Differential contribution of PBP occupancy and efflux on the effectiveness of β-lactams at their target site in clinical isolates of Neisseria gonorrhoeae

Abstract

Neisseria gonorrhoeae exhibits alarming antibiotic resistance trends and poses a significant challenge in therapeutic management. This study aimed to explore the association of penA alleles with penicillin-binding protein (PBP) occupancy patterns and reduced outer membrane permeability, impacting susceptibility to last-line cephalosporins and potential β-lactam candidates. The whole genome sequence, the MICs and PBP IC50s were determined for 12 β-lactams and β-lactamase inhibitors in 8 clinical isolates with varying β-lactam sensitivity, 2 ATCC, and 3 WHO cephalosporin-resistant reference strains. The genetic analysis identified diverse determinants of β-lactam resistance including penA, ponA, porB, and mtrR alterations. Mosaic penA alleles were confirmed to be key determinants of cephalosporin resistance, with notable impacts on PBP2 IC50 affinities (in the presence of all PBPs). Substitutions in positions V316 and A501 exhibited significant effects on β-lactam PBP2 occupancy and MICs. PBP1 inhibition showed marginal effect on β-lactam sensitivity and PBP3 acted as a sink target. Ertapenem and piperacillin emerged as potential therapies against cephalosporin-resistant N. gonorrhoeae strains, along with combination therapies involving tazobactam and/or efflux inhibitors. The study determined the β-lactam PBP-binding affinities of last-line cephalosporins and alternative β-lactam candidates in strains carrying different penA alleles for the first time. These findings provide insights for developing new antimicrobial agents and enhancers against emerging resistant strains. Further research is warranted to optimize therapeutic interventions for cephalosporin-resistant N. gonorrhoeae infections.

Fig 1. Typing, genomic, and phenotypic antimicrobial resistance features of the 2 ATCC reference strains, 8 clinical strains, and the 3 WHO N. gonorrhoeae reference WHO X, WHO Y, and WHO Z strains.

Red indicates a resistant allele, while blue indicates a wild type allele. The MIC values are color-coded from dark blue (lower values) to red (higher values) for each drug. MIC values in bold indicate resistance based on EUCAST breakpoints. Clinical resistance breakpoints are not available for azithromycin, ertapenem, ceftazidime/avibactam, ceftolozane/tazobactam, and piperacillin/tazobactam. The sensitive breakpoint values according to EUCAST are: PEN ≤ 0.06, CFM ≤ 0.125, CTX ≤ 0.125, CRO ≤ 0.125, CIP ≤ 0.03, and TET ≤ 0.5. Azithromycin lacks clinical EUCAST breakpoints; therefore, EUCAST’s epidemiological cut-off (ECOFF) value of MIC > 1 mg/L was used to define resistance. ^a Please refer to Table 1 for antibiotic abbreviations.

Fig 2. Heatmap representing the PBP IC₅₀s of β-lactam antibiotics and β-lactamase inhibitors in the studied N. gonorrhoeae strains.

^a PBP, penicillin-binding proteins. Mosaic alleles highlighted in grey. ^b a fixed concentration of the BLI avibactam or tazobactam was used (4 mg/L). The median values from 3 experiments are shown. Please see Table 1 for drug abbreviations. When the main PBP target was not inhibited using the regular concentrations (0.001 to 0.125 mg/L) an extended range (0.016 to 2 mg/L or 1 to 512 mg/L) was used.

Fig 3. Scatter plots were employed to illustrate the correlation between PBP2 IC₅₀ values and the minimum inhibitory concentrations (MIC) of β-lactam antibiotics, β-lactamase inhibitors (BLI), and the combination of β-lactams with BLI.

Each plot depicts the relationship between PBP2 IC50 and the MICs for each drug across all tested strains. The data were represented on a logarithmic scale and analyzed using a robust fitting method in GraphPad Prism. Light grey dots represent outliers that were included in the analyses. The N. gonorrhoeae strains studied were ATCC 19424 and ATCC 49226, as well as clinical strains NG 3, NG 7, NG 12, NG 14, NG 19, NG 20, NG 21, and NG 22, and WHO reference strains NCTC 13820 (WHO X), NCTC 13821 (WHO Y), and NCTC 13822 (WHO Z). Please refer to Table 1 for drug abbreviations. Italicized values indicate low Pearson correlation coefficients (ρ). ^a To determine an estimated correlation, all values below 0.002 were rounded to 0.001.