The role of epistasis in amikacin, kanamycin, bedaquiline, and clofazimine resistance in Mycobacterium tuberculosis complex

Roger Vargas, Luca Freschi, Andrea Spitaleri, Sabira Tahseen, Ivan Barilar, Stefan Niemann, Paolo Miotto, Daniella Cirillo, Claudio U. Köser, Maha FarhatbioRxiv 2021 · DOI

TL;DR

This study reveals that genetic mutations conferring antibiotic resistance in Mycobacterium tuberculosis can be "overridden" by other mutations, leading to inaccurate genotypic drug susceptibility testing (DST). By analyzing over 31,000 genomes, the researchers identified the first definitive example of antibiotic "reversion" in TB, where a resistance mutation is counteracted by a loss-of-function mutation, potentially leading to misdiagnosis and inappropriate treatment.

What Problem Does This Paper Address?

Mycobacterium tuberculosis (MTB) is a major global health threat, and antibiotic resistance is a significant challenge in controlling its spread. Due to MTB's slow growth and the need for biosafety level 3 containment, genotypic drug susceptibility testing (DST) is the primary method for determining appropriate treatment regimens. However, a critical question arises: are individual resistance mutations reliable indicators of phenotypic resistance, or can other genetic interactions (epistasis) mask or even reverse their effects? Such epistatic interactions could lead to major diagnostic errors, such as false resistance results, which have been reported for streptomycin but are less understood for more clinically relevant antibiotics. This paper addresses this gap by systematically investigating epistatic interactions that could affect the interpretation of genotypic DST for key TB drugs.

Methods

The study employed whole-genome sequencing (WGS) of a large cohort of Mycobacterium tuberculosis isolates. Key methods included:

  • Large-scale WGS analysis: Analyzing over 31,000 MTBC genomes to identify genetic variations.
  • Targeted gene analysis: Focusing on specific genes known to be involved in resistance to amikacin, kanamycin, bedaquiline, and clofazimine, including both resistance-conferring and regulatory/efflux pump genes.
  • Phylogenetic analysis: Constructing phylogenetic trees to understand the evolutionary relationships between isolates and the co-occurrence of mutations.
  • Ancestral reconstruction: Used to estimate the timing and number of independent mutational events (homoplasy).
  • Molecular clock dating: Approximating the age of mutations and their emergence.
  • Variant calling and annotation: Rigorous bioinformatic pipelines were used to identify and annotate single nucleotide polymorphisms (SNPs) and insertions/deletions (indels).

Datasets

The study utilized a dataset of 31,440 Mycobacterium tuberculosis complex (MTBC) clinical isolates. This dataset was sourced from NCBI. Additional data from 12 clinical eis C-14T mutants with available KAN and AMK Minimum Inhibitory Concentrations (MICs) were also incorporated.

Key Insights and Ideas

  • "Antibiotic Reversion" is Real: The most striking finding is the definitive demonstration of antibiotic reversion in MTBC. This occurs when a mutation that confers resistance is counteracted by a loss-of-function (LoF) mutation in a related gene, effectively restoring susceptibility.
  • The eis Gene as a Case Study: The eis C-14T promoter mutation, a common marker for kanamycin and amikacin resistance, was shown to not confer resistance if accompanied by LoF mutations in the eis coding region. This is the first clear example of such reversion for these drugs.
  • Implications for Genotypic DST: This discovery has significant implications for genotypic DST. Assays that rely solely on the eis C-14T mutation may produce false resistance results in the presence of eis LoF mutations, leading to incorrect treatment decisions.
  • Potential for mmpR/mmpL5/mmpS5 Epistasis: The study also suggests a similar epistatic interaction might occur with mutations in mmpR, mmpS5, and mmpL5, potentially affecting bedaquiline and clofazimine resistance. LoF mutations in mmpS5 and mmpL5 might negate the resistance conferred by mmpR mutations.
  • Evolutionary Convergence: The analysis revealed instances of parallel evolution (homoplasy) for resistance mutations, indicating that similar genetic solutions to antibiotic pressure can arise independently in different lineages.

Key Results

  • First Definitive Reversion: The eis C-14T promoter mutation, when co-occurring with eis LoF mutations, failed to confer resistance to amikacin and kanamycin. This represents the first definitive example of antibiotic reversion in MTBC.
  • Widespread eis C-14T: The eis C-14T promoter mutation was found in 881 isolates without other key aminoglycoside resistance mutations in rrs.
  • Co-occurrence of eis mutations: 21 out of 179 isolates with the eis C-14T mutation also harbored eis LoF mutations. These double mutants were phenotypically susceptible to kanamycin.
  • Potential mmpR/mmpL5/mmpS5 interaction: 91 fixed LoF variants were detected in mmpL5-mmpS5-mmpR, with 35 occurring in at least two isolates, suggesting potential for epistatic interactions affecting bedaquiline and clofazimine resistance.
  • Evolutionary timing: The MRCA for isolates with both mmpR and mmpL5 LoF mutations was estimated to be 66-132 years ago, predating the clinical use of bedaquiline and clofazimine.

Limitations and Future Work

  • BDQ/CFZ MICs: The study lacked BDQ and CFZ MIC data for the mmpR/mmpL5 double-LoF mutants, limiting definitive conclusions about epistasis for these drugs. Future work should include this data.
  • Clinical Relevance of eis Reversion: While the study provides strong evidence for eis reversion, the authors note that even for amikacin, this phenomenon is rare, with only 12% of isolates harboring eis C-14T without rrs mutations also having concomitant eis LoF mutations. This suggests that eis C-14T might still be a useful marker for AMK resistance, but a cautious approach is warranted, especially if epistasis is suspected.
  • Comprehensive Workup: The authors emphasize the need for comprehensive microbiological workup in clinical trials, including analysis of mmpR, mmpS5-mmpL5, and their intergenic regions, alongside standardized MIC testing.
  • Discordance Resolution: They recommend confirming discordances between genotypic and phenotypic DST results through retesting and specialized testing.
  • Further Investigation of mmpR/mmpL5: The potential epistasis in the mmpR/mmpL5 system warrants further investigation, particularly in relation to antibiotic pressure.
  • Synonymous Mutations: The study highlights that synonymous mutations at start codons can have functional consequences, challenging the a priori exclusion of such mutations from WGS analysis. This suggests a need for more nuanced variant interpretation.

Summary updated

Abstract

ABSTRACT Antibiotic resistance among bacterial pathogens poses a major global health threat. M. tuberculosis complex (MTBC) is estimated to have the highest resistance rates of any pathogen globally. Given the slow growth rate and the need for a biosafety level 3 laboratory, the only realistic avenue to scale up drug-susceptibility testing (DST) for this pathogen is to rely on genotypic techniques. This raises the fundamental question of whether a mutation is a reliable surrogate for phenotypic resistance or whether the presence of a second mutation can completely counteract its effect, resulting in major diagnostic errors (i.e. systematic false resistance results). To date, such epistatic interactions have only been reported for streptomycin that is now rarely used. By analyzing more than 31,000 MTBC genomes, we demonstrated that eis C-14T promoter mutation, which is interrogated by several genotypic DST assays endorsed by the World Health Organization, cannot confer resistance to amikacin and kanamycin if it coincides with loss-of-function (LoF) mutations in the coding region of eis . To our knowledge, this represents the first definitive example of antibiotic reversion in MTBC. Moreover, we raise the possibility that mmpR ( Rv0678 ) mutations are not valid markers of resistance to bedaquiline and clofazimine if these coincide with LoF mutation in the efflux pump encoded by mmpS5 ( Rv0677c ) and mmpL5 ( Rv0676c ).
BibTeX
@misc{vargas2021role,
  author = {Roger Vargas and Luca Freschi and Andrea Spitaleri and Sabira Tahseen and Ivan Barilar and Stefan Niemann and Paolo Miotto and Daniella Cirillo and Claudio U. Köser and Maha Farhat},
  title = {{The role of epistasis in amikacin, kanamycin, bedaquiline, and clofazimine resistance in <i>Mycobacterium tuberculosis</i> complex}},
  year = {2021},
  doi = {10.1101/2021.05.07.443178},
  howpublished = {bioRxiv},
}
Figure