Characterization of novel bacteriophage phiC119 capable of lysing multidrug-resistant Shiga toxin-producing Escherichia coli O157:H7

Bacteriophage, bacterial strain and culture conditions

Bacteriophage phiC119 was previously isolated from horse feces in Sinaloa, Mexico with an enrichment technique. The bacteriophage was isolated from horse feces collected from five different farms located in the region of located in Northwestern Mexico. Briefly, 5 g of horse feces was diluted 1:10 in sterile distilled water (pH 7.0) and gently mixed by inversion. The mixture was cleared by low-speed centrifugation at 6,500 g for 20 min and filtered through a cellulose acetate syringe filter (0.45 μm pore size, GVS filter technology, USA). The 1 mL of filtered supernatant was then mixed with 20 mL exponential phase bacterial culture, and incubated at 37 °C for 18–24 h. After incubation, the bacterial cells were centrifuged and the supernatant was filtered through a 0.22 μm pore size cellulose acetate syringe filter (GVS filter technology, IN, USA). Then, 100 μl of filtrate and 1 mL of the host strain were mixed with soft agar and poured onto an TSA agar plate. After 24 h incubation at 37 °C, plates were checked for a clear zone of bacterial lysis. Single plaques were picked with a sterile glass Pasteur pipette and suspended in 1 mL of sterile distilled water, and each individual plaque was re-isolated three times to ensure the purity of the phage isolate. The phage was stored at −20 °C in tryptic soy broth (TSB, Bioxon, Mexico) containing 30% (v/v) glycerol for further characterization. E. coli O157 EC-48 (63-Fv18-1) was previously isolated from fecal samples from domestic animals collected from farms located in the Culiacan Valley and was used as the host for phage propagation in this study. Bacterial strains and phage stocks were obtained from the culture collection maintained by the Food Safety National Research Laboratory (LANIIA) at the Research Center in Food & Development (CIAD), Culiacan station. E. coli was grown on TSB at 37 °C; the overnight culture was used in the assays described below.

Host range

The host range of phage phiC119 was determined with a spotting assay using strains previously described as pathogenic in mammalian cells (Amézquita-López et al., 2014). Additionally, 44 environmental Salmonella strains were also included in the study (Jiménez et al., 2014; Estrada-Acosta et al., 2014) (Table 1). On the surface of TSA plates (TSA media with 1.2% agar), 1 mL of overnight culture of each strain and 3 mL of soft agar (TSA media with 0.4% agar) were poured and allowed to solidify. Then, a 10 μL aliquot of several phage dilutions were spotted onto each bacterial overlay and incubated at 37 °C for 18–24 h. After incubation, the presence of phage lysis zones was evaluated in the drops. All testing was performed in triplicate. Bacterial strains used for the bacteriophage host-range investigation were obtained from the LANIIA at the CIAD.

One-step growth curve

E. coli O157 EC-48 was inoculated into 40 mL TSB broth medium and incubated at 37 °C with shaking to reach an OD600 of 0.5. The phage and host cells were mixed with a MOI of 0.01 and allowed to adsorb for 2 min at room temperature. After incubation, the mixture was harvested by centrifugation at 10,000 × g for 1 min at 4 °C. Subsequently, the supernatant was discarded to remove the free phages. The pellet containing infected host cells was gently re-suspended in equal volume of pre-warmed TSB and shake culture at 37 °C. Samples were taken at 5 min intervals (up to 60 min), and phage titer was calculated by double agar plates. The experiment was carried out in triplicated to estimate burst size and latency.

Bacteriophage propagation and DNA extraction

Bacteriophage propagation was performed using the double-layer plaque technique described by Carey-Smith et al. (2006). Briefly, 100 μL of phage stock was mixed with 1 mL of overnight cultured E. coli (CECT 4076) and 2.8 mL of TSB agar (0.4%) preheated to 50 °C. The mixture was poured onto tryptic soy agar (TSA, Bioxon, México) plates (100 × 15 mm Petri dishes) and incubated for 18–24 h at 37 °C under aerobic conditions. Six milliliters of sterile SM buffer (100 mm NaCl, 25 mm Tris-HCl (pH = 7.5), 8 mm MgSO₄ and 0.01% (w/v) gelatin) was added to the surface of each plate, and the top agar was recovered using a sterile loop. Then, the eluate was centrifuged at 4,500 × g for 10 min at 4 °C, and the supernatant was recovered; the procedure was repeated twice. The final pooled supernatant was filtered through a cellulose acetate syringe filter with a 0.45 μm pore size (GVS filter technology, IN, USA). The phage filtrate was concentrated by centrifugation at 40,000 × g for 2 h, and then the pellet was gently resuspended by pipetting in 10 mL of SM buffer and filtered using a cellulose acetate syringe filter with a 0.20 μm pore size. The bacteriophage titer was determined by a double-layer plaque technique with serial decimal dilutions of phage concentrate. The final purified phages were stored at 4 °C.

One milliliter of purified phage suspension (approximately 1 × 10¹² plaque forming units (PFU) per mL) was incubated with 10 μL of DNase I/RNase A (10 mg/mL) (Sigma-Aldrich, MO, USA) for 1 h at 37 °C. Phage DNA was extracted using SDS-proteinase K method as previously described (Sambrook & Russell, 2001). Phage DNA was stored at 4 °C until use. The nucleic acid extract was subjected to digestion with DNase I and RNase according to the manufacturer’s instructions.

Transmission electron microscopy and plaque characteristics

Thirty microliters of purified phage suspension was adsorbed to carbon-coated copper grids (400-mesh) in a vacuum evaporator (JEE400, JEOL Ltd. Tokyo, Japan), allowed to air dry and then negatively stained with 2% phosphotungstic acid (pH 7.2). The excess solution was absorbed with filter paper, and samples were observed with a transmission electron microscope (JEM-1011, JEOL Ltd. Tokyo, Japan) operating at 80 kV (López-Cuevas et al., 2011).

Bacteriophage plaques formed on a TSA plate during the process of propagation (using dilutions that generated 15–30 plaques per plate) were analyzed according to the procedure described by Gallet, Kannoly & Wang (2011) with minor modifications. Briefly, images of ten plates were captured by a supersensitive high-resolution 16-bit camera that was deeply cooled for faint image detection (Bio-Rad Laboratories), and the image of five plaques for each plate were displayed with the ImageJ software (developed at the National Institutes of Health, Bethesda, Maryland). The plates were then incubated for 18–24 h at 37 °C before plaque size determination. To calculate the surface area (expressed in square millimeters) corresponding to each pixel, a graticule of 1 mm² was used as the reference scale for the simplified measurement of the lysis plaques. According to the analysis, each pixel corresponded to 0.5 mm².

PCR to identify stx₁ and stx₂ encoding bacteriophage

Multiplex PCR using a GoTaq® PCR Core System I (Promega, WI, USA) was performed to determine the presence of the stx₁ and stx₂ genes in the genome of phage phiC119. PCR assays were performed using the protocol previously described by Paton & Paton (1998). In addition, E. coli O157:H7 (CECT 4076) DNA was included in the PCR screen as a positive control. All primers used in the PCR assays were commercially synthesized by Sigma–Aldrich (Toluca, México).

Genome size estimation and analysis of the cohesive ends

The genome ends were determined as described by Casjens & Gilcrease (2009). Briefly, 1 μg of phage genetic material was digested with the restriction enzyme EcoRV according to the manufacturer’s specifications, followed by heating for 15 min at 75 °C. Subsequently, the reaction mixture was divided into two equal parts. One was rapidly cooled by immersion into an ice-water bath for 10 min, and the other was cooled to room temperature prior to electrophoresis on a 1% agarose gel at a voltage of 75 V for 90 min. They were then stained with ethidium bromide (1 μL mL⁻¹), and images were captured using a ChemiDoc™ MP imaging system with Image Lab™ software (Bio-Rad Laboratories). The lambda phage DNA was used as a positive control. Lambda DNA digested with the HindIII endonuclease was used as a standard molecular weight marker (Promega, WI, USA).

Genome sequencing and annotation

DNA sequencing was performed at the National Laboratory of Genomics for Biodiversity (LANGEBIO) using the MiSeq sequencing system (Illumina, Inc.) (150-bp single-end reads). In total, 4,832,127 reads were generated and assembled into one contig using Geneious v8.1.2 (the final sequence coverage was approximately 50×). The sequence assembly was validated by a comparative restriction profile (Promega, WI, USA). Potential open reading frames (ORFs) longer than 100 bp were predicted by GeneMark (http://exon.gatech.edu/) and ORF Finder (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). The putative ORFs were analyzed by BLAST at the National Center for Biotechnology Information (NCBI) (http://blast.ncbi.nlm.nih.gov/Blast.cgi) against the database of non-redundant protein sequences using a significant E-value of 10⁻³. Moreover, all identified ORFs were compared against the virulence factor database (http://www.mgc.ac.cn/VFs/) (Chen et al., 2012) and the ResFinder database (http://cge.cbs.dtu.dk/services/ResFinder/) (Kleinheinz, Joensen & Larsen, 2014). The predicted phage protein sequences were searched to identify proteins that were potentially allergenic using tools available at http://www.allergenonline.com from the Food Allergy Research. This analysis was complemented with a search for conserved protein domains using InterProScan, HMMER, Prosite, Motif Search and SMART. Hypothetical isoelectric points and the molecular weights of putative proteins were predicted using the ExPASy server (http://us.expasy.org/tools/protparam.html). Potential tRNA genes in the genome sequence were predicted using tRNAscan-SE and ARAGORN. Promoters and potential rho-independent terminators were identified using the Neural Network Promoter Prediction tool of the Berkeley Drosophila Genome Project (http://www.fruitfly.org/seq_tools/promoter.html) and the FindTerm program (http://linux1.softberry.com/berry.phtml?topic=findterm&group=programs&subgroup=gfinb)(energy threshold value: −11), respectively. The nucleotide genome sequence of phage phiC119 has been deposited in the GenBank database under accession number KT825490.

The lifestyle of the phages was predicted using the PHACTS program (http://www.phantome.org/PHACTS/upload.php). Statistical analysis was performed using Minitab statistical software version 14 (Minitab Inc., State College, PA, USA). Hierarchical clustering analysis was used to determine the relationship between genome size, gene density, and lifestyle.

Furthermore, the amino acid sequences of terminase large subunits of phiC119 and others phages were obtained from GenBank. Twelve bacteriophages, including the phiC119, were selected for phylogenetic analysis, these phages were selected as being the most well-known representatives of each important family of phages. The amino-acid-sequences were aligned using the program ClustalW, and the neighbor-joining phylogenetic tree was generated using Geneious v8.1.2.

Bacteriophage, bacterial strain and culture conditions

Electron microscopic analysis revealed that phage phiC119 was non-enveloped with an icosahedral capsid of approximately 43–45 nm in diameter and a tail of 168–172 nm in length and 7–9 nm in width. These characteristics suggest that phage phiC119 is a member of the Siphoviridae family. The flexibility and the uniformity of the tail lengths indicated that it was non-contractile (Fig. 1). Phage phiC119 produced very large (1.0–1.5 mm in diameter), clear and uniform-sized plaques after 18–24 h incubation at 37 °C with E. coli O157:H7 EC-48 (63-104 Fv18-1) using the double-agar overlay technique.

Host range

The bacteriophage phiC119 was recently isolated by our lab from horse feces and to determine the susceptibility of bacterial strains to lysis by phage, thirty-three environmental isolates of E. coli, previously isolated at the CIAD, were used for determine the host range of phage phiC119 (Table 1). A high proportion (75.75%, n = 25) of E. coli strains were sensitive to phage phiC119, which formed plaqueson a broad spectrum of E. coli serogroups O157, including Stx-producing E. coli. These E. coli isolates were previously characterized as highly virulent because they exhibit toxicity against mammalian cells and have high levels of antibiotic resistance (Amézquita-López et al., 2014; Amézquita-López et al., 2016).

Additionally, we determined the host range of the phage phiC119 with a collection of 44 Salmonella strains. Interestingly, the phage was also able to infect only some strains of certain Salmonella serotypes (Oranienburg, Agona, Luciana, and Minnesota). However, the phage was not able to lyse the other bacterial species used in this study.

One-step growth curve

One-step growth curve experiment was performed to determine the latent time period and burst size of the phage, as these are two of the most important characteristics of phage infection process. According to the results obtained, the entire phiC119 life cycle takes about 60 min to complete. phiC119 had approximately 20 min of latent period and the average burst size is 210 phage particles per infected cell after 55 min at 37 °C (Fig. 2).

Detection of the stx genes

The phage was tested for the presence of the stx₁ and stx₂ genes (Fig. 3). PCR screening for the stx genes using DNA isolated from bacteriophage phiC119 was negative. However, other virulence factors may be encoded in the bacteriophage genome, and therefore, genome sequencing and in silico analyses are required to ensure the absence of virulence, antibiotic resistance or lysogenic genes because lysogenic conversion can increase the pathogenic potential of the bacteria towards their hosts. Hence, bacteriophages suitable for biocontrol purposes should not encode virulence genes or potential immunoreactive allergens.

Analysis of the cohesive ends

The nucleic acid of phage phiC119 was resistant to RNase, sensitive to DNase and digested by restriction enzymes. These results indicate that the phage genome is double-stranded DNA and is approximately 47 kb in size (genome size estimated from the digested fragments). Moreover, enzymatic digestion of the genome suggested that phage phiC119 utilizes the pac-mechanism of DNA packaging because heating/cooling of DNA after enzymatic digestion did not alter the restriction patterns (Casjens & Gilcrease, 2009) (Fig. 4). There, was no evidence for the existence of cohesive ends in the bacteriophage genome. In addition, the analysis revealed a close a phylogenetic relationship between the phagephiC119 and other pac-type phages (Fig. 5).

Bacteriophage genome features

Overall, the bacteriophage genome contained 75 putative ORFs (90.4% of the genome consists of a coding region) (Fig. 6), 21 of which are transcribed from the complementary strand. Based on sequence similarities and protein domains/motifs and BLAST searches, 42 genes were assigned to conserved sequences and 33 were sorted into known functional categories. Furthermore, bioinformatics analysis revealed an organization of the phage genome into four functional modules, coding for structural proteins, DNA packaging, replication and host lysis (a detailed description of gene functions is shown in Table S1).

The genome sequence of phiC119 consisted of 47,319 bp with an average GC content of 44.20%, which is significantly lower than that of E. coli (average 50%). Furthermore, a tRNA gene was identified (Arg-tRNA (anti-codon CCT)) between positions 42,465–42,540 in an adjacent region to the morphogenetic cluster, indicating probable involvement in phage morphogenesis.

Bacteriophage phiC119 genome possesses a high gene density (1.60 genes per kilobase), it contains a large proportion of genes that overlap with coding regions of neighboring genes. Similarly, different authors have indicated that the genes of coliphages (bacteriophages that infect E. coli hosts) are usually tightly packed together with small intergenic regions and a high gene density (Miller et al., 2003; Santos & Bicalho, 2011). Moreover, the genome of phage phiC119 contains several overlapping sets of genes; 20 ORFs overlap with an adjacent ORF, thus generating an increase in the density of genetic information.

Genomic analysis showed that phage phiC119 does not have lysogenic genes, such as integrase and repressor genes. In addition, lifestyle prediction using the PHACT program suggested that phiC119 is a virulent bacteriophage. Furthermore, the bioinformatics analysis of the phiC119 phage did not find any undesired genes in its genome, indicating the lack of known genes coding for potential allergens and virulence genes. Therefore, bacteriophage phiC119 has two of the desirable features of candidate phages used for biocontrol.