Novel N4-Like Bacteriophages of Pectobacterium atrosepticum

Pectobacterium atrosepticum is an economically important phytopathogen that is responsible for potato blackleg and soft rot, and for which current control strategies are limited. In this study, stem samples of potato crops exhibiting blackleg were taken from three farms in Co. Cork, Ireland, and they were found to be infected with P. atrosepticum. Three closely related bacteriophages (phages) that are specific to this phytopathogen were isolated and characterized, namely vB_PatP_CB1, vB_PatP_CB3, and vB_PatP_CB4 (abbreviated as CB1, CB3, and CB4). Both CB1 and CB3 were determined to infect 12 strains and CB4 10 strains of the 19 strains of P. atrosepticum tested. Morphology, latent periods, burst sizes, and their stability at various temperatures and pHs were also examined. Genome sequencing of the three phages revealed that they shared a minimum nucleotide identity of 93% with each other. Their genomes exhibited an Enquartavirinae genome organization, possessing several conserved proteins that were associated with phages of this group, like the type species Escherichia virus N4. Tandem electrospray ionization-mass spectrometry (ESI-MS/MS) allowed for the identification of ten structural proteins that form the virion of CB1, six that are conserved in phage N4. Biocontrol experiments demonstrated that the phages suppress soft rot formation upon co-inoculation with P. atrosepticum on whole tubers. The results of this study indicate that CB1 related phages could be good candidates for phage-based control.


Introduction
The genera of Pectobacterium and Dickeya, which are often referred to collectively as the soft rot Enterobacteriaceae (SRE), are phytopathogens that cause economically important losses in a wide range of arable and ornamental crops. These bacteria are all Gram-negative, non-spore forming, facultative anaerobic rods that are typified by the production of extracellular pectinolytic enzymes during plant infection [1,2]. Both genera are considered to be among the top ten most important plant pathogens [3].

Isolation of Phages, Host Range and General Characteristics
Thirteen phage isolates were obtained from soil samples that were collected from potato grading machinery and potato fields from two of the three farms that are mentioned above. These were subjected to genomic DNA comparison. Restriction digestion analysis, employing BamHI, allowed for the identification of ten isolates that produced three similar band patterns (Figure 1). Phages producing these patterns were found in both the grading machinery and the field soils. An example of each was taken for further study, namely CB1, CB3, and CB4. These isolates produced clear plaques with an approximate diameter of 2-3 mm (overlay concentration 0.4% w/v agar in LB) on their respective host strains of P. atrosepticum, with narrow halos occasionally being observed to the surrounding plaques.
The host ranges of these phages were examined using 31 bacterial strains (local Irish isolates and reference strains) from five different species belonging to SRE, namely P. atrosepticum (19 strains), P. carotovorum subsp. carotovorum (four strains), D. chrysanthemi bv. chrysanthemi (one strain), D. dianthicola (three strains), and D. solani (four strains). These phages were able to form plaques on 15 strains of their host species, P. atrosepticum, with no plaque formation or inhibition of growth being observed on strains of the other species. Slight variations in lytic ability were found among the three phages, with CB1 and CB3 determined to form plaques on 12 strains, while CB4 was only found to infect 10 of the 19 strains that were tested (Table 1). Figure 1. Genomic DNA of Pectobacterium phages CB1, CB3, and CB4, BamHI-digested (lanes 3, 5, and 7, respectively) and undigested (lanes 2, 4, and 6, respectively). Lane 1, DNA marker (Hyperladder 1 kb, Bioline). Gel concentration 1% w/v agarose. Examination of the morphology of the three phages by transmission electron microscopy showed that they can be classified as members of the Podoviridae family [24]. They featured a C1 morphotype with isometric capsids (ca. 70 nm) and short non-contractile tails (length: ca. 25 nm) ( Figure 2 and Table 2). Head and tail measurements are consistent with previously reported N4-like phages [21]. Additionally, a set of (putatively six) short whiskers (length: ca. 25 nm) that were attached to a collar structure (width: ca. 19 nm) were observed. At their distal ends, the whiskers terminate with elongated globular appendices (ca. 12 nm × 7 nm). The three phages were named in accordance with the nomenclature set out by Kropinski et al. [25].  25.6 ± 1.7 (n = 7) 12.0 ± 1.5 (n = 9) 7.4 ± 0.9 (n = 9)
The one-step-growth curve assay, under standard conditions using LB medium, showed that the latent period of CB1 was 60 min with an approximate burst size of 207 plaque forming units (PFU)/cell. For CB3, a latent period of 65 min with an approximate burst size of 246 PFU/cell was observed. For CB4, a latent period of 65 min with an approximate burst size of 158 PFU/cell observed (Supplementary Information 1, Figure S1). Phage viability under different environmental conditions was also examined. Over a duration of one hour, all of the phages were found to be stable between −18 • C and 50 • C. They were also found to be stable between pH 5 and 11 for 24 h ( Supplementary  Information 1, Figures S2 and S3). The genome sequences obtained for phages CB1, CB3, and CB4 were 75,394 bp, 75,522 bp, and 75,973 bp with coverage of 2057×, 1882×, and 1765×, respectively. The genomes of these phages are likely linear with direct terminal repeats (DTRs) with estimated sizes of 647 bp (for CB1 and CB3) and 648 bp (for CB4). This estimation is based on the identification of a localized region with roughly double the read depth in comparison to average read depth across the whole genome of each phage, a similar finding to the previously reported N4-like phages [26,27]. However, it is noteworthy that termini at the ends of the DTRs for these phages can be asymmetric, as seen in the case of Escherichia phage N4 itself [28]. It is not known whether DTR asymmetry exists in phages CB1, CB3, and CB4. As well as this, the average GC content of these genomes was found to be 49%, which was just below the GC content associated with the host bacterium P. atrosepticum at 50-51% [29,30].

Genome and Proteome Analysis
The number of predicted open reading frames (ORFs) that is determined on the non-redundant genome of these phages were 97, 102 and 100 for CB1, CB3, and CB4, respectively ( Supplementary  Information 2-4). Proteins that were predicted to play roles in transcription, DNA replication, virion morphogenesis, and host lysis were identified in all three phages as well as putative genes for homing endonucleases (Figure 3). No integrase, excisionase, or repressor genes were detected, suggesting that all three phages have an exclusively lytic lifecycle. Moreover, no ORFs were identified for pathogenicity or known toxins. Phages CB3 and CB4 exhibited a 97% genome-wide nucleotide identity (BLASTN algorithm) and can thus be considered different isolates of the same phage species [31]. The major feature variations found between them being their ORFs for thymidylate synthase and N4 gp32-like protein. The only major difference identified with a host range between these phages was the ability of CB3 to infect two extra P. atrosepticum strains (CB BL5-1 and CB BL18-1) ( Table 1). In addition, phage CB4 also possesses two genes for tRNAs (for asparagine & glutamine). In contrast, phage CB1 was only 95% similar to CB3 and 93% similar to CB4. This places CB1 on the boundary of speciation with the other two phages. CB1 possesses the same predicted thymidylate synthase gene and it has no tRNA genes like CB3. However, the predicted early gene region possesses variations in its ORF content not shared with CB3 and CB4. These differences include the presence of different hypothetical proteins and the presence of two additional HNH endonucleases. There is also notable sequence variation of the rIIB protein of CB1 to that of the other two phages. Interestingly, CB1 shares its N4 gp32-like gene with CB4 (Supplementary Information 1, Table S2).
The overall genomic architecture of the CB1-like phages resembles those that are belonging to the N4-like group (Figure 4), possessing all 18 core proteins of these phages that were identified by Li et al. [32] (Supplementary Information 1, Table S3), thus allowing the N4-like designation. When comparing these phages to Escherichia phage N4, the gene order of most structural proteins is well conserved with major variations associated with the genes that are involved in host lysis and the genes for structural proteins believed to play roles in tail morphogenesis. This observation has been noted among other N4-like phages [33,34]. These phages do not fall within any genera currently defined in the N4-like group to date. Phylograms that are based on the vRNA and DNA polymerase proteins show that they form their own distinct clade among these phages ( Figure 5), with their closest evolutionary relationship appearing to be with the phages of the genera of Lit1virus and Luz7virus. Additionally, when protein homology was calculated with Gegenees analysis (TBLASTX), they show limited identity to other N4-like members ( Figure 6).    . TBLASTX heat map generated using Gegenees with accurate parameters -fragment length: 200 bp; and step size: 100 bp; threshold: 0%. The map includes the genomes of 38 N4-like phages with phages representing the genera G7cvirus (red), Lit1virus (brown), Ea92virus (green), Luz7virus (pink), and N4virus (yellow) and proposed genera Sp58virus (orange) and Dss3virus (grey), with the Pectobacterium phages CB1, CB3 and CB4 coded in blue.

Transcription
Transcription of the CB1-like phage genomes likely happens in a similar manner to that of Escherichia phage N4. All three phages possess a vRNA (CB1_77, CB3_82, CB4_81), which conducts the transcription of single-stranded DNA that is initiated at hairpin promoters that are located within the early gene region. These promoters are composed of a five-nucleotide hairpin with a three-nucleotide loop possessing a central purine [19]. Inspection of the early gene region of phages CB1, CB3, and CB4 reveal the presence of three potential hairpin loops for all three, which appear analogous to hairpin loops in phage N4 (Supplementary Information 1, Table S4). Middle gene transcription in phage N4 is conducted by a heterodimeric RNA polymerase that is related to the T7-like RNA polymerase family [35], with the CB1-like phages also possessing homologs (CB1_22, 23, CB3_24, 25, and CB4_24, 25). Finally, late gene transcription of phage N4 involves its ssDNA-binding protein activating the E. coli sigma 70 RNA polymerase directing it towards the N4 late promoters [36], with the ssDNA-binding protein also having homologs among the CB1-like phages (CB1_72, CB3_77, and CB4_76). Within the non-redundant parts of the genomes of the CB1-like phages, 18 putative rho-independent terminators were identified for CB1 and CB3, with 17 being identified for CB4 (Supplementary Information 1, Tables S5-S7). Comparison between these phages showed that 16 of these terminators were located at similar locations among their respective genomes (Supplementary Information 1, Table S8).

DNA Replication, Metabolism and Methylation
The CB1-like phages have several components of a DNA replication system, including a DNA polymerase I (CB1_46, CB3_50, CB4_49), a helicase (CB1_40, CB3_42, CB4_42), and a primase (CB1_67, CB3_72, CB4_71). Their ability to alter the nucleotide pool of their host exploiting a thymidylate synthase (CB1_63, CB3_68, CB4_67) allows for the conversion of dUMP to dTMP and a putative nucleoside triphosphate pyrophosphohydrolase (CB1_87, CB3_92, CB4_91), which allows for the conversion of nucleoside triphosphates to their monomer form. These phages also possess a DNA adenine methylase (CB1_53, CB3_57, CB4_56) (IPR012327). Indeed, restriction digestion patterns of the genomic DNA of CB1, CB3 and CB4 using ClaI indicate that their DNA is likely deoxyadenosine-methylated ( Supplementary Information 1, Figures S4-S6). Such methylation is known to occur on the DNA of a number of other phages, such as P1 and T4 of E. coli. It can have regulatory functions that are involved in phage DNA packaging and transcription, but additionally, can provide resistance against host restriction endonucleases [37].

Cell Lysis
Escherichia phage N4 possesses a signal-anchor-release (SAR) endolysin (N-acetylmuramidase, pfam05838) [38]. Such endolysins are transported to the inner membrane by the host sec system and they depend on pin holins to cause the collapse of membrane potential to induce their cell wall degrading activity [39]. A typical feature of such endolysins is the possession of an N-terminal transmembrane domain. However, the CB1-like phage endolysin (Endolysin lambda type, IPR034691) (CB1_60, CB3_64, CB4_63) lacks this feature, thus indicating that their endolysin likely depends on the predicted class II holin (CB1_59, CB3_63, CB4_62) for release into the host cell periplasm to reach cell wall peptidoglycan [39]. Situated next to these predicted ORFs for the endolysin and holin of the CB1-like phages are two overlapping ORFs (CB1_58, 58a, CB3_62, 62a, CB4_61, 61a), which possess one of the typical gene arrangements of a spanin rz and rz1 pair [40]. In this spanin pair arrangement, the gene for the rz protein is the larger of the two, encoding a protein with a transmembrane domain. The smaller gene encodes a lipoprotein. However, in the CB1-like example, the ORF that is predicted to encode the lipoprotein is the larger of the two with the smaller protein lacking the predicted transmembrane domain.

Structural Proteome of Phage CB1
There is a minimum of ten proteins that have been identified to form the virion of Escherichia phage N4 [41], with in silico analysis showing six to be shared with phage CB1. These six are the vRNA polymerase (CB1_77), the major capsid (CB1_83), the portal protein (CB1_86), and structural proteins of unknown function resembling those of gp52 (CB1_79), gp54 (CB1_81), and gp67 (CB1_90) of N4. Those that were not found to be shared with CB1 were the tail sheath (gp65), the tail appendage (gp66), the head decorating protein (gp17), and one other structural protein (gp51) that has previously been suggested to be an internal virion protein [42]. Using in silico analysis, three proteins were identified to play potential roles in the morphogenesis of the tail structure of CB1 not shared with N4, namely CB1_57, 61, and 62. The CB1_61 gene product is likely a tail spike given that it possesses an SGNH hydrolase-type esterase domain (IPR013830), suggesting enzymatic activity, like the tail spike (gp63.1) of N4-like Escherichia phage G7C, which deacetylates host surface polysaccharides [43]. Mass spectrometry (ESI-MS/MS) that was conducted on the structural proteome of CB1 verified the presence of the shared N4 structural proteins along with the three identified putative tail proteins and one hypothetical protein (CB1_78) ( Table 3 and Supplementary Information 5, Table S1). This latter protein of the CB1 virion mirrors the position of gp51 of phage N4, but shares no homology. All of the CB1 structural proteins that were identified possess homologs in phages CB3 and CB4, with the putative tail spike protein CB1_61 being split into two ORFs on the genome of CB4 (CB4_64, 65). Furthermore, the large terminase responsible for virion DNA packaging was identified for all three phages (CB1_91, CB3_97, CB4_95). 2.3.6. Selfish Genetic Elements Homing endonucleases are mobile genetic elements with endonuclease activity that only promote the spread of their own encoding gene [41]. Several homing endonucleases of the HNH family (IPR003615) have been identified on the genomes of the three CB1-like phages. Phage CB1 itself was found to possess six homing endonucleases (CB1_24, 33,39,43,52,71), whereas phages CB3 (CB3_36, 47,56,76) and CB4 (CB4_36, 46, 55, 75) both possess four. Such genes have also been identified among other N4-like phages [26,34].

Phage Biocontrol on Whole Tubers
The phage mixture consisting of phages CB1, CB3, and CB4 was assessed for its ability to suppress soft rot (in a whole tuber assay) caused by a mixed infection by their host P. atrosepticum strains DSM 18077 and DSM 30186. The potato cultivar Rooster was selected for these experiments, given that it is the predominant variety that is grown in the Republic of Ireland (it comprised 60% of the Irish crop in 2014) [44].
The whole tuber rot assay was carried out independently in triplicate, and the results were averaged. Assay involved two sets of ten tubers that were treated with bacteria (designated sets (a) and (b), 100 µL at approx. 1.0 × 10 7 CFU/mL), which were allowed to absorb into the tuber tissue. A third set (c) was treated with water. Sets (a) and (c) were treated with SM, while (b) was treated with the phage mixture (100 µL at 1 × 10 7 PFU/mL). Following incubation, the average weight of rotten tissue from each set of tubers was recorded. Set (a) (bacteria + SM buffer) was 5.39 ± 3.121 g; Set (b) (bacteria + phage) was 0.311 ± 0.498 g, thus the average weight of infected tubers that were treated with phage was less than that treated without phage, with this result being statistically significant (p < 0.0005). Therefore, indicating that phage treatment limited soft rot formation. No rot was observed for set (c) (Figure 7). Look to Supplementary Information 1, Table S7 for the visual outcome of treated tuber sets (a), (b) and (c).

Discussion
Of the sixteen samples of potato crops symptomatic of blackleg that were obtained from three potato farms in Co. Cork, from which pectolytic isolates were found, fifteen were identified to be infected with P. atrosepticum with the remaining infection being identified to be caused by a mixture of P. atrosepticum with P. carotovorum subsp. carotovorum. This study focused on the identification and isolation of exclusively lytic phages of P. atrosepticum from the potato crop environment, followed by an assessment of their potential for biocontrol applications against soft rot and/or blackleg disease. P. atrosepticum is highly relevant to Irish potato horticulture, and to date, relatively few reports focusing on phage biocontrol of this pathogen have appeared in the scientific literature.
During an enrichment screening for P. atrosepticum phages, three closely related phage isolates were obtained, namely CB1, CB3, and CB4. Genome sequencing revealed them to be N4-like, sharing a similar gene order and possessing all 18 core genes that were found among all N4-like members. For phage therapy applications involving the treatment of Pseudomonas aeruginosa murine infections, N4-like phages have been deemed to be safe and effective [45]. This suggested that the three identified phages could be good candidates for biocontrol applications.
Phages CB1, CB3, and CB4 appear to possess relatively broad host ranges within their host species, collectively spanning 15 of the 19 tested P. atrosepticum strains (Table 1). By comparison, among the other members of the N4-like phages (infecting Escherichia, Pseudomonas, Vibrio, and Roseovarius), many have been found to propagate only on the strain used in the original isolation [34,46,47]. The relative broad host range that was observed is desirable for their use in biocontrol applications. Nevertheless, the application of these phages would have limited biocontrol potential where other SRE species may be involved in disease. Additionally, P. atrosepticum strains that were tested consisted predominantly of those from an Irish environment. It is unknown if strains from a different geographical location would have a similar susceptibility to these phages. However, resistance could be overcome by supplementing the phage mixture with additional phages targeting resistant P. atrosepticum strains, as well as other SRE species. The generation of phage cocktails is an important approach being adopted for phage preparations for use in the food industry and veterinary medicine as it ensures the widest possible host range against the targeted bacteria [48,49]. Phages with host ranges that were limited within their respective host species have been observed for a number of other studies focusing on SRE phage biocontrol [13,14,17]. Nevertheless, a few notable exceptions have been described, such as Dickeya phages D4 and D5 being able to infect several Dickeya species and Dickeya phages ϕPD10.3 and ϕPD23.1 being able to infect D. solani, P. carotovorum subsp. carotovorum and P. parmentieri [12,13]. To date, no broad host range phage infecting multiple SRE species, including P. atrosepticum, has been described.
The tuber rot assay in this study indicated that phages CB1, CB3, and CB4 appeared to possess some potential for the inhibition of soft rot formation of potato tubers (Figure 7). A similar finding was previously demonstrated using phages against P. carotovorum subsp. carotovorum, D. solani, and P. parmentieri [13][14][15]. An approximate multiplicity of infection (MOI) of 1 was used in our experiments to achieve this effect, lower than that reported with Dickeya phage vB_DsoM_LIMEstone1 at a MOI of 10 and 100 [14], but higher than that with Dickeya phages ΦPD10.3 and ΦPD23.1 against P. carotovorum subsp. carotovorum, D. solani, and P. parmentieri at an MOI of 0.01 [15]. However, further work is necessary to examine the inhibitory effect of soft rot on potato tubers by the CB1-like phages in more detail as well as additional studies to explore if the phages could inhibit blackleg formation on whole plants. Obviously, other environmental conditions that could potentially affect the viability of these phages in field applications, such as UV light, salinity, soil composition, and agrochemical factors (including fertilizers etc.) would also require investigation.
Of the ten structural proteins that were identified to form the CB1 virion (Table 3), four were not shared with that of Escherichia phage N4 (CB1_57, 61, 62, and 78). These proteins appear to be associated with tail components of the CB1 virion. Their lack of homology is likely due to their adaption to allow for the recognition of its host bacterium and enable subsequent DNA injection. For N4-like Escherichia phage G7C to recognize its host, its tail spike esterase domain must first deacetylate the O-antigen of its host lipopolysaccharide [43]. This may also be the case for the CB1-like phages with their putative tail spike (CB1_61, CB3_66, and CB4_75, 76) possessing an SGNH hydrolase-type esterase domain (IPR013830).
Phylogenetic analysis (based on phylograms and Gegenees (TBLASTX) of CB1-like phages show that they are distinct members within the N4-like group, with their closest evolutionary relationship being found with phages of the genera Lit1virus and Luz7virus. This analysis also indicates the existence of higher order taxonomic relationships between genera of the N4-like phages. Such relationships are becoming apparent due to the increasing number of N4-like genomes being added to public databases. Such a relationship was identified between the phages belonging to the genera of N4virus, G7cvirus, and Ea92virus with Achromobacter phage JWAlpha, which has been proposed to form the subfamily Enquartavirinae [21]. A similar connection was also found with the genera of Luz7virus and Lit1virus (Figures 5 and 6). These relationships are indicated with clades that are closely situated within phylograms and the sharing of protein sequence identity of~30-40% in Gegenees analysis.

Phage Isolation
Phages were isolated using an enrichment method. Five grams of soil was weighed out and were placed into 30 mL of LB broth, along with the addition of 300 µL of an overnight culture of P. atrosepticum (DSM 18077, DSM 30184, DSM 30185, and DSM 30186) and then incubated for 18 h at 25 • C. This was centrifuged to pellet soil matter with supernatant then being filtered (0.45 µm pore-size filter, Sarstedt, Nümbrecht, Germany). The supernatant was spotted (10 µL) onto LB overlays that were seeded with different strains of Pectobacterium. Phages were isolated by picking individual plaques and then replating and reisolating to ensure purity [55].

Host Range and General Characterization
Host range of phages was tested by spotting a serial dilution (neat to 10 −7 ) of a phage suspension (titer of 10 7 PFU/mL) onto LB overlays that were seeded with bacteria of interest. EOP values were determined for sensitive strains by dividing phage titer on target bacterium by phage titer on host bacterium. Bacterial strains used in host range study are listed in Supplementary Information 1, Table S1, and Supplementary Information 6, Table S1.
One-step-growth curve assay was performed in a similar manner as previously described [56,57]. Phage host strains were grown to an OD600 of 0.20-0.23 (approx. 1 × 10 8 CFU/mL), followed by centrifugation of 2 mL in a microfuge to pellet bacteria. The pellet was resuspended in 1 mL of phage suspension to yield an approx. MOI of 5 × 10 −4 following incubation at 25 • C for one minute. This was then centrifuged to pellet bacteria and the supernatant was removed, thus separating bound from unbound phages. The bacterial pellet with bound phage was then resuspended in 10 mL of LB and incubated aerobically in a water bath at 25 • C with agitation at 60 rpm. At five-minute intervals, aliquots were removed to measure phage titer by the overlay method. Based on the number of PFU/mL, the latent period and the burst size were determined, by dividing the average PFU/mL of the latent period by the average PFU/mL of the last four time points of the experiment.
Phage stability was tested by incubating phage suspension of 10 6 PFU/mL in SM buffer (50 mM Tris.HCl pH 7.5, 100 mM NaCl, 8 mM MgSO 4 ) at different temperatures for one hour and incubating phage suspension in pH buffer ranging from 2 to 12 (10 mM trisodium citrate, 10 mM boric acid, and 150 mM KCl, adjusted with NaOH or HCl) for 24 h [14].

CsCl Gradient Purification
Isopycnic centrifugation through CsCl gradients was performed, as previously described [58], with a number of modifications. A high titer phage lysate (>1 × 10 9 PFU/mL), was precipitated using polyethylene glycol (15% w/v PEG8000, 1 M NaCl) at 4 • C overnight and centrifuged, after which the pellet was resuspended in TMN buffer (10 mM Tris-HCl pH 7.4, 10 mM MgSO 4 ·7H 2 O, 0.5 M NaCl), and where necessary a chloroform phase separation step (1:1) was conducted to remove debris. The resulting phage preparation was placed onto a CsCl step gradient composed of 1.3, 1.5, and 1.7 g/mL layers and spun in a 100 Ti rotor (Beckman Coulter, Brea, CA, USA) at 200,480 g for 3 h at 4 • C. Resulting phage bands were collected and subjected to dialysis with two changes of Tris-HCl buffer (10 mM, pH 7.5) at 4 • C.

Transmission Electron Microscopy
Phages were negatively stained on freshly prepared ultra-thin carbon films with 2% (w/v) uranyl acetate and with 1% phosphotungstic acid, as described in detail earlier [59]. Micrographs were taken using a Tecnai 10 transmission electron microscope (FEI Thermo Fisher, Eindhoven, The Netherlands) at an acceleration voltage of 80 kV with a MegaView G2 CCD-camera (EMSIS, Muenster, Germany). Measurements of phage dimensions were calculated using samples that were stained with uranyl acetate (head sizes) or with phosphotungstic acid (all other dimensions).

DNA Isolation, Restriction and Sequencing
DNA extraction was performed as previously described [60]. CsCl purified phage particles were treated with DNase and RNase, followed by treatment with 10% SDS and proteinase K followed by DNA extraction with phenol: chloroform: isoamyl alcohol (25:24:1 v/v) and chloroform: isoamyl alcohol (24:1 v/v). DNA samples were digested with BamHI and ClaI, according to manufacturer's protocols (New England BioLabs, Ipswich, MA, USA). The digested DNA was analyzed by agarose gel electrophoresis.
Prior to sequencing, DNA quality and quantity were estimated using both a Nanodrop (ND-1000, Thermo Fisher, Waltham, MA, USA) and by visualization after agarose gel electrophoresis. Genomic sequencing was outsourced to the Centre for Genomic Research at the University of Liverpool, Liverpool, UK. Illumina MiSeq system (Illumina, San Diego, CA, USA), with a TruSeq DNA Nano LT library sample preparation kit for library preparation. Library quality was assessed using the Agilent Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and Qubit measurements prior to being sequenced with paired-end reads of 2 × 250 bp. Reads were assembled using Spades genome assembler v.3.10 (St. Petersburg, Russia).

Comparative Genomics
The linear genomic comparison maps of CB1 and other N4-like phages were created with the use of either BLASTN or TBLASTX to determine genome homology and it was visualized with Easyfig [72]. Artemis Comparison Tool (ACT) was used for the identification of feature variations between the genomes of the CB1-like phages, with homology being assessed with BLASTN [73]. Phylogenetic analysis employed the use of the DNA polymerase and vRNA polymerase proteins of 38 N4-like phages as well as those of CB1-like phages (Supplementary Information 6, Table S2) using MEGA7 [74], involving the use of MUSCLE for sequence alignment [75], with the construction of phylograms using the maximum likelihood (ML) method that was based on the Jones-Thornton-Taylor model [76], with the robustness of the trees being assessed with bootstrapping (1000). The heat map comparing the genomes of 38 N4-like phages and CB1-like phages was generated using Gegenees, using accurate parameters (fragment length: 200 bp; step size: 100 bp; threshold: 0%) [77].

Phage CB1 Virion ESI-MS/MS Proteome Analysis
Phage capsid proteins were extracted from high titer CsCl purified phage (>1 × 10 9 PFU/mL) using chloroform:methanol extraction (1:1:0.75, v/v/v). The resulting protein pellet was resuspended in loading buffer (1% SDS, 6% sucrose, 100 mM dithiothreitol, 10 mM Tris pH 6.8, 0.0625% w/v bromophenol blue) and heated to 95 • C for 5 min to resuspend the pellet. This was subsequently loaded onto a 12% SDS-PAGE gel, after which gel electrophoresis was conducted. The resulting gel was then stained using Gelcode™ Blue Safe Protein Stain (Thermo Fisher) to visualize virion proteins. Gel fragments were extracted and subjected to tyrosination, which were analyzed using ESI-MS/MS exactly, as described previously [78].

Biocontrol Assays to Determine Biocontrol Potential CB1, CB3 and CB4 Mixture
The potato whole tuber rot assay was used to assess the potential of a phage mixture comprised of phages CB1, CB3, and CB4 to prevent infection of potato tissue by P. atrosepticum and was conducted in a similar manner, as previously described [15]. The phage mixture was made by adding an equal number of phages to a total titer approx. 1 × 10 7 PFU/mL to SM buffer. P. atrosepticum strains, DSM 18077 and DSM 30186, were resuspended in deionized water with cell numbers of approx. 1 × 10 7 CFU/mL and mixed in equal quantities. Ware potato tubers were obtained at a local supermarket, these were first washed with tap water, then surface sterilized for 10 min in 70% isopropanol, then again washed with tap water, and allowed to dry on tissue paper at room temperature.
Whole tuber assay involved tubers being incised at the rose end (opposite the stolon end) to remove a 0.5 cm transverse slice. To which, 100 µL of either bacteria or water was added, left to absorb and then followed with the addition of 100 µL of SM buffer or phage mixture. These were then allowed 30 min to sit at room temperature before the detached potato slices were reattached using sterile toothpicks to tubers. Tubers were then incubated or 72 h at 25 • C in a humid box. To assess the protective effect of the phage mixture, the weight of rotten tissue was determined for each tuber.
Supplementary Materials: The following are available online at http://www.mdpi.com/1424-8247/11/2/45/s1, Supplementary information 1 Figure S1: One-step-growth growth analysis of phage CB1 infection of P. atrosepticum strain DSM18077, phage CB3 infection of P. atrosepticum strain DSM30186 and phage CB4 infection of P. atrosepticum strain DSM30186, Supplementary information 1 Figure S2: Stability of Pectobacterium phages CB1, CB3 and CB4 to various pH values upon 24 h of exposure, Supplementary information 1 Figure S3. Stability of Pectobacterium phages CB1, CB3 and CB4 to various temperatures upon one-hour exposure, Supplementary information 1 Figure S4: Genomic DNA of Pectobacterium phage CB1, which had been digested with restriction enzyme ClaI, Supplementary information 1 Figure S5: Genomic DNA of Pectobacterium phage CB3, which had been digested with restriction enzyme ClaI, Supplementary information 1 Figure S6: Genomic DNA of Pectobacterium phage CB4, which had been digested with restriction enzyme ClaI, Supplementary information 1 Figure S7: Pictures of the typical observed outcomes for the tuber rot assays, Supplementary information 1 Table S1. Results of physiological, biochemical, Pectobacterium genus specific and Pectobacterium atrosepticum and Pectobacterium carotovorum subsp. carotovorum species specific PCRs and MALDI-TOF mass spectrometry on isolates obtained from potato stem samples symptomatic for blackleg from farms in Co. Cork, Ireland, Supplementary information 1, Table S2. Identified ORFs and tRNA gene variations between the genomes of Pectobacterium phages CB1, CB3 and CB4, Supplementary information 1, Table S3. Homologs of the eighteen core proteins described by Li et al. 2016 found present in the genomes of Pectobacterium phages CB1, CB3 and CB4, Supplementary information 1, Table S4. Putative single-stranded hairpin promoters predicted in the genomes of Pectobacterium phages CB1, CB3 and CB4 identified assisted with QuikFold, Supplementary information 1, Table S5. High ∆G rho-independent terminators predicted in the genome Pectobacterium phage CB1 identified using ARNold and QuikFold, Supplementary information 1, Table S6. High ∆G rho-independent terminators predicted in the genome Pectobacterium phage CB3 identified using ARNold and QuikFold, Supplementary information 1, Table S7. High ∆G rho-independent terminators predicted in the genome Pectobacterium phage CB4 identified using ARNold and QuikFold. Supplementary information 1, Table S8. Shared high ∆G putative rho-independent terminators among Pectobacterium phages CB1, CB3 and CB4. Supplementary information 2, Table S1. Genome annotation of Pectobacterium phage CB1, Supplementary information 3, Table S1. Genome annotation of Pectobacterium phage CB3, Supplementary information 4, Table S1. Genome annotation of Pectobacterium phage CB4, Supplementary information 5, Table S1. Output from ESI-MS/MS from analysis of Pectobacterium phage CB1 virion proteins, Supplementary information 6, Table S1. Bacteria strains used in the Isolation and the testing of host range of Pectobacterium phages CB1, CB3 and CB4, Supplementary information 6, Table S2. Table S2. Genbank details of N4-like phages used in phylograms and Gegenees analysis.
Author Contributions: C.B. conducted the majority of work and wrote the manuscript, H.H. prepared phage virion proteins for MS and analyzed resulting output, A.L. assembled phage genomes from sequence reads, H.N. and C.F. produced TEM imagines of phages and critically read the manuscript, J.-P.N. provided resources for MS, R.L. provided bacterial strains for host range study and critically read the manuscript, J.O. critically read the manuscript and A.C. conceived, funded and guided the study.