Isolation, characterization, and alginate hydrogel delivery of a Klebsiella aerogenes bacteriophage and its impact on biofilm degradation Authors: Istiaque Zaeem, Shakhinur Islam Mondal, Raphael Kabir Niloy, Nurnabi Azad Jewel, Mohiminur Rahman Mahin, Rita Bhatta, Daniyal Karim, Mohimenul Haque Rolin, Mohsina Khatoon, Arzuba Akter Published: 2026 ABSTRACT Highlights•Lytic phage KA_SGEB_01 targets multidrug-resistant Klebsiella aerogenes.•WGS confirms a safe genomic profile lacking virulence and AMR genes.•Phage KA_SGEB_01 achieves 60% biomass reduction in mature biofilms.•Stable across physiological temperature and pH conditions.•Alginate hydrogels enable sustained phage release and 14-day stability.Keywords: Bacteriophage, Phage therapy, Antimicrobial resistance, K. aerogenes, Phage delivery, Hydrogel Introduction Antimicrobial resistance (AMR) represents one of the most critical threats to global public health. In 2021, bacterial AMR was associated with an estimated 4.71 million deaths globally, of which 1.14 million deaths were directly attributable to AMR. Without effective interventions, forecasts project that AMR-associated deaths could rise to 8.22 million annually by 2050, with 1.91 million deaths directly attributable to AMR (Naghavi et al., 2024; Okeke et al., 2024). This crisis is particularly acute in chronic wound management, where approximately 6.5 million patients annually in the United States incur treatment costs exceeding $25 billion (Sen, 2025). Bacterial colonization and biofilm formation occur in approximately 80% of chronic wounds, representing primary impediments to healing (Malone et al., 2017; Zhao et al., 2016). The challenge is compounded by a severely limited antibiotic pipeline, with only six new antibacterial agents receiving FDA approval between 2017 and 2021, necessitating urgent development of alternative antimicrobial strategies (Theuretzbacher et al., 2020). Among the pathogens driving this crisis, multidrug-resistant (MDR) K. aerogenes has emerged as a high-priority clinical threat within the broader ESKAPE group of organisms. Formerly classified as Enterobacter aerogenes, the species was reassigned to the genus Klebsiella on the basis of genome-based phylogeny (Feng et al., 2024; Morgado et al., 2024; Sakalauskienė et al., 2025). Recent global genomic surveys reveal that K. aerogenes is taxonomically more diverse than previously recognized, comprising at least four phylogroups and several internationally distributed high-risk clonal lineages, most notably ST4 and ST93 (Feng et al., 2024; Morgado et al., 2024). Although its epidemiological characterization remains less extensive than that of Klebsiella pneumoniae, the organism is clinically significant in its own right, frequently implicated in healthcare-associated pneumonia, bloodstream infection, urinary tract infection, and device-associated infection, particularly among ICU and immunocompromised patients (Ranta et al., 2024; Rodrigues et al., 2024). National surveillance in Japan identified K. aerogenes as the leading carbapenem-resistant Enterobacteriaceae (CRE) pathogen from 2017 onward, accounting for approximately 40% of CRE cases (Kamio and Espinoza, 2022). While genomic surveillance in China and clinical isolates from Brazilian ICUs have documented convergence of blaKPC and blaNDM carbapenemase genes within the same MDR background, underscoring an international trajectory of resistance dissemination rather than isolated local emergence (X. Li et al., 2024; Rodrigues et al., 2024). This organism utilizes a multifaceted arsenal of resistance mechanisms, including carbapenemases (blaKPC, blaNDM, blaIMP, blaVIM, blaGES), extended-spectrum β-lactamases, AmpC dysregulation, porin loss, efflux pumps, and robust biofilm formation (Feng et al., 2024; X. Li et al., 2024; Morgado et al., 2024; Rodrigues et al., 2024). Notably, carbapenem resistance in K. aerogenes frequently arises independently of carbapenemase acquisition, instead emerging from chromosomal β-lactamase dysregulation combined with permeability defects such as omp36 alterations (Morgado et al., 2024). In burn infections, where K. aerogenes is frequently isolated, MDR organisms are associated with mortality rates exceeding 50% despite aggressive therapy (Azzopardi et al., 2014; Roy et al., 2024). The organism's thick capsular polysaccharides serve dual protective functions: they impede phagocytosis and create biofilm matrices that prevent antimicrobial penetration, rendering conventional antibiotics ineffective (Intra et al., 2023). These features collectively highlight the inadequacy of conventional therapies and the necessity for innovative approaches capable of targeting both antimicrobial resistance and biofilm-associated virulence. Bacteriophage therapy has re-emerged as a promising alternative in this regard. Lytic phages undergo destructive replication cycles, releasing progeny phages capable of infecting neighbouring cells in a self-amplifying manner proportional to bacterial density (Hesse and Adhya, 2019; Pelfrene et al., 2016; Strathdee et al., 2023). Their exquisite host specificity minimizes collateral damage to commensal microbiota, addressing a critical limitation of broad-spectrum antibiotics. The phage works specific to K. aerogenes remains comparatively limited; however, it includes two mechanistically distinct candidates. Phage pK4–26, a lytic podoviral phage of the genus Przondovirus isolated from hospital sewage, displayed serotype-specific, depolymerase-mediated capsule degradation and biofilm disruption, and reduced mortality and inflammation in a murine pneumonia model (Cui et al., 2023). A second isolate, fENko-Kae01, is a lytic jumbo myovirus with a 360 kb genome that exploits flagellum-dependent adsorption rather than capsular receptors, illustrating an alternative route for phage targeting despite its narrow host range and susceptibility to resistance emergence (Ranta et al., 2024). More broadly, phages infecting Klebsiella spp. span a wide taxonomic range, including members of Ackermannviridae, Autographiviridae, Demerecviridae, Drexlerviridae, and Straboviridae,with siphoviral and podoviral morphotypes also represented (Cui et al., 2023; Le Bris et al., 2025; Townsend et al., 2021; Zaki et al., 2023). Among these, capsule depolymerase-bearing phages and multi-phage cocktails have been the most extensively used as therapeutic agents, with cocktail strategies generally outperforming single phages in broadening capsule-type coverage and delaying the emergence of phage resistance (H. Chen et al., 2024; Townsend et al., 2021; Yoo et al., 2024). Importantly, phage-encoded depolymerases can degrade bacterial capsules and biofilm matrices, key structural components underlying K. aerogenes pathogenicity, thereby enhancing bacterial susceptibility (Oliveira et al., 2017; Pires et al., 2022; Tyagi et al., 2024). Clinical applications, including compassionate use cases, have demonstrated encouraging outcomes, with approximately 80% of patients showing improvement in severe MDR infections, and bacterial eradication rates near 61% across a multinational retrospective series, particularly when phage therapy was combined with concomitant antibiotics (Corbellino et al., 2020; Leitner et al., 2017; Pirnay et al., 2024; Uyttebroek et al., 2022). The transition of phage therapy from experimental success to standardized clinical practice is hindered by significant delivery obstacles. Phages are susceptible to environmental degradation, often fail to maintain effective concentrations at infection sites, and typically require repeated administration, particularly in topical applications (Loc-Carrillo and Abedon, 2011). To address these limitations, hydrogel-based delivery systems have gained increasing attention as a means of enhancing phage stability and therapeutic efficacy. Hydrogels are three-dimensional, cross-linked polymer networks capable of protecting embedded phages from environmental stressors. Hydrogels enable sustained and controlled release and additionally, their inherent ability to maintain a moist wound environment and conform to irregular surfaces makes them particularly suitable for wound care applications (González-Menéndez et al., 2018; Zhao et al., 2020). Advanced hydrogel systems can further incorporate stimuli-responsive features, allowing phage release to be triggered by infection-associated conditions such as changes in pH, temperature, or enzymatic activity (Y. Chen et al., 2023; Mo et al., 2022; Tao et al., 2026). Beyond conventional alginate-based formulations, several advanced platforms have demonstrated relevant utility for phage delivery. Hyaluronic acid–polyethylene glycol (HA-PEG) in situ-forming hydrogels can gel within minutes while sustaining bioactive release for periods exceeding one year, illustrating the extended release windows achievable with click-chemistry-based crosslinking (Ilochonwu et al., 2022). Thermo-responsive systems, including PCL-PEG and Pluronic F-127/HPMC hydrogels, enable injectable, in situ-gelling delivery with tuneable release kinetics. One such formulation maintained sustained phage and colistin release against MDR Acinetobacter baumannii, achieving substantial reductions in both planktonic and biofilm bacterial burden while remaining stable for nine months at 4 °C (Dethe et al., 2022; Mukhopadhyay et al., 2024). Gelatine methacryloyl (GelMA) and HA/GelMA composite hydrogels have additionally supported antibacterial loading, controlled injectability, and infected-wound repair through promotion of angiogenesis and favourable macrophage polarization, though low-energy crosslinking strategies are required to preserve phage viability during gel formation (Wei et al., 2024; Xiong et al., 2023; Xue et al., 2022). Preclinical studies have already begun to validate this integrated approach. Formulations using chitosan-alginate or gelatine methacryloyl have demonstrated superior bacterial reduction and accelerated wound closure compared to conventional antibiotic treatments (Shafigh Kheljan et al., 2023; Yan et al., 2021). By providing targeted, sustained antimicrobial activity, such platforms hold the potential to drastically reduce the hospitalizations and surgical interventions that currently cost healthcare systems between $6000 to $96,000 per wound episode (Strathdee et al., 2023). Building upon this technological synergy, we hypothesized that a potent lytic, depolymerase-producing bacteriophage encapsulated within an alginate hydrogel matrix would retain its infectivity and depolymerase activity while exhibiting sustained release, and that this combined platform would achieve significant antibiofilm efficacy against MDR K. aerogenes by inhibition of the biofilm matrix. To test this hypothesis, the present study aimed to develop and evaluate an alginate hydrogel-based delivery system for a lytic bacteriophage targeting MDR K. aerogenes. Specifically, we isolated and genomically characterized a phage with depolymerase activity, optimized its infection parameters and environmental stability, incorporated it into alginate hydrogels with defined release kinetics, and assessed the antibiofilm efficacy of this integrated platform. Materials and methods Bacterial strains and culture conditions A multidrug-resistant (MDR) K. aerogenes strain (MAD946) was obtained from a local hospital as a hospital leftover in 2025 in Sylhet, Bangladesh, following standard protocols for bacterial culture and identification. This strain was used as the primary host for phage isolation and all subsequent experiments due to its clinical relevance and was identified by the laboratory based on colony morphology, Gram staining, and biochemical tests, and further confirmed by 16S rRNA gene sequencing. Antibiotic susceptibility of K. aerogenes MAD946 was determined using the Kirby–Bauer disk diffusion method. Inhibition zone diameters were interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines (34th edition, 2024), as summarized in Table 1. Table 1Antibiotic susceptibility profile of K. aerogenes MAD946.Table 1 dummy alt textAntibioticCLSI Breakpoint (mm)Zone Size (mm)InterpretationAztreonam (ATM)S ≥ 21, I 18–20, R ≤ 1711RErythromycin (E)No breakpoint-R (intrinsically resistant)Doxycycline (DOX)S ≥ 14, I 11–13, R ≤ 10-RTobramycin (TOB)S ≥ 15, I 13–14, R ≤ 128RChloramphenicol (C)S ≥ 18, I 13–17, R ≤ 1237SPenicillin (P)No breakpoint-R (intrinsically resistant)Amoxicillin (AX)No breakpoint-R (intrinsically resistant)Tetracycline (TE)S ≥ 15, I 12–14, R ≤ 11-RAmpicillin (AM)S ≥ 17, I 14–16, R ≤ 13-RNalidixic Acid (NA)S ≥ 19, I 14–18, R ≤ 1313RTrimethoprim (SXT)S ≥ 16, I 11–15, R ≤ 109RCefotaxime (CTX)S ≥ 26, I 23–25, R ≤ 22-RGentamicin (CN)S ≥ 15, I 13–14, R ≤ 12-RCefoxitin (FOX)S ≥ 18, I 15–17, R ≤ 1410RBacitracin (B)No breakpoint-R (intrinsically resistant)Meropenem (MEM)S ≥ 23, I 20–22, R ≤ 1934SVancomycin (VA)No breakpoint-R (intrinsically resistant)Amikacin (AK)S ≥ 17, I 15–16, R ≤ 1415I*S= Susceptible, R=Resistant, I= Intermediate. Antibiotic susceptibility profile of K. aerogenes MAD946. *S= Susceptible, R=Resistant, I= Intermediate. For host range determination, 20 additional bacterial strains were obtained from our laboratory collection and are listed in Supplementary Table S1. These strains were selected based on their clinical relevance, taxonomic diversity, and multidrug-resistance profiles. To assess both intra- and inter-genus specificity of the phage, the panel encompassed taxonomically diverse Gram-negative genera as well as Gram-positive genera. This panel allowed a comprehensive evaluation of the phage's lytic spectrum across phylogenetically and clinically diverse bacterial species that co-circulate in the same hospital environment as the host strain. All strains were routinely cultured in Luria–Bertani (LB) broth or agar at 37 °C under aerobic conditions. This study did not involve human subjects. Clinical samples were anonymized hospital leftovers collected for routine diagnostic purposes; therefore, ethical approval was not required. Bacteriophage isolation, enrichment, and purification Sewage samples were collected from the wastewater treatment outlet of MAG Osmani Medical College Hospital, Sylhet, Bangladesh (Geolocation: 24.90137995448862, 91.85424329698328) over a four-week period in April 2025, using a single-scoop grab sampling strategy under aseptic conditions. For each collection, a 500 mL aliquot was obtained directly from the outlet, transported to the laboratory on ice, and processed within 4 h of collection for bacteriophage isolation. Samples were centrifuged at 10,000 × g for 10 min at 4 °C to remove debris, and the supernatant was filtered through 0.22 μm membrane filters. For enrichment, 10 mL of filtered sewage was mixed with 10 mL of LB broth and 1 mL of overnight bacterial culture, followed by incubation at 37 °C with shaking (180 rpm) for 18–24 h. The enriched culture was centrifuged again under the same conditions, and the supernatant was filtered to obtain phage lysate, which was stored at 4 °C. Phage presence was confirmed using spot assays, where 10 μL of serially diluted lysate was applied onto bacterial lawns and incubated overnight. Clear lysis zones indicated lytic activity. Among the phage isolates recovered, a representative isolate was selected and designated KA_SGEB_01 based on quantitative performance across independent enrichment replicates. The selected isolate produced the highest plaque-forming efficiency, exhibiting the maximum phage titer (PFU/mL) among all recovered isolates, and consistently formed large, clear, well-defined plaques on K. aerogenes MAD946 lawns with no visible turbidity, indicating complete lytic activity without evidence of lysogeny or partial lysis. These characteristics collectively identified KA_SGEB_01 as the most productive phage isolate and thus the most suitable candidate for further characterization. Plaque assay and phage enumeration Double-layer agar (DLA) plaque assays were employed for phage enumeration and plaque morphology characterization (Sharifi et al., 2024). Bacterial host culture in mid-exponential growth phase (OD₆₀₀ = 0.4–0.6) was mixed with appropriate dilutions of phage suspension in 5 mL of molten soft agar (0.4% agar in LB medium maintained at 50 °C). The mixture was immediately overlaid onto pre-solidified LB agar base plates (1.4% agar) and allowed to solidify at room temperature. Plates were incubated in an inverted position at 37 °C for 18 h. Individual plaques were enumerated and plaque morphology, including size and clarity, was recorded using digital imaging. Phage titers were calculated and expressed as plaque-forming units per milliliter (PFU/mL). Determination of depolymerase activity by reducing sugar (DNS) assay Extraction of extracellular polymeric substances (EPS) Extracellular Polymeric Substances (EPS) was extracted from an overnight culture of K. aerogenes using a modified protocol described before (Yuan et al., 2024). Briefly, K. aerogenes was inoculated into LB medium and incubated at 37 °C in an incubator for 18 h. A 50 mL bacterial culture was used for total EPS extraction. Bacterial culture was centrifuged at 10,000 × g for 20 min at 4 °C to remove cells. The resulting supernatant was collected and mixed with three volumes of ice-cold absolute ethanol. The mixture was incubated at 4 °C overnight to precipitate EPS. Following incubation, the precipitated EPS was recovered by centrifugation at 12,000 × g for 20 min and resuspended in sterile distilled water. The EPS concentration was adjusted to approximately 1 mg mL⁻¹ prior to use. DNS (3,5-Dinitrosalicylic acid) assay For the depolymerase enzyme activity by phage assay, 500 μL of EPS solution was mixed with 500 μL of filter-sterilized phage lysate and incubated at 37 °C for 8 h. Negative control reactions consisted of EPS mixed with sterile SM buffer, while heat-inactivated phage lysate served as an additional control. Following incubation, 500 μL of each reaction mixture was transferred to a fresh tube and mixed with an equal volume of DNS reagent. The reaction mixtures were heated in a boiling water bath (100 °C) for 10 min to allow color development and subsequently cooled to room temperature. Absorbance was determined at 540 nm using a UV–Vis spectrophotometer. Quantification of reducing sugars A glucose standard curve was generated using glucose solutions ranging from 0 to 1000 μg mL⁻¹ treated identically to the test samples. Reducing sugar concentrations released from EPS were calculated from the standard curve and expressed as glucose equivalents. Increased reducing sugar production in phage-treated samples relative to controls was interpreted as evidence of polysaccharide degradation and depolymerase activity. Data are presented as mean ± standard deviation (SD) from three independent biological replicates. Prior to statistical analysis, the normality of residuals was assessed using the Shapiro–Wilk test. Differences among treatment groups (EPS + Buffer, EPS + Phage, and EPS + Heat-activated Phage) were evaluated using one-way analysis of variance (ANOVA) when assumptions of normality were met. Pairwise comparisons were performed using Tukey’s honestly significant difference (HSD) post hoc test. When normality assumptions were not satisfied, a Kruskal-Walli’s test followed by Dunn’s multiple comparison test was applied. Statistical significance was defined as * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. Graphical visualization of the glucose standard curve included mean ± SD values, a fitted linear regression line, and the corresponding 95% confidence interval. Treatment comparisons were visualized using scatter plots displaying individual biological replicates together with mean ± SD and significance annotations. Host range determination Host range specificity of phage KA_SGEB_01 was evaluated using spot assay against a panel of 21 bacterial strains (Montso et al., 2019). Each strain was grown to mid-exponential phase and uniformly spread onto LB agar plates to form bacterial lawns. Ten-fold serial dilutions of phage stock (10⁻¹ to 10⁻⁸) were prepared, and 10 μL of each dilution was spotted onto the bacterial lawns. Plates were incubated at 37 °C for 18 h and examined for zones of clearing indicative of bacterial lysis. Optimization of multiplicity of infection and time-kill kinetics Time-Kill kinetics of K. aerogenes at different multiplicity of infection (MOI) of phage concentration were assessed as previously described (Sivakumar et al., 2025). Briefly, K. aerogenes MAD946 was grown to mid-exponential phase (OD₆₀₀ = 0.4–0.6) and diluted to approximately 10⁷ CFU/mL in fresh LB broth. Phage KA_SGEB_01 was added at multiplicities of infection (MOI) of 0.0001, 0.001, 0.01, 0.1, 1, 10, and 100, defined as the ratio of phage particles to bacterial cells. An uninfected bacterial control was maintained in parallel. Cultures were incubated at 37 °C with shaking at 180 rpm for 6 h. Bacterial growth was monitored by measuring optical density at 600 nm (OD₆₀₀) at hourly intervals to assess time-kill kinetics. Statistical analyses were performed by a one-way ANOVA was applied to OD₆₀₀ values, followed by Tukey's HSD post-hoc test to identify MOI treatment groups that differed significantly from the positive control. Data are presented as mean ± SD of three biological replicates. Significance is denoted as * p < 0.05, ** p < 0.01, and *** p < 0.001. For phage propagation analysis, samples from different MOI levels (100, 10, 1, 0.1, and 0.01) were incubated with the host K. aerogenes MAD946 overnight and then serially diluted in SM buffer (100 mM NaCl, 8 mM MgSO₄, 50 mM Tris–HCl pH 7.5), and enumerated by double-layer agar (DLA) plaque assay to determine maximum phage titre across different MOI values. Phage titres were expressed as log₁₀ plaque-forming units per millilitre (PFU/mL) (Jintasakul et al., 2025). Statistical analysis was performed using a two-way ANOVA to assess the effects of MOI level and titer type (initial vs. final) on phage titre. Post hoc pairwise comparisons between initial and final titres at each MOI were conducted using Tukey's Honest Significant Difference (HSD) test. Significance brackets indicate differences between initial and final titres at each MOI, where * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. Data are presented as mean ± standard deviation (SD). One-Step growth curve analysis One-step growth kinetics were performed following a previously described method (Molina-López et al., 2025) with slight modifications to determine the latent period and burst size of phage KA_SGEB_01. K. aerogenes MAD946 was grown to mid-exponential phase and infected at a multiplicity of infection (MOI) of 0.01. After adsorption for 10 min at 37 °C without shaking, the culture was centrifuged at 10,000 × g for 1 min to remove unabsorbed phages. The pellet was resuspended in fresh pre-warmed LB broth and diluted 1000-fold to prevent further adsorption. The culture was incubated at 37 °C with shaking (180 rpm), and 100 μL samples were collected at 5–10-minute intervals over 90 min. Phage titers were determined using the double-layer agar method. The latent period was calculated from the resulting growth curve. Thermal and pH stability assessment Phage stability was evaluated under varying temperature and pH conditions with slight alteration as previously described (Abd-Allah et al., 2022). For thermal stability, phage suspensions (1 mL; 10⁸ PFU/mL in SM buffer) were distributed into sterile microcentrifuge tubes and incubated at temperatures ranging from −20 °C to 80 °C for 12 h. After incubation, samples were either cooled on ice (heat-treated) or thawed at room temperature (frozen samples), followed by serial dilution and enumeration using the double-layer agar (DLA) method. Residual phage titers were expressed as PFU/mL. For pH stability, phage suspensions were diluted (1:10) into buffers ranging from pH 1 to 12 and incubated at 25 °C for 12 h. Samples were then neutralized with SM buffer (pH 7.5), serially diluted, and quantified by plaque assay. All assays were performed in triplicate, and results are presented as mean ± standard deviation (SD; n = 3). Antibiofilm activity of phage SGEB_KA_01 Biofilm formation and phage-mediated disruption assay Biofilm formation and treatment were performed as previously described with modifications (Mulani et al., 2022). Briefly, biofilms of K. aerogenes MAD946 were established in flat-bottom 96-well polystyrene microtiter plates. Overnight cultures were diluted 1:100 in LB broth supplemented with 1% glucose, and 200 μL (∼10⁶ CFU/mL) was added to each well. Plates were incubated statically at 37 °C for 48 h, with media refreshed every 24 h to promote mature biofilm formation. Following incubation, wells were gently washed three times with sterile phosphate-buffered saline (PBS, pH 7.4) to remove planktonic cells. Biofilms were then treated with 200 μL of phage KA_SGEB_01 at MOIs of 0.01, 1, and 10, while control wells received SM buffer. Plates were incubated at 37 °C for 24 h. After treatment, wells were washed, air-dried, and stained with 0.1% crystal violet for 15 min. The bound stain was solubilized with 95% ethanol, and biofilm biomass was quantified at OD₅₉₅. All experiments were performed in triplicate, and results are expressed as mean ± standard deviation (SD; n = 3). Statistical differences between phage-treated groups and the positive control were evaluated using one-way ANOVA followed by Tukey's HSD post hoc test, with significance set at * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. To assess biofilm disruption, total viable counts (TVC) were determined for both control and phage-treated biofilms using previously described methods (Mulani et al., 2022) . Preformed biofilms were treated with phage KA_SGEB_01 at MOIs of 0.001, 0.01, and 1, and incubated at 37 °C. Following treatment, biofilms were scraped and resuspended in phosphate-buffered saline (PBS). Ten-microliter droplets of 10-fold serial dilutions were spotted onto LB agar plates and incubated at 37 °C for 24 h. Colony-forming units per millilitre (CFU/mL) were calculated from the resulting colony counts. Statistical differences between the control and phage-treated groups were analysed using one-way ANOVA followed by Tukey's HSD post hoc test, with significance defined as * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. Scanning electron microscopy (SEM) analysis of biofilm formation and phage-mediated disruption Biofilm formation by K. aerogenes and its subsequent disruption by phage SGEB_KA_01 treatment were evaluated using scanning electron microscopy (SEM), following a modified protocol described by (Wintachai et al., 2020). Sterile polypropylene surfaces and glass coverslips (1 × 1 cm) in 24-well plates were inoculated with a 1:100 dilution of an overnight K. aerogenes culture (∼10⁸ CFU/mL) in Tryptic Soy Broth (TSB) supplemented with 1% (w/v) glucose. The plates were incubated statically at 37 °C for 48 h without medium replacement to allow biofilm maturation. Following incubation, the culture supernatant was removed, and mature biofilms were treated with bacteriophage lysate (∼1 × 10⁵ PFU/well) at the previously determined optimal multiplicity of infection (MOI = 0.001) for 24 h at 37 °C. Untreated biofilms served as controls. After treatment, samples were gently rinsed twice with sterile phosphate-buffered saline (PBS, pH 7.4) to remove non-adherent cells and fixed with 2.5% (v/v) glutaraldehyde for 2 h at room temperature. The samples were then washed twice with PBS and dehydrated through a graded ethanol series (30%, 50%, 70%, 90%, and 100% v/v) for 10 min at each step. Chemical drying was performed using a graded hexamethyldisilazane (HMDS) series consisting of 33%, 66%, and 100% HMDS in ethanol, with samples incubated for 15 min at each step, following a modified HMDS drying protocol adapted from previous studies (Araujo, 2003; Mallick et al., 2021). Excess HMDS was decanted, and the specimens were air dried overnight in a fume hood. Dried samples were mounted onto aluminium stubs using conductive carbon tape and sputter-coated with gold for 40 s prior to SEM imaging. The surface morphology and biofilm architecture were examined using an Apreo SEM (Thermo Fisher Scientific, USA) at an accelerating voltage of 1 kV under high vacuum. Genomic characterization of phage KA_SGEB_01 Phage DNA extraction Phage KA_SGEB_01 was propagated on K. aerogenes MAD946 to produce high-titer lysate (10⁹−10¹⁰ PFU/mL). Lysate was clarified by centrifugation (10,000 × g, 15 min, 4 °C) and filtered (0.22 μm). Phage particles were precipitated with 10% (w/v) PEG-8000 and 1 M NaCl overnight at 4 °C, collected by centrifugation (12,000 × g, 2 h, 4 °C), and resuspended in magnesium sulphate buffer. Concentrated phage was treated with DNase I and RNase A (1.25 µl, 20 μg/mL each, 37 °C, 1 h), followed by 20 µl EDTA (pH 8.0, 0.5 M final concentration). Phage lysis was performed with proteinase K (1.25 µl, 20 μg/mL) and 10% (w/v) SDS (25 µl) at 60 °C for 1 h. DNA was extracted using Invitrogen PureLink viral DNA extraction kit according to manufacturer's protocol with minor modifications, and stored at –80 °C. Phage genome sequencing and assembly The DNA library was prepared using Illumina DNA Prep (Illumina, San Diego, CA, USA) and sequenced in the Illumina MiSeq platform with 600 cycles, yielding a total of 706,904 paired-end (PE) reads with an average length of 146 bp. Raw sequencing reads were quality-assessed using FastQC v0.12.1. Trimming and filtering were performed using FastP v1.1.0 (mean quality >Q20, length >50 bp) and Trimmomatic v0.40. (LEADING:20, TRAILING:20, SLIDINGWINDOW:4:20, MINLEN:50). De novo assembly was performed using SPAdes v4.2.0 with careful mode and automatic k-mer selection. Assembly quality was evaluated using QUAST v5.3.0 using the default parameters default parameters (minimum contig length = 500 bp, minimum alignment length = 65 bp effective, minimum alignment identity = 95%). SAMtools analysis revealed a high average sequencing depth of ∼685× across the genome. Genome quality was assessed using CheckV v1.0.3. Genome completeness and topology were assessed using CheckV (v1.0.3), which identified a complete linear genome with Direct Terminal Repeats (DTR; high-confidence, 100% completeness, kmer_freq = 1.0). Terminal analysis with PhageTerm indicated redundant ends on both strands, consistent with the expected cross-mapping behaviour of identical DTR sequences rather than circular permutation. The genome was therefore retained as a linear sequence in the GenBank submission. Phage genome annotation and functional analysis Taxonomic assignment was performed using TaxMyPhage v0.3.4 (https://github.com/amillard/tax_myPHAGE). Genome annotation was conducted using PHAROKKA, integrating PHROGs, CARD, and VFDB databases to identify coding sequences, tRNA genes, tmRNA genes, and CRISPR arrays. To complement the phenotypic evidence for depolymerase activity (plaque halo formation), the predicted open reading frames (ORFs) of phage KA_SGEB_01 were screened computationally for depolymerase-associated sequence and structural signatures. ORF-encoded proteins were first analysed using DepoScope (Concha-Eloko et al., 2024). Each predicted protein was assigned a depolymerase classification score (range 0–1). The two ORFs flagged by DepoScope as high-confidence depolymerase candidates, KA_gp4 and KA_gp5, were subsequently submitted independently to Phyre2 (Protein Homology/analogY Recognition Engine v2) (Powell et al., 2025) for remote-homology-based structural modelling in normal modelling mode. Phylogenetic and comparative genomic analysis Proteomic-based phylogenetic analysis was performed using ViPTreeGen v1.1.3 with 5633 reference phage genomes. Trees were visualized using iTOL. Pairwise genome comparisons were conducted using VIRIDIC with 20 closely related Klebsiella phages and five Streptococcus phage outgroups (selection criteria: query coverage ≥90%, amino acid identity ≥90%). Pan-genome analysis of 51 Klebsiella phage genomes was performed using Roary with parameters: minimum percentage identity 90% (-i 90), core gene threshold 99% (-cd 99). Development of phage-loaded alginate hydrogel formulations Preparation of phage-coated and phage-embedded sodium alginate hydrogels Phage-loaded alginate hydrogels were prepared following an adapted protocol from Shiue et al., 2022. Briefly, 100 μL of phage stock solution (KA_SGEB_01; 1 × 10⁹ PFU/mL) was first added to individual wells of a 96-well microplate to form a bottom layer. For the phage-embedded formulation, 200 μL of phage suspension was mixed with 800 μL of sterile 4% (w/v) sodium alginate solution. Subsequently, 100 μL of this mixture was carefully added to each well. An additional 100 μL of phage stock was then overlaid to form a top coating layer. Crosslinking was initiated by adding 200 μL of 5% (w/v) calcium chloride solution (pH 7.5) and allowing gel formation for 10 min. The formed hydrogels were washed twice with 200 μL phosphate-buffered saline (PBS, pH 7.5) to remove unbound phages. Assessment of phage viability following alginate encapsulation Phage stability and infectivity following alginate encapsulation were evaluated at 7- and 14-days post-incorporation. Briefly, Phage KA_SGEB_01 (1 × 10⁹ PFU/mL) was mixed with 4% (w/v) sodium alginate (100 μL phage + 400 μL alginate), and 100 μL of the mixture was dispensed into 96-well plates. Gelation was achieved using calcium chloride (final concentration 5%, pH 7.5). At each time point (7 day and 14 day), gels were dissolved with 100 μL of 0.5 M sodium citrate dihydrate (pH 6.0), and released phages were serially diluted in SM buffer and quantified using the double-layer agar assay. Plaque morphology (clarity and halo formation) was assessed to evaluate lytic activity. Control samples treated with calcium chloride and sodium citrate alone were included to verify that processing did not affect viability. Characterization of phage release kinetics Phage release from phage-coated and phage-embedded alginate hydrogels was evaluated over 72 h. Hydrogels were prepared in 24-well plates by adding 400 μL of phage stock (KA_SGEB_01; 1 × 10⁹ PFU/mL) as a bottom layer. For the embedded formulation, 260 μL of phage suspension was mixed with 1040 μL of 4% (w/v) sodium alginate, and 400 μL of this mixture was added to each well. An additional 400 μL of phage stock was overlaid on top, followed by crosslinking with calcium chloride (final concentration 5%, pH 7.5). After 30 min, gels were washed twice with PBS (pH 7.5) and incubated with 1 mL SM buffer. At 0, 2, 4, 6, 12, 24, 36, 48, 60 and 72 h, 1 mL supernatants were collected and replaced with fresh 1 mL buffer. Phage titers were quantified using plaque assay, and cumulative release (%) was calculated relative to the initial total phage content (N₀). Evaluation of antibiofilm efficacy of phage-hydrogel formulations The antibiofilm activity of phage-embedded and coated alginate hydrogels was evaluated against K. aerogenes biofilms. Overnight bacterial cultures were adjusted to an optical density at 600 nm (OD₆₀₀) of 0.2, and 200 μL aliquots were inoculated onto the hydrogel surfaces. Non-phage-embedded alginate hydrogels served as negative controls. Microplates were incubated statically at 37 °C for 24 h to allow biofilm formation. Following incubation, the bacterial inoculum was carefully removed, and hydrogels were gently rinsed twice with 100 μL of PBS per wash to eliminate planktonic cells. Viable biofilm-associated bacteria were enumerated by dissolving the alginate gel through the addition of 200 μL of 0.5 M sodium citrate dihydrate solution (pH 6.0) directly into each well. The resulting suspension was transferred to 800 μL of 0.5 M sodium citrate dihydrate (pH 6.0) and homogenized by vigorous vortexing. Serial dilutions of the disrupted biofilm suspensions were prepared in PBS, and viable cell counts (CFU mL⁻¹) were determined using the Miles and Misra drop plate method. This experiment was performed in triplicate, and statistical analysis was conducted using an unpaired Welch's t-test to compare viable bacterial counts between phage-embedded and coated hydrogels and non-phage controls. Differences were considered statistically significant at p < 0.05. Statistical significance was denoted as follows * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. Results Phage isolation and purification Bacteriophage KA_SGEB_01 was successfully isolated from sewage samples using K. aerogenes strain MAD946 as the bacterial host. Initial screening by spot assay demonstrated distinct and well-defined zones of complete bacterial lysis, confirming the presence of active lytic phage particles (Fig. 1A). Plaque morphology was further characterized using the double-layer agar (DLA) method. The phage produced uniform, small, and clear plaques with an average diameter of 1 ± 0.1 mm. Notably, each plaque was surrounded by a translucent halo (Fig. 1B), indicating potential depolymerase activity associated with the degradation of capsular or biofilm-associated polysaccharides. Spot test and plaque morphology of KA_SGEB_01 phage. (A) Spot test result of KA_SGEB_01 showing lytic activity against MDR K. aerogenes(B) KA_SGEB_01 produced uniform, small (∼1 ± 0.1 mm), clear plaques with translucent halos. Host range determination Host range analysis was performed against a panel of 21 bacterial strains comprising diverse gram-negative and gram-positive species (Table 2). Phage KA_SGEB_01 exhibited a highly specific infection profile, producing clear zones of complete lysis exclusively against its primary host strain, K. aerogenes MAD946; no detectable lytic activity was observed against any other species or genus represented in the panel. A K. aerogenes strain CHRF1 produced a turbid zone of clearing on spot testing, indicative of partial rather than complete lysis. This turbid response suggests reduced, but detectable, phage activity against CHRF1 strain, potentially attributable to a partially resistant subpopulation, restricted adsorption efficiency, or another strain-specific limiting factor. Collectively, these findings confirm that KA_SGEB_01 maintains a narrow host range with strong specificity for K. aerogenes, while the partial susceptibility observed in CHRF1 indicates that the phage may retain limited lytic potential against other K. aerogenes strains beyond its primary host strain. Table 2Host range analysis of KA_SGEB_01 phage.Table 2 dummy alt textNo.SpeciesStrain IDPlaque Morphology in Direct Spot Test1A. baumanniiMAD994No Zone2A. baumanniiMAD807No Zone3A. baumanniiICD005No Zone4Acinetobacter lwoffiiMAD546No Zone5Enterococcus faecalisMAD291No Zone6Escherichia coliICD004No Zone7E. coliECNR01No Zone8K. aerogenesMAD946Clear Zone9K. aerogenesMAD285No Zone10K. aerogenesCHRF1Turbid Zone11K. aerogenesCHRF2No Zone12K. pneumoniaeMAD058No Zone13K. pneumoniaeMAD748No Zone14K. pneumoniaeICD001No Zone15K. pneumoniaeICB001No Zone16Salmonella sp.CC2No Zone17Salmonella typhiSTBMB_01No Zone18S. aureusMAD918No Zone19Staphylococcus epidermidisMAD3242No Zone20Streptococcus agalactiaeSAG_BMB01No Zone21Vibrio choleraeVC_BMB-01No Zone Host range analysis of KA_SGEB_01 phage. Depolymerase activity and identification of putative depolymerase-encoding genes The depolymerase activity of phage KA_SGEB_01 was evaluated using 3,5-dinitrosalicylic acid (DNS) assay to quantify reducing sugars released from K. aerogenes extracellular polymeric substances (EPS). A glucose standard curve generated using glucose concentrations ranging from 100 to 1000 μg mL⁻¹ showed a strong linear relationship between glucose concentration and absorbance at 540 nm, enabling quantification of reducing sugars released during EPS degradation (Fig. 2A). DNS assay quantification of EPS depolymerase activity. (A) Glucose standard curve (y = 0.0025x + 0.5012, R² = 0.9808) used to calculate reducing sugar equivalents; shaded band, 95% CI. (B) Reducing sugars released from EPS following treatment with phage, heat-activated phage, or buffer, expressed as glucose equivalents. Data are mean ± SD (n = 3). Normality was assessed by Shapiro–Wilk test; group differences were analyzed by one-way ANOVA with Tukey's HSD post hoc test (or Kruskal–Wallis with Dunn's test if normality was violated). *p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant. Active phage treatment released substantially higher levels of reducing sugars (∼240 μg/mL glucose equivalents) than either the buffer control (∼30 μg/mL) or heat-activated phage treatment (∼60 μg/mL) (Fig. 2B). Statistical analysis revealed a significant treatment effect (one-way ANOVA, F(2,6) = 13.94, p = 0.0056), with the active phage group differing significantly from both control groups, while no significant difference was observed between the buffer and heat-activated phage treatments. The elevated release of reducing sugars following exposure to active phage indicates enzymatic degradation of EPS polysaccharides. In contrast, the absence of a significant increase in reducing sugar release following heat activation suggests that the observed activity is associated with heat-labile phage components, consistent with the action of a phage-associated depolymerase. To identify the genetic basis of this activity, all predicted ORFs of KA_SGEB_01 were screened using DepoScope. Two potential ORFs, KA_gp4 and KA_gp5, were identified as high-confidence depolymerase candidates, each receiving the maximum classification score of 1.0 (Supplementary Table S2). Structural modelling of KA_gp4 using Phyre2 recovered a hyaluronate lyase-related fold based on the tail fibre hyaluronate lyase HylP2 template (PDB: 2WH7) (Supplementary Figure 1), supporting a direct polysaccharide-degrading function. Structural modelling of KA_gp5 identified a C-terminal intramolecular chaperone domain homologous to that of the Bacillus phage GA-1 neck appendage protein (PDB: 3GUD) (Supplementary Figure 2), a structural feature commonly associated with tail-spike and tail-fibre depolymerases. Together with the pronounced halo morphology observed around phage plaques, the DNS assay and genomic analyses provide complementary phenotypic and molecular evidence that KA_SGEB_01 encodes depolymerase-associated proteins capable of degrading extracellular polysaccharides produced by K. aerogenes. Multiplicity of infection (MOI) assays Time-kill assays were performed across a range of MOIs (0.0001, 0.001, 0.01, 0.1, 1, 10 and 100) against K. aerogenes MAD946 over 360 min (Fig. 3A). All phage-treated cultures tracked the positive control through 60–90 min (OD₆₀₀ 0.14–0.17). Beyond 90–120 min, OD₆₀₀ declined sharply across all MOIs, while the untreated positive control continued rising to 0.264 by 360 min. (A) Time killing curve (MOI) of host K. aerogenes by KA_SGEB_01 phage by measurement of OD600.(B) MOI assay of host K. aerogenes by KA_SGEB_01 phage showing maximum phage titer at different MOI values. Statistical analysis was performed using a two-way ANOVA to assess the effects of MOI level and titer type (initial vs. final) on phage titer. Significance brackets indicate differences between initial and final titers at each MOI, where * p < 0.05, ** p < 0.01, *** p < 0.001, ns = not significant. Data are presented as mean ± standard deviation (SD). Low-to-mid MOIs (0.0001.0.001 and 0.01) produced fast and significant clearance, reaching a stable plateau (OD₆₀₀ ≈ 0.115–0.125) by 120–150 min that persisted through 360 min, closely approaching the negative control (≈0.10). This indicates that phage KA_SGEB_01 is able to retain full lytic efficacy at low MOIs. Phage propagation efficiency was further evaluated by plating-based enumeration (Fig. 3B). Significant differences between initial and final phage titres were observed across all tested MOI values. Two-way ANOVA revealed significant effects of MOI, titre type (initial versus final), and their interaction (all p < 0.0001), indicating that phage amplification was strongly influenced by the initial phage-to-host ratio. At higher MOI values (100 and 10), final titres were lower than initial titres, suggesting phage-host dynamics dominated by lytic activity and possible phage particle loss. Conversely, at lower MOI values (0.1 and 0.01), final titres exceeded initial titres, indicating net phage amplification during overnight incubation, with the greatest fold-increase observed at MOI 0.01, Based on these findings, MOI 0.01 was identified as optimal for phage KA_SGEB_01 propagation. One-Step growth curve One-step growth curve analysis revealed classical lytic replication dynamics for phage KA_SGEB_01 (Fig. 4). An initial latent period of approximately 15 min was observed, during which extracellular phage titres remained stable at ∼10³ PFU/mL. This phase was followed by a rapid rise period between 15 and 40 min, characterized by a sharp exponential increase in phage titer, reaching approximately 10⁸ PFU/mL. After 40 min, the growth curve entered a plateau phase, indicating completion of the primary lytic cycle and stabilization of phage production. One step growth curve of KA_SGEB_01 phage. Extracellular PFU/mL were quantified following synchronized infection at defined intervals. A discrete latent period (∼0–15 min) preceded the lysis phase (∼15–40 min), during which phage titres increased exponentially. A subsequent plateau (>40 min) reflects completion of the lytic cycle. Temperature and pH stability The stability of phage KA_SGEB_01 was evaluated under varying temperature and pH conditions following 12-hour incubation. KA_SGEB_01 remained highly stable at −20 °C and 4 °C, maintaining titres of approximately 4 × 10⁸ PFU/mL (Fig. 5A). Optimal stability was also observed at 37 °C, with peak titres reaching ∼7 × 10⁸ PFU/mL. A moderate reduction in titre occurred at 50 °C, whereas exposure to temperatures ≥60 °C resulted in complete loss of phage viability. Stability of phage KA_SGEB_01. (A) Thermal stability profile of phage KA_SGEB_01 across a temperature gradient assessed by PFU/ml. Thermal stability of phage KA_SGEB_01 was evaluated across −20 °C to 80 °C following 12-hour incubation. Data indicate optimal stability under refrigeration and physiological temperatures, with thermal inactivation threshold between 50–60 °C. (B) pH stability profile of phage KA_SGEB_01 across pH gradient assessed by PFU/mL. pH stability of phage KA_SGEB_01 was evaluated across pH 1–12 following 12-hour incubation. Optimal stability occurred at pH 6–9 with maximum titres at neutral pH 7. Moderate conditions (pH 3, pH 11) showed slight reduction while extreme acidic (pH ≤2) and alkaline (pH 12) conditions caused substantial to complete inactivation. Error bars represent standard deviation from triplicate measurements. Phage stability across a pH range of 1–12 demonstrated that KA_SGEB_01 retained high stability between pH 6 and 9, with titres around 5 × 10⁸ PFU/mL and maximum stability at neutral pH (Fig. 5B). Moderate reductions were observed at pH 3 and 11, while extreme conditions (pH ≤2 or ≥12) led to near-complete inactivation. Biofilm biomass reduction activity Reduction of biofilm biomass by phage KA_SGEB_01 The antibiofilm activity of phage KA_SGEB_01 was evaluated against 48-hour-old K. aerogenes biofilms formed on polystyrene 96-well plates. Pre-formed biofilms were exposed to phage at MOIs of 10, 1, and 0.01, and biomass was quantified using crystal violet staining following the treatment period (Fig. 6A). All phage-treated groups showed statistically significant reductions in biofilm biomass relative to untreated positive controls (P < 0.05), indicating that phage exposure inhibited further biofilm biomass accumulation. Among the tested conditions, MOI 0.01 produced the greatest reduction in biofilm biomass, with OD₅₉₅ values decreasing from approximately 2.0–2.22 in untreated controls to 0.82–0.85 in treated wells, corresponding to an approximately 60% reduction in crystal violet or CV-stained biomass relative to controls. Treatment at MOI 1 also produced a substantial reduction in biofilm biomass; however, increasing the MOI beyond this did not further enhance the antibiofilm effect. Antibiofilm activity of phage KA_SGEB_01 against pre-formed K. aerogenes biofilm. (A) Biofilm biomass (OD₅₉₅, crystal violet staining) of 48 h biofilms following treatment with phage at MOI 10, 1, or 0.01, versus untreated positive control and negative control. (B) Bacterial viability (log₁₀ CFU/mL) of treated biofilms following phage exposure at MOI 1, 0.01, and 0.001, versus untreated control. Data are mean ± SD (n = 3). Group differences were assessed by one-way ANOVA with Tukey's HSD post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant. Treatment of preformed biofilms with phage KA_SGEB_01 resulted in significant reductions in bacterial viability across all tested multiplicities of infection (MOIs). The untreated control biofilms maintained a mean bacterial load of approximately 8.3 log₁₀ CFU/mL. Phage treatment at MOI 1, 0.01, and 0.001 reduced viable counts to approximately 5.0, 4.9, and 4.8 log₁₀ CFU/mL, respectively, corresponding to reductions of ∼3.3, ∼3.4, and ∼3.5 log₁₀ CFU/mL relative to the control (Fig. 6B). One-way ANOVA revealed a statistically significant overall effect of phage treatment (p < 0.001). Tukey's HSD post hoc analysis confirmed that all three MOIs differed significantly from the control (*** p < 0.001), while bacterial counts among the phage-treated groups remained comparable across the tested MOI range, suggesting that low MOIs retain substantial biofilm-disrupting activity. Scanning electron microscopy (SEM) was performed to visualize architectural and morphological changes in K. aerogenes biofilms following treatment with phage SGEB_KA_01. Untreated control biofilms (Fig. 7A) displayed a dense, confluent, multilayered architecture composed of tightly packed, smooth-surfaced cells interconnected by an extensive extracellular matrix, consistent with a structurally intact, mature biofilm. This biofilm was embedded within an extracellular polymeric substance (EPS) matrix, a structural framework that acts as a protective barrier for the bacteria against external stressors (Fig. 7A). In contrast, phage-treated biofilms (Fig. 7B) exhibited markedly reduced cell density and surface coverage, with large denuded regions of exposed substratum and loss of the continuous multilayered structure. A substantial proportion of the residual cells showed collapsed, shrivelled, and irregularly deformed envelopes rather than the smooth, turgid morphology seen in untreated controls, indicative of membrane damage and cell lysis. These qualitative observations are consistent with the significant reductions in biofilm biomass and viable cell counts obtained by crystal violet and bacterial viability assays mentioned before, providing direct evidence of the inhibitory effect of phage SGEB_KA_01 on K. aerogenes biofilm. Scanning electron microscopy of K. aerogenes biofilms untreated and treated with phage SGEB_KA_01. (A) Untreated control biofilm showing dense, multilayered aggregates of intact, smooth-surfaced cells embedded in extracellular matrix. (B) Phage-treated biofilm showing reduced cell density, loss of multilayered architecture, and deformed cell envelopes consistent with phage-induced lysis. Images acquired at 10,000× magnification and 1.00 kV accelerating voltage; scale bars, 5 μm. Alginate hydrogel encapsulation and controlled release Phage release kinetics The in vitro release profile of KA_SGEB_01 from alginate hydrogel was assessed over a 72-hour period to characterize the release kinetics (Fig. 8). An initial burst release was observed, with approximately 15% of the total phage titer released within the first 6 h, increasing to 37% by 12 h and 45% by 24 h. Following this initial burst phase, the release rate progressively decelerated, reaching 54% at 36 h and 65% at 48 h, after which the profile plateaued, with cumulative release reaching 67% at 60 h and 67% at 72 h (Fig. 8). In vitro release kinetics of phage KA_SGEB_01 from alginate hydrogel. Cumulative percentage of phage released over a 72 h period at 0, 3, 6, 12, 24, 36, 48, 60, and 72 h. Phage release was rapid during the initial 12 h, followed by a gradual increase and plateau after 48 h. Data are mean ± SD (n = 3). Release monitoring was concluded at 72 h, by which point the release profile had reached a clear plateau, with cumulative release stabilizing at approximately 67% of the total encapsulated titer between 48 and 72 h. Plaque assays performed on aliquots collected at each time point confirmed that released phages retained infectivity, evidenced by clear, characteristic halo formation, indicating that encapsulation and the release process did not compromise phage viability or lytic activity. Phage viability within alginate hydrogels Long-term phage stability of encapsulated phage was evaluated at 7 and 14 days (Fig. 9). The initial phage loading was ∼3.5 × 10⁹ PFU/mL. After 7 days, titres remained unchanged at approximately 3.5 × 10⁹ PFU/mL with no alteration in plaque morphology. At 14 days, only a slight reduction to ∼3.1 × 10⁹ PFU/mL was observed. Control treatments exposed to calcium chloride and sodium citrate alone showed comparable stability, confirming that encapsulation and recovery processes did not adversely affect phage viability. Stability of phage KA_SGEB_01 following alginate encapsulation. Phage titers (PFU/mL, log scale) for untreated stock phage, alginate-encapsulated phage, and CaCl₂/citrate control, assessed at 7 and 14 days. Individual data points represent biological replicates with mean ± SD (n = 3). Phage titers remained stable across all groups and time points, with no substantial loss in infectivity observed following encapsulation. Antibiofilm activity of phage-embedded hydrogels The antibiofilm efficacy of phage-embedded and phage-coated alginate hydrogels was evaluated against K. aerogenes biofilm formation (Fig. 10). Following 24-hour incubation at 37 °C, phage-embedded hydrogels demonstrated significant antibiofilm activity compared to phage-free controls (**P < 0.005). Biofilm-associated bacterial counts were reduced from ∼10¹⁰ CFU/mL in control hydrogels to ∼10⁴ CFU/mL in phage-loaded formulations, representing a substantial decrease in bacterial burden. In contrast, alginate hydrogels without phage showed extensive bacterial growth, confirming that the observed antibiofilm effect was attributable to phage activity rather than the hydrogel matrix. Antibiofilm efficacy of phage KA_SGEB_01-embedded and -coated alginate hydrogels against K. aerogenes biofilm formation. Biofilm-associated viable bacterial counts (CFU/mL, log scale) recovered from phage-free control hydrogels versus phage-embedded and coated hydrogels following 24 h incubation at 37 °C. Data points represent individual biological replicates with mean ± SD (n = 3). Statistical significance was assessed by unpaired Welch's t-test. Genomic characterization phage of KA_SGEB_01 Genomic features and organization of phage KA_SGEB_01 Whole-genome sequencing of phage KA_SGEB_01 revealed a linear double-stranded DNA genome spanning 175,095 bp, with a GC content of 41.86% (Table 3; Fig. 11). Structural annotation of the genome identified 289 predicted coding sequences (CDS) and two transfer RNA (tRNA) genes. Notably, no CRISPR arrays, transfer-messenger RNA (tmRNA) genes, virulence factors, or antimicrobial resistance (AMR) determinants were detected within the genome. However, seven putative lytic genes were identified, suggesting a lytic replication strategy consistent with potential phage biocontrol applications. Table 3Gene count for KA_SGEB_01 phage.Table 3 dummy alt textDescriptionCountCDS289Connector3DNA, RNA and nucleotide metabolism42Head and packaging15Integration and excision0Lysis7Moron, auxiliary metabolic gene and host takeover3Others14Tail30Transcription regulation5Unknown function170tRNAs2CRISPRs0tMRNAs0VFDB_virulence_factors0CARD_AMR_genes0 Gene count for KA_SGEB_01 phage. Genomic organization of bacteriophage KA_SGEB_01. Circular representation of the complete 175,095 bp genome showing predicted coding sequences (CDS) color-coded by functional category in the outermost ring. unknown function (gray), other function (cyan), transcription regulation (yellow), DNA/RNA/nucleotide metabolism (orange), lysis (blue), moron/auxiliary metabolic genes and host takeover (purple), head and packaging (pink), tail (brown), and connector (green). No virulence factors or antimicrobial resistance genes were detected. The second ring displays GC content (average 41.86%), the third ring shows GC skew, and the innermost ring indicates tRNA/tmRNA/CRISPR elements (two tRNA genes detected). Gene names and predicted functions are labelled radially. Taxonomic and phylogenetic analysis Taxonomic classification using TaxMyPhage placed KA_SGEB_01 within the family Straboviridae, genus Slopekvirus, species Slopekvirus matisse (Supplementary Figure 3). Phylogenetic analysis based on proteomic comparisons clustered KA_SGEB_01 within a monophyletic group of Klebsiella-infecting phages (Fig. 12). The phage showed close evolutionary relatedness to other members of the Slopekvirus genus and was clearly distinct from phages infecting non-Klebsiella hosts (Supplementary Figure 3). Phylogenetic analysis of bacteriophage KA_SGEB_01 based on viral proteome comparison. Proteomic tree generated using VIPtree showing phylogenetic relationships between KA_SGEB_01 and related bacteriophages. The tree was constructed by comparing proteome sequences, with phages colored by virus family (inner ring) and host group (outer ring). Green represents Straboviridae family members, with various other families shown in pink, orange, blue and gray. KA_SGEB_01 (starred) clusters within the Straboviridae family alongside other Klebsiella-infecting phages, forming a distinct monophyletic clade. The phylogenetic analysis demonstrates clear host-associated clustering, with KA_SGEB_01 positioned among Klebsiella phages and showing substantial evolutionary distance from phages infecting other bacterial genera such as Escherichia, Salmonella, and Citrobacter. Scale bar indicates genetic distance based on proteome similarity. Intergenomic and Pan-genome analysis Intergenomic comparison using VIRIDIC demonstrated high nucleotide identity (>95%) and genome coverage (>95%) with closely related Klebsiella phages, while showing minimal similarity to distantly related Streptococcus phages (Supplementary Figure 4). Comparative analysis of KA_SGEB_01 (175,095 bp; 289 CDS) against 50 publicly available Slopekvirus genomes revealed that both genome size and coding capacity fall centrally within the genus-typical range (171,960–176,329 bp; 258–291 CDS among complete genomes; mean ≈175.2 kb, ≈276 CDS), indicating a highly conserved genome architecture across the genus. Pan-proteome analysis of 51 complete Klebsiella phage genomes (Supplementary Table S3) identified 497 gene clusters, with KA_SGEB_01 clustering most closely with eight Klebsiella phages of similar genome architecture: OR234023.1, OM103621.1, OX335396.1, OR256021.1, OR256022.1, LR881140.1, PP978606.1, and PP993148.2 (Fig. 13). Of the 289 genes encoded by KA_SGEB_01, 54 (18.7%) belonged to the core genome present in all 51 genomes, encoding essential structural and replication functions including terminase, major capsid protein, DNA polymerase, and helicase, alongside core components of the tail fiber apparatus. A further 151 genes (52.2%) were soft-core (present in 48–50 genomes) and 76 (26.3%) were shell genes (present in 6–47 genomes); together these accessory categories included additional tail fiber and baseplate-associated proteins implicated in host receptor recognition, alongside numerous hypothetical proteins of unknown function (Supplementary Table S4). The remaining eight genes (2.8%) were each shared with five or fewer of the 51 genomes and represent the most clade-restricted component of the KA_SGEB_01 genome. Six of these encoded hypothetical proteins of unknown function, while the remaining two encoded a tail fiber protein implicated in host receptor-binding specificity and a homing endonuclease implicated in DNA recombination and mobile genetic element processing. This dendrogram shows the clustering patterns of 50 Klebsiella phage and KA_SGEB_01 phage (marked with red dot) proteomes. The analysis categorized the phage genomes into 497 gene clusters including core, accessory and unique components, revealing the common essential genes (core) and highlighting the diversity of accessory genes encoding auxiliary functions. Discussion The global escalation of antimicrobial resistance (AMR), particularly among Gram-negative ESKAPE pathogens, has intensified the need for alternative therapeutic strategies capable of addressing both planktonic and biofilm-associated infections. K. aerogenes is increasingly recognized as an opportunistic, multidrug-resistant, biofilm-forming pathogen that significantly limits antibiotic efficacy (Rodrigues et al., 2024). In this study, bacteriophage KA_SGEB_01 was isolated, characterized at the phenotypic and genomic levels, and successfully integrated into an alginate hydrogel delivery system, supporting its potential as a therapeutic candidate against multidrug-resistant K. aerogenes infections. A distinctive phenotypic feature of KA_SGEB_01 is its plaque morphology, comprising small, sharply defined lysis zones surrounded by translucent halos, a hallmark of depolymerase activity reflecting enzymatic degradation of extracellular polysaccharide beyond the immediate zone of lysis (Pires et al., 2016). Plaque size was consistent with previously reported Klebsiella-infecting phages (Lin et al., 2014; Zurabov and Zhilenkov, 2021), and the halo phenotype aligned with well-characterized capsule-depolymerizing phages (Wu et al., 2023), a feature particularly relevant given the role of K. aerogenes capsular polysaccharides in immune evasion, antibiotic resistance, and biofilm stability (Cui et al., 2023). The substantially higher reducing sugar release observed with active KA_SGEB_01 relative to buffer and heat-inactivated controls provides biochemical evidence of EPS-degrading depolymerase activity associated with the phage. These findings corroborate the genomic identification of candidate depolymerase genes (KA_gp4, KA_gp5) and support a mechanistic basis for the phage's biofilm-disruptive potential against K. aerogenes. The strict specificity of KA_SGEB_01 for its primary host reflects receptor-level selectivity typical of lytic phages within Enterobacteriaceae, supporting its potential as a targeted antibacterial with minimal off-target effects. The partial lytic activity in K. aerogenes strain CHRF1, contrasted with full resistance in strains MAD285 and CHRF2, suggests intraspecies variation in receptor accessibility or resistance determinants among K. aerogenes strains. This rigid specificity likely reflects precise recognition of host surface receptors, particularly capsular polysaccharides, which are highly variable among Klebsiella strains (Beamud et al., 2023). However, capsular typing of the tested K. aerogenes isolates, including CHRF1, MAD285, and CHRF2, was not performed in this study. Consequently, the proposed capsule-dependent basis for the observed host specificity and the partial susceptibility of CHRF1 remains inferential rather than empirically confirmed, and should be verified in future work through capsule serotyping and receptor-binding protein characterization. This partial susceptibility indicates residual lytic potential beyond the primary host but warrants further analysis to confirm its biological significance. Genomic analysis supports a strictly lytic and safe profile, with no virulence and antimicrobial resistance genes detected. Combined with the narrow host range, this indicates that KA_SGEB_01 is genetically constrained to targeted, self-limiting lytic activity rather than horizontal gene transfer or broad-spectrum disruption of commensal microbiota. While such narrow host-specificity may limit standalone therapeutic application, this limitation can be addressed clinically through phage cocktails to broaden host coverage. The phage possesses a large double-stranded linear DNA genome with a modular organization typical of members of the Straboviridae family. Phylogenetic analysis placed KA_SGEB_01 within the genus Slopekvirus, showing high nucleotide similarity to other Klebsiella-infecting phages, while Streptococcus phages, included only as a distantly related outgroup for tree resolution, showed negligible similarity. This supports the validity of this genus-level placement. KA_SGEB_01 falls within the genome size and gene content range typical of other Slopekvirus members, reflecting the conserved genome architecture characteristic of this genus. Pan-genome analysis of the 51 Klebsiella-clade phage genomes showed that the core genome predominantly encoded structural and replication machinery essential for particle assembly, genome propagation, and host recognition, including the tail fiber and baseplate apparatus. This conservation reflects a shared evolutionary trajectory of phage KA_SGEB_01 among phages infecting strains within the same bacterial genus. Variability within tail fiber and depolymerase-encoding regions likely underlies the host specificity and adaptive potential of KA_SGEB_01, consistent with selective pressure to recognize diverse capsular structures. The infection kinetics of KA_SGEB_01 include a short latent period and efficient replication at low multiplicity of infection. This supports rapid, self-amplifying bacterial clearance at the site of infection and reduces the need for high initial dosing, consistent with the depolymerase-encoding capacity identified genomically. The SEM imaging provides morphological confirmation of the antibiofilm activity of phage KA_SGEB_01, complementing the quantitative biomass and viability data. The sparse, dispersed distribution of surviving cells and the extensive denuded areas of substratum in treated samples indicate that phage predation disrupts not only individual cells but also the structural continuity of the biofilm matrix, limiting recolonization. Substantial biofilm biomass reduction at low MOIs further confirms the functional contribution of depolymerase activity to matrix degradation and enhanced phage penetration (Hesse et al., 2021). Unlike purely lytic phages, depolymerase-producing phages can extend activity beyond infected cells, increasing the effective therapeutic range (Harper et al., 2014), a property particularly valuable for chronic wounds and device-associated infections where biofilms are a major barrier to treatment. Environmental stability is a further determinant of the clinical applicability of this phage. KA_SGEB_01 retained infectivity across physiologically relevant temperatures and a near-neutral to mildly alkaline pH range typical of skin, wounds, and blood (Jończyk et al., 2011), with loss of viability only under extreme temperature and highly acidic conditions, a limitation common among bacteriophages generally (Blazanin et al., 2022; Rezk et al., 2022) that can be mitigated through appropriate formulation strategies (Malik et al., 2017). Building on these favourable biological and genomic characteristics, the incorporation of KA_SGEB_01 into an alginate hydrogel delivery system allowed us to address two persistent challenges in phage therapy: stability and localized delivery. Encapsulation preserved phage viability, infectivity, and depolymerase activity, and produced a sustained, gradual release profile over 72 h. Titres remained above 10⁹ PFU/mL for at least 14 days, and phage-loaded hydrogels showed significantly greater antibiofilm activity than phage-free controls, confirming that the effect was phage-driven rather than matrix-driven. KA_SGEB_01 has not previously been characterized in hydrogel-encapsulated form, so its survival and action can only be inferred from the broader phage-alginate literature. Alginate's biocompatibility, biodegradability, and high water content make it a common delivery platform (Ren et al., 2024; Tan et al., 2023; Tomić et al., 2023), and its mild ionic gelation favours capsid integrity relative to chemically crosslinked systems, although crosslinking alone can reduce phage titre by more than one log in some cases (Kim et al., 2021). That KA_SGEB_01 retained titres above 10⁹ PFU/mL through encapsulation and 14 days of storage suggests good tolerance of this process, consistent with reports of 85–90% lytic activity retention in similarly mild alginate fabrication routes (Shen et al., 2021; Tao et al., 2026). This likely reflects physical protection within a hydrated polymer mesh rather than chemical stabilization (Khazani Asforooshani et al., 2024). The observed gradual release shows slower diffusion- or degradation-driven release, with even partial early release sufficient for substantial killing because phage amplify biologically once they reach susceptible hosts (Kim et al., 2021; Shen et al., 2021; Vater et al., 2026). Phage-hydrogel systems are best suited to localized infections such as wounds, burns, and implant- or bone-associated infections, where sustained local concentration matters more than systemic distribution (Rotman et al., 2022; Shafigh Kheljan et al., 2023; Vater et al., 2026), with reported release durations ranging from hours to weeks depending on formulation (Chang et al., 2021; Moghtader et al., 2024). Stimuli-responsive designs and phage-antibiotic or phage-probiotic combinations represent promising directions for limiting resistance and broadening efficacy (Y. Chen et al., 2023; Gao et al., 2024; Liang et al., 2026; Tao et al., 2026), and the 14-day stability observed here may be useful where cold-chain storage is unreliable (Abed et al., 2024; Moghtader et al., 2024). However, preclinical hydrogel systems often reduce but do not fully sterilize infections, with weaker efficacy in less accessible compartments such as bone compared to soft tissue (Shafigh Kheljan et al., 2023; Vater et al., 2026; Wroe et al., 2020). Stability is strongly phage- and matrix-dependent, so these results cannot be generalized without direct testing (Moghtader et al., 2024; Shen et al., 2021). Manufacturing scale-up, host immune responses, narrow phage host range, and regulatory uncertainty remain unresolved, and human evidence is largely limited to case reports and ex vivo models rather than randomized trials. Accordingly, the present findings represent in vitro proof-of-concept rather than evidence of clinical efficacy. The significant reduction in biofilm-associated bacterial burden relative to phage-free controls supports a phage-driven mechanism, and while resulting viable counts fell below in vitro benchmarks reported for chronic wound infection (Colom et al., 2015; Harper et al., 2014). This comparison reflects in vitro performance rather than a demonstrated clinical outcome, as no in vivo data are presented here. The combination of lytic activity, depolymerase-mediated biofilm disruption, and controlled release may offer a potential alternative to conventional antimicrobial strategies, pending in vivo validation. Future work should prioritize in vivo efficacy and safety studies, refinement of release kinetics relative to infection burden, and incorporation of KA_SGEB_01 into phage cocktails to broaden host range and limit resistance. Conclusion This study reports the isolation and characterization of a lytic bacteriophage, KA_SGEB_01, targeting multidrug-resistant K. aerogenes, and demonstrates its therapeutic potential against biofilm-associated infections. The phage showed strong lytic activity, rapid replication, and high specificity, along with depolymerase-associated antibiofilm effects, and remained stable under physiologically relevant conditions. Whole-genome sequencing revealed that KA_SGEB_01 possesses a 175,095 bp linear double-stranded DNA genome with a GC content of 41.86%, encoding 289 predicted CDS and lacking any virulence factors, or AMR determinants, consistent with a strictly lytic lifestyle suitable for biocontrol applications. Taxonomic classification placed the phage within the genus Slopekvirus, with phylogenetic analysis confirming its clustering among Klebsiella-infecting phages and clear distinction from phages of non-Klebsiella hosts. Genomic analysis further identified potential two depolymerase-encoding genes consistent with the biofilm-degrading activity observed experimentally. Incorporation into an alginate hydrogel enabled sustained release, preserved viability, and significant antibiofilm efficacy, reducing in vitro bacterial burden to levels comparable with clinically relevant infection thresholds. Overall, KA_SGEB_01 represents a promising candidate for phage-based therapy, and hydrogel-based delivery offers a promising approach for treating multidrug-resistant infections. Future studies should focus on in vivo validation, formulation optimization, and development of phage cocktails to support clinical translation. Funding This work was primarily supported by the SUST Research Center (Grant no: LS/2024/2/21) and the University Grants Commission of Bangladesh (Grant no: 37.01.0000.073.04.030.23.2135). CRediT authorship contribution statement Istiaque Zaeem: Methodology, Validation, Formal analysis, Data curation, Writing – original draft. Shakhinur Islam Mondal: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition. Raphael Kabir Niloy: Methodology, Validation. Nurnabi Azad Jewel: Methodology, Validation, Formal analysis, Software, Project administration. Mohiminur Rahman Mahin: Validation, Formal analysis. Rita Bhatta: Methodology, Validation, Formal analysis. Daniyal Karim: Methodology, Validation. Mohimenul Haque Rolin: Methodology, Validation. Mohsina Khatoon: Resources. Arzuba Akter: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing – review & editing, Visualization, Supervision, Project administration. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.