Single-cell transcriptomics identify cell-type specific transcriptomic signatures in response to CdCl2 in human airway bronchial epithelial cells cultured at the air-liquid interface

Abstract

Introduction

The aim of this project is to determine whether single-cell transcriptomics (sc-RNAseq) can provide information susceptible to enrich classical bulk transcriptomics experiments in toxicology. In the present work, we studied the effects of a non-genotoxic carcinogen (CdCl2) on human airway pseudostratified epithelium.

Methods

The identification of cell-type-specific gene expression signatures was performed through scRNA-seq analyses using PARSE technology on human airway bronchial epithelial cells (HAEC) cultured in 3D at an air-liquid interface (ALI) and exposed to 10 μM CdCl2, an heavy metal pollutant present in cigarette smoke. These cell-type specific responses signatures were validated through RNA-Fluorescence In Situ Hybridizations.

Results

Single-cell gene profiling indicated that different regulons are modulated in a cell-type-specific manner in response to CdCl2 exposure. For instance, we showed that treatment with CdCl2 stimulated the expression of MT1G, MT1 M and to a lesser extent SLC30A1 and HMOX1 preferentially in multiciliated and mucus-secreting cells, while AKT3 and IL1RN are rather induced in basal cells. Among the cadmium-regulated genes, we found that the long non-coding RNA LUCAT1 is induced in specific differentiated cell types. We previously showed1 that LUCAT1 is i) upregulated by hypoxia in lung adenocarcinomas, ii) correlated with poor prognosis in patients and iii) involved in the regulation of oxidative stress in cancer cells. However, its cellular sourcing and physiological role in the differentiated mucociliary airway epithelium have not yet been addressed. We found that LUCAT1 is slightly expressed in the untreated epithelium, but in response to hypoxia or CdCl2, it is markedly induced in suprabasal and deuterosomal cells and in differentiated secretory and multiciliated cells but not in basal cells. Using a CRISPR-Cas9-RNP approach2 to knockdown LUCAT1 in basal bronchial stem cells, we will decipher the cell state-specific functions of this gene i) in the regeneration steps of a fully differentiated mucociliary human airway epithelium and, ii) in hypoxia- and CdCl2-responses.

Conclusion

ScRNA-seq reveals distinct transcriptional responses to CdCl2 among airway epithelial cell subtypes. This method could be applied to other molecules to classify gene expression variations and develop predictive tests for non-genotoxic carcinogens compounds. This study also precise the physiological role of LUCAT1 in the homeostasis, regeneration and stress-response of HAEC.