Background The impact of renal injury in lupus nephritis is widespread with consequences to resident cells in other tissue beds, even non-lesional non-sun exposed skin. Faithful reflection of a relevant renal tissue pathway in a more readily accessible compartment would allow for less invasive diagnostic alternatives. Single-cell transcriptional states as performed in this study may provide a framework for understanding how in vivo biological function emerges from complex cell ensembles, thus allowing for a clearer understanding of potential mutual pathways.
Methods Patients with proteinuria and known ISN/RPS Class and controls were recruited to discovery 1 and 2 cohorts. Single cell RNAseq was performed on cell suspensions prepared from ~2 mm punch biopsies of non-lesional non sun-exposed skin from the buttocks. The libraries were prepared on the Fluidigm C1 platform (discovery 1) and 10X Genomics platform (discovery 2) along with Illumina HiSeq 2500 sequencing.
Results Sorting based on COL1A1, COL1A2, COL3A1, MFAP5 and MFAP4 expression yielded 12 fibroblasts from 3 patients. The 1 Class II subject yielded 5 single-cell transcriptomes. The other 2 subjects (1 Class IV,V; 1 Class III,V) yielded 7 single-cell transcriptomes. 22 transcriptomes were derived from 3 controls. The aggregate data were used to determine the top upregulated genes in cases versus controls, most of which belonged to the interferon-stimulated gene category and the extracellular matrix category (DAVID databases). Fewer cells were obtained using Fluidigm C1 (36 single-cell) than 10X Genomics (7280 single-cell). For the latter, the major biopsy classes were represented (Class III, III/IV, III/V, V and no LN). We applied graph-based clustering and identified 12 major clusters of cells from the patient skin as visualized by t-distributed stochastic neighbor embedding (t-SNE; figure 1). Differential gene expression analysis guided by established lineage markers revealed three keratinocyte clusters (KC1-KC3), two fibroblast clusters (FB1, FB2), smooth muscle cells (SMC), two endothelial cell clusters (VEC, LEC), melanocytes (MEL), sweat gland cells (SG), macrophages/dendritic cells (MAC-DC) and T cells (TC). Ranked by abundance, patient skin exhibited KC>FB>EC>MAC-DC>SMC>TC>SG> MEL.
Conclusions Single-cell RNAseq is both feasible and informative in cell-specific transcriptome analysis of fresh non-lesional non-sun exposed LN skin biopsies. 10X genomics significantly increases cell numbers and facilitates identification of major cell clusters compared to Fluidigm C1. The expression of fibroblasts and genes reflective of fibrotic and interferon-related pathways support the application of this approach to study readily accessible tissue for biomarkers related to disease progression.
Funding Source(s): Judith and Stewart Colton Center for Autoimmunity at NYU Langone Health
Differential gene expression analysis guided by established lineage markers revealed three keratinocyte clusters (KC1-KC3), two fibroblast clusters (FB1, FB2), smooth muscle cells (SMC), two endothelial cell clusters (VEC, LEC), melanocytes (MEL), sweat gland cells (SG), macrophages/dendritic cells (MAC-DC) and T cells (TC). Ranked by abundance, patient skin exhibited KC>FB>EC>MAC-DC>SMC>TC>SG> MEL.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.