Bcl-2/Bcl-xL inhibition predominantly synergistically enhances the anti-neoplastic activity of a low-dose CUSP9 repurposed drug regime against glioblastoma

Abstract

Background and purpose: Drug repurposing represents a promising approach to safely accelerate the clinical application of therapeutics with anti-cancer activity. In this study, we examined whether inhibition of the anti-apoptotic Bcl-2 family proteins Bcl-2 and Bcl-xL enhances the biological effects of the repurposed CUSP9 regimen in an in vitro setting of glioblastoma.

Experimental approach: We applied 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide assays to assess cellular proliferation. Annexin V/propidium iodide and tetramethylrhodamine, ethyl ester staining were used to examine apoptosis. Western blotting, RT-PCR, and specific knockdown experiments using siRNA were employed to examine molecular mechanisms of action.

Key results: Bcl-2/Bcl-xL inhibition exerted synergistic anti-proliferative effects across established, primary cultured, and stem-like glioblastoma cells when combined with CUSP9 which had been reduced to only one tenth of its proposed original concentration (CUSP9-LD). The combination treatment also led to enhanced apoptosis with loss of mitochondrial membrane potential and activation of caspases. On the molecular level, CUSP9-LD counteracted ABT263-mediated up-regulation of Mcl-1. Silencing of Mcl-1 enhanced ABT263-mediated apoptosis which indicates that down-regulation of Mcl-1 is crucial for the induction of cell death by the combination treatment.

Conclusion and implications: These data suggest that Bcl-2/Bcl-xL inhibition enhances the susceptibility of glioblastoma cells towards CUSP9, allowing dramatic dose reduction and potentially decreased toxicity when applied clinically. A clinical trial involving the original CUSP doses (CUSP9v3) is currently ongoing in our institution (NCT02770378). The Bcl-2/Bcl-xL inhibitor ABT263 is in clinical trials and might represent a valuable adjunct to the original CUSP.

Figure 1

(a) Established, (b) primary cultured, and (c) stem‐like glioma cells were treated for 144 hr under serum starvation (1.5% FBS) as indicated. Cell viability was determined by MTT assay. Data shown are means with SEM from at least four independent experiments. *P < .05, significantly different as indicated; n.s., non‐significant; one‐way ANOVA with Newman–Keuls test. (d) Chemical structure of ABT263. (e) T98G established and ULM‐GBM‐SC40 stem‐like glioma cells were treated for 72 hr as indicated. Representative microphotographs were taken at 40× magnification. White arrows mark cellular blebs

Figure 2

(a, b) U87MG and T98G glioblastoma cells were treated for 48 hr as indicated under serum starvation (1.5% FBS). Staining with Annexin V/propidium iodide was performed prior to flow cytometric analysis. Representative density plots are shown. Quantitative representation of (c) U87MG and (d) T98G glioblastoma cells treated as described for (a) and (b). Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test . (e) T98G and ULM‐GBM‐PC38 glioblastoma cells were treated for 48 hr with solvent, CUSP9‐LD, 1‐μM ABT263, or the combination under serum starvation (1.5% FBS). Whole‐cell extracts were collected, and Western blot analysis was performed for caspase 3 (C3), cleaved caspase 3 (cC3), caspase 9 (C9), and cleaved caspase 9 (cC9). Actin served as a loading control. Densitometric analysis was performed using ImageJ (NIH, Bethesda, MD; http://imagej.nih.gov/ij). (f) T98G glioblastoma cells were treated for 72 hr with CUSP9‐LD/ABT263 in the presence or absence of the pan‐caspase inhibitor zVAD.fmk. Annexin V/propidium iodide staining was performed prior to flow cytometric analysis. Data shown are means with SEM (N = 5). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test. Representative histograms of (g) T98G and (h) ULM‐GBM‐PC38 cells treated for 48 hr with solvent, CUSP9‐LD, 1‐μM ABT263, or the combination prior to staining with TMRE and flow cytometric analysis. (i, j) Quantitative representation of the fraction of cells treated as described for (g) and (h) showing a reduced mitochondrial outer membrane potential (MOMP). Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test

Figure 3

(a) U87MG and A172 cells were treated for 72 hr as indicated under serum starvation (1.5% FBS). Whole‐cell extracts were collected, and Western blot analysis was performed for Mcl‐1, Bcl‐2, and Bcl‐xS/L. Actin served as a loading control. The arrow head marks the specific band of Mcl‐1. (b) Quantitative representation of the Mcl‐1 expression in U87MG and A172 cells treated as described for (a). Densitometric analysis was performed using ImageJ (NIH, Bethesda, MD; http://imagej.nih.gov/ij). Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test. (c) A172 cells were treated with non‐targeting (n.t.)‐siRNA or Mcl‐1‐siRNA in the presence or absence of 1‐μM ABT263. Staining with Annexin V/propidium iodide was performed followed by flow cytometric analysis. Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test. (d) A172 cells were treated as described for (c). Western blot for Mcl‐1 was performed to confirm sufficient knockdown. (e) T98G cells were treated with non‐targeting (n.t.)‐siRNA or Bcl‐xL‐siRNA in the presence or absence of CUSP9‐LD prior to staining with Annexin V/propidium iodide and flow cytometric analysis. Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; one‐way ANOVA with Newman–Keuls test. (f) T98G cells were treated as described for (e). Western blot analysis for Bcl‐xL was performed to confirm sufficient knockdown. (g) T98G glioblastoma cells were treated for 48 hr as indicated under serum starvation (1.5% FBS). Staining with Annexin V/propidium iodide was performed prior to flow cytometric analysis to determine the percentage of viable (non‐apoptotic) cells. Data shown are means with SEM (N = 4). n.s., non‐significant; one‐way ANOVA with Newman–Keuls test. (h) A172 glioblastoma cells were treated for 7 or 48 hr with solvent, CUSP9‐LD, 1‐μM ABT263, or the combination. RNA was isolated, and RT‐PCR was performed for Mcl‐1. Data shown are means with SEM (N = 4). (i) A172 cells were pretreated for 5 hr with solvent or CUSP9‐LD prior to adding 10 μg·ml⁻¹ of cycloheximide. At indicated time points, whole‐cell extracts were collected, and Western blot analysis was performed for Mcl‐1. Densitometric analysis was performed using ImageJ (NIH; http://imagej.nih.gov/ij). (j) U87MG and A172 cells were treated for 72 hr as indicated under serum starvation (1.5% FBS). Whole‐cell extracts were collected, and Western blot analysis was performed for Mcl‐1. Actin served as a loading control. Densitometric analysis was performed using ImageJ (NIH; http://imagej.nih.gov/ij). Data shown are means with SEM (N = 4). n.s., non‐significant; one‐way ANOVA with Newman–Keuls test. (k) U87MG, A172, and T98G glioblastoma cells were treated for 144 hr under serum starvation (1.5% FBS) as indicated. Cell viability was determined by MTT assay. Data shown are means with SEM (N = 4). *P < .05, significantly different as indicated; n.s., non‐significant; one‐way ANOVA with Newman–Keuls test

Figure 4

(a) U87MG, (b) T98G, and (c) ULM‐GBM‐PC38 cells were seeded on 24‐well plates followed by sequential microscopic imaging (magnification, 10×) over a total time period of 24 hr. Single‐cell tracking was performed using the MtrackJ software (see Section 2). Wind‐rose plots displaying the paths of 15 single cells per treatment condition during the 24‐hr observation period are shown. The tracks were aligned to start from the same initial position to facilitate comparisons. (d–f) Total distance of 15 cells covered within 24 hr per treatment condition. Data shown are means with SEM (N = 4). P<.05, significantly different as indicated; one‐way ANOVA with Newman‐Keuls test (d); Student's t test (e, f). (g) Monolayers of subconfluent ULM‐GBM‐PC38 cells were scratched prior to treatment with either solvent or CUSP9‐LD/1‐μM ABT263 (ABT)/10‐μM NSC23766 (NSC). Microscopic images were taken at time zero and 18 hr after infliction of the scratch. Magnification, 10×. (h) ULM‐GBM‐PC38 cells were treated as described for (g). The area of the open scratch was determined after 6, 12, 18, and 24 hr using ImageJ (NIH, Bethesda, MD; http://imagej.nih.gov/ij). Data presented are means with SD (N = 4). P<.05, significantly different from treated cells; one‐way ANOVA with Newman–Keuls test