RESUMEN
Background: Acute kidney injury (AKI) is frequently observed in critically ill patients and is associated with a poor prognosis. AKI has recently moved into the focus of interest during the SARS-CoV-2 pandemic as high rates of AKI have been reported in severe COVID-19. We aimed to delineate cell type-specific molecular phenotypes associated with human AKI, including COVID-associated AKI. Methods: We analyzed human kidney tissues using histology and single-nuclei RNA sequencing. Samples included kidney biopsies obtained within 2 hours post mortem from patients who succumbed to critical illness with and without evidence of AKI. Samples also included tumor-adjacent normal kidney tissues obtained during surgeries. AKI cases included patients with severe courses of COVID-19 (COVID AKI) and patients with other types of critical illness associated with systemic inflammation (Non-COVID AKI). Postmortem kidney tissues obtained 30 min, 1 hour and 2 hours after death from a brain-dead patient without AKI were analyzed to assess the impact of post-mortem effects. Results: Single-nuclei sequencing from kidney tissues yielded data of high transcriptional depth, which allowed transcriptome-based identification and de-novo spatial reconstruction of kidney cells. Principal component and differential gene expression analyses indicated that the presence of clinically confirmed AKI was the primary driver of global kidney transcriptomes and that different molecular subtypes of AKI existed. In contrast, the sampling time post-mortem and the presence of COVID-19 had minor effects. Subclustering analyses of different kidney cell types identified subclasses of cells representing injured kidney tubular cells, which were marked by distinct biomarker expression and expression signatures signifying intrinsic responses to inflammation, an induction of epithelial-to-mesenchymal transition, and an upregulation of hitheto unrecognized novel receptor-ligand pairs. Conclusions: We provide the first cell type-specific molecular atlas of human AKI, revealing unanticipated disease subtypes and cell type-specific injury patterns.