DESIGN AND IMPLEMENTATION OF A 3D BONE MARROW HYDROGEL

Background: The extracellular (ECM) matrix is a key mediator in the preferential metastasis of breast cancer cells to bone, lung, liver and brain tissues. Though in vivomouse models are widely used in breast cancer studies, metastasis in mice does not mirror the same preferential spread observed in humans. This discrepancy, in addition to others, limits our ability to parse out the important features that make these tissues hospitable secondary sites. Here, we combined bioinformatics and mechanical tissue characterization to adapt a 3D polyethylene glycol hydrogel (PEG) to recapitulate the integrin binding, matrix degradability, and bulk stiffness of different tissues. We have used these methods to make a bone marrow-like hydrogel to better study the ECM features driving breast-to-bone metastasis. Methods:An 8-arm star PEG is functionalized with 20 different peptide sequences that can degrade in the presence of cell-secreted enzymes or that bind to cell surface integrins. These biochemical features were identified and quantified using an algorithm developed with data from the Protein Atlas. A competitive binding assay validated cell binding to each integrin peptide and an outgrowth assay validated degradation for each MMP peptide in human mesenchymal stem cells (hMSCs) from three donors. The effective Young’s modulus of the hydrogel was matched to the stiffness of intact porcine marrow (4.4±1.0kPa). Cell differentiation was quantified using Oil Red O and OsteoImage stains. Results:Both marrow and our hydrogel can be modeled as elastic materials, validating the use of PEG to mimic the bulk mechanics of marrow tissue. Mass spectrometry on human marrow had an 85% match to the proteins found using our algorithm, indicating it can be used to identify ECM signatures in tissue. We show that hMSCs encapsulated in our hydrogel for 21 days do not spontaneously differentitate. When provided differentiation media, hMSCs in our hydrogel have the best capacity to differentiate into both fat and bone cells, where other culturing platforms have a preference for one lineage of differentiation (TCPS, 2D PEG hydrogel, 3D-RGD functionalized hydrogel). This highlights a biological need for systems that capture both physiological protein and tissue stiffness. Conclusion: We used fundamental tissue characterization to create a novel in vitro platform. This platform can be used to better understand the role of the ECM in breast-to-bone metastasis, representing a significant opportunity to advance the cancer field by coupling advanced biomaterials with quantitative biology.