UCLA

Stem cell technologies, especially patient-specific, induced stem cell pluripotency and directed differentiation, hold great promise for changing the landscape of medical therapies, and will usher in a new era of personalized medicine. Induced pluripotent stem cells (iPSCs) are patient derived cells that may be expanded indefinitely and differentiate into every known cell type in the body providing the potential basis for personalized organ transplants and disease models. In order to meet these challenges, organoids and 3D tissue engineering approaches are being developed though there is still a large technical gap between promise and current technological expertise.

The work presented in this dissertation is founded on the development of a personalized medicine process flow for modeling Idiopathic Pulmonary Fibrosis (IPF). The basis of this technology is the development of the lung organoid, a 3D cell/hydrogel composite that mimics the alveolar geometry of human distal lung. We generated a model of IPF by culturing organoids with TGF-β1 and showed the resulting scarring in a dish was phenotypically similar to that seen in IPF histology. By optimizing the lung organoid process flow for producing large numbers of uniform organoids and inclusion of IPF patient derived mesenchymal cells we demonstrated how this method could be used for high throughput drug discovery. Finally, we developed an artificial neural network for the classification of high throughput drug screening data and showed its applicability in classifying the complex phenotypic patters organoids demonstrate when treating organoids with dimethyl sulfoxide. In total, this work provides a blueprint for 3D phenotypic drug discovery in the context of the lung organoid model of IPF.