Fig 1: Ligand prediction of HSC activation and identification of parathyroid hormone as a novel ligand.(A) Schematic overview: The NicheNet ligand prediction is based on prior knowledge of ligand-receptor interactions, intermediate signalling steps and TF target gene interactions from all possible cell types. This prior knowledge was complemented with HSC-specific TF target gene interactions as found by the pySCENIC algorithm to predict the ligands inducing the differentially expressed genes of myofibroblasts compared to qHSCs. (B) Downstream target genes for the top 10 ligands inducing the differentiation from qHSCs to myofibroblasts. (C) Top 10 predicted ligands based on merged HSC activation atlas (y-axis) and the rank of these ligands when top ligands are predicted in each liver injury dataset separately (x-axis). (D) Normalized expression of the parathyroid hormone receptor Pth1r in HSCs and hepatocytes based on GSE192742.26 (E) Schematic overview of experimental setup testing the effect of parathyroid hormone and TGF-β on primary mouse liver spheroids. HNF4A and Desmin expression at day 7 in mouse HSC-hepatocyte spheroids. (F) Sirius-red staining of spheroids treated with 80 nM parathyroid hormone, 5 ng/ml TGF-β or solvent. (G) Quantification of Sirius red-positive area in the spheroids. The difference between the treatments was tested with a one-tailed paired t test. ∗p <0.05, ∗∗p <0.01, n = 5. (H) Relative expression of HSC activation genes in spheroids treated with 80 nM parathyroid hormone or 5 ng/ml TGF-β. The difference between the treatments was tested with a one-tailed paired t test. ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, n = 5. (q)HSCs, (quiescent) hepatic stellate cells; TFs, transcription factors; PTH, parathyroid hormone.