Dynamic mechanical cues in 2D and 3D matrix mechanobiology: from static to time-resolved bioengineered platforms

The extracellular matrix (ECM) is a dynamic regulator of cell behaviour, continuously changing its biophysical state, including mechanical properties, architecture and ligand presentation, during development, homeostasis, repair and disease. However, conventional in vitro models stimulate cells through static and simplified signals, limiting investigation of how cells decode the temporal features of their surroundings. This work examines dynamic two-dimensional (2D) and three-dimensional (3D) bioengineered platforms designed to reproduce time-changing biophysical signals and their mechanobiological consequences. In dynamic 2D interfaces, tuneable stiffness, cyclic deformation, fluid flow and shear stress, reversible ligand presentation, and switchable topographies or curvature enable controlled delivery of mechanical, biochemical and architectural cues. These systems show that cellular responses, including adhesion organization, cytoskeletal remodelling, cell polarity, nuclear morphology, chromatin organization and mechanoresponsive signalling, depend not only on cue magnitude, but also on rate, frequency, duration, waveform, persistence and mechanical history. We also discuss how dimensionality reshapes the interpretation of dynamic cues. In 3D matrices, cells experience confinement, multidirectional force transmission, matrix degradability, pore architecture and cell-mediated remodelling, making mechanotransduction fundamentally more complex than in planar cultures. Organ-on-chip devices, bioprinted matrices and responsive hydrogels integrate ECM-like architectures with perfusion, mechanical actuators and functional readouts, enabling tissue remodelling to be reconstructed as a controllable experimental variable. Overall, dynamic matrix mechanobiology shifts the focus from fixed material properties to the temporal logic of cell–ECM interactions. More predictive in vitro models will require not simply increasingly complex matrices, but platforms that combine physiological dynamics with sufficient experimental control to identify mechanobiological mechanisms.