Poster Presentation 37th Lorne Cancer Conference 2025

Playing by Different Rules: Heterogeneity in Epithelial and Mesenchymal Cancer Cell States Pre-determines Differences in Acute EGF Signalling (#163)

Felix V Kohane 1 , Chantelle Johnstone 1 , Tim Huang 1 , Daniel Neumann 1 , Andrew Gunawan 1 , Christine L Chaffer 2 , John G Lock 1
  1. School of Biomedical Sciences, UNSW, Sydney, NSW, Australia
  2. The Kinghorn Cancer Centre, Garvan Institute of Medical Research, Darlinghurst, NSW, Australia

Oncogenic signalling pathway responses to growth factor stimulation often appear noisy across cancer cell populations, yet the extent to which this variation is predetermined by prior cell state heterogeneity remains unclear1,2. While epithelial-mesenchymal plasticity (EMP) is known to be regulated by oncogenic signals, how diversity in epithelial-mesenchymal (E/M) states influences growth factor signalling responses is not well defined3. We investigated the influence of E/M states on short-term signalling responses in epidermal growth factor (EGF)-stimulated NSCLC cells using quantitative imaging. Cells were stained with DAPI and phalloidin (common markers), and subsets were immunolabelled for p-AKT-473 or -308, p-ERK1, or p-S6 (variable markers). Using representation learning and trajectory inference, we mapped cells across an EMP cell state landscape based on morphological and contextual profiling. Comparisons of EGF signalling in epithelial and mesenchymal extrema revealed that epithelial cells exhibit more robust and sustained signalling responses. Notably, differences in p-AKT response were linked to intracellular localisation, with epithelial cells showing membrane-bound expression, while in mesenchymal cells, p-AKT localised to actin-rich membrane ruffles. To evaluate single cell positions and progression within dynamically evolving, multi-molecular signalling response landscapes, we integrated independently labelled phospho-signalling data across all cells using multivariate regression to achieve ‘computational multiplexing’. Pseudotime analysis of this multiplexed data projected into a signalling state space confirmed significant differences in response magnitudes and kinetics between epithelial and mesenchymal cells. These findings indicate that signalling heterogeneity is largely driven by pre-existing differences in E/M states, particularly through distinct intracellular signal localisation and variations in multi-molecular signal propagation. This has significant implications for therapeutic efficacy in tumours, where heterogeneous E/M states imply varied oncogenic signalling responses to incident growth factor signals, resulting in diverse sensitivities to targeted therapies that modulate these pathways. Our study underscores the necessity of modelling and considering such variation when developing targeted therapeutic strategies.

  1. Filippi S., et al. Robustness of MEK-ERK Dynamics and Origins of Cell-to-Cell Variability in MAPK Signaling. Cell Reports, 15(11), 2524-2535, 2016.
  2. Kramer B., et al. Multimodal perception links cellular state to decision-making in single cells. Science, 377(6606), 642-648, 2022.
  3. Gonzalez D. & Medici, D. Signaling mechanisms of the epithelial-mesenchymal transition. Science Signaling, 7(344), re8, 2015.