Dual β-lactam combination therapy for multi-drug resistant Pseudomonas aeruginosa infection: enhanced efficacy in vivo and comparison with monotherapies of penicillin-binding protein inhibition

Abstract

The aim of the study was to determine the efficacy of dual β-lactam combination treatments derived from eight approved drugs against Galleria mellonella larvae infected with MDR strains of P. aeruginosa. Carbapenem-resistant P. aeruginosa NCTC 13437 and an unrelated clinical isolate were used to infect G. mellonella larvae and the efficacy of twenty-eight dual β-lactam combination therapies were compared to their constituent monotherapies. For the most potent combinations identified, penicillin-binding protein (PBP) inhibition profiles were measured and compared with each constituent antibiotic. Five of the dual β-lactam combinations resulted in greater than 70% survival of infected G. mellonella. Two combinations showed potent, enhanced efficacy versus both strains - ceftazidime + meropenem and aztreonam + meropenem. Comparison of PBP inhibition profiles revealed that the enhanced efficacy of these two dual β-lactam combinations could not be explained by more potent inhibition of PBPs or inhibition of a broader range of PBPs. A possible contribution to the enhanced efficacy of the combinations could be stimulation of innate immunity via increased haemocyte numbers compared to their constituent monotherapies. Combinations of β-lactam antibiotics show promise in overcoming MDR P. aeruginosa and are worthy of additional study and development.

Figure 1

Effect of treatment with β-lactam monotherapies or dual combinations on survival of G. mellonella larvae infected with 2.5 × 10³ cfu/mL P. aeruginosa NCTC13437 and treated with PBS or: (a) MEM (2.5 mg/kg), CAZ (5 mg/kg) or MEM + CAZ (2.5 + 5 mg/kg); (b) ATM (50 mg/kg), CTX (100 mg/kg) or ATM + CTX (50 + 100 mg/kg); (c) ATM (50 mg/kg), MEM (2.5 mg/kg) or ATM + MEM (50 + 2.5 mg/kg); (d) ATM (50 mg/kg), PIP (100 mg/kg) or ATM + PIP (50 + 100 mg/kg). A single dose of the antibiotic treatments was administered 2 h p.i. The uninfected group represents larvae sham-infected with sterile PBS and treated with sterile PBS. *Indicates significantly enhanced survival compared to both monotherapies (p < 0.05, log rank test with Holm correction for multiple comparisons); n = 30 (pooled from duplicate experiments).

Figure 2

Survival of G. mellonella larvae 96 h p.i with increasing inoculum sizes (2.5 × 10³, 10⁴, 10⁵ or 10⁶ cfu/mL) of P. aeruginosa NCTC13437 and treated with PBS or dual β-lactam combinations of: (a) MEM + CAZ (2.5 + 5 mg/kg); (b) ATM + CTX (50 + 100 mg/kg); (c) ATM + MEM (50 + 2.5 mg/kg); (d) ATM + PIP (50 + 100 mg/kg). A single dose of the antibiotic treatments was administered 2 h p.i. The uninfected group represents larvae sham-infected with sterile PBS and treated with sterile PBS. Error bars represent the mean ± SEM of replicate experiments.

Figure 3

Effect of treatment with β-lactam monotherapies or dual combinations on survival of G. mellonella larvae infected with 2.5 × 10³ cfu/mL of a carbapenem-resistant clinical isolate of P. aeruginosa and treated with PBS or: (a) MEM (0.5 mg/kg), CAZ (0.5 mg/kg) or MEM + CAZ (0.5 + 0.5 mg/kg); (b) MEM (0.5 mg/kg), PIP (50 mg/kg) or MEM + PIP (0.5 + 50 mg/kg); (c) ATM (2.5 mg/kg), MEM (0.5 mg/kg) or ATM + MEM (2.5 + 0.5 mg/kg); (d) MEM (0.5 mg/kg), CTX (100 mg/kg) or MEM + CTX (0.5 + 100 mg/kg). A single dose of the antibiotic treatments was administered 2 h p.i. The uninfected group represents larvae sham-infected with sterile PBS and treated with sterile PBS. *Indicates significantly enhanced survival compared to both monotherapies (p < 0.05, log rank test with Holm correction for multiple comparisons); n = 30 (pooled from duplicate experiments).

Figure 4

Titration of P. aeruginosa PA01 PBPs by ATM, CAZ or MEM, or ATM + MEM, or CAZ + MEM. Live cells were treated with 10-fold increasing concentrations of each of the β-lactam treatments listed above prior to labelling of PBPs with a fluorescent analogue of penicillin V (Boc-FL). Cells were sonicated and the protein content in each sample equalised to 8 mg/mL. PBPs were separated via SDS-PAGE on 10% Tris-HCl polyacrylamide gels and visualised using a fluorescent scanner (excitation at 473 nm with a 520 nm emission filter). Gel images shown are representative images from duplicate experiments.

Figure 5

Quantitation of Boc-FL labelling of each P. aeruginosa PA01 PBP gel band in the presence of increasing concentrations of either ATM, CAZ or MEM, or the dual combinations of ATM + MEM or CAZ + MEM. The density of Boc-FL labelling of each PBP gel band was quantified relative to the no-antibiotic exposure reference sample and expressed as %. Relative Boc-FL labelling values from two independent experiments were averaged and error bars represent the mean ± SEM.

Figure 6

Effect of exposure to β-lactams on overall numbers of circulating G. mellonella haemocytes. Larvae were sham-infected with PBS and then 2 h p.i administered either PBS, ATM (50 mg/kg), CAZ (5 mg/kg) or MEM (2.5 mg/kg), or combinations of ATM + MEM or CAZ + MEM at the same doses as the single drugs. Numbers of haemocytes in the larval haemolymph were then counted at 5, 20.5 and 24 h p.i. Haemocyte numbers are presented relative to larvae sham-infected with PBS and subsequently treated with PBS. Data is from 9 larvae per time-point showing mean ± SEM.