Cell-type-specific gene expression will be changed in the process of cell fate transformation and response to external signals, and dynamic changes in RNA levels are regulated by the interaction of RNA transcription, processing and degradation. In complex tissues and systems with multiple cell types, it is important to accurately determine the dynamic changes of these mRNA at the single-cell level for understanding the regulation of gene expression.
Recent advances in single-cell RNA sequencing (scRNA-seq) are resulting in a more complete understanding of heterogeneity in cell types and states. However, conventional single-cell transcriptome sequencing technology can reveal the stable transcriptome of different cell types, but it cannot accurately distinguish the newly generated and existing mRNA at a specific time. Commonly used approaches for distinguishing new from old RNAs within the same population of transcripts depend on RNA metabolic labeling that utilizes exogenous nucleoside analogs 4-thiouridine (4sU) and biochemical enrichment of labeled RNAs. But they require ample starting material and pose challenges to enrichment normalization. Recently, several methods were developed to chemically convert 4sU into cytidine analogs, generating uracil-to-cytosine substitutions that label newly transcribed RNAs after reverse transcription. These chemical approaches allow direct measurement of temporal information about cellular RNAs without biochemical enrichment. However, they are costly and time-consuming, lack of unique molecular identifiers (UMIs), and thereby preventing accurate quantification of new transcript levels.
To overcome these constraints, Wu Hao Laboratory of the University of Pennsylvania has developed single-cell metabolically labeled new RNA tagging sequencing (scNT-seq), a high-throughput and UMI-based scRNA-seq method, to detect the dynamic changes of single-cell mRNA. This method innovatively integrates metabolic labeling of mRNA, high-throughput single-cell transcriptome analysis technology based on droplet microfluidics and chemically induced recoding of 4sU to a cytosine analog to simultaneously measure new and old transcriptomes from the same cell. In terms of data analysis, the author constructed a statistical model based on unique molecular identifiers (UMIs) to more accurately analyze the proportion of newly generated mRNA at the single-cell level. Relevant research was published online in Nature Methods, titled "Massively parallel and time-resolved RNA sequencing in single cells with scNT-seq".
The specific technological process of scNT-seq is as follows: Firstly, the cells were metabolically labeled with 4sU. Co-encapsulate individual cells with a barcoded oligo (dT) primer-coated bead in a nanoliter-scale droplet and perform one-pot 4sU chemical conversion on pooled barcoded beads. Then reverse transcription, cDNA amplification, tagmentation and indexing PCR were performed. Finally, using a UMI-based statistical model analyze T-to-C substitutions within transcripts and infer the new transcript fraction.
Figure 1. Overview of scNT-seq. (Qiu Q, et al., 2020)
Using this technique, researchers analyzed the single-cell gene regulatory network during the rapid activation of mouse neurons and predicted the trajectory of cell state changes. They further studied the regulation mechanism of mRNA dynamic changes during the state transition of different embryonic stem cells. Integrating scNT-seq with genetic perturbation identifies DNA methylcytosine dioxygenase as an epigenetic barrier into the two-cell embryo (2C)-like cell state.
In general, scNT-seq is a new technology that can accurately quantify new transcripts at the single-cell level. Compared with the existing single-cell metabolic marker analysis methods, this new technology has the advantages of high accuracy, high throughput and low cost. Besides, dual-labeling of cells with 4sU and 6-thioguanine followed by scNT-seq can enable two independent transcriptomic recordings in single cells, allowing time-series experimental designs to untangle complex RNA regulatory mechanisms and to predict past and future cell states over an extended period. Thus, high-throughput time-resolved single-cell transcriptomics provides a widely applicable strategy to investigate dynamic biological systems.