Transcriptional Landscape of Chronic Lung Allograft Rejection in Humans

Abstract

<h3>Purpose</h3> Chronic lung allograft dysfunction (CLAD) remains the main barrier to long-term survival after lung transplant. The main phenotypes of CLAD are bronchiolitis obliterans syndrome (BOS) with small airway fibrosis, and restrictive allograft syndrome (RAS) with extensive lung parenchymal fibrosis. Mechanisms of CLAD and the pathways active in BOS/RAS remain poorly understood. We hypothesized that bulk RNA sequencing (seq) of human CLAD lungs would help uncover pro-fibrotic pathways and phenotype-specific mechanisms. <h3>Methods</h3> At time of retransplantation, we obtained CLAD lung tissue from 27 BOS and 18 RAS. Negative controls included donor lungs (18) and lobectomy samples done for suspected cancers (14); positive controls came from patients with idiopathic pulmonary fibrosis (IPF) (19). RNA was extracted and submitted for bulk RNAseq at a depth of 50 million paired-end 100-base-pair reads. Differentially expressed genes were detected using DESeq2 and NOIseq. <h3>Results</h3> Top differentially expressed genes in CLAD vs. negative controls included genes involved in cell-matrix interactions (COL10A1, ADAM9) and cell proliferation (CST, MDK). Gene Ontology analysis showed higher expression of genes associated with fibrosis and motility, whereas genes involved in defense and immunity were decreased due to immunosuppression in CLAD vs. negative controls. Analyses to compare RAS to BOS are ongoing. As we anticipated an imperfect correlation between clinical phenotypes and transcriptomic endotypes, we performed unsupervised clustering of all samples. A subset of RAS/mixed samples clustered together and in close proximity to IPF, while other distinct clusters appeared independent of clinical phenotypes: These transcriptomic endotypes will merit further exploration. <h3>Conclusion</h3> This large cohort of lung samples allows a detailed analysis of the lung allograft transcriptome in different phenotypes of CLAD, enabling identification of the top differentially expressed genes and pathways in this poorly understood condition.