Isolation and therapeutic potential of phage vB_EcoM_GXW16 against a drug-resistant avian pathogenic Escherichia coli strain
Ting Xu, Wenwen Yang, Jia Cao, Xiaofang Wei, Huixin Liu, Yiming Li, Sijia Pan, Nihar Ali, Hongbin Si
TL;DR
A new bacteriophage was isolated that effectively targets drug-resistant Escherichia coli in poultry, offering a potential alternative to antibiotics.
Contribution
The discovery and characterization of a novel lytic phage with therapeutic potential against drug-resistant APEC strains.
Findings
The phage vB_EcoM_GXW16 showed broad lytic activity against 70.91% of avian-source E. coli isolates.
It demonstrated environmental stability and significantly inhibited drug-resistant E. coli in vitro.
In vivo tests showed improved survival rates and reduced bacterial loads in infected chickens.
Abstract
Avian pathogenic Escherichia coli (APEC) is a major pathogenic subset of E. coli responsible for avian colibacillosis, representing one of the most common and economically damaging bacterial threats to the poultry industry globally. Currently, clinical treatment mainly relies on antibiotics. However, the widespread prevalence of drug-resistant strains poses a major challenge to global public health. Bacteriophages (phages) have regained significant attention as promising alternatives to antibiotics. In this study, a lytic bacteriophage vB_EcoM_GXW16 was isolated and purified from wastewater samples collected at a poultry farm. The phage exhibited broad lytic activity against a panel of avian-source Escherichia coli isolates (70.91%, 39/55). Its optimal multiplicity of infection (MOI) was 0.001, and it had a latent period of 10 min. The phage also demonstrated tolerance to a range of pH…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsBacteriophages and microbial interactions · Monoclonal and Polyclonal Antibodies Research · Bacterial Genetics and Biotechnology
Introduction
Avian pathogenic Escherichia coli (APEC) is an extraintestinal pathogenic E. coli (ExPEC) that causes a variety of localized and systemic infections in chickens, turkeys, ducks, and other avian species (Dho-Moulin and Fairbrother, 1999). The most common infections in chickens caused by APEC include perihepatitis, airsacculitis, and pericarditis, which are collectively referred to as avian colibacillosis (Mageiros et al., 2021). Furthermore, APEC is a significant contributor to high mortality rates (up to 53.5%) in chickens (Kathayat et al., 2021).The primary strategy for combating APEC infections in poultry has been the use of various antibiotics (Thomrongsuwannakij et al., 2020). However, antimicrobial resistance (AMR) has emerged as a major global threat to both public and veterinary health (Naghavi et al., 2022; Sevilla-Navarro et al., 2022). The emergence of antibiotic-resistant strains severely impacts animal health and food safety (Nawaz et al., 2024). This situation not only leads to treatment failures and economic losses but also facilitates the horizontal transfer of resistance genes, posing potential risks to human health (Hu et al., 2022). It has been hypothesized that antibiotic-resistant human ExPEC strains may originate from poultry through direct contact with birds and the consumption of contaminated poultry products (Mellata, 2013). An analysis of 81 APEC strains collected from eastern China between 2008 and 2023 by Shikai Song et al. revealed that 76.5% of the isolates exhibited multidrug resistance, posing a serious challenge to infection control (Song et al., 2025). Although some vaccines against APEC are available, their efficacy is primarily limited to specific serogroups and has not been proven to offer broad protection (Jamali et al., 2024). Consequently, there is an urgent need to develop novel therapeutic alternatives to antibiotics for the treatment of avian colibacillosis caused by APEC (Kazibwe et al., 2020a).
The escalating crisis of antibiotic resistance has revitalized interest in phage therapy (Strathdee et al., 2023). Phages, which are viruses that specifically infect and replicate within bacterial cells (Hatfull et al., 2022), are the most abundant and ubiquitous biological entities on Earth (Hassan et al., 2021). They represent a natural antimicrobial agent (Soontarach et al., 2022) and are considered a promising alternative for treating bacterial diseases (Wang et al., 2024). In comparison with conventional antibiotics, phage therapy presents several distinct advantages. Phages are abundant and widely distributed in nature, allowing for rapid and cost-effective screening, unlike the development of new antibiotics (Liu et al., 2024). Furthermore, phages exhibit remarkable specificity for bacterial infections, enabling selective targeting of pathogenic bacteria without harming other bacterial species, thereby preserving the balance of the host's microbiota (Gordillo Altamirano and Barr, 2019). The safety and efficacy of phage therapy for treating severe infections have been confirmed in numerous clinical trials (Kaikabo et al., 2017). These studies highlight the potential of phages as a viable solution to the global antibiotic resistance crisis (Subramanian, 2024).
In the present study, a lytic phage designated vB_EcoM_GXW16 was isolated from poultry farm wastewater, with its biological characteristics, genomic sequence and in vitro bacteriostatic effects subsequently characterized. We further evaluated its lytic activity against a panel of avian-derived E. coli isolates and assessed its therapeutic potential against the APEC strain (Escherichia coli O117:H25_E5) in a chicken model of colibacillosis.
Materials and methods
Bacterial strain and drug resistance phenotype detection
A total of 55 avian-source Escherichia coli strains were employed for phage isolation and host range determination. Of these, 47 strains were obtained from laboratory stock (Li et al., 2022a; Zhang et al., 2022a), while the remaining 8 strains were isolated from liver samples of deceased layer or broiler chickens collected from farms in Guangxi, China, or from laboratory settings. All strains were cultured on selective media. The eight E. coli isolates obtained in this study were confirmed by PCR assay using primers reported in a previous study (Saeed et al., 2022). One isolate was subjected to whole-genome sequencing and identified as an APEC strain, designated Escherichia coli O117:H25_E5.
The antimicrobial resistance phenotype of Escherichia coli O117:H25_E5 was assessed using the Kirby-Bauer disk diffusion method (Webber et al., 2022). Antimicrobial susceptibility test disks were purchased from Changde Bikeman Biotechnology Co., Ltd. (Changsha, China). All experimental procedures, including quality control and result interpretation, were performed strictly in accordance with the Clinical and Laboratory Standards Institute (CLSI) VET01S, 5th Edition (2024) guidelines (Yu et al., 2025).
Isolation and purification of bacteriophages
Bacteriophages were isolated and purified as previously described (Rajab et al., 2024) with some modifications. Wastewater and fecal samples were collected from chicken farms in Guangxi Province, China. An aliquot of each sample was mixed with 5 mL of Luria-Bertani (LB) broth and incubated at 37°C with shaking for 4 h. The mixture was centrifuged at 10,000 rpm for 5 min, and the supernatant was filtered through a 0.22 µm pore-size membrane to remove bacterial cells. The filtrate, designated as the crude phage lysate, was collected. Phages were isolated using the double-layer agar spot assay (Mirzaei and Nilsson, 2015; Hyman, 2019). Briefly, 200 µL of log-phase Escherichia coli culture was mixed with 5 mL of molten soft agar maintained at approximately 55°C. The mixture was immediately poured onto a pre-prepared LB solid agar plate and allowed to solidify at room temperature. Subsequently, 5 µL aliquots of the crude phage lysate were spotted onto designated areas of the solidified double-layer plates. The plates were incubated at 37°C for 6 - 12 h and examined for plaque formation. Individual, well-isolated plaques were picked and subjected to 4-6 rounds of plaque purification using the same double-agar overlay method to obtain purified phage stocks. The purified phage stocks were mixed with an equal volume of sterile 50% (v/v) glycerol, and stored at -80°C for long-term preservation and further studies.
Determination of phage host range
The host range of the phage was determined using the double-layer agar spot assay (Bilhman et al., 2025a). Briefly, a 5 µL aliquot of the purified phage lysate was spotted onto the surface of bacterial lawns overlaid with different Escherichia coli strains. The plates were then incubated overnight at 37°C and examined for plaque formation. The appearance of clear plaques served as an indicator of bacterial lytic by the phage. Among them, vB_EcoM_GXW16 exhibited the broadest lytic spectrum, and was therefore selected for subsequent studies.
Morphological observation of phage vB_EcoM_GXW16
Phage morphology was examined by transmission electron microscopy (TEM) following established protocols (Ackermann, 2009; Hudson et al., 2013). In accordance with the International Committee on Taxonomy of Viruses (ICTV) guidelines (Adriaenssens and Brister, 2017), the phage was identified and classified based on its morphological features (Zerbini et al., 2023, 2025).
Determination of the optimal multiplicity of infection (MOI)
The optimal MOI for the phage vB_EcoM_GXW16 was determined as previously described with some modifications (Lu et al., 2003). The concentration of the host bacteria was adjusted to 10⁸ CFU/mL. Log-phase bacterial cultures were mixed with equal volumes of serially diluted phage suspensions (1 × 10⁴-1 × 10⁸ PFU/mL) to obtain MOIs of 0.0001, 0.001, 0.01, 0.1, and 1. The mixtures were incubated at 37 °C with shaking at 180 rpm for 4 h. Subsequently, the cultures were centrifuged at 12,000 rpm for 2 min and filtered through a 0.22 μm membrane to collect phage lysates. The lysate titers were determined using the double-layer agar plaque assay. The MOI yielding the highest lysate titer was defined as the optimal MOI for the phage.
One-step growth curve of phage vB_EcoM_GXW16
The method described by Tian et al. (2025)) was employed and adapted for this study as follows: 500 µL of the host bacterial culture and 500 µL of the phage suspension were mixed at the optimal multiplicity of infection (MOI). After incubation in a 37 °C water bath for 5 min, the mixture was centrifuged at 12,000 rpm for 5 min. The supernatant was removed, and the pellet was washed 2-3 times with pre-warmed LB broth. The pellet was then resuspended in 1 mL of pre-warmed LB broth, which was added to 100 mL of fresh pre-warmed LB broth. The culture was then incubated in a constant-temperature shaker at 37°C. Samples were taken every 10 min for the first 120 min, and then every 20 min until 180 min. Phage titers were determined using the double-layer agar method (Waturangi, 2024). The burst size was calculated according to the following formula: Burst size = Final bacteriophage titer / Initial number of the host bacteria.
Determination of phage vB_EcoM_GXW16 adsorption rate
The phage adsorption assay was performed as described previously with minor modifications (Al-Zubidi et al., 2019). Phages and host bacteria were mixed at the optimal multiplicity of infection (MOI) and incubated at 37°C for 12 min. Samples were collected every 2 min, filtered through a 0.22 µm membrane, and serially diluted to determine the titer of unadsorbed phages in the supernatant. The adsorption percentage at each time point was calculated using the formula: [(Initial phage titer - Phage titer in supernatant) / Initial phage titer] × 100%.
Temperature stability, pH stability, and storage experiments
To assess the thermal stability of the phage vB_EcoM_GXW16, the phage lysate was incubated in water baths at 40 °C, 50 °C, 60 °C, 70 °C, and 80 °C. Samples were taken at 20, 40, and 60 min (Toaquiza-Vilca et al., 2025), and the phage titer was determined by the double-layer agar method. Furthermore, the phage suspension was stored at room temperature and 4°C for 4 weeks, with the phage titer measured weekly (Yao et al., 2023a). To evaluate pH stability, LB medium was adjusted to pH values ranging from 1 to 13 using HCl or NaOH. Take 1 mL of LB medium with different pH values, add 10 µL of phage, mix well, and incubated at 37°C for 1 h (Acharya and Chinnasamy, 2025). The phage titer under different pH conditions was then determined using the double-layer agar plate method.
In vitro bacteriostatic activity of phages vB_EcoM_GXW16
The experimental protocol was slightly modified based on the methods described by Chenglin Tao and Zhaohui Tang (Tao et al., 2021; Tang et al., 2023). To evaluate the inhibitory effect of phages on host bacteria in vitro, the host bacteria were infected with phages at three different multiplicities of infection (MOI: 1, 0.1, and 0.01). The host bacterial strain was adjusted to an OD_600_ of 0.2 (approximately 10⁸ CFU/mL). Phages were then added to the host bacteria at MOIs of 0.01, 0.1, and 1. The mixtures were incubated at 37°C for 12 h. Bacterial growth was assessed by measuring the absorbance at 600 nm in a 96-well plate at 2 h intervals. Concurrently, the number of viable Escherichia coli was determined by colony counting on MacConkey agar plates. Bacterial cultures without phage treatment served as the growth control, and LB broth alone served as the blank control.
Genome sequencing and analysis of phage
The genomic DNA of phage vB_EcoM_GXW16 was extracted using the E.Z.N.A.® Bacterial DNA Kit (Omega Bio-tek, USA) The genome of the phage vB_EcoM_GXW16 was sequenced using the Illumina novaseq 6000 platform at Shanghai Biozeron Biotechnology Co, Ltd. (Shanghai, China.). To ensure the reliability of the subsequent data analysis, the raw sequencing data were filtered and quality-controlled using the Fastp (https://github.com/OpenGene/fastp). The final assembly results were obtained using ABySS (v2.2.0) and GapCloser (v1.12) (Simpson et al., 2009; Xu et al., 2020). Putative open reading frames (ORFs) were predicted using ORF fnder (https://www.ncbi.nlm.nih.gov/orffnder/), and all gene models were blastp against non-redundant (NR in NCBI) database, SwissProt (http://uniprot.org), KEGG (http://www.genome.jp/kegg/), and COG (http://www.ncbi.nlm.nih.gov/COG) for functional annotation using the BLASTP algorithm. Putative tRNA genes were identified using the tRNAscan-SE (v2.0.4,http://lowelab.ucsc.edu/tRNAscan-SE) and rRNA were determined using the RNAmmer (v1.2, http://www.cbs.dtu.dk/services/RNAmmer/). To assess potential safety risks, the genome was screened against the Virulence Factors Database (http://www.mgc.ac.cn/cgi-bin/VFs/v5/main.cgi) and the Comprehensive Antibiotic Resistance Database (https://card.mcmaster.ca/analyze/rgi) to confirm the absence of virulence and antibiotic resistance genes. A genomic map was generated using Proksee (https://proksee.ca/). MEGA 12 was used for multiple phages sequences alignment, iTOL (https://itol.embl.de/personal_page.cgi) was used to infer phylogenies (Lang et al., 2023).
In vivo experiment on the treatment of avian colibacillosis with bacteriophage vB_EcoM_GXW16
To determine the Median Lethal Dose (LD_50_) (Lorke, 1983) of the APEC strain Escherichia coli O117:H25_E5, the following experiment was conducted. Sixty healthy white-feathered broilers were raised to 7 days of age and randomly divided into 6 groups with 10 birds per group. Broilers in each group were intraperitoneally injected with different doses of Escherichia coli O117:H25_E5 bacterial suspension (10¹⁰, 10⁹, 10⁸, 10⁷, 10⁶ CFU in 0.1 mL), while the control group received an equal volume of sterile saline. Chickens survival was recorded for 7 days post-challenge, and the median lethal dose (LD_50_) was calculated using SPSS software (Long et al., 2022).
To assess the therapeutic potential of phage vB_EcoM_GXW16 against Escherichia coli O117:H25_E5 infection, 96 one-day-old healthy white-feathered broiler chickens were housed under controlled laboratory conditions for six days. On day seven, the chickens were randomly allocated into eight groups (Phage Prophylaxis Group A, Phage Prophylaxis Group B, Phage Treatment Group C, Phage Treatment Group D, E. coli Control Group, Antibiotic Treatment Group, Blank Control Group, and Phage Control Group), with 12 chickens per group. The animal experimental groups and dosing regimen are shown in Table 1. Chickens in Groups A-D, the E. coli Control Group, and the Antibiotic Treatment Group were intraperitoneally injected with a 50% lethal dose (LD₅₀) of Escherichia coli O117:H25_E5, whereas the Blank Control Group and Phage Control Group received an equivalent volume of physiological saline instead. Phage Prophylaxis Groups A and B were orally administered a single dose of vB_EcoM_GXW16 at 10⁵ PFU and 10⁸ PFU, respectively, 3 h before bacterial challenge. Phage Treatment Groups C and D received oral administration of vB_EcoM_GXW16 at 10⁵ PFU and 10⁸ PFU, respectively, starting at 3 h post-infection and continuing once daily for three consecutive days (Oliveira et al., 2009). Florfenicol for the Antibiotic Treatment Group (Shaheen and El‑Far, 2013; Ghahramani et al., 2024) was purchased from Hubei Yuanhao Biotechnology Co., Ltd.(Hubei, China) and administered according to the manufacturer's instructions The Phage Control Group was orally administered 10⁸ PFU of vB_EcoM_GXW16 once daily for three days. The Blank Control Group and the E. coli Control Group were given an equivalent volume of physiological saline as a control. All groups were housed under identical conditions with a consistent light/dark cycle and had ad libitum access to feed and water throughout the experiment.Table 1. Animal experiment grouping and dosage regimen.Table 1: dummy alt textGroup (n=12)Treatment (oral administration)Phage Prophylaxis Group A3 h before infection, 10^5^PFU phagePhage Prophylaxis Group B3 h before infection, 10^8^PFU phagePhage Treatment Group C3 h post-infection, 10^5^PFU phage, daily for 3 daysPhage Treatment Group D3 h post-infection, 10^8^PFU phage, daily for 3 daysE. coli Control GroupEqual volume of physiological salineAntibiotic Treatment GroupFlorfenicol, according to the manufacturer's instructionsBlank Control GroupEqual volume of physiological salinePhage Control Group10^8^PFU phage, daily for 3 daysExcept for the blank control group and phage control group, all other groups received intraperitoneal injections of LD50 doses of Escherichia coli O117:H25.The phage used in this study was vB_EcoM_GXW16, and it was administered orally.Florfenicol was purchased from Hubei Yuanhao Biotechnology Co., Ltd. (Hubei, China)
On day 7 post-challenge, surviving chickens were fasted for 3 h and then individually weighed before being euthanized. Subsequently, the thoracic and abdominal cavities were opened to fully expose the heart and liver. The degree of organ damage was evaluated after modification based on the previously established severity scoring criteria for pericarditis and perihepatic inflammation (Wu et al., 2024). Detailed scoring criteria are provided in Table 2. Each condition was scored from 0 to 3, and the scores for pericarditis and perihepatitis were summed to obtain a total pathology score ranging from 0 to 6 (Lau et al., 2010). Chickens that died during the treatment period were assigned a score of 7, the mean lesion score was then calculated. To ensure objectivity, lesion scoring for surviving birds was performed in a double-blind manner, where the scorer was unaware of the group assignment of each bird. Two independent scorers evaluated the lesions. Their scores were compared for consistency, and any discrepancies were resolved by a third scorer, whose matching score was adopted as the final result (Ask et al., 2006; Lynne et al., 2012). Subsequently, a section of liver was aseptically collected, weighed, homogenized in sterile PBS, and plated on MacConkey plates agar for bacterial enumeration of the challenge strain (Green et al., 2017). In addition, the liver, heart and spleen were taken out under aseptic conditions, fixed in a 4% paraformaldehyde, the specimens were embedded and sectured, and then stained with hematoxylin and eosin (H&E) for histopathological analysis.Table 2. Scoring criteria for pericarditis and perihepatitis.Table 2: dummy alt textScoreDescriptionPericarditis Score0NormalPericardium is thin and transparent, without exudate. The cardiac surface is smooth and shows no adhesions.1MildPericardium is slightly thickened with minimal translucent or milky-white fibrinous exudate covering <30% of the pericardial surface area.2ModeratePericardium is noticeably thickened. Fibrinous exudate covers 30%-60% of the pericardial surface, possibly accompanied by mild adhesions.3SeverePericardium is substantially thickened. Extensive yellowish-white or purulent exudate covers >60% of the surface, with adhesions present between the heart and sternum/liver.Perihepatitis Scoring0NormalLiver capsule is smooth and free of exudate, with uniform coloration.1MildMinimal fibrinous exudate, covering less than 25% of the liver surface area, is present on the liver surface. Localized slight thickening is observed, with no pseudomembrane formation.2ModerateExudate covers 25%-50% of the liver surface, forming patchy pseudomembranes. The liver edge is mildly swollen or congested.3SevereExudate covers >50% of the liver surface. The pseudomembrane is thick and easily detachable. The liver is markedly swollen, exhibits fragile texture, or shows signs of necrosis.Individuals who died during the treatment period were recorded as 7 points.
Statistical analysis
All data are presented as the mean ± standard deviation (SD) from at least three independent biological replicates conducted under identical conditions. Statistical analyses were performed using GraphPad Prism version 9.5. Student's t-test and one-way analysis of variance (ANOVA) were used to assess statistical significance. The significance levels are denoted as follows: *P < 0.05; **P < 0.01; ***P < 0.001. "ns" indicates no statistically significant difference (P > 0.05).
Ethical approval
All study procedures were approved by the Animal Care& Welfare Committee of Guangxi University (approval numbers: GXU-2024-309 and GXU-2024- 310). All efforts were made to minimize animal suffering.
Results
Isolation of Escherichia coli and detection of antibiotic resistance phenotypes
Eight avian-derived Escherichia coli strains were isolated in this study from diseased chicken liver samples collected from farms in Guangxi Province and archived samples preserved in our laboratory. The PCR results are shown in Fig. 1. The essential information for these 8 isolates and 47 previously archived avian-derived E. coli strains in our lab is summarized in Table 3.Fig. 1. Gel electrophoresis analysis of PCR-amplified Escherichia coli phoA gene fragments (720 bp). Lane M, DNA marker (8000 bp); Lane +, positive control strain (E. coli ATCC 25922); Lane -, negative control; Lanes 1-8, avian-origin E. coli E1, E2, E3, E4, Escherichia coli O117:H25_E5, E6, E7 and E8.Fig 1 dummy alt textTable 3Basic information of 55 avian-origin Escherichia coli strains.Table 3: dummy alt textE. colisourceReferenceE. colisourceRefsE1chicken, liverThis studyEC8chicken, fecesZhang et al., 2022E2chicken, liverThis studyEC9chicken, fecesZhang et al., 2022E3chicken, liverThis studyEC10chicken, fecesZhang et al., 2022E4chicken, liverThis studyEC11chicken, fecesZhang et al., 2022Escherichia coli_O117:H25_E5chicken, liverThis studyEC12chicken, fecesZhang et al., 2022E6chicken, liverThis studyEC13chicken, fecesZhang et al., 2022E7chicken, liverThis studyEC15chicken, fecesZhang et al., 2022E8chicken, liverThis studyEC16chicken, fecesZhang et al., 2022ATCC25922laboratoryThis studyEC22chicken, fecesZhang et al., 2022EC002chickenLi et al., 2022EC25chicken, fecesZhang et al., 2022EC003chickenLi et al., 2022EC27chicken, fecesZhang et al., 2022EC005chickenLi et al., 2022EC29chicken, fecesZhang et al., 2022EC008chickenLi et al., 2022EC32chicken, fecesZhang et al., 2022EC013chickenLi et al., 2022EC34chicken, fecesZhang et al., 2022EC016chickenLi et al., 2022EC38chicken, fecesZhang et al., 2022EC017chickenLi et al., 2022EC39chicken, fecesZhang et al., 2022EC018chickenLi et al., 2022EC40chicken, fecesZhang et al., 2022EC019chickenLi et al., 2022EC42chicken, fecesZhang et al., 2022EC020chickenLi et al., 2022EC43chicken, fecesZhang et al., 2022EC022chickenLi et al., 2022EC44chicken, fecesZhang et al., 2022EC023chickenLi et al., 2022EC45chicken, fecesZhang et al., 2022EC037chickenLi et al., 2022EC46chicken, fecesZhang et al., 2022EC046chickenLi et al., 2022EC47chicken, fecesZhang et al., 2022EC2chicken, fecesZhang et al., 2022EC48chicken, fecesZhang et al., 2022EC3chicken, fecesZhang et al., 2022EC50chicken, fecesZhang et al., 2022EC4chicken, fecesZhang et al., 2022EC51chicken, fecesZhang et al., 2022EC5chicken, fecesZhang et al., 2022EC52chicken, fecesZhang et al., 2022EC6chicken, fecesZhang et al., 2022Partial archived strains were from our laboratory’s prior studies (Li et al., 2022b; Zhang et al., 2022b).
Whole-genome sequencing was conducted on one representative strain. The sequencing results were analyzed using the EcOH database to predict the serotype as O117:H25, and the pubmlst database was used to predict the sequence type (ST) as ST57. This Escherichia coli strain was designated Escherichia coli O117:H25_E5, and its complete genome sequence has been deposited in GenBank under accession number PRJNA1335934.
Further analysis of whole-genome sequencing data revealed that Escherichia coli O117:H25_E5 harbors multiple virulence genes associated with APEC, including those involved in iron uptake (irp2, fyuA, iutA, iroN, iucD), serum resistance (iss), adhesion and invasion (tsh), and toxin secretion (ompT). Detailed information for these genes is summarized in Table 4. Based on established criteria in the literature (Rodriguez-Siek et al., 2005; Johnson et al., 2008), the presence of these virulence genes strongly supports classifying this strain as APEC. The remaining 54 E. coli isolates used for host range determination were confirmed as avian-source but were not further characterized for APEC-specific virulence genes.Table 4. Major virulence genes associated with APEC in Escherichia coli O117:H25_E5.Table 4: dummy alt textSymbolStrandStartStopLength (aa)NameGene typeiss-2427272097increased serum survival lipoprotein Issprotein-codingiutA+960111802733ferric aerobactin receptor IutAprotein-codingiucA+38315555574aerobactin synthase IucAprotein-codingiucB+55566503315N (6)-hydroxylysine O-acetyltransferase IucBprotein-codingiucC+65038245580NIS family aerobactin synthetase IucCprotein-codingiucD+82429575444uncharacterized genepseudogeneiroN+56117788725siderophore salmochelin receptor IroNprotein-codingompT-71128065317omptin family outer membrane proteaseprotein-codingirp2–5753763642035yersiniabactin non-ribosomal peptide synthetase HMWP2protein-codingsitA+1810269plus iron/manganese ABC transporter substrate-binding protein SitAprotein-codingtsh+21543481377plus temperature-sensitive protease autotransporter hemagglutininprotein-codingfyuA-4232844349673siderophore yersiniabactin receptor FyuAprotein-codingphoA+84289843471alkaline phosphataseprotein-coding
Antimicrobial susceptibility of Escherichia coli O117:H25_E5 to 24 antibiotics was assessed by the Kirby-Bauer disk diffusion method following CLSI guidelines. The results of the antimicrobial susceptibility testing are shown in Table 5. It can be observed that Escherichia coli O117:H25_E5 exhibited resistance to most β-lactam and aminoglycoside antibiotics.Table 5. Antimicrobial resistance phenotype of Escherichia coli_O117:H25_E5.Table 5: dummy alt textDrug ClassAntimicrobialAgentDisc Potency(µg)Zone Diameter(mm)Interpretationβ-lactamsAmpicillin100RPiperacillin10013.2RAmoxicillin250RImipenem1026.2SCefazolin300RCephalexin307.2RCefuroxime Sodium300RCeftazidime3022.1SCefepime3020.8ICeftriaxone3010.9RCefoperazone7513.5RsulfonamidesTrimethoprim-Sulfamethoxazole250RAminoglycosidesAmikacin3018.1SStreptomycin100RGentamicin1010.3RNeomycin309.8RKanamycin300RTetracyclinesTetracycline309.8RDoxycycline3015.9SMinocycline3017.2SFluoroquinolonesCiprofloxacin523SLevofloxacin520.9SNorfloxacin1024.3SDetermined according to CLSI VET01 standards: S, Susceptible; I, Intermediate; R, Resistant.
Isolation, purification of phages and determination of lytic spectra
Using the aforementioned 55 avian-source Escherichia coli strains as host bacteria, a total of 32 phage strains were isolated from sewage, fecal, and bedding samples collected from farms across Guangxi Province. The information on the isolated phages is presented in Table 6. The lytic spectra of the phages were assessed by the double-layer agar spot assay (Fig. 2). Among them, vB_EcoM_GXW16 exhibited the broadest lytic spectrum, with a host range coverage of 70.91% (39/55), and was therefore selected for subsequent studies.Table 6. Basic information on isolated phages.Table 6: dummy alt textPhageNumber of Susceptible Strains (n=55)Coverage(%)SourceP12240.00NanningP22545.45P511.81ChongzuoP8712.73P93054.55BeihaiP103054.55GuigangP1135.45P122749.09ChongzuoP133258.18QinzhouP142443.63GuigangP1547.27ChongzuovB_EcoM_GXW163970.91GuigangP173258.18BeihaiP203665.45ChongzuoP212443.63GuigangP222952.72QinzhouP233156.36GuigangP2823.63P2935.45P3011.81NanningP3111.81P3247.27P3323.63P3447.27P3511.81P3635.45P3859.09P413156.36P42712.72P433054.54P442952.72P4511.81All phages were isolated from Guangxi Province, China.Fig. 2. Lytic spectrum of bacteriophages presented as a heat map. Lytic activity increases sequentially from 0 to 3,with higher values corresponding to increased plaque clarity. The E5 in the figure refers to Escherichia coli O117:H25_E5. The P16 in the figure refers to vB_EcoM_GXW16.Fig 2 dummy alt text
Morphology of the phage vB_EcoM_GXW16
Phage vB_EcoM_GXW16 formed clear, circular plaques approximately 1 mm in diameter after 12 h of incubation at 37°C on double-layer agar plates (Fig. 3-A). Transmission electron microscopy (TEM) showed that phage vB_EcoM_GXW16 exhibited an icosahedral head with a diameter of 86 ± 3 nm and a contractile tail measuring 87 ± 3 nm in length, along with distinct tail fibers (Fig. 3-B). Based on its morphological features, phage vB_EcoM_GXW16 was classified as a member of the class Caudoviricetes and the family Myoviridae.Fig. 3. Morphological characterization of bacteriophage vB_EcoM_GXW16.(A) Plaque morphology of bacteriophage vB_EcoM_GXW16 on a double-layer agar plate.(B) Transmission electron micrograph of bacteriophage vB_EcoM_GXW16.Fig 3 dummy alt text
Optimal multiplicity of infection (MOI) for phage vB_EcoM_GXW16
When the MOI is 0.001, phage vB_EcoM_GXW16 exhibits optimal proliferation effects on the host bacteria, with a titer of 3.47 × 10⁹ PFU/mL (Fig. 4-A). Thus, the optimal MOI for phage vB_EcoM_GXW16 was determined to be 0.001.Fig. 4. Biological characteristics of bacteriophage vB_EcoM_GXW16(A) Optimal multiplicity of infection (MOI) of bacteriophage vB_EcoM_GXW16. Maximal phage yield was achieved at an MOI of 0.001 following complete host lysis.(B) Adsorption rate.Adsorption of bacteriophage vB_EcoM_GXW16 to Escherichia coli O117:H25_E5 is presented as the percentage of total phage particles adsorbed.(C) One-step growth curve of bacteriophage vB_EcoM_GXW16. Phage titer (PFU/mL) was monitored over 180 min, illustrating characteristic phases of infection: the latent period, rise period, and plateau.(D) Thermal stability assay. The stability of vB_EcoM_GXW16 was assessed after 1 h of exposure to temperatures ranging from 40°C to 80°C. The phage remained stable up to 60°C, while a significant reduction in titer was observed at 70°C and 80°C.(E) Storage stability assay. The phage titer of vB_EcoM_GXW16 remained stable for 4 weeks when stored at room temperature and 4°C.(F) pH stability assay. The stability of vB_EcoM_GXW16 was evaluated across a pH gradient of 1 to 13, with phage titer (PFU/mL) measured at each pH value.Fig 4 dummy alt text
One-step growth curve and adsorption rate
The adsorption rate of the phage vB_EcoM_GXW16 was 25.52% within 2 min, 56.92% within 6 min, 81.43% within 10 min, and 78.98% within 12 min, indicating that adsorption reached saturation at approximately 10 min (Fig. 4-B). These results demonstrate that the phage exhibits a high and rapid adsorption rate. According to the one-step growth curve (Fig. 4-C), phage vB_EcoM_GXW16 exhibited a latent period of 10 min, followed by a steady rise in titer during a burst period lasting 110 min, after which a plateau phase was observed. The burst size was approximately 81 PFU/cell. These data provide evidence for the replication kinetics and lytic activity of phage vB_EcoM_GXW16.
Thermal, pH, and storage stability of phage vB_EcoM_GXW16
The stability of phage vB_EcoM_GXW16 was assessed under varied conditions. Thermal stability tests indicated that vB_EcoM_GXW16 maintained stable activity across a temperature range of 40-60°C. However, the phage titer gradually decreased when incubated at 70°C, and was completely inactivated within 20 min at 80°C (Fig. 4-D). Furthermore, storage experiments showed that vB_EcoM_GXW16 retained stable activity at room temperature and at 4°C (Fig. 4-E), demonstrating its tolerance to ambient conditions and ease of storage, which supports its therapeutic potential for practical applications. The pH stability assay found that phage vB_EcoM_GXW16 was completely inactivated at pH 1 and pH 13. Phage viability declined at pH below 4, while the phage titer remained stable within the pH range of 4 to 12 (Fig. 4-F). These results suggest that vB_EcoM_GXW16 exhibits greater tolerance to alkaline than to acidic environments.
In vitro bacteriostatic activity assay of phage vB_EcoM_GXW16
The results of the antibacterial assay demonstrated that the growth of Escherichia coli O117:H25_E5 was completely inhibited within 4 h of treatment with vB_EcoM_GXW16 at all tested multiplicities of infection (MOIs). The OD_600_ values within 12 h still showed significant differences from the positive control (p < 0.001) (Fig. 5-A). Relative to the control group, the viable bacterial counts in the three phage-treated groups (MOI = 1, 0.1, and 0.01) declined rapidly within 2 h, dropping from 1.55 × 10⁸ CFU/mL to no more than 5.2 × 10⁵ CFU/mL, after which a gradual increase was observed. Although bacterial counts in all treatment groups exceeded 10⁸ CFU/mL by 12 h, they remained significantly lower (p < 0.001) than those of the positive control, (Fig. 5B). These results demonstrate that vB_EcoM_GXW16 significantly inhibits the growth of the host bacterium at MOIs of 1, 0.1, and 0.01, with the most potent bactericidal activity observed at an MOI of 1.Fig. 5. Bacteriostatic effect of phage vB_EcoM_GXW16 ex vivo.(A) OD_600_ measurement of the host bacteria. Phage vB_EcoM_GXW16 significantly inhibited the OD_600_ of the host bacterium Escherichia coli O117:H25_E5 at different multiplicities of infection (MOIs; 1, 0.1, and 0.01) (*** p < 0.001).(B) Bacterial count of the host bacteria. Phage vB_EcoM_GXW16 significantly reduced the bacterial count of the host bacterium Escherichia coli O117:H25_E5 at different MOIs (1,0.1, and 0.01) (*** p < 0.001).Fig 5 dummy alt text
vB_EcoM_GXW16 whole genome analysis
Whole-genome sequencing revealed that vB_EcoM_GXW16 is a double-stranded DNA (dsDNA) phage with a genome length of 170,605 bp and a G+C content of 39.51%. The coding sequences accounted for 95.2% of the entire genome. A total of 267 open reading frames (ORFs) were predicted, with 41 located on the positive strand and the remaining 226 on the negative strand. Among these, 127 ORFs (47.57%) were annotated as genes with known functions, including proteins involved in DNA replication and metabolism, structural/packaging proteins, and host lytic proteins. The remaining 140 ORFs (52.43%) were annotated as hypothetical proteins. No tRNA genes were identified within the genome, indicating that vB_EcoM_GXW16 is completely dependent on the host for protein synthesis. Furthermore, no virulence factors, toxin genes, or antibiotic resistance genes were detected, indicating the potential safety of phage vB_EcoM_GXW16 for potential clinical application (Fig. 6-A). The genome sequence of phage vB_EcoM_GXW16 has been deposited in the GenBank database under accession number PX436236.Fig. 6. Whole-genome analysis of bacteriophage vB_EcoM_GXW16. (A) Circular genome map of phage vB_EcoM_GXW16. The map displays key genomic features, including coding sequences (CDS), regulatory elements, and GC content. (B) Phylogenetic tree constructed from the phage terminase large subunit gene sequences. The tree illustrates the evolutionary relationships between phage vB_EcoM_GXW16 and other related phages, with branch values indicating bootstrap confidence levels and the extent of evolutionary divergence.Fig 6 dummy alt text
A phylogenetic tree was constructed based on the sequence of the phage terminase large subunit to elucidate the evolutionary relationships among different phages (Fig. 6-B). The phylogenetic tree indicates that vB_EcoM_GXW16 (marked in red) is closely related to other Escherichia coli phages and belongs to the genusDhakavirus, family Myoviridae, class Caudoviricetes. This analysis clarifies the phylogenetic placement and evolutionary relationships of vB_EcoM_GXW16 among its bacteriophage relatives.
In vivo therapeutic efficacy of phage vB_EcoM_GXW16
First, through LD_50_ experiments, we determined the LD_50_ of Escherichia coli O117:H25_E5 to be 4.6 × 10⁸ CFU/mL. Using this dose, we successfully established an E. coli infection model in chickens. Infected chickens displayed characteristic clinical signs including lethargy, depression, closed-eye immobility, and wing drooping (Fig. 7-A). Necropsy revealed the peritoneal surfaces were covered with a layer of fibrinous exudate, and adhesions were noted between the heart and the sternum/liver (Fig. 7-B).Necropsy of dead chickens showed varying degrees of liver enlargement, with a layer of fibrinous exudate covering the liver surface,the capsule was thick and easy to peel off. The pericardium is significantly thickened, covered with a large amount of yellowish-white or purulent exudate. The pathological changes are shown in the figure below (Fig. 7-C).Fig. 7. Clinical signs and gross pathological alterations in chickens following infection with Escherichia coli O117:H25 E5 (A) Overall appearance of infected chickens. Clinical manifestations included drooping wings, lethargy, and depression. (B) Abdominal dissection findings in the infected group. The visceral surfaces were covered with fibrinous exudate, and adhesions were observed between the heart and sternum/liver. (C) Comparison of heart, liver, and spleen tissues from infected and normal groups. Characteristic lesions in the infected group comprised pericarditis, perihepatitis, and splenomegaly.Fig 7 dummy alt text
The therapeutic efficacy of phage vB_EcoM_GXW16 against Escherichia coli O117:H25_E5 infection was evaluated in an avian colibacillosis model. Following the application of different treatment plans, the survival rates of all groups were improved to varying degrees. The highest survival rate (83.33%) was observed in chickens treated with 10^8^ PFU/mL of phage for three days, starting at 3 h post-infection. Survival rates of 75% were recorded in the group treated with 10^5^ PFU/mL of the phage for three days starting 3 h post-infection, as well as in the group receiving a single dose of 10^8^ PFU/mL of the phage 3 h prior to infection. Additionally, the phage-only control group showed a 100% survival rate, indicating that vB_EcoM_GXW16 at 10⁸ PFU caused no adverse effects in broilers (Fig. 8-A).Fig. 8. Therapeutic efficacy of phage vB_EcoM_GXW16 in a chick model infected with Escherichia coli O117:H25_E5.(A) The therapeutic effect of phage vB_EcoM_GXW16 on chicks infected with Escherichia coli O117:H25_E5. Survival probability of chicks was monitored over time under experimental conditions. In the E. coli-infected group (red curve), survival rates declined significantly over time. In contrast, phage treatment group D (pink curve), showed significantly improved survival compared to the infected group.(B) Average body weight of surviving chicks at the endpoint of the experiment. Compared with the infected group, surviving chicks in phage treatment groups B, C, and D exhibited a significant increase in body weight (*P < 0.05; **P < 0.01; ***P < 0.001).(C) Bacterial load in the livers of surviving chicks at the end of the study. Phage-treated groups B, C, and D significantly reduced the bacterial burden in the livers compared to the infected group (*P < 0.05; ***P < 0.001).(D) Histopathological examination of cardiac, hepatic, and splenic tissues across experimental groups (HE, 200 ×).Fig 8 dummy alt text
At the end of the experiment, the average body weights of surviving chickens were compared among groups. Compared with the blank control group, chickens in the Escherichia coli-infected group showed a significant reduction in body weight (p < 0.001). In contrast, phage treatment groups B, C, and D exhibited significantly higher body weights than the Escherichia coli-infected group (p < 0.01), indicating that phage vB_EcoM_GXW16 mitigated the weight loss associated with Escherichia coli O117:H25_E5 infection (Fig. 8-B).
At the end of the experiment, viable bacterial counts in liver homogenates were determined. The liver bacterial loads in phage-treated groups B, C, and D were significantly lower than those in the bacterial control group (p < 0.001), demonstrating that phage vB_EcoM_GXW16 effectively reduced the hepatic bacterial burden caused by Escherichia coli O117:H25_E5 (Fig. 8-C). Moreover, the degree of organ damage was assessed by macroscopic lesion scoring, with the results presented in Table 7. Compared with the infection group, the mean pathological score of phage treatment group D was significantly decreased (p < 0.05). (Scores from individuals that died during the treatment period were included in the calculation of the group averages.)Table 7. Clinical lesion scores for pericarditis and hepatic peritonitis.Table 7: dummy alt textGroupAverage score of lesions±SDGroupAverage score of lesions±SDBlank Control0.00±0.00Phage Prophylaxis B3.33±2.90Phage Control0.00±0.00Phage Treated C2.75±2.62Bacterial Control5.33±2.13Phage Treated D2.42±2.69*Phage Prophylaxis A4.42±2.49Antibiotics Treated3.92±2.96The asterisk in the upper right corner of the numbers in the table indicates a statistically significant difference between that group and the E. coli control group.(*p < 0.05).
Furthermore, we examined the pathological changes in the heart, liver, and spleen tissues of chickens from each group (Fig. 8-D). In the bacterial control group, heart tissue exhibited extensive necrosis of the epicardium and subjacent myocardium. Additionally, there was marked proliferation of fibrous connective tissue accompanied by dense infiltration of lymphocytes, macrophages, and heterophils. Liver tissue displayed localized capsular thickening and fibrous connective tissue proliferation, accompanied by considerable lymphocyte infiltration, Extensive hepatocellular edema and cellular swelling were evident, and focal lymphocyte infiltration was commonly observed. In the spleen, necrosis of both red and white pulp cells was frequently observed. In contrast, hearts from phage-treated group C and D exhibited clear epicardial and endocardial structures without significant abnormalities, with only minor infiltration of lymphocytes within the myocardium. The liver showed mild hepatocellular edema and lymphocyte infiltration. No significant pathological alterations were detected in the spleen.
These results demonstrate that phage vB_EcoM_GXW16 effectively alleviated tissue lesions induced by Escherichia coli O117:H25_E5 infection, reduced hepatic bacterial load, and improved the survival rate and average body weight of chickens.
Discussion
Avian pathogenic Escherichia coli (APEC) is a major cause of extraintestinal infections in poultry, leading to significant economic losses worldwide (Goudarztalejerdi et al., 2022). The increasing prevalence of multidrug-resistant (MDR) strains has further complicated the control of these infections (Zych et al., 2024). In this context, phage therapy has garnered considerable attention as a promising alternative for combating colibacillosis (O’Flaherty et al., 2009; Kapoor et al., 2024).
In this study, a drug-resistant APEC strain, Escherichia coli O117:H25_E5, showing resistance to β-lactam and aminoglycoside antibiotics, was isolated. Subsequently, a bacteriophage capable of lysing this E. coli strain, designated vB_EcoM_GXW16, was isolated from sewage at a poultry farm. vB_EcoM_GXW16 exhibited significant bactericidal activity against Escherichia coli O117:H25_E5 both in vitro and in vivo.
The current findings indicate that vB_EcoM_GXW16 exhibits a relatively broad host range, lysing 39 out of 55 avian-source Escherichia coli strains, corresponding to a lytic rate of 70.91%. Although not all isolates have been confirmed as APEC, they all originate from poultry. Its coverage is higher than that of UPEC04, which inhibited only 17.9% of strains (10/56) (Kazibwe et al., 2020b),fEg-Eco19, which infected only two strains (2/137) (Badawy et al., 2022) and ɸEcM-vB1, which infected 51% of strains (33/65) (Abozahra et al., 2025). The broad host range of vB_EcoM_GXW16 may be attributed to its efficient recognition of bacterial surface receptors or to a broader host adaptation capacity acquired through evolution (Subramanian et al., 2022). Phages with a broad host range can more effectively combat drug-resistant infections by lysing several related bacterial strains, thereby reducing the population of resistant bacteria (Viertel et al., 2014; Zhang et al., 2025). These results demonstrate the superior lytic potential of vB_EcoM_GXW16 against drug-resistant strains.
Transmission electron microscopy (TEM) observation and genomic sequencing confirmed that vB_EcoM_GXW16 belongs to the class Caudoviricetes. Previous studies have shown that bacteriophages belonging to the class Caudoviricetes exhibit the highest diversity, prevalence, and distribution among all phages, representing the most common phage group discovered to date (Gulyaeva et al., 2022; Yao et al., 2023b). Phages also serve as one of the most extensive mediators of horizontal gene transfer (HGT) and can act as vectors for antibiotic resistance genes (Tzipilevich et al., 2017). Specifically, phage particles can carry genetic material from donor bacterial cells to recipient cells, thereby facilitating the dissemination of antibiotic resistance and virulence factors (Balcazar, 2014; Touchon et al., 2017). Whole-genome analysis of vB_EcoM_GXW16 identified it as a double-stranded DNA phage with a genome of 170,605 bp. Notably, no genes associated with virulence, toxins, antibiotic resistance, or lysogeny were identified in its genome. This absence significantly reduces the risk of horizontal gene transfer (HGT) during therapeutic application, providing molecular-level assurance of its biosafety and indicating its safety for clinical use as a bacterial control agent. vB_EcoM_GXW16 exhibits outstanding biological characteristics. The optimal MOI of phage vB_EcoM_GXW16 was 0.001, a value lower than those reported for several other APEC-specific phages, including vB_EcoM_P3322 with an MOI of 1 (Huang et al., 2025) and phage CABI-SEA 2301 with an MOI of 0.01 (Bilhman et al., 2025a). A lower MOI indicates that a smaller inoculum of phage vB_EcoM_GXW16 can generate a substantial progeny burst, as it allows sufficient time for productive infection cycles without excessive inter-phage competition, thereby maximizing progeny yield. In contrast, a high MOI often results in abortive infection, where simultaneous infection by multiple phages leads to rapid host cell death and consequently reduced viral production (Brown and Bidle, 2014). Adsorption, the process by which a phage attaches to specific receptors on the host bacterial cell surface, constitutes the initial and critical step of phage infection (Yao et al., 2023c; Oluwarinde et al., 2024). Approximately 81.43% of vB_EcoM_GXW16 particles adsorbed to the host bacterium within 10 min. This adsorption rate is higher than the 77% adsorption rate reported for phage CABI-SEA 2301 within 10 min (Bilhman et al., 2025b) and is similar to the 84.1% adsorption rate of phage PEC9 within the same timeframe (Yao et al., 2023c), indicating a high and rapid adsorption capacity for vB_EcoM_GXW16. A higher adsorption rate implies that phages can attach to and infect host bacteria more rapidly, which in turn can accelerate bacterial killing during phage therapy (Abedon, 2023).
The latent period, burst size, and stability in diverse environmental conditions are pivotal factors for the therapeutic application of bacteriophages (Huang et al., 2025). Growth curve analysis revealed that vB_EcoM_GXW16 exhibited a latent period of approximately 10 min, followed by a burst period of 110 min, and reached a plateau after 120 min, with a burst size of about 81 PFU per cell. The latent period of vB_EcoM_GXW16 is shorter than the 40-20 min reported for three phages in a study by Śliwka et al. (2025)). Compared with vB_EcoM_JS09 (30 min, 79 PFU/cell) (Zhou et al., 2015) and phage PEC9 (20 min, 68 PFU/cell) (Yao et al., 2023c), vB_EcoM_GXW16 demonstrated a shorter latent period and a larger burst size. These findings indicate that vB_EcoM_GXW16 possesses rapid replication kinetics and high reproductive activity, facilitating the release of a substantial number of progeny phages within a short time frame (Zhu et al., 2024). Regarding environmental adaptability, vB_EcoM_GXW16 retained stability for 60 min at temperatures ranging from 40 to 60°C, and remained stable for four weeks at room temperature and 4°C. Furthermore, its activity remained stable across a pH range of 4 to 12. Compared to the report by Lau GL, vB_EcoM_GXW16 demonstrated superior stability under both acidic/alkaline conditions and temperature variations (Lau et al., 2012). These results suggest that this phage can effectively kill host bacteria and retain its activity under diverse environmental conditions, which is a prerequisite for successful phage application (Karami et al., 2024). However, a marked reduction in the activity of vB_EcoM_GXW16 was observed at temperatures above 70 °C, pH < 4, and pH = 13. This loss of activity is likely attributable to denaturation of the proteinaceous capsid or degradation of the internal nucleic acid structure caused by high temperature, strong acid, or strong alkali, thereby compromising structural integrity and infectivity (Jończyk-Matysiak et al., 2019).
Phage therapy represents a promising approach for combating APEC infections, particularly those caused by multidrug-resistant strains (Yao et al., 2023c).To assess the in vivo therapeutic potential of the phage, we first evaluated the antibacterial activity of vB_EcoM_GXW16 against planktonic cells. At MOIs of 0.01, 0.1, and 1, vB_EcoM_GXW16 significantly inhibited (P < 0.001) the growth of planktonic bacterial cells, demonstrating strong antibacterial efficacy and suggesting its potential to countrolling bacterial infections. Subsequently, a chick model of Escherichia coli infection was successfully established via intraperitoneal injection, and phage prophylaxis and treatment were administered orally. Experimental results indicated that all treatment groups improved the survival rates of infected chickens to varying degrees. Furthermore, a more significant reduction in E. coli loads in the liver, as well as a decrease in the severity of pericarditis and perihepatitis, was observed in the treated chickens. Histopathological examination further confirmed the therapeutic effect of phage vB_EcoM_GXW16. Pathological lesions in the heart, liver, and spleen of chickens in the phage-treated groups were less severe compared to those in the untreated infection group. Importantly, no mortality was observed in the group receiving phage alone, indicating the safety and efficacy of vB_EcoM_GXW16 as an in vivo antibacterial agent and suggesting that phage treatment did not cause significant adverse effects.
In this study, the high-dose phage treatments in Groups B and D exhibited superior efficacy to the low-dose treatments in Groups A and C, likely as an adequate phage titer is essential for effective therapeutic activity. According to Abedon, for phage therapy to be effective, a sufficient concentration of phages must reach the infection site to act on susceptible bacteria. Regardless of bacterial density, a phage density of 10^8^ PFU is considered reasonable to exert prompt antibacterial effects (Abedon, 2016). The experiment by W.E. Huff also confirmed that the highest titer (10^8^ PFU) of phage SPR02 was the only consistently effective phage treatment regimen (Huff et al., 2006). This study included both phage prophylaxis and treatment groups, which together cover the core application scenarios of phage intervention, providing key insights into the optimal timing of phage administration. However, the therapeutic outcomes in the prophylaxis groups did not meet expectations, which may be attributed to multiple mechanisms. First, as a foreign entities, phages may activate the host's innate immunity, leading to their nonspecific clearance and thereby indirectly reducing their lytic efficiency against Escherichia coli (Hodyra-Stefaniak et al., 2015; Dąbrowska and Abedon, 2019). Second, studies have indicated that the host may produce phage-specific neutralizing antibodies. Such anti-phage immunity can accelerate phage clearance from tissues, especially in the absence of host bacteria, and this immune response may further compromise subsequent therapeutic effects (Berkson et al., 2024). Additionally, prophylactic administration often coincides with low bacterial loads, which may be insufficient to support robust phage replication. Consequently, phages may fail to undergo normal lytic cycles and instead be cleared or inactivated, reducing prophylactic effectiveness (Stone et al., 2019; Stacey et al., 2022). Similarly, previous studies have also indicated that when the target bacterial load is low, effective in vivo phage amplification is difficult to achieve, leading to suboptimal therapeutic outcomes (Sarker et al., 2016).
In this study, we used the intraperitoneal route to establish the APEC infection model in chickens. This route was chosen based on previous reports demonstrating that intraperitoneal inoculation consistently reproduces the systemic lesions characteristic of avian colibacillosis, including pericarditis, perihepatitis, and bacterial colonization of internal organs (Shi et al., 2023). Compared to other routes such as intratracheal or intravenous administration, intraperitoneal injection has been shown to produce more reproducible pathology and higher mortality rates without the inconsistency or lameness observed with intravenous challenge (Cox et al., 2021). Meanwhile, this route has been extensively used in recent APEC infection models for evaluating therapeutic efficacy (Yuan et al., 2020; Mao et al., 2025).
Despite these advantages, the use of the intraperitoneal route for bacterial challenge still has limitations. APEC infection typically initiates as a localized infection of the air sacs, commonly known as airsacculitis or air sac disease, which may subsequently spread to other internal organs and lead to systemic infection (Dho-Moulin and Fairbrother, 1999). The intraperitoneal route bypasses the natural mucosal barriers and the early stages of host-pathogen interaction, potentially affecting disease progression, immune responses, and the subsequent evaluation of therapeutic efficacy (Antão et al., 2008). Furthermore, compared with respiratory acquired infections, phages administered orally may exhibit distinct pharmacokinetic and pharmacodynamic characteristics in intraperitoneally induced infection models. Therefore, while the intraperitoneal injection model provides a reproducible and well-established system for evaluating therapeutic interventions, caution is warranted when extrapolating our findings to field conditions.
Meanwhile, the therapeutic efficacy of phages against APEC infections can vary significantly depending on the infection and administration routes used. Tsonos et al. (2014)) reported that a cocktail of four APEC specific phages, despite strong in vitro activity, failed to achieve therapeutic success in an intratracheal infection model, regardless of the administration route. However, Moreno et al. (2025)) demonstrated that oral administration of the UPWr_E124 phage cocktail effectively reduced bacterial loads in the lungs, bursa of Fabricius, and blood of chickens infected via the respiratory route, a finding that directly supports the relevance of orally administered phages for respiratory infections and aligns with our study using oral phage administration.
In summary, these studies indicate that future research should employ more physiologically relevant infection routes and administration methods to validate the therapeutic potential of phages under conditions closer to natural disease transmission.
This study has several limitations that should be acknowledged. First, while the challenge strain Escherichia coli O117:H25_E5 was confirmed as APEC by whole-genome sequencing, the other 54 E. coli isolates used for host range determination were not screened for APEC specific virulence genes. Therefore, the host range data presented in this study should be interpreted as the lytic activity of phage vB_EcoM_GXW16 against a panel of avian-source E. coli isolates, rather than exclusively against confirmed APEC strains. Future studies should include molecular characterization of all strains using accepted criteria to definitively establish their APEC status. Second, the prophylactic efficacy of phage vB_EcoM_GXW16 did not meet expectations, and the underlying mechanisms warrant further investigation. Despite these limitations, the robust in vitro and in vivo efficacy of vB_EcoM_GXW16 against a confirmed APEC strain supports its potential as a therapeutic candidate for controlling avian colibacillosis.
Conclusion
In this study, a novel myoviridae Escherichia coli phage, vB_EcoM_GXW16, was isolated and characterized. The phage displayed broad-spectrum lytic activity against a panel of avian-source E. coli isolates, potent antibacterial efficacy against a confirmed APEC strain, and a favorable safety profile. In vitro assays demonstrated its significant inhibitory effect on host bacterial growth. In vivo trials further confirmed that the phage could effectively alleviate symptoms of Escherichia coli infection in chickens and significantly enhance their survival rate. These findings indicate that phage vB_EcoM_GXW16 exhibits promising therapeutic potential against drug-resistant Escherichia coli infections and could serve as a potential antibiotic alternative for the control of avian colibacillosis.
CRediT authorship contribution statement
Ting Xu: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Data curation, Conceptualization. Wenwen Yang: Writing – review & editing. Jia Cao: Software, Resources, Formal analysis. Xiaofang Wei: Validation, Supervision, Investigation, Formal analysis. Huixin Liu: Validation, Supervision, Investigation, Formal analysis. Yiming Li: Validation, Supervision, Investigation, Formal analysis. Sijia Pan: Validation, Supervision, Investigation, Formal analysis. Nihar Ali: Validation, Supervision, Investigation, Formal analysis. Hongbin Si: Writing – review & editing, Writing – original draft, Funding acquisition, Conceptualization.
Disclosures
The authors of the manuscript entitled “Isolation and Therapeutic Potential of vB_EcoM_GXW16, a Broad-Spectrum Phage Targeting Drug-Resistant Avian Pathogenic Escherichia coli” hereby declare that there are no conflicts of interest to disclose. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. All authors have directly participated in the planning, execution, and analysis of this study. The authors have read and approved the final version submitted. The materials within have not been and will not be submitted for publication elsewhere. There were no relationships with any entities that could be perceived to influence, or that give the appearance of potentially influencing, what was written in the submitted work. The authors affirm that they have no financial affiliation (e.g., employment, direct payment, stock holdings, retainers, consultantships, patent licensing arrangements, or honoraria), or involvement with any commercial organization with direct financial interest in the subject or materials discussed in this manuscript, nor have any such arrangements existed in the past three years. Any other potential conflicts of interest have been disclosed to the best of our knowledge.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Abedon S.T.Phage therapy dosing: the problem(s) with multiplicity of infection (MOI)Bacteriophage 62016 e 122034810.1080/21597081.2016.1220348 PMC 505677927738558 · doi ↗ · pubmed ↗
- 2Abedon S.T.Bacteriophage adsorption: likelihood of virion encounter with bacteria and other factors affecting rates Antibiotics 122023 Available athttps://www.mdpi.com/2079-6382/12/4/723(verified 28 November 2025)10.3390/antibiotics 12040723 PMC 1013536037107086 · doi ↗ · pubmed ↗
- 3Abozahra R.Shlkamy D.Abdelhamid S.M.Isolation and characterization of ɸEc M-v B 1 bacteriophage targeting multidrug-resistant Escherichia coli BMC Res. Notes 18202533975415410.1186/s 13104-024-07033-x PMC 11699686 · doi ↗ · pubmed ↗
- 4Acharya S.Chinnasamy T.Therapeutic potential of ECP-SA 1: a novel lytic phage against multidrug-resistant Escherichia coli Microb. Pathog.208202510795610.1016/j.micpath.2025.10795640769226 · doi ↗ · pubmed ↗
- 5Ackermann H.-W.Basic phage electron microscopy Methods Mol. Biol. Clifton NJ 501200911312610.1007/978-1-60327-164-6_1219066816 · doi ↗ · pubmed ↗
- 6Adriaenssens E.Brister J.R.How to name and classify your phage: an informal guide Viruses 92017702836835910.3390/v 9040070 PMC 5408676 · doi ↗ · pubmed ↗
- 7Al-Zubidi M.Widziolek M.Court E.K.Gains A.F.Smith R.E.Ansbro K.Alrafaie A.Evans C.Murdoch C.Mesnage S.Douglas C.W.I.Rawlinson A.Stafford G.P.Identification of novel bacteriophages with therapeutic potential that target Enterococcus faecalis Infect. Immun.87201910.1128/iai.00512-19PMC 680332531451618 · doi ↗ · pubmed ↗
- 8Antão E.-M.Glodde S.Li G.Sharifi R.Homeier T.Laturnus C.Diehl I.Bethe A.Philipp H.-C.Preisinger R.Wieler L.H.Ewers C.The chicken as a natural model for extraintestinal infections caused by avian pathogenic Escherichia coli (APEC)Microb. Pathog.4520083613691884898010.1016/j.micpath.2008.08.005 · doi ↗ · pubmed ↗