Prediction of Antibiotic Resistance Evolution by Growth Measurement of All Proximal Mutants of Beta-Lactamase.

The antibiotic resistance crisis continues to threaten human health. Better predictions of the evolution of antibiotic resistance genes could contribute to the design of more sustainable treatment strategies. However, comprehensive prediction of antibiotic resistance gene evolution via laboratory approaches remains challenging. By combining site-specific integration and high-throughput sequencing, we quantified relative growth under the respective selection of cefotaxime or ceftazidime selection in ∼23,000 Escherichia coli MG1655 strains that each carried a unique, single-copy variant of the extended-spectrum ß-lactamase gene blaCTX-M-14 at the chromosomal att HK022 site. Significant synergistic pleiotropy was observed within four subgenic regions, suggesting key regions for the evolution of resistance to both antibiotics. Moreover, we propose PEARP and PEARR, two deep-learning models with strong clinical correlations, for the prospective and retrospective prediction of blaCTX-M-14 evolution, respectively. Single to quintuple mutations of blaCTX-M-14 predicted to confer resistance by PEARP were significantly enriched among the clinical isolates harboring blaCTX-M-14 variants, and the PEARR scores matched the minimal inhibitory concentrations obtained for the 31 intermediates in all hypothetical trajectories. Altogether, we conclude that the measurement of local fitness landscape enables prediction of the evolutionary trajectories of antibiotic resistance genes, which could be useful for a broad range of clinical applications, from resistance prediction to designing novel treatment strategies.

Assessment of the reversibility of resistance in the absence of antibiotics and its relationship with the resistance gene's fitness cost: a genetic study with mcr-1.

BACKGROUND: The intensive use of antibiotics has resulted in strong natural selection for the evolution of antimicrobial resistance (AMR), but whether, and under what circumstances, the removal of antibiotics would result in a rapid reduction in AMR has been insufficiently explored. We aimed to test the hypothesis that in the simple, yet common, case of AMR conferred by a single gene, removing antibiotics would quickly reduce the prevalence of resistance if the AMR gene imposes a high fitness cost and costless resistance is extremely rare among its proximal mutants. METHODS: In this genetic study, to test our hypothesis, we used the mcr-1 gene in Escherichia coli, which confers resistance to the last-resort antibiotic colistin, as a model. A high-throughput reverse genetics approach was used to evaluate mcr-1 variants for their fitness cost and resistance levels relative to a non-functional construct, by measuring relative growth rates in colistin-free media and at 2 µg/mL and 4 µg/mL colistin. We identified costless resistant mcr-1 mutants, and examined their properties within the context of the sequential organisation of mcr-1's functional domains as well as the evolutionary accessibility of these mutations. Finally, a simple population genetic model incorporating the measured fitness cost was constructed and tested against previously published real-world data of mcr-1 prevalence in colonised inpatients in China since the 2017 colistin ban in fodder additives. FINDINGS: We estimated the relative growth rates of 14 742 mcr-1 E coli variants (including the wild type), 3449 of which were single-nucleotide mutants. E coli showed 73·8% less growth per 24 h when carrying wild-type mcr-1 compared with the non-functional construct. 6252 (42·4%) of 14 741 mcr-1 mutants showed colistin resistance accompanied by significant fitness costs, when grown under 4 µg/mL colistin selection. 43 (0·3%) mcr-1 mutants exhibited costless resistance, most of which contained multiple mutations. Among the 3449 single mutants of mcr-1, 3433 (99·5%) had a fitness cost when grown in colistin-free media, with a mean relative growth of 0·305 (SD 0·193) compared with the non-functional variant. 3059 (88·7%) and 1833 (53·1%) of 3449 single mutants outgrew the non-functional mcr-1 in the presence of 2 µg/mL and 4 µg/mL colistin, respectively. Single mutations that gave rise to costless mutants were rare in all three domains of mcr-1 (transmembrane domain, flexible linker, and catalytic domain), but the linker domain was enriched with cost-reducing and resistance-enhancing mutations and depleted with cost-increasing mutations. The population genetics model based on the experimental data accurately predicts the rapid decline in mcr-1 prevalence in real-world data. INTERPRETATION: Many identified costless resistant variants that consist of multiple mutations are unlikely to evolve easily in nature. These findings for colistin and mcr-1 might be applicable to other cases in which AMR entails a substantial fitness cost that cannot be mitigated in proximal mutants. FUNDING: National Natural Science Foundation of China, and National Key Research and Development Program of China.

Genome-wide probing of eukaryotic nascent RNA structure elucidates cotranscriptional folding and its antimutagenic effect.

The transcriptional intermediates of RNAs fold into secondary structures with multiple regulatory roles, yet the details of such cotranscriptional RNA folding are largely unresolved in eukaryotes. Here, we present eSPET-seq (Structural Probing of Elongating Transcripts in eukaryotes), a method to assess the cotranscriptional RNA folding in Saccharomyces cerevisiae. Our study reveals pervasive structural transitions during cotranscriptional folding and overall structural similarities between nascent and mature RNAs. Furthermore, a combined analysis with genome-wide R-loop and mutation rate approximations provides quantitative evidence for the antimutator effect of nascent RNA folding through competitive inhibition of the R-loops, known to facilitate transcription-associated mutagenesis. Taken together, we present an experimental evaluation of cotranscriptional folding in eukaryotes and demonstrate the antimutator effect of nascent RNA folding. These results suggest genome-wide coupling between the processing and transmission of genetic information through RNA folding.

Expression level is a major modifier of the fitness landscape of a protein coding gene.

The phenotypic consequence of a genetic mutation depends on many factors including the expression level of a gene. However, a comprehensive quantification of this expression effect is still lacking, as is a further general mechanistic understanding of the effect. Here, we measured the fitness effect of almost all (>97.5%) single-nucleotide mutations in GFP, an exogenous gene with no physiological function, and URA3, a conditionally essential gene. Both genes were driven by two promoters whose expression levels differed by around tenfold. The resulting fitness landscapes revealed that the fitness effects of at least 42% of all single-nucleotide mutations within the genes were expression dependent. Although only a small fraction of variation in fitness effects among different mutations can be explained by biophysical properties of the protein and messenger RNA of the gene, our analyses revealed that the avoidance of stochastic molecular errors generally underlies the expression dependency of mutational effects and suggested protein misfolding as the most important type of molecular error among those examined. Our results therefore directly explained the slower evolution of highly expressed genes and highlighted cytotoxicity due to stochastic molecular errors as a non-negligible component for understanding the phenotypic consequence of mutations.