The phage was recovered from sewage water collected in Karabük, Türkiye, using the standard laboratory strain E. coli ATCC 25922 as the indicator and propagation host. After enrichment and four consecutive rounds of single-plaque purification, the researchers obtained a homogeneous phage preparation producing circular, clear plaques with well-defined edges. Transmission electron microscopy of negatively stained particles revealed a polyhedral head roughly 45 nanometers in diameter and a long, flexible, non-contractile tail of about 176 nanometers, giving a total particle length of approximately 220 nanometers. The absence of a contractile sheath places KBU_SV1 firmly in the siphovirus morphotype within the class Caudoviricetes, the broad group of tailed double-stranded DNA phages that dominates the bacterial virosphere.

Host-range testing against a panel of 106 bacterial strains revealed a pattern of selective rather than broad lytic activity. Spot assays produced visible clearance on 16 strains, corresponding to a host range index of 0.15. Notably, clearance was detected on 7 of 20 carbapenem-resistant E. coli clinical isolates and 4 of 20 carbapenem-susceptible E. coli isolates, meaning that lysis was actually more frequent among the resistant strains. Four of 40 carbapenem-resistant Klebsiella pneumoniae isolates also showed spot clearance, while none of the 20 Enterococcus faecalis clinical isolates or the other reference strains tested were affected. The researchers caution that spot clearance alone does not prove productive infection, since clearing can arise through mechanisms that do not generate progeny phages, such as lysis from without, and efficiency-of-plating measurements will be needed to confirm true susceptibility, particularly for K. pneumoniae.

In liquid culture, KBU_SV1 suppressed the growth of E. coli ATCC 25922 across the entire range of multiplicities of infection tested, from 100 down to 0.01. During the first 14 hours of incubation, optical density in all phage-treated cultures remained below 0.10, a reduction of at least 93 percent relative to the untreated control, which reached an optical density of 1.70 by 24 hours. Regrowth was eventually observed under all conditions, a common feature of phage-bacterium systems that is often attributed to the emergence of phage-resistant subpopulations. Interestingly, the late growth response showed an apparent non-monotonic relationship with the initial phage dose: cultures treated at multiplicities of 0.1 and 0.01 ended up with lower optical densities at 24 hours than those treated at 10 and 1. The authors emphasize that the mechanism behind this pattern remains unresolved and that it should not be taken as evidence that lower doses are generally superior.

Physicochemical stability is a key practical consideration for any phage envisaged as a therapeutic candidate, and here KBU_SV1 performed well. The phage retained infectivity after one hour of incubation across a pH range from 4 to 12, with only modest declines of roughly 0.25 log units at the extremes of that interval compared with the neutral pH reference. No infectivity was detected at pH 2, however, a point worth noting given the acidic environment of the stomach. Temperature testing showed stable titers at 4, 23 and 37 degrees Celsius, but exposure to 65 degrees Celsius eliminated detectable infectivity. Taken together, these results suggest that KBU_SV1 is robust under moderate conditions relevant to storage, formulation and physiological temperature, although long-term storage stability was not evaluated.

Biofilms, structured bacterial communities embedded in an extracellular polymeric matrix, are a major driver of treatment failure because they dramatically reduce susceptibility to both antibiotics and immune defenses. KBU_SV1 demonstrated meaningful antibiofilm activity in crystal violet assays. Treatment of established biofilms of E. coli ATCC 25922 reduced biomass by 84.5 to 86.0 percent across all tested multiplicities of infection, while inhibition of biofilm formation reached 78.0 to 83.7 percent. Against the carbapenem-resistant clinical isolate Ec59, the effects were smaller but still substantial, with biomass reductions of 22.3 to 54.2 percent and inhibition of formation ranging from 24.6 to 57.5 percent, with the strongest effects at the highest multiplicity. The authors note that crystal violet staining measures total attached biomass rather than viable cells, so these findings reflect reductions in biofilm mass rather than proof of bacterial killing or complete eradication.

The in vivo experiments used the greater wax moth Galleria mellonella, a widely used invertebrate infection model. Larvae infected with a lethal dose of E. coli ATCC 25922 died rapidly, with 40 percent mortality at 12 hours and complete mortality by 18 hours. Larvae that received phage treatment one hour after infection survived significantly longer: mortality was distributed across the 48-hour observation period, and 40 percent of treated larvae were still alive at 24 hours. Kaplan-Meier survival analysis confirmed a statistically significant delay in mortality compared with the infection control, with a log-rank test yielding a p-value of 1.2 × 10⁻⁷. Importantly, no deaths occurred in larvae receiving phage alone or mock injections, indicating no detectable toxicity of the preparation under the tested conditions. The authors stress, however, that the phage delayed rather than prevented death, and that G. mellonella does not fully reproduce mammalian physiology, so these results should be viewed as preliminary evidence supporting further investigation rather than demonstration of therapeutic efficacy.

Genomic sequencing provided the most reassuring safety-relevant findings. The KBU_SV1 genome is a 45,190-base-pair double-stranded DNA molecule with a G plus C content of 53.56 percent, confirmed as complete by CheckV with 100 percent estimated completeness, zero contamination and a high-confidence 272-base-pair direct terminal repeat at both ends. Sequencing depth averaged nearly 24,000-fold with complete positional coverage. The genome encodes 71 putative open reading frames organized into five functional modules covering DNA and RNA replication, DNA metabolism, structural proteins, assembly and lysis, and contains no tRNA genes. Critically, screening with ResFinder, VirulenceFinder, the Comprehensive Antibiotic Resistance Database and the Virulence Factor Database detected no known antimicrobial resistance genes, no virulence determinants and no lysogeny-associated genes such as integrases, excisionases or phage repressors. Machine-learning-based lifestyle prediction tools PhageAI and PhageLeads both classified the phage as virulent, meaning strictly lytic, a property generally considered desirable for therapeutic candidates because it rules out the temperate lifestyle that could mediate horizontal gene transfer.

Comparative genomics clarified where KBU_SV1 fits in the phage taxonomy. BLASTn searches revealed 99.12 percent nucleotide identity with Escherichia phage EPIMAM03, and VIRIDIC intergenomic similarity calculations showed values above the 95 percent species-level threshold with several related genomes, including Dhillonvirus LM40P1 at 98.7 percent. Independent taxonomic analysis with taxMyPhage assigned KBU_SV1 to the genus Dhillonvirus within the class Caudoviricetes, and phylogenetic trees based on the large terminase subunit and whole-proteome comparisons were congruent with this placement. The complete genome has been deposited in GenBank under accession number PZ050807.1. Rather than representing an entirely new species, KBU_SV1 belongs to an existing species-level genomic cluster, which the authors note is consistent with contemporary genome-based approaches to phage classification.

The study has limitations that the authors themselves enumerate candidly. Efficiency-of-plating measurements were not performed on the spot-test-positive isolates, one-step growth experiments to determine adsorption kinetics, latent period and burst size were not carried out, the mechanism of bacterial regrowth was not investigated, and the DNA packaging mechanism implied by the direct terminal repeat was not experimentally resolved. The growth-reduction and in vivo experiments used the reference strain rather than carbapenem-resistant clinical isolates, and endotoxin levels in the phage preparation were not quantified. Nevertheless, the overall profile of KBU_SV1, combining lytic activity against carbapenem-resistant E. coli, broad pH and temperature stability, antibiofilm effects, delayed mortality in an infection model and a genome apparently free of resistance, virulence and lysogeny genes, marks it as a promising candidate for further development, whether deployed alone or as a component of phage cocktails. Before any clinical application, the authors conclude, future work must address productive host range, replication parameters, resistance development, optimized dosing and efficacy in mammalian infection models.

Subject of Research: Isolation and genomic characterization of the bacteriophage KBU_SV1 as a candidate therapeutic against carbapenem-resistant Escherichia coli

Article Title: Biological and genomic characterization of Escherichia phage KBU_SV1 active against carbapenem-resistant Escherichia coli

Article References: Biological and genomic characterization of Escherichia phage KBU_SV1 active against carbapenem-resistant Escherichia coli. (n.d.). https://doi.org/10.3389/fmicb.2026.1927801

Image Credits: AI Generated

DOI: 10.3389/fmicb.2026.1927801

Keywords: bacteriophage, phage therapy, carbapenem-resistant E. coli, antimicrobial resistance, Dhillonvirus, biofilm, Galleria mellonella, Caudoviricetes, siphovirus, phage genome, antibiofilm activity, wastewater