Investigation of the Mesenchymal Manifestations of Tuberous Sclerosis Complex using Tissue-Engineered Disease Models

Abstract

Tuberous sclerosis complex (TSC) is a multisystem tumor-forming disorder caused by biallelic inactivation of TSC1 or TSC2. The primary cause of mortality arises from mesenchymal manifestations in the lung and kidney: pulmonary lymphangioleiomyomatosis (LAM) and renal angiomyolipomas (RAMLs). Despite a well-described monogenic etiology, there remains an incomplete understanding of disease pathogenesis. Consequentially, tractable models which fully recapitulate disease characteristics are lacking. Here, I develop and study novel tissue-engineered models of TSC lung and kidney disease. In my first chapter, I demonstrate that lung-mimetic hydrogel culture of pluripotent stem cell-derived diseased cells more faithfully recapitulates human LAM biology compared to conventional culture on two-dimensional plastic. Leveraging this culture system, I conducted a three-dimensional drug screen using a custom 800-compound library, tracking cytotoxicity and invasion modulation phenotypes at the single cell level. I identified histone deacetylase (HDAC) inhibitors as a group of anti-invasive agents that are also selectively cytotoxic towards TSC2-/- cells. HDAC inhibitor therapeutic effects remained consistent in vivo upon xenotransplantation of LAM cellular models into zebrafish. In my second chapter, I develop a genetically-engineered human renal organoid model which recapitulates pleiotropic features of RAMLs in vitro and upon orthotopic xenotransplantation. I find that loss of TSC1/2 affects multiple developmental processes in the renal epithelial, stromal, and glial compartments. First, loss of TSC1/2 leads to an expanded stroma by favouring stromal cell fate acquisition and alters terminal stromal cell identity. Second, epithelial cells in the TSC1/2-/- organoids exhibit a rapamycin-insensitive epithelial-to-mesenchymal transition. Third, a melanocytic population forms exclusively in TSC1/2-/- organoids, branching from MITF+ Schwann cell precursors of a bona fide neural crest-to-Schwann cell differentiation trajectory. Through these two thesis chapters, I realize the power of tissue-engineered models for the study of TSC. This work offers novel insights into the pathogenesis of RAMLs and identifies a new class of therapeutics suitable for trialing in patients with pulmonary LAM.