Abstract LB-028: Mechanically tunable 3D microenvironments modulate tumor cell phenotype: Models of mechanotransduction and drug resistance in osteosarcoma

Abstract

Abstract The past few decades have seen marked improvements in survival rates of osteosarcoma due to the advent of modern chemotherapy, radiotherapy, and surgical techniques. However, for patients presenting with metastatic disease, five-year survival rates have remained dismal, hovering around 20%. Repeated failures of clinical trials to confirm potential therapeutic options highlight the need for more accurate pre-clinical testing. Currently, the field of cancer research relies heavily on monolayer culture methods on hard plastic or glass in vitro models; there is a dearth of pre-clinical models that accurately recapitulate the tumor-microenvironment interactions. Matrix stiffness has been implicated in modulating intracellular signaling pathways that promote cancer cell survival, proliferation, and stem cell fate. We have developed a novel three-dimensional (3D) tumor model with variable mechanical properties in order to determine the effect of substrate stiffness and tissue architecture on osteosarcoma cell phenotype, plasticity, and response to therapy. We employed coaxial electrospinning techniques to fabricate highly porous fibrous mesh scaffolds that mimic the bone microenvironment. By controlling the ratio of poly(ϵ-caprolactone) (PCL) and gelatin (PCL:gelatin, core:shell, respectively) in constituent fibers, we were able to manipulate the range of tensile moduli of individual fibers over three orders of magnitude, from 68.91 ± 8.77 kPa to 66.05 ± 7.61 MPa. Osteosarcoma cells cultured in these variable mechanical environments responded by modulating the localization and expression of Hippo pathway regulators. YAP downregulation correlated with decreasing fiber stiffness while both YAP and TAZ had decreasing nuclear:cytoplasmic ratio in less stiff environments. Furthermore, the IGF-1/mTOR axis was downregulated in 3D conditions compared to monolayers and a strong upregulation of Sox2, a stem cell transcription factor, was observed in all 3D conditions. Correspondingly, in the presence of agents targeting the IGF-1/mTOR axis, dose response curves to doxorubicin indicated that IC50 values increase with decreasing substrate stiffness. These phenotypic changes indicate that osteosarcoma cells respond to both stiffness and architecture by modulating the Hippo and IGF-1R/mTOR pathways and increasing cancer stem cell qualities and chemoresistance. We sought to validate our model using osteosarcoma patient biopsies. Analysis of tumor samples from 36 osteosarcoma patients confirmed that YAP/TAZ localization and nuclear pIGF-1R/IGF-1R in our 3D models recapitulated phenotypes observed in patient samples. Our models highlight the need for incorporation of mechanical and architectural cues in the preclinical study of cancer biology as these signals have drastic impacts on osteosarcoma phenotypes and responses to therapy. Citation Format: Eric R. Molina, Letitia K. Chim, Maria C. Salazar, Shail M. Mehta, Brian A. Menegaz, Salah-Eddine Cherradi-Lamhamedi, Tejus Satish, David McCall, Sana Mohiuddin, Ana Maria Zaska, Katherine Jane Grande-Allen, Joseph A. Ludwig, Antonios G. Mikos. Mechanically tunable 3D microenvironments modulate tumor cell phenotype: Models of mechanotransduction and drug resistance in osteosarcoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-028.