Transcriptomic profiling during normothermic machine perfusion of human kidneys reveals a pro-inflammatory cellular landscape and gene expression signature associated with prolonged delayed graft function after transplantation [bulk RNA-seq]

Assessment and treatment of severe ischaemia-reperfusion-injury (IRI) remains an unmet challenge in kidney transplantation. Normothermic machine perfusion (NMP) recapitulates IRI ex situ, but there is limited understanding of the transcriptional pathways, and the associated cellular landscape, driving IRI during NMP and determining its severity. Such knowledge is essential for therapeutic targeting and organ resuscitation during machine perfusion. Using tissue obtained at the time of NMP from kidneys subsequently transplanted as part of a randomised controlled trial, we undertook in-depth transcriptomic analyses comparing kidneys suffering severe IRI, manifesting clinically as the development of delayed graft function (DGF), to kidneys with mild IRI (defined by immediate graft function, IGF) post-transplantation. We validated upregulation of previously described pro-inflammatory and immune transcriptomic pathways, including TNFa via NFkB signaling, Allograft Rejection and Inflammatory Response. Going further, we identified innate immune system driven processes at the core of the transcriptional signature in kidneys suffering severe IRI, such as recruitment and migration of myeloid leucocytes, macrophage activation, phagocytosis and inflammasome activation. Deconvolution using single-cell-RNAseq data showed kidneys with severe IRI and post-transplant DGF were enriched for pro-inflammatory mononuclear phagocytes, myofibroblasts and fibroblasts, but depleted of tubuloepithelial, cell signatures. These transcriptional findings were recapitulated in tissue biopsies obtained during NMP from an external cohort comparing kidneys with high acute tubular injury and severe IRI to kidneys with low acute tubular injury and mild IRI; these kidneys were histologically similar to the DGF/IGF kidneys, respectively. Together, our study characterises the transcriptional signature of severe IRI during NMP, suggesting the role of pro-inflammatory innate/pro-fibrotic cells in this process. We describe a transcriptomic signature that may support future prospective therapeutic trials as a potential efficacy endpoint, and highlight potential cellular targets for therapeutic intervention during NMP in an era of precision medicine.