ROLE OF NATURAL KILLER CELLS IN THE PATHOGENESIS OF PRIMARY BILIARY CHOLANGITIS

Background: Primary biliary cholangitis (PBC) is the most prevalent autoimmune liver disease and involves the selective destruction of small intrahepatic biliary epithelial cells (BECs), potentially leading to biliary cirrhosis and liver transplantation, if left untreated. As of now, anticholestatic therapy, in particular Ursodeoxycholic acid (UDCA), stands as the primary treatment approach for PBC, serving as the established first-line therapy irrespective of the disease stage. Unfortunately, up to 40% of PBC patients experience treatment failure with UDCA, facing an elevated risk of disease progression, while the survival rate among biochemical "responders" aligns with that of the general population. Therefore, there is urgent need to better characterize the features of PBC patients and find tools that allow the identification of “non-responders” early in the history of the disease. Research problem: PBC is characterized by immune dysregulation and increasing evidence suggests that innate immune cells play an active role early in the disease. Despite recent studies demonstrated a central role for Natural Killer (NK) cells in the destruction of BECs, a deep and comprehensive characterization of these cells in PBC patients is still lacking, especially in those patients that fails UDCA treatment. Methodology: In this project, we characterized the transcriptomic profile of circulating NK cell subsets in PBC patients and healthy donors (HDs) by single-cell RNA sequencing (scRNAseq); we also investigated differences in the transcriptomic profile of NK cells from PBC patients stratified based on their response to UDCA. By using a 23-color flow cytometry panel, we characterized and compared the frequencies and phenotype of NK cell subsets in the PB and liver from PBC patients and HDs. We also investigated differences in the immunophenotype of circulating NK cells from PBC patients based on their response to UDCA. We also set-up an “in vitro” experimental model to investigate the pathogenesis of the disease, by obtaining iPSC-derived BECs to be co-cultured with autologous PBMCs and LMNCs from end-stage “non-responder” patients and HDs. Key results: When we evaluated the transcriptomic profile of PB NK cells from PBC patients, we observed an overall activated transcriptomic profile compared with HDs, suggestive of higher cytotoxic activity, production of lytic granules and migration to inflammatory sites. We confirmed these results by flow cytometry, demonstrating a higher expression of activation markers and chemokine receptors on circulating NK cells from PBC patients than HDs. We also demonstrated that all subsets of NK cells in “non-responder” patients were characterized by a transcriptomic signature suggestive of higher cellular activation, inflammation, cytotoxic functions and pro-fibrotic profile compared with “responder” patients. Notably, we confirmed an overall activation of PB NK cells from “non-responder” patients also when we analyzed the NK cell phenotype by flow cytometry, as assessed by both manual gating strategies and unbiased PhenoGraph analysis. When we applied PhenoGraph analysis in the liver, we observed that the liver of end-stage “non-responder” PBC patients was infiltrated by a high proportion of liver resident (lr) CD56Bright NK cells that may be likely actively involved in the process of liver cirrhosis though the release of inflammatory cytokines. We also identified a cluster of adaptive-like lr-CD49a+ CD56Bright NK cells enriched in PBC patients that have been reported to have low cytotoxic properties but high cytokine production activity upon stimulation. Conclusions: By demonstrating an activated and pro-fibrotic profile of NK cells in PBC patients who failed first line treatment with UDCA, the results of this study suggest that NK cells, and in particular the CD56Bright NK cell subset, may play a relevant role in the active uncontrolled disease likely through unleashed production of pro-inflammatory cytokines that may sustain and exacerbate the process of fibrosis. If confirmed in further studies, this result may pave the way to novel therapeutic strategies for the treatment of PBC patients unresponsive to UDCA.