Collateral susceptibility-guided alternation of ceftolozane/tazobactam with imipenem prevents resistance development in XDR Pseudomonas aeruginosa biofilms

Abstract

Objectives: New combinations of β-lactams and β-lactamase inhibitors, such as ceftolozane/tazobactam could be useful to combat biofilm-driven chronic infections by extensively resistant (XDR) Pseudomonas aeruginosa but resistance development by mutations in the Ω-loop of AmpC has been described. However, these mutations confer collateral susceptibility to carbapenems. Thus we aimed to evaluate the therapeutic efficacy and the prevention of resistance development of regimen alternating ceftolozane/tazobactam and imipenem.

Methods: A carbapenem-resistant XDR P. aeruginosa clinical strain (ST175, 104-B7) and its isogenic imipenem-susceptible ceftolozane/tazobactam-resistant mutant derivative (AmpC T96I, 104-I9) were used. Experiments of single strains and mixed (104-B7 and 104-I9, 1:0.01 ratio) biofilms were performed. 48h biofilms (flow cell system) were treated for 6 days with either ceftolozane/tazobactam, 4/4 mg/L or the alternation of ceftolozane/tazobactam (2 days)-imipenem 4 mg/L (2 days) - ceftolozane/tazobactam (2 days). After treatment, biofilms were collected and plated on Mueller-Hinton agar± ceftolozane/tazobactam 4/4 mg/L. Structural dynamics were monitored using confocal laser scanning microscopy and images were processed with IMARIS software. At least, three independent triplicate experiments per condition were performed. Emerging resistant mutants were characterized through whole genome sequencing (Illumina).

Results: Ceftolozane/tazobactam monotherapy failed to reduce the biofilms of the 104-B7 XDR strain and led to the selection of resistant mutants that showed AmpC Ω-loop mutations (T96I, L244R or aa236Δ7). On the contrary, alternation with imipenem enhanced activity (3 Logs reduction at day 6) and prevented the emergence of ceftolozane/tazobactam-resistant mutants. Likewise, treatment with ceftolozane/tazobactam dramatically amplified the resistant strain 104-I9 in mixed biofilms (>90 % of the population), while the alternation regimen counterselected it.

Conclusions: Collateral susceptibility-guided alternation of ceftolozane/tazobactam with imipenem effectively prevented the selection of resistant mutants and thus could be a potential therapeutic strategy for the treatment of P. aeruginosa XDR chronic infections.

Keywords: Antagonistic resistance mechanisms to antibiotics; Biofilms; Pseudomonas aeruginosa; XDR.

Fig. 1

Diagram of control and treatments. After the inoculation (t-2), 48 h-old biofilms (t0), were either followed as controls without treatment during 6 days (t2, t4 and t6) or challenged with ceftolozane/tazobactam 4/4 mg/L (t2), to be divided, afterwards, in two subbranches: with ceftolozane/tazobactam in monotherapy during 4 days (t4 and t6), or with the alternation of imipenem 4 mg/L (2 days, t4) and ceftolozane/tazobactam (2 days, t6). Cell counts and images were monitored at time points t0, t2, t4 and t6 with, at least, three independent experiments per condition. INOC stands for inoculation; CTRL stands for control biofilms (without treatment); TL/TZ stands for ceftolozane/tazobactam and IM stands for imipenem.

Fig. 2

Dynamics over time of the bacterial populations of 104-B7 biofilms: (a) control; (b) treated with 4/4 μg/ml ceftolozane/tazobactam (TL/TZ), 6 days; and (c) treated with 4/4 μg/ml TL/TZ (2 days), 2 μg/ml imipenem (IM) (2 days) and TL/TZ (2 days). The results represent the averages (symbols) and standard deviations (error bars) from at least three independent experiments. ∗Stands for statistically significant differences (p < 0.05) in total CFU reduction between treatments (Mann Whithney U test).

Fig. 3

Biomass (μm³/μm²) analysis for control (black bars) and treated with ceftolozane/tazobactam (TL/TZ) (dark grey bars) and alternation regime TL/TZ-imipenem (IM)-TL/TZ (light grey bars) biofilms formed by 104-B7 obtained with the COMSTAT program for the quantification of three-dimensional biofilm structures. The results represent the averages (bars) and standard deviations (error bars) from at least three independent experiments. ∗Stands for statistically significant differences (p < 0.05) in biomass (ANOVA).

Fig. 4

Three-dimensional images and transversal sections of 104-B7 control biofilms, treated with ceftolozane/tazobactam (TL/TZ) and with alternation regime (TL/TZ-imipenem (IM)-TL/TZ), stained with propidium iodide (red). The images obtained at four time points (t0, t2, t4, and t6) are shown. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

a) Dynamics over time of mixed biofilms viable cells (initial 1:0.01 ratio; 104-B7:104-I9, respectively); control (black bars) and treated with ceftolozane/tazobactam (TL/TZ) (dark grey bars) and alternation regime TL/TZ-imipenem (IM)-TL/TZ (light grey bars). The results represent the averages (bars) and standard deviations (error bars) from at least three independent experiments. ∗Stands for statistically significant differences (p < 0.05, ANOVA). b) Proportion (%) of TL/TZ resistant mutants in control, treated with TL/TZ and with the alternation regime TL/TZ-IM-TL/TZ from mixed biofilms. The results represent the averages (symbols) and standard deviations (error bars) from at least three independent experiments. ∗Stands for statistically significant differences (p < 0.05) between treatments vs controls and ^•stands for statistically significant differences (p < 0.05) between treatments (Student’ t-test).

Fig. 6

Three-dimensional images and transversal sections of mixed control biofilms [initial 1:0.01 ratio; 104-B7 (EYFP, yellow):104-I9 (ECFP, blue), respectively] and along treatment with ceftolozane/tazobactam (TL/TZ) and with alternation regime (TL/TZ-imipenem (IM)-TL/TZ). The images obtained at four time points (t0, t2, t4, and t6) are shown. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)