Rules for penetrating the Gram-negative bacterial envelope

University Of Kentucky, Lexington KY

Investigators

Linked publications & trials

Abstract

The development of new antimicrobials has drastically lagged behind the fast growth of drug resistant pathogenic microbes, especially for Gram-negative (GN) bacteria. To be effective in GN bacteria, entering and staying in the cells is a challenge. Insight into the physicochemical properties that enable the entrance of compounds is critical to direct future effort in the development of effective treatment against GN pathogens. Toward this goal, we will develop methods to accurately quantify the accumulation of small molecule compounds, without requiring fluorescence or other properties for detection, and independent of their antibacterial activity. In addition, we will develop a novel computational algorithm to describe and predict physicochemical properties that favors entry and evade efflux in GN bacteria. The computer model will be validated and refined through the selection and characterization of ~1,500 of compounds. Specifically, we will pursue the following aims: Aim 1) To develop and validate LC-MS/MS based assays to quantify compound partition. We will use a panel of antibiotics as gold standards to develop and validate assays to accurately determine the accumulation of small molecule compounds in different cellular compartments. We will develop LC-MS/MS assays and experiment with the kinetics of entrance, optimizing the conditions to enable reproducible and accurate quantification. We will use a CRE strain of E. coli and a MDR strain of Acinetobacter baumannii in this study. Aim 2) To develop a computational algorithm for predicting compounds that are potentially good penetrators for GN bacteria. We will develop a new and highly accurate computational algorithm through machine learning which can systematically recognize all favorable structural factors of the compounds effective for GN bacteria. Selected compounds will be experimentally characterized in Aim 3, the knowledge from which will be incorporated back into refining the computational algorithm to generate a new round of learning/prediction. This cycle will be repeated, with the testing of ~1,500 compounds in total. Aim 3) To measure the accumulation and subcellular distribution of selected compounds and to validate rules on compound accumulation. With the assays from Aim 1, we will measure the accumulation data of compounds selected through Aim 2. Data from these analyses will be used to polish the computer model in Aim 2. Finally, the models will be evaluated through measurement of accumulation data of selected compounds and examine if the experimental data match the prediction. The knowledge gap about penetrating the cellular envelope has become the biggest hurdle to the rational design of new classes of antimicrobials against all GN pathogens. Outcome from the proposed study will help bridge the gap between a potent inhibitor and an effective antibacterial drug. The combined use of the detailed experimental assays and the new computational algorithm will empower drug discovery efforts that aim to identify promising drug candidates targeting GN bacteria.