Polymyxin Combinations Combat Escherichia coli Harboring mcr-1 and blaNDM-5: Preparation for a Postantibiotic Era

Abstract

The rapid increase of carbapenem resistance in Gram-negative bacteria has resurrected the importance of the polymyxin antibiotics. The recent discovery of plasmid-mediated polymyxin resistance (mcr-1) in carbapenem-resistant Enterobacteriaceae serves as an important indicator that the golden era of antibiotics is under serious threat. We assessed the bacterial killing of 15 different FDA-approved antibiotics alone and in combination with polymyxin B in time-killing experiments against Escherichia coli MCR1_NJ, the first reported isolate in the United States to coharbor mcr-1 and a New Delhi metallo-β-lactamase gene (bla_NDM-5). The most promising regimens were advanced to the hollow-fiber infection model (HFIM), where human pharmacokinetics for polymyxin B, aztreonam, and amikacin were simulated over 240 h. Exposure to polymyxin B monotherapy was accompanied by MCR1_NJ regrowth but not resistance amplification (polymyxin B MIC from 0 to 240 h [MIC_0h to MIC_240h] of 4 mg/liter), whereas amikacin monotherapy caused regrowth and simultaneous resistance amplification (amikacin MIC_0h of 4 mg/liter versus MIC_240h of >64 mg/liter). No MCR1_NJ colonies were observed for any of the aztreonam-containing regimens after 72 h. However, HFIM cartridges for both aztreonam monotherapy and the polymyxin B-plus-aztreonam regimen were remarkably turbid, and the presence of long, filamentous MCR1_NJ cells was evident in scanning electron microscopy, suggestive of a nonreplicating persister (NRP) phenotype. In contrast, the 3-drug combination of polymyxin B, aztreonam, and amikacin provided complete eradication (>8-log₁₀ CFU/ml reduction) with suppression of resistance and prevention of NRP formation. This is the first comprehensive pharmacokinetic/pharmacodynamic study to evaluate triple-drug combinations for polymyxin- and carbapenem-resistant E. coli coproducing MCR-1 and NDM-5 and will aid in the preparation for a so-called "postantibiotic" era.IMPORTANCE A global health crisis may be on the horizon, as the golden era of antibiotics is under serious threat. We recently reported the first case in the United States of a highly resistant, Escherichia coli so-called "superbug" (MCR1_NJ), coharboring two of the most worrying antibiotic resistance genes, encoding mobile colistin resistance (mcr-1) and a New Delhi metallo-β-lactamase (bla_NDM-5). Worryingly, the medical community is vulnerable to this emerging bacterial threat because optimal treatment strategies are undefined. Here, we report the activity of an optimized combination using simulated human doses of commercially available antibiotics against MCR1_NJ. A unique triple combination involving a cocktail of polymyxin B, aztreonam, and amikacin eradicated the MCR-1- and NDM-5-producing E. coli Each antimicrobial agent administered as monotherapy or in double combinations failed to eradicate MCR1_NJ at a high inoculum. To our knowledge, this is the first study to propose 3-drug therapeutic solutions against superbugs coharboring mcr-1 and bla_NDM, seeking to prepare clinicians for future occurrences of these pathogens.

Keywords: Enterobacteriaceae; MCR-1; NDM-5; amikacin; aztreonam; carbapenem-resistant; polymyxins.

FIG 1

E. coli MCR1_NJ total population bacterial counts (black lines) and antibiotic-resistant subpopulations (colored lines) quantified during the HFIM at a starting inoculum of either ~10⁶ CFU/ml (A) or ~10⁸ CFU/ml (B). Antibiotic-resistant subpopulations, which are fractions of the respective total population, were quantified using MHA plates imbued with the specified concentrations of polymyxin B (blue lines), aztreonam (red lines), or amikacin (green lines). Humanized regimens of polymyxin B with front loading (3.33 mg/kg for 1 dose followed by 1.43 mg/kg q12h starting 12 h later), aztreonam (2 g q8h), and amikacin (15 mg/kg q24h) were simulated over 240 h.

FIG 2

Scanning electron microscopy images of representative E. coli MCR1_NJ cells at a magnification of ×10,000. Images were obtained from HFIM samples at ~10⁸-CFU/ml inoculum taken prior to antibiotic exposure (A), after 24-h exposure to aztreonam (B), after 240-h exposure to aztreonam (C), after 24-h exposure to polymyxin B plus aztreonam (D), and after 240-h exposure to polymyxin B plus aztreonam (E). Bacterial samples were prepared and imaged on 0.2-μm filter paper, which is seen behind the cells for reference.