Isolation and characterization of a lytic bacteriophage against Pseudomonas aeruginosa

Plaque and phage morphology of DRL-P1

Isolated phage was initially screened against P. aeruginosa through spot tests. A clear zone over a bacterial lawn was observed due to the lytic activity of the phage (Fig. 1a). This phage was named ‘DRL-P1’. On double-layer agar (DLA) plates, DRL-P1 produced small but clear plaques of approximately 2 ± 0.23 mm diameter (n = 56) of similar morphology (Fig. 1b). For subsequent characterization, bacteriophage enrichment was performed by repeated plaque purification method and a stock of 10⁹ PFU mL⁻¹ was prepared (Fig. 1c). Purified phage particles were examined under a transmission electron microscope (TEM) and following International Committee on Taxonomy of Viruses (ICTV) guidelines, classified as a member of the bacteriophage family Myoviridae, signified by a head, neck, contractile tail, base plate, tail fiber geometry (Fig. 2). The average (n = 9) particle length (head to the base plate), head diameter, head length, tail length, tail diameter, and base-plate diameter was measured to be 197.47 ± 1.72, 68.89 ± 2.37, 93.03 ± 2.85, 94.54 ± 2.52, 16.82 ± 1.80, and 27.92 ± 3.03 nm, respectively.

Antibiotic sensitivity and host range

The antibiotic response pattern of the bacterial isolate used in this study is tabulated as supplementary data (Table S1). The isolate showed resistance against Ceftazidime (CAZ), Nitrofurantoin (NIT), Nalidixic acid (NA), Ampicillin (AMP), Co-Trimoxazole (COT). However, it was found to be sensitive to Ciprofloxacin (CIP), Amikacin (AK), Amoxyclav (AMC), Cefotaxime (CTX), and Gentamicin (GEN), while intermediate sensitivity was noted against Netillin (NET), Tobramycin (TOB).

The host range of the phage DRL-P1 was examined against various standard bacterial cultures, obtained from the Microbial Type Culture Collection and Gene Bank (MTCC, Institute of Microbial Technology, Chandigarh, India). Phage DRL-P1 did not show any lytic activity against Escherichia coli (MTCC 443), Vibrio cholera (MTCC 3904), Bacillus megaterium (MTCC 428), Shigella flexneri (MTCC 1457), Bacillus subtilis (MTCC 1305), Salmonella enterica Typhimurium (MTCC 1251, MTCC 1252), Streptococcus pyogenes (MTCC 442) and Klebsiella pneumoniae (MTCC 8911). However, DRL-P1 showed clear lytic activity against P. aeruginosa (MTCC 1688) and also against nine other P. aeruginosa isolates (IS1-IS9), field-collected from the soil of Arunachal Pradesh, a northeastern state of India (Supplementary data, Table S2).

Features of the DRL-P1 genome

Next-Generation Sequencing of the phage DNA resulted in the generation of a total of 1,295,948 raw reads (read length 150) amounting to 194.4 Mb bases. After sequence QC, a total of 1,221,536 reads (178.93 Mb bases) were used to assemble a terminally redundant genome of 66,243 nts having GC content of 54.9%, consisting of 22.75% A, 22.31% T, 27.52% G, and 27.40% C. The genome sequence was predicted to be ‘intact’ (completeness score 120) in the PHASTER analysis. In Blastn (Megablast) search, the DRL-P1 genome sequence was found to be significantly similar (up to 97.77% nucleotide identity over 99% query coverage) to Pseudomonasphage genome sequences belonging to the genus Pbunavirus (Order Caudovirales; Family Myoviridae), with top 10 hits namely being isolates- ‘DL52’ (KR054028), ‘misfit’ (MT119367), ‘zikora’ (MW557846), ‘R26’ (NC_048663), ‘datas’ (NC_050143), ‘Epa 14’ (NC_050144), ‘billy’ (MT133563), ‘elmo’ (MT119364), ‘kraken’ (KT372692), ‘Jollyroger’ (KT372691).

A total of 93 phage-hit ORFs were identified in the genome, of which 36 were functionally annotated based on homology with similar phage proteins available in the databases, while 57 were annotated as phage hypothetical proteins. Predicted ORFs were found to encode proteins ranging from 31 to 1035 aa in length, the largest being the DNA polymerase (Table 1). Identified ORFs included genetic regions, responsible for encoding proteins related to virion structure, genome replication, assembly & packaging, DNA synthesis & repair, regulation of gene expression, host identification & infection, host lysis, and recombination that are essential for the phage cycle. Among the 93 ORFs, 54 (58%) and 39 (42%) ORFs were encoded on each of the strands of the dsDNA, respectively. The strand with most of the ORFs was considered as the plus strand in further analyses. A genome map showing predicted ORFs (with definite phage-related proteins) is presented in (Fig. 3). Together, all the ORFs were encoded within 65,495 bps (from nts 634 to 66,128), resulting in an extremely high coding density of 98.87%. Notably, the start codon of 25 ORFs (26.88%) overlapped with the stop codon of the previous gene, suggesting transcriptional interactions among these neighboring genes. No putative tRNA encoding gene was identified in the genome. A total of 83 promoter regions and 27 Rho-independent terminators across the genome were predicted (Supplementary data, Tables S3 and S4). No toxin or antimicrobial resistance-related gene was predicted in the genome. Despite in-silico prediction as a temperate phage (averaged probability ± SD, 0.534 ± 0.03), DRL-P1 always presented highly lytic activity in in vitro assays.

Neighbor-Joining (NJ) phylogenetic trees were reconstructed for terminase and DNA polymerase III genetic sequences with top 100 BLAST hit sequences (including RefSeq sequences). In the terminase sequence phylogeny (Fig. 4), DRL-P1 clustered most closely with a Pbunavirus RefSeq sequence ‘DL60’ (NC_028745), and an unclassified Pbunavirus sequence ‘zikora’ (MW557846), having a divergence of 0.0124 base substitutions per site with both the sequences. Conversely, in the DNA polymerase III sequence phylogeny (Fig. 5), DRL-P1 clustered most closely with two unclassified Pbunaviruses, ‘zikora’ and ‘elmo’ (MT119364), both showing the divergence of 0.001 base substitutions per site, followed by close affiliation to RefSeq Pbunavirus ‘datas’ (NC_050143) and ‘DL52’ (KR054028) having divergences of 0.0068 and 0.0133 base substitutions per site, respectively. Similar to the DNA polymerase III gene sequence phylogeny, in the complete genome phylogeny, DRL-P1 clustered with Pbunavirus isolates ‘zikora’, ‘DL52’, ‘datas’, ‘steven’ (MT119370), and ‘elmo’, supported by high bootstrap values (Fig. 6).

To further elucidate the taxonomic position of DRL-P1, we used the VIRIDIC program to calculate pairwise intergenomic similarities between DRL-P1 and 100 top BLAST hit Pbunavirus genomes including 37 RefSeq and 63 other complete genome sequences retrieved from the GenBank. In concurrence with the complete genome phylogeny, the VIRIDIC program clustered DRL-P1 with ‘DL52’, ‘zikora’, ‘elmo’, and ‘steven’ as a separate species under the genus Pbunavirus (Supplementary data, Tables S5,S6), presenting with 96.0% to 97.5% nucleotide identity among themselves. On the other hand, DRL-P1 showed 93.3% and 92.8% nucleotide identity with Pbunavirus RefSeq sequences ‘DL60’ and ‘datas’, respectively, which were suggested as close relatives in the phylogenetic analysis of the terminase and DNA polymerase III genetic regions. Although DRL-P1 showed high intergenomic similarities and close phylogenetic relatedness to these phage genomes isolated from different parts of the world, results of progressive multiple genome alignment analysis demonstrated that the arrangement of locally colinear blocks (LCBs) in the DRL-P1 genome was substantially distinct (Fig. 7).

Minor differences in terminase and DNA polymerase III phylogenetic tree topologies along with the observed differences in the arrangement of LCBs in the DRL-P1 genome as compared to closely related Pbunavirus genomes indicated the possibility of horizontal gene transfer or recombination. Therefore, a NN was reconstructed (Fig. 8) using the complete genome dataset previously used to reconstruct NJ phylogeny. The clustering pattern of sequences in the NN principally agreed to the NJ phylogeny, yet there were signals of genetic exchange in the form of extensive reticulation at the base of the DRL-P1 stock. The reticulation between DRL-P1 and other related genomes indicated multiple events of genetic exchanges, suggesting to rationalize the observed subtle divergences in terminase and the DNA polymerase III tree topologies. Subsequent analysis of recombination revealed the existence of genetic fragments similar to various other Pbunaviruses in the DRL-P1 genome, which seem to suggest the evolution of the DRL-P1 genome through the frequent exchange of genetic material (Fig. 9, Table 2).

Phage adsorption and growth kinetics

A maximum of 76% phage adsorption was documented within 20 min without any supplements. In comparison, when supplemented with MgCl₂, approximately 82% of the phages were adsorbed within 5 min, while a maximum of 90% adsorption took place within 20 min. These results indicate a positive effect of Mg²⁺ ions on phage infectivity, probably by accelerating phage adsorption rate, thereby ensuring lysis of a maximum number of bacterial cells (Fig. 10a).

A single-step growth curve (Fig. 10b) was prepared to evaluate the latent period and the burst size of the phage DRL-P1. The latent period was found to be approximately 30 min which signifies the time interval between phage adsorption and the start of the first burst. On the other hand, the duration of the rise period was estimated to be around 50 min with a burst size of approximately 100 PFU per infected cell was recorded in the experiments.

Stability of phage at different temperatures and pH conditions

The temperature vs. phage stability was studied at six different temperatures viz. 25 °C, 37 °C, 40 °C, 50 °C, 60 °C, and 70 °C. Storage at 4 °C was considered as the control for the temperature stability experiments (Fig. 11a). Our results demonstrated that the phage DRL-P1 was substantially stable at 25 °C, 37 °C temperatures, while moderately stable at 40 °C and 50 °C temperatures. However, at a temperature of 60 °C and above, phage stability decreased significantly. Specifically, as compared to control, 72% of the phages survived 60 °C temperature (*P < 0.05), whereas, only 14% of the phages could survive 70 °C temperature (***P < 0.001), after the experiment.

Our study demonstrated that after 18 h of incubation under different pH conditions, phage DRL-P1 was most stable at pH 6.0, 7.0, and 8.0, and no significant loss in the titer was observed. However, below pH 3.0 and beyond pH 10.0, only minute fractions of the viable phages were found. Approximately, 70% of the phage population was viable between pH 5.0 to 10.0. As compared to the initial titer, a significant reduction resulting in the survival of 65% and 72% phages was observed at pH 4.0 (**P < 0.01) and pH 5.0 (*P < 0.05), respectively. No viable phages were recovered after incubation at extreme pH 1.0, 2.0, 13.0, and 14.0 conditions (Fig. 11b).

Decontamination of fomites through phage preparations

In the present work, a glass coverslip was used to represent the contaminated solid surface. The decontamination potential of the DRL-P1 phage was determined at different MOIs. A 90% reduction in the bacterial count was recorded at MOI:1.0 (***P < 0.001), whereas at MOI:0.1, 52% (***P < 0.001) reduction was observed. As compared to no-phage control, a significant (***P < 0.001) reduction in the bacterial count was recorded at even low MOIs of 0.01 and 0.001, demonstrating 42% and 37% reduction, respectively. Data (mean ± SD) obtained from the experiments are represented in Fig. 12.

Phage action on bacteria

Phage action on bacterial growth was observed through a change in OD₆₀₀ over 8 h of incubation. Data from the no-phage control experiment (MOI:0) showed a typical sigmoid curve representing an uninhibited bacterial growth, whereas experiments set up with different MOIs (100, 10, 1, 0.1, 0.01, and 0.001) indicated inhibition of bacterial growth due to phage action (Fig. 13).

Biofilm degradation assay

In our experiments, DRL-P1 showed degradation potential against pre-established P. aeruginosa biofilms (Fig. 14). Using the standard method of phage-bacteria co-cultures set-up at various MOIs, we could observe a considerable decrease in bacterial biomass after 12, 24, and 48 h of incubation. In co-cultures set-up at MOI:10, significant loss (***P < 0.001) of approximately 45%, 59%, and 67% biomass were observed after 12 h, 24 h, and 48 h incubation, respectively. Similarly, at a MOI:1.0, significant (***P < 0.001) loss of approximately 40%, 59%, and 64% biomass were observed after 12 h, 24 h, and 48 h incubation, respectively, as compared to the untreated phage control. DRL-P1 also exhibited a significant biofilm degradation efficiency even at relatively lower MOIs 0.01, 0.001 (***P < 0.001 and ***P < 0.001, respectively) after 24 and 48 h of incubation (Fig. 14).

Stability of phage after lyophilization and encapsulation within alginate

Lyophilization of bacteriophage (~ 10⁹ PFU mL⁻¹) resulted in an initial drop in the phage titer (~ 10⁸ PFU mL⁻¹). For experiments, lyophilized phages were reconstituted in 2 mL TM buffer and plaque assay was performed to determine PFU. Our results indicate that once lyophilized, DRL-P1 retained its lytic activity without any significant drop in titer up to 12 months. However, after 18 months of storage (tested so far), the PFU of the lyophilized sample was found to be significantly reduced (~ 10⁷ PFU mL⁻¹, ***P < 0.001) (Fig. 15a).

The lytic activity of phage-loaded and mock-loaded (only TM buffer) alginate beads (5–6 mm) was tested by placing the beads over a lawn of P. aeruginosa. A clear zone was measured to determine the lytic potential of the DRL-P1 phage loaded within the alginate beads. Each of the freshly prepared phage-loaded beads showed ~ 13 ± 1.1 mm zones of clearance. No statistically significant change in the zone of clearance was noted in assays performed with phage-loaded beads stored for 6 months and 12 months, showing ~ 13 ± 1.0 and ~ 11 ± 1.1 mm zones of clearance, respectively. However, after 18 months of storage, the beads showed a significant decrease (*P < 0.05) in a zone of clearance (~ 10 ± 2 mm) (Fig. 15b).