A population-scale atlas of blood and tissue in lupus nephritis

Lupus nephritis (LN), a severe manifestation of Systemic lupus erythematosus (SLE), is a heterogeneous disease driven by diverse immune and tissue cell types. Current treatments of LN are non-specific, increase risk of infection, and have high rates of relapse. To better understand LN, we obtained 538K single-cell and 143K single-nuclear profiles from kidney biopsies of 156 LN patients and 30 pre-implantation transplant biopsy controls, along with 327K single-cell blood profiles from many of these patients. We built an atlas of kidney tissue cell types in LN encompassing 55 immune and 36 tissue states. Integrating renal immune single cell data with blood data enabled the identification of tissue-specific immune cell states. For example, within the myeloid compartment, plasmacytoid dendritic cells (DC20) were shared across tissue and blood, while macrophage cell states were tissue-specific (M5-14). We identified LN pathological features associated with cell states. Specifically, we examined the chronicity index, which quantifies irreversible tissue damage, and the activity index, which measures active reversible inflammation. Increasing chronicity index tracked with the most dramatic changes in both tissue and immune cell states. The proportion of injured or degenerating proximal tubule cells expanded with chronicity, reflecting ongoing tissue damage. Increasing chronicity index was also associated with an expansion of tissue-specific CD56 ^bright NK cells (NK1), GPNMB ^high NUPR1 ^high Macrophages (M5), and CLEC10A ^low cDC2 (DC17). Importantly, most expanded immune populations associated with chronicity index were tissue-specific states. After adjusting for chronicity effects, we identified specific cell states in the myeloid compartment that associated with activity. Highly tissue-specific macrophage populations expressing known Scar Associated Macrophage (SAM) gene signatures were particularly expanded, including M5, GPNMBhigh LYVE1low (M11), C1Qlow SPP1high, SPP1highFABP5high (M7), and MERTKhighFABP5high (M9). We hypothesize that M5 may be a profibrotic pathogenic population that could be targeted for therapeutic benefit. We further demonstrate that blood only partially reflects renal tissue. Our results define key renal cell states in the tissue and immune compartments and argue that profound changes in both compartments are associated with increasing chronicity, while specific changes in the myeloid compartment track with activity. In aggregate, these observations strongly support that therapeutic targeting of myeloid populations, in addition to B cells, may offer an unproved strategy to prevent progression of renal inflammation and ongoing kidney damage in LN.