those displayed in Figure   4A. Cells were treated with 200 nM ouabain or 200 µM MB7. Revert staining: total protein staining.

B.
As assessed by western blotting. Protein extracts are identical to those displayed in Figure   4B. Cells were treated with 200 nM ouabain with or without 10 µM dasatinib for 3 h. Revert staining: total protein staining. A. Cardioglycosides as senolytic screening hits of the Prestwick library (ouabain:

Supporting
supplementary control) in both BRafSen and EtoSen BJ cells. Cells were incubated with drugs at 20 µM for 72 h. Surviving cell count was determined after fixation and nuclei staining using a CX5 screening microscope. Strictly standardized mean difference (SSMD) was calculated for each compound, and a score of SSMD<1.645 (red dotted line) was considered a hit.
B. Dose-response toxicity assay of ouabain in IMR90 cells. Prolif: proliferating (red), RasVSen: RasVal12 senescent (green), EtoSen: etoposide senescent (blue). We expressed survival as the percentage of viable cells remaining attached to the well after incubation with the drug for 72 h, normalized to the initial number of cells at the time of drug addition. Data were aggregated from three independent biological replicates.
C. Dose-response toxicity assay of digoxin in IMR90 cells. Data were aggregated from three independent biological replicates.
For all panels, colored overlapping dots represent independent replicates.