Mycobacterium abscessus (Mab) causes serious infections that often require over 18 months of antibiotic combination therapy. There is no standard regimen for the treatment of Mab infections and the multitude of combinations that have been used clinically have had low success rates and high rates for toxicities. With β-lactam antibiotics being safe, double β-lactam and β-lactam/β-lactamase inhibitor combinations are of interest for improving treatment of Mab infections and minimizing toxicity. However, a mechanistic approach for building these combinations is lacking since little is known about which penicillin-binding protein (PBP) target receptors are inactivated by different β-lactams in Mab. We determined the preferred PBP targets of 13 β-lactams and two β-lactamase inhibitors in two Mab strains and identified PBP sequences by proteomics. The Bocillin FL binding assay was used to determine the β-lactam concentrations (IC50) that half-maximally inhibited Bocillin binding. Principal component analysis identified four clusters of PBP occupancy patterns. Carbapenems inactivated all PBPs at low concentrations (0.016-0.5mg/L; cluster 1). Cephalosporins (cluster 2) inactivated PonA2, PonA1, and PbpA at low (0.031-1mg/L; ceftriaxone and cefotaxime) or intermediate concentrations (0.35-16mg/L; ceftazidime and cefoxitin). Sulbactam, aztreonam, carumonam, mecillinam and avibactam (cluster 3) inactivated the same PBPs as cephalosporins, but required higher concentrations. Other penicillins (cluster 4) specifically targeted PbpA at 2-16mg/L. Carbapenems, ceftriaxone and cefotaxime were the most promising β-lactams since they inactivated most or all PBPs at clinically relevant concentrations. These first PBP occupancy patterns in Mab provide a mechanistic foundation for selecting and optimizing safe and effective combination therapies with β-lactams.

Multidrug-resistant bacteria are causing a serious global health crisis. A dramatic decline in antibiotic discovery and development investment by pharmaceutical industry over the last decades has slowed the adoption of new technologies. It is imperative that we create new mechanistic insights based on latest technologies, and use translational strategies to optimize patient therapy. Although drug development has relied on minimal inhibitory concentration testing and established in vitro and mouse infection models, the limited understanding of outer membrane permeability in Gramnegative bacteria presents major challenges. Our team has developed a platform using the latest technologies to characterize target site penetration and receptor binding in intact bacteria that inform translational modeling and guide new discovery. Enhanced assays can quantify the outer membrane permeability of β-lactam antibiotics and β-lactamase inhibitors using multiplex liquid chromatography tandem mass spectrometry. While β-lactam antibiotics are known to bind to multiple different penicillin-binding proteins (PBPs), their binding profiles are almost always studied in lysed bacteria. Novel assays for PBP binding in the periplasm of intact bacteria were developed and proteins identified via proteomics. To characterize bacterial morphology changes in response to PBP binding, high-throughput flow cytometry and time-lapse confocal microscopy with fluorescent probes provide unprecedented mechanistic insights. Moreover, novel assays to quantify cytosolic receptor binding and intracellular drug concentrations inform target site occupancy. These mechanistic data are integrated by quantitative and systems pharmacology modeling to maximize bacterial killing and minimize resistance in in vitro and mouse infection models. This translational approach holds promise to identify antibiotic combination dosing strategies for patients with serious infections.

No abstract