Single-cell analysis reveals M. tuberculosis ESX-1-mediated accumulation of anti-inflammatory macrophages in infected mouse lungs

Zheng, Weihao1 ; Borja, Michael2; Dorman, Leah2 ; Liu, Jonathan2; Zhou, Andy2; Seng, Amanda2; Arjyal, Ritwicq2; Sunshine, Sara1; Nalyvayko, Alina1; Pisco, Angela2; Rosenberg, Oren1; Neff, Norma2; Zha, B. Shoshana1

Published Dec 21, 2024 on Dryad. https://doi.org/10.5061/dryad.0p2ngf28s

Data files

Dec 21, 2024 version files 292.10 GB

202305231432_TB-CTRL1-MB_VMSC02401.tgz
31.72 GB
202305251147_TB-CTRL2-MB_VMSC02401.tgz
56.57 GB
202305261458_TB-TEST3-MB_VMSC02401.tgz
34.06 GB
202309251210_20230925-RitwicqA-TB_VMSC02401.tgz
50.61 GB
202310091253_AmandaS-TB-02_VMSC02401.tgz
60.08 GB
202310111428_RitwicqA-TB-4b_VMSC02401.tgz
56.69 GB
Annotated_h5ad_individual.tgz
776.66 MB
Annotated_h5ad.tgz
1.37 GB
avg_signal_df.csv

8.21 MB
cell_annotations.tgz
1.29 MB
postprocessed_counts.tgz
214.73 MB
README.md
6.26 KB
Stats.tgz
64.70 KB

Abstract

Mycobacterium tuberculosis (MTB) ESX-1, a type VII secretion system, is a key virulence determinant contributing to MTB’s survival within lung mononuclear phagocytes (MNPs), but its effect on MNP recruitment and differentiation remains unknown. Here, using multiple single-cell RNA sequencing techniques, we studied the role of ESX-1 in MNP heterogeneity and response in mice and murine bone marrow-derived macrophages (BMDM). We found that ESX-1 is required for MTB to recruit diverse MNP subsets with high MTB burden. Further, MTB induces an anti-inflammatory transcriptional signature in MNPs and BMDM in an ESX-1-dependent manner. Spatial transcriptomics revealed an upregulation of anti-inflammatory signals within MTB lesions, where monocyte-derived macrophages concentrate near MTB-infected cells. Together, our findings suggest that MTB ESX-1 facilitates the recruitment and differentiation of anti-inflammatory MNPs, which MTB can infect and manipulate for survival. Importantly, we provide a comprehensive transcriptomic dataset across various models and methods, which could contribute to the broader understanding of recruited cell heterogeneity during MTB lung infection.

README: Single-cell analysis reveals M. tuberculosis ESX-1-mediated accumulation of anti-inflammatory macrophages in infected mouse lungs

https://doi.org/10.5061/dryad.0p2ngf28s

This dataset contains spatial transcriptomics collected using the Vizgen MERFISH and preprocessed using their standard data processing pipeline. This dataset contains all the raw data used for spatial analysis in this study.

Description of the data and file structure

The data folder contains nested subfolders including a tar archive of the Merscope output for each section of the image data pertaining to each experiment, internally organized by experiment, tissue section, and Z-slice.

The code necessary to process this data as shown in the paper is presented in this archive as TBSpatial_Analysis_code.ipynb (python notebook) and an html file (TBSpatial_Analysis_code.html), and can also be found on Github at https://github.com/bspeco/ESX1_macrophageheterogeneity_scRNA/blob/main/TBSpatial_Analysis_code.ipynb. After downloading the entire data folder, unzip the .tgz files with the command tar -xzvf filename.tgz. You should end up with a folder containing the following subfolders:

Stats

all the statistics calculated in this script, including genewise permutation statistics comparing monocyte/macrophage TB infected cells to their neighbors (AllMonoMacInfectedVsNeighbors.csv); comparing cells within 100 micrometers of a TB lesion to those farther away in bulk (statsAllTB_regionAll.csv) and divided by replicate (statsAllTB_individualReplicates-excludingTB.csv), which were compiled from the statistics shown in each included region-specific csv file; and comparing cells within 100 micrometers of a comparable lesion-like location within a control uninfected sample to those further away (statsAllControl_regionAll.csv).

Annotated_h5ad_individual

Annotated H5AD (anndata) files for each spatial section. These files can be opened in python using the Anndata or Scanpy software packages. Each file is named according the experiment (listed above) and region within that experiment.

Annotated_h5ad

the mononuclear phagocyte subset of single cell (Smartseq2) sequencing data published in this paper for comparison (scRNAseq_MNP_SS2only.h5ad) as well as an annotated copy of all the vizgen data used in the form of an anndata (h5ad) file (Total_annotatedvizgenTB20231120.h5ad). A raw merged copy of all the files in Annotated_h5ad_individual is also included (Total_concatenated_vizgen_h5ad.h5ad). All h5ad files can be opened using anndata or scanpy in python.

cell_annotations

csv files with cell annotations defining whether a cell is positive for TB or not, as defined by the script, to load into the Merfish Vizualizer along with the .vzg files included in DOI:10.5061/dryad.08kprr5bx. Each file is named to correspond with a particular mouse lung section as detailed below.

avg_signal_df.csv

csv file with all the TB antibody staining values per cell, provided so that you can skip running the most time-intensive code chunk generating this file. This file was used to determine the TB status of each cell.

postprocessed_counts

counts generated by Merscope post-processing. Each folder corresponds to one of the mouse lung samples listed below, and the h5ad file inside is the result of the automated Vizgen processing pipeline containing a summary of counts per gene for each identified cell.

Vizgen .tif image files and accompanying metadata are found in the following named folders:

202305231432_TB-CTRL1-MB_VMSC02401: Control uninfected mouse lung sample processed on 05/23/2023 by Michael Borja.
202305251147_TB-CTRL2-MB_VMSC02401: Control uninfected mouse lung sample processed on 05/25/2023 by Michael Borja.
202305261458_TB-TEST3-MB_VMSC02401: TB infected mouse lung sample processed on 05/26/2023 by Michael Borja.
202309251210_20230925-RitwicqA-TB_VMSC02401: TB infected mouse lung sample processed on 09/25/2023 by Ritwicq Arjyal.
202310091253_AmandaS-TB-02_VMSC02401: TB infected mouse lung sample processed on 10/09/2023 by Amanda Seng.
202310111428_RitwicqA-TB-4b_VMSC02401: TB infected mouse lung sample processed on 10/11/2023 by Ritwicq Arjyal.

Each folder contains the following:dapi.png, polyt.png: Image files showing DAPI signal and Poly-T signal for the prepared sample, including multiple regions. experiment.json: JSON file including details of image capture for all channels. region_0, region_1, region_2: Folders containing TIF images and information for each captured section. Only the regions used in this analysis were included. Regions were excluded if they were damaged or showed no evidence of MTB lesions. Each region folder contains the cell boundary coordinates in a csv and parquet format (cellboundaries.parquet and cell_metadata.csv), counts per gene per cell (cell_by_gene.csv), coordinates of all detected transcripts (detected_transcripts.csv), and a snapshot of the cell boundaries, transcripts detected, and automated qc graphs (summary.png). Each nested images folder contains a json (manifest.json) with image information, a csv file detailing how to convert pixel measurements into distance (micron_to_mosaic_pixel_transform.csv), and one compressed tif file per z-section with the TB antibody staining image ("Cellbound3"; mosaic_Cellbound3_z[0,1,2,3,4,5,6]_lzw.tif)

Sharing/Access information

The transcriptomic data associated with this study can be accessed on GEO, Series GSE263892, at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE263892. The vizgen .vzg files are available on Dryad at DOI: 10.5061/dryad.08kprr5bx.

Code/Software

The code associated with the processing of this dataset for the linked manuscript is available within this dataset and on github at https://github.com/bspeco/ESX1_macrophageheterogeneity_scRNA/releases/tag/v1.0.0

Methods

Spatial Transcriptomics

Lungs were perfused with 10 mL of PBS/2 mM EDTA via right ventricle, followed by inflation through the trachea with 1mL of optimal cutting temperature (OCT) mixed with PBS at 1:1. C57BL/6 mice aerosol infected with H37Rv-zsGreen at ~75 CFU/mouse for 28 days were sacrificed. Lungs were embedded in OCT, flash frozen, and stored at -80°C for later cryosectioning. To confirm quality, RNA integrity number (RIN)126 was obtained by extracting RNA from each block (QIAGEN RNeasy Mini Kit, Cat. No. 74104) and analyzing quantity on an Agilent Tapestation system. OCT-embedded tissue blocks were later cryosection at -20°C to a thickness of 10 μm and mounted onto MERSCOPE Slides (Vizgen, PN 20400001). The mounted tissues were immediately fixed with 4% PFA at 37°C for 30 minutes, washed with 3x PBS and stored in 70% ethanol at 4°C for no more than 1 mo before proceeding following the manufacture’s protocol (MERSCOPE) as previously described83 with the following exceptions.

First, cell boundaries of mounted infected tissues including MTB bacterium were stained with the Vizgen Cell boundary Staining Kit (Cat. no. 10400009) in conjunction with a polyclonal MTB antibody (Abcam, ab905) raised in rabbit diluted to 1:100. After washing, the Secondary Staining Mix (Vizgen, PN 20300011) was diluted with a blocking solution in two step dilution (1st - 1:100, 2nd - 1:33) to ensure correct ratio of primary to secondary staining since a custom primary MTB antibody was used.

Imaging: The MERFISH gene panel consisted of 122 genes, designed by selecting top differentially expressed genes in each cell type from the myeloid scRNA sequence atlas, canonical markers, and genes of interest. The probes were hybridized, and samples washed prior to gel embedding without deviation from manufacter protocol. Samples were then treated with a clearing solution and imaged 1 day later per MERSCOPE User Guide 91600001. Only one gel-embedded and cleared sample was prepared per imaging run. The gel-embedded and cleared sample was washed prior to adding the DAPI / Poly T Staining Reagent, and again prior to assembling the onto the MERSCOPE Flow Chamber (Vizgen, PN 10300003). Once fully thawed, the MERSCOPE 140 Gene Imaging Cartridge was activated and a layer of mineral oil was carefully added on top of the imaging buffer. The assembled flow chamber with sample and imaging cartridge was placed in the MERSCOPE, image mosaic was acquired from the regions of interest, and imaging of the sample was performed with multiple rounds of hybridization of fluorescent probes which resulted in a raw images stack that contained about 1 TB of data.