Abstract
Background
Inflammation- and metabolism-related transcriptional signals in lupus nephritis (LN) vary among renal cell populations. We examined these features in a murine kidney single-cell dataset, with additional comparisons in human renal cancer and drug-response experiments in human cell lines.Methods
Four GSE201932 kidney-cortex libraries from female MRL/lpr and C57BL/6 mice at 6 and 10 weeks were reanalyzed. Cell composition, marker expression and gene-set scores were summarized by library, with one library per age-strain combination. Seven programs were also examined in TCGA-KIRC bulk RNA-sequencing data from clear cell renal cell carcinoma (ccRCC). THP-1 and Raji cells were exposed to dexamethasone (100 nM) plus cyclophosphamide (10 μM) for 48 h in three independent experiments, and selected immune transcripts were measured by GAPDH-normalized qRT-PCR.Results
Of 39,074 raw barcodes, 20,761 cells passed quality control. Tubular cells predominated, the Other category comprised 783 cells (3.8%), and several immune populations were small. Mean Stat1, Ifit1 and Ccl5 expression was higher in the LN libraries at both ages, whereas Isg15 changes varied. Interferon (IFN) scores showed no clear overall LN-control separation. Mean oxidative phosphorylation (OXPHOS), fatty-acid oxidation (FAO) and reactive oxygen species (ROS) scores were lower in LN libraries. Five of seven program differences had matching signs in the mouse and TCGA-KIRC comparisons; glycolysis and ROS had opposite signs. Drug exposure was associated with lower CD74 and ACP5 and higher CCL5 expression in THP-1 cells, and lower CD79A and CD74 expression in Raji cells.Conclusions
The murine kidney libraries showed heterogeneous inflammation- and metabolism-related transcriptional features, without a consistent overall difference in IFN scores. Program directions overlapped partly with those in ccRCC, while selected immune transcripts responded differently to drug exposure. Their relevance to LN requires confirmation in independent renal cohorts and matched cell models.