Delineating genetic determinants of polymyxin resistance in Serratia marcescens

This proposal aims to define the molecular mechanisms of resistance to polymyxins in Serratia marcescens. The emergence of carbapenem resistance in this nosocomial pathogen poses a treatment dilemma as few treatment options exist. S. marcescens has long been considered intrinsically resistant to polymxyins, a mainstay for treatment of carbapenem-resistant Enterobacteriales. However, we unexpectedly noted that a substantial proportion of carbapenem-resistant S. marcescens (CR-SM) are fully susceptible to polymyxin. In addition, resistant isolates displayed two distinct phenotypes (low-level (R1) and high-level resistance (R2). Polymyxins are cationic peptides targeting negatively charged components of the bacterial outer membrane, notably Lipid A. Chemical modification of Lipid A through enzymatic transfer of L-Ara4N to Lipid A is the major contributor to polymyxin resistance and is catalyzed by the aminoarabinose transferase ArnT. Lipid A analysis of isolates representing the 3 phenotypic groups demonstrated a complete lack of L-Ara4N modification in susceptible isolates and distinct lipid A profiles between R1 and R2 phenotypes. Through whole genome sequencing we have begun to establish putative genetic determinants of polymyxin susceptibility. This includes a unique arnC-like gene only present in susceptible isolates and a flippase related to arnE only encoded in highly-resistant R2 isolates. In addition, R1 and R2 isolates harbored unique arnD and eptA alleles compared to susceptible isolates. Here, we aim to define the mechanisms underlying polymyxin susceptibility in detail to provide critical information on the potential suitability of polymyxins as a treatment option for CR-SM infections. We have assembled a highly skilled team with complementary expertise in microbiology, bacterial genomics and membrane protein biochemistry. In Aim 1, we will determine the prevalence of polymyxin susceptibility in a comprehensive collection of clinical multi-drug resistant (MDR) and non-MDR S. marcescens isolates. Through whole genome sequencing we will refine putative genetic markers of PR phenotypes and through phylogenetic reconstruction determine if the loss or PR occurred repeatedly or represents a stable sublineage. We will then validate candidate genetic markers on PR through a combination of complementation, gene editing and deletion experiments. In Aim 2, we will assess LPS structure and lipid A modifications of isogenic mutants generated in Aim 1 using radiolabeling techniques and mass spectrometry. We will evaluate gene expression of lipid A modifying enzymes and overall differences in transcriptional profiles across phenotypes. Lastly, we will examine the impact of PR mutants on bacterial fitness and virulence. Findings from our study will fill significant gaps in knowledge of the genomic and molecular determinants of intrinsic PR, provide information on the mechanisms of disruption of lipid A modifications, deliver molecular markers for rapid screening of polymyxin susceptibility phenotypes in CR-SM; and improve our understanding of the evolutionary trade-offs underlying the loss of the intrinsic PR phenotype.