The strain used in the present study, Pss (strain number CGMCC 1.3070, GenBank entry number: AF094749), was purchased from the China General Microbiological Culture Collection Center (CGMCC). The strain was cultured in Luria–Bertani (LB) broth at 30° C with shaking at 180 rpm/min. It was preserved in 50% glycerol at −80° C for long-term storage.

Using Pss as the host bacteria, bacteriophages were isolated from sewage using a double-layer agar plate method, as previously described (Kyrkou et al., 2020). Briefly, sewage collected around hospitals in Kunming, Yunnan, was centrifuged at 15,000 × g for 15 min, and the resulting supernatant was filtered through a 0.22-μm microporous filter. A total of 1 mL of the filtrate was added to 5 mL of a log-growing host Pss culture, and the mixture was incubated overnight at 30 °C with shaking at 180 rpm and 4 °C. After centrifugation at 15,000 × g for 15 min, the supernatant was filtered again through a 0.22-μm microporous filter, and a phage stock was obtained. This phage stock solution was serially diluted with LB liquid medium, and 100 μL of different dilutions of this bacteriophage solution was mixed with 300 μL of Pss host cell culture. This mixture was incubated at 20–25 °C for 15 min, mixed with 4.5 mL LB semi-solid medium, and poured onto the surface of pre-prepared LB solid media. The phages were cultured at 30 °C overnight and monitored for plaque formation. Clear single plaques were selected and transferred into a host bacterial suspension [optical density at 600 nm (OD₆₀₀) = 0.5–0.8] in the logarithmic growth stage for phage amplification. The mixture was incubated overnight at 30 °C with shaking at 180 rpm/min. Subsequently, the culture was centrifuged at 4 °C and 15,000 × g for 15 min, and the supernatant was filtered through a 0.22-μm microporous filter. Phage plaques were then assessed using the double-layer agar plate method. This purification step was repeated multiple times until only a single plaque morphology with consistent size appeared on the plates. The obtained filtrate was considered a pure bacteriophage solution, and its titer was subsequently determined using the double-layer agar plate method. The purified bacteriophage was stored in 50% glycerol at −80 °C and was regularly revived to maintain viability.

An optimal MOI represents the ratio of phages to bacteria at the time of infection. To determine the MOI, host bacteria were cultured to the logarithmic growth phase until they reached a colony number of 3 × 10⁷ colony-forming units [CFU]/mL, as previously described (Liu et al., 2024). Phage stock with a known titer and host cells at the exponential stage were mixed with a gradient of MOI ranging from 0.001–100 and cultured in a shaker at 180 rpm/min at 30 °C for 12 h. The phage titer was then determined using the double-layer agar plate method after centrifugation (15,000 × g for 15 min at 4° C), followed by filtration through a 0.22-μm membrane. The experiment was repeated three times, and the average value was obtained, with the highest value representing the optimal MOI.

The parameters of the one-step growth curve were determined based on a previously described method (Liu et al., 2021). Briefly, phages were added to a log-phase culture of Pss (3 × 10⁷ CFU/mL) to achieve an MOI of 1, and the mixture was incubated for 15 min at room temperature. After centrifugation at 15,000 × g for 5 min at 4° C, the supernatant was discarded. The resulting phage-infected bacteria pellet was washed three times with LB medium at 30° C. The final pellet was resuspended in 30 mL of LB medium and incubated at 30° C with shaking at 180 rpm/min. Samples (600 μL) were collected every 15 min, centrifuged, and filtered. The phage titer was then determined using the double-layer agar plate method. Control groups without phages or bacteria were included for comparison. The experiment was repeated three times, and the results were plotted to determine the phage latent period, lysis period, and steady phase.

To assess the effects of temperature and pH on the survival of the PSV6 and PSV3 phages, 10 μL of PSV6 and PSV3 lysates (1–3 × 10⁹ plaque-forming units [PFU]/mL) were added to 990 μL of various aseptic buffers, including citrate, phosphate, Tris-HCl, glycine, carbonate, and sodium hydrogen phosphate buffers, with pH levels ranging from 3–12 and incubated at 30 °C for 2 h. After dilution, the phage titers were determined using the double-layer agar plate method. For the thermal stability experiments, 4 mL of phage stock solution was incubated at varying temperatures (4, 30, 40, 50, 60, and 70 °C) for 180 min. Samples were collected every 30 min, and phage titers were determined. All experiments were performed in triplicate.

The effect of organic solvents on phage survival was determined using a previously reported method (Pallavi et al., 2021) with some modifications. Briefly, 900 μL of PSV6 or PSV3 (1–3 × 10⁹ PFU/mL) was mixed with ether, chloroform, acetone, or ethanol. The mixture was incubated for 30 min, followed by centrifugation at 15,000 × g for 10 min. The supernatant was then diluted, and the phage titers were measured using the double-layer agar method, with the LB medium used as a control. The experiment was repeated three times.

The host range of the Pss phages PSV6 and PSV3 was assessed using the spot assay method. Bacterial strains included P. syringae ATCC 19304, P. aeruginosa ATCC 10145, Bacillus cereus ATCC 14579 and GDMCC 1.3589, Serratia marcescens ATCC 13880, and Escherichia coli ATCC 11303, which were standard strains purchased from the Strain Preservation Center. Additionally, Serratia marcescens KMR-3 was an isolated strain. For each assay, 200 μL of bacterial cells were mixed with 4.5 mL of a semi-solid medium and transferred onto the surface of a double-layer agar plate. A total of 10 μL of the phage suspension (10⁹ PFU/mL) was spotted onto the double-layer agar surface of different bacterial lawns. The plates were incubated at 28° C overnight, and plaque formation within the spotted areas was observed. The experiment was repeated three times.

Phage genomic DNA was extracted using an OMEGA-Viral DNA kit 50 (Guangzhou Feiyang Biological Engineering Co., Ltd, China) following the manufacturer's instructions. The extracted phage genomes were treated with DNase I and RNase A to determine the type of nucleic acid present. To verify whether the genomes of PSV6 and PSV3 are circular or linear, three pairs of reverse primers for PSV6 and two pairs of reverse primers for PSV3 were designed, along with one pair of forward primers as a positive control. The sequences of primers are listed in Table 1. PCR amplification was performed using the phage DNA as the template. The presence of bands indicated circular DNA, whereas their absence suggested linear DNA.

The extracted PSV6 and PSV3 genomic DNA were subjected to next-generation sequencing by Shenggong Bioengineering (Shanghai) Co., Ltd. using an Illumina Novaseq6000 sequencing platform (Illumina Inc., San Diego, CA, USA). Raw data were evaluated using FastQC, and the sequences were trimmed using Trimmomatic to obtain high-quality and reliable data. After quality control, genome assembly was performed using SPAdes software (https://github.com/ablab/spades). Subsequent bioinformatic analyses were conducted on the assembled sequences to obtain and characterize the complete the whole genome of PSV6 and PSV3. Whole-genome was mapped using GCView (https://cgview.ca/) (Grin and Linke, 2011), and a whole-genome phylogenetic tree was constructed using the VICTOR database (https:genomes//ggdc.dsmz.de/victor.php#). The phylogenetic trees of the terminase and major capsid proteins of PSV6 and PSV3 were generated using the neighbor-joining method in MEGA-X software (Kumar et al., 2018). The VFDB (http://www.mgc.ac.cn/VFs/) virulence factor database and antibiotic resistance gene database (https://card.mcmaster.ca/) were screened for genes associated with virulence factors and antibiotic resistance genes (Chen et al., 2016).

To assess the effects of PSV6 and PSV3 on Pss and generate lysis curve experiments, a 1–2% inoculum of Pss was cultured in 50 mL LB liquid medium at 30 °C with shaking at 180 rpm/min until it reached the logarithmic growth phase. PSV6 and PSV3 phages were then added to the culture at MOIs of 0.001, 0.01, 0.1, 1, 10, and 100, and incubation continued under the same conditions. A total of 200 μL of culture was sampled hourly, and the OD₆₀₀ was measured using a microplate reader. The control group included a host bacterial suspension without phages or a phage suspension without host bacteria. Three replicates were performed for each group, and the experiment was repeated three times.

To determine whether Pss was pathogenic to Arabidopsis thaliana, a pathogenicity model was established.

The main challenge in the biological control of plant bacterial diseases lies in demonstrating the efficacy of bacteriophages for plant control or treatment. In this study, A. thaliana was used as a plant model to evaluate the biocontrol efficacy of PSV6 and PSV3. A positive control was prepared using 15 mL of 10 mmol/L MgCl₂ buffer, and a negative control was established using 15 mL of the Pss suspension. Based on the in vitro experimental results, the three highest MOI values (1, 10, and 100) were selected for the experimental groups. Mixtures of Pss bacterial suspension and either single phages or a phage mixture (prepared by mixing PSV3 and PSV6 at equal titers in a 1:1 ratio) were prepared at a final volume of 15 mL, based on the different MOI values, and incubated for 15 min. The mixtures from each group were then poured onto 3 week old Arabidopsis seedlings for 1 min. The mixtures were then removed, and the seedlings were transferred to a sterile plant growth chamber for cultivation, with six replicates per group. Disease symptoms on the leaves were observed and recorded every 24 h. For each treatment, three randomly selected leaf discs (with diameters of 0.6 cm) were placed in 2 mL centrifuge tubes containing 1 mL of 10 mmol/L MgCl₂ buffer. The leaf discs were ground in a sterile mortar, and 300 μL of the homogenate was serially diluted in 10 mmol/L MgCl₂ buffer to concentrations of 10⁻¹, 10⁻², 10⁻³, 10⁻⁴, and 10⁻⁵. Subsequently, 100 μL of each dilution was spread on LB agar plates and incubated at 30° C for 24 h to count the colony-forming units (CFUs). This experiment was repeated three times.

Statistical analyses were performed using GraphPad Prism software (GraphPad Inc., La Jolla, CA, USA). One-way analysis of variance (ANOVA) was used to evaluate significant differences in phage titer under different MOIs and organic solvents and determine significant differences in biocontrol efficacy under different experimental conditions. P < 0.05 was considered statistically significant, and three independent replicates were performed for each condition.

To identify phages that can control Pss infections, we used Pss as a host for two strongly lytic phages isolated from wastewater, PSV6 and PSV3. Using the double-layered plate, we observed clear plaques on the lawn of Pss. The plaques formed by PSV3 were significantly smaller than those of PSV6, with diameters of 0.5–1 mm and 1–3 mm, respectively. Both plaques were transparent without any halos (Figures 1A, C). TEM revealed that the morphologies of the two phages differed. Specifically, the head of the PSV6 phage was icosahedral with a diameter of 54 ± 1 nm, and it had a short, non-contractile tail with a diameter of 22 ± 1 nm (Figure 1B). In contrast, the head of PSV3 was polyhedral with a diameter of 45 ± 1 nm, and it had a long, non-contractile tail with a diameter of 186 ± 1 nm, where the tail fibers were clearly visible (Figure 1D).

When the MOI of PSV6 and PSV3 was 1, the number of progeny phages released by Pss infected by phages was the highest (1.74 × 10¹⁰ PFU/mL and 7.40 × 10⁹ PFU/mL, respectively), indicating that the optimal infection complex numbers of both PSV6 and PSV3 were 1 (Figures 2A, 3A). The incubation period, ascent period, and burst size are key parameters used to characterize the life cycle of phages (Sillankorva et al., 2008). Our one-step growth curve demonstrated that these three values for PSV6 were 40 min, 40 min, and 135 PFU/cell (Figure 2B), whereas they were 45 min, 75 min, and 16 PFU/cell for PSV3, respectively (Figure 3B). The pH stability tests showed that PSV6 remained stable within the pH range of 6–9 (Figure 2C), while PSV3 was stable between pH 5 and 10 (Figure 3C); however, both phages were inactivated at extreme pH values. PSV3 exhibited greater pH stability than PSV6. In terms of temperature stability, PSV6 maintained its titer between 4–30° C (Figure 2D), while PSV3 was stable over a broader range of 4–60° C (Figure 3D). Both phages showed a gradual decline in titer with increasing temperature, with PSV3 demonstrating better thermal stability than PSV6. Moreover, the sensitivities of PSV6 and PSV3 to organic solvents slightly differed. Specifically, PSV6 was resistant to acetone but was sensitive to chloroform, ether, and ethanol (Figure 2E). PSV3 was resistant to chloroform and ether but sensitive to acetone and ethanol (Figure 3E). Among the seven different bacterial strains tested, only P. syringae ATCC 19304 was susceptible to PSV6 and PSV3, whereas the others were not infected by these phages (Table 2).

The genomic nucleic acids of both phages could only be digested by DNase I but not by RNase A (Figure 4A). Moreover, the PSV6 reverse primer successfully amplified a band using genomic DNA as the template, whereas the PSV3 reverse primer failed to produce any amplification (Figure 4B). Therefore, the nucleic acid type of PSV6 is circular DNA, while that of PSV3 is linear DNA.

To better understand phages and their applicability for biological control, PSV6 and PSV3 were subjected to whole-genome sequencing analysis. Our TEM and whole-genome phylogenetic tree analyses indicated that phages PSV6 and PSV3 both belong to the class Caudoviricetes, with PSV6 classified under Autotranscriptaviridae and PSV3 under Jondennisvirinae. The PSV6 genome (GenBank entry number: OP485124) has a length of 40,070 bps and 57.96% G + C content (Figure 5A). It also contains 48 open reading frames (ORFs), among which 31 are known functional coding sequences, and 17 are hypothetical proteins. The PSV3 genome (GenBank entry number: OP712460) has a total length of 2,29,871 bps and 49.08% G+C content (Figure 5B). It also contains 376 ORFs, among which 34 are known functional coding sequences and 341 are hypothetical proteins. Blast comparison analysis revealed that PSV6 is closely related to phages UNO-SLW1 to UNO-SLW4, PCS5, and PCS4, belonging to the same genus as them but not the same species (Figure 6A). PSV3 is located on the same branch as phages PSN9, Kakheyi25, PaeS C1, PaeS SCH Ab26, and PA73, sharing a common evolutionary origin and belonging to the same genus but not the same species (Figure 7A). The large subunit of the phage terminal enzyme and the major capsid protein are important for phage evolution. Therefore, we constructed evolutionary trees of the PSV3 and PSV6 terminal enzyme large subunits and major capsid proteins. The results indicated that the terminal large subunit of PSV6 exhibits high homology with UNO-SLW1, PPpW-4, 22PfluR64PP, PFP1, phiIBB-PF7A, Pf-10, and Phi-S1 in the existing NCBI database (Figure 6B). The main capsid proteins demonstrated high homology with PCS4, UNO-SLW1, 22PfluR64PP, PFP1, 71PfluR64PP, BIM BV-46, and Pf-10 in the database (Figure 6C). The terminal large subunit of PSV3 is closely related to the bacteriophages vB_SmaS-DLP_1 and BUCT-PX-5 (Figure 7B). The main capsids are closely related to the bacteriophages vB_PaeS_SCH_Ab26 and vB_PaeS_SCUT-S3 (Figure 7C) and belong to the same evolutionary clade. Moreover, our analysis of virulence factor and antibiotic resistance databases revealed that the PSV6 and PSV3 genomes did not encode virulence factors or antibiotic resistance-related genes; therefore, they could be used for subsequent biological control research.

To investigate the growth inhibitory effect of PSV6 and PSV3 on the host bacteria Pss, we determined the absorbance values of PSV6 and PSV3 at OD₆₀₀ under different infection conditions. The two tested bacteriophages effectively reduced the OD value of the bacteria. At high titers (MOIs of 10 and 100), PSV6 and PSV3 decreased the OD value within a short period. However, the time required for bacterial OD was also short, with durations of 3 and 10 h for PSV6 and PSV3, respectively. At low titers, the OD value decreased slowly. At MOI 1 and MOI 0.1, PSV6 started to decrease the bacterial OD value from 2 h, whereas at an MOI 0.001, the bacterial OD value started decreasing at 3 h. However, at MOI 1, PSV6 notably resulted in a more rapid and pronounced decrease in bacterial OD (Figure 8A). At an MOI of 1, PSV3 notably decreased the bacterial OD value after 5 h, whereas at MOIs of 0.001 and 0.1, the bacterial OD value decreased after 11 h and 9 h, respectively (Figure 8B). Additionally, the OD value increased at low phage concentrations.

To evaluate the in vivo efficacy of PSV6 and PSV3, we developed a plant infection model using Pss and A. thaliana (Figure 9). Analysis of the effect of phage treatments at different MOIs for 15 min revealed that treatment with PSV6 at an MOI of 1 significantly decreased the number of Pss cells at 0–96 h of treatment, whereas PSV3 only significantly decreased the number of Pss cells at 24 h. Notably, longer sampling times exerted no significant effect on the number of Pss cells. The combination treatment with PSV6 and PSV3 resulted in a higher bacterial abundance at 24 h than that in the Pss group (negative control), and the number of bacteria was also significantly reduced at the remaining time points (Supplementary Figure 1A). We also observed leaf symptoms. Specifically, disease severity was lower in the PSV6 treatment group than in the PSV3 alone and combined phage treatment groups. The PSV3 alone and combined phage treatment groups exhibited symptom levels comparable to those in the Pss group from 48 to 72 h (Supplementary Figure 1C). At an MOI of 10, the three treatment groups exhibited significantly reduced numbers of Pss cells from 0 to 96 h; this effect was more significant than that observed at MOI 1. Moreover, PSV3 reduced the number of Pss more than PSV6 or the combination treatment (Supplementary Figure 1B). After 15 min of treatment at an MOI of 10, fewer A. thaliana leaves turned yellow at 96 h in the PSV3 treatment group than in the Pss group. From 48 to 96 h of treatment, the disease symptoms were more severe in the PSV6 and mixed treatment groups than in the PSV3 treatment group, but less severe than those in the Pss group (Supplementary Figure 1C). After treatment with phages at an MOI of 100 for 15 min, plants in the PSV6, PSV3, and mixed treatment groups demonstrated significantly reduced numbers of Pss cells compared to those treated at an MOI of 1 and 10. This finding indicates that an increase in phage concentration resulted in a more pronounced inhibitory effect on Pss in vivo. Moreover, PSV3 reduced Pss abundance more than PSV6 and the mixed phage treatment (Figure 10A). Hence, leaf disease observations indicated that at an MOI of 100, the severity of leaf symptoms followed the order of PSV6 > mixed treatment > PSV3. Moreover, the leaf disease symptoms in each group at an MOI of 100 were less severe than those observed at MOI 1 and MOI 10 and those in the negative control group (Figure 10B).