- HTML 99.1%
- Jupyter Notebook 0.7%
- R 0.2%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| code | ||
| data | ||
| figures | ||
| supplementary-data | ||
| supplementary-figures | ||
| .gitignore | ||
| LICENSE | ||
| README.md | ||
Host oxidative stress primes mycobacteria for rapid antibiotic resistance evolution
Abstract
The rapid emergence of multidrug-resistant Mycobacterium tuberculosis (Mtb) threatens global TB control, yet the mechanisms enabling rapid evolution of resistance in Mtb remain poorly understood. Here, we show that pre-existing mutations in oxidative stress response genes create permissive genomic backgrounds that accelerate high-level isoniazid resistance (INHR), challenging the paradigm that resistance mutations must precede compensatory adaptation. Using Mycobacterium smegmatis mc2155 (Msm) as a model, we demonstrate that brief exposure to sublethal isoniazid (INH) enriches for “low-level resistance and tolerance” (LLRT) mutants in a single step. LLRT mutants, particularly those with ohrR loss-of-function mutations, acquire high-level resistance (>500× IC50) at ~6-fold higher rates than wildtype, primarily through otherwise deleterious mycothiol biosynthesis mutations that become tolerable in an oxidative stress-buffered background.
Crucially, sublethal oxidative stress alone, mimicking host immune pressure, nearly tripled the rate of INHR evolution. Analysis of 1,578 clinical Mtb isolates revealed significant enrichment of oxidative stress-related loci among those associated with INHR. Reanalysis of genome-wide CRISPRi data further linked oxidative stress response pathways to survival under multiple antibiotics. Together, these findings suggest that host-imposed oxidative stress and suboptimal drug exposure may prime Mtb populations for rapid resistance evolution, highlighting oxidative stress defenses as potential targets to limit resistance emergence.
Repository Overview
This repository contains all code and data required to reproduce the analysis and figures in the associated article, published in Nature Communications: Host oxidative stress primes mycobacteria for rapid antibiotic resistance evolution.
Directory structure
Mycobacterium_PreR/
├── code/
├── data/
├── figures/
├── supplementary-figures/
├── supplementary-tables/
├── LICENSE
└── README.md
How to reproduce the entire analysis performed in this manuscript
-
Clone the repository:
git clone https://github.com/evanpepper/Mycobacterium_PreR.git cd Mycobacterium_PreR/code -
Install dependencies for each R Markdown file, found at the beginning of each script.
-
The Fig. 6 notebook will require downloading two datasets:
Dataset 1: Bayes Null Distributions from Zenodo
Dataset 2: Li et al 2022, Source Data Fig. 1
Unzip Dataset 1:
cd Mycobacterium_PreR/ mkdir Mycobacterium_PreR/data/Fig6/LDF/bayes-null-distributions/INH-global-download tar -xvf ~/Downloads/INH-global.tar.gz -C Mycobacterium_PreR/data/Fig6/LDF/bayes-null-distributions/INH-global-downloadCopy Dataset 2 into the following directory:
cd Mycobacterium_PreR/ mkdir Mycobacterium_PreR/data/published-data mv ~/Downloads/41564_2022_1130_MOESM4_ESM.xlsx ~/Downloads/li-et-al-2022-table-SD-F1.xlsx cp ~/Downloads/li-et-al-2022-table-SD-F1.xlsx /data/published-data/li-et-al-2022-table-SD-F1.xlsx -
Run the R Markdown file, either cell by cell or all at once. Reproduced figures will be deposited into
figures/and generated results will appear indata/
Dependencies
The analyses in this repository were performed using the following software and R packages:
- R (version 4.1.1)
- scales (version 1.4.0)
- ggpubr (version 0.6.1)
- rstatix (version 0.7.2)
- naniar (version 1.1.0)
- factoextra (version 1.0.7)
- knitr (version 1.50)
- drc (version 3.0-1)
- lubridate (version 1.9.4)
- forcats (version 1.0.0)
- stringr (version 1.5.1)
- dplyr (version 1.1.4)
- purrr (version 1.1.0)
- readr (version 2.1.5)
- tidyr (version 1.3.1)
- tibble (version 3.3.0)
- tidyverse (version 2.0.0)
- growthcurver (version 0.3.1)
- ggridges (version 0.5.7)
- ggplot2 (version 3.5.2)
- ggrepel (version 0.9.6)
- ggupset (version 0.4.1)
- see (version 0.11.0)
- viridis (version 0.6.5)
- seqinr (version 4.2-36)
- Biostrings (version 2.72.1)
Citation
Link to Nature Communications article
Contact
Evan Pepper -- epepper@systemsbiology.org or epepper@uw.edu