An integrated multiomic single-vesicle atlas delineates graft exosome heterogeneity in chronic lung allograft dysfunction

Abstract The underlying mechanisms of exosomes in chronic lung allograft dysfunction (CLAD) remain poorly understood. The primary challenge lies in the substantial functional heterogeneity of exosomes in CLAD. In this study, lung transplantation was initially performed, followed by the generation of a chronic rejection (CR) model in rats to simulate CLAD in humans. Exosomes from lung graft tissues were extracted, purified, and subjected to proximity barcoding assay-based single-vesicle membrane proteomic profiling. Additionally, bulk proteomic and metabolomic profiling of the graft tissues was conducted. Exosomes were categorized into 14 clusters. Compared with the syngeneic group, the CR group had significantly lower levels of clusters 7 (Nphs1 high ), 13 (Aqp1 high ), and 14 (Csf3r high ), and higher levels of clusters 2 (Lrp2 high Mrc1 high ) and 4 (Nt5e high ). The abundance of cluster 12 (Ilk high ) exosomes was strongly correlated with pleural thickness (r = −0.90, P < 0.001). Cluster 12 exosomes exhibited the greatest number of related differentially expressed proteins and metabolites. The proteins correlated with cluster 12 were enriched in the “complement and coagulation cascades”, whereas the metabolites were enriched in the “glycerophospholipid metabolism”. Further verification in patients with restrictive allograft syndrome revealed the same decrease in abundance of cluster 12 exosomes as that observed in the rat model. Finally, the core molecular network of cluster 12 that contributes to pleural thickening in CLAD was constructed. In this study, the heterogeneous composition of exosomes in lung grafts was mapped to expand the understanding of their effects and to drive their application in CLAD.