Multiplexed detection of proteins, transcriptomes, clonotypes and CRISPR perturbations in single cells

Abstract

Multimodal single-cell assays provide high-resolution snapshots of complex cell populations, but are mostly limited to transcriptome plus an additional modality. Here, we describe expanded CRISPR-compatible cellular indexing of transcriptomes and epitopes by sequencing (ECCITE-seq) for the high-throughput characterization of at least five modalities of information from each single cell. We demonstrate application of ECCITE-seq to multimodal CRISPR screens with robust direct single-guide RNA capture and to clonotype-aware multimodal phenotyping of cancer samples.

Figure 1:

ECCITE-seq allows simultaneous detection of transcriptome, proteins, clonotypes and CRISPR perturbations. a. Schematic overview of the multiple cellular modalities captured by ECCITE-seq. cDNAs derived from mRNA and sgRNA transcripts acquire cell barcodes and unique identifiers through a template switch reaction with the bead-bound oligo, whereas protein tags harbor sequences that allow direct annealing to the bead oligo. Reverse transcription and amplification yields products with distinct sizes that can be separated and amplified independently (right panel). b. Species-mixing experiment. Left: number of transcripts associated with each cell barcode (red, >90% human reads; green, >90% mouse reads; blue, >10% human and mouse (multiplet). Right: sgRNA reads associated with each cell barcode. Points are colored based on species classifications using transcripts. c. Transcriptome-based clustering of single-cell expression profiles of the mixed human and mouse sample, illustrating the 5 modalities of ECCITE-seq: Transcriptome, Cell Hashing [I], Protein [II], T cell antigen receptors (α/β: red, γ/δ: blue) [III] and sgRNAs [IV]. d. Number of cell barcodesbefore (purple) or after (blue) removal of cell doublets assigned zero, one or two unique guides per cell in two independent experiments and cell lines. e. K562 cells were clustered based on normalized and scaled sgRNA counts. Highlighted are counts for sgRNAs (black) targeting the indicated gene and counts for respective mRNA (blue) and protein tags (green) where applicable.

Figure 2:

ECCITE-seq couples clonotype determination with immunophenotyping. a. Transcriptome-based clustering of PBMCs from healthy donor (top) and CTCL patient (bottom) after removing cell doublets. Projected is the mRNA (blue) and protein (green) signal for CD3, CD4 and CD8, as well as the most abundant CD4+ or CD8+ TCR α/β or two most abundant TCR γ/δ clonotypes (red). b. Transcriptome-based clustering of the combined dataset after merging, depth normalization and cell alignment. Highlighted from left to right is sample identity, productive TCR rearrangement and most abundant CD4+ TCRα/β clonotype. c. Differentially expressed genes between CD3+ CD4+ T clusters of control and CTCL PBMCs, as defined by in silico gating based on CD3 and CD4 protein counts coupled with clonotype determination.