The Kinase Inhibitor GNF-7 Is Synthetically Lethal in Topoisomerase 1-Deficient Ewing Sarcoma

Abstract

Background/objectives: Ewing sarcoma (ES), a highly aggressive bone and soft tissue cancer occurring in children and young adults, is defined by the ETS fusion oncoprotein EWS::FLI1. Although event-free survival rates remain high in ES patients with localized disease, those with metastatic or relapsed disease face poor long-term survival odds. Topoisomerase 1 (TOP1) inhibitors are commonly used therapeutics in ES relapse regimens.

Methods: In this work, we used a genome-wide CRISPR knockout library screen to identify the deletion of the TOP1 gene as a mechanism for resistance to topoisomerase 1 inhibitors. Using isogenic cell line models, we performed a high-throughput small-molecule screen to discover a small molecule, GNF-7, which had an IC50 that was 10-fold lower in TOP1-deficient cells when compared to the wild-type cells.

Results: The characterization of GNF-7 demonstrated the molecule was highly active in the inhibition of CSK, p38α, EphA2, Lyn, and ZAK and specifically downregulated genes induced by the EWS::FLI1 fusion oncoprotein.

Conclusions: Together, these results suggest that GNF-7 or small molecules with a similar kinase profile could be effective treatments for ES patients in combination with TOP1 inhibitors or for those patients who have developed resistance to TOP1 inhibitors.

Keywords: Ewing sarcoma; GNF-7; focal adhesion; irinotecan; kinase inhibitor; topoisomerase.

Figure 1

A genome-wide CRISPR screen reveals TOP1 deletion as a primary resistance mechanism to camptothecin. (a) Distribution of RRA scores of sgRNAs enriched in EW8 cells treated with CPT (8 nM) vs. DMSO from a genome-wide CRISPR library screen (n = 4 biological replicates); (b) Western blot of TOP1 expression in EW8 and EW8 TOP1 KD cells; (c–e) dose response curves for EW8 (c), RDES (d), and TC71 (e) cells and their corresponding TOP1 KD cells treated with various doses of CPT for 3.25 d (EW8 and RDES) or 4.25 d (TC71). Data shown are mean ± SD (n = 3–4 replicates). Graphs shown are representative data from one of three to four independent experiments; (f) Volcano plot showing differentially expressed genes in TOP1 KD cells vs. parental EW8 cells (n = 2–3 biological replicates). Blue and red dots represent genes associated with the extracellular matrix and focal adhesions, respectively (DAVID enrichment analysis). Dashed lines at FDR = 0.05 and fold change = ±2; (g) bar graph of normalized enrichment scores (NES) of gene sets related to cell adhesion that are positively enriched in TOP1 KD cells vs. EW8 cells by GSEA analysis. All scores are considered significant (FDR q-value < 0.25).

Figure 2

GNF-7 is preferentially cytotoxic to EW8 TOP1 KD cells compared to parental cells. (a) Scatterplot of areas under the curve (AUC) of dose response curves for EW8 vs. EW8 TOP1 KD cells treated with compounds in the MIPE small-molecule library (n = 2480 compounds); (b) scatterplot highlighting TOP1 inhibitors from the data shown in (a). Dashed line at y = x. (c) Table of the AUC values for each cell line for the ten compounds with the greatest difference in AUC for TOP1 KD cells compared to EW8 cells. Also included are the putative targets of these compounds. (d) Chemical structure of GNF-7 (upper left). (e–g) Dose response curves for EW8, RDES, and TC71 cells and their corresponding TOP1 KD cells treated with various doses of GNF-7 (upper right, lower left, and lower right, respectively) for 4.5 d (EW8 and RDES) or 4.25 d (TC71). Data shown as mean ± SD (n = 3–5 replicates; representative data from one of three to five independent experiments).

Figure 3

GNF-7 induces G1/S cell cycle arrest and reverses ES-associated gene signatures. (a) Levels of caspase-3 and -7 activity in EW8 and TOP1 KD cells after 24 h of treatment with the indicated compounds. Error bars represent SD of two biological replicates. *, p < 0.05; ns, not significant (two-way ANOVA). (b) Cell cycle distribution of EW8 (left) and TOP1 KD (right) cells treated with either DMSO, 4 nM CPT, or 40 nM GNF-7 for 24 h. Numbers within the bars represent the percentage of cells in that phase. (c) Heat map of GSEA NES for gene sets related to ES in EW8 and TOP1 KD cells based on RNA expression after treatment with 40 nM GNF-7 for various times. For (a,b), data shown are representative data from one of two independent experiments.

Figure 4

GNF-7 potently inhibits p38α and several tyrosine kinases. (a) Kinase trees mapping in situ kinase inhibition data from EW8 (left) and TOP1 KD (center) cells treated with 40 nM GNF-7 for 1 h. Circle color and size indicate percent inhibition. Table (right) displays all kinases inhibited >50% in either cell line, with the actual percent inhibition values for both cell lines, as well as the difference in inhibition between the two lines. Kinase tree illustrations reproduced courtesy of Cell Signaling Technology, Inc. (b) Protein expression of phosphorylated p38α (upper left), JNK (lower left), and SRC (upper and lower right) in EW8 and TOP1 KD cells after treatment with 40 nM GNF-7 for various lengths of time. All data normalized to the corresponding unphosphorylated protein and shown as mean ± SD (n = 3). * or ⁺, p < 0.05; ** or ⁺⁺, p < 0.01; *** or ⁺⁺⁺, p < 0.001; **** or ⁺⁺⁺⁺, p < 0.0001 (value compared to 0 h for EW8 (*) or TOP1 KD (⁺) by two-way ANOVA with Dunnett’s multiple comparisons test). ^###, p < 0.001 (TOP1 KD value compared to EW8 at that time point by two-way ANOVA with Šídák’s multiple comparison test). (c) Heat map of the percentage of viable EW8 or TOP1 KD cells 72 h after transfection with siRNAs targeting the kinases identified in (a). Each data point represents the average value of three individual siRNAs.

Figure 5

GNF-7 induces focal adhesions. (a) Phase microscopy images of EW8 cells treated with DMSO or 40 nM GNF-7 for 72 h. Scale bar, 200 μm. (b) Western blot showing FAK phosphorylation at Tyr397 after various lengths of treatment with 40 nM GNF-7. Numbers indicate densitometry values for p-FAK/FAK/GAPDH normalized to EW8 0 h. (c–e) IF staining of actin (red) and vinculin (green) in EW8 and TOP1 KD cells treated with either DMSO for 48 h (c), or with 40 nM GNF-7 for 4 h (d) or 24 h (e). Scale bar, 10 μm.

Figure 6

Combining GNF-7 and a TOP1 inhibitor improves survival in an orthotopic mouse model. (a) Diagram of the treatment schedule used for the in vivo experiment. (b,c) Average tumor volumes (b) and Kaplan–Meier survival curves (c) of mice orthotopically implanted with EW8 cells. Mice were treated with vehicle, irinotecan (1.25 mg/kg), GNF-7 (10 mg/kg) or the combination (n = 8–11 per treatment group). *, p = 0.0388 (Gehan–Breslow–Wilcoxon test). Data shown as average ± SD (b). (d) Representative histology images of EW8 tumors harvested 8 d after treatment initiation. (e) Representative IHC images for γ-H2AX staining from tumors harvested on day 8 of treatment. (f) Quantification of the γ-H2AX IHC data shown in (e) (n = 2–3). *, p < 0.05 (one-way ANOVA with Tukey’s multiple comparison test).