Bacteria were isolated from pond water and diseased fish (12–15 cm in length) near Varanasi, Uttar Pradesh, India. A total of 65 diseased fish were identified by swimming abnormality, pale gills, skin ulceration, and wounds on the body and tail. Specimens for isolation of bacteria were taken from the pond water sample, organs (gills, liver, kidney, ovary, and skin) of diseased fish during postmortem, and swabs from wounds. The culture was performed using blood agar (HiMedia Laboratories, Maharashtra, India) and MacConkey agar (HiMedia Laboratories) and incubated at 37°C overnight. The next day, the plates were examined for growth.

The circular off-gray colonies from blood agar and non-lactose fermenting colonies from MacConkey were subjected to Gram staining and biochemical testing for such things as motility, oxidase production, sugars fermentation, gas production, susceptibility to vibriostatic agent O/129 (decarboxylase testing), and esculin hydrolysis for the identification of A. hydrophila (Samal et al., 2014).

Biochemically identified Aeromonas species were subcultured on Mueller–Hinton agar (MHA; HiMedia Laboratories), and then DNA extraction was performed using the phenol–chloroform–isoamyl alcohol (25:24:1) method (Sambrook and Russell, 2006). The isolated genomic DNA was subjected to amplification using 16S rDNA-specific primers targeting genus-specific sequences of Aeromonas species (forward 5′-GGG AGT GCC TTC GGG AAT CAG A-3′ and reverse 5′-TCA CCG CAA CAT TCT GAT TTG-3′) (Hussain et al., 2014). The positive strains for Aeromonas species were further amplified by using A. hydrophila species-specific primers Ahh1 (forward 5′-GCC GAG CGC CCA GAA GGT GAG-3′ and reverse 5′-GAG CGG CTG GAT GCG GTT GT-3′) (Hussain et al., 2014). The master mix (Sigma-Aldrich, St. Louis, MO, USA) consisted of 50 ng of bacterial genomic DNA. The reaction mixture (25 µL) was subjected to 35 cycles on the following program: an initial melting temperature of 95°C for 5 min, with denaturation at 94°C for 45 s, annealing at 59°C for 45 s, and extension at 72°C for 1 s. The final extension was carried out at 72°C for 7 min in a thermal cycler (Bio-Rad Universal Hood II, Hercules, CA, USA). Then, PCR products were visualized on 1% agarose gel electrophoresis.

Phages were isolated from the pond, river, and sewage water by using the soft agar (0.8%) overlay method, following the method described by Kutter (2009), with some modifications. A. hydrophila was plated in a lawn culture (1.5 × 10⁸ CFU mL^–1) to isolate bacteriophages on MHA and incubated for 6 h to reach the log phase. Water specimens from different sources were collected and treated with 1% chloroform (Sigma-Aldrich) for 10 min and centrifuged three times for 15 min at 10,778 × g. One milliliter of the supernatant was flooded on the 6-h bacterial lawn of the A. hydrophila on an MHA plate and incubated overnight at 37°C. The next day’s lawn was washed with Tris-HCl magnesium gelatin (TMG, pH 7.4) buffer and centrifuged at 10,778 × g for 15 min.

Isolated phages were purified by using plaque counting by applying the 0.8% soft agar overlay method. The single isolated plaque was picked up for further processing (Orlova, 2012). The phage count was increased by lawn and spot methods (Marcó et al., 2012). For purification (toxin-free), the harvested fluid was subjected to membrane dialysis (membrane pore size 20 nm, HiMedia Laboratories) against poly ethylene glycol (PEG) 20% in 2.5 M sodium chloride (NaCl) solution (HiMedia Laboratories) overnight at 4°C and then washed with phosphate buffer saline. The dialysis and washing process was repeated twice (Gangwar et al., 2021).

For their bacteriolytic activity, different isolated phages were subjected against different isolates of A. hydrophila. In addition, the lawn culture of A. hydrophila (1.5 × 10⁸ CFU mL^–1, 0.5 McFarland) was prepared on MHA. Ten microliters of each phage with the concentration of 1 × 10⁹ PFU mL^–1 was spotted on the MHA (Montso et al., 2019). The MHA plates were observed for the lysis of bacteria (clear zone) after incubation at 37°C overnight. Then, the three most active phages (i.e., φAHBHU12, φAHBHU16, and φAHBHU19) were selected for further characterization and experiments.

To see the spectrum of activity of these three phages (φAHBHU12, φAHBHU16, and φAHBHU19), we used different American Type Culture Collection strains of bacteria. The lawn culture of Pseudomonas aeruginosa, Aeromonas sobria, Escherichia coli, Plesiomonas shigelloides, Enterococcus faecalis, Salmonella Typhi, Acinetobacter lwoffii, Enterobacter cloacae, and Staphylococcus aureus (1.5 × 108 CFU mL^–1) were made on MHA. The lytic activity of phages against all the bacterial species was determined by spot assay. Spot assay was used to determine the lytic spectrum activity of phage isolates, as described previously.

The bacteriophage strains were screened for their lytic activity at different pH ranges (3-12). Bacteriophages (1 × 10⁹ PFU mL^–1 final concentrations) were incubated in equal amounts of TMG buffer (1:1) of different pH at 4°C for 2 h. After incubation, the lytic activity of bacteriophages was checked by the spot assay method.

Activities of bacteriophages were screened at different temperatures. For this screening, phages (1mL; 1 × 10⁹ PFU mL^–1 final concentrations) were incubated at different temperatures (–80°C, –20°C, 4°C, 28°C, 37°C, 45°C, and 55°C) for 48 h, and then spot assay was performed to check its activity.

The most virulent phages (i.e., φAHBHU12, φAHBHU16, and φAHBHU19) with titer 1.0 × 10¹¹ PFU mL^–1 were filtered through a 0.22-µm syringe filter. The phage suspension was centrifuged at 21,124 × g for 90 min, and the supernatant was decanted. The pellet was washed three times with 0.1 M ammonium acetate (HiMedia Laboratories, pH 7.0). At the final step, the pellet was resuspended in ammonium acetate. The samples were treated with 2% uranyl acetate for negative staining and carbon-coated formvar films and examined under transmission electron microscopy (TALOS, Thermo Scientific, AIIMS, New Delhi, India). The diameter of the head and tail of these phages was determined by ImageJ software.

Bacteriophage DNA was extracted and purified using a Norgen Biotek Corp. kit (cat. no 46800). Further phage DNA was subjected to restriction digestion using AluI (10 U µL^–1) in accordance with the manufacturer’s instructions (Thermo Fisher Scientific) for 60 min at 37°C in a 20-μL reaction mixture containing 1 μL of DNA (concentration 1 μg μL^–1), 2 μL of 10 × buffer, 1 μL of restriction enzyme, and 16 μL of sterile water. The digested products were electrophoresed with 1% agarose gel at 80 V for 1 h. The gel images were captured under ultraviolet light using a gel documentation system (BioRad Universal Hood II). The sizes of the DNA bands were estimated using 1 kb and 100 bp DNA ladders (Invitrogen, Thermo Fisher Scientific).

The genomic DNA of the phages against A. hydrophila was subjected to enterobacterial repetitive intergenic consensus (ERIC) PCR. The master mix contained 50 ng of phage genomic DNA. The primers were used to amplify phage genomic DNA (forward 5′-ATG TAA GCT CCT GGG GAT TCA-3′ and reverse 5′-AAG TAA GTG ACT GGG GTG AGC G-3′) (Ranjbar et al., 2017). The PCR mixture was subjected to the following program for amplification: melting temperature at 94°C for 7 min with denaturation for 45 s at 92°C, annealing at 31.8°C for 45 s, and extension at 72°C for 1 s. A final extension step at 72°C for 7 min was performed at the end of the 34 cycles. The PCR product was run on 1% agarose gel electrophoresis with 100 bp and 1 kb DNA ladder.

The experimental fish, P. buchanani, weighing 12–16 g, 12–15 cm in size, and 6–8 weeks old, were procured from commercial fish farms in Varanasi. The fish were reared in well-aerated 40-L rectangular glass aquariums in the Department of Microbiology, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India. The permission of the institutional ethics committee was obtained (reference number Dean/2016/CAEC/70/dated 30 March 2017). Before the experiment, the fish were acclimatized for 1 week. Then, the fish were immersed in 0.01% potassium permanganate (HiMedia Laboratories) for 10 min to remove the parasitic infection (Prasad et al., 2011). The aquatic environment of the tank was maintained with dissolved oxygen 8.0 ± 0.5 mg^-1, ammonia 0.6 ± 0.05 mg^-1, pH 7.2 ± 0.2, and temperature 27°C ± 2°C without chlorination. Each aquarium was aerated with an air pump, and one-third of the water was replaced daily, dead fish were removed, and debris was siphoned from the bottom of the aquarium.

Two different routes of administration [i.e., intramuscular (IM) and water immersion] were used to establish A. hydrophila infection in the fish.

The phage cocktail was used for prophylactic and therapeutic purposes. The fish experiments were set up according to the following plan:

All the experiments were repeated three times for robustness of data.

Experimental groups were challenged with 100 μL of A. hydrophila by injecting 8.0 × 10⁵ CFU/fish IM. The bacteriophage cocktail at the doses of 1.0 × 10³, 1.0 × 10⁴, 1.0 × 10⁵, 1.0 × 10⁶, 1.0 × 10⁷, and 1.0 × 10⁸ PFU/fish were given IM simultaneously at a different site from that of the bacterial injection.

Each group of 10 fish was placed in the 10-L aquarium. Fish were challenged with 100 μL of A. hydrophila at the dose of 8.0 × 10⁵ CFU/fish through the IM route, and phage cocktails at the quantity of 1.0 × 10⁴ PFU/fish were given simultaneously after 6, 12, and 24 h of bacterial challenge IM. The water was changed daily, as described elsewhere, and the experiment was monitored for 7 days.

Ten groups of P. buchanani were placed in separate aquaria. The fish were challenged with 100μL of A. hydrophila at a concentration 8.0 × 10⁵ CFU/fish through the IM. The phage cocktail at different concentrations (i.e., 1.0 × 10⁴, 1.0 × 10⁵, 1.0 × 10⁶, 1.0 × 10⁷, 1.0 × 10⁸, and 1 × 10⁹ PFU mL^–1) was administered through water immersion simultaneously.

The groups, comprising 10 fish each, were challenged with 100 μL of A. hydrophila (8.0 × 10⁵ CFU/fish) through the IM route. However, phage cocktails containing 1.0 × 10⁸ PFU mL^–1 were added into the water tank after 6, 12, and 24 of bacterial challenge.

In water immersion, 1.0 × 10⁷ CFU mL^–1 bacterial suspension was added to the 2-L water aquaria with 10 mL of LB broth, and phages cocktails were given simultaneously at a concentration of 1.0 × 10³, 1.0 × 10⁴, 1.0 × 10⁵, 1.0 × 10⁶, 1.0 × 10⁷, 1.0 × 10⁸, and 1.0 × 10⁹ PFU mL^–1, with 10 mL of LB broth in 2 L by adding in water. Furthermore, the phage cocktail (1.0 × 10⁶ PFU mL^–1) was given at 6, 12, and 24 after adding bacteria into the water.

A total of 18 strains of A. hydrophila were tested against 23 bacteriophages by spot assay. In detail, the most virulent bacteriophages, φAHBHU12 (72.2%), φAHBHU16 (66.6%), and φAHBHU19 (83.3%), were used for further experiment (see Table 1 ).

The most virulent bacteriophages [φAHBHU12 (lysing 72.2%), φAHBHU16 (66.6%), and φAHBHU19 (83.3%)] were selected for further characterization. The three phages could not lyse Pseudomonas aeruginosa, Aeromonas sobria, Escherichia coli, Salmonella Typhi, Acinetobacter lwoffii, Enterobacter cloacae, Plesiomonas shigelloides, Enterococcus faecalis, and Staphylococcus aureus.

Good lytic activity of the bacteriophages against A. hydrophila was observed at pH 3–12 (see Figure 2A ), and the three phages could survive well at temperatures –80°C, –20°C, 4°C, 28°C, and 37°C. However, φAHBHU12 and φAHBHU19 had satisfactory activity (approximately 50%) at 45°C (see Figure 2B ).

The bacteriophages φAHBHU12, φAHBHU16, and φAHBHU19 were examined under transmission electron microscopy cryo-TEM (Talos, Thermo Fisher Scientific, Waltham, MA, USA). Based on previous classification (Dion et al., 2020), the φAHBHU12 could be placed in the Podoviridae family, as the diameter of the isometric head was 52.47 nm, and a non-contractile short tail could not be visualized (see Figure 3A ). In contrast, bacteriophage φAHBHU16 belongs to the Siphoviridae family, having an icosahedral head (47.82 nm in width) and a long non-contractile tail ranging from 93.55 to 117.30 nm in length (see Figure 3B ). The φAHBHU19 could be classified as belonging to the Corticoviridae family, having an icosahedral head (42.17 nm in width) and no tail (see Figure 3C ).

The isolated three phages (φAHBHU12, φAHBHU16, and φAHBHU19) were further characterized at the level of whole-genome fingerprinting by restriction digestion and ERIC PCR of their genomic DNA. Different banding patterns were indicated in the restriction digestion with AluI ( Figure 4A ). Also, in ERIC PCR, distinct band patterns were observed in all three phages ( Figure 4B ). The results delineate that these three phages have different genetic characteristics.

The isolated potent phages were further characterized at the genomic level using RAPD-PCR and restriction digestion by isolating their genomic DNA. Different banding patterns were observed in the RAPD-PCR analysis ( Figure 3A ). Also, the restriction digestion with EcoRI shows a distinct band pattern in all three phages ( Figure 3B ). The results delineate that the isolated potent phages have different genetic makeup.

The lethal dose of A. hydrophila killing the P. buchanani fish weighing 12–16 g on intramuscular injection in 7 days was 8.0 × 10⁵CFU/fish. However, the lethal dose through the water immersion route was 1.0 × 10⁷ CFU mL^–1 within 7 days when mixed with 10 mL of LB broth containing 2 L of the water tank as organic contamination. Interestingly, no death could be observed in the absence of LB broth (organic matter).