Identification of Intercellular Signaling Changes Across Conditions and Their Influence on Intracellular Signaling Response From Multiple Single-Cell Datasets.

Mengqian Hao,Suoqin Jin,Xiufen Zou

doi:10.3389/fgene.2021.751158

Abstract

Identification of intercellular signaling changes across multiple single-cell RNA-sequencing (scRNA-seq) datasets as well as how intercellular communications affect intracellular transcription factors (TFs) to regulate target genes is crucial in understanding how distinct cell states respond to evolution, perturbations, and diseases. Here, we first generalized our previously developed tool CellChat, enabling flexible comparison analysis of cell–cell communication networks across any number of scRNA-seq datasets from interrelated biological conditions. This greatly facilitates the ready detection of signaling changes of cell–cell communication in response to any biological perturbations. We then investigated how intercellular communications affect intracellular signaling response by inferring a multiscale signaling network which bridges the intercellular communications at the population level and the cell state–specific intracellular signaling network at the molecular level. The latter is constructed by integrating receptor-TF interactions collected from public databases and TF-target gene regulations inferred from a network-regularized regression model. By applying our approaches to three scRNA-seq datasets from skin development, spinal cord injury, and COVID-19, we demonstrated the capability of our approaches in identifying the predominant signaling changes across conditions and the critical signaling mechanisms regulating target gene expression. Together, our work will facilitate the identification of both intercellular and intracellular dysregulated signaling mechanisms responsible for biological perturbations in diverse tissues.

Highlights

Cell–cell communication means that one cell sends a message to another cell through a medium to initiate cellular response of the target cell
Compared to the original CellChat that was limited to the comparison analysis of only two datasets, the updated CellChat generalizes many existing functions, which enables systematical comparison analysis of intercellular communications across any number of scRNA-seq datasets
We demonstrated the effectiveness of our proposed approaches by studying the signaling changes across three mouse embryonic developmental stages, four time points after mouse spinal cord injury, and patients with different COVID-19 severities

Summary

Introduction

Cell–cell communication means that one cell sends a message to another cell through a medium to initiate cellular response of the target cell. The communication between cells plays a vital role in the development, physicology, and pathology of muticellular organisms In this process, cells can communicate with and respond to neighboring or distant cells through ligand-receptor interactions by utilizing biochemical molecules, such as cytokines and growth factors. A number of computational methods have been recently developed to infer cell–cell communication by integrating scRNA-seq data with a prior ligand–receptor interaction database, most of these methods only focus on the intercellular communications in one biological condition (Almet, et al, 2021; Armingol, et al, 2021), lacking the capability of identifying signaling changes across conditions. We have recently developed a computational tool CellChat (Jin, et al, 2021) to identify dysregulated interactions by comparing cell–cell communication networks across conditions. With the increasing number of scNRA-seq datasets collected from multiple conditions, time points, and disease states, easy-to-use tools that can seamlessly identify signaling changes across any biological conditions from multiple scRNA-seq datasets are highly needed

Objectives

Methods

Results

Conclusion