Machine learning reveals mesenchymal breast carcinoma cell adaptation in response to matrix stiffness

Epithelial-mesenchymal transition (EMT) and its reverse process, mesenchymal-epithelial transition (MET), are believed to play key roles in facilitating the metastatic cascade. Metastatic lesions often exhibit a similar epithelial-like state to that of the primary tumour, in particular, by forming carcinoma cell clusters via E-cadherin-mediated junctional complexes. However, the factors enabling mesenchymal-like micrometastatic cells to resume growth and reacquire an epithelial phenotype in the target organ microenvironment remain elusive. In this study, we developed a workflow using image-based cell profiling and machine learning to examine morphological, contextual and molecular states of individual breast carcinoma cells (MDA-MB-231). MDA-MB-231 heterogeneous response to the host organ microenvironment was modelled by substrates with controllable stiffness varying from 0.2kPa (soft tissues) to 64kPa (bone tissues). We identified 3 distinct morphological cell types (morphs) varying from compact round-shaped to flattened irregular-shaped cells with lamellipodia, predominantly populating 2-kPa and >16kPa substrates, respectively. These observations were accompanied by significant changes in E-cadherin and vimentin expression. Furthermore, we demonstrate that the bone-mimicking substrate (64kPa) induced multicellular cluster formation accompanied by E-cadherin cell surface localisation. MDA-MB-231 cells responded to different substrate stiffness by morphological adaptation, changes in proliferation rate and cytoskeleton markers, and cluster formation on bone-mimicking substrate. Our results suggest that the stiffest microenvironment can induce MET. Author summary The epithelial-mesenchymal transition (EMT), a process by which epithelial cells lose their cell polarity and cell-cell adhesion and gain migratory and invasive properties to become mesenchymal cells, is believed to play the key role in the initiation of the metastatic cascade. The ability of mesenchymal cells to survive, transform and establish colonies in distant organs is well known but not well understood. To gain insight into this process, we developed a workflow using machine learning and image analysis of cells to examine how morphology (or physical form and structure) and biochemical properties of individual (triple negative) breast carcinoma cells vary in response to their interaction with a substrate of variable stiffness. We found that mesenchymal breast cancer cells responded to different substrate stiffness by morphological adaptation, changes in proliferation rate and cytoskeleton markers such as cell-cell adhesion protein E-cadherin. Surprisingly, the cell properties on the stiffest tested substrate that mimicked bone stiffness changed dramatically-they formed multicellular clusters with distinct E-cadherin localisation at the cell-cell adhesion surfaces. Our results suggest that the stiffest microenvironment can induce cell transitioning from mesenchymal to epithelial cell phenotype.

Automated summary

The study used machine learning and image analysis to examine how individual breast cancer cells respond to substrates of varying stiffness, revealing significant changes in cell morphology and molecular expression. The results suggest that the stiffest microenvironment can induce MET, providing new insights into the process of metastasis.