Curative Treatment of Severe Gram-Negative Bacterial Infections by a New Class of Antibiotics Targeting LpxC

Abstract

The infectious diseases caused by multidrug-resistant bacteria pose serious threats to humankind. It has been suggested that an antibiotic targeting LpxC of the lipid A biosynthetic pathway in Gram-negative bacteria is a promising strategy for curing Gram-negative bacterial infections. However, experimental proof of this concept is lacking. Here, we describe our discovery and characterization of a biphenylacetylene-based inhibitor of LpxC, an essential enzyme in the biosynthesis of the lipid A component of the outer membrane of Gram-negative bacteria. The compound LPC-069 has no known adverse effects in mice and is effective in vitro against a broad panel of Gram-negative clinical isolates, including several multiresistant and extremely drug-resistant strains involved in nosocomial infections. Furthermore, LPC-069 is curative in a murine model of one of the most severe human diseases, bubonic plague, which is caused by the Gram-negative bacterium Yersinia pestis Our results demonstrate the safety and efficacy of LpxC inhibitors as a new class of antibiotic against fatal infections caused by extremely virulent pathogens. The present findings also highlight the potential of LpxC inhibitors for clinical development as therapeutics for infections caused by multidrug-resistant bacteria.IMPORTANCE The rapid spread of antimicrobial resistance among Gram-negative bacilli highlights the urgent need for new antibiotics. Here, we describe a new class of antibiotics lacking cross-resistance with conventional antibiotics. The compounds inhibit LpxC, a key enzyme in the lipid A biosynthetic pathway in Gram-negative bacteria, and are active in vitro against a broad panel of clinical isolates of Gram-negative bacilli involved in nosocomial and community infections. The present study also constitutes the first demonstration of the curative treatment of bubonic plague by a novel, broad-spectrum antibiotic targeting LpxC. Hence, the data highlight the therapeutic potential of LpxC inhibitors against a wide variety of Gram-negative bacterial infections, including the most severe ones caused by Y. pestis and by multidrug-resistant and extensively drug-resistant carbapenemase-producing strains.

Keywords: LpxC; animal models; antimicrobial drug; antimicrobial resistance; plague.

FIG 1

LpxC inhibitors LPC-058 and LPC-069 display antibiotic activity against Y. pestis in vitro. The molecular structures of LPC-058 and LPC-069, including common features and structural differences, are depicted. The MICs of LPC-058 and LPC-069 were determined for two strains of each of three Y. pestis biovars, grown at the optimal growth temperature (28°C) or at the host temperature (37°C) in Mueller-Hinton medium (MH) supplemented or not with serum albumin (SA). The MICs of LPC-058 and LPC-069 determined with E. coli grown at 37°C in MH medium are also reported. The MICs were determined from three independent biological replicates.

FIG 2

Crystal structure of LPC-069 bound to A. aeolicus LpxC. The structural features of LpxC are depicted as a ribbon diagram, and the catalytic zinc ion is shown as a sphere. LPC-069 and the side chains of zinc-binding residues and important catalytic residues of LpxC are shown as a stick model. Electrostatic interactions between the fluorine atoms and the positively charged K227 and H253 residues are shown as blue dashed lines (right panel).

FIG 3

Effect of LPC-058 and LPC-069 regimens on hematologic, renal, and hepatic parameters in mice. Groups of four 8- to 9-week-old OF-1 female mice were inoculated intravenously with LPC-058 (10 mg/kg q8h, 20 mg/kg q12h, or 40 mg/kg q12h) or LPC-069 (20 mg/kg q12h, 40 mg/kg q12h, 40 mg/kg q8h, or 200 mg/kg q8h) for 5 consecutive days. On the day after the last injection, blood was collected and levels of protein (Prot), urea, alkaline phosphatase (AP), and alanine aminotransferase (ALT) were measured. The data are shown as the mean ± standard deviation. *, P < 0.05 in a Mann-Whitney U test.

FIG 4

LPC-069 confers protection against plague. Eight- to 9-week-old OF-1 female mice were intradermally inoculated with 100 CFU Y. pestis. Eighteen hours later, mice were given a 5-day course of intravenous LPC-058 at 10 mg/kg q8h, LPC-069 at 40 mg/kg q8h (A and B) or 200 mg/kg q8h (C), or doxycycline (Dox) at 50 mg/kg q12h (C). (A) The survival of mice treated with LPC-058 (gray squares) or LPC-069 (white squares) and nontreated mice (black circles) was recorded every 8 h. The survival curves plotted for treated mice differed significantly from the curve plotted for untreated mice (P < 0.0001 in a log rank test). Bacterial loads in the survivors’ skin, draining lymph nodes, spleen, and blood were determined 18 h after the end of the treatment. Horizontal bars indicate the median of the individual data points. A total of 15 mice were used. (B) Bacterial loads in the skin, draining lymph nodes, spleen, and blood of eight mice at the start of treatment and in eight mice treated with LPC-058 (light gray boxes) or LPC-069 (dark gray boxes) and nontreated mice (white boxes) 24, 48, and 72 h after the start of the treatment (i.e., 42, 66, and 90 h postinfection). The bacterial loads given at the 138-h time point are those from the experiment shown in panel A. The boxes represent the interquartile range of the bacterial load, and the lines crossing the boxes indicate the median bacterial load. The whiskers represent the lowest and highest bacterial loads. Dead mice are indicated by skulls and crossbones. A two-way analysis of variance revealed that treatments with LPC-069 or LPC-058 were associated with significantly (P < 0.02) lower colonization of the lymph nodes, spleen, and blood, relative to nontreated controls. (C) Survival of nontreated mice (black circles) or mice treated intravenously with LPC-069 at 200 mg/kg q8h (white squares) or doxycycline (black squares) at 50 mg/kg q12h for 5 consecutive days. LPC-069 protected mice against plague as efficiently as doxycycline did (with 15 and 14 out of 15 survivors at day 21 postchallenge, respectively; P > 0.5). Skin, lymph nodes, spleen, and blood from all survivors were sterile. A total of 15 mice were used.