Posted on 11.12.26

Hundreds of new resistance mutations uncovered with key antibiotics and metagenomic sequencing

Tags

Hundreds of new resistance mutations uncovered with key antibiotics and metagenomic sequencing

Antimicrobial resistance (AMR) is a global health crisis.  A huge obstacle is to identify mutations that lead to resistance in a given genomic background and environment.

Predicting susceptibility from genomic data is becoming integral to antimicrobial stewardship.  However, the accuracy of the predictions is limited by our knowledge of which mutations contribute to AMR and via what mechanisms.

Mapping mutations to AMR phenotypes is typically achieved through sequencing.  This tool is often biased towards mutations that confer high-level resistance and struggle to discern how AMR evolution varies in different contexts.

Researchers at the University of Manchester (Manchester, UK) developed a high-throughput technique – called Quantitative Mutational Scan sequencing (QMS-seq) – that allows quantitative comparison of which genes are under antibiotic selection and finds how genetic background can influence resistance evolution.   It uses metagenomic sequencing to characterize the mutational landscape of resistance.

In this technique, a genetically homogeneous population is left to accumulate random mutants over 24 hours under minimal selection (growing in rich media without antibiotics) to generate a heterogeneous population where most of the variants have a single mutation .

Four E. coli strains were used in this study and were exposed to three different antibiotics: Ciprofloxacin, Cycloserine, and Nitrofurantoin (from TOKU-E) using 3 replicates for each of the 12 conditions, for a total of 36 samples. 

The population is then spread across ten selective agar plates, containing the minimum inhibitory concentration (MIC) of an antibiotic, which is the lowest concentration of a given antibiotic that prevented colonies from forming.  The MIC was 0.005 μg/mL for Ciprofloxacin, 100 μg/mL for Cycloserine and 10 μg/mL for Nitrofurantoin.  After resistant colonies have grown, they are then mixed and sequenced with sufficient depth to allow detection of low-frequency resistance mutations. 

A key strength of QMS-seq is the ability to simultaneously identify hundreds of mutations throughout the genome, which comes from pooling mutant colonies and sequencing collectively.  A custom bioinformatic pipeline was developed to capture a landscape of mutations under selection in a given condition, enabling comparison across different environments and genetic backgrounds.

These antibiotics have distinct mechanisms of action: Ciprofloxacin targets DNA topoisomerase II (gyrase), which causes replication fork blockage and DNA damage, Cycloserine inhibits growth by interfering with cell wall biosynthesis, and Nitrofurantoin is reduced into various metabolites, some of which are highly reactive and non-specifically target ribosomes to prevent protein synthesis.

Authors identified 812 resistance mutations, in 251 genes and 49 regulatory features.  Many of the mutations were in genes and regulatory regions not previously associated with resistance, including 11 of the 30 most targeted genes.  They considered these resistance mutations because they emerged under strong selection in colonies that were able to grow at 1x MIC.  This indicates how much of the mutational landscape for resistance remains uncharacterized and emphasizes QMS-seq’s utility in exploring it. 

To verify that the mutations were also relevant in non-laboratory settings, authors searched for matches in 21,711 genomes of clinical and environmental E. coli isolates from the BV-BRC database.  There were 50 genes (of 251 they identified) whose mutations matched the sequence of one or more isolates, including several well-characterized genes associated with clinical resistance.

Their experimental design could differentiate between multi-drug resistance (MDR) mutations that conferred resistance to all three antibiotics and mutations that conferred antibiotic-specific resistance (ASR).  They discovered the most common mutations were active site modifications and regulatory changes that provide MDR, in contrast to ASR which typically arises from knockout mutations.

Among the top ASR genes were well-known: For Ciprofloxacin: gyrA and icd;  for Cyclosporin: ispA/B and ubiA/E/G/H, and Nitrofurantin:ompC .  Authors found ompC was unique as it is typically considered a target for MDR, rather than being specific to Nitrofurantoin.

This powerful technique can provide insights into microbe biology and investigate how resistance to one antibiotic influences the evolution of resistance to another one.  Its broad applicability could help transition to pre-emptive strategies for AMR by looking at diverse strains’ evolvability under a variety of conditions.

Reference

Jago MJ et al (2025)  High-throughput method characterizes hundreds of previously unknown antibiotic resistance mutations. Nat Commun 16 (780)  Link