#1004 Single-nucleus RNA sequencing identifies upregulated antigen presentation, apoptosis, mTORC1 signaling and EMT in primary FSGS podocytes

Abstract

Abstract Background and Aims Primary focal segmental glomerulosclerosis (FSGS) is associated with irreversible podocyte damage, but the underlying cause and pathophysiology are incompletely understood. Furthermore, diagnostic markers that allow reliable differentiation between primary and secondary FSGS subtypes are lacking. We applied single-nucleus RNA-sequencing (snRNA-seq) on podocytes of patients with primary or secondary (maladaptive) FSGS, to identify subtype-specific differentially expressed genes and pathways. Method We used droplet-based snRNA-seq (10X Genomics Chromium) on cryopreserved kidney biopsy cores from human patients with newly diagnosed primary FSGS (n = 9, all nephrotic), maladaptive FSGS (n = 9, not nephrotic), proteinuric controls (PLA2R-positive membranous nephropathy (PLA2R+ MN), n = 3), and healthy controls (pre-perfusion biopsies, n = 4). Differential gene expression (DGE) analysis was done with a pseudobulk method provided by the EdgeR R-package to reduce bias and false discoveries. The fGSEA R-package was used for subsequent gene set enrichment analysis. Results We identified a total of 120 751 high-quality nuclei, of which 2 471 were podocytes (Fig. 1A). DGE analysis identified the top 20 differentially expressed genes (DEGs) per diagnosis (Fig. 1B). Remarkably, primary FSGS podocytes showed more transcriptional similarity to PLA2R+ MN podocytes, compared to maladaptive FSGS podocytes. This underscores the heterogeneity of patients with FSGS and shows that heavy proteinuria, regardless of the underlying diagnosis, may be associated with a distinct transcriptional profile in podocytes. Upregulated genes in primary FSGS podocytes included TIMP1, BTG2, ZNF44, EHD4, ZNF250, ECM1, SEMA4G, PHLDB1, and HDAC5. Upregulation of the matrix metalloproteinase inhibitor TIMP1 may possibly contribute to the development of glomerulosclerosis. Upregulation of BTG2 has previously been found to be pathogenic in FSGS by promoting epithelial-mesenchymal transition (EMT)-induced podocyte injury [1]. Next, gene set enrichment analysis on the identified DEGs in primary FSGS podocytes showed upregulation of pathways involved in antigen presentation (driven by upregulation of HLA-A, HLA-B, HLA-C, HLA-E and B2M), mTORC1 signaling, apoptosis and EMT (Fig. 1C). In maladaptive FSGS, upregulated pathways were mainly involved in extracellular matrix interaction. This may correlate with the significantly higher degree of chronic damage on kidney biopsy in the maladaptive FSGS patients (median Mayo Clinic Chronicity Score of 6.0, IQR [4.0-7.0]) vs. primary FSGS (2.0, [1.0-4.0]) and PLA2R+ MN (1, IQR [1.0-1.5]) (Kruskal-Wallis test, P = 0.007). Conclusion Our results corroborate previous studies that implicate apoptosis, mTORC1 signaling, EMT and auto-immunity in primary FSGS. Our future work will focus on cell-cell interactions within the glomerular niche and validation studies of candidate podocyte-specific genes.