Applying single cell profiling to assess drug anti fibrotic properties in the human precision cut lung slice model of fibrosis

Abstract

Much of our understanding of pulmonary fibrosis has been derived from mechanistic observations from animal models and many of the drugs currently studied were developed with very limited preclinical evidence for relevance to the human disease. Here, as part of the Pulmonary Fibrosis Connectome Project we used human precision lung cut slices (PCLS) and single nuclear RNA sequencing (snRNAseq) to determine the effect of several compounds targeting key fibrotic pathways and validating drugs predicted as anti-fibrotic.

Methods

Four PCLS slices per control donor (n = 2) at day 5 after DMSO, FC, FC + treatment were washed in cold 1X PBS and snap frozen. Fibrotic Cocktail (FC) contained TGFb, TNFa, PDGF and LPA, for 5 days, as previously described 1. Treatment included IPF FDA-approved drugs (Nintedanib), as well as drugs currently in clinical trials. Single nuclei suspensions were barcoded using the 10x Chromium Single Cell platform, cDNA libraries generated and sequenced on Illumina platform and integrated together using RCPA. To build the heatmap we calculated the average mean expression of gene of interest within each cell type and across the different conditions.

Results

250,000 single nuclei transcriptomes obtained from 2 donors (2 PCLS per condition, 9 conditions) were analyzed. Fibrotic cocktail stimulation induced the emergence of aberrant basaloid cells (Panel A). No treatment stimulation influenced cell type proportions (Panel B). However, several treatments significantly affected the gene signature of fibroblast and aberrant basaloid cells signature, including Verteporfin, and FDA approved drug targeting YAP-TEAD signaling. Verteporfin led to a significant decrease of ECM or profibrotic genes such as MMP1, TNC, FN1, COL6A6, SERPINE1 and CDH₂ (Panel C). In addition to the gene expression changes, preliminary and ongoing ligand receptor analysis suggests treatment condition significantly modified cellular connectivity of the aberrant basal cells in particular with the alveolar epithelial cells, as compared to the fibrotic condition (Panel D) (Fig. 1).

Conclusion

Applying SnRNA seq to PCLS treated with drugs is an innovative and promising model to assess and potentially predict drug efficacy in human tissues, at a cellular resolution on both abnormal cell populations and cell specific fibrotic gene expression patterns and connectivity.