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

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

MDR gram-negative bacterial undergo selection and evolution in the natural context of antibiotic treatment in a human host, though important features of host context are often not included in studies of AMR. Additionally, other features underlying bacterial resilience in the context of infection including the ability to evade host defenses often synergize with specific AMR mechanisms and consequently have linked evolutionary relationships. 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 three specific projects: Project 1: Mechanisms by which mismatch repair (MMR) deficiencies can facilitate the rapid evolution of resistance to antibiotics in P. aeruginosa. P. aeruginosa is an important pathogen responsible for significant morbidity and mortality. We previously demonstrated that evolved MMR deficiencies may be dynamically exploited by P. aeruginosa to facilitate rapid acquisition of mutations mediating resistance to ceftazidime-avibactam (CZA), a new class of broad-spectrum beta lactam/beta lactamase inhibitor in the context of acute infection over a period of days (Khil et al, mBio, 2019). Using a combination of directed in vitro microevolution and highly parallel genomic analysis, we now have characterized the detailed mutational and transcriptional events underlying the development of CZA resistance in wild type (WT) and MMR-deficient P. aeruginosa. While the development of resistance in WT isolates was dominated by alterations in gene targets known to be involved in beta lactam resistance, MMR-deficient isolates were able to develop high-level CZA resistance surprisingly without mutations or transcriptional alterations in these previously described targets, implicating novel mechanisms. Current work described below includes functional characterization of novel targets involved in resistance to broad spectrum cephalosporin/beta lactamase inhibitor combination antibiotics identified with this approach. Project 2: In vivo evolution of an emerging zoonotic pathogen (Bordetella hinzii) in an immunocompromised host. This project applies system biology approaches to understanding the adaptive evolution in the emerging pathogen Bordetella hinzii following presumptive zoonotic transfer from an animal reservoir to an individual with interleukin-12 receptor deficiency. Initial work involved whole genome sequencing of 24 B. hinzii isolates cultured from blood and stool over a period of 45 months, which revealed a single clonal lineage that had undergone extensive within-host genetic change reflected in a remarkable degree of observed phenotypic diversity. Analysis demonstrated that the majority of isolates shared a founding mutation in the DNA Polymerase III Epsilon-subunit active site, resulting in a replicative DNA proofreading deficiency. Within these proofreading-deficient lineages, secondary compound hypermutators with complex alterations in mutational spectra emerged and dominated clinical cultures for a period of 12 months, demonstrating their superior in vivo fitness. Statistical signatures of mutational targeting were present in multiple sequential enzymes of the tricarboxylic acid cycle and gluconeogenesis pathways, suggesting specialized metabolic adaptation to the host environment (Launay et al, Nature Communications, 2021). Current work described below employs highly parallel RNA-seq in combination with deep metabolic phenotyping to characterize the detailed transcriptional landscape and metabolic physiology of adaptation. Project 3: Comprehensive whole genome sequencing and genomic analysis of a historical collection of clinical Bacteroides fragilis group (BFG) isolates spanning decades. Members of the BFG are important constituents of the human microbiota, but they can also behave as significant pathogens in certain contexts. Historically, antimicrobial susceptibility patterns in BFG isolates were largely predictable, allowing effective use of empiric treatment regimens. Alarming increases in AMR have recently necessitated reconsideration of empiric strategies. To understand the genomic basis of these AMR trends, we have initiated an effort to sequence a large group of clinical BFG isolates spanning a period of five decades. This work has so far resulted in the construction of 400 long read genomes. Current analysis is focused on AMR gene content of the BFG pan-genome. Work completed during FY21: Project 1. Sequencing and analysis of more than 100 P. aeruginosa transcriptomes from isolates generated from the in vitro adaptive evolution studies described above was performed. This analysis allowed mapping of detailed evolutionary trajectories of lineages at single transcriptome resolution as resistance to ceftazidime-avibactam (CZA) evolved. Integration of genomic and transcriptomic data revealed that MMR-deficient hypermutators evolved CZA resistance through distinct evolutionary pathways involving novel mechanisms. Further work in 2021 included the detailed study of one such mechanism mediating resistance to important cephalosporin/beta lactamase inhibitor combination antibiotics based on the MexVW efflux pump. Genetic engineering of the specific identified MexVW mutations in an isogenic background conferred resistance to multiple cephalosporins and cephalosporin/beta lactamase inhibitor combination antibiotics, defining a new mechanism for resistance to CZA and ceftolozane/tazobactam in P. aeruginosa. Project 2. Work completed during FY 2021 focused on three aims. (1) Completion of genomic and phylogenetic analysis of the B. hinzii isolates along with a functional analysis of the molecular mechanisms and mutational spectra driving adaptation to the human host. This work was published as Launay A, Wu CJ, Dulanto Chiang A, Youn JH, Khil PP, Dekker JP. In vivo evolution of an emerging zoonotic bacterial pathogen in an immunocompromised human host. Nat Commun. 2021 Jul;12(1):4495. (2) Characterization of the transcriptional landscape in the isolate set. The goal of this work is to understand how networks controlling gene expression are reprogrammed to facilitate evolution of AMR and host adaptation. More than 180 transcriptomes were sequenced, allowing detailed analysis of the transcriptional landscape. Additionally, nanopore-based direct long read RNA sequencing was used to facilitate annotation of operons in the transcriptome. Unsupervised hierarchical clustering grouped trancriptomes of proofreading-deficient isolates separately from proofreading-intact lineages and revealed differential expression in a variety of important functional gene classes. (3) A program to perform array-based metabolic phenotyping of the entire isolate set was initiated. Metabolic data will be integrated with transcriptome data to inform a genome-scale metabolic/physiologic model to understand the more general biology of pathoadaptation. Project 3. Long-read nanopore-based sequencing of 400 BFG genomes was completed to allow assembly of end-to-end contiguous assemblies of chromosomes, episomes, and plasmids. Detailed phylogenetic reconstruction and exhaustive annotation of AMR elements in both genome and plasmids has been performed. Comparison with phenotypic susceptibility data has allowed genotype-phenotype analysis and development of an approach to the prediction of antimicrobial resistance based on gene content.