Isolation and identification of a newly discovered broad-spectrum Acinetobacter baumannii phage and therapeutic validation against pan-resistant Acinetobacter baumannii

Abstract

The treatment of Acinetobacter baumannii (A. baumannii) poses significant clinical challenges due to its multidrug/pan-drug resistance. In this study, we isolated a broad-spectrum lytic A. baumannii phage, named P425, from medical wastewater, targeting nine multidrug-resistant A. baumannii (MDRAB) with diverse capsular types. Biological characterization revealed that P425 maintains activity at pH range of 3–12 and temperature range of 4–50 ​°C. It resists UV irradiation for 20 ​minutes, and had an optimal multiplicity of infection (OMOI) is 0.00001. The adsorption kinetics showed that P425 achieves > 90% within 10 ​minutes of incubation, and the one-step growth curve indicated a 10-min latent period, with a burst size of 184 ​PFU/cell. The genome sequencing results indicated that it harbors a double-stranded DNA genome of 40,583 bp with a GC content of 39.39%. Intergenomic similarity analysis classified it as a novel species within the Friunavirus genus, while electron microscopy results showed that it belongs to the Podoviridae family. Notably, P425 exhibits potent 24-h in vitro inhibitory activity against MDRAB, and demonstrates synergistic effect at an MOI of 0.001 when combined with five classes of antibiotics targeting distinct antimicrobial mechanisms. Safety evaluations confirmed the absence of cytotoxicity, hemolytic activity, or systemic toxicity both in vitro and in vivo. In mouse infection models, P425 can significantly improve the survival rates of mice infected with Ab25 (ST1791/KL101). When co-administered with levofloxacin, it achieved 100% protection against mortality and promoted immune recovery. Collectively, P425 is a prospective lytic phage that could offer novel strategies for combating MDRAB infections.

Highlights

• •An Acinetobacter baumannii (A. baumannii) phage named P425 was isolated and characterized.
• •P425 is a broad-spectrum lytic phage exhibiting potent inhibitory activity against multidrug-resistant A. baumannii.
• •P425 synergizes with five antibiotics of distinct mechanisms to enhance antibacterial effects.
• •P425 exhibits potent therapeutic effect and provides preventive protection against A. baumannii infections in mice.

Introduction

Acinetobacter baumannii (A. baumannii) is a major opportunistic pathogen and poses one of the greatest treatment challenges among hospital-acquired infections globally. It is responsible for a wide spectrum of serious conditions, including bacteremia, skin and soft-tissue infections, wound infections, urinary tract infections, meningitis, endocarditis, and pneumonia (Ibrahim et al., 2021). The management of A. baumannii infections is particularly difficult due to the bacterium's potent virulence factors, such as outer membrane proteins, lipopolysaccharides, and phospholipases (Dehbanipour and Ghalavand, 2022; Maure et al., 2023; Müller et al., 2023).

Bacteriophages, also known as phages, are viruses that infect and replicate within bacteria, and exhibit high host specificity, typically targeting single species or specific strains (Kasman LM, 2022; Lai et al., 2020; Salmond and Fineran, 2015). Phages have been employed for therapeutic purposes since the early 1900s, with their golden era spanning the 1920s–1940s. However, interest in phage therapy waned during the rise of antibiotics (Gordillo Altamirano and Barr, 2019; Summers, 2024). In the current post-antibiotic era, phage therapy has regained attention as a promising alternative to combat antimicrobial resistance (Santini, 2024; Taati Moghadam et al., 2020; Tamma and Suh, 2021).

Despite their potential, the effectiveness of phages can be limited by the emergence of phage-resistant bacterial strains, analogous to the development of antibiotic resistance. However, recent studies have shown that phage resistance may increase bacterial susceptibility to antibiotics, and that antibiotics may enhance phage-mediated bacterial eradication (Luong et al., 2020). For instance, Kirby (2012) demonstrated that Staphylococcus aureus (S. aureus) phages act synergistically with gentamicin; antibiotic-induced bacterial aggregates increased survival during antibiotic exposure but also rendered the bacteria more susceptible to phage attack, likely due to upregulation of phage receptors. Luo wt al. (2024) reported synergistic effects between phage pB23 and meropenem in both a pig skin explant model and a zebrafish infection model, successfully treating carbapenem-resistant A. baumannii infections. Choi et al. (2024) showed that the combination of phage vB_AbaSi_W9 and rifampicin improved mouse survival rates in infected mice to 100%. Similarly, the combined use of phages (AbKT21phi3 and KpKT21phi1) and antibiotics (mucomycin and meropenem) effectively eradicated multidrug-resistant Klebsiella pneumoniae and multidrug-resistant A. baumannii (MDRAB), accelerating wound healing in clinical cases (Grygorcewicz et al., 2020; Tu et al., 2023).

In this study, we isolated a lytic A. baumannii phage, P425, and evaluated its biological and genomic characteristics, in vitro bacteriostatic ability, biofilm-disrupting capability, synergistic interactions with antibiotics, and therapeutic efficacy in a mouse model of bacteremia. These findings support the potential of P425 as a promising therapeutic phage, offering innovative strategies to combat MDRAB infections.

Results

Isolation and biological characterization of P425

P425 is a lytic bacteriophage that formed clear plaques with a surrounding translucent halo that expanded over time when co-cultured with its host bacteria on double-layer agar plates at 37 ​°C (Fig. 1A). Transmission electron microscopy classified P425 within the Podoviridae family (Valencia-Toxqui and Ramsey, 2024), based on its morphological features (Fig. 1B). Host spectrum analysis revealed that P425 possessed broad-spectrum lytic activity, lysing 79.59% (39/49) of tested A. baumannii clinical isolates and targeting nine different capsular types (K3, K7, K9, K14, KL72, KL77, KL101, KL104, and KL160) (Table 1), indicating its potential for broad therapeutic application.

Fig. 1.

The morphological features and biological characterizations of phage P425. A The plaque morphological feature of phage P425. B The morphological feature of P425 was observed by transmission electron microscopy. The tail structure is marked by the red arrow. C pH stability of P425. Phages at an initial concentration of 2 ​× ​10⁸ ​PFU/mL were incubated at 37 ​°C for one ​hour at the pH range of 1–12, phage titers were determined using the double-layer agar method. D Temperature stability of P425. Phages at an initial concentration of 2 ​× ​10⁸ ​PFU/mL were incubated at different temperature (4 ​°C, 25 ​°C, 37 ​°C, 50 ​°C, 60 ​°C, and 70 ​°C) for one ​hour, phage titers were determined using the double-layer agar method. E UV tolerance at an initial concentration of 2 ​× ​10⁸ ​PFU/mL of P425. Phages were exposed to UV light for 10 ​minutes and sampled at 10-min intervals for a total of 60 ​minutes. Phage titers were determined using the double-layer agar method. Viability (%) = (Post-treatment titer/Initial titer) ​× ​100. F Optimal MOI assay. Phage P425 and host bacterium Ab25 were co-cultured under different MOIs, and phage titers were determined after eight ​hours of infection. G Adsorption rate assay. Phage P425 was co-cultured with host bacterium Ab25 and the supernatant was collected at different time intervals to ascertain free phage titer. Adsorption (%) ​= ​[1 - (Free phage titer/Initial phage titer)] ​× ​100. H One-step growth curve assay. Phage P425 was co-cultured with host bacterium Ab25 at an MOI of 0.1, and the phage titers were determined at the indicated time points after infection. The experiment was repeated three times. Data are presented as mean ​± ​SD. Statistical analysis was performed using one-way ANOVA. ∗∗∗, P ​< ​0.001.

Table 1.

A. baumannii isolates used to examine the host range of phage.

A. baumannii	Sample	ST type	K type	Phage sensitivity
Ab01	Sputum	ST136	KL77	+
Ab02	Sputum	ST540	KL160	–
Ab03	Lavage fluid	ST195	K3	+
Ab04	Sputum	ST195	K3	+
Ab05	Sputum	ST1791	KL101	+
Ab07	Sputum	ST2499	KL104	+
Ab08	Sputum	ST191	KL72	+
Ab09	Sputum	ST191	KL72	+
Ab10	Sputum	ST208	K2	–
Ab11	Sputum	ST369	K9	+
Ab12	Sputum	ST1791	KL101	+
Ab13	Hydrothorax and abdominal fluid	ST191	KL72	+
Ab14	Sputum	ST938	KL210	–
Ab15	Lavage fluid	ST195	K3	–
Ab16	Sputum	ST208	K7	+
Ab17	Sputum	ST195	K3	–
Ab18	Sputum	ST540	KL160	–
Ab19	Sputum	ST195	K3	+
Ab21	Sputum	ST195	K3	+
Ab22	Sputum	ST191	KL72	+
Ab23	Sputum	ST1791	KL101	+
Ab24	Lavage fluid	ST195	K3	+
Ab25	Sputum	ST1791	KL101	+
Ab26	Sputum	ST195	K3	+
Ab27	Other	ST195	K3	+
Ab28	Sputum	ST1968	K14	+
Ab29	Lavage fluid	ST195	K3	+
Ab32	Sputum	ST208	K7	–
Ab33	Sputum	ST195	K3	+
Ab40	Conduits	ST1968	K14	+
Ab41	Sputum	ST136	KL77	+
Ab51	Blood	ST1791	KL101	+
Ab52	Lavage fluid	ST2499	KL104	+
Ab53	Sputum	ST1791	KL101	+
Ab55	Sputum	ST191	KL72	+
Ab56	Sputum	ST191	KL72	+
Ab57	Sputum	ST1791	KL101	+
Ab59	Lavage fluid	STnew1	KL101	+
Ab63	Lavage fluid	ST195	K3	+
Ab71	Pus	ST136	KL77	+
Ab108	Sputum	ST1968	K14	+
Ab111	Lavage fluid	ST938	KL210	–
Ab116	Lavage fluid	ST540	KL160	+
Ab118	Other	ST208	K7	–
Ab119	Other	ST208	K7	–
Ab122	Lavage fluid	ST2499	KL104	+
Ab123	Lavage fluid	ST2499	KL104	+
Ab125	Pus	ST136	KL77	+
Ab126	Sputum	ST2499	KL104	+

The samples were collected from the Second Hospital of Nanjing.

The pH stability test showed that phage activity remained above 70% at pH 5–11, with optimal pH at 8 and 9. The phage remained stable at temperatures up to 50 ​°C but was fully inactivated at 60 ​°C. After ten ​minutes (mins) of UV irradiation, only 10.7% of phage activity remained. The optimal multiplicity of infection (OMOI) was determined to be 0.00001. Adsorption assays indicated that more than 90% of phage particles attached to host cells within 10 ​mins. A one-step growth curve analysis showed a 10-min latent period and a burst size of 184 ​PFU/cell, with the lytic cycle reaching a plateau at 70 ​mins post-infection (Fig. 1C–H).

Genomic analysis of P425

The genome of phage P425 consists of a linear double-stranded DNA molecule comprising 40,583 bp, with a GC content of 39.39%. Genome annotation using Prokka identified 47 coding sequences (CDSs), of which 41 encode proteins with known functions (e.g. endolysin and phage holin), while 6 remain hypothetical with unknown functions (Fig. 2A). Notably, the tail fibronectin protein (CDS40) was predicted to encode a depolymerase with 99.2% confidence. The phage with the highest homology to P425 was MRABP9 (OP727261.1), sharing 91% coverage and 98.95% homology. Blast analysis of the four phages with the highest homology showed distinct gene organizations, suggesting the presence of genetic recombination (Fig. 2B). Intergenomic similarity analysis using VIRIDIC among 50 related A. baumannii phages showed that P425 shared the highest nucleotide identity (89.3%) with SH-Ab15519 (NC_041905.1) and the lowest (71.4%) with AB3 (NC_021337.1) (Supplementary Fig. S1). According to established taxonomy thresholds, phages are classified as the same species if they share over 95% nucleotide identity across the entire genome, while genus-level boundaries are typically set at 70% identity (Simmonds et al., 2023; Turner et al., 2021; Valencia-Toxqui and Ramsey, 2024). Based on this analysis, P425 represents a novel species within the Friunavirus genus. Phylogenetic trees based on the large terminase subunit and endolysin proteins placed P425 in a distinct branch, while analyses based on capsid protein and whole genome sequences indicated a closer evolutionary relationship between P425 and MRABP9 (Supplementary Fig. S2). Importantly, no virulence or antibiotic resistance genes were identified in the genome, supporting the safety and therapeutic potential of P425 as a lytic phage.

Fig. 2.

The genomic analysis of phage P425. A Genomic circle plot of phage P425 was plotted using CGView. GC content, GC skew, annotated proteins of known function, and putative proteins are marked. B Comparative genomic analysis of bacteriophage P425 with vB_ApiP_P2, vB_AbaP_APK87 W, MRABP9, and VB_Ab-P-7. Comparative phage genomics visualized with Easyfig.

In vitro bacterial inhibition and biofilm inhibition/removal by P425

Twenty-four-hour in vitro inhibition assays demonstrated that the antibacterial activity of phage P425 against the host strain Ab25 increased with higher MOIs. At MOIs ranging from 100 to 0.01, P425 effectively inhibited bacterial growth for 24 ​hours, although approximately 1 ​× ​10⁶ ​CFU/mL of viable bacteria remained. To assess resistance development, residual bacteria were streaked, and 30 individual colonies were isolated and tested for sensitivity to P425. Most isolates remained susceptible; however, nine exhibited resistance, suggesting that resistant strains can emerge even under high phage pressure (Supplementary Fig. S3). At lower MOIs (0.001–0.00001), P425 suppressed bacterial growth for approximately nine hours before resistant strains began to proliferate (Fig. 3A).

Fig. 3.

In vitro inhibitory activity and effect on biofilm of phage P425. A The in vitro inhibitory effect of phage P425 at different MOIs within 24 ​hours. Bacterial Ab25 in the logarithmic phase (1 ​× ​10⁸ ​CFU/mL) were incubated with various MOIs of phage P425 at 37 ​°C for 24 ​hours. The absorbance of culture at different time points was determined at 600 ​nm. B The inhibition of biofilm by phage P425. The inhibitory effect of phage P425 on biofilm was determined at different concentrations, and the biofilm was quantified by crystal violet staining. Phage P425 was diluted with liquid LB medium, and LB was used as a negative control, and the bacterial solution alone was used as a positive control. C The removal effect of phage P425 on biofilm. The Ab25 was incubated in 37 ​°C incubator for 24 ​hours. After the formation of biofilm, the biofilm was treated with different concentrations of P425 for 24 ​hours. The biofilm was quantitatively analyzed by crystal violet staining, and the control group was the same as above. The experiment was repeated three times. Data are presented as mean ​± ​SD. Statistical analysis was performed using one-way ANOVA. ∗∗∗∗, P ​< ​0.0001; ns: no significant.

Regarding biofilm inhibition, P425 demonstrated over 80% inhibition at concentrations ranging from 10¹⁰ to 10⁶ ​PFU/mL, with no further improvement at higher doses. Even at lower concentrations (10⁵–10³ ​PFU/mL), the inhibition rates remained above 60% (Fig. 3B). In biofilm removal assays, P425 exhibited over 80% removal efficiency at concentrations of 10¹⁰ to 10³ ​PFU/mL, but this effect was not dose-dependent, with no statistically significant differences observed between concentrations (Fig. 3C).

Synergistic activity of P425 with antibiotics

Combination therapy experiments revealed that P425 exhibited synergistic effects with all eight tested antibiotics. Co-administration of P425 reduced the minimum inhibitory concentrations (MICs) of ceftriaxone sodium to 1/2, meropenem to 1/4, and tigecycline, tetracycline, ciprofloxacin, polymyxin B, and sulfamethoxazole to 1/8 of their original MICs. Notably, the MIC of levofloxacin was reduced to 1/16 of its original value when combined with P425 (Fig. 4). At concentrations above these reduced MIC breakpoints, although the combination did not completely inhibit bacterial growth, the growth curves were consistently lower than those with either agent alone (Supplementary Fig. S4). This suggests that the synergistic effect of bacteriophages and antibiotics may enhance the antibacterial effect and potentially reduce the required drug dosage.

Fig. 4.

In vitro synergistic interactions of phage P425 with key antibiotic concentrations. The 24-h co-inhibitory effect of phage P425 (MOI ​= ​0.001) with different concentrations of antibiotics was determined by checkerboard method. Antibiotic MIC determination group, antibiotic-treated group, phage-treated group, bacterial solution alone group (growth control) and LB group were set as the control. The experiment was repeated three times. Data are presented as mean ​± ​SD. The full concentration-range data are provided in Supplementary Fig. S4, which excludes the key concentrations shown here for clarity.

In vitro and in vivo safety of P425

Cytotoxicity analysis using the CCK-8 assay showed that P425 at concentrations from 10¹⁰ to 10³ ​PFU/mL had no adverse effects on cell viability (Fig. 5A). The in vitro hemolysis assay confirmed that P425 did not cause hemolysis at concentrations below 0.25 ​× ​10¹⁰ ​PFU/mL (Fig. 5B). In addition, we also verified the safety of phage P425 in vivo. To assess in vivo safety, mice were administered 1 ​× ​10¹⁰ ​PFU of P425 daily via intraperitoneal or tail vein injection for seven days. No abnormalities were observed, in contrast to in vitro hemolysis trials——potentially due to the systemic dilution of the phage. Hematological analysis conducted 168 ​hours post-injection revealed no statistically significant differences in white blood cell or neutrophil counts between phage-treated and control groups (Fig. 5C and D). Additionally, daily clinical scoring over the seven-day period showed no abnormalities (Fig. 5E), indicating that P425 was safe in vitro and in vivo.

Fig. 5.

In vivo and in vitro safety assessment of phage P425. A Phage toxicity assay on LO2 cells. Cytotoxicity against LO2 human liver cells seeded at 1 ​× ​10⁴ ​cells per well in RPMI-1640 medium supplemented with 20% fetal bovine serum was evaluated using CCK-8 assays. Cells were treated with phage P425 across a concentration range of 10³ to 10¹⁰ ​PFU/mL for 24 ​hours, using PBS as negative control and 2% Triton X-100 as positive control; viability was quantified by measuring optical density at 450 ​nm after 2-h CCK-8 incubation. B Phage hemolysis assay on mouse erythrocytes. Murine erythrocytes isolated from Kunming mice blood via centrifugation (1000×g, 10 ​min, 4 ​°C) were exposed to endotoxin-removed phage at serial two-fold dilutions in PBS for 3 ​hours at 37 ​°C. Test groups contained 800 ​μL phage solution mixed with 200 ​μL erythrocyte suspension; controls received 800 ​μL of 0.1% Triton X-100 (positive) or PBS (negative) with 200 ​μL erythrocytes. Post-incubation centrifugation at equivalent parameters enabled hemoglobin release quantification at 540 ​nm. Hemolysis percentage was calculated as: [(OD _sample – OD _PBS)/(OD _{0.1% Triton X-100} – OD _PBS)] ​× ​100. C–E Mice were injected with P425 (1 ​× ​10¹⁰ ​PFU) continuously for seven days via intraperitoneal cavity (IP) or tail vein (IV). The changes in white blood cells (C), neutrophils (D), and clinical scores (E) of the mice were monitored. Data are presented as mean ​± ​SD.

P425 effectively treats a murine bacteremia model

To evaluate the therapeutic efficacy of phage P425 against A. baumannii, we established the Ab25 (ST1791/KL101)-infection model in mice. Intraperitoneal injection of mice with 1 ​× ​10⁸ ​CFU of Ab25 (ST1791/KL101) led to complete mortality within 24 ​hours, whereas a dose of 1 ​× ​10⁷ ​CFU per mouse resulted in a 40% mortality rate (Supplementary Fig. S5A). Autopsy of the deceased and PBS-treated control mice revealed significantly higher bacterial loads in the liver, spleen, lungs, kidneys, and blood of infected animals (Supplementary Fig. S5B). Clinical scores were recorded at 2-h intervals across three experimental groups. At the 1 ​× ​10⁷ ​CFU dose, some mice displayed transient symptoms but gradually recovered and survived (Supplementary Fig. S5C). Based on these findings, a challenge dose of 1 ​× ​10⁸ ​CFU per mouse was selected for subsequent experiments.

Treatment with phage P425 at varying MOIs significantly improved survival rates in mice infected with Ab25 (ST1791/KL101). Specifically, an MOI of 1 resulted in a 100% survival rate, while MOIs of 0.001 and 0.00001 achieved 80% survival (Fig. 6A). At the 10-h checkpoint, organs (liver, spleen, lungs, and kidneys) and blood were collected for colony and plaques quantification. Bacterial load was significantly reduced in all treated groups compared to the infected group, with the most significant reduction observed in blood samples. The colony count in blood decreased by approximately 7 logs at MOIs of 1 and 0.001, and by approximately 6.6 logs at the MOI of 0.00001 (Fig. 6B). Phages were detected in all analyzed tissues, confirming systemic distribution (Fig. 6C). Clinical scores indicated that the mice in the treatment groups returned to normal activity levels within 72–96 ​hours (Fig. 6D), although the body weight remained lower than the control group (Fig. 6E). In conclusion, phage P425 may effectively tackle the bacteremia caused by Ab25 (ST1791/KL101) infection in mice, significantly reducing bacterial burden and restoring physiological function.

Fig. 6.

Therapeutic efficacy evaluation of phage P425 with different MOIs in mice. Female Kunming mice aged 4–5 weeks were randomized into five experimental groups (n ​= ​8/group): negative control (PBS injection), positive control (bacterial infection only), and therapeutic groups receiving single-dose intraperitoneal administration of endotoxin-removed phage P425 at MOIs of 1 (1 ​× ​10⁸ ​PFU), 0.001 (1 ​× ​10⁵ ​PFU), and 0.00001 (1 ​× ​10³ ​PFU). Lethal infection models were established in mice via intraperitoneal injection with 1 ​× ​10⁸ ​CFU/mouse of Ab25 (ST1791/KL101). Therapeutic intervention was initiated at one hour post-infection. A The survival curves of mice treated with phage P425 at MOIs of 1, 0.001, and 0.00001. B The colony counts in the liver, spleen, lungs, kidneys, and blood of the mice at the checkpoint of 10 ​hours. C The counts of phage plaques in the above organs and in the blood samples. D The clinical scores within 168 ​hours. E The successive changes in body weights of the surviving mice for seven days. Data are presented as mean ​± ​SD. Statistical analysis was performed using one-way ANOVA. ∗∗∗∗, P ​< ​0.0001.

Combination therapy of P425 with levofloxacin outperforms monotherapy

At 1-h post-infection with Ab25 (ST1791/KL101), we administered an intraperitoneal combination of P425 (MOI = 0.001) and levofloxacin to infected mice and compared survival rates to P425 monotherapy, levofloxacin monotherapy, and untreated control groups. Results revealed that P425 ​+ ​levofloxacin, P425, and levofloxacin increased the survival rate of infected mice to 100%, 83.3%, and 41.67%, respectively, compared to untreated controls (Fig. 7A). At the 10-h checkpoint, bacterial and phage loads were assessed in organ and blood samples. The combination therapy group showed the most pronounced reduction in bacterial load across all tested organs and blood (Fig. 7B). Notably, bacterial colonies were undetectable in the blood of all treated groups. Phage titers were measurable in the liver, spleen, lungs, and kidneys, confirming systemic distribution. In the bloodstream, phage levels reached 1.25 ​× ​10⁴ ​PFU/mL in the P425 group and 1.12 ​× ​10⁴ ​PFU/mL in the combination therapy group (Fig. 7C). Clinical scoring indicated that all treatment groups returned to baseline health within 72–96 ​hours, with the combination therapy group exhibiting the fastest recovery (Fig. 7D). However, all infected groups maintained lower body weights compared to uninfected controls (Fig. 7E).

Fig. 7.

Therapeutic efficacy evaluation of phage P45 in combination with levofloxacin in mice. Female Kunming mice aged 4–5 weeks were randomized to five groups (n ​= ​15/group): PBS control, infection control, levofloxacin (30 ​mg/kg), phage P425 (1 ​× ​10⁵ ​PFU), phage P425 (1 ​× ​10⁵ ​PFU) ​+ ​levofloxacin (30 ​mg/kg). Lethal infection models were established in mice via intraperitoneal injection with 1 ​× ​10⁸ ​CFU/mouse of Ab25 (ST1791/KL101). Therapeutic intervention was initiated at one hour post-infection. A The survival curve. B The colony counts in liver, spleen, lungs, kidneys, and blood of mice at the 10-h checkpoint, C The plaque counts of phage in the above mentioned organs and in the blood. D The clinical scores. E The change in body weights of the surviving mice for seven consecutive days. F, G The change in white blood cells and neutrophils in mice under different treatments, respectively. Data are presented as mean ​± ​SD. Statistical analysis was performed using one-way ANOVA. ∗, P ​< ​0.05, ∗∗, P ​< ​0.01, ∗∗∗, P ​< ​0.001, ∗∗∗∗, P ​< ​0.0001; ns: no significant.

Routine blood analysis showed no statistically significant differences in total white blood cell counts between treated and control mice. However, only the combination therapy group restored neutrophil counts to control levels, while the P425 and levofloxacin monotherapies showed significantly reduced neutrophil counts (Fig. 7F and G), suggesting enhanced immune recovery in the combination group.

Inflammatory cytokine assays revealed a slight increase in IL-1β and TNF-α levels in the levofloxacin-treated and phage P425-treated groups compared to the control group, indicating the presence of residual inflammation (Fig. 8A and B). There was no difference of serum IL-6 levels between all treatment groups and the control group (Fig. 8C).

Fig. 8.

The inflammatory cytokine and pathologic changes after combination treatment. For combined efficacy assessment at the 10-h checkpoint, three mice per group were randomly euthanized. Serum levels of cytokines IL-6, IL-1β, and TNF-α were quantified using cytokine ELISA kits. Liver, spleen, lung, and kidney tissues were collected, fixed in 4% paraformaldehyde, paraffin-embedded, sectioned, and stained with hematoxylin and eosin (H&E). A–C The changes of IL-1β, TNF-α and IL-6 inflammatory cytokines at the 10-h checkpoint, respectively. Data are presented as mean ​± ​SD. Statistical analysis was performed using one-way ANOVA. ∗, P ​< ​0.05, ∗∗, P ​< ​0.01, ∗∗∗, P ​< ​0.001, ∗∗∗∗, P ​< ​0.0001; ns: no significant. D Pathologic changes in major organs of mice in each experimental group. Arrows indicate key pathologies: blue ​= ​inflammatory infiltration; black ​= ​architectural alteration; red ​= ​vascular abnormalities.

Histopathological analysis of the liver, spleen, lungs, and kidneys revealed reduced inflammation and hemorrhage in all treatment groups compared to the infected group (Fig. 8D). Collectively, these results demonstrate that combination therapy with phage P425 and levofloxacin confers superior anti-inflammatory and tissue-protective effects compared to either agent alone.

P425 effectively prevents bacteremia in mice within a specific time window

Prophylactic administration of phage P425 demonstrated time-dependent protective efficacy against Ab25 (ST1791/KL101) infection in mice. Immediate administration post-infection provided complete protection, with 100% survival. When administered six ​hours prior to infection, the protection rate remained high at 75%. However, efficacy dropped sharply with earlier administration (only 12.5% of mice survived when P425 was given 12 ​hours before infection, and no mice survived when it was administered 24 ​hours in advance). This decline in protective effect is likely due to the progressive metabolic clearance of the phage over time (Fig. 9A). Mice that survived early prophylaxis regained normal activity levels within 72–84 ​hours post-infection (Fig. 9B). However, their body weights remained lower than those of uninfected controls (Fig. 9C). These results suggest that prophylactic administration of P425 is highly effective when delivered within a 6-h window prior to infection, offering significant protection against A. baumannii-induced bacteremia.

Fig. 9.

The preventive effect of phage P425 in mice. Female Kunming mice aged 4–5 weeks were randomly allocated to six groups (n ​= ​8/group): PBS control, infection control (1 ​× ​10⁸ ​CFU/mouse), and P425-treated groups (1 ​× ​10⁵ ​PFU/mouse) administered at −24, −12, −6, or 0 ​h relative to bacterial infection. A The survival curves at different times of prevention. B The clinical scores. C The body weight changes of surviving mice for seven consecutive days. Data are presented as mean ​± ​SD.

Discussion

In this study, we isolated phage P425, a broad-spectrum lytic phage with a lysis rate of approximately 79.59% (39/49). The host bacterium used for isolation, Ab25, belongs to the KL101 capsular type (ST1791). However, P425 does not exhibit strict capsular type specificity. It effectively targets strains with capsular types K3, K7, K9, K14, KL72, KL77, KL101, KL104, and KL160, corresponding to ten distinct sequence types, including novel ones: ST195, ST208, ST369, ST1968, ST191, ST136, ST1791, ST2499, ST540, and ST new1. Typically, A. baumannii phages are known for their narrow host range and high target specificity (Schooley et al., 2017). For instance, phage AB1 selectively lyses certain A. baumannii strains, but fails to infect others, while phage P1068 is specific to strains with the K3 capsular type (Yang et al., 2010; Zheng et al., 2024). Similarly, phage vB_Ab4_Hep4 targets only a limited range of A. baumannii strains, although occasional genetic mutations may broaden its host range (He et al., 2024).

Host recognition and attachment represent the initial stage of phage infection. Tail fiber proteins, which are classified as receptor-binding proteins (RBPs), play a crucial role in recognizing and binding to the host, thereby determining the phage's host specificity (de Leeuw et al., 2020; Ouyang et al., 2024). Mutations in these tail fiber proteins can lead to alterations in host profiles. For example, a spontaneous mutation in the putative tail fiber gene of phage PaP1 resulted in a broader host range compared to its parental phage, JG004 (Le et al., 2013). In addition to tail fibers, point mutations in the tail stinger protein of the Shigella phage Sf6 have also been shown to facilitate host range expansion (Subramanian et al., 2022). Consequently, alterations in tail fibers or tail spike proteins may drive phage host range evolution and enable escape from bacterial defense mechanisms (Zheng et al., 2024). In our preliminary analysis, we identified four phages with high sequence similarity to P425 using BLAST against the NCBI database for covariance analysis. The results revealed that P425 shared minimal similarity in the tail fiber and tail stinger gene regions with phages vB_ApiP_P2 (NC_042007.1), vB_AbaP_APK87 (MN604239.1), MRABP9 (OP727261.1), and vB_Ab-P-7 (OQ982387.1), which may explain the differences in their host ranges. We also predicted the putative tail fibronectin protein (CDS40) of P425 and found it may encode a depolymerizing enzyme, with a modeling confidence of 99.2%. Genome annotation of P425 revealed the presence of lysis-related genes, including endolysin and holin. Whether these components contribute to the phage's ability to lyse different capsular types remains to be clarified in future studies. Importantly, no virulence factors or antibiotic resistance genes were identified in the P425 genome, supporting its potential as a safe and broadly applicable therapeutic agent. Similarly, other phages such as vB_AbaSi_W9 (identified by Choi et al. (2024)) and Abp95 (reported by Huang et al. (2023)) have demonstrated lytic activity against A. baumannii strains of various sequence types.

Secondly, a 24-h in vitro study on bacterial suppression by P425 demonstrated that its inhibitory effect on bacterial growth was stage-dependent. At MOIs ranging from 100 to 0.01, P425 effectively inhibited bacterial proliferation for up to 24 ​hours. However, purification and susceptibility testing of colonies from the 96-well plates indicated that a small fraction of resistant bacteria persisted. At lower MOIs (0.001–0.00001), bacterial growth was initially suppressed for approximately 9 ​hours, after which resistant strains began to emerge. This pattern differs from the findings of Han et al. (2023), who reported that phage BUCT631 exhibited similar inhibition across different MOIs, with resistant bacterial growth observed after just 6 ​hours. In contrast, Choi et al. (2024) found that the concentration-dependent inhibition pattern observed with phage vB_AbaSi_W9 varied by strain: A. baumannii ATCC 17978 and KBN10P05982 (ST369) showed MOI-dependent suppression, while strains such as KBN10P04948 (ST191), LIS2013230 (ST208), and KBN10P05231 (ST369) exhibited minimal differences in inhibition across MOIs, with resistant bacteria eventually emerging in all cases. These variations suggest that differences in phage and bacterial strain characteristics play a significant role in inhibitory dynamics. Notably, P425 demonstrated strong and sustained bactericidal activity, along with exceptional efficacy in both inhibiting and eradicating biofilms.

Furthermore, previous experimental results demonstrated that phage P425, even at low concentrations, can synergize with five antibiotics that act via different mechanisms. Notably, it reduced the MIC of levofloxacin to only 1/16 of its original value. This phenomenon, known as phage-antibiotic synergy (PAS), refers to the ability of sub-inhibitory antibiotic concentrations to enhance phage replication and, consequently, phage-mediated bacterial killing (Comeau et al., 2007). A key advantage of PAS is its potential to slow the development of antimicrobial resistance and shift evolutionary pressures in a favorable direction by intensifying bacterial suppression (Burmeister et al., 2020; Chan et al., 2016; Fujiki et al., 2023; Torres-Barceló and Hochberg, 2016). Several PAS mechanisms have been proposed, largely based on phenotypic observations, including antibiotic-induced cell elongation/filamentation and increased phage replication (Bulssico et al., 2023; Diallo and Dublanchet, 2022). However, the precise mechanisms underlying the synergy between P425 and antibiotics remain to be fully elucidated. In vivo, combining P425 with levofloxacin led to a 100% survival rate in infected mice, significantly reduced bacterial loads and inflammatory markers, and promoted faster recovery. Notably, this combination was more effective in restoring neutrophil counts than either agent alone. Supporting this observation, Rodriguez-Gonzalez et al. (2024) reported that phages and neutrophils can work synergistically to clear bacterial infections. For example, phage-neutrophil interactions effectively eliminated Pseudomonas aeruginosa lung infections in wild-type mice (Roach et al., 2017), while phage treatment was markedly less effective in neutropenic mice (Roach et al., 2017; Tiwari et al., 2011). Finally, we evaluated the prophylactic potential of P425 and found that its protective effect diminished with increased time between administration and infection. This reduction in efficacy is likely due to the metabolic clearance of phages in the absence of host bacteria. Because phages do not persist long-term in the body without their target, the timing of prophylactic administration is critical for maximizing effectiveness.

Conclusions

In summary, we identified and characterized P425, a broad-spectrum lytic phage capable of targeting nine distinct A. baumannii capsular types, demonstrating a wide host range. Genomic analysis confirmed that P425 represents a novel member of the Friunavirus genus. In vitro, P425 exhibited potent bacteriostatic activity, effectively inhibiting bacterial growth and eradicating biofilms formed by A. baumannii. P425 also showed strong synergistic effects when combined with five antibiotics of differing mechanisms, significantly reducing their minimum inhibitory concentrations. Notably, the MIC of levofloxacin was reduced to 1/16 of its original value. Both in vivo and in vitro safety evaluations confirmed the phage's biosafety. In a murine infection model, P425 achieved excellent therapeutic outcomes. When used alone or in combination with low-dose levofloxacin, it increased survival rates up to 100%. The combination therapy not only enhanced bacterial clearance and clinical recovery but also reduced the required phage dosage, thereby lowering potential treatment costs. Moreover, prophylactic administration within an optimal time window provided effective protection against infection. Overall, P425 demonstrates strong potential as a therapeutic candidate. Phage-based treatments such as this may offer a promising strategy to combat antibiotic-resistant bacterial infections; however, further studies and preclinical evaluations are essential to support future clinical applications.

Materials and methods

Bacterial strains and growth conditions

This study utilized 49 MDRAB strains collected from the Second Hospital of Nanjing, China, as listed in Table 1. All A. baumannii strains were isolated and purified using Luria-Bertani (LB) solid medium, and subsequently cultured in LB broth at 37 ​°C with shaking.

Phage isolation and purification and host range determination

To purify the phage isolated from sewage, individual plaques were repeatedly screened on double-layer agar plates using Acinetobacter baumannii strain Ab25 as the host until uniform plaque morphology was achieved. Phage nomenclature adhered to the guidelines established for bacterial and archaeal viruses (Kropinski et al., 2009,b).

The host range of phage P425 was determined using the spot-drop method against all 49 MDRAB strains (Khan Mirzaei and Nilsson, 2015). Phage titers were assessed using the double-layer agar method (Kropinski et al., 2009,b). Briefly, 100 ​μL of diluted phage suspension and host bacteria in the logarithmic growth phase were mixed with 5 ​mL of semi-solid LB (0.8% agar), poured onto solid LB plates, and incubated overnight at 37 ​°C.

Morphological observation by transmission electron microscopy

Phage particles were enriched and purified following the protocol described by Nasukawa et al. (2017). A drop of purified phage suspension was placed onto a copper grid, dried under an incandescent lamp, and negatively stained with 2% (w/v) phosphotungstic acid (PTA). Phage morphology was examined, and electron micrographs were captured using a Hitachi H-7650 transmission electron microscope in Japan.

Biological characterization of P425

Assays evaluating the temperature, pH, and UV stability of phage P425 were assessed following established protocols with slight modifications (Laemmli, 1970). To evaluate the temperature stability, phage (100 ​μL, 2 ​× ​10⁹ ​PFU/mL) was mixed with 900 ​μL of SM buffer and incubated at various temperatures (i.e., 4 ​°C, 25 ​°C, 37 ​°C, 50 ​°C, 60 ​°C, and 70 ​°C) for one ​hour. Phage titers were then determined using the double-layer agar method. To assess the pH stability, phage P425 was adjusted to a starting concentration of 2 ​× ​10⁸ ​PFU/mL in SM buffer and incubated at 37 ​°C for one ​hour at pH values ranging from 1 to 12, then phage titers were assessed by the above method. To determine the UV stability, phage (2 ​× ​10⁸ ​PFU/mL) samples were exposed to UV light and samples were collected every 10 ​mins over a 60-min period. All samples were subsequently diluted, and phage titers was evaluated using the double-layer agar method. In all stability assays, the relative phage viability was represented as the percentage of surviving phages relative to the initial titer (2 ​× ​10⁸ ​PFU/mL), calculated by: 100% ​× ​(post-treatment titer/initial titer). The untreated control was defined as 100% viability.

To determine the optimal multiplicity of infection (OMOI), phage P425 was co-cultured with host bacteria in logarithmic phase at various MOIs (100, 10, 1, 0.1, 0.01, 0.001, 0.0001, and 0.00001). The mixtures were incubated on a shaking platform at 37 ​°C for 8 ​hours. After centrifugation at 2390×g for 10 ​mins at 4 ​°C, the supernatants were filtered and diluted, and the phage titers were quantified. The MOI that yielded the highest phage titer in the liquid culture was considered the OMOI.

To evaluate the adsorption rate, phage P425 was mixed with log-phase host bacteria at an MOI of 0.1. Samples were taken at various time intervals, and the titers of unadsorbed (free) phages in the supernatants were determined. Adsorption efficiency was calculated as the percentage of free phages remaining over time.

To ascertain the one-step growth curve of P425, the log-phase bacterial solution was mixed with phage at an MOI of 0.1 and co-incubated at 37 ​°C for 5 ​mins. After centrifugation to remove unbound phages, the pellet was resuspended in 10 ​mL of fresh LB broth and incubated at 37 ​°C with shaking at 180 ​rpm. Samples were taken at defined intervals, and phage titers were determined using the double-layer agar method. The burst size was calculated by dividing the number of phages at the plateau phase by the initial number of infected bacterial cells (Lee et al., 2019). All experiments were performed in triplicate to ensure reproducibility.

Whole genome sequencing and biological analysis of P425 and host bacteria

Purified DNA from phage P425 was extracted using the λ Phage Genome DNA Rapid Extraction Kit (Beijing Zhuangmeng International Bio-genetic Science and Technology Co.) and delivered to Shanghai Parsonage Bio-technology Co. for sequencing. Log-phase cultures of the host bacterium were centrifuged at 4 ​°C, 5000×g for 10 ​mins, and the bacterial pellet was submitted to Beijing Novozymes Ltd. for whole genome sequencing. Both libraries were constructed on the Illumina NovaSeq platform and sequenced using paired-end (PE). Raw sequencing data were quality-filtered to remove adapter sequences and de novo assembled using A5-MiSeq and SPAdes (Bankevich et al., 2012; Coil et al., 2015).

Assembled host genome data were uploaded to Pathogenwatch v22.3.8 (https://pathogen.watch) for multi-locus sequence typing (MLST) and capsular polysaccharide locus (KL) typing prediction. The assembled phage genome was analyzed as follows: genome circular mapping was performed using CGView; gene function annotations were conducted with Prokka; and potential virulence factors and antibiotic resistance genes were identified using the VFDB and CARD databases, respectively. Phage phylogenetic and comparative analyses were carried out using MEGA11 (for evolutionary tree construction) and Easyfig (for visual genome comparison). Genomes of 50 phages with high sequence similarity to P425 were retrieved from the NCBI database. Intergenomic similarities and pairwise distances at the nucleotide level were computed using the Viral Intergenomic Distance Calculator (http://viridic.icbm.de) (Moraru et al., 2020). The complete genome sequence of P425 has been deposited in the NCBI database (accession number: PQ211117) and ScienceDB (https://doi.org/10.57760/sciencedb.20929).

Antibacterial activity of P425 in vitro

The host strain Ab25 was cultured to the logarithmic phase and diluted to a final concentration of 1 ​× ​10⁸ ​CFU/mL. Aliquots were dispensed into 96-well plates, followed by the addition of phage at various concentrations to achieve MOIs of 100, 10, 1, 0.1, 0.01, 0.001, 0.0001, and 0.00001. The final volume per well was adjusted to 200 ​μL. The 96-well plates were placed in a microplate reader set to measure absorbance at 600 ​nm and incubated continuously at 37 ​°C for 24 ​hours. Each experiment was repeated three times to ensure reproducibility.

Inhibition and removal of biofilms by P425

To assess the inhibitory effect of phage P425 on biofilm formation, A. baumannii strain Ab25 (OD₆₀₀ ​= ​0.5), grown to the logarithmic phase, was incubated with various concentrations of phage P425 at 37 ​°C for 24 ​hours. Biofilm quantification was performed using crystal violet staining as described previously (Kakian et al., 2024). For biofilm removal assays, logarithmic-phase Ab25 cultures were incubated in a PVC 96-well plate at 37 ​°C for 24 ​hours to allow biofilm formation. After incubation, planktonic cells were removed by washing with PBS, and phage P425 was added at different dilutions. Plates were incubated at 37 ​°C for an additional 24 ​hours. The biofilm removal assessment was conducted as previously described (Kakian et al., 2024). LB medium alone was used as the negative control, while untreated bacterial cultures served as the positive control. All experiments were conducted in triplicate.

Combined administration of P425 and antibiotics in vitro

The checkerboard assay was employed to evaluate the synergistic effect between phage P425 and various antibiotics. Antibiotics tested included meropenem, ceftriaxone sodium, tigecycline, tetracycline, levofloxacin, ciprofloxacin, polymyxin B, and sulfamethoxazole. These were serially diluted in 96-well plates starting from their respective MIC values. The MIC and minimum bactericidal concentration (MBC) values of each antibiotic against strain Ab25 are summarized in Table 2. Log-phase Ab25 bacterial cultures (∼1 ​× ​10⁷ ​CFU/mL) were added to the wells, followed by the addition of phage P425 at an MOI of 0.001. Control conditions included: (1) LB group (blank), (2) bacterial solution alone group (growth control), (3) phage-treated group (MOI = 0.001), (4) antibiotic-treated group, and (5) antibiotic MIC determination group. The absorbance at 600 ​nm was measured using a microplate reader every hour for 24 ​hours at 37 ​°C. Growth curves were generated by averaging the recorded values across replicates.

Table 2.

The sensitivity of Ab25 to different antibiotics.

Antibiotic	Mechanisms of action	Classifications	MIC (μg/mL)	MBC (μg/mL)
Sulfamethoxazole	Inhibition of folate synthesis	Sulfonamide	>512	>512
Ceftazidime	Inhibition of cell wall synthesis	Beta-lactams	512	>512
Ceftriaxone sodium	Inhibition of cell wall synthesis	Beta-lactams	>512	>512
Ciprofloxacin	Inhibition of DNA synthesis	Quinolone	256	>512
Tetracycline	Inhibition of protein synthesis	Tetracycline	512	>512
Tigecycline	Inhibition of protein synthesis	Tetracycline	2	4
Levofloxacin	Inhibition of DNA synthesis	Quinolone	16	32
Imipenem cilastatin sodium	Inhibition of cell wall synthesis	Beta-lactams	128	256
Meropenem	Inhibition of cell wall synthesis	Beta-lactams	64	128
Polymyxin B	Increased plasma membrane permeability	Polypeptide	2	4

Notes: MIC, minimum inhibitory concentration; MBC, minimum bactericidal concentration.

Cytotoxicity assay and hemolysis assay

The cytotoxicity of phage on human liver cells (LO2) purchased from Nanjing Cobioer Biosciences Co., Ltd. (Cat# CBP60224) was assessed using the CCK-8 Cell Proliferation and Cytotoxicity Assay Kit (Biosharp). Briefly, LO2 cells were cultured in RPMI-1640 medium (Gibco) supplemented with 20% fetal bovine serum (FBS, SUNNCELL). Cells (1 ​× ​10⁴ per well) were seeded into 96-well plates and incubated at 37 ​°C in a humidified incubator with 5% CO₂ for 24 ​hours. Phage solutions at varying concentrations were prepared in RPMI-1640 medium and added to the wells. PBS was used as the negative control and 2% Triton X-100 (Solarbio) served as the positive control. After 24 ​hours of incubation, CCK-8 reagent was added and incubated for an additional two ​hours. Absorbance was measured at 450 ​nm using a microplate reader.

To evaluate the hemolytic activity of phage P425 in vitro, erythrocytes from mice were exposed to phage preparations, following the method described by Wang et al. (2021). All experiments were conducted in triplicate.

In vivo safety assessment of P425

Female Kunming mice (4–5 weeks old) were obtained from Suzhou Specific Pathogen-Free Animal Technology Co. Following a period of acclimatization, mice were randomly divided into three groups (n ​= ​5 per group). The negative control group received 200 ​μL of PBS, while the treatment groups received 200 ​μL of phage P425 (1 ​× ​10¹⁰ ​PFU/day) via intraperitoneal or tail vein injection for seven consecutive days. Body weight and survival were monitored daily, and clinical assessments were performed using the Mouse Sepsis Scale, which ranges from 1 (normal) to 5 (moribund) (Shrum et al., 2014). Evaluations were conducted over a 168-h observation period. At the end of the experiment, mice were anesthetized with isoflurane and euthanized by cervical dislocation. Blood samples were collected via retro-orbital puncture into anticoagulant tubes, and analyzed using a Sysmex XN-2800 hematology analyzer.

Mouse bacteremia model

To establish a mouse model of bacteremia, female Kunming mice (4–5 weeks old) were randomly divided into three groups (n ​= ​5 per group). The negative control group received 200 ​μL of PBS, while the experimental groups were intraperitoneally injected with 1 ​× ​10⁷ or 1 ​× ​10⁸ ​CFU/mouse of A. baumannii strain Ab25 (ST1791/KL101), respectively. The mice were monitored every two ​hours for clinical signs and survival, with clinical scores recorded over a 48-h observation period. In cases of mortality, the liver, spleen, lungs, kidneys, and blood were collected for bacterial colony enumeration.

P425 treatment in bacteremia mouse model

To establish a lethal infection model, mice were challenged with 1 ​× ​10⁸ ​CFU of A. baumannii strain Ab25 (ST1791/KL101) per mouse, which led to fatal outcomes within 24 ​hours. To evaluate the therapeutic potential of phage P425, mice were randomly divided into five groups (n ​= ​8 per group). The negative control group received PBS only. The infection control group was inoculated with 1 ​× ​10⁸ ​CFU of bacteria. The treatment groups received de-endotoxinized P425 phage one ​hour post-infection at different MOIs: 1, 0.001, and 0.00001. At the 10-h checkpoint, three mice from each group were euthanized, and tissues including the liver, spleen, lungs, kidneys, and blood were collected to determine bacterial load and phage titers. The remaining mice were monitored daily for 168 ​hours, with assessments of survival, body weight, and clinical condition using the Mouse Sepsis Score.

Based on the treatment outcomes, we further assessed the combined therapeutic efficacy of phage P425 and levofloxacin at an MOI of 0.001 when treatment was initiated at one ​hour post-infection. Fifteen mice were randomly assigned to five groups: PBS, infection control (1 ​× ​10⁸ ​CFU of bacteria per mouse), or three therapy groups (30 ​mg/kg levofloxacin; 1 ​× ​10⁵ ​PFU P425; 1 ​× ​10⁵ ​PFU P425 ​+ ​30 ​mg/kg levofloxacin). At the 10-h checkpoint, three mice from each group were randomly euthanized for bacterial and phage quantification in the liver, spleen, lungs, kidneys, and blood. Tissue sections were fixed in 4% paraformaldehyde, embedded in paraffin, and stained with hematoxylin and eosin (H&E) for histopathological examination. Serum levels of IL-6, IL-1β, and TNF-α were measured using cytokine ELISA kits (ELK Biotech) according to the manufacturer's instructions. The remaining mice were monitored daily over a 168-h period for survival, body weight, and clinical scores using the Mouse Sepsis Scale.

Preventive treatment with P425

To evaluate the prophylactic efficacy of phage P425 against bacteremia in mice, animals were randomly assigned to six groups (n ​= ​8 per group). The negative control group received PBS, while the positive control group was infected with A. baumannii Ab25 (ST1791/KL101) at a dose of 1 ​× ​10⁸ ​CFU/mouse, consistent with previous experiments. The experimental groups were administered de-endotoxinized P425 phage solution (1 ​× ​10⁵ ​PFU/mouse) immediately after infection (0 ​hour post-infection), or prophylactically at 24, 12, and 6 ​hours prior to infection. Survival was monitored for 168 ​hours following bacterial challenge.

Statistical analysis

Statistical analyses were performed using GraphPad Prism version 7. Differences between experimental and control groups were evaluated using one-way analysis of variance (ANOVA). A P-value of less than 0.05 (P ​< ​0.05) was considered statistically significant.

Data availability

All the data generated during the current study are included in the manuscript. The genomic data of P425 has been uploaded to the NCBI database (PQ211117) and ScienceDB (https://doi.org/10.57760/sciencedb.20929).

Ethics statement

All animal experimental studies were approved by the Laboratory Animal Welfare and Ethics Committee of Kangtai Medical Laboratory Services Hebei Co Ltd (Ethical Approval No.: MDL2024-05-21-01) and were conducted in accordance with the guidelines of the animal welfare organization.

Author contributions

Miaomiao Lin: conceptualization, methodology, validation, formal analysis, investigation, resources, data curation, writing-original draft, visualization, funding acquisition. Lele Xiong: validation, investigation, resources. Wen Li: visualization, resources. Lingyan Xiao: resources, funding acquisition. Wei Zhang: resources. Xiaogui Zhao: resources. Yishan Zheng: conceptualization, funding acquisition, resources, supervision, project administration, writing-review and editing.

Conflict of interest

The authors declare that they have no conflict of interest.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Figure S1.

Genome-wide similarity analysis of phage P425. The similarity of intergenic distances between phage P425 and the genomes of 50 Acinetobacter baumannii phages in the NCBI database was calculated from the nucleic acid level using VIRIDIC, and the similarity between P425 and them ranged from 71.4 ​% to 89.3 ​%, and it was defined as a new species in the genus Friunavirus.

Supplementary Figure S2.

The phylogenetic tree constructed for phage P425. The phylogenetic trees are constructed terminal large subunits (A), whole genome sequences (B), capsid proteins (C), and endolysin proteins (D), respectively. The phylogenetic tree was drawn using MEGA (version 11.0.11) using the Neighbor-Joining (N-J) method with a Bootstrap value of 1000, and the percentage of nodes in subfigure A represents site coverage.

Supplementary Figure S3.

The sensitivity of residual bacteria colonies to phage P425. After 24 ​h' inhibition, the residual bacteria colonies were purified and their sensitivity to phage P425 was determined. Blue color indicates sensitivity to P425 and white color indicates insensitivity.

Supplementary Figure S4.

Checkerboard assays showing 24-h combinatorial effects of phage P425 (MOI ​= ​0.001) with the full range of antibiotic concentrations (data not shown in Fig. 4). Controls identical to Fig. 4. Data are presented as mean ​± ​SD.

Supplementary Figure S5.

The virulence assay of Ab25 in mice. Four to five week-old female Kunming mice were randomly assigned to three groups (n ​= ​5/group). The negative control group was administered 200 ​μL of PBS, while the experimental groups received intraperitoneal injections with 1 ​× ​10⁷ or 1 ​× ​10⁸ ​CFU/mouse of bacteria Ab25. A The survival curves. B The colony counts in liver, spleen, lung, kidney and blood of death mice were determined. C The clinical scores. Data are presented as mean ​± ​SD.