Introduction

Pseudomonas syringae pv. actinidiae is a gram-negative phytopathogen that causes bacterial canker disease in kiwifruit trees [23]. This pathogen damages both Actinidia deliciosa (green kiwifruit) and A. chinensis (gold kiwifruit), resulting in severe worldwide economic losses. P. syringae pv. actinidiae was first isolated in Japan in 1984 as the causative agent of bacterial canker in A. deliciosa [32] and was subsequently recorded in Korea [22], Italy [29], Portugal [1], Spain [2], France [33], and Turkey [33]. Worst hit by the recent global outbreaks are the kiwifruit industries in Italy and New Zealand. In New Zealand, kiwifruit canker disease was first discovered in 2010 [12]; three years later, kiwifruit canker disease had spread to 1,400 orchards, contaminating 52% of all kiwifruit orchards in the country [11]. In addition, the kiwifruit industries in Japan, Spain, Portugal, and France have also been severely affected [30]. P. syringae pv. actinidiae can be classified into four biovars. Psa1 (biovar 1) was discovered in Japan and produces and secretes phaseolotoxin. Psa2 (biovar 2) was discovered in Korea and secretes coronatine. Psa3 (biovar 3) is a newly discovered strain that is highly pathogenic and secretes many effector proteins. Finally, Psa4 (biovar 4) is an avirulent strain [6].

Bacterial canker disease is very problematic in all kiwifruit-cultivating regions, including Korea. Effective control methods for kiwifruit canker disease have been investigated [5], and the common control methods of P. syringae pv. actinidiae are to spray crops with streptomycin or copper compounds. However, streptomycin is not a viable control option in many countries because of antibiotic residues in the fruits [15]. Furthermore, copper and streptomycin resistance genes have been detected in P. syringae pv. actinidiae [18,28]. Therefore, alternative methods to control this disease should be developed.

Bacteriophages are viruses that specifically infect target bacteria. Structurally, the bacteriophage head is composed of a nucleic acid genome within a protein capsid. Some bacteriophages have a tail consisting of lipids or proteins [8]. Most bacteriophages contain double-stranded DNA as their genetic material. More than 95% of known bacteriophages belong to the order Caudovirales [27]. The three main families of bacteriophages are distinct based on morphological characteristics, in particular, tail shape. Myoviridae phages have double-layered contractile tails, Siphoviridae phages have long flexible tails, and Podoviridae phages have short stubby tails [9]. Bacteriophages are abundant in all environments, including water, soil, and inside and outside of plants. Bacteriophages attach to their target bacteria, inject their DNA into the host, and self-replicate inside the host, leading to bacterial cell death [14]. Bacteriophages are relatively safe because they have no activity against animal or plant cells [25] and do not affect other beneficial microorganisms. In addition, they can be easily isolated from the native host bacterial environment.

Currently, bacteriophages have been characterized and studied as agents for phage therapy of plant diseases caused by plant-pathogenic bacteria [14], including Dickeya solani [7], Erwinia amylovora [4], Pectobacterium carotovorum [26], and Ralstonia solanacearum [34]. Recently, bacteriophages of P. syringae pv. actinidiae have been reported and characterized [10,15]. These bacteriophages were isolated from soil, water, and leaves of kiwifruit orchards infected by P. syringae pv. actinidiae in New Zealand, and some of them were further characterized. Interestingly, bacteriophages belonging to the Myoviridae were predominant.

In this study, we isolated bacteriophages against P. syringae pv. actinidiae from kiwifruit orchard soil in Korea and characterized five of them in detail. This study will provide more insights into the potential use of bacteriophages for phage therapy against bacterial canker disease in kiwifruit.

Materials and Methods

Bacterial Strains and Growth Conditions

A total of 18 P. syringae pv. actinidiae strains were used in this study, as listed in Table 1. Ten strains were isolated from various regions of Korea, and the Korean Agricultural Culture Collection (KACC) at the Rural Development Administration provided eight additional strains. All strains, including other bacteria listed in Table 2, were grown in tryptic soy broth (TSB; Difco, USA) or on agar plates at 26℃.

Table 1.KACC, Korean Agricultural Culture Collection.

Table 2.Host ranges of the five selected bacteriophages against other bacteria.

Identification of P. syringae pv. actinidiae Strains

All P. syringae pv. actinidiae strains used in this study were re-identified, and their biovars were re-confirmed using multiplex-polymerase chain reaction (m-PCR). Two primer pairs were used in this study: Psa-F (5’-CAGAGGCGCTAACGAGGAAA-3’) and Psa-R (5’-CGAGCATACATCAACAGGTCA-3’) and Tac-F (5’-CGGGCTAGACAGTACGCTGT-3’) and Tac-R (5’-CAGGCCCTTCTACCGCTAC-3’) [3,21]. The m-PCR mixture consisted of 25 μl of PCR smart mix 2 (Solgent, South Korea), 5 μl of bacterial suspension (108 CFU/ml), 1 μM primer pair, and sterile distilled water in a 50 μl final volume. Amplifications were carried out in a T100 thermal cycler (Bio-Rad, USA) and optimized to the following conditions: an initial denaturation step at 94℃ for 5 min, followed by 30 cycles of a denaturation step of 94℃ for 30 sec, annealing at 65℃ for 30 sec, and extension at 72℃ for 30 sec, with a single final extension step at 72℃ for 5 min. The amplification products were detected using electrophoresis of 5 μl amplification mixture in a 1.5% agarose gel in 1× Tris-acetate-EDTA buffer. Gels were stained with EcoDye DNA Staining Solution (Biofact, South Korea).

Bacteriophage Isolation and Purification

An approximately 100 g soil sample was collected from each of 82 kiwifruit orchards in South Korea. The bacteriophages were isolated as described previously, with minor modifications [20]. Five grams of soil sample was homogenized in 45 ml of distilled water by shaking for 30 min. After centrifugation of the culture (9,000 ×g, 10 min, 4℃) and filtration of the clear supernatant (0.2-μm-pore-size filter; Pall, USA), the filtrate (5 ml) was mixed with 25 ml of TSB and a 500 μl overnight culture of P. syringae pv. actinidiae. The bacteria-phage mixture was shaken at 200 rpm with incubation at 26℃ for 18 h, followed by centrifugation and filtration as described above. Ten-fold serial dilutions of these filtrates were used in the spotting assay with P. syringae pv. actinidiae to confirm the presence of bacteriophages. To isolate and purify the bacteriophages, the overlay assay was carried out as previously described [24]. Each plaque showing a unique morphology was collected with a sterile tip, and bacteriophages were eluted with sodium chloride-magnesium sulfate (SM) buffer (50 mM Tris-HCl (pH 7.5), 100 mM NaCl, and 10 mM MgSO4). For purification, this process was repeated at least three times per plaque.

Spotting and Overlay Assays

The spotting assay was carried out as described previously with minor modifications [20]. The host bacteria were cultured overnight, and then 100 μl (109 CFU/ml) of bacterial culture was added to 5 ml of soft TSA agar (0.4% agar). After gentle mixing using a vortex mixer, the culture was poured onto prepared TSA agar plates and left to solidify at room temperature for 20 min. Then, 10 μl of bacteriophage stock dilutions was spotted on the top agar layer, and the plates were dried at room temperature for 30 min. The cultures were incubated overnight at 26℃ to produce bacterial lawns and were inspected the next day for plaques or bacterial growth inhibition zones.

The overlay assay was performed as described previously with modifications [20]. Briefly, bacteriophage stock dilutions (100 μl) were mixed with 100 μl of overnight bacterial culture (109 CFU/ml) in 5 ml of soft TSA agar (0.4% agar), and the mixture was poured onto plates. After the medium was left to solidify for 30 min at room temperature, the plates were incubated at 26℃ and plaques were examined the next day.

Propagation and Purification of Bacteriophages

Propagation of bacteriophages was carried out as described previously with minor modifications [20]. Briefly, the lysate of a single bacteriophage plaque was added to a P. syringae pv. actinidiae culture (optical density at 600 nm (OD600) = 0.5 to 0.6), which was then incubated at 26℃. The culture cleared of cellular debris (host bacterial lysate) was treated with chloroform (1% of final volume), incubated at 26℃ for 5 min, and then centrifuged (15,000 ×g, 10 min, 4℃). Bacteriophage particles in the filtered supernatant (0.22-μm-pore-size filter) were precipitated with 10% (w/v) polyethylene g lycol (PEG) 6000 in 1 M NaCl at 4℃ f or 10 h. After centrifugation (10,000 ×g, 15 min, 4℃), precipitated bacteriophages were resuspended in SM buffer and separated by CsCl density gradient ultracentrifugation (78,500 ×g, 2 h, 4℃). Each bacteriophage band was extracted from the ultracentrifuge tubes and dialysis was performed in SM buffer. Finally, the phage stock was stored in glass tubes at 4℃ until use.

Morphology of Bacteriophages

A solution containing bacteriophages was placed on carbon-coated copper grids, and 2% aqueous uranyl acetate (pH 4.0) was added for 20 sec to negatively stain the phage particles. Purified bacteriophages were examined using transmission electron microscopy (TEM; LEO 912AB, Carl Zeiss, Germany), and images were scanned with a Proscan 1,024 × 1,024 pixel charge-coupled device camera. Bacteriophages were classified according to the guidelines of the International Committee on Taxonomy of Viruses [13].

Extraction of Bacteriophage Genomic DNA and Restriction Enzyme Digestion

Bacteriophage genomic DNA was extracted using a phage DNA isolation kit (Norgen Biotek, Canada). Restriction enzyme digestions with 1 μg of bacteriophage genomic DNA were carried out using EcoRI, EcoRV, NcoI, StuI, SphI, XhoI, RsaI, DraI, HinfI, Acc65I, KpnI, Sau3AI, SphI, Tsp45I, MnlI, or NheI according to the manufacturer’s instructions (Promega Korea, Seoul, Korea). After digestion, electrophoresis of the samples was performed in 1% agarose containing EcoDye.

Lytic Activity of Bacteriophages

Propagation of bacteriophages was conducted as described previously with minor modifications [26]. TSB inoculated with an overnight culture of P. syringae pv. actinidiae was incubated at 26℃ with shaking. When the culture reached the early exponential phase (108 CFU/ml), bacteriophage lysate was added at a multiplicity of infection (MOI) of 0.01. The OD600 was measured for 80 h using a TECAN microplate reader (TECAN, Männedorf, Switzerland). As a negative control, a bacterial culture was inoculated with the same volume of SM buffer instead of bacteriophage lysate.

Stability of Bacteriophages Under Various Conditions

To investigate bacteriophage stability at various temperatures, phage aliquots (106 pfu/ml) in SM buffer were incubated at 30℃, 40℃, 50℃, or 60℃ for 1 h. After incubation, viable phage titers were enumerated using a plaque assay. For pH stability, bacteriophage aliquots were added to SM buffer (final concentration 106 pfu/ml) adjusted with HCl or NaOH to a pH range of 3-12 and incubated at room temperature for 1 h. After incubation, phage viability was determined using a plaque assay. For ultraviolet (UV) sensitivity, bacteriophage aliquots in SM buffer were incubated in an empty Petri dish and placed over a 306 or 365 nm UV lamp (8W, dual UV transilluminator; Core Bio, South Korea). Bacteriophages were sampled every 15 min for 1 h, and their viable phage titers were enumerated using a plaque assay.

Results and Discussion

Isolation and Purification of Bacteriophages Against P. syringae pv. actinidiae

Bacteriophages are present in the surrounding soil, water, and plants of kiwifruit orchards, where host bacteria are also most likely present. Therefore, we collected soil from kiwifruit orchards in order to isolate bacteriophages against P. syringae pv. actinidiae. A total of 82 soil samples from inside or outside the orchards of 11 different regions (mostly located in the southern part of the Korean peninsula) were collected and screened for bacteriophages against 10 different P. syringae pv. actinidiae strains isolated in Korea using plaque assays. A total of 261 plaques were observed and, depending on plaque appearance, were divided into two groups; ++ (clear plaques) and + (turbid plaques). Seventy plaques belonging to the ++ group were used to purify individual bacteriophages through three repetitions of the plaque assay. As a result, bacteriophages were successfully isolated and purified from 70 of the 82 soil samples collected from kiwifruit orchards.

Restriction Enzyme Digestion Patterns of Bacteriophage Genomic DNA

Genomic DNA was extracted from the 70 purified bacteriophages using phage DNA isolation kits. The genomic DNA of each bacteriophage was first digested with EcoRI, SphI, and NheI restriction enzymes. Most of the bacteriophage genomic DNAs were digested with EcoRI, and showed 4 to 5 unique bands with different sizes (Fig. 1). Based on the sizes of these bands, their genome sizes were estimated (Table 3). Their digested patterns were compared, and three bacteriophages, KHUφ38, KHUφ59, and KHUφ74, showed different size patterns and were selected for further assays (Fig. 1). Genomic DNAs of some bacteriophages, including KHUφ34 and KHUφ44, were not digested with the first three restriction enzymes (Fig. 1). The following enzymes were tested but failed to produce results; EcoRV, NcoI, StuI, XhoI, RsaI, DraI, HinfI, Acc65I, KpnI, Sau3AI, SphI, Tsp45I, and MnlI. Thus, KHUφ34 and KHUφ44 were also selected for further assays. These results indicate that more than five groups of bacteriophages effective against P. syringae pv. actinidiae are present in the soils of kiwifruit orchards in Korea.

Fig. 1.Genomic DNA band patterns of five selected bacteriophages digested with EcoRI. Genomic DNA (1 μg) of each bacteriophage was treated with EcoRI for 24 h, and restriction patterns were visualized using gel electrophoresis. M, 1 kb ladder.

Table 3.aBased on restriction enzyme digestion with EcoRI.

Host Ranges of the Five Selected Bacteriophages

The host ranges of the five selected bacteriophages were first determined with 18 strains of P. syringae pv. actinidiae, which were all isolated in Korea (Table 1). Biovars of these strains were confirmed, and 14 strains belonged to biovar 2 and four belonged to biovar 3. We then determined the effectiveness of each bacteriophage to each host strain. Bacteriophage KHUφ44 was very effective against all 14 strains of P. syringae pv. actinidiae belonging to biovar 2 and was also effective against SYS1 and SYS4 belonging to biovar 3 (Table 1), although it had limited effects on SYS2 and SYS3. Some bacteriophages were very specific to certain strains, and only KHUφ44 and KHUφ34 were effective against the KGY4 strain, belonging to biovar 2, whereas KHUφ34 was specific to the SYS3 biovar 3 strain (Table 1). These results indicate that the bacteriophages isolated in this study have specificity to their target bacterium, P. syringae pv. actinidiae, and the specificity persists at the strain level of this bacterium.

We next determined whether the isolated bacteriophages were effective against the 13 other bacteria, including three different P. syringae species pathovars (Table 2). Interestingly, most of the bacteriophages were also effective against other P. syringae pathovars, and none showed any effect on other bacteria. These results indicate that the isolated bacteriophages target P. syringae species, regardless of pathovar. This bacterial species includes more than 50 pathovars causing plant disease in many important crops, and these pathovars are unique to plant pathogens [17]. Therefore, the bacteriophages isolated in this study might be applicable to all P. syringae pathovars, resulting in a broad application for these bacteriophages.

Frampton et al. [15] showed that some bacteriophages are effective against other P. syringae pathovars. Interestingly, they found that some bacteriophages effective against P. syringae species could also target other Pseudomonas species, such P. viridiflava and P. fluorescens. The effectiveness of the isolated bacteriophages in our study was tested against P. fluorescens, Acidovorax species, and Ralstonia solanacearum; however, none of the isolated bacteriophages were effective against these bacterial species (Table 2), further demonstrating that these bacteriophage isolates are specific to P. syringae species.

Morphology of the Five Selected Bacteriophages

The morphology of each of the five selected bacteriophage isolates was observed using transmission electron microscopy. All bacteriophages belonged to the order Caudovirales. Based on tail shape and head size, bacteriophages KHUφ34 and KHUφ44 belong to the Myoviridae family, and the other three bacteriophages belong to the Podoviridae family (Fig. 2). Their features, including isolation location, head size, and total length, are listed in Table 3. Based on restriction enzyme digestion patterns, many of the 70 bacteriophages isolated in this study showed the same restriction pattern as KHUφ38, KHUφ59, or KHUφ74 (data not shown), indicating that the bacteriophages belonging to Podoviridae are dominant in Korea. These results are different from those of New Zealand bacteriophages, most of which belong to the family Myoviridae [15].

Fig. 2.Morphology of the five selected bacteriophages observed using transmission electron microscopy. (A) KHUφ34, (B) KHUφ38, (C) KHUφ44, (D) KHUφ59, and (E) KHUφ74. These photos were taken using a LEO 912AB transmission electron microscope (Carl Zeiss) after negative staining. Arrows indicate the representative phages.

Lytic Activity of the Five Bacteriophages Against P. syringae pv. actinidiae

Lytic activity of bacteriophages is one of the most important factors for their application as phage therapy to control plant diseases. Therefore, the lytic activity of the five selected bacteriophages was confirmed. In this assay, P. syringae pv. actinidiae strain KBE9 was first used because all five bacteriophages were effective against this strain. Bacteriophage KHUφ38 showed lytic activity slightly later than did KHUφ59 and KHUφ74, and their lytic activity increased during the first 24 h after treatment but then was slightly reduced (Fig. 3). Bacteriophages KHUφ34 and KHUφ44 did not show lytic activity until 24 h. The lytic activities of KHUφ34, KHUφ38, and KHUφ44 at 80 h after treatment were very similar. Bacteriophages KHUφ59 and KHUφ74 showed significant lytic activity within the first 24 h after treatment, but the OD600 value of the culture increased after 24 h (Fig. 3), indicating that bacterial cells very likely gained tolerance or resistance to this bacteriophage. Two other strains of P. syringae pv. actinidiae, KGY4 (biovar 2) and SYS2 (biovar 3), were tested, and results were very similar to those of the KBE9 strain. These results indicate that the lytic activity of these three bacteriophages was sustained for at least 80 h after treatment, and the pattern of lytic activity depends on bacteriophage type. Compared with previously described bacteriophages, the lytic activities of KHUφ34, KHUφ38, and KHUφ44 appear very effective [16,26,34], which will be very useful for the management of kiwifruit canker disease as a phage therapy.

Fig. 3.Lytic activity of the five bacteriophages against P. syringae pv. actinidiae strain KBE9. Each bacteriophage (0.01 MOI) was added to a bacterial suspension (108 CFU/ml) in the early exponential phase, and then OD600 was measured at the designated time points. Con, bacterial growth in the same medium with SM buffer. The error bars indicate standard error.

Stability of the Five Bacteriophages to Temperature, pH, and UV Light

For the application of bacteriophages as a phage therapy to control plant diseases, the stability of bacteriophages to certain environmental factors such as temperature, soil pH, and UV light is critical. Therefore, the stability of the five bacteriophages at diverse temperatures and pH and under UV light exposure was determined by counting the number of viable bacteriophages after treatment. The five tested bacteriophages remained stable at 40℃ for 1 h, their viability then began to decrease at 50℃, and they were completely inactivated at 60℃ (Fig. 4A). The decrease in viability varied depending on the specific bacteriophage. All bacteriophages, except KHUφ74, were stable at pH 3-11 for 1 h but were inactivated at pH 12 (Fig. 4B). KHUφ44 was stable at pH 4-11, viability was decreased at pH 11, and the bacteriophage was inactive at pH 3 and 12. UV-A (320~400 nm) and UV-B (280~320 nm) are the solar wavelengths that reach Earth's surface [19]. Thus, the 365 nm (UV-A) and 306 nm (UV-B) wavelengths of UV light were selected for testing. All bacteriophages, except KHUφ34, retained lytic activity for at least 60 min under 365 nm UV light (320 mW/m2), but the activity of bacteriophage KHUφ34 was reduced by 20% (Fig. 4C). The lytic activity of all bacteriophages was reduced by 50% or more under 306 nm UV light for 60 min (320 mW/m2) (Fig. 4D). These results indicate that the selected bacteriophages are very stable under various temperature and pH values and to UV-A light.

Fig. 4.Stability of the five bacteriophages to various temperatures (A) and pH values (B), and exposure to 365 nm (C) or 306 nm (D) UV light. After treatment, titers of living bacteriophages (PFU/ml) were determined using plaque assays. The error bars indicate standard error.

Currently, the Korea Meteorological Administration measures the intensity of UV-B but not UV-A, and the annual average intensity of UV-B in Korea is 103.6 mW/m2 (Korea Meteorological Administration, 2014). In addition, kiwifruit trees optimally grow at pH 5.5−6.0, and the high temperature in their growing area in southern Korea is around 34℃ [31]. Based on these environmental conditions, all five selected bacteriophages are likely to be stable in the natural growth environment of kiwifruit trees.