Colistin resistance in extensively drug-resistant Gram-negative pathogens

点击翻译按钮获取中文摘要

Extensive drug resistance (XDR) in Gram-negative bacteria is widely recognized as a top priority public health issue. Colistin is often the only viable treatment option for infections caused by such pathogens, but morbidity and mortality are substantial even with colistin treatment. Furthermore, clinical use of colistin has led to the emergence of colistin-resistant strains. We have shown, for Acinetobacter baumannii and Klebsiella pneumoniae, that colistin resistance arises during therapy primarily due to modifications of lipid A, the membrane anchor component of the Gram-negative lipopolysaccharide, with amino-containing moieties such as aminoarabinose and phosphoethanolamine, and alternatively galactosamine, primarily mediated through specific mutations in the PmrAB or PhoPQ two-component regulatory systems. We have also demonstrated that rapid profiling of lipid A can serve as a reliable method to predict colistin susceptibility and resistance, an important advance given challenges associated with conventional susceptibility testing of colistin. Meanwhile, pharmacokinetic-based dosing of colistin has been widely implemented in the hope of improving its efficacy but without tangible clinical benefits so far. Our previous investigations have deepened understanding of colistin resistance mechanisms and their implications in A. baumannii, but also identified new knowledge gaps. These include wide ranges of resistance observed in the presence of lipid A modifications with colistin MICs ranging from 4 mg/L to >256 mg/L, difficulties with defining colistin MICs in some strains, and apparent disconnect between in vitro activity and suboptimal clinical outcome, especially among patients with pneumonia. Importantly, in vitro MICs may not always reflect in vivo efficacy of colistin due to remodeling of lipid A that occurs in the host, with or without selective pressure from colistin, and this may account for clinical failure in patients who are treated with ostensibly appropriate doses of colistin. This project aims to address these issues by quantifying resistance-conferring amino-containing lipid A moieties in clinical strains with a wide range of MICs among XDR Gram-negative pathogens, elucidating lipid A remodeling of XDR Gram-negative bacteria in the BAL fluid and the lung tissue of mice with pneumonia, determining lipid A modification directly in BAL specimens of patients infected with XDR Gram-negative pathogens, and correlating the ability of colistin adjuvants to abrogate colistin resistance to levels of Lipid A modification. The proposal therefore advances our understanding of colistin resistance across various XDR pathogens, fills a critical knowledge gap that lies between in vitro and in vivo resistance to colistin, and explores novel colistin adjuvants in abrogating lipid A modifications and colistin resistance. This comprehensive effort will be realized by close and ongoing multidisciplinary collaboration representing clinical microbiology, glycolipidomics, and synthetic chemistry.