Multiomic analysis of human kidney disease identifies a tractable inflammatory, pro-fibrotic tubular cell phenotype

Abstract

Maladaptive proximal tubular cells have been implicated in failure of repair following renal injury in rodent models, however whether this translates to human kidney disease is unknown. Hence, we integrated snRNA-ATAC-seq with single-cell molecular imaging to generate a multiomic atlas of human kidney disease. In injured kidneys, a subset of tubular epithelial cells acquired an inflammatory phenotype, enriched with pro-fibrotic and senescence markers, analogous to maladaptive cells in mice. Cell neighborhood analysis positioned the inflammatory phenotype adjacent to leucocytes and myofibroblasts and ligand-receptor analysis highlighted paracrine signaling from inflammatory tubular cells to mediate leucocyte recruitment and myofibroblast activation. Loss of an HNF4α-driven gene regulatory network and activation of NF-κβ and AP-1 transcription factors epigenetically imprinted the inflammatory phenotype. Targeting these inflammatory tubular cells by administration of an AP-1 inhibitor or a senolytic agent ameliorated inflammation, expression of senescence-associated transcripts and fibrosis in murine models of kidney injury suggesting these as therapies for human kidney disease.