Background/Purpose: Lupus nephritis (LN) is a hallmark of SLE, affecting 50-60% of patients within 10 years of diagnosis. Current treatments for LN have suboptimal response rates and considerable side effects. T cells comprise the majority of infiltrating cells in LN kidney biopsies and contribute to end-organ damage in both SLE patients and murine models. Paradoxically, our recent work showed that kidney-infiltrating T cells (KITs) in murine LN exhibit classic features of T cell exhaustion, including expression of inhibitory receptors, suppressed cytokine production, reduced proliferation and altered metabolism. What remains unknown is whether KITs are a heterogenous or homogenous population. Additionally, prior work suggests that KITs are oligoclonal; however, these studies were completed using low resolution methods only assessing β-chain sequences of the T cell receptor (TCR). Thus, in this study using high throughput single cell RNAseq (scRNAseq) and TCR sequencing, we aim to more definitely define KITs in murine lupus nephritis.

Methods: To explore whether there is heterogeneity amongst KITS we performed scRNAseq using the 10X Genomics platform on flow-sorted kidney and splenic CD4 and CD8 T cells from lupus (MRL/lpr) prone mice. We completed two different comparisons in this study: the first comparing splenic T cells and KITs from MRL/lpr mice, with B6 splenocytes used as naïve controls. In the second study we evaluated KITs from 5 MRL/lpr mice using 5’ libraries to obtain traditional scRNAseq transcripts along with TCR sequences, and cite-seq for PD-1 and Tim3 expression to better evaluate the exhausted T cell population. In all we obtained reads from nearly 30,000 individual cells.

Results: Despite KITs exhibiting a homogenous functional phenotype in our prior work, we found that there is significant heterogeneity in the transcriptional profile of KITs. CD8 KITs were mainly defined by four subgroups: activated (T_EM), exhausted (T_EX), tissue resident (T_RM), and a transitional group. Interestingly, all of these populations were transcriptionally distinct from splenic T cells. Similarly, CD4 KITs exhibited significant heterogeneity. One specific population of CD4 KITs may represent cytotoxic CD4 T cells defined by granzyme and perforin expression. Similar to the CD8 KITs, the CD4 KITs expressed more exhaustion related genes than did their splenic or naïve CD4 counterparts. Pseudotime trajectory analysis suggested that kidney infiltrating cells and specifically T_EX KITs could represent a terminal differentiation state. TCR sequencing analysis confirmed that there is an oligoclonal population of KITs, with single clones representing up to 10% of the CD4 or CD8 T cell compartment in a single mouse. Notably, these clones were more often associated with transcriptional clusters associated with T_EX.

Conclusion: These findings will improve our understanding of T cell function in lupus nephritis. We have uncovered that the KIT population is heterogenous, but also oligoclonal in nature. Furthermore, since clonal T cells are associated with an T_EX transcriptional profile, this supports the hypothesis that chronic antigen exposure and immune activation in the kidney drives T cell exhaustion.

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/kidney-infiltrating-t-cells-in-murine-lupus-nephritis-exhibit-transcriptional-heterogeneity-and-oligoclonal-expansion/