MCR-1-dependent lipid remodelling compromises the viability of Gram-negative bacteria

Abstract

The global dissemination of the mobilized colistin resistance gene, mcr-1 , threatens human health. Recent studies by our group and others have shown that the withdrawal of colistin as a feed additive dramatically reduced the prevalence of mcr-1 . Although it is accepted that the rapid reduction in mcr-1 prevalence may have resulted, to some extent, from the toxic effects of MCR-1, the detailed mechanism remains unclear. Here, we found that MCR-1 damaged the outer membrane (OM) permeability in Escherichia coli and Klebsiella pneumonia and that this event was associated with MCR-1-mediated cell shrinkage and death during the stationary phase. Notably, the capacity of MCR-1-expressing cells for recovery from the stationary phase under improved conditions was reduced in a time-dependent manner. We also showed that mutations in the potential lipid-A-binding pocket of MCR-1, but not in the catalytic domain, restored OM permeability and cell viability. During the stationary phase, PbgA, a sensor of periplasmic lipid-A and LpxC production that performed the first step in lipid-A synthesis, was reduced after MCR-1 expression, suggesting that MCR-1 disrupted lipid homeostasis. Consistent with this, the overexpression of LpxC completely reversed the MCR-1-induced OM permeability defect. We propose that MCR-1 causes lipid remodelling that results in an OM permeability defect, thus compromising the viability of Gram-negative bacteria. These findings extended our understanding of the effect of MCR-1 on bacterial physiology and provided a potential strategy for eliminating drug-resistant bacteria.

Identification of a novel MCR-1 variant displaying low level of co-resistance to β-lactam antibiotics that uncovers a potential novel antimicrobial peptide

“…Consistent with our previous observation [30], MCR-1-expressing cells displayed increased sensitivity to detergents compared with that of the empty plasmid control, whereas M6 expression to some extent recovered host permeability (Figure 1D). We further quantified the degree of membrane permeability by staining the indicated strains with the small molecule n-phenyl-1-napthylamine (NPN), reflecting leakiness in the outer membrane, or propidium iodide (PI), a marker to represent the integrity of the bacterial inner membrane.…”

“…The bacterial outer membrane is the first line of defence, forming a strong permeability barrier against antibiotics, whose permeabilization has been suggested to facilitate the entry of many antibiotics [31][32][33]. Given that recent evidence indicates that MCR-1 causes defect in the permeability of outer membranes [30,34], we wondered if certain MCR-1 variants inducing significant OM defect may exhibit a co-resistance phenotype to other classes of antibiotics. We tested this hypothesis using our previous mcr-1 mutant library, which included 171,769 mutation genotypes [35].…”

“…Subsequently, it is accepted that the reduced mcr-1 prevalence after the ban on colistin-added fodders is firmly associated with the MCR-1-mediated fitness cost in the absence of antibiotic selective pressure [6,7,9]. Notably, our recent work suggested that MCR-1 causes lipid remodelling that results in a defect in the outer membrane permeability, thus compromising the viability of Gram-negative bacteria [30]. These observations indicate that MCR-1 may possess a potential lipid binding pocket that disrupts lipid A distribution in the outer membrane.…”

The fitness landscape of the mobilized colistin resistance gene mcr-1

“…Several reports have shown the relevance of MCR-1 expression on bacteria fitness [11][12][13] which was caused by the embedding of MCR-1 in outer membrane and the modification of lipid A 14 . More recently, we further demonstrated that MCR-1 causes lipid remodeling that resulted in an outer membrane permeability defect, thus compromising the viability of Gram-negative bacteria 15 .…”

“…Indeed, when we overlaid the maximum relative growth observed for the substitutions of each amino acid on the three-dimensional structure of the wild-type MCR-1 (Figure 2C and 2D), we again noticed more beneficial mutations in the transmembrane domain. Interestingly, our recent work demonstrated that a potential lipid A binding pocket of MCR-1 was essential for colistin resistance and maintaining bacterial viability 15 . These observations suggest that MCR-1 variants, at least all possible single nucleotide substitutions, could not lead to a substantial increase in colistin resistance.…”