Amanda R. Beyer

GSEA results in ESR1m models of diferentially expressed siATR and siChk1 RSR defect signature GSEA normalized enrichment scores and significance values in ESR1m models. Gene signatures were developed utilizing differentially expressed genes from siATR and siChk1 knockdown models from McGrail et al as described in Materials and Methods.

Representative gating for flow cytometry analyzing percent BrdU uptake relative to PI in MCF-7 cells analyzed via flow cytometry All treatments were performed in charcoal stripped media after 48 hours of estrogen deprivation in charcoal stripped media followed by 36 hours of Lovastatin (10 µM) to synchronize cell cycles. Gating for BrdU is shown for both WT and Y537S cells untreated or treated with 0.5 mM Hydroxyurea (HU).

Proximity ligation assay of ER-PARP1 co-localization in Y537C cells Representative images of proximity ligation assay ER-PARP1 co-localization puncta in Y537C cells treated for 48 hours with Fulv (100 nM), Olap (5 µM), or Fulv+Olap compared to untreated control (Ctrl) following 5-day estrogen deprivation in charcoal stripped media.

Confirmation of Y537C mutation in LTED cell line a. Scatter plot of individual droplets for Y537C or WT ESR1 confirmed by ddPCR b. Quantitation of ddPCR droplets for Y537C and WT alleles. Positive mutations were detected at a threshold of 3 per 1000 droplets.

ESR1m defective RSR signature shows prognostic association with disease specific survival (DSS) in the METABRIC ER+ breast cancer cohort Kaplan-Meier plot of high (top 50%) versus low (bottom 50%) of patients expressing the ESR1 mutant defective replication stress response signature in the METABRIC ER+ patient cohort. P-values were determined utilizing the log-rank test.

Western blot analysis of cell cycle checkpoint and DDR proteins in Y537S and Y537C cell line models control (Ctrl) vs 24 h treatment with 0.5 mM HU Cells were maintained in charcoal stripped serum media for 48 hours before harvest. Densitometry is indicated under each band and relative to WT cells after normalization to loading control (GAPDH). GAPDH densitometry is displayed relative to WT cells.

MTT growth assay of single treatments with Chk1i (treated with concentrations ranging from 12.5 nM - 5 µM) and Olap (treated with concentrations ranging from 0.25 µM- 64 µM) in WT, Y537S and Y537C models Percent viability was normalized to untreated controls. Graphs represent mean + standard deviation with minimum n=4 at each concentration. IC50s were determined using CompuSyn. Statistics comparing %viability of ESR1m versus WT cells at indicated concentrations are denoted by asterisks in corresponding colors (Y537S=red, Y537C=gold). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001

Development of Fulv and Abema resistant cell lines

Lower concentrations of Olaparib treatment inhibit PARylation but do not show effect on estrogen regulated proteins in ESR1m models Western blot analysis of WT or Y537S cells after 48 hours of estrogen deprivation in charcoal stripped serum media followed by Olap (300 nM, 1 µM, 2 µM) treatment harvested at indicated timepoints in hours. Densitometry is relative to untreated controls within each cell line due to differences in protein expression between cell lines. Densitometry is indicated under each band and was normalized to lane 1 WT control.

The ESR1m defective RSR signature does not predict disease specific or overall survival in the METABRIC ER-negative breast cancer cohort a. Kaplan-Meier plots of disease specific survival (DSS) and b. overall survival (OS) for the ESR1m defective RSR signature in the top 50% versus bottom 50% of signature expressing ER-negative patients (n=440) from the METABRIC cohort.