Mechano-metabolism of adherent cells in 2D and 3D microenvironments.

Cells regulate their shape and metabolic activity in response to the mechano-chemical properties of their microenvironment. To elucidate the impact of matrix stiffness and ligand density on a cell's bioenergetics, we developed a non-equilibrium, active chemo-mechanical model that accounts for mechanical energy of the cell and matrix, chemical energy from ATP hydrolysis, interfacial energy, and mechano-sensitive regulation of stress fiber assembly through signaling. By integrating the kinetics and energetics of these processes we introduce the concept of the metabolic potential of the cell that, when minimized, gives experimentally testable predictions of the cell contractility, shape, and the ATP consumption. Specifically, we show that MDA-MB-231 breast cancer cells in 3D collagen gels follow a spherical to spindle to spherical change in morphology with increasing matrix stiffness consistent with experimental observations. This biphasic transition in cell shape emerges from a competition between increased contractility accompanied by ATP hydrolysis enabled by mechano-sensitive signaling, which lowers the volumetric contribution to the metabolic potential of elongated cells and the interfacial energy which is lower for spherical shapes. On 2D hydrogels, our model predicts a hemispherical to spindle to disc shape transition with increasing gel stiffness. In both cases, we show that increasing matrix stiffness monotonically increases the cell's contractility as well as ATP consumption. Our model also predicts how the increased energy demand in stiffer microenvironments is met by AMPK activation, which is confirmed through experimental measurement of activated AMPK levels as a function of matrix stiffness carried out here in both 2D and 3D micro-environments. Further, model predictions of increased AMPK activation on stiffer micro-environments are found to correlate strongly with experimentally measured upregulation of mitochondrial potential, glucose uptake and ATP levels. The insights from our model can be used to understand mechanosensitive regulation of metabolism in physiological events such as metastasis and tumor progression during which cells experience dynamic changes in their microenvironment and metabolic state.