Two Novel Enterobacter Phages Show EDTA-Boosted Antibiofilm Activity

Bacteriophage research has taken a notable step forward with the discovery and characterization of two novel lytic viruses capable of attacking multidrug-resistant bacteria from the Enterobacter cloacae complex, one of the more troubling groups of hospital-associated pathogens. The newly isolated phages, named vB_EhoIP_HHH and vB_EluM_RZH, were recovered from a stretch of the River Chelt in England that had recently received sewage discharge through an overflow pipe, an environment in which the constant arms race between bacteria and their viruses concentrates particularly formidable phage populations. What makes this work especially compelling, however, is not merely the isolation of two new viruses but the way the research team paired them with EDTA, a common chelating agent, and mapped exactly where that combination helps and where it hurts.

The Enterobacter cloacae complex, which includes species such as E. hormaechei and E. ludwigii, has become a significant cause of nosocomial and community-acquired infections ranging from soft tissue infections and osteomyelitis to urinary tract infections, respiratory infections and septicemia. These organisms frequently carry extended-spectrum beta-lactamase and carbapenemase genes, including blaKPC, blaNDM, blaOXA, blaVIM and blaIMP, and they overexpress the AcrAB-TolC efflux pump system, making them difficult to treat with conventional antibiotics. Phage therapy has been explored over the past decade as an alternative or adjunct, but phage-resistant bacteria can emerge, and while phage cocktails broaden the lytic range and slow the rise of resistant populations, they do not fully suppress the re-emergence of resistant cells. Combining phages with antibiotics has shown promise, yet synergy often collapses when biofilms are mature, dosing is poorly timed, or the antibiotic is antagonistic to the phage.

To isolate the phages, the researchers processed sewage-contaminated river water, centrifuging samples to remove debris, enriching them with calcium chloride and magnesium sulfate, and incubating them with laboratory cultures of E. hormaechei and E. ludwigii grown to exponential phase. Chloroform lysis, polyethylene glycol precipitation, and filtration through 0.22-micron membranes yielded purified phage stocks. Electron microscopy then revealed their distinct morphologies: phage HHH belongs to the Autographiviridae family with podovirus-like architecture, possessing a short non-contractile tail, a head of roughly 39 by 42 nanometers, and a tail length of about 12 nanometers, while phage RZH displays myovirus-like features with a longer contractile tail measuring approximately 73 nanometers and a head of about 47 by 52 nanometers. Both produced clear, lytic plaques, confirming their virulent lifestyle.

Genomic sequencing on the Illumina NovaSeq platform revealed sharply different genome architectures. Phage HHH carries a linear double-stranded DNA genome of 39,582 base pairs with 51.2 percent GC content and 63 predicted open reading frames, including the DNA-directed RNA polymerase that is a hallmark of T7-like phages, allowing transcription independent of the host machinery. Phage RZH is far larger at 174,197 base pairs with 39.4 percent GC content and 314 coding sequences, encoding an arsenal characteristic of T4-like myophages: DNA replication and repair proteins, topoisomerases, ribonucleotide reductase subunits, thymidylate synthase, extensive baseplate and tail fiber proteins, and a sophisticated lysis module with holin, spanin components and lysis-inhibition proteins. Notably, RZH also carries phage-encoded tRNAs matching its preferred codons, an adaptive feature that may help the virus overcome host translational constraints and sustain late-stage replication. Safety screening through multiple bioinformatic databases detected no antimicrobial resistance genes, toxins or virulence factors in either genome, and no integrases or repressors were found, supporting their suitability for further therapeutic development.

Phylogenetic analysis using BLASTn, the VICTOR genome-based distance phylogeny tool, and VIRIDIC intergenomic similarity calculations confirmed that both phages are novel species. Phage RZH clusters with Enterobacter-infecting phages including vB_ECC_CW742 and CC31, but no pairwise comparison exceeded the 95 percent nucleotide identity threshold for species demarcation, placing it as a novel species within the genus Karamvirus. Phage HHH shares up to 84.2 percent intergenomic similarity with its closest relative, Escherichia phage vB_EcoP_SSK1, and was classified as a novel species within the genus Kayfunavirus. Host range testing showed complementary lytic profiles: HHH was most effective against E. ludwigii and several ECC isolates, while RZH lysed E. hormaechei and other complex members. The multidrug-resistant ECC strain BCU-17 proved susceptible to both phages, making it the ideal shared host for subsequent combination testing.

Growth kinetics favored practical deployment. At a multiplicity of infection of 0.01, both phages reached titers of 10^9 plaque-forming units per milliliter. Phage HHH showed a latent period of 24 minutes and a burst size of 60.8 virions per infected cell, while RZH had a 30-minute latent period and a burst of 73.5 virions per cell. Adsorption was rapid and efficient: 95 percent of HHH particles attached to their host within ten minutes with an adsorption constant of 1.5 × 10^-8 milliliters per cell per minute, and 93 percent of RZH particles adsorbed with a constant of 1.3 × 10^-8. Both phages remained stable at temperatures between 4 and 37 degrees Celsius and at pH values between 5 and 8, with peak survival at pH 7, though viability declined sharply above 45 degrees Celsius and under strongly acidic or alkaline conditions.

The centerpiece of the study was a checkerboard assay combining the HHH/RZH cocktail with EDTA concentrations from 0.5 to 2.25 millimolar against 24-hour and 48-hour biofilms of the MDR ECC strain. Crystal violet staining showed biomass reductions of up to 59.5 percent at 24 hours, achieved with phage at MOI 10 combined with 0.5 millimolar EDTA. Bliss independence analysis, computed across the full checkerboard matrix using SynergyFinder+, revealed a crucial nuance: while localized synergy appeared in a narrow window around MOI 10 with 0.5 to 1.0 millimolar EDTA, the overall interaction across all dose combinations was significantly antagonistic, with mean Bliss scores of -6.13 for 24-hour biofilms and -7.72 for 48-hour biofilms. Higher EDTA concentrations, particularly at or above 1.25 millimolar, shifted the interaction toward additivity or outright antagonism.

Colony-forming unit enumeration on selected combinations confirmed the biological significance of the biomass data. The MOI 10 plus 0.75 to 1.0 millimolar EDTA combination reduced viable bacteria in 24-hour biofilms to approximately 5.6 log10 CFU, a 1.4 log10 reduction from the untreated 7.0 log10 CFU control, while 48-hour biofilms showed a more modest 0.8 log10 reduction, reflecting the greater tolerance of mature biofilms. Mechanistically, the biphasic EDTA effect is plausible: at low concentrations EDTA chelates calcium and magnesium ions that stabilize the extracellular polymeric matrix and destabilizes the lipopolysaccharide layer of Gram-negative outer membranes, increasing matrix porosity and phage receptor access, but at higher concentrations excessive chelation may deplete the free metal ions that phage replication and host interaction depend upon. Interestingly, MOI 10 outperformed MOI 100, consistent with host-depletion trade-offs and lysis-from-without phenomena in which an overabundance of virions overwhelms individual cells before progeny can be produced.

Safety data rounded out the translational picture. Lactate dehydrogenase release assays on human dermal fibroblasts showed minimal membrane damage from phage cocktail, EDTA, or the combination at 2 and 4 hours, with values comparable to negative controls and significantly lower than bacteria-infected positive controls. In a fibroblast infection model, treatment with MOI 10 plus 0.75 millimolar EDTA reduced recoverable E. cloacae from approximately 6.4 to 5.1 log10 CFU at 2 hours and from 6.5 to 4.6 log10 CFU at 4 hours, corresponding to reductions of 1.3 and 1.9 log10 respectively. These findings collectively establish MOI 10 with 0.75 millimolar EDTA as the most promising condition, and they carry an important lesson for formulation design: EDTA acts as a dose-sensitive adjunct rather than a universally beneficial enhancer of phage activity, meaning that concentration windows must be carefully defined for any phage-chelator product aimed at wound-associated or device-associated Enterobacter biofilms. Future work, the authors note, should examine EDTA-mediated effects on extracellular polymeric substance structure, divalent-cation availability, phage penetration and calcium and magnesium rescue experiments under high-EDTA conditions to fully resolve how matrix disruption and phage activity interact within established biofilms.