Genome-wide analysis captures the determinants of the antibiotic cross-resistance interaction network

Viktória Lázár, István Nagy, Réka Spohn, Bálint Csörgő, Ádám Györkei, Ákos Nyerges, Balázs Horváth, Andrea Vörös, Róbert Busa‐Fekete, Mónika Hrtyan, Balázs Bogos, Orsolya Méhi, Gergely Fekete, Balázs Szappanos, Balázs Kégl, Balázs Papp, Csaba PálNature Communications 2014 · DOI

Abstract

Understanding how evolution of antimicrobial resistance increases resistance to other drugs is a challenge of profound importance. By combining experimental evolution and genome sequencing of 63 laboratory-evolved lines, we charted a map of cross-resistance interactions between antibiotics in Escherichia coli, and explored the driving evolutionary principles. Here, we show that (1) convergent molecular evolution is prevalent across antibiotic treatments, (2) resistance-conferring mutations simultaneously enhance sensitivity to many other drugs, and (3) 27% of the accumulated mutations generate proteins with compromised activities, suggesting that antibiotic adaptation can partly be achieved without gain of novel function. By using knowledge on antibiotic properties, we examined the determinants of cross-resistance and identified chemogenomic profile similarity between antibiotics as the strongest predictor. In contrast, cross-resistance between two antibiotics is independent of whether they show synergistic effects in combination. These results have important implications on the development of novel antimicrobial strategies.

Introduction

As bacteria evolve resistance to one antibiotic, they often simultaneously change their sensitivity to other drugs—a phenomenon known as cross-resistance (increased resistance to a second drug) or collateral sensitivity (increased vulnerability). While these interactions are clinically critical, a systematic map of these evolutionary trajectories and the rules governing them has been lacking. Specifically, it was unknown whether the physiological interactions of drugs (e.g., synergy) could predict how resistance would evolve. This study seeks to move from phenomenological observations to a predictive, genome-wide understanding of the antibiotic interaction network.

Methods

The researchers employed a multi-stage experimental pipeline:

  1. Experimental Evolution: 120 independent lines of E. coli K12 were evolved under increasing concentrations of 12 antibiotics representing various modes of action (e.g., cell wall inhibitors, protein synthesis inhibitors, gyrase inhibitors).
  2. Phenotyping: Minimum Inhibitory Concentrations (MIC) were measured for all lines against all 12 antibiotics to chart the cross-resistance map.
  3. Genomics: Whole-genome sequencing (WGS) was performed on 63 lines to identify specific mutations (SNPs, indels, and structural variants).
  4. Chemogenomic Integration: Evolutionary data were compared with existing chemogenomic datasets—screens identifying genes that alter baseline sensitivity—to define the "intrinsic resistome" of the bacteria.

Results: The Cross-Resistance Landscape

The study revealed a dense network of cross-resistance, with 52% of all investigated antibiotic pairs showing interactions in at least one direction. Evolution was found to be highly convergent; 66% of parallel mutated genes occurred in lines adapted to different antibiotics, highlighting a shared set of multidrug resistance functional modules like efflux pumps and membrane permeability regulators.

Figure 1: Map of antibiotic cross-resistance interactions in E. coli. Arrows indicate that resistance to one drug confers resistance to another.

Notably, aminoglycosides (Kanamycin and Tobramycin) acted as outliers, showing very few cross-resistance interactions with other drug classes. Furthermore, roughly 27% of the accumulated mutations were putative loss-of-function (LoF) mutations, suggesting that antibiotic adaptation is frequently achieved by compromising existing protein activities rather than gaining new ones.

Determinants of Cross-Resistance

A major finding of this work is the decoupling of physiological synergy and evolutionary cross-resistance. The data demonstrated that whether two drugs work well together in a cocktail (synergy) does not predict if resistance to one will protect against the other.

Figure 2: Cross-resistance frequency is independent of physiological drug interactions (antagonistic, no interaction, or synergistic).

Instead, the strongest predictor of cross-resistance between two drugs is the similarity of their chemogenomic profiles. If two drugs rely on the same set of "intrinsic" genes for their efficacy, resistance to one is highly likely to confer resistance to the other. Using only the set of accumulated mutations, the authors could predict the resistance phenotype of an evolved strain with 76% average accuracy.

Discussion

This research provides a mechanistic framework for understanding how resistance evolves across the antibiotic landscape. While limited by its laboratory setting and the absence of horizontal gene transfer (HGT), the study identifies fundamental principles of bacterial adaptation. The discovery that chemogenomic profiles are superior to physiological interactions for predicting cross-resistance offers a new path for designing antimicrobial strategies that avoid the pitfalls of multi-drug resistance.

Summary updated

Abstract

Understanding how evolution of antimicrobial resistance increases resistance to other drugs is a challenge of profound importance. By combining experimental evolution and genome sequencing of 63 laboratory-evolved lines, we charted a map of cross-resistance interactions between antibiotics in Escherichia coli, and explored the driving evolutionary principles. Here, we show that (1) convergent molecular evolution is prevalent across antibiotic treatments, (2) resistance conferring mutations simultaneously enhance sensitivity to many other drugs and (3) 27% of the accumulated mutations generate proteins with compromised activities, suggesting that antibiotic adaptation can partly be achieved without gain of novel function. By using knowledge on antibiotic properties, we examined the determinants of cross-resistance and identified chemogenomic profile similarity between antibiotics as the strongest predictor. In contrast, cross-resistance between two antibiotics is independent of whether they show synergistic effects in combination. These results have important implications on the development of novel antimicrobial strategies.
BibTeX
@inproceedings{lazar2014genome,
  author = {Viktória Lázár and István Nagy and Réka Spohn and Bálint Csörgő and Ádám Györkei and Ákos Nyerges and Balázs Horváth and Andrea Vörös and Róbert Busa‐Fekete and Mónika Hrtyan and Balázs Bogos and Orsolya Méhi and Gergely Fekete and Balázs Szappanos and Balázs Kégl and Balázs Papp and Csaba Pál},
  title = {{Genome-wide analysis captures the determinants of the antibiotic cross-resistance interaction network}},
  booktitle = {Nature Communications},
  year = {2014},
  doi = {10.1038/ncomms5352},
}
Figure