Abstract PO1-06-07: EMSY enhances cancer stem cell self-renewal and tumorigenesis by reshaping methionine metabolism in triple-negative breast cancer

Abstract Treatment of triple-negative breast cancer (TNBC) remains challenging. Cancer stem cells (CSCs) are the most intractable subpopulation of TNBC cells, which has associated with a high risk of relapse and poor prognosis. However, eradication of CSCs continues to be difficult. Here, we integrated the multiomics data of our large TNBC cohort (n=360) to identify vital markers of CSCs. We discovered that EMSY, inducing a BRCAness phenotype, was preferentially expressed in breast CSCs, promoted the enrichment of ALDH+ cells and was positively correlated with poor relapse-free survival. Mechanistically, EMSY competitively bound to the Jmjc domain, which was critical for the enzyme activity of the KDM5B-specific H3K4 demethylase to reshape methionine metabolism and promote CSC self-renewal and tumorigenesis in an H3K4 methylation-dependent manner. Moreover, EMSY accumulation in TNBC cells sensitized them to PARP inhibitors against bulk cells and methionine deprivation against CSCs. These findings indicate that clinically relevant eradication of CSCs could be achieved with a strategy that targets CSC-specific vulnerabilities in amino acid metabolism. EMSY was highly expressed in BCSCs and positively correlated with TNBC malignancy a. Heatmap with mRNA expression characteristics of non-CSC and CSC subtypes. The top 2,000 most differentially expressed genes used for clustering are plotted. Samples are also annotated on top by the FUSCC subtypes5. b. Volcano plot showing genes targeted by a siRNA inducing a significant change (Fold change (FC) represented the proportion of ALDH+ cells compared to the control)34. c. Overlap analysis of the upregulated genes in CSC subtype (Gene cluster in A) and the genes via siRNA (Gene cluster in B) inducing a significant reduction of the ALDH+ cell ratio. d. The mRNA expression of EMSY in non-CSC and CSC TNBC transcriptomic subtypes (n = 360, Mann-Whitney’s test). Each boxplot showed the median and 95% confidence intervals (95% CI). e. The CNV frequency of EMSY in non-CSC and CSC subtypes. f, g. EMSY mRNA expression of samples with different EMSY copy number in FUSCC (n=302) and TCGA (n=71) TNBC cohorts. h, i. Immunohistochemistry analysis of EMSY expression in specimens of patients between non-CSC (n=45) and CSC (n=20) subtypes. Representative images (h) and the graph of H Scores (i, mean ± SEM, Mann-Whitney’s test) were shown. Scale bar, 4μm (left), 20μm (right). j, k. Kaplan–Meier RFS plots of EMSY in FUSCC (n=360) and GSE234669 (n=442) TNBC (https://kmplot.com/analysis/). Accumulation of EMSY enhanced CSC self-renewal and tumorigenesis a. The mRNA (left) and protein (right) expression of EMSY in FACS sorted ALDH+ and ALDH− cells. b-d. ALDH activity detected by ALDEFLUOR assay and shown as mean ± SEM (n=3/group, one-way ANOVA’s test) in EMSY-knockdown cells (b, c) and EMSY-overexpressing cells (d). e, f. Self-renewal ability determined by primary mammosphere formation (e, n=6 biological independent experiments, one-way ANOVA’s test) and secondary mammosphere formation (f, n=6 biological independent experiments, one-way ANOVA’s test) in EMSY-knockdown SUM149 and MDA-MB-453 cells. Scale bar, 200 μm. g. Self-renewal ability determined by primary mammosphere formation and secondary mammosphere formation in EMSY-overexpressing SUM159 (n=6 biological independent experiments, one-way ANOVA’s test). Scale bar, 200 μm. h-l. EMSY-knockdown SUM149 were injected into mammary fat pads with extreme limiting dilutions (5 x 104, 5 x 103 cells/fat pad). Tumor growth curves (n=5/group, two-way ANOVA’s test) (h, i) were shown. Tumor cell ALDH activity was determined by ALDEFLUOR assay and the bar plot was shown as mean ± SEM (n=5/group, one-way ANOVA test) (j). Frequency of breast CSC was calculated based on the positive tumor sites per group by ELDA (ELDA: Limiting Dilution Analysis for stem cell research (wehi.edu.au)) (k, l). Dairy methionine deprivation, alone or in combination with PARPi, as a therapeutic strategy for EMSY-amplification tumors. a. Immunohistochemistry analyses of EMSY expression in specimens of patients and representative images were shown. b. Patient-derived organoid (PDO) models were performed with fresh isolated primary tissues derived from four independent TNBC patients according to the mRNA level of EMSY. PDO was cultured with fresh medium with (Complete) or without methionine (methionine deprivation, MD) and tumor sphere numbers were determined (n=6/group, two-way ANOVA test). c-e. Effect of MD and olaparib on CSCs and tumor progression. Mice bearing patient-derived xenograft model (PDX, FD-009) were treated with PARPi (50mg/kg, i.p., once every day), dairy MD, and their combination. Tumor volume (c), tumor cell ALDH activity (d) and absolute ALDH+ cancer cells in tumor tissue (e) were assessed (n=5/group, one/two-way ANOVA test). f. Single breast cancer cells from the primary tumors were implanted for a second tumor formation in a limited dilution assay (1000 and 10,000 cells/fat pad); frequency of breast CSC was calculated based on the positive tumor sites per group by ELDA (ELDA: Limiting Dilution Analysis for stem cell research (wehi.edu.au)). Citation Format: Cui-Cui Liu, Ke-da Yu. EMSY enhances cancer stem cell self-renewal and tumorigenesis by reshaping methionine metabolism in triple-negative breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-06-07.