Single-cell RNA sequencing for the study of lupus nephritis

Abstract

Single-cell RNA sequencing (scRNA-seq) has recently undergone rapid advances in the development of this technology, leading to high throughput and accelerating discovery in many biological systems and diseases. The single-cell resolution of the technique allows for the investigation of heterogeneity in cell populations, and the pinpointing of pathological populations contributing to disease. Here we review the development of scRNA-seq technology and the analysis that has evolved with the ever-increasing throughput. Finally, we highlight recent applications of scRNA-seq to understand the molecular pathogenesis of lupus and lupus nephritis.

Keywords: inflammation; lupus nephritis; systemic lupus erythematosus.

Figure 1

Bulk sequencing versus scRNA-seq. In bulk sequencing, a pool of extracted RNA from a population of cells is sequenced, providing expression data representing the average expression of a particular gene across all cells. In contrast, scRNA-seq retains the originating cell-specific transcript information. If many cells of the same cell type were identified, the averaging of sequence reads across all cells yields cell-type-specific expression information similar to bulk RNA-seq profiles. Hypothetical outputs for a gene in bulk sequencing and scRNA-seq are shown. scRNA, single-cell RNA sequencing; tSNE, t-Distributed Stochastic Neighbour Embedding.

Figure 2

ScRNA-seq workflows. The basic workflow of scRNA-seq follows similar steps regardless of the platform. Tissue is disaggregated into single-cell suspensions and loaded onto the scRNA-seq platform of choice (ie, Dropseq or Fluidigm C1). Lysis, reverse transcription, PCR amplification and sequencing are followed by downstream analyses. scRNA, single-cell RNA sequencing.

Figure 3

Droplet-based scRNA-seq integration of cell barcodes and UMIs. Droplet-based scRNA-seq platforms use barcoded beads to retain information both on the cell of origin and mRNA sequence. This is accomplished by use of beads conjugated to oligonucleotides-containing sequences for the PCR priming, a cell barcode unique for each bead and a UMI randomised sequence distinguishing oligonucleotides per bead. The end of the oligo is a polyT tail which binds to polyA tails of mature mRNAs. This allows reverse transcription of the mRNA into a cDNA molecule which now contains the PCR handle, cell barcode and UMI. in this way, during PCR amplification, the information regarding input cell of origin and mRNA of origin is retained. When the library is prepared and sequenced, both ends are sequenced. The first read will sequence the cell barcode and UMI, whereas the second read sequences the cDNA. Together these reads provide the gene encoding the mRNA, the cell it was expressed in, and further, by summing UMIs, the total number of mRNAs for the specific gene in that cell. scRNA, single-cell RNA sequencing; UMI, unique molecular identifiers.

Figure 4

Representative plots of downstream scRNA-seq analysis. (A) An illustrative tSNE plot, where cells form clusters based on similarities and differences in gene expression. Different colours designate different clusters which differentiate between cell types and cell states. (B) An illustrative pseudotime plot, the line indicating a continuum of hypothetical differentiation from least differentiated (red colour) to several branches of distinct differentiated cell types. scRNA, single-cell RNA sequencing; tSNE, t-Distributed Stochastic Neighbour Embedding.