Cefiderocol heteroresistance associated with mutations in TonB-dependent receptor genes in Pseudomonas aeruginosa of clinical origin

Abstract

The siderophore-cephalosporin cefiderocol (FDC) presents a promising treatment option for carbapenem-resistant (CR) P. aeruginosa (PA). FDC circumvents traditional porin and efflux-mediated resistance by utilizing TonB-dependent receptors (TBDRs) to access the periplasmic space. Emerging FDC resistance has been associated with loss of function mutations within TBDR genes or the regulatory genes controlling TBDR expression. Further, difficulties with antimicrobial susceptibility testing (AST) and unexpected negative clinical treatment outcomes have prompted concerns for heteroresistance, where a single lineage isolate contains resistant subpopulations not detectable by standard AST. This study aimed to evaluate the prevalence of TBDR mutations among clinical isolates of P. aeruginosa and the phenotypic effect on FDC susceptibility and heteroresistance. We evaluated the sequence of pirR, pirS, pirA, piuA, or piuD from 498 unique isolates collected before the introduction of FDC from four clinical sites in Portland, OR (1), Houston, TX (2), and Santiago, Chile (1). At some clinical sites, TBDR mutations were seen in up to 25% of isolates, and insertion, deletion, or frameshift mutations were predicted to impair protein function were seen in 3% of all isolates (n = 15). Using population analysis profile testing, we found that P. aeruginosa with major TBDR mutations were enriched for a heteroresistant phenotype and undergo a shift in the susceptibility distribution of the population as compared to susceptible strains with wild-type TBDR genes. Our results indicate that mutations in TBDR genes predate the clinical introduction of FDC, and these mutations may predispose to the emergence of FDC resistance.

Keywords: Gram-negative resistance; Pseudomonas aeurginosa; TonB-dependent receptor; beta-lactamases; cefiderocol; heteroresistance; multi-drug resistance.

Fig 1

TonB-dependent receptor mutant prevalence across four clinical populations of Pseudomonas aeruginosa. (A) Prevalence of mutations in the genes encoding the PirA and PiuA/D cefiderocol import. (B) Distribution of major mutations, defined as those resulting in a frameshift (fs), insertion (ins), deletion (del), and/or early stop codon.

Fig 2

Heteroresistance phenotype associated with TBDR mutations is present independent of the iron state of growth media upon cefiderocol exposure and with controlled inoculum. (A) Comparison of starting inoculum after adjusting to 0.5 McFarland (OD_{600 nm} 0.08–0.1) for (B) population analysis profile performed by agar dilution on ascending concentrations of cefiderocol in Muller-Hinton agar. (C) Comparison of starting inoculum for (D) population analysis performed by growth of strains with ascending cefiderocol concentrations in iron-depleted Mueller-Hinton broth. Dashed line marks the cutoff for heteroresistance (hR). Error bars represent standard deviation of three independent runs. Asterisk above each point represents the statistical difference from PAO1 at each indicated concentration. **, P < 0.01; ****, P < 0.0001; ns, non-significant.

Fig 3

PAP-AUC correlates with a shift in MIC distribution of population after a single FDC exposure. Stacked bar chart with cefiderocol (FDC) MICs of individual colonies recovered from PAP plates after FDC exposure to 0 µg/mL (bottom, light gray), 4 µg/mL (middle, mid-gray), and 16 µg/mL(top, black). Heteroresistant isolates and AUC are shown in the top panel, with susceptible controls in bottom panel.

Fig 4

Correlation of heteroresistance phenotype by PAP-AUC to Kirby-Bauer disk diameter. AUC values versus Kirby-Bauer disk diameters at (A) 24 h and (B) 48 h. Resistant (R, dash line) and susceptible (S, dot-dash line) CLSI disk breakpoints indicated. Linear regression with standard error is overlaid on KD disk diameters.