SPF Balb/c mice at 6 weeks of age were purchased from Chengdu Chengda Shuo Experimental Animal Co., Ltd. in Sichuan, China. All animal experiments in this study complied with the guidelines of the Chinese Animal Welfare Committee and were approved by the Animal Welfare and Research Ethics Committee of Northwest A&F University in Shaanxi, China.

The P. aeruginosa strains used in this study were previously isolated and kept in the Epidemic Prevention and Control Laboratory of Northwest Agriculture and Forestry University. All 165 strains were previously cultured by cetyltrimethylamine bromide (Qingdao Haibo Biological Company, China) and PCR. Other isolates were isolated, identified and preserved by the Epidemic Prevention and Control Laboratory of Northwest A&F University. All strains were cultured in LB broth at 37°C with 50% glycerol in sterile LB broth and stored in a −80°C refrigerator.

The isolation and purification of the phage was performed using a bilayer plate method (Wang et al., 2017). Manure, soil and sewage were collected from a forest musk farm in Shaanxi Province, China. Manure and soil were soaked in saline overnight and initially filtered through three layers of gauze, and the supernatant was retained. The effluent samples were centrifuged at 6,000 rpm for 10 min and the supernatant was taken. The two supernatants were mixed with 300 μL of indicator bacteria of P. aeruginosa 27 (PA27) in 100 mL LB broth, incubated overnight at 37°C, and then centrifuged at 12,000 rpm for 10 min. The supernatant was collected and filtered through a 0.22 μm membrane to obtain the phage enrichment solution. One hundred microliters of the stock solution was mixed with the same volume of PA27 bacterial solution and incubated at 37°C for 5 min. Finally, 200 μL of the mixture was mixed with 5 mL of melted LB agar at 50°C and poured on top of the LB plates. The plates were left to stand for 10 min and placed to incubate at 37°C for 12 h to produce plaques. Individual phage plaques were selected and dissolved in 100 μL of SM buffer for 30 min at 37°C, then centrifuged at 12,000 rpm for 10 min. The supernatant was taken for 10-fold ratio dilution and purified at least 5 times by double layer plate method to obtain phage plaques of uniform size and shape.

The morphology of phages YL1 and YL2 was determined by transmission electron microscopy (TEM) observation (Ackermann, 2009). Specifically, 20 μL of phage suspension (3.3 × 10¹¹ PFU/mL) was dropped on the surface of a copper mesh and adsorbed for 15 min. Then, the phage was stained with 2% phosphotungstic acid for 5 min, washed three times with ddH₂O, and dried before observing the phage and recording the images using a HT7700 transmission electron microscope (TEM, Hitachi, Japan).

The host range of phage YL1 and YL2 was tested using a double-layer agar plate method with 165 strains of P. aeruginosa and other strains. The tested bacterial broth was freshly prepared at 10⁸ CFU/mL. A 100 μL volume of phage YL1 and YL2 was taken and mixed with the same volume of the tested bacterial broth, and then incubated at 37°C for 5 min. Finally, the phage titers were measured as above.

One-step growth curves were experimented as previously reported, with some modifications (Lee et al., 2021). Phages YL1 and YL2 were mixed with the host strain of PA27 at optimal infection rate (MOI) respectively and then incubated at 37°C for 5 min. The mixture was centrifuged at 12,000 rpm for 30 s and washed twice with sterile LB broth to remove unabsorbed phages. The sediment was resuspended in 5 mL of LB broth and incubated at 37°C with shaking at 200 r/min. After inoculation with phage, samples were taken every 5 min for the first 30 min and every 10 min for the next 170 min. The titer of phage was measured after centrifugation at 12,000 rpm for 5 min.

To investigate the effect of different relevant environmental conditions on phage infection, phage stability at various conditions should also be assessed. We incubated aliquots of the two phage proliferators with determined titers in a water bath at a range of temperatures (−20°C, 4°C, 30°C, 40°C, 50°C, 60°C, and 70°C) for 1 h. Three parallel controls were set up at each temperature and titers were measured using the double layer agar plate method. In addition, two strains of phage were exposed to sterile petri dishes, which were then placed in an ultra-clean bench for 2 h of UV irradiation, and samples were taken at 10 min intervals during the first 1 h and at 20 min intervals during the second 1 h and titers were measured using the double-layer agar plate method to determine the stability of the phage. Prepare 1 mol/L of NaOH solution and HCl solution and use a burette to configure sterile LB broth with different pH values of 2, 4, 6, 8, 10 and 12, add 1 mL of phage proliferation solution of known potency to it, adjust its pH to 7 using NaOH solution and HCl solution immediately after 1 h of water bath at 37°C and measure the titer using the double-layer agar plate method. The experiment was repeated three times.

Genomic DNA of the two phage strains was extracted by phenol-chloroform method (Zhao F. et al., 2019). We analyzed phage genome sequences using NCBI ORF finder and GeneMarks for prediction and annotation of phage open reading frames (ORFs) (Besemer et al., 2001). The predicted ORFs were identified by comparison with BLASTp in NCBI. Phage tRNAs were predicted using tRNAscan-s (Schattner et al., 2005). Lysogenic genes in the genome were analyzed using the online tool PHASTER. The online tool PhageScope server allows analysis of phage drug resistance genes and virulence genes, and the whole genome of phage was mapped in circles using CGview online software. The phylogenetic tree of the two phage strains was constructed using MEGA 10 (Tamura et al., 2013) based on conserved and meaningful sequences.

To understand the bactericidal activity of the two strains of P. aeruginosa in vitro, PA27 was used as the host bacterium and incubated at 37°C until the logarithmic growth phase of the bacteria, and the concentration of the bacteriophage solution was measured, and the potency of the phage proliferation solution was also determined; the proliferation solution was centrifuged at 8,000 rpm for 5 min, the supernatant was discarded, and new sterile LB broth was added and resuspended three times. In a 96-well bacterial culture plate, 50 μL of the above resuspension solution was added to each well, and then 50 μL of single phage or phage cocktail was added to each well, and the test was performed according to phage OMOI, and the phage cocktail was mixed in a 1:1 volume ratio, and three parallel tests were set up for each group. The control group was spiked with 100 μL of bacterial solution and 100 μL of sterile LB broth and incubated in a 37°C incubator, and OD₆₀₀ was measured every 1 h. The experiment was performed three times. All data were statistically analyzed using GraphPad Prism 8.0 software.

To assess the therapeutic ability of two strains of phages against infection with P. aeruginosa haemorrhagic pneumonia, a mouse model of haemorrhagic pneumonia was constructed and given treatment. To determine the half lethal dose (MLD₅₀) of the host bacterium PA27, five Balb/c mice in each group were intranasally challenged with 50 μL PA27 (3.5 × 10⁹, 3 × 10⁸, 3.5 × 10⁷, 3 × 10⁶ or 3.5 × 10⁵ CFU/mL) in a nasal drip after respiratory anesthesia via isoflurane. As a control, another group of mice was treated with PBS. The number of dead mice was recorded daily during the 7-day follow up period. All of the assays were performed in triplicate. According to the measured median lethal dose in mice, the experimental group was inoculated with 50 μL of P. aeruginosa PA27 (10⁸ CFU/mL). After a 2 h challenge with PA27, mice were respiratory anesthetized with isoflurane and injected intranasally with either YL1 (10⁸ PFU/mL in a volume of 50 μL), YL2 (10⁸ PFU/mL in a volume of 50 μL) or a cocktail (YL1 and YL2,10⁸ PFU/mL in a volume of 50 μL). Each dose group consisted of five Balb/c mice. As a control, one group of infected Balb/c mice was treated with 1 mL of PBS. Each group was treated for 24 h, after which euthanasia was performed. All assays were performed in triplicate.

To monitor the amount of bacteria in the lungs of mice, lung tissue samples were also collected from euthanized mice. The lungs were weighed, suspended in 1 mL of aseptic PBS, and homogenized using a tissue grinder. Viable PA27 cell counts (CFU/g tissue) were determined by serial dilution and coating on LB agar plates. Histopathological analysis was performed on the lungs of PA27-infected mice treated with phage or PBS and on the lungs of uninfected mice. Mice were euthanized 24 h after PA27 challenge. Lung tissues were removed and immediately placed in 4% paraformaldehyde. Paraformaldehyde-fixed tissues were processed and analyzed by microscopy.

The clear plaques produced by the phage YL1or YL2 that infected PA27 strain had well-defined boundaries with a diameter of about 3 mm (Supplementary datasheet 1). TEM images showed that the phage YL1 possessed an isometric polyhedral head of about 70.3 nm in diameter and an uncontracted short tail of approximately 33 nm, and the phage YL2 possessed an same head like YL1 about 65 nm in diameter and an uncontracted short tail of approximately 35 nm (Figures 1A,B). According to the current classification system developed by the International Committee on the Taxonomy of Viruses (ICTV), they belong to Podoviridae family and designated as vB_PaeP_YL1 and vB_PaeP_YL2 (abbreviated of YL1 and YL2). The state of P. aeruginosa PA27 after cracking by phage can be lysed seen under SEM (Figure 1C).

YL1 and YL2 exhibited lytic activity. The results showed that the lytic rate of phage YL1 to different sources of P. aeruginosa strains was 40.6% (67/165), while that of phage YL2 to different sources of P. aeruginosa strains was 32.1% (53/165) (Supplementary datasheet 2, PA12 is the standard P. aeruginosa strain). There was no cross-species cleavage in the two bacteriophages. To broaden the bactericidal spectrum, a bacteriophage cocktail was constructed by considering the optimal MOI of the two phages, as well as the principle of host range complementarity among bacteriophages.

The thermal stability test showed that YL1 was stable below 50°C, but it was completely inactivated at 70°C after 1 h (Figure 2A). In addition, the titer of YL1 could be maintained above 10⁶ PFU/mL at 60°C. The survival rate of YL2 is same like YL1. The stability of YL1 and YL2 under ultraviolet light was studied. The results showed that the titers of YL1 and YL2 had been in a downward trend until 120 min, but the titers of the two bacteriophages dropped to 0, suggesting that they were sensitive to UV irradiation (Figure 2B). YL1 and YL2 could maintain activity over a broad range of pHs (4–10) (Figure 2C), but it was inactivated at extreme pHs (below pH 4 or above pH 10), indicating that YL1 and YL2 has a good acid–base tolerance. Additionally, the one-step growth curve of YL1 revealed that the latent period and lysis period were 10 and 90 min, respectively, and the average burst size was 125 PFU/infected cell. The latent period and lysis period of YL2 were 15 and 80 min, and the average burst size was 94 PFU/infected cell (Figure 2D). The MOI for both YL1 and YL2 was 0.1 (Figure 2E).

As shown by the trend of the curve (Figures 3A,B), the OD₆₀₀ values of the positive control increased continuously within 24 h, and the OD₆₀₀ values of the negative control remained unchanged, the OD₆₀₀ values of the phage cocktail group continued to rise within 1–3 h regardless of whether the MOI was 0.1 or 0.001. Between 3–9 h, the OD₆₀₀ values of each individual phage group began to decline, with the phage cocktail group showing a more significant decline. Between 9–15 h, the OD₆₀₀ values of each phage group tended to stabilize. Between 15–20 h, the OD₆₀₀ values of the YL1 and YL2 groups slowly increased, suggesting that this may be due to a buildup of bacterial fragments in the sample or the emergence of phage-resistant strains, leading to an increase in OD₆₀₀ values. In short, both single phages and the phage cocktail group effectively inhibited the growth of the host bacteria PA27 in vitro, with the phage cocktail group exhibiting a superior inhibitory effect compared to the single phage groups.

Each mouse was injected with 3.5 × 10⁹ CFU/mL of PA27, which was sufficient to induce hemorrhagic pneumonia and resulted in 100% mortality within two days. However, mice injected with 3.5 × 10⁷ CFU/mL of P. aeruginosa PA27 showed a survival rate of 80%. Based on the Reed–Muench formula, the mouse median lethal dose (MLD₅₀) was determined to be 1.5 × 10⁸ CFU/mL (Figure 6A). To determine the protective effect of phage and phage cocktail against hemorrhagic pneumonia in mice, mice were intranasally administered with phage YL1 (10⁸ PFU/mL), phage YL2 (10⁸ PFU/mL), and phage cocktail (10⁸ PFU/mL) 2 h after challenge with P. aeruginosa. Compared to PBS, phage significantly protected mice from P. aeruginosa infection. After 24 h of treatment, the mice were euthanized and the bacterial counts in their lungs were measured. In comparison to the PBS group, the single phage-treated mice exhibited a reduction in bacterial load by 2 to 3 log values, while the phage cocktail-treated group showed a decrease by 4 log values. Notably, no bacterial colonies were detected in the lungs of mice from the normal control group. It was found that the administration of phage significantly reduced the bacterial load in the lungs, with the phage cocktail group showing the most significant reduction (Figure 6B). Phage load assay revealed a slight decrease in the number of individual phage in the lungs, while the phage cocktail showed an increase, confirming the effective infection of host strain PA27 by both individual phage and phage cocktail in mice (Figure 6C).

Evaluate the effect of a single phage and a phage cocktail on the improvement of lung pathology in mice. On the surface, the lung tissue of normal mice appears pink, with a pale pink and smooth surface (Figure 7D). In contrast, the lungs of mice infected with PA27 appear dark red and reach the stage of red hepatoid characteristic of pneumonia (Figure 7E). In comparison, the lung tissue of mice treated with a single phage (YL1: Figure 7A, YL2: Figure 7B) in hemorrhagic pneumonia generally appears pink, with only certain areas showing dark red spots of hemorrhage. The lung tissue of mice treated with the phage cocktail (Figure 7C) in hemorrhagic pneumonia shows a significantly reduced presence of hemorrhage compared to the single phage treatment group. Based on histopathological analysis, normal lung tissue displays thin alveolar septa, with no infiltration of inflammatory cells in the alveolar spaces. In contrast, the lungs of mice with hemorrhagic pneumonia 24 h after infection with PA27 show severe inflammation, dilated capillaries, thickened alveolar septa, and congested alveolar walls, with almost no visible alveolar spaces. In comparison, the pathological changes in the lung tissue of mice treated with a single phage are significantly reduced, and the mice in the phage cocktail protection group exhibit the mildest lung pathology, indicating the most significant protective effect.