Antimicrobial resistance evolution in Pseudomonas aeruginosa in an artificial sputum in-vitro model

Saulawa, Ummasalma (2023) Antimicrobial resistance evolution in Pseudomonas aeruginosa in an artificial sputum in-vitro model. Doctor of Philosophy thesis, University of Liverpool.

Text 201132142_AUG2023.pdf - Author Accepted Manuscript Access to this file is embargoed until 1 January 2025. Download (7MB)

Abstract

The increase in antimicrobial resistance is a growing public health concern. Pseudomonas aeruginosa is a bacterium associated with high levels of antimicrobial resistance, that has been identified by the World Health Organisation as a priority one pathogen. It can cause a variety of different infections, including both acute and chronic lung infections. During chronic lung infections, P. aeruginosa evolves through a series of genetic and lifestyle changes that promote resistance to multiple antibiotics. This makes these bacterial infec-tions difficult to eradicate. P. aeruginosa can cause chronic lung infections in people with cystic fibrosis. These individuals can suffer from frequent pulmonary exacerbation which is commonly treated using combination antibiotic therapy. However, the choice of anti-biotics is usually empirical. Despite intensive antibiotic treatment, infections are rarely eradicated. In this study, the impact of sub-inhibitory antibiotics on the P. aeruginosa reference strain PAO1 was investigated. PAO1 was evolved in artificial sputum media using tobra-mycin and ceftazidime in single and dual combination therapy. Following this, the impact of cycling the two antibiotics was also explored. The overall aim was to investigate whether the antibiotics drove resistance to a panel of antibiotics and modelled which treatment might be most effective at limiting the evolution of resistance. In these exper-iments, the bacterial populations were treated with sub-lethal antibiotics for 2 weeks (transfer 4; T4) to mimic treatment during exacerbation and then recovery was studied by evolving the populations for a further 2 weeks (transfer 8; T8) in the absence of antibiot-ics. Bacterial growth across the treatment groups was consistent throughout the study. Antimicrobial resistance profiles of the evolved isolates were evaluated at T4 and T8. The percentage of bacteria able to grow at double the tobramycin (0.125µg/ml) concentration used in the experiment significantly increased over time, including when antibiotic pres-sure was removed, while for ceftazidime, there were no significant changes in resistance over time. Therefore, this shows that P. aeruginosa develops low-level resistance to the antibiotics used in treatment during the experiment, but this does not reach the break-point to be considered resistant. Furthermore, the use of tobramycin or ceftazidime in single or dual therapy does not drive resistance to ciprofloxacin, aztreonam or tazobac-tam/piperacillin. However, the most striking finding was that the use of tobramycin in either a single treatment or in combination with ceftazidime drove resistance to mero-penem. In the tobramycin single treatment group, 63.3 % of isolates were sensitive to meropenem, 2.5% of isolates displayed intermediate resistance and 34.2% were resistant at T4. Meropenem resistance then decreased during the recovery period to only 3.3% of isolates displaying resistance. Surprisingly, the dual combination treatment showed a higher level of meropenem resistance (75% at T4), which then persisted longer and was found to be at 57% after withdrawal of antibiotic treatment during the recovery period (T8). Cycling the antibiotics was found to alter the resistance dynamics. Furthermore, the phenotypes of isolates from different treatment groups, both meropenem resistant and susceptible were studied in greater detail. P. aeruginosa can display a variety of different colony morphologies, and diversity in morphotype is often associated with chronic infec-tion and adaptation to the lung environment. Wrinkly morphologies have been associat-ed with alterations in biofilm formation and redox-driven adaptation resulting in maxi-mal oxygen accessibility to increase the surface area under oxygen limitation. The input (wild type) PAO1 had a smooth, green-pigmented morphology, whereas single tobramycin or ceftazidime antibiotic treatment resulted in changes in both forms (wrinkly) and pig-mentation. However, wrinkly colonies were also detected when P. aeruginosa was grown in artificial sputum in the absence of antibiotics. The use of cycling antibiotics of ceftazidime followed by tobramycin limited changes in colony morphology and all colo-nies appeared like the wild-type PAO1. Analysis of growth revealed that resistance does not carry a fitness cost in nutrient-limited or nutrient-rich environments and meropenem resistant isolates showed a significant increase in area under the curve (AUC). Five re-sistant isolates displayed significantly increased β-lactamase activity and were therefore a minor driver of resistance in the meropenem-resistant isolate group. However, increased efflux pump activity was identified in resistant isolates from the dual treatment group. Whole genome sequencing was performed on the subset of isolates selected across the treatment groups. Genome sequences were determined for 53 selected isolates. For all the sequenced isolates, mutations arose in genes known to be associated with resistance but also in genes whose functions have not been identified yet. These include transcrip-tional regulatory factors such as lasR, mvfR, aguR, pqsA and mexT, type 6 secretory sys-tem genes, metabolic genes, efflux pump proteins and other small molecules transporta-tion proteins. In addition, genes associated with biofilm formation (fha1, dlt, bifA) and prophage (pf1) were also identified. The ftsY gene, which promotes resistance to mero-penem, was also identified in the resistant isolates. In the context of chronic infection, whereby P. aeruginosa is rarely eradicated, determining which antibiotic combinations either enhance or limit resistance is of paramount importance. Antibiotic choice is often not determined using a logical basis and over time both resistance and antibiotic-associated toxicities limit the choice available to people with CF. This study provides in-sight into the importance of understanding resistance dynamics and suggests that both individual antibiotic choice and treatment type (combination or cycling) may have a key impact on collateral resistance.

Item Type:	Thesis (Doctor of Philosophy)
Divisions:	Faculty of Health and Life Sciences
Depositing User:	Symplectic Admin
Date Deposited:	29 Jan 2024 15:43
Last Modified:	29 Jan 2024 15:43
DOI:	10.17638/03173404
Supervisors:	Fothergill, Joanne Floyd, Rachel Winstanley, Craig
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3173404