Chimeric protein EWS::FLI1 drives cell proliferation in Ewing Sarcoma via aberrant expression of KCNN1/SK1 and dysregulation of calcium signaling

Abstract

Ewing sarcoma (ES) is characterized by EWS::FLI1 or EWS::ERG fusion proteins. Knowing that ion channels are involved in tumorigenesis, this work aimed to study the involvement of the KCNN1 gene, which encodes the SK1 potassium channel, in ES development. Bioinformatics analyses from databases were used to study KCNN1 expression in patients and cell lines. Molecular approaches and in vitro assays were used to study the transcriptional regulation of KCNN1 and its involvement in the regulation of ES cell proliferation. KCNN1 is overexpressed in ES patient biopsies, and its expression is inversely correlated with patient survival. EWS::FLI1, like EWS::ERG, promotes KCNN1 and SK1 expression, binding to GGAA microsatellites near the promoter of KCNN1 isoforms. KCNN1 is involved in the regulation of ES cell proliferation, with its silencing being associated with a slowing of the cell cycle, and its expression modulates membrane potential and therefore calcium flux. These results highlight that KCNN1 is a direct target of EWS::FLI1 and EWS::ERG and demonstrate that KCNN1 is involved in the regulation of intracellular calcium activity and ES cell proliferation, making it a promising therapeutic target in ES.

Fig. 1. Expression of KCNN1 in biopsies from ES patients and in ES cell lines.

A Gene expression (Log2 normalized expression) of genes encoding SKCa channels (KCNN1, KCNN2 and KCNN3) in ES patients (117 patients) (Brown–Forsythe and Welch Anova test). B Gene expression (Log2 normalized expression) of KCNN1 in biopsies from patients with different tumors (Kruskal–Wallis test). C Overall survival probability (Kaplan Meier curve) of ES patients according to KCNN1 expression (high or low). The Bonf p corresponds to the p-value after the Bonferroni correction for multiple comparisons. Data are derived from the GSE17679 (see Supplementary Table S2). The method used to define high or low expression of KCNN1 is performed using the Kaplan Scan where an optimum survival cut-off is established based on statistical testing (see R2: Genomics Analysis and Visualization Platform)”. D Clinical tumor samples of 18 ES patients were fixed, embedded in paraffin, sectioned and immunostained with SK1 antibody. One representative photomicrograph is shown. E Expression of KCNN1 in normal tissues from GTEx database. TPM transcripts per million. F Gene expression (Log2 normalized expression) of KCNN1 in Mesenchymal Stem Cells (MSCs) coming from thirty healthy donors (30 donors) and in patients suffering from ES (117 patients) (Unpaired t-test). G Expression of KCNN1 in tumor-derived cell lines. *p < 0.05; **p < 0.01; ***p < 0.001 and ****p < 0.0001. (In A, B, F, the boxes boundaries indicate the first and third quartiles - and the midline represents the median – of genes expression. The whiskers indicate the fifth and the ninety-fifth percentiles. In E, G, boxes represent the interquartile range, upper and lower whiskers the largest and smallest values, respectively. TPM transcript per million).

Fig. 2. Regulation of KCNN1 by the chimeric protein EWS::FLI1.

A KCNN1 expression (Log2 normalized expression) in A-673 cell line stably transduced with a doxycycline-inducible shRNA targeting EWS::FLI1 in control condition (−Doxycycline) or after doxycycline treatment (+Doxycycline). B. Normalized and relative EWS::FLI1 (left panel) and KCNN1 (middle panel) gene expressions in A-673/TR/shEF cell line after doxycycline treatment (1 µg/mL) (N = 5). Western-blots of SK1 and EWS::FLI1 (right panel) in A-673/TR/shEF cell line after doxycycline treatment (1 µg/mL) (N = 2). C Normalized and relative EWS::FLI1 and KCNN1 gene expressions in A-673 (N = 3), RD-ES (N = 4), and SK-ES-1 (N = 3) cell lines after 48 h of transient transfection with a pool of siRNAs targeting EWS::FLI1 (siRNA EWS::FLI1) or a siRNA control (siRNA CT) (upper panels). D Western blots of SK1 and EWS::FLI1 in A-673, RD-ES and SK-ES-1 cell lines after 72 h of transient transfection with a pool of siRNA targeting EWS::FLI1 (siRNA EWS::FLI1) or a siRNA control (siRNA CT) (N = 2) (lower panels). E Different KCNN1 transcripts according to their position on the genome. Brown rectangles correspond to transcribed and translated exons, while the salmon rectangles correspond to transcribed but not translated exons. White spaces are intronic sequences. F Normalized expression of KCNN1 transcripts (FPKM) in A-673 cell line (N = 3). G Normalized expression of KCNN1 transcripts (FPKM) in A-673/TR/SHEF cell line at day 0 and day 7 of doxycycline treatment (1 µg/mL). *: p < 0.05; **: p < 0.01 (Mann–Whitney tests). (In A, the boxes boundaries indicate the first and third quartiles - and the midline represents the median – of genes expression. The whiskers indicate the fifth and the ninety-fifth percentiles. In B, C, bars indicate means ± SD of relative and normalized EWS::FLI1 or KCNN1 mRNA expression determined by RT-qPCR. In E, F, FPKM Fragment Per Kilobase Per Million reads.).

Fig. 3. Fixation of EWS::FLI1 and EWS::ERG near KCNN1 promoter.

A H3K4me3, H3K27ac and FLI1 ChIP-Seq data with quantitative value of peaks around KCNN1 promoters and sequences of EWS::FLI1 binding sites in A-673 (left panel) and TC106 (right panel) cell lines. B H3K27ac ChIP-Seq of A-673/TR/shEF cell line at day 0 and day 7 of doxycycline treatment with quantitative value of peaks around KCNN1 promoters. C FLI1 ChIP-Seq of A-673/TR/shEF cell line with quantitative value of FLI1 peaks around KCNN1 promoters, at day 7 of doxycycline treatment, and 3 days (day 7 + 3), 4 days (day 7 + 4), 7 days (day 7 + 7) and 10 days (day 7 + 10) after discontinuation of doxycycline treatment. D H3K27ac and FLI1 ChIP-Seq data on MSCs transfected with control plasmid, FLI1 overexpression plasmid or EWS::FLI1 overexpression plasmid around KCNN1 promoters.

Fig. 4. Effect of KCNN1 knockdown on ES cell proliferation.

A KCNN1 expression in A-673 cell line stably transduced with doxycycline-inducible shRNA control cells (shRNA CT) or 3 shRNA cells targeting KCNN1 (shRNA 511, shRNA 908 and shRNA 576) after 72 h of treatment with doxycycline (0.5 µg/mL). Bars indicate mean ± SD of relative and normalized KCNN1 mRNA expression obtained after RT-qPCR (N = 5 for shRNA CT and shRNA 908, N = 4 for shRNA 511 and N = 3 for shRNA 576). B Clonogenicity assays of A-673 cell line stably transduced with shRNA control (shRNA CT) or three shRNA targeting KCNN1 cells (shRNA 511, shRNA 908 and shRNA 576) after 72 h of doxycycline treatment (0.5 µg/mL), with the colonies numbers observed and the relative surface area of clones. Bars indicate mean ± SD of counted colonies or the clones surface area observed on the left panel relative to the untreated cells (N = 6 for shRNA CT, N = 3 for shRNA 511 and shRNA 908 and N = 7 for shRNA 908). C Cell cycle assays on shRNA 908 cells treated (shRNA 908 +Doxycycline) or not (shRNA 908 −Doxycycline) with doxycycline 0, 6, 8 or 10 h after thymidine double block. Dark blue: G0/G1 phase, Light blue: S phase, Purple: G2/M phase. D Percentage of cell distribution 6, 8, and 10 h after thymidine double block (N = 5 for 6 and 10 h after thymidine double block; N = 6 for 8 h after thymidine double block, Wilcoxon test). E Western blots of p21 and Aurora A and their quantification normalized to β-tubuline 10 h after the thymidine double block (N = 3). *p < 0.05; **p < 0.01 (A, B, D, E: Mann–Whitney test).

Fig. 5. Effect of KCNN1 knockdown on membrane potential and calcium flux.

A Measures of the plasma membrane depolarization in A-673 cell line stably transduced with shRNA control cells (shRNA CT) or three shRNA cells targeting KCNN1 after 7 days of doxycycline treatment (0.5 µg/mL). (N = 48 for shRNA CT and shRNA 908, N = 46 for shRNA 511, N = 45 for shRNA 576). B Membrane potential measures of shRNA CT or shRNA 908 spheroids treated 7 days with doxycycline (0.5 µg/mL) (N = 7). C Plasma membrane depolarization measures in A-673 cells treated with the SK1 channel activator GW542573X (N = 50). D Effect of KCNN1 knockdown on SOCE in A-673 cell line stably transduced with shRNA control (shRNA CT) or three shRNA targeting KCNN1 and the relative fluorescence measurement of SOCE induced by TG (N = 26). E Effect of KCNN1 knockdown on CCE using Mn²⁺ quenching assay and the normalized Mn²⁺ quenching slope (N = 21 for shRNA CT, N = 20 for shRNA 511 and N = 22 for shRNA 908 and shRNA 576). F Effect of the SK1 channel activator GW542573X on SOCE in A-673 cells and the relative fluorescence measurement of SOCE induced by TG in GW542573X conditions (N = 14 for Control and N = 21 for G542573X condition). G Effect of GW542573X on CCE using Mn²⁺ quenching assay and the normalized Mn²⁺ quenching slope (N = 28 for Control and N = 20 for GW542573X condition). *p < 0.05; **p < 0.01; ***p < 0.001 (boxes indicate the first and third quartiles—and the median inside—of the different measures. The whiskers indicate the fifth and the ninety-fifth percentiles. A, D, E Anova test; B, F, G Mann–Whitney test and C: Unpaired t-test).

Fig. 6. Effect of plasma membrane potential variation on calcium influx and colony formation in ES cells.

A Measurement of plasma membrane potential in A-673 cells cultured in potassium-enriched (40 mM KCl) or non-potassium-enriched (40 mM NaCl) medium (N = 25). B Effect of 40 mM KCl (high K⁺) on SOCE in A-673 cell line and the relative fluorescence measurement of SOCE induced by TG (N = 16). C Effect of 40 mM KCl (high K⁺) on CCE using Mn²⁺ quenching assay and the normalized Mn²⁺ quenching slope (N = 15). D Clonogenicity assays of A-673 cell line in presence of high concentration of extracellular potassium (40 mM, High K⁺) or low concentration of extracellular potassium (5 mM, low K⁺), with the colonies numbers observed and the relative surface area of clones. Bars indicate mean ± SD of counted colonies or the clones surface area observed on the left panel relative to the untreated cells (N = 3). Bars indicate means ± SD of colonies numbers and the surface occupation. *: p < 0.05; **: p < 0.01; ****: p < 0.001. In B and C, boxes indicate the first and third quartiles - and the median inside - of the different measures. The whiskers indicate the fifth and the ninety-fifth percentiles. A, B, C, and D: Mann–Whitney test.