T Cell-Receptor (TCR)-Independent Activation of CD8 Effector T Cells by IL-15 and IFNβ Drives Kidney Damage in Lupus Nephritis

Background: Lupus nephritis (LN) implies morbidity in Systemic Lupus Erythematosus (SLE). A T cell-rich infiltrate parallels deposition of antibodies. How infiltrating T cells contribute to kidney damage is unclear. They can also be observed in urine, which thus can be used as a “window into the kidney” to investigate the cellular pathogenesis of LN. We report analysis of kidney infiltrating T cells found in urine to test if a renal autoimmune reaction contributes to tubulointerstitial inflammation (TII). Methods: Urinary T and tubular epithelial cells (TEC) were quantified by flow cytometry (FC) in LN patients. T cells were subjected to single cell RNA sequencing after FC sort from urine and blood of five LN patients at flare. Comparison of gene transcription revealed putatively pathogenic cells. 67 T cell receptors (TCR) from expanded clonotypes were cloned to test specificity; T cell subsets from blood were subjected to in-vitro experiments. T cell stimulating cytokines were determined in urine from LN and healthy donors, reflecting the renal milieu. Results: Majority of urinary T cells in LN are of CD8 lineage, the amount of which correlates with urinary TEC, a proxy for kidney damage. Mainly being of effector memory phenotype they are recruited partially from a CX3CR1 positive subset in blood which has a of type I interferon (IFN) signature. TCR analysis revealed expanded CD8 clones, none of which reacted with autologous TEC (0/18). Reactivity to viral CMV and EBV epitopes was shown for 15/67 clones. CMV or EBV was not detectable in respective kidney biopsies. In kidney tissue and/or urine we detected elevated levels of T cell stimulating cytokines, like IL-15. In-vitro, stimulation of CD8 T cells with IL-15 and IFNβ caused TCR-independent activation, degranulation and secretion cytokines TNF and IFNγ, almost exclusively by CX3CR1 positive CD8 effector T cells. Conclusions: Contribution of CD8 T cells to TII and TEC damage in LN is antigen and TCR independent. In-vivo stimulation with IL-15 and type I IFN may be sufficient to trigger effector functions of a CX3CR1 positive subset. Targeting these cells in the blood, their migration, their response to cytokines or reducing renal cytokine levels may prevent LN flares and represent a tailored treatment.