Dissecting Context-Specific Effects of ERK5 Signaling in Triple-Negative Breast Cancer

Triple negative breast cancer (TNBC) is an aggressive subtype of cancer with poor clinical outcomes. There is a critical need to identify novel, druggable targets for TNBC to improve therapy response and patient outcomes. Kinases are attractive therapeutic targets due to their roles in cell fate, and irregular kinase activity drives tumor progression. The role of extracellular signal-regulated kinase 5 (ERK5) in mediating TNBC extracellular matrix (ECM) has previously been described in 2D culture and in vivo. Here, we characterized the impact of ERK5 on breast cancer biology in 2D culture, 3D spheroids, and our 3D breast adipose-macrophysiological system (BA-MaPS). Methods. We assessed migration changes in MDA-MB-231 parental and ERK5-knockout (ERK5-ko) cells cultured in the three in vitro models using transwell, scratch, and spheroid pseudo-migration assays. Differential gene expression among these cell lines in the three platforms was evaluated by RNA sequencing and pathway analysis. Stromal remodeling of adipocytes and matrix was evaluated by H&E and Masson's Trichrome. Results Across the in vitro models, ERK5 deletion impaired TNBC cell migration. ERK5-mediated transcriptomic changes included genes associated with epithelial-to-mesenchymal transition (EMT) and migration, with further analysis showing significant alterations in core and associated matrisome. Histological staining corroborated downregulation of collagen with ERK5 depletion in the BA-MaPS. The NF?B pathway was significantly upregulated only in the ERK5-ko 2D-cultured cells, not in 3D spheroids nor the BA-MaPS model. Conclusions. These results indicate a link between ERK5 and TNBC progression through regulation of TME remodeling, EMT, and cell motility. Differences in 2D culture, 3D spheroid, and BA-MaPS underscore the importance of using physiologically relevant models in breast cancer research. Overall design: TNBC cell line MDA-MB-231 was acquired from American Type Culture Collection (ATCC). ERK5-ko cells were previously generated using a pU6-driven guide strand with dual expression cassettes for Cas9/EGFP approach (Hori-zon, Cambridge, UK). For all studes MDA-MB-231 cells and MDA-MB-231-ERK5-KO cells were used. For 2D culture, all cell lines were grown to 80% confluency in a T25 flask with Dulbecco's Modified Eagle Medium (Invitrogen, Carlsbad, California, USA, DMEM; pH 7.4) supplemented with 10% Fetal Bovine Serum (Thermo Scientific, Waltham, Massachusetts, USA, FBS), 1% nonessential amino acids, minimal amino acids, sodium pyruvate, antibiotic/anti-myotic, and insulin (complete media) under mycoplasma-free conditions at 37°C in humidified 5% CO2 (n=3). At 80% confluence, cells pelleted for RNA extraction. For 3D studies, two million MDA-M-231 cells of each cell line were seeded in an AggreWell 400 24 well plate (Stem Cell Technologies, Vancouver, BC, Canada) that was prepped with Anti-Adherence solution (Stem Cell Technologies, Vancouver, BC, Canada) in accordance with the manufacturer's protocol and grown overnight in the same media used in 2D experiments to form 1200 3D spheroids containing about 1700 cells/spheroid (n=3). 24 h later, spheroids were pelleted for RNA extraction. To assemble the BA-MaPS, 500,000 cells were seeded into 250µL of minced breast tissue (n=4 different donor breast tissues) and assembled using a plunger method to sandwich the breast tissue between two confluent brASC cell sheets (n=4). BA-MaPS was collected after 5 days of cultured in MEM Alpha (Thermo Scientific, Waltham, Massachusetts, USA, #12571063) supplemented with 10% Fetal Bovine Serum Premium Select (R&D Systems, Minneap-olis, Minnesota, #S11550) and 1% antibiotic/anti-myotic (ASC media) at 37°C in hu-midified 5% CO2. BrASCs were used between passages 1-5. At end point, whole construct was collected and RNA was extracted.