TY - JOUR
AU - Azeredo, Joana
AB - ABSTRACT Strictly lytic phages are considered powerful tools for biocontrol of foodborne pathogens. Safety issues needed to be addressed for the biocontrol of Shiga toxin-producing Escherichia coli (STEC) include: lysogenic conversion, Shiga toxin production through phage induction, and emergence/proliferation of bacteriophage insensitive mutants (BIMs). To address these issues, two new lytic phages, vB_EcoS_Ace (Ace) and vB_EcoM_Shy (Shy), were isolated and characterized for life cycle, genome sequence and annotation, pH stability and efficacy at controlling STEC growth. Ace was efficient in controlling host planktonic cells and did not stimulate the production of the Stx prophage or Shiga toxin. A single dose of phage did not lead to the selection of BIMs. However, when reintroduced, BIMs were detected after 24 h of incubation. The gain of resistance was associated with lower virulence, as a subset of BIMs failed to agglutinate with O157-specific antibody and were more sensitive to human serum complement. BIM's biofilm formation capacity and susceptibility to disinfectants was equal to that of the wild-type strain. Overall, this work demonstrated that phage Ace is a safe biocontrol agent against STEC contamination and that the burden of BIM emergence did not represent a greater risk in environmental persistence and human pathogenicity. Shiga Toxin-producing Escherichia coli, bacteriophages, biocontrol, bacteriophage insensitive mutants, human safety, environmental persistence INTRODUCTION Escherichia coli is an important commensal of the mammalian gastrointestinal tract; however, this species also includes several classes of human pathogens. The latter includes a variety of diarrheagenic E. coli (Fratamico et al. 2016) which range from self-limiting watery diarrheal strains to highly virulent Shiga-toxigenic strains of E. coli (STEC). STEC strains can be divided into two categories: Enterohemorrhagic E. coli (EHEC) and the non-EHEC. EHEC causes bloody diarrhoea that can lead to life-threatening haemolytic uremic syndrome (HUS) (Sheng et al. 2008). In 2017, the European Union (EU) reported 6703 confirmed STEC infections with 20 deaths (Food and Authority 2018). In USA, STEC is implicated in ∼270 000 cases of human illness per year (Koudelka, Arnold and Chakraborty 2018). STEC includes a variety of serotypes, but the O157:H7 serotype is responsible for the majority of reported cases of HUS (Food and Authority 2018). Healthy ruminants are the reservoir host for STEC; therefore, the source of infection is often food with direct or indirect contact with ruminant faeces (Wang et al. 2017). Biocontrol strategies using specific bacteriophages (phages) have been proposed as a means to limit EHEC contamination of food products during processing (Anany et al. 2011; Coffey et al. 2011; Viazis et al. 2011; Howard-Varona et al. 2018; Son et al. 2018) but such measures must be proven safe and efficient. There are two important considerations when selecting new candidate EHEC biocontrol phages. First, lytic phages can induce host cell stress responses, e.g. DNA damage, which can signal resident prophage induction. Induction of Stx prophage can promote Shiga toxin production. This is because the genes for Shiga toxin are encoded on the prophage. The combined toxin gene induction and increased gene copy number during phage replication can significantly amplify toxin concentration (Balasubramanian et al. 2019). The second important aspect of using phages as biocontrol agents is the selection of bacteriophage insensitive mutants (BIMs) (European Food Safety Authority (EFSA) 2011). Relevant virulence phenotypes associated with persistence in environmental or industrial settings include biofilm formation. Therefore, phage control agents that select for phage resistance must be examined for potential changes in STEC phenotypes that promote persistence in industrial and environmental settings. STEC O157 can be naturally resistant to harsh conditions including high temperature and high acid. To guarantee the complete safety of phages in an industrial setting, it is important to examine if BIMs have an enhanced biofilm capability and therefore persist for longer in these settings. This is particularly relevant to EHEC because the infectious dose is very low (∼50 CFU) (Galié et al. 2018). This study used two new STEC bacteriophages isolated from effluents of urban wastewater treatment plants from the north of Portugal. These phages designed Ace and Shy, have potential as biocontrol agents. The use of phages that belong to conserved clusters of the Demerecviridaea and Myoviridae families, will generate more data regarding the safety of established phages used for the biocontrol of foodborne pathogens. There is a need to accelerate phage approval and public acceptance to control STEC. Phage host specificities were examined as well as safety issues including Shiga toxin production by EHEC target strains, BIM changes in virulence, and environmental persistence. METHODS Bacterial strains and culture conditions Two non-toxigenic E. coli O157:H7 strains were used for phage isolation and propagation: CECT 5947 (Δstx2::cat) and NCTC 12 900. A panel of 66 E. coli isolates from Ireland, Spain, and Portugal, E. coli 43 895 (O157+), and the mutant E. coli 43895Δper (O157–) (Sheng et al. 2008), as well as 16 non-O157 E. coli reference strains were used for phage host-range characterization (supplementary Table S1). To complete this panel, the following Enterobacteriaceae species were also used: Salmonella typhimurium SGSC 2523 and 3029, Salmonella enteritis ATCC 13076, Citrobacter freundii SGSC 5345, Citrobacter koseri SGSC 56110, Shigella sonnei ATCC 2593, Shigella flexneri ATCC 12022, Shigella boydii ATCC 9207, Enterobacter aerogenes CECT 4078, Enterobacter kobei CCM 8576, Morganella morganii CDC 4195–69, Cronobacter sakazakii NCTC 11467. The toxigenic E. coli strain CECT 4783 was used for phage safety assays. All strains were grown at 37°C in LB (Miller) broth (VWR Chemicals, PA, USA) supplemented with 5 g/L of NaCl, with or without agar (Merck) (1.2% for the bottom layer and 0.3% for soft agar overlays). STEC strains used for phage isolation and production (CECT 5947 and NCTC 12900) as well as strain ATCC 43895 belong to risk group 2 (consulted at collections websites: www.cect.org, www.phe-culturecollections.org.uk/, and www.lgcstandards-atcc.org, respectively). These strains were handled and manipulated in a Class-2 Biohazard safety cabinet (BSL-2). Handling and manipulation of strain CECT 4783 (risk group 3) and all isolates was performed in an authorized facility. Phages isolation, propagation, and purification Phages were isolated as previously described (Melo et al. 2019). Briefly, wastewater from ETAR of Paço de Sousa (Penafiel—Portugal) or ETAR of Serzedo (Guimarães—Portugal) was centrifuged (9500 × g, 4°C, 10 min), and the supernates were used for phage enrichment. Twenty-five mL of each wastewater supernate were mixed with an equal volume of double-strength LB and inoculated with 50 µL of an overnight culture of each propagation strain of E. coli O157 noted above. The cultures were incubated for 16 h at 37°C at 120 rpm (BIOSAN ES-20/60, Riga, Latvia). After incubation, each culture was centrifuged (9500 × g, 4°C, 10 min), and the supernates filtered (PES, 0.22 µm, ThermoFisher Scientific, Massachusetts, USA). The presence of phage in enriched supernates was detected by spotting 10 µl onto soft agar overlay plates freshly seeded with 100 µL of the E. coli test strains. Plates were then incubated at 37°C overnight (Oliveira et al. 2017; Melo et al. 2019). Clear or turbid plaques were indicative of phage presence. Each plaque was purified with toothpicks and paper strips to obtain single phage plaques. A single plaque was selected, and at least three passages were performed to guarantee phage isolation (single plaque morphology). The isolated phages were propagated as described previously (Melo et al. 2014). Briefly, each phage was spread on a bacterial lawn of the respective host using a paper strip, and the plates were incubated for 16 h at 37°C. Then, Saline-Magnesium (SM) Buffer (5.8 g.L−1 NaCl, 2 g.L−1 MgSO4.7H2O, 50 mL.L−1 1 M Tris pH 7.5) supplemented with 0.002% (w/v) gelatine was added to all plates and incubated at 4°C, 50–90 rpm (BIOSAN PSU-10i, Riga, Latvia) for 16 h. Both the liquid and the soft overlay agar were collected, centrifuged (9500 × g, 4°C, 10 min), and filtered (PES 0.22 µm) to remove cell debris. Phage suspensions were further purified as described by Sambrook et al. (Sambrook and Russell 2001). Sodium chloride was added to phage lysate solutions (0.584 g/10 mL), and the mixtures were incubated at 4°C (50-90 rpm (BIOSAN PSU-10i) for 1 h), centrifuged (9500 × g, 4°C, 10 min) and the supernates recovered. To the supernates, 1 g of PEG 8000 per 10 mL was added and gently mixed until dissolved, and further incubated at 50–90 rpm for at least 5 h at 4°C. The mixture was centrifuged (9500 × g, 4°C, 10 min) and the pellet resuspended in SM Buffer (final volume was 1/10 of initial phage suspension volume). To guarantee thorough phage resuspension this mixture was shaken (50 rpm, BIOSAN PSU-10i) at 4°C for 1 h. Chloroform was added (25% (v/v)) to the suspension and mixed by vortexing. The solution was centrifuged (3500 × g, 4°C, 10 min), and the top aqueous layer was collected, filtered (PES 0.22 µm), and stored at 4°C until use. Phage quantification was performed using the double-agar overlay technique (Adams 1959). Briefly, 100 µL of 10-fold serially diluted phage suspensions were mixed with 100 µL of an overnight host cell culture in 3 mL of soft agar, mixed, and overlaid on an LB agar plate. The plates were incubated for 16 h at 37°C, and the plaque forming units (PFUs) were determined from plates yielding 30–300 plaques. Host range and efficiency of plating The ability of phages Shy and Ace to lyse different bacteria (several E. coli strains and related Enterobacteriaceae species) was first tested by performing spot assays (as described above). To determine the Efficiency of Plating (EOP) for both phages, 10 µL from a 10-fold dilution series was spotted on bacterial lawns of 13 strains (including E. coli 43895 (O157+) and the mutant E. coli 43895Δper (O157–)). Plates were incubated at 37°C overnight, and phage titres (PFU.mL–1) determined for each strain. The EOP was calculated by dividing the phage titre obtained from each strain by the titre derived from the original host (Melo et al. 2014). One-step growth curve One-step growth curves (OSGC) were obtained as described previously (Sillankorva, Neubauer and Azeredo 2008). Briefly, 10 mL of a mid-exponentially growing host culture was centrifuged (5000 × g, 4°C, 5 min) and the cells resuspended in up to 5 mL of fresh LB to an OD620 of 1.0. The same volume of a titred phage suspension was added to achieve a Multiplicity of Infection (MOI) of 0.005. The culture was incubated at 37°C on a shaker, 120 rpm (BIOSAN ES-20/60), for 5 min to allow phage adsorption. The culture was then centrifuged (5000 × g, 4°C, 5 min) and the pellet resuspended in 10 mL of LB medium. Samples were taken every 10 min for one hour (for phage Shy) or 90 min (for phage Ace), and the number of plaques quantified by the pour-plate assay. Three experiments were performed in duplicate. pH stability Phage stability to pH was assessed using a universal buffer (150 mM potassium chloride, 10 mM potassium dihydrogen phosphate, 10 mM sodium citrate, 10 mM boric acid) adjusted to pH values of 1, 3, 5, 7 (control), 9, 11 and 13. A defined concentration of phage was added to each tube and incubated at 4°C for 24 h. Phage titrations were performed to assess the effect of pH on phage survival (Melo et al. 2019). Three experiments were perf ormed in duplicate. Transmission electron microscopy Phages were centrifuged (20 000 × g, 4°C, 1 h) and washed twice using tap water. Then, phages were deposited on copper grids provided with carbon-coated Formvar films and stained with 2% uranyl acetate (pH 4.0). Electron microscopy was performed using a Jeol JEM 1400 transmission electron microscope (Tokyo, Japan) (Melo et al. 2019). DNA isolation, genome sequencing, genome annotation and genome comparisons Phage genomic DNA was extracted using the phenol-chloroform alcohol method (Sambrook and Russell 2001). The library construction of phage samples was performed using the Illumina Nextera XT library preparation kit, and DNA libraries were sequenced with the Illumina MiSeq platform using 250 bp paired-end sequencing reads. The raw sequence data was treated to trim out adapters and low-quality reads. A single contig was obtained by de novo assembly using Geneious R9 (Melo et al. 2019). Assembled genomes were annotated using MyRAST (Aziz et al. 2008) to search for coding regions; tRNAs were identified with tRNAscan-SE (Schattner, Brooks and Lowe 2005); and BLASTP (Altschul et al. 1990) was used to find protein homologs. All proteins were analyzed through TMHMM (Krogh et al. 2001) and SignalIP (Petersen et al. 2011) for the prediction of transmembrane domains and signal peptide cleavage sites. Comparative genome analyses were performed using BLASTN (Altschul et al. 1990) and tbBlastX within EasyFig (Sullivan, Petty and Beatson 2011). Phage proteomes were compared using the Orthovenn2 (Xu et al. 2019). Genomes of phages Ace and Shy were deposited in the NCBI database with the accession numbers MT833283.1 and MT833282.1, respectively. Whole-genome comparative analysis To analyze the diversity of all E. coli O157:H7-infecting phage genomes, a search at NCBI Virus database was made for phages at GenBank based on their nucleotide completeness and E. coliO157: H7 hosts (taxid: 83 334 and taxid: 155 864). After manual inspection, a resulting database composed of 81 E. coli O157:H7 phages was obtained and re-annotated using Glimmer on Geneious R9 software (Kearse et al. 2012). Phamerator was used to compare all genomes based on shared gene content using an alignment-free K-clust algorithm (Cresawn et al. 2011). Heat map visualizing of shared gene content was created in excel. SplitsTree was then used to visualize the resulting protein repertoire relatedness obtained by Phamerator (Huson 1998). Phage groups (clusters) were defined based on members sharing at least 40% of all genes (phams). Phages were marked as individuals (singletons) if having fewer shared genes against any phage of the data set, as similarly described for other studied phage populations (Oliveira et al. 2019). Infection of planktonic cells E. coli strains CECT 5947 and NTCC 12 900 were grown in LB medium at 37°C, 120 rpm (BIOSAN ES-20/60) for 16 h. Cultures were diluted 1/100 and grown at 37°C with aeration at 120 rpm (BIOSAN ES-20/60) until an OD620nm of 0.2 was reached (early exponential growth phase). At this point, phages Ace and Shy were added to the respective hosts at MOIs of 0.1, 1 and 10. Controls were performed by adding SM buffer to cultures instead of phage. Quantification of bacterial cell survival (colony forming units (CFU) counts) were performed at 0, 2, 4 and 24 h after phage infection. To avoid over-estimation of the phage lysis activity, a virucide was added to the samples before CFU count. The combination of plant extracts with ferrous sulphate compounds was demonstrated to be able to destroy free phages without damaging metabolically active cells, avoiding further infection of host cells (Kropinski, Clokie and Lavigne 2018). Virucide was prepared as described (de Siqueira, Dodd and Rees 2006), with minor modifications. Briefly, a loose-leaf of Black tea (7.5% w/v) was boiled for 10 min and then filtered (Acetate cellulose, 0.45 µm, Whatman, Maidstone, United Kingdom). The filtered solution was autoclaved (121°C, 15 min) and stored at 4°C until used. Just before use, 330 µL of tea solution was mixed with 700 µL of freshly prepared 5 mM Ammonium Iron (II) Sulfate hexahydrate. The mixture was added to each sample at a ratio of 1/10, and incubated at room temperature (RT) for at least 5 min before making 10-fold serial dilutions. Three experiments were performed in duplicate. Stx prophage induction assays E. coli strain CECT 4783 (also known as CDC EDL 933) was used to assess the ability of lytic phage Ace to induce Stx prophage. This strain carries a prophage that encodes for the Shiga toxin Stx2a (Plunkett et al. 1999). For a positive control of prophage induction, Mitomycin C (Mit C) (Amresco-VWR, PA, USA) was used (Bonanno et al. 2016). To perform the induction assays, strain 4783 was grown at 37°C, 120 rpm (BIOSAN ES-20/60) for 16 h. A 1/50 dilution of this culture was made in fresh LB, and bacteria grown at 37°C, 120 rpm (ES-20/60) to an OD620nm of 0.2. At this point, the culture was divided into five new flasks, and phage Ace was added to three of the flasks at MOI of 0.01, 1 and 50; Mit C at 5 µg.mL–1 was added to flask 4, and SM buffer was used as negative control at flask 5. Samples of 10 mL were taken every hour up to 6 h, centrifuged (5000 × g), and filtered (PES 0.2 µm) to remove bacterial cells. Five mL of the supernatant were frozen for Shiga toxin quantification, and 5 mL were kept at 4°C for immediate use. Four independent experiments were performed. DNA extraction of Stx phages and qPCR Two hundred µL of the filtered supernatant were treated with DNase I (Thermo Scientific) and RNase (Thermo Scientific), as recommended by the manufacturer. Samples were heated (100°C, 10 min) to inactivate the enzymes, followed by addition of EDTA to further inactivate DNase I. Plasmid pUC19 was added to each tube (final concentration of 5 fg.µL–1) as an Internal Amplification Control (IAC) (Fricker et al. 2007) for qPCR reaction (ISO/TS 13136). Stx2a prophage genomic DNA was extracted by heating the samples for 15 min at 100°C. For relative quantification by qPCR, primers and probes specific for the stx2a gene and for the pUC19 plasmid were used (Table S2, Supporting Information). qPCR was performed as follows: 12.5 µL SsoAdvanced Universal Probes Supermix (Bio-Rad, California, USA), 0.4 ρmol.μL–1 each forward and reverse primer, 0.2 ρmol.μL–1of each probe, 1 µL of template DNA plus IAC, and nuclease free water up to 25 µL. qPCR reactions were performed in a CFX96 real-time PCR System (Bio-Rad) with the following parameters: 3 min at 95°C, 5 cycles of 15 s at 95°C, 25 s at 56°C and 30 s at 65°C; followed by 35 cycles of 15 s at 95°C, 25 s at 56°C and 30 s at 65°C. Relative fold increase of the gene stx2a in the STEC cultures infected with phage Ace and Mit C were compared to the control (without inductive agent) and the differences calculated using the mathematical model of Pfaffl (2-ΔΔCt method with E correction) (Pfaffl 2001). pUC 19 amplification data was used to normalize the DNA concentration of the samples. For each sample, 3 qPCR reactions were performed. Shiga toxin quantification To obtain a relative toxin quantification, an Arbitrary Unit (AU) was obtained by diluting a positive sample (E. coli 4783 treated with Mit C) until a linear correlation was obtained. With these dilutions it was possible to obtain the following equation: AU = (OD450nm-0.0505)/0.1393, where OD450nm represents the OD value obtained for each sample tested. Shiga toxin protein samples, previously collected and frozen (above), were thawed at RT, diluted to obtain OD450 values within the ones used to obtain the AU equation, and analysed using the kit RIDASCREEN® Verotoxin (R-Biopharm AG, Darmstadt, Germany) following the manufacturer's instructions. Generating BIMs To generate BIMs of E. coli 4783 an experimental design based on the model ‘Host Evolution’ from Brockhurst and Koskella was used (Brockhurst and Koskella 2013) to enrich for phage resistant strains. Strain 4783 was grown to the beginning of exponential phase (OD620nm 0.2), then phage Ace was added at an MOI of 100 (to increase phage pressure on bacterial cells). The culture was incubated overnight at 37°C, 120 rpm (BIOSAN ES-20/60). Ten % of the grown culture (bacteria and phage) was mixed with fresh LB and fresh phage Ace (same volume as before). The new culture mixture was incubated overnight at 37°C, 120 rpm (BIOSAN ES-20/60). The culture was streaked onto LB agar plates for colony purification and another round of phage infection was performed. Ten colonies from the 2nd and 3rd rounds of infection were randomly picked and further purified at least six times using LB agar phage-free plates. To guarantee that colonies selected were O157 serotype strains, after the third round of purification, the colonies were streaked to a MacConkey Sorbitol Agar plate. Colonies were confirmed as BIMs if phage Ace was unable to infect as determined by an EOP assay. BIM characterization: reactivity to O157-specific antibody, biofilm formation, and tolerance to disinfectants The ability of BIMs to react with the O157-specific antibody was confirmed by anti-O157 latex agglutination (Pro-Lab Diagnostics, Toronto, Canada) as described elsewhere (Sheng et al. 2008). For biofilm formation the methodology described by Lajhar et al. (Lajhar, Brownlie and Barlow 2018) was followed with some modifications. Briefly, wild-type (WT) strain (E. coli 4783), and all BIMs were grown overnight at 37°C, 120 rpm (BIOSAN ES-20/60), and 100-fold dilutions were made in LB. Two hundred µL per well of WT and of each BIM bacterial suspension was added to a 96-well flat bottom polystyrene plate (Orange Scientific, Braine-l'Alleud, Belgium). LB was used as negative control. The plates were incubated for 24 and 48 h at 25°C without shaking. After the incubation period, the bacterial suspension was removed, and the wells were washed two times with NaCl 0.9% to remove cells that were not attached to the wells. Biofilm mass formation was quantified using the crystal violet (CV) method (Borges et al. 2017). Experiments were performed three times in triplicate. BIMs’ biofilm tolerance to disinfectants was assessed as described by Lajlar et al. (Lajhar, Brownlie and Barlow 2018). In brief, 48 h biofilms (prepared as explained above) were washed twice with NaCl 0.9%. Two wells of each BIM and WT strains were filled with 200 µL of NaCl 0.9% as negative control, and another two wells were filled with 200 µL of sodium hypochlorite (1% v/v), SumaBac D10 (1% v/v), or lactic acid (4% v/v), and plates were incubated at RT for 10, 5 and 10 min, respectively. After the incubation period, the disinfectants were removed by aspiration, and 200 µL of a neutralizer described in ISO 18593 was added to each well. Biofilm detachment was accomplished using an ultra-sonic water bath (Grupo Selecta, Barcelona, Spain) for 6 min. CFU counts estimated cell survival. Experiments were performed three times in duplicate. BIM survival in human serum Even though STEC infections are usually associated with the GI tract, a few cases of STEC bacteremia have been reported (Chiurchiu et al. 2003; Buvens et al. 2013; Kato et al. 2019). BIM survival in human serum was tested as previously described (Oliveira et al. 2018) as a way to study BIMs virulence towards humans in comparison to WT. WT and BIMs were grown until exponential phase, and diluted in LB to achieve ∼104 CFU mL–1. Human serum was added to each culture in a ratio of 1:3 and incubated at 37°C for 1 h. Bacterial survival was determined by plate count. Experiments were performed three times in duplicate. RESULTS Phages Ace and Shy are highly specific against STEC O157: H7 strains Several phages were isolated using two wastewater effluents from ETAR of Paço de Sousa (Penafiel—Portugal) and ETAR of Serzedo (Guimarães—Portugal). The two phages with the widest lytic spectrum and best EOP were selected for further characterization. Phages vB_EcoS_Ace (Ace) and vB_EcoM_Shy (Shy) were isolated using E. coli strains CECT 5947 and NCTC 12 900 as hosts, respectively. Examination of each phage by transmission electronic microscopy classified both phages morphologically to the order Caudovirales (tailed bacteriophages). Phage Ace was further classified into the family Demerecviridae because it had a non-contractile tail of 168±12 nm length and 10±1 nm width (Fig. 1A). The capsid diameter was 81±2 nm (Fig. 1A). In contrast, phage Shy had a contractile tail and was classed to the family Myoviridae. The tail measured 108±10 nm length and 21±4 nm width, and the capsid had a diameter of 77±1 nm (Fig. 1C). Figure 1. Open in new tabDownload slide Morphological observation and Genome maps of phages Ace and Shy. TEM micrographs (1 500 000 X) show isolated phage particles of (A) phage Ace and (C) phage Shy, negatively stained with 2% (w/v) uranyl acetate (scale bar = 100 nm). Program tbBLASTX within EasyFig was used to compare the genomes of (B)phage Ace with phage AKFV33 and (D) phage Shy with phage WV8. Arrows indicate open reading frames that are coloured according to predicted function; similarity is indicated in grayscale. Figure 1. Open in new tabDownload slide Morphological observation and Genome maps of phages Ace and Shy. TEM micrographs (1 500 000 X) show isolated phage particles of (A) phage Ace and (C) phage Shy, negatively stained with 2% (w/v) uranyl acetate (scale bar = 100 nm). Program tbBLASTX within EasyFig was used to compare the genomes of (B)phage Ace with phage AKFV33 and (D) phage Shy with phage WV8. Arrows indicate open reading frames that are coloured according to predicted function; similarity is indicated in grayscale. Phages Ace and Shy had a broad lytic spectrum of E. coli O157 strains, infecting 51 out of 52 O157 strains. When non-O157 strains were tested, each phage displayed different host specificities. Phage Shy lysed 12 out 26 non-O157 strains with the majority being O26 strains. Phage Ace had a narrower host range and lysed only three of the 26 non-O157 strains (one strain with unknown serotype, O125:H19, and OR:H48) (Table S1, Supporting Information). Phage Ace also lysed some strains of Citrobacter freundii and Shigella sonnei. Moreover, both phages showed a high EOP on several O157 strains tested (Table 1). It is of note, that the relative EOP methodology used may be influenced by the bacterial host strain or, the culture conditions, such as pH, temperature, or the presence of optimal levels of bivalent cations. Nevertheless, this methodology was previously shown to be useful to select phages for biocontrol (Gutiérrez et al. 2017). Both phages lysed E. coli 43895Δper, a perosamine deletion mutant that does not express the O157 antigen, (Table 2). Figure 7. Open in new tabDownload slide E. coli O157 bacteriophage-insensitive mutants (BIMs) lose or retain anti-O157 reactivity. BIMs isolated at 48 h or 72 h post-phage exposure were tested with the anti-O157 latex agglutination assay and displayed two phenotypes. Panels 1 and 2 are E. coli positive (O157+) and negative (Δper O157–) agglutination controls. Representative BIM reactions (Panels 3 and 4) show the two different phenotypes found among all 20 independent BIM isolates. Panel 3 is negative (BIM 48.6) and Panel 4 is positive (BIM 48.7) for reaction in the anti-O157 latex agglutination test. Figure 7. Open in new tabDownload slide E. coli O157 bacteriophage-insensitive mutants (BIMs) lose or retain anti-O157 reactivity. BIMs isolated at 48 h or 72 h post-phage exposure were tested with the anti-O157 latex agglutination assay and displayed two phenotypes. Panels 1 and 2 are E. coli positive (O157+) and negative (Δper O157–) agglutination controls. Representative BIM reactions (Panels 3 and 4) show the two different phenotypes found among all 20 independent BIM isolates. Panel 3 is negative (BIM 48.6) and Panel 4 is positive (BIM 48.7) for reaction in the anti-O157 latex agglutination test. Table 1. Efficiency of plating for bacteriophages Ace and Shy with O157 E. coli strains. A. Comparation of phage titers on O157 E. coli strains and E. coli CECT 5947 a . . . O157 strains Titerc(PFU/mL) EOPd CECT 4267 1.20×106 1.62 CCC-1–12 7.40×105 1.00 CCC-5–12 2.00×106 2.70 CCC-10–12 1.70×106 2.30 CCC-18–12 2.10×106 2.84 CCC-26–12 1.80×106 2.43 ZTA09/00760EO1A6H 2.50×105 1.08 ZTA09/01068EO1A6H 5.00×104 0.54 ZTA09/01317EO1A6H 8.00×105 0.58 ZTA09/02270EO1A24H 4.00×105 0.72 ZTA09/03804EO1A6H 4.30×105 0.59 CECT 5947 5.30×105 1.00 B. Comparation of phage titers on O157 E. coli strains and E. coli NCTC 12 900b O157 strains Titer (PFU/mL) EOP CECT 4267 1.70×106 2.15 CCC-1–12 8.00×105 1.01 CCC-5–12 2.20×106 2.78 CCC-10–12 2.30×106 2.91 CCC-18–12 1.50×106 1.90 CCC-26–12 1.10×106 1.39 ZTA09/00760EO1A6H 8.00×105 1.01 ZTA09/01068EO1A6H 5.50×105 0.70 ZTA09/01317EO1A6H 1.90×106 2.41 ZTA09/02270EO1A24H 1.00×106 1.27 ZTA09/03804EO1A6H 5.20×105 0.66 NTCT 12 900 7.90×105 1.00 A. Comparation of phage titers on O157 E. coli strains and E. coli CECT 5947 a . . . O157 strains Titerc(PFU/mL) EOPd CECT 4267 1.20×106 1.62 CCC-1–12 7.40×105 1.00 CCC-5–12 2.00×106 2.70 CCC-10–12 1.70×106 2.30 CCC-18–12 2.10×106 2.84 CCC-26–12 1.80×106 2.43 ZTA09/00760EO1A6H 2.50×105 1.08 ZTA09/01068EO1A6H 5.00×104 0.54 ZTA09/01317EO1A6H 8.00×105 0.58 ZTA09/02270EO1A24H 4.00×105 0.72 ZTA09/03804EO1A6H 4.30×105 0.59 CECT 5947 5.30×105 1.00 B. Comparation of phage titers on O157 E. coli strains and E. coli NCTC 12 900b O157 strains Titer (PFU/mL) EOP CECT 4267 1.70×106 2.15 CCC-1–12 8.00×105 1.01 CCC-5–12 2.20×106 2.78 CCC-10–12 2.30×106 2.91 CCC-18–12 1.50×106 1.90 CCC-26–12 1.10×106 1.39 ZTA09/00760EO1A6H 8.00×105 1.01 ZTA09/01068EO1A6H 5.50×105 0.70 ZTA09/01317EO1A6H 1.90×106 2.41 ZTA09/02270EO1A24H 1.00×106 1.27 ZTA09/03804EO1A6H 5.20×105 0.66 NTCT 12 900 7.90×105 1.00 a Host of phage Ace. b Host of phage Shy. c Titer, PFU/ml are means of triplicate determinations, all S.E. <0.05. d EOP, efficiency of plating. Open in new tab Table 1. Efficiency of plating for bacteriophages Ace and Shy with O157 E. coli strains. A. Comparation of phage titers on O157 E. coli strains and E. coli CECT 5947 a . . . O157 strains Titerc(PFU/mL) EOPd CECT 4267 1.20×106 1.62 CCC-1–12 7.40×105 1.00 CCC-5–12 2.00×106 2.70 CCC-10–12 1.70×106 2.30 CCC-18–12 2.10×106 2.84 CCC-26–12 1.80×106 2.43 ZTA09/00760EO1A6H 2.50×105 1.08 ZTA09/01068EO1A6H 5.00×104 0.54 ZTA09/01317EO1A6H 8.00×105 0.58 ZTA09/02270EO1A24H 4.00×105 0.72 ZTA09/03804EO1A6H 4.30×105 0.59 CECT 5947 5.30×105 1.00 B. Comparation of phage titers on O157 E. coli strains and E. coli NCTC 12 900b O157 strains Titer (PFU/mL) EOP CECT 4267 1.70×106 2.15 CCC-1–12 8.00×105 1.01 CCC-5–12 2.20×106 2.78 CCC-10–12 2.30×106 2.91 CCC-18–12 1.50×106 1.90 CCC-26–12 1.10×106 1.39 ZTA09/00760EO1A6H 8.00×105 1.01 ZTA09/01068EO1A6H 5.50×105 0.70 ZTA09/01317EO1A6H 1.90×106 2.41 ZTA09/02270EO1A24H 1.00×106 1.27 ZTA09/03804EO1A6H 5.20×105 0.66 NTCT 12 900 7.90×105 1.00 A. Comparation of phage titers on O157 E. coli strains and E. coli CECT 5947 a . . . O157 strains Titerc(PFU/mL) EOPd CECT 4267 1.20×106 1.62 CCC-1–12 7.40×105 1.00 CCC-5–12 2.00×106 2.70 CCC-10–12 1.70×106 2.30 CCC-18–12 2.10×106 2.84 CCC-26–12 1.80×106 2.43 ZTA09/00760EO1A6H 2.50×105 1.08 ZTA09/01068EO1A6H 5.00×104 0.54 ZTA09/01317EO1A6H 8.00×105 0.58 ZTA09/02270EO1A24H 4.00×105 0.72 ZTA09/03804EO1A6H 4.30×105 0.59 CECT 5947 5.30×105 1.00 B. Comparation of phage titers on O157 E. coli strains and E. coli NCTC 12 900b O157 strains Titer (PFU/mL) EOP CECT 4267 1.70×106 2.15 CCC-1–12 8.00×105 1.01 CCC-5–12 2.20×106 2.78 CCC-10–12 2.30×106 2.91 CCC-18–12 1.50×106 1.90 CCC-26–12 1.10×106 1.39 ZTA09/00760EO1A6H 8.00×105 1.01 ZTA09/01068EO1A6H 5.50×105 0.70 ZTA09/01317EO1A6H 1.90×106 2.41 ZTA09/02270EO1A24H 1.00×106 1.27 ZTA09/03804EO1A6H 5.20×105 0.66 NTCT 12 900 7.90×105 1.00 a Host of phage Ace. b Host of phage Shy. c Titer, PFU/ml are means of triplicate determinations, all S.E. <0.05. d EOP, efficiency of plating. Open in new tab Table 2. Efficiency of plating for bacteriophages with O157+ and O157–E. coli strains. A. Comparison of phage titers on E. coil 43895Δper (O157–) a and 4783 (O157+) b . . . Phage and Bacteria Pairsc Titerd(PFU/mL) EOPe Phage 12 900 Shy on E. coli 43895Δper 1.82×1010 Phage 12 900 Shy on E. coli 4783 3.47×1010 0.52 Phage 5947 Ace on E. coli 43895Δper 4.0×107 Phage 5947 Ace on E. coli 4783 1.69×108 0.24 B. Comparison of phage titers on E. coli 43895Δper (O157–) and 43 895 (O157+)f Phage and Bacteria Pairs Titer (PFU/mL) EOP Phage 12 900 Shy on E. coli 43895Δper 1.25×1010 Phage 12 900 Shy on E. coli 43895 1.42×1010 0.88 Phage 5947 Ace on E. coli 43895Δper 7.30×107 Phage 5947 Ace on E. coli 43895 7.10×107 1.0 A. Comparison of phage titers on E. coil 43895Δper (O157–) a and 4783 (O157+) b . . . Phage and Bacteria Pairsc Titerd(PFU/mL) EOPe Phage 12 900 Shy on E. coli 43895Δper 1.82×1010 Phage 12 900 Shy on E. coli 4783 3.47×1010 0.52 Phage 5947 Ace on E. coli 43895Δper 4.0×107 Phage 5947 Ace on E. coli 4783 1.69×108 0.24 B. Comparison of phage titers on E. coli 43895Δper (O157–) and 43 895 (O157+)f Phage and Bacteria Pairs Titer (PFU/mL) EOP Phage 12 900 Shy on E. coli 43895Δper 1.25×1010 Phage 12 900 Shy on E. coli 43895 1.42×1010 0.88 Phage 5947 Ace on E. coli 43895Δper 7.30×107 Phage 5947 Ace on E. coli 43895 7.10×107 1.0 a E. coil 43895Δper (O157–), perosamine deletion mutant that is negative for the O157 antigen. b E. coli 4783 (O157+), wild-type strain that is positive for the O157 antigen. c Phage and Bacteria Pairs, the designated phage was plated on the designated bacteria. d Titer, PFU/mL are means of triplicate determinations, all S.E. <0.05. e EOP, efficiency of plating. f 43895 (O157+), E. coli 43895, isogenic parental strain positive for the O157 antigen. Open in new tab Table 2. Efficiency of plating for bacteriophages with O157+ and O157–E. coli strains. A. Comparison of phage titers on E. coil 43895Δper (O157–) a and 4783 (O157+) b . . . Phage and Bacteria Pairsc Titerd(PFU/mL) EOPe Phage 12 900 Shy on E. coli 43895Δper 1.82×1010 Phage 12 900 Shy on E. coli 4783 3.47×1010 0.52 Phage 5947 Ace on E. coli 43895Δper 4.0×107 Phage 5947 Ace on E. coli 4783 1.69×108 0.24 B. Comparison of phage titers on E. coli 43895Δper (O157–) and 43 895 (O157+)f Phage and Bacteria Pairs Titer (PFU/mL) EOP Phage 12 900 Shy on E. coli 43895Δper 1.25×1010 Phage 12 900 Shy on E. coli 43895 1.42×1010 0.88 Phage 5947 Ace on E. coli 43895Δper 7.30×107 Phage 5947 Ace on E. coli 43895 7.10×107 1.0 A. Comparison of phage titers on E. coil 43895Δper (O157–) a and 4783 (O157+) b . . . Phage and Bacteria Pairsc Titerd(PFU/mL) EOPe Phage 12 900 Shy on E. coli 43895Δper 1.82×1010 Phage 12 900 Shy on E. coli 4783 3.47×1010 0.52 Phage 5947 Ace on E. coli 43895Δper 4.0×107 Phage 5947 Ace on E. coli 4783 1.69×108 0.24 B. Comparison of phage titers on E. coli 43895Δper (O157–) and 43 895 (O157+)f Phage and Bacteria Pairs Titer (PFU/mL) EOP Phage 12 900 Shy on E. coli 43895Δper 1.25×1010 Phage 12 900 Shy on E. coli 43895 1.42×1010 0.88 Phage 5947 Ace on E. coli 43895Δper 7.30×107 Phage 5947 Ace on E. coli 43895 7.10×107 1.0 a E. coil 43895Δper (O157–), perosamine deletion mutant that is negative for the O157 antigen. b E. coli 4783 (O157+), wild-type strain that is positive for the O157 antigen. c Phage and Bacteria Pairs, the designated phage was plated on the designated bacteria. d Titer, PFU/mL are means of triplicate determinations, all S.E. <0.05. e EOP, efficiency of plating. f 43895 (O157+), E. coli 43895, isogenic parental strain positive for the O157 antigen. Open in new tab Both phages belong to highly conserved clusters The phage Ace genome is a linear dsDNA molecule of 112,791 bp with a 40% G + C ratio, coding for 163 open reading frames (ORF) on both strands (Fig. 1B). Phage Ace had up to 98% nucleotide identity (BlastN analysis) with phages belonging to the genus Tequintavirus. All predicted hypothetical proteins had homology to proteins of Tequintavirus phages, except for one protein (gp9). Using the Orthovenn2 tool, phage Ace was found to share 152 of 163 proteins with four representatives of the genus Tequintavirus (T5, AFKV33, Stitch, and Seaber). Sequence analysis identified proteins related to DNA replication, recombination, and modification (for example: H-N-H endonuclease—gp66, helicase—gp129, recombinase—gp132), the lysis cassette (lysin—gp40, holin—gp41, o-spanin—gp43A, and i-spanin—gp44), and several structural proteins, such as tail (gp138 to 143), major capsid (gp152), and portal protein (gp155). The protein responsible for receptor binding (gp159) was annotated. Curiously, this ORF was homologous to a protein of a Salmonella phage Sw2 (high identity–99.33%), only sharing 67.85% identity to a strictly O157-specific phage (AKFV33) (Niu et al. 2012). Moreover, analogous to other Tequintavirus phages of the Markadamsvirinae subfamily, a receptor-binding/receptor-blocking module was identified, having a receptor-blocking protein (gp160) following the receptor-binding protein (Table S3, Supporting Information). Phage Shy has a smaller genome of 88 760 bp, with a G + C ratio of 39%, coding 134 ORFs on both strands (Fig. 1D). Phage Shy had high homology (>90% nucleotide identity) with phages from the genus Felixounavirus that belongs to the subfamily Ounavirinae. It shared 129 of 134 ORFs with other phages in the genus Felixounavirus (FelixO1, LMP25, PHB11, WV8). Most proteins were hypothetical; nevertheless, some structural proteins were identified as major capsid (gp56), tail proteins (gp41, gp74, gp75, gp98), lysin (gp40), and proteins related to DNA replication, recombination and modification (DNA ligase—gp88, DNA polymerase—gp96, DNA primase/helicase—gp101) (supplementary Table S4). To place phages Ace and Shy into a wider context, their genomes were compared with genomes of all phages infecting E. coli O157: H7 deposited at GenBank with complete genomes. Our dataset included 81 phages ranging from 33.6 kb (Enterobacteria phage fiAA91-ss) to 353.1 kb (Escherichia phage UB) in genome size and encoding between 46 and 574 predicted genes. Based on average shared gene content, E. coli O157: H7 phages could be grouped into 11 clusters (A to K), and five singletons (Escherichia phages SP15, HUN/2013, HY02, vB_EcoM_005 and VT2-Sakai) and logically divided according to their predicted virion morphologies (family and subfamily) (Figure 2 and Table S2, Supporting Information). Major clusters were G (n = 22) and K (n = 14) being the remaining groups with eight or fewer members. All clusters were formed with closely related phages, with a high shared gene content (>64%), except for cluster H, containing phages that shared fewer genes. Phages Ace and Shy were found within the conserved clusters C composed of four members of Demerecviridae (Markadamsvirinae subfamily) and cluster D comprising members of Myoviridae (Ounavirinae subfamily), respectively, both groups sharing high gene content similarity (>72%). Figure 2. Open in new tabDownload slide Diversity of E. coli O157: H7 phages genomes. Genomes of 81 phages were compared using Phamerator based on shared gene content, and the resulting relationship visualized in Splitstree. For each cluster, the family and subfamily are indicated. Clusters were assigned when sharing >40% gene products and are highlight by different colours. The scale bar indicates 0.01 substitution. Figure 2. Open in new tabDownload slide Diversity of E. coli O157: H7 phages genomes. Genomes of 81 phages were compared using Phamerator based on shared gene content, and the resulting relationship visualized in Splitstree. For each cluster, the family and subfamily are indicated. Clusters were assigned when sharing >40% gene products and are highlight by different colours. The scale bar indicates 0.01 substitution. Both phages are highly stable at low pH It is known that many STEC strains can tolerate very acidic conditions found in several food products, such as cheese. Some STEC strains can tolerate a pH as low as 2.5, and even grow when the pH is as low as 4.4 (Kotrola 1995; Molina, Parma and Sanz 2003; Oh et al. 2014). Phages intended to be used in the food industry should be able to survive in such harsh conditions, so both phages were tested for survivability over a pH range from 1 to 13. Both phages had a high tolerance to various pH values (Fig. 3), only losing viability completely at the pH extremes of 1 and 13. Phage Ace viability decreased by about 2 logs after 24 h of incubation at pH 3. Figure 3. Open in new tabDownload slide Phage survival in buffers ranging from pH 1 to 13. Phages were incubated for 24 h at 4°C. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences (*P-value <0.001) between assays and control condition (pH 7) were analysed using two-way ANOVA with a Tukey's multiple comparison test. UDL—under detection limit (100 CFU mL–1). Figure 3. Open in new tabDownload slide Phage survival in buffers ranging from pH 1 to 13. Phages were incubated for 24 h at 4°C. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences (*P-value <0.001) between assays and control condition (pH 7) were analysed using two-way ANOVA with a Tukey's multiple comparison test. UDL—under detection limit (100 CFU mL–1). Phage Ace has a superior in vitro lytic activity To further characterize both phages, the growth parameters were determined by recording one-step growth curves. Phage Ace had a long latency period of 55 min and a low burst size of only 19 PFU per infected cell (Fig. 4A). Phage Shy had a similar low burst size (20 PFU per infected cell) and a shorter, yet long, latent period of 35 min (Fig. 4B). Figure 4. Open in new tabDownload slide One-step growth curves of phage. (A) Ace replicating on host E. coli CECT 5947 and (B) Shy replicating on host E. coli NCTC 12900. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Figure 4. Open in new tabDownload slide One-step growth curves of phage. (A) Ace replicating on host E. coli CECT 5947 and (B) Shy replicating on host E. coli NCTC 12900. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Both phages were tested for their ability to lyse planktonic cells of their respective hosts. Phage Ace had a higher lytic activity against planktonic cells than phage Shy (Fig. 5). Using phage Shy, even with a higher MOI of 10, a reduction of cell viability was not achieved (Fig. 5B). Phage Shy was also tested against planktonic E. coli strain 5947, and the same outcome, i.e. no reduction of cell viability, was observed, indicating that the lack of lytic activity in liquid culture is not strain specific for phage Shy. Lytic activity of phage Ace at all MOIs tested was similar, with reductions of about 5 logs after 4 or 6 h post-phage infection (Fig. 5A). A further 1 log reduction of host cells occurred after 24 h at MOIs of 1 and 10, a good feature for biocontrol. Figure 5. Open in new tabDownload slide Host cell survival after phage treatment. (A) Strain CECT 5947 infected with phage Ace, and (B) strain NCTC 12 900 infected with phage Shy. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between MOIs and Control (*P-value <0.001) were analysed using two-way ANOVA between control with a Tukey's multiple comparison test. Figure 5. Open in new tabDownload slide Host cell survival after phage treatment. (A) Strain CECT 5947 infected with phage Ace, and (B) strain NCTC 12 900 infected with phage Shy. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between MOIs and Control (*P-value <0.001) were analysed using two-way ANOVA between control with a Tukey's multiple comparison test. Phage infection does not induce Stx2a production or Stx prophage release Different MOIs were tested to simulate different scenarios of cell stress, ranging from 0.01 to 50 PFU.cell–1 (Fig. 6). Phage Ace at lower MOI (0.01) did not reduce the growth of bacteria, as it was similar to the growth of the negative control (without any inducer), but at higher MOIs (1 and 50) the phage prevented bacterial proliferation and caused a reduction of the OD after 2 h of infection. For the Stx prophage induction assays, Mitomycin C (Mit C) at 0.5 µg.mL–1 was used as a positive control. An increase of prophage production was observed after 3 h of incubation with Mit C (Fig. 6A), and bacterial growth started to decrease. The prophage induction was demonstrated by the increase of the stx2a gene present in the prophage particles being released (Fig. 6B). Note, the culture supernates were filtered and treated with DNase and RNase, so only DNA inside phage capsids remained for qPCR quantification. As phage at MOIs of 1 and 50 reduced bacterial growth, there was a concomitant decrease of the stx2a gene relative quantification, signifying reduced Stx prophage induction. Phage at lower MOIs did not reduce bacterial cell multiplication. Nevertheless, the amount of Stx prophages in the supernates was equivalent to or slightly reduced as compared to the control culture (without phage or Mit C) (Fig. 6B). Figure 6. Open in new tabDownload slide Stx prophage excision and Shiga toxin production. (A) Growth curves of E. coli CECT 4783 exposed to Mitomycin C (Mit C, 0.5 µg/mL) and phage Ace (MOI 0.01, 1, and 50). OD reduction with Mit C is associated with prophage excision and bacterial lysis. Error bars represent standard deviations from three independent experiments performed in duplicate. (B) Fold-change of extracellular prophage stx2a gene from E. coli CECT 4783 after exposure to Mit C (0.5 µg/mL) and phage Ace (MOI 0.01, 1 and 50). Relative quantification was performed by qPCR comparing the control (supernate of bacterial culture not exposed to the agents) and the other variables (supernate of phage- and Mit C -treated cells). The relative gene quantification was calculated using the 2-ΔΔCt method with E correction. Error bars represent standard deviations from 4 independent experiments. Statistical differences between Mit C and different MOIs (*P-value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. C) Shiga toxin quantification using an ELISA kit RIDSCREEN® Verotoxin. The Arbitrary Unit (AU) was determined by the equation: AU = (OD450nm-0.0505)/0,1393, where OD450nm represents the OD value obtained for each sample tested. The samples of two independent assays were used. (t0–beginning of infection and t6–6 h post infection; M—MOI and MC and Mit C—Mitomycin C) Figure 6. Open in new tabDownload slide Stx prophage excision and Shiga toxin production. (A) Growth curves of E. coli CECT 4783 exposed to Mitomycin C (Mit C, 0.5 µg/mL) and phage Ace (MOI 0.01, 1, and 50). OD reduction with Mit C is associated with prophage excision and bacterial lysis. Error bars represent standard deviations from three independent experiments performed in duplicate. (B) Fold-change of extracellular prophage stx2a gene from E. coli CECT 4783 after exposure to Mit C (0.5 µg/mL) and phage Ace (MOI 0.01, 1 and 50). Relative quantification was performed by qPCR comparing the control (supernate of bacterial culture not exposed to the agents) and the other variables (supernate of phage- and Mit C -treated cells). The relative gene quantification was calculated using the 2-ΔΔCt method with E correction. Error bars represent standard deviations from 4 independent experiments. Statistical differences between Mit C and different MOIs (*P-value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. C) Shiga toxin quantification using an ELISA kit RIDSCREEN® Verotoxin. The Arbitrary Unit (AU) was determined by the equation: AU = (OD450nm-0.0505)/0,1393, where OD450nm represents the OD value obtained for each sample tested. The samples of two independent assays were used. (t0–beginning of infection and t6–6 h post infection; M—MOI and MC and Mit C—Mitomycin C) The concordance between the stx2a gene relative quantification and Shiga toxin production was assessed using the commercial ELISA kit RIDSCREEN® Verotoxin (Fig. 6C). In fact, the quantity of Shiga toxin increased only when the Mit C (MC_t6) was added. After 6 h incubation with phage (M1_t6 and M50_t6), the quantity of toxin was equal to that in the control sample at time zero (C_t0). Even with MOI 0.01 (M0.01_t6), the amount of toxin reached levels equivalent to the control culture at 6 h post infection (C_t6). Toxigenic E. coli strain 4783 gains resistance to phage Ace after two rounds of infection To mimic industrial use, a culture of E. coli 4783 was continually infected with phage Ace at high concentration (to increase selective pressure), and after two and three rounds of 24 h of infection, 20 colonies were selected and confirmed to possess a stable resistance to phage Ace. These BIMs were tested for resistance to phage Shy and 19 of 20 remained sensitive. An anti-O157 latex agglutination test showed that some BIMs lost the ability to react with the O157-specific antibody (Table 3). The BIMs that lost this ability were classified as negative since their phenotype is the same as Δper O157– control, as represented in Fig. 7. Only one of the 48h BIMs did not agglutinate with the O157-specific antibody; however, six colonies isolated after 72 h lost reactivity. Table 3. Anti-O157 latex agglutination reactions of Bacteriophage-Insensitive Mutants. 48 h BIMastrain . Agglutinationb . 72 h BIMcstrain . Agglutination . 48.1 +d 72.1 + 48.2 + 72.2 -e 48.3 + 72.3 - 48.4 + 72.4 + 48.5 + 72.5 + 48.6 - 72.6 - 48.7 + 72.7 + 48.8 + 72.8 - 48.9 + 72.9 - 48.10 + 72.10 - 48 h BIMastrain . Agglutinationb . 72 h BIMcstrain . Agglutination . 48.1 +d 72.1 + 48.2 + 72.2 -e 48.3 + 72.3 - 48.4 + 72.4 + 48.5 + 72.5 + 48.6 - 72.6 - 48.7 + 72.7 + 48.8 + 72.8 - 48.9 + 72.9 - 48.10 + 72.10 - a 48 h BIM, E. coli O157 bacteriophage-insensitive mutants collected after 48 h incubation with phage. b Agglutination, agglutination reaction with the anti-O157 latex agglutination test. c 72 h BIM, E. coli O157 bacteriophage-insensitive mutants collected after 72 h incubation with phage. d +, agglutination positive. e -, agglutination negative. Open in new tab Table 3. Anti-O157 latex agglutination reactions of Bacteriophage-Insensitive Mutants. 48 h BIMastrain . Agglutinationb . 72 h BIMcstrain . Agglutination . 48.1 +d 72.1 + 48.2 + 72.2 -e 48.3 + 72.3 - 48.4 + 72.4 + 48.5 + 72.5 + 48.6 - 72.6 - 48.7 + 72.7 + 48.8 + 72.8 - 48.9 + 72.9 - 48.10 + 72.10 - 48 h BIMastrain . Agglutinationb . 72 h BIMcstrain . Agglutination . 48.1 +d 72.1 + 48.2 + 72.2 -e 48.3 + 72.3 - 48.4 + 72.4 + 48.5 + 72.5 + 48.6 - 72.6 - 48.7 + 72.7 + 48.8 + 72.8 - 48.9 + 72.9 - 48.10 + 72.10 - a 48 h BIM, E. coli O157 bacteriophage-insensitive mutants collected after 48 h incubation with phage. b Agglutination, agglutination reaction with the anti-O157 latex agglutination test. c 72 h BIM, E. coli O157 bacteriophage-insensitive mutants collected after 72 h incubation with phage. d +, agglutination positive. e -, agglutination negative. Open in new tab BIM and wild-type strains have equal ability to persist in the environment To understand the potential of BIMs persistence in industrial and clinical settings, biofilm formation was analysed and compared to the wild-type strain. Lower temperatures result in weak E. coli biofilms (Ma et al. 2019), so in this study a high room temperature was used. With these requirements in mind, biofilms were formed at 25°C without agitation to better mimic the natural condition of biofilm formation in these settings. In general, none of the 20 BIMs had a greater ability to form biofilm compared to the wild-type (WT), with the exception of one (48.1) that was able to produce significantly (p<0.001) more biofilm during the 24 or 48 h incubation period (Fig. 8A). To further study the capability of BIMs to persist in the environment, the biofilms formed by ten of 20 BIMs (five of 48 h BIMs and five of 72 h BIMs) were challenged with two disinfectants commonly used in the food industry and clinical settings (SumaBac D10 and sodium hypochlorite), as well as Latic Acid (the only decontaminant approved for use in foods of animal origin by the European Union; Regulation EU 101/2013). None of the BIMs tested had a better survival rate than the WT (Fig. 8B). Figure 8. Open in new tabDownload slide BIMs virulence and persistence in the environment. A) Biofilm formation of BIMs 48.1 to 48.10 and 72.1 to 72.10 compared to wild-type (strain E. coli CECT 4783). Biofilms were formed in a 96-well plate for 24 and 48 h at 25°C in static conditions. Biofilm formation was quantified by the crystal violet method. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between different BIMs and the wild-type (* P value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. B) Biofilms of BIMs 48.1 to 48.5 and 72.1 to 72.5 tested for survival in different disinfectants: SumaBac D10 at 1% (v/v); Sodium hypochlorite 1% (v/v) and Lactic Acid 4% (v/v). Assays were incubated at room temperature for 5 min for the first disinfectants, and 10 min for the others. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Figure 8. Open in new tabDownload slide BIMs virulence and persistence in the environment. A) Biofilm formation of BIMs 48.1 to 48.10 and 72.1 to 72.10 compared to wild-type (strain E. coli CECT 4783). Biofilms were formed in a 96-well plate for 24 and 48 h at 25°C in static conditions. Biofilm formation was quantified by the crystal violet method. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between different BIMs and the wild-type (* P value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. B) Biofilms of BIMs 48.1 to 48.5 and 72.1 to 72.5 tested for survival in different disinfectants: SumaBac D10 at 1% (v/v); Sodium hypochlorite 1% (v/v) and Lactic Acid 4% (v/v). Assays were incubated at room temperature for 5 min for the first disinfectants, and 10 min for the others. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. BIMs are not able to survive serum antimicrobial activity Another important aspect of BIM virulence is resistance to the human immune system, including the complement system (Dunkelberger and Song 2010). The human serum assay assesses the ability of serum complement to kill pathogens (Oliveira et al. 2018). All 20 BIMs had a higher susceptibility to human serum antimicrobial activity than the WT (Fig. 9). Six of ten BIMs from 48 h (48.4, 48.5, 48.6, 48.8, 48.9 and 48.10) were significantly (p<0.001) more sensitive to serum than WT (Fig. 9A). The 72 h-generated BIMs did not present the same level of serum sensitivity, and only two (72.7 and 72.9) were significantly (p<0.001) more sensitive than WT (Fig. 9B). Figure 9. Open in new tabDownload slide BIMs are more sensitive to human serum than wild-type E. coli. A) BIMs 48.1 to 48.10 and B) 72.1 to 72.10 compared to wild-type (strain E. coli CECT 4783) survival in human serum after 1 h incubation at 37°C. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between different BIMs and the wild-type (* P value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. Figure 9. Open in new tabDownload slide BIMs are more sensitive to human serum than wild-type E. coli. A) BIMs 48.1 to 48.10 and B) 72.1 to 72.10 compared to wild-type (strain E. coli CECT 4783) survival in human serum after 1 h incubation at 37°C. Boxplots show the median, the upper and lower quartiles and the highest and lowest values; three independent experiments performed in duplicate are represented in each boxplot. Statistical differences between different BIMs and the wild-type (* P value <0.001) were analysed using two-way ANOVA with a Tukey's multiple comparison test. DISCUSSION Phages are the most abundant biological entities in the environment (Melo et al. 2014), and therefore, it is expected that phages will be found in the same environment inhabited by their host bacteria. One of the most common reservoirs of E. coli O157 is the ruminant gastrointestinal tract (Bai et al. 2018). Northern Portugal has traditional farms and cattle husbandry (Monteiro, Mantha and Rouboa 2011; Ballem et al. 2020), and wastewater treatment plant effluents from this region were used as a source for phage isolation. Several phages were isolated for the hosts E. coli CECT 5947 and NCTC 12 900 using different effluent sources. The most promising phages for potential biocontrol of E. coli O157 were vB_EcoS_Ace (Ace) and vB_EcoM_Shy (Shy), isolated using E. coli CECT 5947 and NCTC 12 900, respectively. Phage Ace belongs to the family Demerecviridae, being very similar to other phages belonging to the genus Tequintavirus. In turn, phage Shy belongs to the family Myoviridae, having greatest similarity to phages within the genus Felixounavirus. The EOP revealed that among both phages, they lysed all but one STEC O157 strains tested. Because both Shy and Ace phages showed high specificity for E. coli O157, and because the O-antigen side chain can be recognized as a receptor for some phages, we tested whether Ace and Shy phage could infect E. coli 43895Δper (O157–). Both phages infected this mutant with high efficiency showing that host range specificity is not due to the specificity of the O-antigen. We also showed that only phage Ace was able to infect E. coli K12, suggesting these two phages recognize different receptors. This latter observation is important. The fact that phage Ace replicated in the non-pathogenic E. coli K12 strain will allow propagation of this anti-pathogenic phage in a non-pathogenic host, an important consideration for preparing a safe phage-based product (Pirnay et al. 2015). Bioinformatic analysis of phage Ace host recognition and binding predicted this phage to be equivalent to other Tequintavirus phages which employ a protein-binding/protein-blocking module. The known Tequintavirus host recognizing receptors include the proteins FhuA (for T5 phage), BtuB (for BF23 phage) and FepA (for H8 phage) (Hong et al. 2008). The blocking protein in this module prevents access to the receptor protein in the host and thus, prevents superinfection and binding of progeny phage to cell debris after lysis (Golomidova et al. 2016). Again, this may be an important asset in biocontrol phage production. To compare these newly isolated phages against other E. coli phage genomes, we used whole-genome comparative analysis based on shared gene content. As there are more than 2300 complete E. coli phage genomes deposited at GenBank, we limited our comparisons to 81 phage complete genomes from phages infecting E. coliO157:H7. We showed that this population was highly diverse, composed of 11 clusters and five singletons. The fact that six families (Myoviridae, Ackermannviridae, Drexlerviridae Siphoviridae, Podoviridae and Demerecviridae) and more subfamilies were present, demonstrates that phylogenetically distant phages have evolved to infect E. coli O157:H7 hosts. By limiting the analysis to O157-infecting phages, some interesting related phages may have been missed. Nevertheless, using this dataset it was possible to focus on the phages that evolved to infect O157 strains. Our phages, Ace and Shy, lack known genes associated with virulence (as toxin or antibiotic resistance genes) and lysogeny (such as integrases or transposases), and both share more than 70% of their genes with phages of cluster C and D, respectively. A large set of genes (96 out of 165 genes for phage Ace and 101 of 134 genes of phage Shy) code for hypothetical proteins. Both phages Ace and Shy only have one unique gene. Phages from the genus Tequintavirus (Raya et al. 2011; Niu et al. 2012; Liu et al. 2015), like phage Ace, have known efficacy in phage therapy for bacterial infections, a further indicator that this phage may be safe for STEC biocontrol. Phage Shy belongs to the genus Felixounavirus, and there is less information on the efficacy of these types of phages for biocontrol applications. Both phages have low burst sizes, which could be a problem for their use in biocontrol applications. Low burst sizes and longer lag phases are unusual for E. coli siphoviruses. For example, siphoviruses AKFV33 and CEV2 have bursts of 350 PFU/infected cell and shorter latent periods of 29 and 39 min, respectively (Raya et al. 2011; Niu et al. 2012). However, phage FelixO1, the archetype Felixounavirus, like our two phage isolates, has a long latent period accompanied by a low burst size (O'Flynn et al. 2006). Nevertheless, both Ace and Shy phages have high EOPs (with values ≥1.00 for 7 and 9 out of 11 strains tested, respectively) and lyse a wide range of O157 STEC strains. This fact, combined with tolerances against a wide pH range, are important features for their potential as biocontrol phages. By comparison, the siphovirus AKFV33 is nonviable after only 2 h at pH 3.0 and has maximum survival (76.3%) at pH 9.3 (Niu et al. 2012). To further assess if Ace and Shy phages were good candidates for STEC biocontrol they were tested for their lytic activity against planktonic cells. Contrary to expectations, phage Ace had a higher lytic activity against planktonic cells than phage Shy. Both had high EOPs (agar plates), and phage Shy, with a shorter latent period, would be expected to have a higher efficiency against a planktonic cell culture. However, even using a higher MOI of 10, a reduction in cell viability was not achieved. In contrast, phage Ace, at all MOIs tested, reduced planktonic host cells by 5 logs between 2 or 4 h post-phage infection, and reduced cell numbers by an additional log after 24 h for MOIs of 1 and 10. This is a good feature for phage biocontrol applications. O'Flynn et al. (2006) reported a similar result, with two siphoviruses having better performance in planktonic cells compared to FelixO1 phage (myovirus). The latter was unable to immediately reduce the number of cells even with a MOI of 100, in comparison to other phages. It took ∼6 h for FelixO1 to reduce the number of cells, and an immediate regrowth of the cells was observed (O'Flynn et al. 2006). As in our study, the lytic activity was measured at 2, 4, and 24 h post-phage infection, the maximum lytic activity of phage Shy may have been missed. These results indicate that of the two phages, phage Ace may be a better candidate for the biocontrol of STEC in liquid environments. When considering new and alternative decontamination substances to reduce or eliminate foodborne pathogens it is important to know their effects on the expression of virulence factors. Therefore, any use of lytic phage to control the growth of STEC strains must demonstrate their use will not induce Stx prophage induction with concomitant Shiga toxin production. Our results show phage Ace, at MOIs that significantly reduced E. coli O157 concentrations, did not induce either Stx-prophage induction or Shiga toxin synthesis. Even at low MOIs, conditions for which the growth of E. coli O157 was not affected, phage Ace did not induce the release of Stx prophage. These results are in accordance with a previous study performed employing two strictly lytic Tequatrovirus, p000v and p000y (Howard-Varona et al. 2018). Infection of E. coli O157 with these phages reduced the stx2a gene after 7.5 h of phage infection. Using strictly lytic phage to control STEC has the additional advantage to also reduce free Stx prophage in the environment and thus the potential of lateral gene transfer of Shiga toxin genes to another host bacterium. The appearance of BIMs is a recognized disadvantage for therapeutic and biocontrol employment of phages (Andreoletti et al. 2012) with several studies reporting the appearance of BIMs after single phage applications (Park et al. 2012; Tomat et al. 2013). In this study, the appearance of BIMs did not occur after a single dose (24 h incubation) of phage Ace. However, in an industrial setting, phages and target hosts will be in constant contact, with ‘fresh’ phages being added to the same bacterial community. This will favour the selection of phage-resistant host cells (Brockhurst and Koskella 2013). In line with this assumption, when a second dose was applied to the same bacterial culture, stable BIMs were in fact isolated. BIMs selected by phage Ace exposure are still infected by phage Shy, verifying that there is no cross resistance for these phages and further supporting that these phages recognize different receptors. Importantly, it highlights the potential to reduce BIM selection by employing multiple phages against the same target host. The frequency of BIM selection using a single phage vs. two phages should be equivalent to the reversion rate of a single auxotroph (10–8) vs. a double auxotroph (10–16). BIMs were further characterized in comparison with their parental wild-type strains to address the problem of environmental persistence. BIM capability for increased biofilm formation and resistance to disinfectants was not different compared to their wild-type progenitors. In contrary to other studies, inclusive for other pathogens (Gutiérrez et al. 2015), the BIMs selected post phage Ace infection did not produce significantly less biofilm than the WT. In fact, one BIM (48.1) did produce more biofilm at 24 h and 48 h of incubation. The other 19 BIMs showed a biofilm formation ability similar to the WT, meaning that the resistance gain did not result in a fitness cost (at least for biofilm ability and resistance to disinfectants profiles), however, and more importantly, did not lead to highly persistent variants. BIMs originating in a food industry setting can be an infection source for humans. The first host barrier that a pathogen encounters during human infection is the innate immune system (Luo et al. 2013). The resistance of E. coli O157 BIMs to human serum was analysed. None of the BIM isolates showed increased resistance to human serum compared to their wild-type parental strains. This demonstrates that the BIMs generated by phage Ace do not have an increased ability to evade serum complement or other antibacterial serum components. Variability in results was higher for BIMs of 48h (Fig. 9). Usually, the BIMs reduction was superior to the WT reduction. For BIMs collected at 72h, the reduction was similar to the WT, indicating that the resistant phenotypes that emerged from phage Ace intervention vary in time, however, without resulting in increased virulent phenotypes. Other reports indicate that mutations resulting in phage-resistance carry a fitness cost (Mangalea and Duerkop 2020). Additionally, changes in lipopolysaccharide composition may also effect bacterial fitness to changing environments (Rashid et al. 2006). Therefore, the ability of BIMs to react to the O157-specifc antibody was also analysed. Even though phage Ace infects the E. coli per-deletion mutant (O157–), indicating the O-antigen is not the Ace receptor, we determined that one class of Ace BIMs lost the ability to agglutinate with anti-O157 antibody. This signifies that there has been a significant alteration to the E. coli outer-membrane, potentially rendering these strains less pathogenic towards humans than their wild-type parents. Concordant observations were made in the study reported by Mizoguchi et al. (2003), which shows the co-evolution of phage PP01 and its host, E. coli O157: H7, with the appearance of new and distinct resistant mutants along the experiment. Resistance gain was associated to alterations on the LPS or in the absence of OmpC surface protein (Mizoguchi et al. 2003). Mizoguchi and colleagues studied the co-evolution of phages and host, and in this study, the evolution was only studied from the host point of view. Even though, the experiments were performed differently, the outcome is similar. Both studies show that the interaction between phages and their host, E. coli O157: H7, results in a heterogenous culture of BIMs with different phenotypes. In conclusion, lytic phage Ace is a promising biocontrol agent against the foodborne pathogen STEC O157. The phage is non-lysogenic and does not induce resident Stx prophage replication or Shiga toxin production. In fact, eliminating STEC cells, phage Ace might prevent the dissemination of free Stx prophages. Repeated phage exposure does select for BIMs, but these phage-resistance mutants do not display enhanced environmental persistence attributes nor increased resistance to disinfectants. Therefore, phage Ace, and perhaps in combination with phage Shy, have the potential to be used as effective biocontrol agents of STEC. ACKNOWLEDGEMENTS This study was supported by the Portuguese Foundation for Science and Technology (FCT) under the scope of the strategic funding of UIDB/04469/2020 unit and the project PhageSTEC PTDC/CVT-CVT/29628/2017 [POCI-01–0145-FEDER-029628] funded by FEDER through COMPETE2020 (Programa Operacional Competitividade e Internacionalização) and by National Funds thought FCT. GP acknowledges the FCT grant SFRH/BD/117365/2016. C.J.H and S.A.M. were funded by the University of Idaho Agricultural Experiment Station Hatch projects IDA01467 and IDA01406, respectively, and the National Institute of General Medical Sciences of the National Institutes of Health under Grant #P20GM103408. Conflicts of interest None declared. REFERENCES Adams MH. Enumeration of bacteriophage particles . Bacteriophages . London : Interscience Publishers, Ltd , 1959 , 27 – 34 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Altschul SF , Gish W, Miller W et al. Basic local alignment search tool . J Mol Biol . 1990 ; 215 : 403 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat Anany H , Chen W, Pelton R et al. Biocontrol of Listeria monocytogenes and Escherichia coli O157:H7 in meat by using phages immobilized on modified cellulose membranes . Appl Environ Microbiol . 2011 ; 77 : 6379 – 87 . Google Scholar Crossref Search ADS PubMed WorldCat Andreoletti O , Baggesen DL, Bolton D et al. Scientific Opinion on the evaluation of the safety and efficacy of Listex TM P100 for the removal of Listeria monocytogenes surface contamination of raw fish . EFSA J . 2012 ; 10 : 2615 . Google Scholar OpenURL Placeholder Text WorldCat Aziz RK , Bartels D, Best AA et al. The RAST Server: rapid annotations using subsystems technology . BMC Genomics . 2008 ; 9 : 1 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Bai X , Fu S, Zhang J et al. Identification and pathogenomic analysis of an Escherichia coli strain producing a novel Shiga toxin 2 subtype . Sci Rep . 2018 ; 8 : 1 – 11 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Balasubramanian S , Osburne MS, BrinJones H et al. Prophage induction, but not production of phage particles, is required for lethal disease in a microbiome-replete murine model of enterohemorrhagic E. coli infection . Blanke SR (ed). PLoS Pathog . 2019 ; 15 : e1007494 . Google Scholar Crossref Search ADS PubMed WorldCat Ballem A , Gonçalves S, Garcia-Meniño I et al. Prevalence and serotypes of Shiga toxin producing Escherichia coli (STEC) in dairy cattle from Northern Portugal . PLoS One . 2020 ; 15 : 1 – 12 . Google Scholar Crossref Search ADS WorldCat Bonanno L , Petit MA, Loukiadis E et al. Heterogeneity in induction level, infection ability, and morphology of Shiga toxin encoding phages (Stx phages) from dairy and human Shiga toxin-producing Escherichia coli O26:H11 isolates . Appl Environ Microbiol . 2016 ; 82 : 2177 – 86 . Google Scholar Crossref Search ADS PubMed WorldCat Borges A , Lopez-Romero JC, Oliveira D et al. Prevention, removal and inactivation of Escherichia coli and Staphylococcus aureus biofilms using selected monoterpenes of essential oils . J Appl Microbiol . 2017 ; 123 : 104 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Brockhurst MA , Koskella B. Experimental coevolution of species interactions . Trends Ecol Evol . 2013 ; 28 : 367 – 75 . Google Scholar Crossref Search ADS PubMed WorldCat Buvens G , De Rauw K, Roisin S et al. Verocytotoxin-producing Escherichia coli O128ab:H2 bacteremia in a 27-year-old male with hemolytic-uremic syndrome . J Clin Microbiol . 2013 ; 51 : 1633 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat Chiurchiu C , Firrincieli A, Santostefano M et al. Adult nondiarrhea hemolytic uremic syndrome associated with Shiga toxin Escherichia coli O157:H7 bacteremia and urinary tract infection . Am J Kidney Dis . 2003 ; 41 : E4 . Google Scholar Crossref Search ADS PubMed WorldCat Coffey B , Rivas L, Duffy G et al. Assessment of Escherichia coli O157:H7-specific bacteriophages e11/2 and e4/1c in model broth and hide environments . Int J Food Microbiol . 2011 ; 147 : 188 – 94 . Google Scholar Crossref Search ADS PubMed WorldCat Cresawn SG , Bogel M, Day N et al. Phamerator: a bioinformatic tool for comparative bacteriophage genomics . BMC Bioinformatics . 2011 ; 12 : 395 . Google Scholar Crossref Search ADS PubMed WorldCat de Siqueira RS , Dodd CER, Rees CED. Evaluation of the natural virucidal activity of teas for use in the phage amplification assay . Int J Food Microbiol . 2006 ; 111 : 259 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat Dunkelberger JR , Song WC. Complement and its role in innate and adaptive immune responses . Cell Res . 2010 ; 20 : 34 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat European Food Safety Authority (EFSA) . Scientific Opinion on the evaluation of the safety and efficacy of lactic acid for the removal of microbial surface contamination of beef carcasses, cuts and . EFSA J . 2011 ; 9 : 2317 , 35 . OpenURL Placeholder Text WorldCat Food E , Authority S. The European Union summary report on trends and sources of zoonoses, zoonotic agents and food-borne outbreaks in 2017 . EFSA J . 2018 ; 16 . DOI: 10.2903/j.efsa.2018.5500 . Google Scholar OpenURL Placeholder Text WorldCat Fratamico PM , DebRoy C, Liu Y et al. Advances in Molecular Serotyping and Subtyping of Escherichia coli† . Front microbiol . 2016 ; 7 : 1 – 8 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Fricker M , Messelhäußer U, Busch U et al. Diagnostic real-time PCR assays for the detection of emetic Bacillus cereus strains in foods and recent food-borne outbreaks . Appl Environ Microbiol . 2007 ; 73 : 1892 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Galié S , García-Gutiérrez C, Miguélez EM et al. Biofilms in the food industry: health aspects and control methods . Front microbiol . 2018 ; 9 : 1 – 18 . Google Scholar Crossref Search ADS PubMed WorldCat Golomidova AK , Kulikov EE, Prokhorov NS et al. Branched lateral tail fiber organization in T5-like bacteriophages DT57C and DT571/2 is revealed by genetic and functional analysis . Viruses . 2016 ; 8 : 1 – 21 . Google Scholar Crossref Search ADS WorldCat Gutiérrez D , Rodríguez-Rubio L, Fernández L et al. Applicability of commercial phage-based products against Listeria monocytogenes for improvement of food safety in Spanish dry-cured ham and food contact surfaces . Food Control . 2017 ; 73 : 1474 – 82 . Google Scholar Crossref Search ADS WorldCat Gutiérrez D , Vandenheuvel D, Martínez B et al. Two phages, phiIPLA-RODI and phiIPLA-C1C, lyse mono-and dual-species Staphylococcal biofilms . Appl Environ Microbiol . 2015 ; 81 : 3336 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat Hong J , Kim KP, Heu S et al. Identification of host receptor and receptor-binding module of a newly sequenced T5-like phage EPS7 . FEMS Microbiol Lett . 2008 ; 289 : 202 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Howard-Varona C , Vik D, Solonenko N et al. Fighting fire with fire: phage potential for the treatment of E. coli O157 Infection . Antibiotics . 2018 ; 7 : 101 . Google Scholar Crossref Search ADS WorldCat Huson DH . SplitsTree: analyzing and visualizing evolutionary data . Bioinformatics . 1998 ; 14 : 68 – 73 . Google Scholar Crossref Search ADS PubMed WorldCat Kato H , Yamaguchi H, Ito Y et al. Escherichia coli O157 enterocolitis followed by non-diarrheagenic Escherichia coli bacteremia . The Indian Journal of Pediatrics . 2019 ; 86 : 750 . Google Scholar Crossref Search ADS PubMed WorldCat Kearse M , Moir R, Wilson A et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data . Bioinformatics . 2012 ; 28 : 1647 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Kotrola JS. Growth and survival of Escherichia coli O157:H7 under acidic conditions . Appl Environ Microbiol . 1995 ; 61 : 7 – 11 . Google Scholar OpenURL Placeholder Text WorldCat Koudelka GB , Arnold JW, Chakraborty D. Evolution of STEC virulence: insights from the antipredator activities of Shiga toxin producing E. coli . Int J Med Microbiol . 2018 ; 308 : 956 – 61 . Google Scholar Crossref Search ADS PubMed WorldCat Krogh A , Larsson B, Von Heijne G et al. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes . J Mol Biol . 2001 ; 305 : 567 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Kropinski AM , Clokie MRJ, Lavigne R. Bacteriophages . 2018 . Lajhar SA , Brownlie J, Barlow R. Correction: characterization of biofilm-forming capacity and resistance to sanitizers of a range of E. coli O26 pathotypes from clinical cases and cattle in Australia . BMC Microbiol . 2018 ; 18 : 1 – 15 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Liu H , Niu YD, Meng R et al. Control of Escherichia coli O157 on beef at 37, 22 and 4°C by T5-, T1-, T4-and O1-like bacteriophages . Food Microbiol . 2015 ; 51 : 69 – 73 . Google Scholar Crossref Search ADS PubMed WorldCat Luo S , Skerka C, Kurzai O et al. Complement and innate immune evasion strategies of the human pathogenic fungus Candida albicans . Mol Immunol . 2013 ; 56 : 161 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Ma Z , Bumunang EW, Stanford K et al. Biofilm formation by Shiga toxin-producing Escherichia coli on stainless steel coupons as affected by temperature and incubation time . Microorganisms . 2019 ; 7 . DOI: 10.3390/microorganisms7040095 . Google Scholar OpenURL Placeholder Text WorldCat Mangalea MR , Duerkop BA. Fitness trade-offs resulting from bacteriophage resistance potentiate synergistic antibacterial strategies . Infect Immun . 2020 : 1 – 45 . Google Scholar OpenURL Placeholder Text WorldCat Melo LDR , Ferreira R, Costa AR et al. Efficacy and safety assessment of two enterococci phages in an in vitro biofilm wound model . Sci Rep . 2019 ; 9 : 1 – 12 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Melo LDR , Sillankorva S, Ackermann H-W et al. Isolation and characterization of a new Staphylococcus epidermidis broad-spectrum bacteriophage . J Gen Virol . 2014 ; 95 : 506 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Mizoguchi K , Morita M, Fischer CR et al. Coevolution of bacteriophage PP01 and Escherichia coli O157:H7 in continuous culture . Appl Environ Microbiol . 2003 ; 69 : 170 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Molina PM , Parma AE, Sanz ME. Escherichia coli O157:H7 and non-O157:H7 isolated in Argentina . BMC Microbiol . 2003 ; 3 : 1 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Monteiro E , Mantha V, Rouboa A. Prospective application of farm cattle manure for bioenergy production in Portugal . Renewable Energy . 2011 ; 36 : 627 – 31 . Google Scholar Crossref Search ADS WorldCat Niu YD , Stanford K, Kropinski AM et al. Genomic, proteomic and physiological characterization of a T5-like bacteriophage for control of Shiga toxin-producing Escherichia coli O157:H7 . PLoS One . 2012 ; 7 : e34585 . Google Scholar Crossref Search ADS PubMed WorldCat O'Flynn G , Coffey A, Fitzgerald GF et al. The newly isolated lytic bacteriophages st104a and st104b are highly virulent against Salmonella enterica . J Appl Microbiol . 2006 ; 101 : 251 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Oh J-H , Vinay-Lara E, McMinn R et al. Evaluation of NaCl, pH, and lactic acid on the growth of Shiga toxin-producing Escherichia coli in a liquid Cheddar cheese extract . J Dairy Sci . 2014 ; 97 : 6671 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Oliveira H , Costa AR, Ferreira A et al. Functional Analysis and Antivirulence Properties of a New Depolymerase from a Myovirus That Infects Acinetobacter baumannii Capsule K45 . Pfeiffer JK (ed). J Virol . 2018 ; 93 . DOI: 10.1128/JVI.01163-18 . Google Scholar OpenURL Placeholder Text WorldCat Oliveira H , Pinto G, Oliveira A et al. Characterization and genomic analyses of two newly isolated Morganella phages define distant members among Tevenvirinae and Autographivirinae subfamilies . Sci Rep . 2017 ; 7 : 46157 . Google Scholar Crossref Search ADS PubMed WorldCat Oliveira H , Sampaio M, Melo LDR et al. Staphylococci phages display vast genomic diversity and evolutionary relationships . BMC Genomics . 2019 ; 20 : 357 . Google Scholar Crossref Search ADS PubMed WorldCat Park M , Lee JH, Shin H et al. Characterization and comparative genomic analysis of a novel bacteriophage, SFP10, simultaneously inhibiting both Salmonella enterica and Escherichia coli O157:H7 . Appl Environ Microbiol . 2012 ; 78 : 58 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat Petersen TN , Brunak S, von Heijne G et al. SignalP 4.0: discriminating signal peptides from transmembrane regions . Nat Methods . 2011 ; 8 : 785 – 6 . Google Scholar Crossref Search ADS PubMed WorldCat Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR . Nucleic Acids Res . 2001 ; 29 : e45 . Google Scholar Crossref Search ADS PubMed WorldCat Pirnay J-P , Blasdel BG, Bretaudeau L et al. Quality and Safety Requirements for Sustainable Phage Therapy Products . Pharm Res . 2015 ; 32 : 2173 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Plunkett G , Rose DJ, Durfee TJ et al. Sequence of Shiga toxin 2 phage 933W from Escherichia coli O157:H7: Shiga toxin as a phage late-gene product . J Bacteriol . 1999 ; 181 : 1767 – 78 . Google Scholar Crossref Search ADS PubMed WorldCat Rashid RA , Tabata TA, Oatley MJ et al. Expression of putative virulence factors of Escherichia coli O157:H7 differs in bovine and human infections . Infect Immun . 2006 ; 74 : 4142 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Raya RR , Oot RA, Moore-Maley B et al. Naturally resident and exogenously applied T4-like and T5-like bacteriophages can reduce Escherichia coli O157 . Bacteriophage . 2011 ; 1 : 15 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat Sambrook J , Russell DW. Molecular Cloning: A Laboratory Manual . 3rd(ed). Cold Spring Harbor Laboratory Press , 2001 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schattner P , Brooks AN, Lowe TM. The tRNAscan-SE, snoscan and snoGPS web servers for the detection of tRNAs and snoRNAs . Nucleic Acids Res . 2005 ; 33 : 686 – 9 . Google Scholar Crossref Search ADS WorldCat Sheng H , Lim JY, Watkins MK et al. Characterization of an Eschenchia coli O157:H7 O-antigen deletion mutant and effect of the deletion on bacterial persistence in the mouse intestine and colonization at the bovine terminal rectal mucosa . Appl Environ Microbiol . 2008 ; 74 : 5015 – 22 . Google Scholar Crossref Search ADS PubMed WorldCat Sillankorva S , Neubauer P, Azeredo J. Isolation and characterization of a T7-like lytic phage for Pseudomonas fluorescens . BMC Biotech . 2008 ; 8 : 80 . Google Scholar Crossref Search ADS WorldCat Son HM , Duc HM, Masuda Y et al. Application of bacteriophages in simultaneously controlling Escherichia coli O157:H7 and extended-spectrum beta-lactamase producing Escherichia coli . Appl Microbiol Biotechnol . 2018 : 6 – 10 . Google Scholar OpenURL Placeholder Text WorldCat Sullivan MJ , Petty NK, Beatson SA. Easyfig: a genome comparison visualizer . Bioinformatics . 2011 ; 27 : 1009 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat Tomat D , Migliore L, Aquili V et al. Phage biocontrol of enteropathogenic and shiga toxin-producing Escherichia coli in meat products . Frontiers in Cellular and Infection Microbiology . 2013 ; 3 : 1 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat Viazis S , Akhtar M, Feirtag J et al. Reduction of Escherichia coli O157:H7 viability on leafy green vegetables by treatment with a bacteriophage mixture and trans-cinnamaldehyde . Food Microbiol . 2011 ; 28 : 149 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Wang L , Qu K, Li X et al. Use of Bacteriophages to Control Escherichia coli O157:H7 in Domestic Ruminants, Meat Products, and Fruits and Vegetables . Foodborne Pathogens and Disease . 2017 ; 14 : 483 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat Xu L , Dong Z, Fang L et al. OrthoVenn2: a web server for whole-genome comparison and annotation of orthologous clusters across multiple species . Nucleic Acids Res . 2019 ; 47 : W52 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2021. Published by Oxford University Press on behalf of FEMS. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - The interactions of bacteriophage Ace and Shiga toxin-producing Escherichia coli during biocontrol
JO - FEMS Microbiology Ecology
DO - 10.1093/femsec/fiab105
DA - 2021-08-09
UR - https://www.deepdyve.com/lp/oxford-university-press/the-interactions-of-bacteriophage-ace-and-shiga-toxin-producing-m6Yh5M3AuY
VL - 97
IS - 8
DP - DeepDyve
ER -