Abstract
Development of effective treatment strategies for lung tissue destruction as seen in emphysema would greatly benefit from representative human in vitro models of the alveolar compartment. Studying how cellular cross talk and/or (altered) biomechanical cues affect alveolar epithelial function could provide new insight for tissue repair strategies. Preclinical models of the alveolus ideally combine human primary patient-derived lung cells with advanced cell culture applications such as breathing-related stretch, to reliably represent the alveolar microenvironment. To test the feasibility of such a model, we isolated primary alveolar type 2 cells (AEC2s) from patient-derived lung tissues including those from patients with severe emphysema, using magnetic bead-based selection of cells expressing the AEC2 marker HTII-280. We obtained pure alveolar feeder-free organoid cultures using a minimally modified commercial medium. This was confirmed by known AEC2 markers as well as by detection of lamellar bodies using electron microscopy. Following (organoid-based) expansion, cells were seeded on both cell culture inserts and the Chip-S1 Organ-Chip that has a flexible polydimethylsiloxane (PDMS) membrane enabling the application of dynamic stretch. AEC2s cultured for 7 days on inserts or the chip maintained expression of HTII-280, prosurfactant protein C (SP-C), SP-A and SP-B, and zonula occludens-1 (ZO-1) also in the presence of stretch. AEC2s cultured on the chip showed lower expression levels of epithelial-mesenchymal transition-related vimentin expression compared with static cultures on inserts. The combination of a straightforward culture method of patient-derived AEC2s and their application in microfluidic chip cultures supports successful development of more representative human preclinical models of the (diseased) alveolar compartment.