A18 Culture of Immortalized Human Alveolar Epithelial Cells in 2D and 3D to Model Lung Adenocarcinoma Progression in Vitro

Abstract

Lung cancer is the leading cause of cancer death in the United States. Lung adenocarcinoma (LUAD) is the most common histologic subtype, arising from epithelial cells of terminal respiratory units called alveoli. The overall 5-year survival of lung cancer remains low at 19%. There is an urgent need to understand early events in LUAD development, as well as to develop new 1st- , 2nd- , and 3rd-line targeted therapies. Human organoids are powerful research tools for the molecular and mechanical manipulation of genetically diverse cells without exposing human subjects to treatment. Establishment of normal cell lines from human alveolar epithelial cells (AECs) has remained challenging due to the difficulty of growing primary cells in long-term culture. Human alveolar organoids would provide a powerful tool to 1) study LUAD development, progression, and drug resistance; 2) screen for new therapeutics; and 3) study the effects of environmental exposures on AECs. Optimize growth and genetic conditions to derive human AEC lines from purified primary cells for the stepwise modeling of LUAD. We tested different immortalization strategies using primary purified AECs to determine which condition allowed cells to continue proliferating while retaining their epithelial phenotype in two-dimensional (2D) culture and their ability to form spheroids in three-dimensional (3D) culture. Using purified primary AECs from three deceased, deidentified human subjects, we found that the initial propagation of AECs in media containing Y-27632 and subsequent transduction with Simian virus 40 Large T antigen allowed cells to divide readily in 2D as a monolayer, while expressing epithelial marker E-cadherin but not mature lung genes. When placed in 3D Matrigel culture with fibroblasts, these cells form multilobulated structures expressing mature AEC markers, reminiscent of the peripheral lung. We are currently optimizing this system to allow stepwise modeling of LUAD.