Adaptive response of Pseudomonas aeruginosa under serial ciprofloxacin exposure

Abstract

Understanding the evolution of antibiotic resistance is important for combating drug-resistant bacteria. In this work, we investigated the adaptive response of Pseudomonas aeruginosa to ciprofloxacin. Ciprofloxacin-susceptible P. aeruginosa ATCC 9027, CIP-E1 (P. aeruginosa ATCC 9027 exposed to ciprofloxacin for 14 days) and CIP-E2 (CIP-E1 cultured in antibiotic-free broth for 10 days) were compared. Phenotypic responses including cell morphology, antibiotic susceptibility, and production of pyoverdine, pyocyanin and rhamnolipid were assessed. Proteomic responses were evaluated using comparative iTRAQ labelling LC-MS/MS to identify differentially expressed proteins (DEPs). Expression of associated genes coding for notable DEPs and their related regulatory genes were checked using quantitative reverse transcriptase PCR. CIP-E1 displayed a heterogeneous morphology, featuring both filamentous cells and cells with reduced length and width. By contrast, although filaments were not present, CIP-E2 still exhibited size reduction. Considering the MIC values, ciprofloxacin-exposed strains developed resistance to fluoroquinolone antibiotics but maintained susceptibility to other antibiotic classes, except for carbapenems. Pyoverdine and pyocyanin production showed insignificant decreases, whereas there was a significant decrease in rhamnolipid production. A total of 1039 proteins were identified, of which approximately 25 % were DEPs. In general, there were more downregulated proteins than upregulated proteins. Noted changes included decreased OprD and PilP, and increased MexEF-OprN, MvaT and Vfr, as well as proteins of ribosome machinery and metabolism clusters. Gene expression analysis confirmed the proteomic data and indicated the downregulation of rpoB and rpoS. In summary, the response to CIP involved approximately a quarter of the proteome, primarily associated with ribosome machinery and metabolic processes. Potential targets for bacterial interference encompassed outer membrane proteins and global regulators, such as MvaT.

Keywords: Pseudomonas aeruginosa; antibiotic resistance; ciprofloxacin; iTraq labelling; proteomics.

Fig. 1.. Morphological changes of P. aeruginosa in response to ciprofloxacin. (a) SEM images of the initial unexposed strain (P. aeruginosa ATCC 9027), the strain exposed to ciprofloxacin for 14 days (CIP-E1) and strain CIP-E1 after 10 days of antibiotic-free culture (CIP-E2). The bacterial culture was incubated with PPE stubs and then fixed with 10 % formalin for 24 h. The fixed samples were dried through serial concentrations of ethanol and coated with gold before the image was captured at 2000× magnification. (b) Cell length and width of P. aeruginosa ATCC 9027, CIP-E1 and CIP-E2. The length and width of 30 cells of each sample were measured by ImageJ, analysed by SPSS and presented as mean±sd. *Statistically significant difference (P<0.05).

Fig. 2.. CIP-E1 showing filamented cells under SEM. The CIP-E1 sample was incubated with a PPE stub and then fixed with 10 % formalin for 24 h. The fixed samples were dried through serial concentrations of ethanol and coated with gold before the image was captured at 10 000× (left) and 5000× (right) magnification.

Fig. 3.. Differentially expressed proteins in ciprofloxacin-exposed P. aeruginosa strains. Differentially expressed proteins in CIP-E1 (yellow circle) and CIP-E2 (blue circle) include upregulated proteins (green; normalized fold change >1.5) and downregulated proteins (red; normalized fold change <−1.5). The number is expressed as n (% of total identified proteins) unless otherwise indicated.

Fig. 4.. Protein–protein interaction network of differentially expressed proteins in CIP-E1 using the STRING database. Only known interactions from experimental evidence and existing databases are shown. Disconnected nodes are hidden (https://string-db.org/). Ribosome and translation regulators (red); metabolic pathways (green); ATP metabolic process (purple). Red boxed: efflux pumps and DNA recombination proteins. Protein–protein interactions are shown in evidence view and proteins were linked based on curated, experimental evidence. Network analysis was set at high stringency (STRING score=0.7).

Fig. 5.. Summary of differentially expressed proteins in CIP-E1 by (a) GO-Slim molecular function and (b) GO-Slim biological process. Annotation of the differential expressed proteins was via the PANTHER database (http://pantherdb.org/).

Fig. 6.. Summary of differentially expressed proteins in CIP-E1 and CIP-E2 using (a) GO-Slim molecular function and (b) GO-Slim biological process. Annotation of the differential expressed proteins was via the PANTHER database (http://pantherdb.org/).

Fig. 7.. Gene expression alterations in P. aeruginosa and its ciprofloxacin-exposed strains. (a) Expression changes of selected genes from Table 4 and (b) expression changes of selected global regulators and transcription factors. Fold change was calculated against P. aeruginosa ATCC 9027 and visualized on a log scale, with gene expression of P. aeruginosa ATCC 9027 as 1. Values >1 indicate upregulated genes, while values <1 indicate downregulated genes. Expression of the housekeeping gene rpoD was used as the reference gene value [1953]. Fold change and confidence level (95 % CI, error bar) were calculated in MS Excel according to standard practice [20].