Pectobacterium Phage Jarilo Displays Broad Host Range and Represents a Novel Genus of Bacteriophages Within the Family Autographiviridae

Abstract

Background: Soft rot Pectobacteriaceae includes the genera Pectobacterium and Dickeya, which are important plant pathogens being responsible for diseases in a wide range of plant species, with potatoes as the main group. Both genera cause pre- and postharvest losses of potatoes, resulting in huge economic losses linked with the soft rot diseases.

Materials and Methods: Organic waste was used to isolate phages, with Pectobacterium carotovorum subsp. carotovorum DSM 30170 as host. Complete genome sequencing, comparative genomics, and electron microscopy were used to characterize the phage. An adsorption assay was used to estimate adsorption rate. Twenty-three strains from the genera Pectobacterium and Dickeya were used to examine the host range of the phage.

Results: Pectobacterium phage Jarilo represents a novel genus of bacteriophages within the family Autographiviridae, order Caudovirales. Jarilo possesses a double-stranded DNA genome of 40557 bp with a G+C% content of 50.08% and 50 predicted open reading frames. Gene synteny and products seem to be partly conserved between Pectobacterium phage Jarilo and Enterobacteria phage T7, but limited nucleotide similarity is found between Jarilo and other phages within the family Autographiviridae. The adsorption rate of phage Jarilo increased continuously for 1 h upon infection. Phage Jarilo was not able to infect any strains of P. carotovorum and Dickeya tested with the exception of the P. carotovorum strain used for isolation. However, phage Jarilo infected 10 of 16 Pectobacterium atrosepticum strains tested.

Conclusion: We propose Pectobacterium phage Jarilo as the first member of a new genus of bacteriophages within the family Autographiviridae, order Caudovirales, displaying a broad host range within the genera of Pectobacterium.

Introduction

Pectobacterium carotovorum is part of the family Pectobacteriaceae, which members include important bacterial plant pathogens involved in a wide range of diseases.¹ Pectobacteriaceae are gram-negative facultative anaerobes, which are nonspore forming and produce extracellular enzymes involved in their pathogenicity.² P. carotovorum (formerly Erwinia carotovora) is able to form soft rot disease symptoms in various plant species and is considered to be among the top 10 most important plant pathogens.³

In potatoes, P. carotovorum, together with other members of soft rot Pectobacteriaceae, is the causative agent of the two diseases black leg and soft rot, affecting both potato plants in the growing season and postharvest through rotting of tubers during storage.^4–6 Infection of seed potatoes is an important source of infection. Infected seeds reintroduce the pathogen into the field to constitute a continued disease cycle.⁷ Control mechanisms of bacterial soft rot are currently limited and primarily based on sanitary conditions and good agricultural practice.⁸ Owing to these limitations, increased attention has been directed toward biological control agents.⁹ Phages have been proposed as biocontrol agents toward Pectobacteriaceae in several studies.^10–15

Historically, incorporation of new viruses into the international committee on taxonomy of viruses (ICTV) formal taxonomy has been dependent on knowledge on their phenotypic properties. However, this approach has been challenged during the past decade with an explosion in virus nucleotide data due to high-throughput sequencing methods.^16,17 Consequently, the phage classification system is currently changing toward primarily genomic-based methods.¹⁸ ICTV recently updated the taxonomy system with a change in phage orders, families, subfamilies, genera, and species. In this report, we describe the isolation and characterization of Pectobacterium phage Jarilo, representing a new genus of bacteriophages within the family Autographiviridae.

Materials and Methods

Phage isolation and DNA extraction

Jarilo was isolated from an organic waste sample as described previously,¹⁵ using P. carotovorum subsp. carotovorum DSM 30170 (DSM30170) as host. Before isolation, virus particles in the organic waste sample were concentrated by polyethylene glycol precipitation, as described elsewhere.¹⁹ Phage DNA was isolated and sequencing libraries were prepared using direct plaque sequencing protocol,²⁰ with the modification that sodium dodecyl sulfate was used at a final concentration of 0.1%.²¹

Genome sequencing and bioinformatics

Reads from sequencing were trimmed and de novo assembled using CLC Genomic Workbench (10.1.1). Open reading frame (ORF) prediction was executed using DNA Master (version 5.0.2), with GeneMark²² and Glimmer²³ as gene caller, and was corrected manually. Putative gene functions were manually curated using blastp²⁴ and HHpred²⁵ with the databases Pfam (version 32.0), SCOP70 (version 1.7.5), and pdb70. tRNA sequences were searched for using tRNAscan-SE.²⁶ To obtain highly concentrated phage preparation for transmission electron microscopy (TEM), the phage lysate was purified by CsCl gradient ultracentrifugation as described here.²⁷

Transmission electron microscopy

Bacteriophage Jarilo was prepared for micrographs as follows: 100 μL lysate (10¹⁰ pfu/mL) were microdialyzed for 30 min against SM-buffer (20 mM Tris-HCl [pH 7.2], 10 mM NaCl, 20 mM MgSO₄) on a Millipore VSWP 0.025 μm filter membrane (Merck Darmstadt, DE). Phages were adsorbed to ultrathin carbon films for 30 min. After fixation with 1% (v/v) glutaraldehyde (electron microscopy grade) for 30 min, phages were negatively stained with 2% (w/v) uranyl acetate. Carbon films were picked up with standard 400 mesh copper grids (3 mm diameter). TEM was performed at an accelerating voltage of 80 kV (Tecnai 10; FEI Thermo Fisher Scientific, Eindhoven, The Netherlands). Micrographs were captured with a MegaView G2 CCD camera (Emsis, Muenster, Germany). Number of particles measured to estimate head diameter and tail length was n = 7 and n = 6, respectively.

Adsorption assay

To estimate adsorption rate, an adsorption assay was performed with the host strain DSM30170 and Pseudomonas syringae DC3000 (DC3000) as a test strain, indicating the amount of non-adsorbing phages present in the pellet (Supplementary Fig. S1).

DSM30170 was grown in tryptic soy broth (TSB) at 28°C, DC3000 in LB-Miller broth at 28°C over night (ON). ON cultures were reinoculated in fresh medium and grown to exponential phase, ∼1 × 10⁷ colony forming units (CFU)/mL, DSM30170 was grown to OD_600nm = 0.05, DC3000 was grown to OD_600nm = 0.03. DSM30170 and DC3000 have a doubling time of 60 min and 90 min, respectively. Owing to the observation that DSM30170 entered stationary phase relatively early, the culture was grown to OD_600nm = 0.05 corresponding to ∼1 × 10⁷ CFU/mL. To obtain the same number of cells in both cultures, DC3000 was grown to OD_600nm = 0.03, based on previous studies.²⁸

The bacterial suspensions were aliquoted in equal volumes. Phage Jarilo was added to the bacterial suspensions at a multiplicity of infection of 0.01. Samples of 1 mL were taken every 15 min for 1 h. Samples were centrifuged at 10.000 × g for 1 min, the supernatant was removed, and the pellet was resuspended in 1 mL of TSB. Phage titer of both supernatant and pellet fractions was determined by plaque assay: 10 μL was spotted in serial dilutions, on TSB agarose (4 g/L and 10 mM CaCl₂+MgCl₂) double overlay on tryptic soy agar (TSA) plates, with 100 μL of ON culture of DSM30170. The experiment was carried out in triplicates, resulting in nine samples per time point. Plates were incubated at 28°C ON before the number of plaques was counted.

Adsorption rate was determined as phage titer of resuspended pellet on host strain excluding phage titer of resuspended pellet on the test strain, in relation to the total amount of added phages, calculated by the phage titer of supernatant and pellet combined.

Host range experiment

To determine initial host range, Jarilo was tested against the isolation host DSM30170, 21 Pectobacterium strains²⁹ and 2 Dickeya strains (Supplementary Table S1). All strains were grown in TSB at 28°C. 10 μl of phage Jarilo lysate (∼5 × 10⁷ PFU/mL) were spotted on the bacterial lawn, using TSB agarose (4 g/L and 10 mM CaCl₂+MgCl₂) double overlay on TSA plates, with 100 μL of ON culture of the respective host. Plates were incubated at 28°C ON before presence of clearance zones was noted.

Strains being susceptible toward phage Jarilo, in the initial host range using high titer lysate, were subsequently used to determine relative efficiency of plaquing (EOP) of phage Jarilo. A serial dilution of phage Jarilo with a titer of ∼5 × 10⁷ PFU/mL was spotted on the isolation host DSM30170 and the following hosts as test strains: 10 of 16 isolates of Pectobacterium atrosepticum (CB BL 2-1, 3-1, 4-1, 7-1, 9-1, 11-1, 12-2, 13-1, 14-1, 16-1).²⁹ All strains were grown in TSB at 28°C. Phage titer was determined by plaque assay: 10 μL of phage Jarilo lysate (∼5 × 10⁷ PFU/mL), in dilutions, were spotted on the bacterial lawn, using TSB agarose (4 g/L and 10 mM CaCl₂+MgCl₂) double overlay on TSA plates, with 100 μL of ON culture of the respective host. Plates were incubated at 28°C ON before number of plaques was counted. EOP was determined by phage titer on the given test strain in relation to phage titer on the isolation host.

Results

Genome analysis

Pectobacterium phage Jarilo has a double-stranded DNA genome of 40557 bp with a G+C% content of 50.08%. In the genome there are 50 predicted ORFs, where the majority (30/50) could be assigned with a putative function due to homology with other known phage genomes within the family Autographiviridae. No tRNAs were detected. Based on their predicted functions, ORFs were categorized into five gene categories: hypothetical protein (grey), DNA replication and metabolism (red), lysis proteins (yellow), others/unknown (green), and morphogenesis (blue) (Fig. 1). The gene synteny of the genome of Jarilo is partly conserved among phages within the family Autographiviridae, especially for many of the structural genes (Fig. 1).

FIG. 1.

Comparative genomics of six genomes of bacteriophages within the family Autographiviridae using blastn in the Easyfig software (version 2.2.3).³⁰ Phages showing highest scores, using blastn,²⁴ against the genome of Jarilo were chosen together with phage T7 as a reference genome. Putative gene functions of the genome of Jarilo were divided into five categories: hypothetical proteins (grey), DNA replication and metabolism (red), lysis proteins (yellow), others/unknown (green), and morphogenesis (blue). Following phage genomes were used (with Genbank accession numbers): Yersinia phage AP10 (KT852574.1), Klebsiella phage KOX3 (MN101216.1), Enterobacter phage E-3 (KP791806.1), Citrobacter phage SH2 (KU687348.1), and Enterobacteria phage T7 (V01146.1). Green arrow points out gene encoding endonuclease VII, pink arrow points out gene encoding major capsid protein, and turquoise arrow points out gene encoding tail fiber protein.

Like other phages within the family Autographiviridae, Jarilo encodes an RNA polymerase³⁰ and does not encode any detectable integrase. Despite the shared gene synteny and somewhat shared nucleotide similarity within the gene encoding the DNA polymerase between phage Jarilo, T7 and Yersinia phage AP10, an insertion of a putative endonuclease VII appears to have happened within the DNA polymerase gene, in the genome of Jarilo. This insertion of 768 bp separates the DNA polymerase gene into two ORFs, both ORFs share high sequence similarity with the DNA polymerase of related phages (green arrow in Fig. 1).

Pectobacterium phage Jarilo exposes limited DNA sequence similarity with other phages within the family Autographiviridae, with Yersinia phage AP10 as the closest relative only sharing an intergenomic similarity score of 51.5 (Fig. 2).

FIG. 2.

Heatmap displaying nucleotide-based intergenomic similarities between six genomes of bacteriophages within the family Autographiviridae, using the VIRIDIC web server (default settings).³¹ Phages showing highest scores, using blastn,²⁴ against the genome of Jarilo were chosen together with phage T7 as a reference genome. Colors indicate similarity between phages, from similar (dark blue) to less similar (light blue). The intergenomic similarity scores (sim) are based on the output of blastn²⁴ 2.9.0+ alignment, where the identity score (id) of each genome (A+B) is combined and multiplied by 100, in relation to the sequence length (l) of each genome in total, simAB = ((idAB + idBA) × 100)/(lA + lB)). Following phage genomes were used (with Genbank accession numbers): Yersinia phage AP10 (KT852574.1), Klebsiella phage KOX3 (MN101216.1), Enterobacter phage E-3 (KP791806.1), Citrobacter phage SH2 (KU687348.1), and Enterobacteria phage T7 (V01146.1).

Morphology

TEM of Jarilo showed common Autographiviridae morphological characteristics, resembling the Podoviridae family, order Caudovirales, with an isometric head (head diameter: 56.8 ± 2.7 nm) and a short tail stub without appendages or fibers (tail length: 13.3 ± 2.7 nm) (Fig. 3).

FIG. 3.

Transmission electron micrographs of Pectobacterium phage Jarilo negatively stained with 2% (w/v) uranyl acetate. The isometric-headed phage has a short cone-shaped phage stub without discernible appendages or fibers.

Biological characterization

The phage adsorption rate for bacteriophage Jarilo increased continuously 1 h after infection (Fig. 4), indicating a relatively slow adsorption. A host range experiment revealed phage Jarilo not to be able to infect any of the tested P. carotovorum strains except the isolation host DSM30170. Phage Jarilo was able to infect 10 of 16 tested P. atrosepticum strains. However, 3 out of 10 infected P. atrosepticum strains only showed clearance and no plaque formation (Table 1). In the case of these three strains, clearance was obtained when using high titers (5 × 10⁷ and 5 × 10⁶ PFU/mL) but with no individual plaque formation at lower titer dilutions, indicating no phage propagation under these experimental conditions. Phage Jarilo did not show the ability to infect any of the tested Dickeya strains.

FIG. 4.

Adsorption assay for bacteriophage Jarilo. Adsorption rate is based on phage titer of host sample (Pectobacterium carotovorum DSM30170) excluding phage titer of negative adsorption control sample (Psuedomonas syringae DC3000) in relation to the total amount of added phages. T1, T2, and T3 represent biological replicate 1, 2, and 3. Standard deviation within each replicate is represented by error bars.

Table 1.

Infectivity of Phage Jarilo

Species	Strain	Sensitivity
Pectobacterium carotovorum subs. carotovorum	DSM30170*	1.00
	DSM14775	—
	DSM102074	—
	DSM108906	—
P. carotovorum subs. odoriferum	DSM22556	—
Pectobacterium atrosepticum	CB BL 1-1	—
	CB BL 2-1	+
	CB BL 3-1	0.55
	CB BL 4-1	(+)
	CB BL 5-1	—
	CB BL 7-1	1.44
	CB BL 9-1	0.14
	CB BL 11-1	0.80
	CB BL 12-2	0.14
	CB BL 13-1	0.90
	CB BL 14-1	1.14
	CB BL 15-1	—
	CB BL 16-1	+
	CB BL 18-1	—
	CB BL 19-1	—
	ATCC 33260	—
Dickeya dadantii	NCPPB 4097	—
Dickeya solani	DSM28711	—

^* Represent host organism.

Sensitivity is determined by EOP value. —, no infection; (+), presence of clearance with no plaque formation.

Discussion

As previously mentioned, phage Jarilo shares gene synteny with phages within the family Autographiviridae, but an insertion of a putative endonuclease VII appears to have split the DNA polymerase gene (green arrow in Fig. 1). Yersinia phage YpP-G, also part of the Autographiviridae family, also encodes an endonuclease VII, which gives the highest score against the putative endonuclease VII encoded by Jarilo, using blastp, the nr database.^24,31 Curiously, YpP-G displays a similar gene arrangement with an endonuclease VII within the gene encoding the DNA polymerase (Supplementary Fig. S2). Furthermore, the same phenomenon is described in another phage within the family Autographiviridae, Pf-WMP3.³²

Interestingly, an endonuclease VII insertion into the DNA polymerase has previously been proved to constitute part of a self-splicing intron in Pseudomonas phage LUZ24³³; however, this has not previously been demonstrated for any members of the Autographiviridae family. In phage LUZ24, Ceyssens et al. confirmed the presence of an endonuclease VII by isolation of total RNA and execution of PCR on the cDNA, resulting in smaller products compared with the control reactions based on the DNA template. These results revealed the absence of the endonuclease VII gene in cDNA, which indicates self-splicing of the intron restores the reading frame of the DNA polymerase. The endonuclease VII was classified as a self-splicing group I intron.³³

Group I introns inserted in genes involved in crucial steps of DNA metabolism have been found in a wide variety of phages. It has been speculated whether self-splicing introns could have any selective advantage in form of regulation of the DNA metabolism during certain circumstances. One such regulatory potential has been speculated of phage T4 introns. T4 is known to be infective (injection of phage DNA) during stationary phase, however, does not lyse the infected bacteria during stationary phase. The infection remains dormant and can be resumed if new nutrients are added to the medium. During certain circumstances, limitations of de novo production of DNA would leave more energy available for the cell to survive and hence the production of new phage progeny.³⁴

It has also been speculated whether the presence of multiple introns in a single gene may allow synthesis of several products. However, phage introns have also been hypothesized to be solely selfish, with intron-encoded DNA endonucleases promoting the mobility of the intron.^35,36

The gene encoding the capsid protein in Escherichia phage T7 has two products, a major (10A) and a minor (10B), produced by frameshifting into the -1 frame near the end of 10A.³⁷ The gene encoding the major capsid protein as well as a short gene downstream, categorized as a hypothetical gene, in Jarilo, resembles the capsid gene in T7 with some similarity, which could be the result of a frameshift as well (pink arrow in Fig. 1). The gene encoding head to tail joining protein appears to be divided into two (Fig. 1), which also could be a result of frameshifting, as in the capsid protein of T7, or due to a nonsense mutation.

Genes categorized as being involved in morphogenesis in Jarilo are also conserved among phages within the family Autographiviridae; however, the tail fiber protein is conserved to a lesser extent, which could represent a difference in receptor preferences, which makes sense given their difference in isolation hosts (turquoise arrow in Fig. 1).

Genomic analysis of tailed phages in general indicates that functionally different types of genes are not equally prone to variation.^38,39 Interestingly, Jarilo, T7, and AP10 share a conserved region at the terminal of the genome, not representing an ORF (Fig. 1). However, noncoding genes may not imply the lack of biological function, as some noncoding RNAs or DNA binding motifs, such as promotors and terminators, could be present within these noncoding regions.⁴⁰

Phage Jarilo showed a relatively low adsorption rate. The low rate could be due to the relatively low cell density in the bacterial suspension upon infection (1 × 10⁷ CFU/mL). The relatively low cell density of P. carotovorum upon infection was chosen due to an observation that the cells enter stationary phase early.

The host range results did, surprisingly, show phage Jarilo to be infective toward both P. carotovorum and P. atrosepticum. A possible reason for the ability to infect across species could be due to common receptors on the given host. Three P. atrosepticum strains did, however, show clearance but no signs of plaque formation, which could be due to the presence of antiphage defense systems or lysis from without.⁴¹ The limited host range within P. carotovorum strains together with the broader host range within P. atrosepticum strains suggests phage Jarilo as a P. atrosepticum phage isolated on P. carotovorum by coincidence.

Phage Jarilo can be considered as a broad host range phage with therapeutic potential, targeting both P. atrosepticum and P. carotovorum. These species are both considered important plant pathogens resulting in huge economic losses linked with the soft rot diseases.³ Phages with broader host range have been implied to be of interest concerning phage therapy, being compared with broad spectrum antibiotics.⁴² The host range results indicate that phage Jarilo only affects closely related bacterial species and, therefore, it is not expected to affect overall plant or soil microbiome; however, this has not been investigated.

The Podoviridae subfamily Autographivirinae has recently been changed into the family Autographiviridae within the order Caudovirales. Morphological characteristics of the phages within the family Autographiviridae resemble the Podoviridae family, with a small (ca. 60 nm in diameter) isometric head attached to a short tail.⁴³ Phage Jarilo does likewise possess these morphological characteristics.

Other common characteristics of this family are the presence of an RNA polymerase, conservation of gene arrangement, and a genus-specific lysis cassette.⁴³ Phage Jarilo also encodes an RNA polymerase and has a somewhat conserved gene arrangement compared with the closest related phages, with a lysis cassette existing of holin, endopeptidase Rz, and Rz1-like protein, only interrupted by the gene encoding the small subunit terminase (Fig. 1), whereas an endolysin, N-acetylmuramoyl-l-alanine amidase, is found upstream from the lysis cassette. This arrangement is, nonetheless, somewhat conserved between phage Jarilo and the closest relatives (Fig. 1). Holins are hydrophobic proteins that oligomerize in the cytoplasmic membrane, triggering hole formation, whereas endolysins are peptidoglycan-degrading enzymes. In Jarilo, Rz and Rz1 are found as overlapping genes that share the same DNA in different reading frames, a similar constellation can be found in other Autographiviridae, including T7.

No gene encoding an integrase was found in the genome of phage Jarilo, implying phage Jarilo is a strictly lytic phage.

Conclusion

Based on the findings, we propose Pectobacterium phage Jarilo as the first representative of a new genus of bacteriophages belonging to the family Autographiviridae, order Caudovirales. Phage Jarilo possesses limited DNA sequence similarity, with an intergenomic similarity score of 51.5 to the closest relative (Fig. 2).⁴⁴ The genome of phage Jarilo indicates phage Jarilo to be strictly lytic and can be considered as a broad host range phage infectious toward both P. carotovorum and P. atrosepticum, which makes it interesting to consider as a biocontrol agent.

Authors' Contributions

The following is the contribution of authors: conceptualization was by J.S.P., A.M.D., A.B.C., and L.H.H.; methodology was J.S.P., A.M.D., A.B.C., W.K., H.N., and L.H.H.; validation was by J.S.P., A.M.D., A.B.C., W.K., and L.H.H.; formal analysis was by J.S.P., H.N., and W.K.; investigation was by J.S.P., A.M.D., and A.B.C.; resources were by W.K. and L.H.H.; data curation was by J.S.P., A.M.D., and A.B.C.; writing—original draft preparation was by J.S.P.; writing—review and editing were by J.S.P., A.M.D., A.B.C, W.K., H.N., and L.H.H.; visualization was by J.S.P. and H.N.; supervision was by A.M.D., A.B.C., W.K., and L.H.H.; project administration was by W.K. and L.H.H.; funding acquisition was by W.K. and L.H.H. All coauthors have reviewed and approved on the article before submission.

Nucleotide Sequence Accession Number

The GenBank accession number for Pectobacterium phage Jarilo is MH059637.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by the Danish Research Council for Technology and Production Grants (Grant Nos. DFF-4184-00070 and DFF-0130-00005B), the Human Frontier Science Program (Grant No. RGP0024/2018), and the Villum Experiment (Grant No. 17595).

Supplementary Material

Supplementary Table S1

Supplementary Figure S1

Supplementary Figure S2