Dissecting Pseudomonas aeruginosa β-Lactam Resistance: Insights from PKPD Modeling of ampC and ampD Mutations

Study Background and Research Question

The emergence of multidrug-resistant (MDR) Pseudomonas aeruginosa, particularly strains resistant to carbapenems and advanced β-lactam/β-lactamase inhibitor (BL/BLI) combinations, represents a significant clinical and research challenge. Ceftolozane-tazobactam (C/T) has been a therapeutic mainstay for MDR P. aeruginosa, but resistance arising during therapy has been increasingly reported. Traditional susceptibility assays, such as minimum inhibitory concentration (MIC) determinations, often fail to resolve the complexities of adaptive and acquired resistance mechanisms. The reference study (Deroche et al., 2023) addresses this gap by leveraging semi-mechanistic pharmacokinetic/pharmacodynamic (PKPD) modeling to quantify the contributions of specific ampC and ampD mutations to both initial and time-dependent (adaptive) resistance to C/T, and their effects on imipenem (IMI) susceptibility.

Key Innovation from the Reference Study

The innovative core of this study lies in its application of semi-mechanistic PKPD modeling to experimentally engineered P. aeruginosa strains and clinical isolates. By precisely introducing and reversing mutations in ampC (G183D) and ampD (H157Y), the authors could isolate the effects of each mutation—singly and in combination—on antibiotic susceptibility. Crucially, their approach moves beyond static MIC measurements to dynamically model the kinetics of bacterial killing and the emergence of resistance throughout antibiotic exposure. This allows for discrimination between the immediate impact of genetic mutations and the development of adaptive resistance over time, providing a much-needed mechanistic framework for understanding treatment failure and resistance evolution.

Methods and Experimental Design Insights

The research utilized a combination of whole genome sequencing (WGS), targeted genetic engineering, and time-resolved bacterial killing assays. Key features of the experimental workflow include:

• Identification of clinical P. aeruginosa isolates with differential susceptibility to C/T and IMI, confirmed as isogenic by sequence type (ST252).
• Detection of two resistance-associated mutations—ampC G183D and ampD H157Y—via SNP analysis.
• Generation of isogenic PAO1-derived mutant strains (with single and double mutations, as well as reversions in the clinical background) using homologous recombination.
• Sequential time-kill curve experiments for all strains exposed to C/T and IMI, capturing both early and late-phase responses.
• Development of a semi-mechanistic PKPD model, incorporating adaptation parameters to fit experimental data and distinguish between initial (genetic) and adaptive resistance components.

Core Findings and Why They Matter

The study yielded several critical insights into the nuanced interplay between genetic mutations and antibiotic resistance phenotypes:

• Stepwise Resistance Escalation: Introduction of the ampC G183D and ampD H157Y mutations led to 1.4-fold and 4.1-fold increases in C/T EC₅₀ values, respectively, with the double mutant exhibiting a dramatic 29-fold increase initially—a finding sustained and even magnified over time (up to 320-fold at late time points for some mutants).
• Reversion Experiments: Reversal of these mutations in the clinical background substantially reduced EC₅₀ (from 80.5 mg/L to 6.77 mg/L), confirming causality and the additive nature of these genetic changes for C/T resistance.
• Adaptive Resistance Discrimination: The PKPD model enabled clear separation of baseline resistance from adaptive (time-dependent) resistance, which cannot be resolved by standard MIC testing. Notably, the ampC G183D mutation also prevented the development of adaptive resistance to IMI, suggesting a complex trade-off between β-lactam susceptibility profiles.
• Clinical Implications: The approach demonstrates how resistance mutations can simultaneously increase resistance to one agent while restoring susceptibility to another, underscoring the importance of detailed mechanistic profiling in antibiotic stewardship and the design of combination therapies.

For researchers, these results provide a template for investigating resistance mechanisms using dynamic modeling and engineered strains, greatly improving the resolution of antibiotic efficacy studies.

Comparison with Existing Internal Articles

Several internal reviews highlight the significance of advanced cephalosporins and PKPD modeling in resistance research. For example, "Cefepime (BMY-28142): Unveiling New Horizons in CNS Infection Research" and "Strategic Mechanisms and Frontiers for Cefepime" both emphasize the role of broad-spectrum cephalosporins in multidrug resistance modeling and central nervous system infection studies. However, the reference study distinguishes itself by providing quantitative, mutation-specific PKPD data, allowing direct measurement of resistance adaptation kinetics—not just overall efficacy. This mechanistic approach bridges the gap between static in vitro susceptibility data and the dynamic, clinically relevant scenarios often modeled in translational research.

Limitations and Transferability

While the study's use of isogenic strains and precise mutation engineering provides clear mechanistic insights, some limitations should be acknowledged:

• The PKPD modeling was conducted in vitro; in vivo validation remains necessary for direct clinical translation.
• Only two mutations were investigated, though they are representative of major resistance mechanisms in clinical isolates.
• Adaptive resistance parameters may vary across strain backgrounds and antibiotic exposure regimens.

Nevertheless, the workflow is transferable to other pathogens and resistance determinants, provided similar genetic tools and dynamic modeling frameworks are available.

Protocol Parameters

• Strain engineering: Use homologous recombination to introduce or revert ampC and ampD mutations in a PAO1 or clinical background.
• Time-kill assay setup: Inoculate bacterial cultures at defined starting densities; sample at multiple time points post-antibiotic exposure to capture both immediate and adaptive responses.
• PKPD modeling: Fit time-kill data using semi-mechanistic models with explicit adaptation terms; adjust parameters to distinguish between initial susceptibility and the kinetics of resistance emergence.
• Antibiotic selection: Employ agents relevant to the resistance mechanism under study (e.g., ceftolozane-tazobactam, imipenem, or broad-spectrum cephalosporins such as Cefepime for comparative profiling).
• Genomic confirmation: Validate all engineered strains using WGS or targeted sequencing to ensure desired mutations are present and off-target changes are absent.

Research Support Resources

For investigators seeking to replicate or extend PKPD-based resistance modeling, robust research-grade antibiotics with well-characterized spectra and pharmacological profiles are essential. Cefepime (BMY-28142) (SKU BA1013) is a broad-spectrum cephalosporin antibiotic with established activity against Gram-positive and Gram-negative bacteria and demonstrated penetration into the central nervous system. Its use in controlled bacterial infection models, including neurotoxicity and resistance evolution studies, is supported by a strong evidence base. Researchers can source this compound from APExBIO to support dynamic modeling workflows analogous to those described in the reference study. Solutions should be prepared freshly and handled according to best laboratory practices due to potential neurotoxicity and stability considerations.