Application of Genomic Approaches to Bacterial Pathogenesis and Mechanisms of Antimicrobial Resistance

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications & trials

Abstract

Summary MDR gram-negative bacterial pathogens undergo selection and evolution in the natural context of antibiotic treatment in the human host, though important features of host context are often not included in studies of AMR. Our work employs a systems biology approach to study the evolutionary mechanisms by which resistance - defined broadly to include resistance to antibiotics and to host defenses - emerges in the natural context of host infection. Current work is organized around seven related primary projects which span aerobic and anaerobic MDR pathogens, and from basic biology to new treatments: Project 1: Mechanisms and evolution of antimicrobial resistance in Pseudomonas aeruginosa. We previously demonstrated that evolved mismatch repair (MMR) deficiencies may be dynamically exploited by P. aeruginosa to facilitate rapid acquisition of mutations mediating resistance to two critical broad-spectrum antibiotics, ceftazidime-avibactam (CZA), and ceftolozane-tazobactam (C/T), in the context of acute clinical infection (Khil et al, mBio, 2019). We characterized the detailed mutational and transcriptional events underlying the development of CZA and C/T resistance in wild type (WT) and MMR-deficient P. aeruginosa and identified a set of potential novel resistance mutations unique to the MMR deficient isolates. This work established the MexVW-OprM efflux pump as an important (and previously unrecognized) mediator of resistance to CZA and C/T in Pseudomonas (published in Dulanto et al, PLoS Biology, 2022). Given the importance of the MexVW-OprM efflux pump as demonstrated in our prior studies noted above, work completed in FY25 focused on characterization of MexVW-OprM structure, and investigation of its role in mediating resistance to CZA and C/T in clinical P. aeruginosa isolates. To facilitate structural analysis of MexVW-OprM, we engineered a set of vectors and cell lines that express MexV, MexW, and OprM proteins. In collaboration with Susan Buchanan’s intramural group in NIDDK, we have made significant progress with expression and purification to facilitate cryo-EM structural studies. Progress in FY25 from this collaboration included characterization of a 2.09A cyro-EM structure of OprM, the highest resolution single-particle structure for this important protein, which forms the outer membrane subunit of the major RND-class efflux pumps in P. aeruginosa. A manuscript reporting this structure is currently in preparation. This structure is also being used as the basis of design of protein-based inhibitors of this family of multidrug efflux pumps (see Project 2). To study whether MexVW-OprM mediates resistance to CZA and C/T more broadly in clinical P. aeruginosa infections, we engineered mexVW deletions in a collection of clinical P. aeruginosa strains. Work completed in FY25 involved the genetic construction of more than a dozen such isogenic pairs of clinical isolates in which we assessed the role of MexVW-OprM in conferring resistance. These studies demonstrated that MexVW-OprM contributes to CZA and C/T resistance in a set of clinical isolates where resistance is efflux dominated. Deletion of MexVW in a subset of these isoaltes restored susceptibility to CZA and C/T, indicating that MexVW-OprM may serve as an important pharmacologic target (See Project 2). A manuscript is currently in preparation reporting the results of this work. Other work completed in FY25 has involved further refinement of genome assemblies of a 20-year collection of P. aeruginosa bacteremia isolates (320 isolates from 300 unique patients) from the NIH Clinical Center. Isolates from this collection were used in the work described above. Project 2. Design of novel antimicrobial agents. In collaboration with the Vaccine Research Center (VRC), we initiated a new project in FY25 to discover and develop protein-based biologics as a novel class of antimicrobial agents. This collaboration leverages the scientific and clinical subject matter expertise of our group, opportunities unique to the NIH Clinical Center, and the product development expertise of the VRC. VRC efforts are being led by Dr. Danny Douek (Chief of the Human Immunology Section, VRC). The first discovery campaign is targeting RND-class efflux pumps. This targeted approach is enabled by the critical contributions of Project 1 above in collaboration with the Buchanan lab, which has produced a high-resolution Cryo-EM structure of the OprM protein target. OprM is shared by both the MexAB-OprM and MexVW-OprM efflux pumps, and represents the only surface-exposed component of the complex. This high-resolution structure, particularly the extracellular accessible loops, is being used as the basis for computational design and discovery of protein-based antibodies, nanobodies or minibody inhibitors. The Dekker lab will test the activity of lead compounds in antimicrobial resistance and pump inhibition assays against a large panel of curated historical clinical P. aeruginosa isolates (n=300) from the NIH Clinical Center (see Project 1). This panel captures great genetic diversity of clinical isolates and includes isogenic derivatives with deletions of MexAB-OprM and MexVW-OprM engineered by the Dekker lab to distinguish these two pumps in inhibition assays. Lead molecules with confirmed pump inhibitory activity will be down-selected for further testing, with the ultimate future goal of a clinical trial. Project 3: Principles of Intra-host evolution: Bordetella hinzii. This project applies systems biology approaches, including genomics, transcriptomics, and metabolomics to understanding adaptive evolution that occurred in the emerging pathogen Bordetella hinzii during the course of infection in an individual with IL-12RB1 deficiency. The purpose of this work is to characterize foundational principles underlying to how bacterial populations evolve in the context of human infections. Previous work from the lab (Launay et al, Nature Communications, 2021) demonstrated that a mutation in the DNA Pol III epsilon proofreading subunit resulted in a replicative DNA proofreading deficiency that drove genetic divergence among clonal descendants of an original founder population within this patient over the course of 45 months of persistent infection. Evidence of mutational targeting and positive selection were present in multiple sequential enzymes of the tricarboxylic acid cycle and gluconeogenesis pathways, suggesting specialized metabolic adaptation to the host environment. Work begun during FY24 and completed during FY25 (published in Ghosh et al, Nature Communications, 2025) included analysis of B. hinzii transcriptomes (n=175) from the serial clinical isolates from this patient and other human and animal comparators to understand how adaptive changes were reflected in the transcriptomes of the isolates. Functional analysis revealed that multiple niche-specific transcriptome modifications occurred in the lineages recovered from the intravascular and gastrointestinal compartments that involved a variety of metabolic processes as well as loss of the flagella, possibly mediating immune escape. These findings demonstrated substantial unappreciated plasticity in the transcriptome of this model pathogen during host adaptation, with more general implications for how pathogens evolve in the context of chronic infection. Work completed in FY25 also included ongoing study of the potential role of evolved metabolic auxotrophies in generating antibiotic tolerance, a critical and understudied phenomenon of general importance to clinical susceptibility testing and antibiotic selection. Phenylalanine auxotrophies were identified in the clinical B. hinzii isolates that evolved multiple times independently during the course of the chronic infection. Further experiments revealed that these auxotrophies may confer tolerance to meropenem under defined conditions in vitro. A manuscript reporting these findings is in preparation. Other ongoing work completed in FY25 under this project involved establishment of an in vivo human phagocyte infection system to study host-pathogen interactions between the human adapted B. hinzii isolates from this patient and human macrophages. Dedicated eukaryotic culture equipment and techniques were brought into the laboratory, and a vector-based strategy for labeling bacterial cells with fluorescent proteins was developed. Live imaging of infected macrophages, flow cytometry, and dual RNA-seq in conjunction with the generation of isogenic knock out mutations and cytokine panel analysis are planned to study the consequences of potential adaptive changes identified in the prior genomic and metabolic work described above. Among other identified adaptive changes, we will investigate the roles that hypermutation and phenylalanine auxotropy may play in the context of intracellular infection of host phagocytes. Project 4: Principles of intra-host evolution: Burkholderia vietnamiensis. A second project to study within-host evolution and adaptation of a different pathogen – B. vietnamiensis - from another patient with IL-12RB1 deficiency. Previous work performed in the lab under this project involved short read and long-read sequencing of 183 genomes from this patient, which facilitated contiguous genome assemblies to resolve complex structural rearrangements involving all three chromosomes. Work in FY25 involved optimization of new nanopore assembly approaches to refine the long-read assemblies above. We published this work separately to make the approach available to the community (Vereecke at al, Journal Clinical Microbiology, 2025). Analysis of these refined genomes revealed that ongoing insertional mutagenesis by a highly active IS element in combination with homologous recombination drove genomic plasticity and diversification during the course of the infection. This process appears to have resulted in the repeated disruption of genes mediating the stringent/starvation response, and next steps involve studies of whether these modifications facilitated persistence or antibiotic resistance. To test these hypotheses, we will complement functional copies of the disrupted genes and use an optical reporter of (p)ppGpp, a principal metabolite that controls this response for live, non-invasive imaging. We will specifically examine whether activation of the stringent/starvation response by these mutations facilitated persistence of the infection, using an in vivo human phagocyte infection system. Project 5: Antimicrobial resistance and host-pathogen interactions in the Bacteroides fragilis group (BFG). Previous work in the lab involved the construction of a historical collection of clinical BFG isolates spanning four decades. Unlike most of the published isolate collections that are derived from human stool, the isolates in this collection were cultured from infections, which included bacteremia, wound, and intraabdominal abscess sources. In previous work in the lab more than 350 isolates in this collection underwent long-read sequencing from which highly contiguous assemblies were generated. Prior and ongoing work has focused the hypothesis that genome methylation by mobile methyltransferases associated with restriction-modification systems may play a role in regulating expression of antimicrobial resistance genes. Previous work involving methylome sequencing of 268 BFG isolates revealed hundreds of methylation motifs and 6000 putative methyltransferases. Many of the observed methylated motifs were located adjacent to, or within, the gene bodies of AMR genes suggesting the possibility that methylation modulates the expression of AMR genes (Tisza et al, Nature Communications, 2023). Work performed in 2025 has involved long-read and methylome sequencing of another >200 isolates to facilitate computational analysis and pan-genome and pan-methylome associations. A Cas9-based approach has been developed to allow directed deep single-molecule (nanopore) methylome sequencing of the cfiA gene (Vereecke et al, submitted 2025). This approach is being used in combination with digital RT- PCR to study how methylation may modulate expression of this gene and influence meropenem MICs in clinical isolates. Additional work seeks to understand how aerotolerance in clinical BFG isolates cultured from aerated sites (ie blood) may contribute to in vivo virulence and reduced susceptibility to phagocyte oxidases. Work completed in FY25 has involved establishing culture of a human neutrophil (NL-60) cell line and protocols for experiments with control of O2 concentration using anaerobic chamber (O2 of 0.01%) and microfluidic bioreactor in which O2 can be controlled between 1- 21%. Live imaging of infected neutrophils, flow cytometry, and dual RNA-seq in conjunction with the generation of isogenic knock out mutations will be planned to study the roles of individual genes in aerotolerance and virulence in this system. Project 6: (Collaboration) Helicobacter pylori. In collaboration with researchers comprising the Helicobacter pylori genome project (HpGP) Research Network we have contributed to genomic analysis of a large international collection of H. pylori isolates. Work published in FY25 involved a study of unique prophages that are integrated into the H. pylori genome, and their role in the H. pylori cell cycle (Vale, Gut Microbes, 2024), and genomic determinants of antibiotic resistance (Martínez-Martínez, Lancet Microbe, in press 2025). Ongoing work includes a study of mobile genetic element insertions and genome plasticity, and evolutionary metabolic adaption that occurred within H. pylori as it adapted to the human gastric environment. Project 7: (Collaboration) Studies of antibacterial activity of marine-derived pyrrole-imidazole alkaloids. In this work, we have collaborated with Carol Bewley’s group within the intramural NIDDK to study the activity of marine derived pyrrole-imidazole alkaloids against multidrug-resistant strains of Staphylococcus aureus, Escherichia coli, and Acinetobacter baumannii. In work completed in F25, the action of sceptrin, a representative pyrrole-imidazole alkaloid, was examined in E. coli. Multiple resistant lineages were generated in vitro and underwent long-read sequencing to look for genomic mechanisms mediating resistance. Sequencing failed to demonstrated mutations, indels, or other rearrangements conferring heritable resistance, and suggested that resistance was likely induced and reversible. Studies performed by the Bewley lab demonstrated that sceptrin perturbed the bacterial lipid membrane. A combination of lipidomics and SPR analysis confirmed direct, preferential binding to anionic membrane phospholipids, supporting a membrane-targeting mechanism. These studies suggest that pyrrole-imidazole alkaloids may represent a new class of antibiotics for treating both gram-positive and gram-negative infections. A manuscript reporting this work has been submitted (Klein et al, submitted, 2025).