ENGINEERING SPATIOTEMPORAL CONTROL OF THE CELL MICROENVIRONMENT: FROM DYNAMIC HYDROGELS TO BIOPRINTED TUMOR TISSUE ARCHITECTURES

Mechanotransduction, the conversion of extracellular matrix (ECM) mechanics into biochemical responses, is a key regulator of cell behavior in health and disease, e.g. cancer. However, dissecting these processes requires biomaterial systems that mimic the complex 3D ECM while isolating the role of single cues through precise spatiotemporal control. In this thesis, we first present a redox-responsive, softening polyacrylamide hydrogel, made by replacing the standard crosslinker with bisacryloylcystamine, adding reducible disulfide bonds. This system allows gradual in situ stiffness modulation with a mild reductant while preserving continuous cell-matrix coupling. Live-cell imaging on these gels revealed an unknown sequence of mechanotransduction: while peripheral adhesions undergo centripetal remodeling during softening, the fast collapse of subnuclear adhesions is the main trigger for YAP/TAZ nuclear export. These results show a more adaptive force-transfer mechanism than modeled before and identify subnuclear dynamics as an early driver of mechanosignaling. To move beyond 2D, we next aimed to reproduce the native 3D ECM of soft tissues, where its main component, type I collagen, is a key regulator of mechanotransduction. In the breast tumor microenvironment (TME), this is exemplified by tumor associated collagen signatures (TACS), collagen spatial patterns clinically prognostic for metastasis. The second key achievement of this thesis was developing a workflow to program the fibrillar structure of pristine, low-concentration collagen using extrusion bioprinting. This combined a short pre-fibrillation (10°C, 15 min) to nucleate alignable fibrils with an optimized FRESH printing strategy; this overcame core processability limits, enabling fabrication of complex collagen shapes with tuned alignment. Using this system, we applied 3D bioprinting to build increasingly advanced, dual-compartment models of the breast TME and probe how stromal architecture drives tumor progression. Initial models via in-air extrusion showed that a type I collagen stroma, unlike non-fibrillar GelMA bioinks, is crucial for fibroblasts to organize into a tangentially aligned peritumoral shell, a hallmark of in vivo early tumorigenesis. A subsequent coaxial printing strategy enabled the use of more physiological, low- concentration collagen to mimic distinct stromal architectures observed in breast cancer patients. These models showed that collagen architecture acts as a mechanotransductive switch: an ordered, core-shell layout, as in metastasis-free TMEs, promoted cytoplasmic YAP/TAZ, while a disordered, mixed form typical of metastatic tumors triggered its nuclear entry. Finally, the most advanced model used our optimized TRACE printing workflow to build a multilayer collagen shell with a highly fibrillar, circumferentially aligned microstructure. This anisotropy guided fibroblasts into a more effective peritumoral barrier than in amorphous collagen, though quantification was biased by fabrication-related interlayer gaps yielding preferential invasion paths. Overall, this thesis presents a set of engineered platforms, from dynamically responsive materials to advanced bioprinting workflows, that offer fine spatiotemporal tuning of the cell microenvironment. This level of control is instrumental in dissecting the interplay between ECM mechanics and cell behavior, from fundamental time-resolved responses to the structural cues driving cancer progression.