Ultrasoft edge-labelled hydrogel sensors reveal internal tissue stress patterns in invasive engineered tumors

Measuring internal mechanical stresses within 3D tissues can provide important insights into drivers of morphogenesis and disease progression. Cell-sized hydrogel microspheres have recently emerged as a powerful technique to probe tissue mechanobiology, as they can be sufficiently soft as to deform within remodelling tissues, and optically imaged to measure internal stresses. However, measuring stresses at resolutions of ∼10 Pa requires ultrasoft, low-polymer content hydrogel formulations that are challenging to label with sufficiently fluorescent materials to support repeated measurements, particularly in optically dense tissues over 100 μm thick, as required in cancer tumor models. Here, we leverage thermodynamic partitioning of hydrogel components to create "edge-labelled" ultrasoft hydrogel microdroplets, in a single polymerization step. Bright and stable fluorescent nanoparticles preferentially polymerize at the hydrogel droplet interface, and can be used to repeatedly track sensor surfaces over long-term experiments, even when embedded deep in light-scattering tissues. We utilize these edge-labelled microspherical stress gauges (eMSGs) in inducible breast cancer tumor models of invasion, and demonstrate distinctive internal stress patterns that arise from cell-matrix interactions at different stages of breast cancer progression. Our studies demonstrate a long-term macroscale compaction of the tumor during matrix encapsulation, but only a short-term increase in local stress as non-invasive tumors rapidly make small internal reorganizations that reduce the mechanical stress to baseline levels. In contrast, once invasion programs are initiated, internal stress throughout the tumor is negligible. These findings suggest that internal tumor stresses may initially prime the cells to invade, but are lost once invasion occurs. Together, this work demonstrates that mapping internal mechanical stress in tumors may have utility in advancing cancer prognostic strategies, and that eMSGs can have broad utility in understanding dynamic mechanical processes of disease and development.