Measuring, modelling, and dissecting the cellular mechanical response

While understanding cells\u27 responses to mechanical stimuli is seen as increasingly important for understanding cell biology, the underlying mechanisms of how cells sense and interpret mechanical variables are still unknown. The ubiquity of mechanosensitive processes suggests a physically based mechanism. The combination of rheological measurements and modeling has been used for centuries to study the response of non-living materials to applied stresses. However cell theological measurements are limited by a myriad of confounding effects including: cytoskeletal heterogeneity, ATP-dependent processes and cell regional variations. Here a novel formalism for interpreting microrheology data in the presence of ATP dependent processes is created and a suite of single cell microrheological techniques to control for all of these variables is developed. This approach allows the determination of the consensus mechanical picture of the cell that is suitable for modeling, isolates mechanically distinct sub-cellular structures, evaluates the predictions of cell mechanical models, and identifies key molecular determinants of cellular mechanics. Specifically, two mechanically distinct structures corresponding to the cell cortex and perinuclear region are found. Both regions exhibit shear moduli with similar weak power-law frequency dependence at low frequency, that transition to ω 3/4 at high frequencies. However the cortical region is actin dominated, while the microtubles are a key component of the perinuclear region. Furthermore by analyzing the observed cytoskeletal fluctuations and heterogeneity we find that no current cell mechanics model is sufficient for describing this data. Based on similarities with in vitro systems of actin and cross-linking proteins, we develop a mechanical model based on protein domain unfolding and evaluate its behavior with simulations. A unique state, where proteins accumulate on the cusp of unfolding, suggesting a mechanosensing role, is found. An analogous state is observed in other models that produce rheology similar to that observed cellularly. As the folding state of a protein is readily detected biochemically, we hypothesize that by modulating binding of signaling species, unfolding cross-link domains function as the fundamental biochemical transducers of cytoskeletal deformation