Describing the Early Innate Immune Response of Mycobacterium Tuberculosis at the Single Cell Level

RATIONALE: Mycobacterium tuberculosis (MTB) is responsible for more deaths than any other infectious disease in the world. Despite decades of investigation, little is known about the early interactions of MTB and the innate immune system. We have previously shown mononuclear phagocytes (MPs) recruited during MTB infection are heterogeneous and dynamic through time. However, these results have been limited by resolution to describe unique cellular subtypes and in vivo mechanistic investigations. Single-cell RNA sequencing (scRNA-seq) is an increasingly utilized method that provides detailed descriptions of cell types and cellular alterations in disease models. We therefore aimed to elucidate unique transitional cells and gene signatures of infected cells using this granular approach. METHODS: C57BL/6 mice were infected by aerosolization with MTB expressing mCherry. After 28 days, allowing for influx of T cells, mice were euthanized, lungs dissociated, stained with a combination of antibodies, and index sorted into 384-well plates containing gentle lysis buffer. Following Smart-Seq2 platform (Picelli et al 2014), RNA was reverse transcribed, amplified, fragmented, and after library preparation, sequenced on NovaSeq6000 (Illumina). Sequences were de-multiplexed using bcl2FASTQ and aligned to the mouse (mm10plus) genome using STAR. Standard filtering, variable gene selection, dimensionality reduction, clustering and differential gene expression was performed using Scanpy (Python). Gene enrichment pathway analysis was performed using Ingenuity (Qiagen). RESULTS: The major subset of cells infected with MTB 28 days after aerosol infection in the mouse are recruited monocytes, with a smaller subset of resident macrophages, dendritic cells, and neutrophils, correlating well with our previous findings. The differential gene expression of infected and uninfected bystander cells continues to be examined for complete portrayal of transitional cells and signatures, with key findings including upregulation in the infected cell of the Nrf2 pathway. CONCLUSIONS: We have successfully used scRNA-seq for the first time to build an atlas of innate immune response in the lung of MTBinfected mice. We observe heterogeneity in the monocyte and macrophage populations and find hundreds of differentially expressed genes between infected and bystander cells of each subset. Current verification includes protein differential expression and immunohistochemistry. We now have a working platform to further study MTB and host perturbations in order to elucidate MTB-host interactions at a granular level and obtain essential information needed for the development of host directed therapies.