1. Introduction

The prevalence of extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli poses a severe global healthcare problem, as these strains are resistant to beta-lactam antibiotics, including penicillins, cephalosporins, and monobactams. Salam et al.¹ demonstrated through recent data that antimicrobial resistance directly leads to more than 1.27 million annual fatalities, suggesting growing global health risks. The genes that produce ESBLs spread via mobile genetic elements, enabling bacteria to readily exchange resistant genes and transfer this resistance to other microorganisms. Resistant E. coli strains demonstrate cross-resistance against antibacterials, including fluoroquinolones, aminoglycosides, and tetracyclines².

The World Health Organization issued its updated list of priority pathogens, which designates E. coli and its related bacteria, Pseudomonas aeruginosa, and Klebsiella pneumoniae as critical Gram-negative pathogens. Medical research has confirmed that E. coli is a top-priority pathogen because it causes severe infections both inside and outside healthcare facilities³.

Frequently referred to as a “Jekyll and Hyde organism,” E. coli is a commensal resident of both humans’ and animals’ gastrointestinal tracts that contributes to gut health⁴. However, several dangerous E. coli strains may cause severe infections, including bloodstream infections, meningitis, and urinary tract infections (UTIs)⁵. E. coli strains have been classified into four different phylogenetic groups (A, B1, B2, and D), as shown in Figure 1. Commensals belong to groups A and B1, whereas pathogenic extra-intestinal bacteria belong to B2 and, to a certain extent, to the D group. These phylogenetic classifications are essential as they provide information about the strains’ origins and pathogenicity, which will help identify an accurate treatment option⁶.

Figure 1. Phylogenetic classification of Escherichia coli.

 According to recent studies, E. coli is the primary cause of more than 404.64 million UTI cases worldwide each year⁷. ESBL-producing E. coli consistently occurs at high levels in hospital settings and also spreads into water bodies, the food supply, and agricultural products⁸. The resistant strains persist in significant reservoirs in downstream water sources because these water bodies receive contamination from hospitals, industries, and residential waste⁹.

Over half to three-quarters of all antibiotics prescribed worldwide are used to treat E. coli infections with beta-lactam drugs, including penicillins, cephalosporins, and carbapenems⁶. Nevertheless, the misuse of antibiotics and repeated use of medications have rapidly led to the emergence of bacterial strains that resist multiple treatments. First reported in 1983, ESBL-producing bacteria have since spread globally, with the cefotaximase-Munich gene family now replacing Temoneira and sulfhydryl variable as the dominant ESBL types. The vital role of the aquatic environment in spreading resistant strains calls for immediate isolation and characterization of these strains to develop novel treatments¹⁰.

The activity of ESBL enzymes continues to contribute to antibacterial resistance across a wide variety of beta-lactam antibiotic classes^8,10,12. The approach using bacteriophage therapeutics holds promise as a replacement for antibiotics following its reintroduction as an alternative treatment mode. Phages specifically recognize and eliminate harmful bacteria without harming the beneficial microbes that exist in humans¹³. They are effective at low doses, evolve alongside bacterial resistance, and replicate at infection sites. The challenges of identifying the causative bacterial strain and ensuring phage efficacy have led to the development of phage cocktails⁵. Multiple phage strains within a single formulation work jointly to achieve maximal bacterial elimination while slowing the occurrence of phage resistance. Researchers now demonstrate that therapeutic phage mixtures show promise in combating antibiotic-resistant microorganisms, particularly those composed of multiple drug-resistant strains^4,5. Sustaining the potential of phage therapy requires extensive preparation to establish both safety procedures and treatment reliability. The removal of bacterial endotoxins, along with virulent genetic elements, is necessary to ensure the safety of phage preparations. The success of clinical phage preparations depends on their development as highly purified non-pyrogenic preparations¹⁴.

The main objective of this investigation is to combine several well-established molecular and microbiological methods into a simplified workflow for the identification, description, and evaluation of phage susceptibility in E. coli-producing ESBL from environmental water samples. Although each technique is conventional on its own, when combined into a single framework, they improve the workflow’s practical usefulness for targeted phage screening and environmental monitoring.

2. Materials and methods

2.1. Materials

The laboratory work used MacConkey agar, along with Mueller–Hinton (MH) agar, Luria–Bertani (LB) broth, and semi-solid agar (Oxoid, UK). Antibiotic disks used in this study included cefotaxime (CTX), ceftazidime (CAZ), clavulanic acid (CLA), and amoxicillin-CLA (AMC; Sigma-Aldrich, USA). Other materials used included sodium thiosulfate, along with crystal violet, iodine, ethanol, safranin, chloroform, and phosphate-buffered saline (PBS) (Merck, Germany). Analytical profile index for Enterobacteriaceae (API 20E) test strips and reagents (BioMérieux, France) were used in this study. Laboratory consumables, including filters and syringes, were procured from Millipore (USA). UV–Vis spectrophotometer (UV-1800, Shimadzu, Japan) was also used in this study.

2.2. Sampling area and sample collection

Water samples (6 L each) were collected from downstream sources in Islamabad and Rawalpindi, including Nala Lai Khayaban-e-Sir Syed, a stream near Chattar Park, Shahdara, and Nescom Hospital H-10. Sterilized jars were used to collect samples from just below the surface. Sodium thiosulfate was added to neutralize chlorine and preserve microbial viability. Samples were transported in a cooling box with ice packs at 4°C.

2.3. Preparation and inoculation of media

Bacterial culture was performed according to previous study ¹⁵. Distilled water (500 mL) was used to suspend 25 g of MacConkey agar, which was then heated to fully dissolve, followed by autoclaving for 15 minutes at 121°C. A stock solution (0.5 mL) containing 5 mg/mL CTX was added to the medium. The mixture (25–30 mL) was then transferred into sterilized 90 mm petri dishes. To resuspend particles, samples kept at 4°C were vortexed. Serial dilutions (10⁻¹ to 10⁻⁵) were performed by diluting a 1 mL aliquot in 9 mL PBS. Each dilution (100 µL) was added to prewarmed MacConkey agar plates (35–55°C, 30 minutes) and evenly distributed with sterile spreaders. The plates were incubated at 35–37°C for 18–24 hours.

2.4. Membrane filtration

A 0.45-micrometer filter was aseptically placed on the porous base using sterile forceps, after which the funnel was carefully placed above it. The highest dilution (10 mL) was filtered under vacuum, and the funnel walls were rinsed with sterile diluent. The membrane filter was then transferred onto an agar plate. Successive dilutions were filtered without intermediate sterilization, with the setup sterilized using alcohol and flame before processing new samples¹⁶.

2.5. Gram staining

A heat-fixed smear was prepared on a glass slide for Gram staining. Crystal violet (1 minute), iodine (1 minute), ethanol (5–10 seconds) decolorization, and safranin (1 minute) counterstaining were applied to the smear consecutively. The slide was then inspected under 10 × magnification for morphology and 100 × (oil immersion) for the Gram-staining response following washing and air drying.¹⁷

2.6. Biochemical tests

Standardized laboratory methods were used for biochemical assessments of the bacterial strain. The bacterium’s catalase activity was assessed using hydrogen peroxide, while its oxidase activity was tested using oxidase reagent. The generation of hydrogen sulfide, gas, and sugar fermentation was assessed using triple sugar iron agar. Kovac’s reagent was used to measure indole synthesis in tryptone broth, while urease activity was assessed in urea broth. The methyl red and Voges–Proskauer (VP) assays were used to detect mixed-acid fermentation and acetoin synthesis, respectively, while citrate utilization was evaluated on citrate agar^18-22.

2.7. Analytical profile index for Enterobacteriaceae

The API 20E test strip was conducted in accordance with the manufacturer’s protocol. In brief, the strip was filled with an isolated Enterobacteriaceae colony suspended in sterile distilled water. Certain compartments (arginine dihydrolase [ADH], lysine decarboxylase [LDC], ornithine decarboxylase [ODC], hydrogen sulfide [H₂S], and urease) received sterile oil. Following 18 hours of incubation at 37°C, color changes were used to record the results. As required, reagents were added, including 40% potassium hydroxide with alpha-naphthol for VP, ferric chloride for tryptophan deaminase (TDA), and Kovac’s for the indole test. The API 20E reading scale was used to determine if the results were positive or negative, and a seven-digit code was generated for organism identification and was then compared against the API 20E catalog²³.

2.8. Antibiotic sensitivity testing

A pure bacterial culture was inoculated into 5 mL of tryptic soy broth (TSB) and cultured at 37°C for 18 hours. The turbidity of 1 mL culture from the TSB tube was adjusted to meet a 0.5 McFarland standard. The standardized inoculum was uniformly swabbed onto MH agar, and antibiotic disks were added within 15 minutes. Plates were incubated at 37°C for 18 hours; inhibitory zones were measured in mm and compared against the Clinical and Laboratory Standards Institute (CLSI) standards²⁴.

2.9. Combination disk test                                                                                                               

The CTX and CAZ disks, alone and combined with CLA (CTX-CLA 30/10, CAZ-CLA 30/10), were placed on an MH plate. After incubation, the test was considered positive if the inhibition zones with CLA were ≥ 5 mm larger than those without²⁵.

2.10. Double-disk synergy test

A cephalosporin disk (CTX 30 or CAZ 30) was placed on an MH plate, with an AMC 30 disk positioned 20 mm center-to-center from the cephalosporin disk. A positive result was indicated if the inhibition zone around the cephalosporin disk expanded toward the AMC disk after incubation²⁶.

2.11. Polymerase chain reaction-based virulence gene detection

The DNA extraction was performed using the DNeasy Blood and Tissue Kit (Qiagen, Germany). The conventional polymerase chain reaction (PCR) was performed utilizing adhesion factor (fimH) primers, pyelonephritis-associated pili (papC) primers, hemolysin (hlyA) primers, and Shiga toxins (stx1/stx2) primers, with previously published primer sequences²⁷.

A 25 µL PCR reaction was prepared, containing nuclease-free water, PCR master mix, gene-specific primers, and template DNA. PCR amplification was performed under the cycling conditions described by Zimoń et al. ²⁷. Agarose gel electrophoresis on 1.5% agarose gels stained with ethidium bromide was used to analyze the PCR products. Expected fragment sizes were confirmed by comparing amplicon sizes against a 100 bp DNA ladder. It was considered a positive detection of the corresponding gene when a band of the expected size was present. To ascertain the accuracy of the amplification results, an E. coli strain that has previously been described as virulence-positive was added as a positive control, while water devoid of nuclease served as a negative control. The negative control was prepared by replacing the template DNA with nuclease-free water.To ensure that all isolates had effective PCR amplification and intact DNA, a housekeeping gene (such as the 16S rRNA gene) was amplified.

2.12. Serogroup determination via multiplex polymerase chain reaction

Using previously published primers, multiplex conventional PCR was used to determine the O-antigen serogroup with a focus on O15, O25, and O157-specific gene sequences²⁷. PCR and electrophoresis were conducted as mentioned in Section 3.10. On 1.5% agarose gels, amplicons were observed, and the results were interpreted based on anticipated product sizes. In each PCR run, nuclease-free water was used as the negative control, and the previously identified serogroup-specific E. coli reference strains (O15, O25, and O157) as positive controls.

2.13. Bacteriophage isolation

2.13.1. Sample collection

Water samples were obtained from different locations across Islamabad and Rawalpindi, starting at Nala Lai, then at Khayaban-e-Sir Syed, followed by streams surrounding Chattar Park, and finally at G-10 and Nescom Hospital H-10. Wastewater samples were collected in autoclaved flasks and processed immediately.

2.13.2. Bacteriophage enrichment

Sewage samples were allowed to settle for 15 minutes, and heavy particles were removed. The sample was centrifuged at 4,000 rpm for 5 minutes. A mixture of 10 mL autoclaved 5 × LB broth, 40 mL centrifuged sewage sample, and 1 mL 24-hour-old bacterial culture was incubated at 37°C for 24 hours with shaking at 120 rpm. After incubation, 1% chloroform was added, followed by centrifugation at 11,000 rpm for 5 minutes to remove debris²⁸.

2.13.3. Double-layer agar method

Phage lysate was serially diluted, mixed with 100 µL of exponentially growing bacteria, incubated at 37°C for 15–20 minutes, and poured onto LB agar plates. Semi-solid LB agar (3–4 mL) was added, mixed, and left to solidify. For phage quantification, a top agar containing 0.7% agar was used to ensure consistency across the assays. Plates were incubated at 37°C for 18 hours. Phage presence was indicated by plaque formation²⁸.

2.13.4. Phage purification

Isolated plaques were picked using sterile micropipette tips, propagated in 25 mL LB broth with 1 mL host bacteria, and incubated at 37°C overnight. After adding chloroform  to a final concentration of 1% (v/v), the mixture was centrifuged at 10,000 rpm for 10 minutes. The supernatant was filtered through a 0.45 µm syringe filter, labeled as phage lysate, and purified over 5–10 cycles. Purity was confirmed via the double-layer agar method, and plaques were stored at 4°C with 5% glycerol. The phage isolated via this procedure was designated as NHE-1 and was propagated on ESBL E. coli NLA-1²⁹.

2.14. Spot assay

To test phage activity, 100 µL of a 24-hour bacterial culture was poured onto LB agar plates containing 3–4 mL of semi-solid agar. After solidifying, 5 µL of phage lysate was spotted and absorbed onto the plates, which were then incubated at 37°C for 24 hours. Clear zones indicated phage efficacy against ESBL-producing E. coli NLA-1³⁰.

2.15. Bacterial growth reduction assay

To evaluate lytic potential, two flasks with 50 mL LB broth were prepared. The experiment contained two flasks, including a bacteria-phage mix at MOI 10, along with a bacterial-only control flask. Both flasks were incubated at 37°C with shaking at 120 rpm. Every two hours, 2 mL samples were collected, and the optical density (OD) at 600 nm was measured. A graph of OD versus incubation time was plotted to assess phage lytic activity³¹.

2.16. One-step growth curve analysis

The lytic ability and replication efficiency of bacteriophage NHE-1 were assessed by analyzing its one-step growth kinetics at multiplicities of infection (MOIs) of 1 and 10. Phage NHE-1 was introduced into exponentially dividing E. coli cells, which were then left to adsorb for 10 minutes at 37°C. The infected cultures were resuspended in fresh media and incubated under standard conditions after the unadsorbed phages were removed via centrifugation. Phage titers were calculated using the double-layer agar method, and samples were taken at regular intervals. Equation 1 was used to determine the burst size:

The results were shown as mean ± standard deviation, and each experiment was conducted in three biological replicates. To ensure measurement consistency, technical duplicates were included with every replication.

2.17. Time kill assay of phages against extended-spectrum beta-lactamase- and non-extended-spectrum beta-lactamase-producing E. coli strains

The effectiveness of phages at various MOI against an ESBL-producing (NLA-1) and a non-ESBL-producing E. coli strain (SHD-2) was assessed using a time kill experiment. Samples were collected at predetermined intervals (0, 1, 2, 4, 6, 8, and 12 hours) after NHE-1 phages were applied to bacterial cultures at MOIs of 1 and 10. To calculate colony-forming units (CFU), the bacterial cultures were serially diluted and plated on LB agar. CFU/mL was determined for each time point after the plates were incubated for 24 hours at 37°C³².

2.18. Statistical analysis

All experiments were repeated twice under controlled conditions in triplicate to validate the results. The results were analyzed using Excel (Version 2016, Microsoft Corporation, USA), and the standard error was calculated. All graphics were produced using Origin software (8.5, OriginLab Corporation, USA). Since the primary goal of the study was to describe the isolates and phage characteristics, no inferential statistical tests were used.

3. Results

Table 1 presents E. coli detection results for 15 water samples collected at different sites, including the number and proportion of positive and negative findings per location.

Based on its source, each isolated strain was provided with a unique designation:

• NES-1 to NES-3 (Nescom Hospital H-10)
• G10-1 to G10-2 (G-10)
• SHD-1 to SHD-2 (Shahdara)
• NLA-1 to NLA-3 (Nala Lai)

The E. coli and ESBL-producing E. coli CFU/mL counts are shown in Table 2, which reveal microbial frequencies and resistance levels in environmental water sources. Standard biochemical tests and selective medium were used to identify presumed E. coli isolates. Using the double-disk synergy test in accordance with CLSI standards, phenotypic confirmation of ESBL production was performed. An enhanced inhibition zone toward the AMC disk following 18 hours of incubation at 37°C was interpreted as positive for ESBL production.

3.1. Gram staining

A peptidoglycan layer combined with an outer membrane defines the composition of E. coli. The staining process allows E. coli to take up safranin dye, which turns the bacterium pink, as shown in Figure S1.

3.2. Biochemical tests

Table 3 provides a clear overview of the biochemical tests conducted to characterize E. coli, confirming the presence of E. coli in the analyzed water samples (samples 1–10). Visual results are shown in Figures S2-S6.

3.3. Analytical profile index for Enterobacteriaceae

Biochemical tests demonstrated the presence of E. coli through valid o-nitrophenyl-β-D-galactopyranoside, LDC, and ODC results and four specific sugar fermentation outcomes, confirming lactose utilization and the activity of decarboxylase and other sugar-metabolizing enzymes. The bacterium showed negative results for ADH, TDA, H₂S production, gelatinase activity, and specific fermentation tests (inositol, amygdalin, melibiose), confirming its absence of enzymatic functions. The API 20E test results identified E. coli with 99.9% accuracy when evaluating the generated code (5144532), as shown in Table 4 and Figure S7.

3.4. Disk diffusion test

E.coli isolate15s exhibited resistance to AMC (30 µg), tazobactam (30 µg), and ciprofloxacin (CIP; 15 µg), while showing sensitivity to cefixime (30 µg), ceftriaxone (CRO; 30 µg), CTX 30 µg, levofloxacin (30 µg), netilmicin (30 µg), aztreonam (20 µg), and imipenem (IMP; 10 µg), as shown in Figure S8 and Table 5.

3.5. Antibiotic sensitivity tests

Isolates of ESBL-producing E. coli were subjected to an antibiotic sensitivity test. A total of 10 agents from different antibiotic classes were used. Several studies have shown an increasing trend in bacterial resistance to these antibiotics. Antibiotics for veterinary and human health were also considered. A summary of the percentages of ESBL-producing E. coli resistant to these antibiotics is shown in Table 6.

3.6. Antibiotic resistance profiles of extended-spectrum beta-lactamase-producing E. coli samples from downstream water samples

The ESBL-producing E. coli showed sensitivity to IMP (10 µg) with inhibition zones ranging from 31 mm to 35 mm and an average resistance of 5.32%, meeting CLSI guidelines for sensitivity. Results showed resistance to CTX (30 µg) when combined with CTX/CLA (30 µg), as well as CAZ (30 µg), when used with CAZ/CLA (30 µg) and CRO (30 µg), as shown in Figures S9 and S10.

3.7. Polymerase chain reaction-based virulence gene detection

The presence of papC (50%), fimH (80%), and hlyA (30%) was detected using conventional PCR virulence testing, as shown in Table 7. The genes hlyA, papC, fimH, and Stx1/Stx2 were identified using conventional PCR. The percentage of isolates that test positive for each gene is shown as detection (%).

3.8. Serogroup determination using multiplex polymerase chain reaction

Genes specific to the O15, O25, and O157 serogroups were identified using multiplex PCR. Among the ESBL-producing Escherichia coli isolates (n = 4), serogroup O15 was detected in 50% of isolates, while O25 and O157 were each detected in 25% of isolates (Table 8).

3.9. Bacteriophage isolation

3.9.1. Double-layer agar results

The double-layer agar method detected various plaque morphologies, which implied the presence of multiple bacteriophage populations in the water. Plaque appearance on agar plates indicates bacterial death due to phage infection, yet plaque sizes depend on the phage strain, bacterial receptor preference, and the physicochemical properties of the growth medium, as shown in Figure 2.

Figure 2. Double-layer agar method indicating the presence of bacteriophages (plaque formation).

3.9.2. Phage purification

A serial dilution of bacteriophage culture was performed to isolate phages of uniform morphology and size. After four-fold dilution and culturing on fresh LB media, uniform morphology was confirmed using the double-layer agar method (Figure 3).

Figure 3. Double-layer agar plates showing uniform morphology.

3.9.3. Spot assay

Fifteen samples were screened for bacteriophages using ESBL-producing E. coli NLA-3 as the host. A phage isolated from stream water near Nescom Hospital in Islamabad was named NHE-1 and tested using a spot assay. Circular plaques measuring 3 mm in diameter were observed, confirming the effectiveness of NHE-1 against ESBL-producing E. coli NLA-3. Larger plaques indicated the phage with the highest lytic activity, as shown in Figures 4 and S11.

Figure 4. Potential phage against extended-spectrum beta-lactamase-producing Escherichia coli NLA-1.

3.10. Reduction assay results

A reduction assay was conducted to assess the lytic activity of NHE-1 phage against ESBL-producing E. coli NLA-1. Initially, both the control and MOI 10 flasks were turbid; over time, the MOI 10 flask became clearer. The OD of the control flask remained high, while the OD of the NHE-1-infected flask stayed constant, indicating bacterial inhibition. The control showed constant bacterial growth as measured in OD₆₀₀ over 24 hours, whereas NHE-1-infected E. coli NLA-1 showed limited growth, indicating the phage’s lytic activity, as shown in Figure 5.

Figure 5. Effect of NHE-1 phage on the log phase of extended-spectrum beta-lactamase-producing Escherichia coli culture compared to the control. Abbreviation: MOI: Multiplicity of infection.

The one-step growth curve (Figure 6) of bacteriophage NHE-1 showed three phases:

• During the eclipse phase (0–30 min), intracellular phage multiplication took place without any discernible extracellular virions.
• In the burst phase (around 30–40 minutes), a sharp rise in phage titer was caused by the quick release of progeny phages.
• Phage counts stabilized at the plateau phase (> 40 min), signifying the completion of the lytic cycle.

The eclipse phase lasted for around half an hour. The calculated burst size at both MOI values (MOI 1 and MOI 10) was approximately 6,000 ± 150 plaque-forming units per infected cell, indicating that NHE-1 replicated efficiently in the host strain. All biological replicates exhibit a high burst size, likely due to the phage’s rapid intracellular genome replication and well-optimized host–phage interactions. Phage therapy relies on this curve as a key tool for studying bacterial virus replication patterns, as shown in Figure 6.

Figure 6. One-step growth curve of bacteriophage NHE-1.

3.10.1. Time kill assay of phages against extended-spectrum beta-lactamase- and non-extended-spectrum beta-lactamase-producing E. coli strains

Data indicated that elevated MOI values led to effective bacterial reduction over the observation period. ESBL-producing E. coli NLA-1 at MOI 10 exhibited complete eradication in eight hours, whereas MOI 1 showed slower bacterial reduction. Non-ESBL-producing strain SHD-2 showed a gradual decline, with CFU remaining detected at 12 hours, as shown in Figure 7.

Figure 7. Time kill assay of phage activity at different MOIs (MOI 1 and MOI 10). Abbreviations: E. coli: Escherichia coli; ESBL: Extended-spectrum beta-lactamase; MOI: Multiplicity of infection.

3.10.2. Host range and applicability

Phage NHE-1 was tested for lytic activity against two E. coli isolates: ESBL- and non-ESBL producing E. coli. Although both cases demonstrated successful lysis, this represents only an initial evaluation of the host range. Therefore, any conclusions regarding broader environmental or therapeutic applicability should be interpreted with caution. For a thorough assessment of the host range of phage NHE-1, a broader and more genetically varied panel of E. coli strains should be used.

4. Discussion

Antibiotic resistance in bacteria is progressing drastically, and these emerging infectious agents pose serious threats to human health. There is a rapid increase in the number of multiple-drug-resistant bacteria due to horizontal gene transfer. Moreover, these bacteria are resistant to third-generation cephalosporins, and the only treatment options are last-resort antibiotics. The current study aims to integrate several well-established molecular and microbiological methods into a streamlined process to identify, characterize, and assess the phage susceptibility of ESBL-producing E. coli from environmental water sources. Although each technique is well-established on its own, when combined into a cohesive approach, they improve the workflow’s practical efficiency and applicability, making it especially useful for environmental surveillance and targeted bacteriophage screening. By carefully combining traditional methods, this integrated approach offers a thorough and rapid assessment platform for monitoring antibiotic resistance.

The isolation of E. coli from five locations indicated significant contamination at sites such as Nala Lai and Nescom H-10, with 100% of samples testing positive for E. coli. Conversely, samples from Chattar Park exhibited no contamination. These findings align with studies by Beshiru et al. ³³, who reported higher contamination rates in urban water bodies compared to recreational or rural areas. The observed distribution underscores the influence of anthropogenic activities on microbial contamination levels.

Results from antibiotic susceptibility testing revealed varying levels of resistance among ESBL-producing E. coli. Resistance to AMC and CIP was consistent with findings by Yanestria et al.³⁴ and Davies and Everett³⁵, who noted similar resistance patterns in waterborne E. coli. However, sensitivity to IMP aligns with previous reports highlighting carbapenem effectiveness against ESBL strains^35,36. These results emphasize the need for stringent monitoring of antibiotic use to curb resistance development. The biochemical characterization of E. coli isolates confirmed their ability to ferment glucose, lactose, and sucrose, with positive results for indole and catalase tests and negative results for the urease and oxidase tests. These findings are consistent with the typical biochemical profile of E. coli reported in earlier studies by Shoaib et al.²². Furthermore, the API 20E test identified ESBL-producing E. coli with a high probability (99.9%), validating the accuracy of biochemical and enzymatic tests in identifying resistant strains. Similar conclusions were drawn by Abdullah and Al-Taee³⁷, who highlighted the reliability of the API 20E test in detecting pathogenic E. coli in environmental samples. Moreover, the findings demonstrated that 80% of isolates possessed the virulence gene fimH and 50% contained papC, indicating their ability to adhere strongly and act as uropathogens. The strains lacked Shiga toxin genes (stx1/stx2), indicating they do not produce enterotoxigenic effects but may still contribute to UTIs³⁸.

The results highlighted that the O15 and O25 serogroups were the most abundant, at 40% and 30%, respectively, as these serotypes frequently emerge as antibiotic-resistant pathogens in medical environments. The surveillance further identified O157 in 30% of bacterial strains, which typically cause foodborne outbreaks, aligning with the findings by Pearse et al.³⁹. The bacteriophage NHE-1 showed significant lytic activity against ESBL-producing E. coli NLA-1, as evidenced by the spot assay and time kill curve, which demonstrated dose-dependent inhibition of bacterial growth. These results are in line with the work of by Siopi et al. ⁴⁰, who reported similar dose-dependent reductions in multidrug-resistant E. coli using bacteriophages. The reduction assay further confirmed the sustained antibacterial activity of NHE-1, as evidenced by the decline in the OD of treated samples compared to controls. Results from double-layer agar tests showed different plaque morphologies that reflect specific phage-host interplay patterns by Stachurska et al. ²⁸, who emphasized the importance of host specificity in phage therapy. The time kill assay experiments demonstrated that treatment with phages effectively suppressed bacterial growth. The results suggest that isolated bacteriophages were more efficient against ESBL-producing E. coli NLA-1 than non-ESBL-producing E. coli SHD-2. This was likely due to a difference in their mechanisms of resistance. This rapid reduction in bacterial counts at higher MOI aligns well with a previous study by Zhang et al.⁴¹ that outlined rapid phage-mediated lysis at elevated MOIs. Nevertheless, the pronounced depletion of non-ESBL-producing E. coli SHD-2 may indicate variability in susceptibility due to strain-specific factors.

Previous studies on bacteriophage therapy have highlighted its potential as a complementary or alternative treatment for antibiotic-resistant infections^28,34,41. A better understanding of why phages exhibit variable effects across different environments, host strains, and bacterial resistance patterns remains an important area of ongoing investigation. Phage NHE-1’s lytic effectiveness was evaluated against a strain of E. coli that produced ESBL and one that did not. This is a preliminary assessment of the phage’s host range, even though both cases showed efficient bacterial clearance. Therefore, broader conclusions regarding its therapeutic or environmental relevance should be drawn with caution. It will be necessary to test NHE-1 against a broader range of genetically diverse E. coli strains to fully determine its host range.

The study provides important findings; however, its primary limitation is the restricted sampling scope, as only five sites across two regions were examined. Future studies should involve larger sample sizes, while researchers should investigate genetic factors that govern phage–host relationship dynamics. The evaluation of environmental factors that affect phage functionality will advance our understanding of the application of bacteriophage therapy across different environments.

5. Conclusion

The analysis identifies extensive E. coli contamination along water sources throughout urban environments, which are shaped by human activities. The antibiotic susceptibility results indicated resistance to several commonly used antibiotics, underscoring the need for rigorous antibiotic management approaches. The isolation of bacteriophage NHE-1 demonstrated its strong ability to eliminate ESBL-producing E. coli, making NHE-1 a promising therapeutic option for combating antibiotic-resistant E. coli. The spot assay, combined with the reduction assay and time kill curve analysis, strengthened the case for phage effectiveness in eradicating bacteria. However, it will be necessary to test NHE-1 against a broader range of genetically diverse E. coli strains to fully determine its host spectrum.

Further research should focus on scaling up phage applications while studying bacterial–host genetic interactions and environmental factors that affect phage effectiveness. Integrated surveillance methods with novel strategies are essential to slow the growing threat of waterborne antimicrobial resistance, according to this investigation.