Isolation, Characterization, and Genomic Analysis of Bacteriophages Against Pseudomonas aeruginosa Clinical Isolates from Early and Chronic Cystic Fibrosis Patients for Potential Phage Therapy

Abstract

Pseudomonas aeruginosa is associated with both community and hospital-acquired infections. It colonizes the lungs of cystic fibrosis (CF) patients, establishing an ecological niche where it adapts and evolves from early to chronic stages, resulting in deteriorating lung function and frequent exacerbations. With antibiotics resistance on the rise, there is a pressing need for alternative personalized treatments (such as bacteriophage therapy) to combat P. aeruginosa infections. In this study, we aimed to isolate and characterize phages targeting both early and chronic P. aeruginosa isolates and evaluate their potential for phage therapy. Four highly virulent phages belonging to myoviral, podviral, and siphoviral morphotypes were isolated from sewage samples. These phages have a broad host range and effectively target 62.5% of the P. aeruginosa isolates with a positive correlation to the early isolates. All the phages have a virulence index of ≥0.90 (0.90-0.98), and one has a large burst size of 331 PFU/cell and a latency period of 30 min. All phages are stable under a wide range of temperature and pH conditions. Genomic analysis suggests the four phages are strictly lytic and devoid of identifiable temperate phage repressors and genes associated with antibiotic resistance and virulence. More significantly, two of the phages significantly delayed the onset of larval death when evaluated in a lethal Galleria mellonella infection model, suggesting their promise as phage therapy candidates for P. aeruginosa infections.

Keywords: Pseudomonas aeruginosa; antimicrobial resistance; bacteriophage; cystic fibrosis; lytic phage.

Figure 1

Midi-throughput phage hunting. Overview of the workflow used for phage hunting. Stage 1: sewage sample preparation through filtration and concentration. Stage 2: stocking of bacterial isolates in 96-well plates, preparation of hunting plates with both bacterial isolates and sewage samples, incubation, and data analysis for the presence of potential phages. Stage 3: mini liquid propagation, followed by testing for potential phage isolation by the double agar layer assay. Abbreviations: F = filtrate, C = concentrate. The figure was created with BioRender.com.

Figure 2

Bar plots showing the spot assay versus PHIDA results. (A) Spot assay results. The numbers represent the lysis pattern in the spot: 4 = complete clearing, 3 = clearing with hazy background, 2 = substantial turbidity through the cleared zone, 1 = few individual plaques, and 0 = no clearing. (B) Phage–host interaction results expressed as C = complete inhibition of the bacterial growth, D+ = more than 5 h delay in bacterial growth compared to control, D = time difference to reach detection threshold between sample and control is ≥1 and <5 h, and N = small effect on bacterial growth endpoint.

Figure 3

Transmission electron micrographs of the four phages. (A) U17 shows a podoviral morphotype, and (B) AA17 shows a siphoviral morphotype, while (C) AA20 and (D) AC20 show a myoviral morphotype. The phages were purified using CsCl density gradient centrifugation, and then for imaging, they were negatively stained with 1% uranyl acetate, and images were captured by TEM HITACHI H-7500 at a scale bar of 50 nm. The virion dimensions were measured by ImageJ v1.53.

Figure 4

pH stability of the four tested phages. Following an hour of exposure to varying pH levels, phage titers were determined. The pH stability of phages was compared to SM control (pH 7.4). SM control is represented in black, and shades of each color represent pH from low to high. Error bars show the standard deviation of a biological duplicate and technical triplicate experiment. Statistical analysis was performed on each phage set by GraphPad Prism (v9) (GraphPad Software Inc., San Diego, CA, USA) using a one-way ANOVA test, followed by Dunnett’s multiple comparisons test. Statistical significance is represented as ** p ≤ 0.01 and *** p ≤ 0.001.

Figure 5

Phage temperature stability. (A) The stability of four phages at 4 °C over 6 months. (B) The phage stability at variable stressing temperatures (black for control tested at 4 °C). Error bars show the standard deviation of a biological and technical triplicate experiment. Statistical analysis was performed on each phage set by GraphPad Prism (v9) (GraphPad Software Inc., San Diego, CA, USA) using a one-way ANOVA test, followed by Dunnett’s multiple comparisons test. Statistical significance is represented as * p ≤ 0.05 and **** p ≤ 0.0001.

Figure 6

Genomic maps of the four phages. Circularized genomic maps (for ease of viewing, genomes are packaged as linear fragments) of (A) U17 and (B) AA17. (C) Linear genomic maps of AA20 and AC20. Functional assignment of the predicted proteins encoded by each CDS is as follows: hypothetical (navy), morphogenesis (green), transcription and translation (teal), virulence (red), lysis (pink), DNA packaging (blue), regulation and metabolism (black), recombination (dark red), tRNA (yellow), and DNA modifications (orange). (D) LASTZ alignment of AC20 genome against AA20 genome. The figure was generated using Geneious Prime.

Figure 7

A rooted tree of the large terminase proteins of (A) U17 and (B) AA17. Each tree was generated using the large terminase protein from each phage against the top 100 BLASTp hits from the NCBI database. A Clustal Omega (v 1.2.3) protein alignment was performed, and the resulting alignment was run with RAxML (v8.2.11) to generate a phylogenetic tree. Green represents large terminase protein of U17 and AA17 and red represents large terminase protein hits with percentage similarity between 95 and 100% to U17 and AA17. The figure was generated using Geneious Prime.

Figure 8

Proteomic tree of phages AA20 and AC20. (A) Circular proteomic tree based on genome-wide similarities of phages AA20 and AC20 (marked with red stars) and closely related reference phage genomes. (B) Rectangular proteomic tree of phage phages AA20 and AC20 (marked with red stars and labeled in red) and 43 phages with the highest ViPTree SG scores.

Figure 9

A clinker gene cluster comparison. A comparison of whole genomes for phages AA20 and AC20 against six other closely related phages, with percent amino acid identity represented by greyscale links between genomes. Each similarity group is assigned a unique color.

Figure 10

One-step growth curves of the four phages at MOI of 0.001. Data were means of three biological and technical replicates, and error bars represent standard deviation.

Figure 11

Growth suppression assay. For the four phages, growth suppression kinetics was tested using a combination of the respective bacterial host and six MOIs, starting from 1000 and as low as 0.001. The mean of biological and technical triplicates is represented at each time point, and error bars represent standard deviation.

Figure 12

G. mellonella phage rescue assays against infection with the corresponding P. aeruginosa strain. SM buffer was used as negative treatment control, and larval survival was monitored for a total of 48 h.

Figure 13

Twitching motility ability of the P. aeruginosa panel. The 56 isolates were tested for twitching motility by stabbing 10 uL of bacterial cultures of OD₆₀₀ of 2.0 into the bottom of Eiken agar plates. Each dot shows the area measured at the end of the experiment for each bacterial strain. The black horizontal line represents the mean area in each group. The experiment was performed in biological triplicates, with four plates for each replicate.