Molecular Characterization and Taxonomic Assignment of Three Phage Isolates from a Collection Infecting Pseudomonas syringae pv. actinidiae and P. syringae pv. phaseolicola from Northern Italy

Abstract

Bacterial kiwifruit vine disease (Pseudomonas syringae pv. actinidiae, Psa) and halo blight of bean (P. syringae pv. phaseolicola, Pph) are routinely treated with copper, leading to environmental pollution and bacterial copper resistance. An alternative sustainable control method could be based on bacteriophages, as phage biocontrol offers high specificity and does not result in the spread of toxic residues into the environment or the food chain. In this research, specific phages suitable for phage-based biocontrol strategies effective against Psa and Pph were isolated and characterized. In total, sixteen lytic Pph phage isolates and seven lytic Psa phage isolates were isolated from soil in Piedmont and Veneto in northern Italy. Genome characterization of fifteen selected phages revealed that the isolated Pph phages were highly similar and could be considered as isolates of a novel species, whereas the isolated Psa phages grouped into four distinct clades, two of which represent putative novel species. No lysogeny-, virulence- or toxin-related genes were found in four phages, making them suitable for potential biocontrol purposes. A partial biological characterization including a host range analysis was performed on a representative subset of these isolates. This analysis was a prerequisite to assess their efficacy in greenhouse and in field trials, using different delivery strategies.

Keywords: Pph; Psa; Pseudomonas syringae; bean; bean halo blight; biocontrol; kiwifruit; kiwifruit canker; phage.

Figure 1

Plaque morphology and electron microscopy images of the three phages representing putative new species: (A) pphageB1 capsid has an icosahedric head with a short tail and clear fibers, typical for a podovirus morphology; (B) psageA1 has an icosahedral head and a putative contractile tail typical for the myovirus morphology; (C) psageB1 shows a long and flexible non-contractile tail linked to an icosahedral head, a typical siphovirus morphology.

Figure 2

Genome organization of pphageB1 phage compared with MR18 (A), psageB1 compared with nickie (B), psageA1 (C), psageK4 compared with psageK4e and phiPsa267 (D), and psageB2 compared with phiPsa1 (E). The arrows indicate annotated ORFs; and the asterisks, annotated tRNA sequences. Yellow arrows indicate ORFs associated with structural proteins, blue DNA- and metabolism-associated ORF, red ORFs that encode for lysis associated proteins, and purple ORFs homologues to terminases. The intensity of the color between two compared sequences indicates percentages of BLASTn similarity.

Figure 2

Genome organization of pphageB1 phage compared with MR18 (A), psageB1 compared with nickie (B), psageA1 (C), psageK4 compared with psageK4e and phiPsa267 (D), and psageB2 compared with phiPsa1 (E). The arrows indicate annotated ORFs; and the asterisks, annotated tRNA sequences. Yellow arrows indicate ORFs associated with structural proteins, blue DNA- and metabolism-associated ORF, red ORFs that encode for lysis associated proteins, and purple ORFs homologues to terminases. The intensity of the color between two compared sequences indicates percentages of BLASTn similarity.

Figure 3

The charts show the viability of pphageB1, psageK4, psageB1 and psageA1 phages under exposure to different pHs (A), or temperatures (B), for one hour, and to different times of UV-C irradiation (C). The number of replicates for every condition is four, and the error bars indicate the standard error. The statistical significance was calculated through the agricolae R package and consists of ANOVA (ANOVA p-values are shown under every plot) and Kruskal–Wallis analysis (significant differences indicated by letters over the bars). Different lower-case letters above each bar plot represent statistically different values.

Figure 4

Phylogenetic analysis of a selected number of Major Capsid Protein (MCP) sequences similar to the MCP of pphageB1 phage (A), of a selected number of DNA ligases similar to the DNA ligase of psageA1, psageK4 and psageK4e phages (B), and of a selected number of MCP similar to the psageB1 MCP (C). The Escherichia virus T7 MCP and DNA ligase were used as an outgroup for psageB1 and psageA1. The N4 phage MCP sequence was used as an outgroup for pphageB1 tree. The maximum likelihood methodology was used to obtain the best tree. After comparisons, the chosen best model of substitution was WAG+G4 for the pphageB1 and psageB1 trees and Blosum62+G4 for the psageA1 tree. Consensus trees were constructed from 1000 bootstrap trees. PphageB1, psageA1 and psageB1 are indicated in bold characters. At the nodes, the bootstrap values, expressed in percentages, are indicated.