Bacteriophages of Erwinia amylovora

Abstract

Fifty bacteriophage isolates of Erwinia amylovora, the causal agent of fire blight, were collected from sites in and around the Niagara region of southern Ontario and the Royal Botanical Gardens, Hamilton, Ontario. Forty-two phages survived the isolation, purification, and storage processes. The majority of the phages in the collection were isolated from the soil surrounding trees exhibiting fire blight symptoms. Only five phages were isolated from infected aerial tissue in pear and apple orchards. To avoid any single-host selection bias, six bacterial host strains were used in the initial isolation and enrichment processes. Molecular characterization of the phages with a combination of PCR and restriction endonuclease digestions showed that six distinct phage types, described as groups 1 to 6, were recovered. Ten phage isolates were related to the previously characterized E. amylovora PEa1, with some divergence of molecular markers between phages isolated from different sites. A study of the host ranges of the phages revealed that certain types were unable to efficiently lyse some E. amylovora strains and that some isolates were able to lyse the epiphytic bacterium Pantoea agglomerans. Representatives from the six molecular groups were studied by electron microscopy to determine their morphology. The phages exhibited distinct morphologies when examined by an electron microscope. Group 1 and 2 phages were tailed and contractile, and phages belonging to groups 3 to 6 had short tails or openings with thin appendages. Based on morphotypes, the bacteriophages of E. amylovora were placed in the order Caudovirales, in the families Myoviridae and Podoviridae.

Erwinia amylovora, a member of the Enterobacteriaceae, is the causal organism of fire blight, a serious disease of the pome fruit (19-21). The disease-causing organism is currently controlled by the antibiotic streptomycin and pruning. Another control measure became apparent when bacteriophages that were able to infect E. amylovora and an avirulent yellow bacterium commonly found in the orchard ecosystem were discovered (8, 12). The yellow bacterium was subsequently identified as Pantoea agglomerans (=Erwinia herbicola). Erskine (11) recognized that E. amylovora phages may play an important role in the epidemiology of fire blight and proposed the use of phages released from the yellow saprophytic bacterium (lysogenic antagonists) as biological control agents (11).

Richie (16, 17) isolated E. amylovora phages from aerial portions of fire blight-infected trees by using as a host strain E. amylovora 110 Rif (16). The phages, named PEa1 and PEa7, belonged to two distinct groups based on chemical and physical data. Recently, E. amylovora phages were collected from orchards with fire blight symptoms and were characterized by plaque morphology, PCR, restriction fragment polymorphisms, pulse-field gel electrophoresis, and host range studies (18).

The objective of this work was to estimate the diversity of bacteriophages collected from orchards in southern Ontario that had active fire blight disease symptoms. To overcome any potential host-induced bias, the initial isolation and enrichment of the phages exploited a six-host system. The host ranges were determined for each phage isolate. Each phage isolate was examined under the electron microscope and was placed in phage families based on its morphology or morphotype as described by Ackermann and colleagues (1-6, 14). In addition, the phages were all grown on a common host, E. amylovora 110R, and further characterized by PCR using the PEa1-specific primers and restriction fragment length polymorphisms (RFLPs).

MATERIALS AND METHODS

Bacterial strains and media.

Strains of E. amylovora and P. agglomerans were cultured on nutrient agar (Difco Laboratories, Detroit, Mich.) amended with 0.25% yeast extract (Difco) and 0.5% food grade sucrose (NASYE) and incubated at 26°C. Liquid culture was carried out with nutrient broth (Difco) amended with 0.25% yeast extract and 0.5% food-grade sucrose (NBSYE) and incubated at 26°C in an orbital shaker. Strains of Pseudomonas spp. were grown on pseudomonas agar F (Difco) and incubated at 26°C. Escherichia coli was grown on Luria-Bertani medium (Difco) and incubated at 37°C. Bacteriophages were plated using the soft agar overlay method described by Adams (7). Phages were diluted and stored in sterile 0.4% nutrient broth at 4°C. Strains of bacteria used in the experiments and their origin are listed in Table 1.

TABLE 1.

Phenotypes, origin, and sources of E. amylovora isolates used in this study^a

Isolate	Phenotype	Origin	Source
E. amylovora
Ea6-4*	wt	Pear	D. Hunterb
Ea17-1-1*	wt	Pear	D. Hunter
EaD-7*	wt	Pear	D. Hunter
EaG-5*	wt	Pear	D. Hunter
Ea 29-7*	wt	Apple	A. M. Svircevb
110R*	Rifr	Apple	A. L. Jonesc
Ea 1-97	wt	Raspberry	G. Braund
Ea 4-96	wt	Raspberry	G. Braun
Ea 6-96b	wt	Raspberry	G. Braun
BC20A	wt	Crabapple	P. Sholberge
BC29	Strr	Pear	P. Sholberg
BC34A	wt	Apple	P. Sholberg
BC1280	Strr	Apple	P. Sholberg
P. agglomerans
31420			ATCCf
49018			ATCC
P. fluorescens A506			Plant Health Technologiesg
P. syringae MB-4			D. Cupplesh
E. coli DH5-α			A. Castlei

^aAsterisk denotes host strain during phage isolation and propagation. wt, wild-type isolate. Rif^r, rifampicin-resistant isolate. Str^r, streptomycin-resistant isolate.

^bAgriculture and Agri-Food Canada, Southern Crop Protection and Food Research Centre, Vineland Station, Canada.

^cDepartment of Botany and Plant Pathology, Michigan State University, East Lansing, Mich.

^dAgriculture and Agri-Food Canada, Atlantic Food and Horticulture Research Centre, Kentville, Canada.

^eAgriculture and Agri-Food Canada, Pacific Agri-Food Research Centre, Summerland, Canada.

^fATCC, American Type Culture Collection.

^gBlightBan A506; Boise, Id.

^hAgriculture and Agri-Food Canada, Southern Crop Protection and Food Research Centre, London, Canada.

ⁱDepartment of Biological Sciences, Brock University, St. Catharines, Canada.

Phage isolation.

Collections were made from mid-June to late August 1998 from sites in and around the Niagara region and Hamilton, Ontario, Canada. At each collection site, cuttings were taken from the aerial portions of trees and soil samples were taken from the bases of trees by using a stainless-steel soil corer (diameter, 2 cm; length, 35 cm) driven to a depth of 10 to 20 cm approximately 1 m from the base of the tree. All soil and aerial samples were enriched in liquid cultures in a procedure modified from that of Crosse and Hingorani (10). Flasks containing 60 ml of NBSYE were inoculated with 200-μl overnight cultures of each of the six E. amylovora propagation hosts listed in Table 1. Into each flask was placed 50 to 60 g (wet weight) of soil or 10 to 20 g (fresh weight) of aerial tissue and incubated for 18 to 20 h. The resulting slurry was agitated thoroughly with 500 μl of chloroform and centrifuged at 4°C and 8,000 × g for 20 min. The supernatant was removed with a pipette and stored at 4°C over chloroform. The supernatant was diluted and plated onto six lawns each, seeded with one of the propagation hosts listed in Table 1. Lawns were checked for the formation of plaques after 24 and 48 h. Single plaques were picked from these lawns and placed into microcentrifuge tubes containing 1 ml of NBSYE and 2% (vol/vol) chloroform. Tubes were centrifuged at 8,000 × g and stored at 4°C. Bacteriophage isolates were purified by passage through this single-plaque isolation procedure three times. Bacteriophage PEa1(h) (ATCC 29780-B1) was obtained from the American Type Culture Collection. Representative phages in the collection are available upon request.

Host range analysis.

The host ranges of all phage isolates were tested against 13 E. amylovora strains. Host ranges of a limited number of phages were also tested against five bacterial strains representing four species other than E. amylovora. Bacterial lawns were prepared by seeding 3 ml of top agar with 10⁷ CFU bacteria suspended in 10 mM sodium phosphate buffer (pH 6.8). Phage lysates were diluted to a concentration of 10⁷ PFU/ml, and 10 μl was spotted onto lawns. Plates were dried in a laminar flow hood for 10 min and incubated at 26°C for 18 to 20 h. Areas of clearing under points of phage application were scored as positive, while areas which looked no different than the surrounding untreated lawn were scored as negative. Experiments were repeated three times.

DNA extraction.

Phage DNA was extracted using a method modified from that of Manfioletti and Schneider (13). To each 10-ml volume of lysate was added DNase I (Boehringer Mannheim, Laval, Canada) to a final concentration of 20 μg/ml and RNase A (Boehringer Mannheim) to a final concentration of 100 μg/ml. After incubation at room temperature for 15 min, 0.8 ml of 0.5 M EDTA (Sigma) (pH 8) and proteinase K (Boehringer Mannheim) to a final concentration of 50 μg/ml were added, followed by incubation at 45°C for 15 min. DNA was precipitated with 0.2% (wt/vol) hexadecyl trimethyl ammonium bromide (Sigma) and 20 mM NaCl and incubated at 65°C for 3 min, followed by cooling on ice. The DNA-hexadecyl trimethyl ammonium bromide complex was pelleted by centrifugation at 4°C and 8,000 × g for 10 min. The pellet was resuspended in a minimal volume (usually 1 ml) of 1.2 M NaCl, and the DNA was precipitated with two volumes of 95% ethanol. DNA was resuspended in a minimal volume of sterile distilled water and stored at −20°C.

PCR.

Primer sequences specific for bacteriophage PEa1 were generously donated by A. L. Jones of Michigan State University. Primer sequences were 5′ AATGGGCACCGTAAGCAGT 3′ for PEa1-A and 5′ TAATGGGTATGATAGAAGGCAGAC 3′ for PEa1-B. Primers were expected to amplify a 304-bp product.

Reactions were run in 50-μl volumes using a 0.2 μM concentration (each) of primers PEa1-A and PEa1-B (Norgen Biotek, St. Catharines, Canada), 1× PCR buffer, 0.2 mM (each) deoxynucleoside triphosphates, 1.5 mM MgCl₂, and 1.5 U of Taq polymerase (MBI Fermentas). One μl of a 10⁸-PFU/ml phage suspension in phosphate buffer was used as a template. Reactions were run in a GeneAmp 9600 thermocycler (Perkin-Elmer, Norwalk, Conn.) under the following conditions: 95°C, 2 min; 95°C, 30 s, 53°C, 30 s, and 72°C, 30 s, 30 cycles.

RFLPs.

Bacteriophage DNA was digested with EcoRI, BglII, BamHI (all from Promega), or ThaI (Life Technologies, Rockville, Md.) according to the supplier's instructions and using 0.5 to 1 μg of DNA, 3 U of enzyme, and 0.1-mg/ml acetylated bovine serum albumin per 50-μl reaction mixture. Samples were digested with EcoRI, BglII, or BamHI overnight at 37°C or with ThaI for 1 to 2 h at 60°C. Fragments were separated on a 1% agarose gel in Tris-acetate-EDTA and stained in 1 μg of ethidium bromide/ml.

Transmission electron microscopy.

High-titer phage liquid cultures were prepared and centrifuged at 8,000 × g to separate phage from the host cells. The supernatant containing the phages was centrifuged at 16,000 × g for 1 h at 5°C. The phage pellet was resuspended in sterile distilled water. One drop of the suspension was placed onto a 200-by-200-mesh nickel grid coated with formvar and allowed to sit for 2 min. The phages on the grid were negatively stained with 2% uranyl acetate, 3% sodium phosphotungstate, or 1% ammonium molybdate. To each stain solution, 4 mM MgCl₂ was added in a 1:1 (vol/vol) ratio. This was allowed to sit for 3 min before the solution was wicked off and the grid was air dried. The electron microscope was calibrated with catalase crystals at the same magnification used for viewing the phages (5). Each phage isolate was enriched a minimum of three times on its original host, and each replicate was examined under the electron microscope. Specimens were viewed using a Philips CM10 transmission electron microscope with an accelerating voltage of 80 kV.

RESULTS

Phage isolation.

A total of 50 bacteriophage isolates were collected from the Botanical Gardens and southern Ontario apple and pear orchards. Eight bacteriophages isolates failed to flourish under the laboratory storage and enrichment regimes. The majority of the 42 phages were collected from soil samples adjacent to infected trees, while five phages were isolated from diseased aerial tissue (Table 2).

TABLE 2.

Phage grouping by RFLP pattern, bacterial host, origin, and morphotypes

Groupa	Isolateb	Bacterial hostc	Crop	Sourced	Plaque type (diameter, mm)	Head (Length,e nm)	Morphotypesf
1	PEa10-2	Ea6-4	Apple	Aerial	1-3	NTg	A1
	PEa10-3	EaD-7	Apple	Aerial	1-3	NT	A1
	PEa10-4	EaG-5	Apple	Aerial	1-3	57-69	A1
	PEa21-1	110R	Pear	Soil	1-3	60-71	A1
	PEa21-2	EaG-5	Pear	Soil	1-3	NT	A1
	PEa21-3	EaD-7	Pear	Soil	1-3	53-71	A1
	PEa21-4	Ea6-4	Pear	Soil	1-3	NT	A1
	PEa31-1	110R	Apple	Soil	1-3	86-110	A1
	PEa35-2	Ea17-1-1	Pear	Soil	1-3	63-72	NT
2	PEa10-6	Ea29-7	Apple	Aerial	0.5	NT	NT
	PEa31-2	EaD-7	Apple	Soil	0.5	84-114	A1
	PEa31-4	Ea29-7	Apple	Soil	0.5	NT	NT
	PEa35-7	Ea29-7	Pear	Soil	0.5	NT	NT
3A	PEa10-7	Ea29-7	Apple	Soil	4-6	46-54	C1
	PEa10-8	Ea29-7	Apple	Soil	4-6	NT	C1
	PEa10-9	110R	Apple	Soil	4-6	NT	C1
	PEa10-10	110R	Apple	Soil	4-6	NT	C1
	PEa10-11	Ea17-1-1	Apple	Soil	4-6	NT	NT
	PEa10-13	Ea6-4	Apple	Soil	4-6	NT	NT
	PEa10-14	Ea110	Apple	Soil	4-6	NT	NT
	PEa10-15	110R	Apple	Soil	4-6	NT	NT
	PEa31-3	Ea29-7	Apple	Soil	4-6	NT	C1
3B	PEa46-2	EaD-7	Crabapple	Aerial	4-6	NT	C1
3C	PEa1	110R	Apple	Aerial	4-6	53-63	C1
4	PEa10-12	EaG-5	Apple	Soil	0.5	60-71	C1
	PEa10-16	Ea17-1-1	Apple	Soil	0.5	NT	C1
5	PEa9-2	Ea17-1-1	Pear	Soil	1-3	55-65	C1
	PEa9-4	EaG-5	Pear	Soil	1-3	NT	NT
	PEa9-5	Ea6-4	Pear	Soil	1-3	NT	C1
6	PEa51-1	Ea17-1-1	RBGh	Soil	1-2	NT	NT
	PEa51-2	110R	RBG	Soil	1-2	66-77	C1
	PEa51-4	Ea29-7	RBG	Soil	1-2	NT	C1
	PEa51-6	Ea6-4	RBG	Soil	1-2	NT	NT
	PEa51-7	Ea29-7	RBG	Soil	1-2	NT	NT
	PEa51-8	Ea17-1-1	RBG	Soil	1-2	NT	NT

^aGroup designation based on PCR and RFLP analyses. PEa9-3, PEa10-1, PEa35-3, PEa35-4, PEa35-5, PEa35-6, PEa45-1, and PEa45-3 were left ungrouped. The first six phages listed could not be digested with the four restriction enzymes used in this study, and the remaining two phages were not tested by RFLP.

^bPhage isolates were collected from seven different sites in southern Ontario, Canada. Designation for each phage is PEa, followed by site number-isolate number.

^cBacterial host E. amylovora isolate designations are described in Table 1.

^dSource of phages in the orchard.

^eLength measurement represents a range of values obtained by measuring a minimum of five phages/isolate.

^fPhage morphology for phage families as described by Ackermann, Fig. 1 (3). Tailed phages belonging to morphotype A1 to A3 are Myoviridae with contractile tails; morphotype B1 to B3 are Siphoviridae with noncontractile tails; C1 to C3 are Podoviridae with short tails.

^gNT, not tested.

^hRBG, Royal Botanical Gardens, Hamilton, Canada. Collected from trees, Sorbus spp., demonstrating fire blight symptoms.

Plaque morphology.

The bacteriophage isolates in the collection produced one of four general plaque sizes on lawns of host E. amylovora (Table 2). Those phages, identified as related to phage PEa1 by PCR and RFLP analysis (groups 3A, 3B, and 3C), produced distinctive 4- to 6-mm plaques with a distinct translucent halo which continued to expand after the plaque itself stopped growing. The remaining groups of phages produced plaques of variable diameters ranging from 0.5 to 3 mm (Table 2).

Host range.

The phages in groups 1, 2, 4, and 5 were all able to form plaques on 11 or more of the 13 E. amylovora stains tested (Table 3). Phages in Groups 3A, 5, and 6 exhibited narrower host range patterns (Table 3). The phages in these groups could not infect the British Columbia Erwinia isolates and some of the Ontario isolates. Phages belonging to groups 3C (phage PEa1) and 3B showed little or no visible lytic activity against E. amylovora strains Ea29-7, BC29, Ea34A, BC34A, and BC1280 or against bacterial strains EaG-5 and Ea6-4, isolated from Harrow, Ontario (data not shown). Host resistance was evident in the bacterial strains EaG-5, Ea6-4, and Ea4-96 of E. amylovora. The phages in groups 1, 4, and 6 were unable to lyse these host bacterial isolates.

TABLE 3.

Phage host range according to RFLP groups and Erwinia host

Erwinia host	No. of phage isolates in RFLP grouph
	Group 1			Group 2			Group 3A			Group 4			Group 5			Group 6			Ungrouped
	+e	(+)f	−g	+	(+)	−	+	(+)	−	+	(+)	−	+	(+)	−	+	(+)	−	+	(+)	−
EaG5a	6	1	3	2	0	2	2	7	0	2	0	0	3	0	0	2	4	0	7	0	1
EaD7	9	0	0	4	0	0	9	0	0	2	0	0	3	0	0	5	1	0	8	0	0
Ea6-4	9	0	0	4	0	0	2	4	3	2	0	0	3	0	0	2	0	4	8	0	0
Ea17-1-1	9	0	0	4	0	0	9	0	0	2	0	0	3	0	0	6	0	0	8	0	0
Ea29-7	4	4	0	4	0	0	9	0	0	1	1	0	3	0	0	5	0	1	8	0	0
110Rb	9	0	0	4	0	0	9	0	0	2	0	0	3	0	0	6	0	0	8	0	0
BC20Ac	8	0	0	4	0	0	9	0	0	2	0	0	3	0	0	6	0	0	8	0	0
BC29	9	0	0	4	0	0	2	0	7	2	0	0	2	0	1	2	0	4	8	0	0
BC34A	4	3	0	4	0	0	2	0	7	1	1	0	0	2	1	0	2	4	5	2	0
BC1280	4	4	0	4	0	0	2	1	6	0	2	0	0	1	2	0	3	3	4	2	2
Ea4-96d	5	1	1	4	0	0	2	2	0	0	0	1	0	2	0	0	0	6	8	0	0
Ea6-96b	8	0	0	4	0	0	9	0	0	2	0	0	3	0	0	6	0	0	8	0	0
Ea1-97	9	0	0	4	0	0	9	0	0	2	0	0	3	0	0	6	0	0	8	0	0

^aBacteria isolated from pear and apple orchards in southern Ontario.

^bIsolate obtained from Michigan.

^cBacteria isolated from apple orchards in British Columbia.

^dBacteria isolated from raspberry plantings in Nova Scotia.

^e+, Visible plaque formation

^f(+), weak plaque formation.

^g−, No plaque formation.

^hNumbers in the columns represent the number of phage isolates that produced a plaque, weak plaque, and/or no plaque on the designated Erwinia host.

A limited number of isolates were selected for evaluation of their ability to form plaques on lawns composed of bacterial species other than E. amylovora (data not shown). Of the 14 phage isolates used in these experiments, none was able to lyse lawns composed of E. coli DH5α, Pseudomonas syringae MB-4, or Pseudomonas fluorescens A506. Two phage isolates from group 1 were able to produce plaques on both P. agglomerans strains evaluated. Of the six members of group 6, five were tested and all were able to lyse P. agglomerans 49018.

PCR.

Using the primers specific for bacteriophage PEa1, phages PEa10-7, PEa10-8, PEa10-9, PEa10-10, PEa10-11, PEa10-13, PEa10-14, PEa10-15, PEa31-3, and PEa46-2 produced a ca. 300-bp PCR product, indicating relatedness to phage PEa1 (data not shown). Phage PEa1(h) obtained from the American Type Culture Collection also produced this fragment.

RFLPs.

DNA was extracted from 40 bacteriophage isolates collected from the field and phage PEa1(h). The arrangement of phage isolates into groups based on RFLP data is shown in Table 2. All of the phages which were identified as related to PEa1 using PCR produced similar restriction patterns; these phages were placed into Group 3. Isolates PEa10-7, PEa10-8, PEa10-9, PEa10-10, PEa10-11, PEa10-13, PEa10-14, PEa10-15, and PEa31 produced identical restriction patterns and were designated group 3A. Phage PEa46-2, which produced two more fragments than the phages in group 3A when digested with BglII, was undigested by EcoRI and produced two different fragments when digested with ThaI. Based on this information, PEa46-2 was placed into group 3B. Phage PEa1(h) was also undigested by EcoRI and produced two fragments which were of differing lengths from those found in group 3A when digested with BglII and one extra fragment when digested with ThaI. Phage PEa1 (h) was placed into group 3C.

Transmission electron microscopy.

E. amylovora bacteriophages were placed into morphotype groups according to the method of Ackermann (3). The phage morphology within each RFLP grouping is described in Table 2. The phages placed into RFLP group 1 consisted of an icosahedral head (variable lengths from 53 to 110 nm), contractile tail, and A1 morphotype. Two distinct phage shapes and sizes were evident in this group: the smaller phages, as seen in (Fig. 1A and C), or the larger phages (Fig. 1B and D). Phages PEa31-2 and -4 belonged to RFLP group 2 and morphotype A1. The contracted stage was never observed for these two phage isolates. The morphology of the distinct rigid tail places the phages in the Myoviridae family (Hans Ackermann, personal communication). All the phages placed in RFLP groups 3A, 3B, 3C, 4, 5, and 6 belonged to morphotype C1 (Fig. 1E).

FIG. 1.

Electron micrographs of representative phages belonging to restriction groups 1 and 3. (A to D) Erwinia bacteriophages belonging to the Myoviridae. Small phage in uncontracted state (A), larger phage in uncontracted state (B) (note the decorations at the bottom of the tail [arrow]), small phage in the contracted state (C), and the larger phage in the contracted state (D) are shown. Panel E shows group 3A phages belonging to the Podoviridae. The arrow points to the tail region. Micron marker, 100 nm.

DISCUSSION

Of the 42 bacteriophages collected from the pear and apple orchards and the Royal Botanical Gardens, 37 phages were isolated from the soil surrounding the infected trees. Five phages were recovered from aerial portions of infected branches that exhibited fire blight symptoms. This finding is in agreement with the work of Erskine (11), who isolated phages from the soil surrounding infected trees and not from aerial samples. Phages have been shown to be stabilized in soil environments due to adsorption to charged colloids, such as clays (15, 22). Additionally, soil provides protection against desiccation and UV light, two factors which will inactivate a wide variety of phages, including those of E. amylovora (7, 11, 12, 16). In this study, E. amylovora phages could be recovered from soil samples after 12 months' storage at 4°C (data not shown). It is unlikely that populations of E. amylovora phages are able to reside permanently in aerial tissues, since Ritchie (16, 17) was unable to isolate E. amylovora phages from aerial tissue in the winter months.

The independent methods used to characterize the phages, molecular identification, plaque morphology, and electron microscopy, proved to be remarkably congruent in organizing phages in this collection. Phages were placed into six groups based on restriction patterns, with one group, group 3, subdivided into three subgroups. Transmission electron microscopy of phages in each of the restriction groupings revealed that the morphology was consistent within the same group. Based on the system devised by Bradley (9) and refined by Ackermann and DuBow (1, 5), the morphology will place these phages into the virus order Caudovirales. In our study all phages belonging to groups 1 and 2 had contractile tails and belonged to the family Myoviridae (Fig. 1A to E). Within the Myoviridae two distinct morphologies were evident, a smaller contractile phage and a larger contractile phage. Both morphotypes, however, are consistent with the A1 type description of Ackerman and Dubow (1, 5), and both produced the same plaque morphology. The phages seen in Fig. 1A and D are similar in appearance to Pseudomonas spp. phages, while those in Fig. 1B and D have not been previously observed in the Enterobacteriaceae (H.-W. Ackermann, personal communication). The phages belonging to groups 3A, 3B, 3C, 4, 5, and 6 were short-tailed phages in the family Podoviridae. In this group the morphology varied from short tails to barely discernible openings with or without decorations. Group 3 phages were all identifiable as PEa1 type both by the PCR protocol and by their production of distinctive, large plaques with expanding haloes. The majority of these phages, placed into group 3A, produced identical restriction patterns which differed slightly from the type phage, PEa1, and from phage PEa46-2. Group 6 phages had a unique restriction pattern and plaque morphology but shared phage morphological characteristics with groups 3, 4, and 5. Based upon these observations, it should be possible to categorize with a fair degree of accuracy any new E. amylovora bacteriophage isolates.

The one characteristic which did not show any consistent correlation with any of the above phenotypes was host range. This characteristic could not be used to place individual phage into any a specific RFLP group. This lack of association of host range with any other characteristic could very likely be the reason for the great diversity of phages isolated in this study. Historically, studies involved in the isolation of E. amylovora bacteriophages have consistently used a single host system (16, 17, 18). Schnabel and Jones (18) isolated 48 bacteriophages on a single host, E. amylovora strain 110R. They recovered five distinct phage types based on differences in genome size and restriction patterns; however, the diversity of phages was not extensive in that 41 of the 48 isolates were of the PEa1 type, 4 of the remaining 7 were PEa7-like, and 3 were novel (18). In the present study, six bacterial hosts, including strain 110R, were used to isolate bacteriophages. The use of the six hosts had a profound influence on the variety of bacteriophages recovered. Analysis of host ranges revealed that phages were able to infect a wide range of host bacteria. We noted that phages in groups 3, 5, and 6, with few exceptions, were unable to infect the E. amylovora bacteria from British Columbia. Therefore, the diversity of our collection may in part be attributed to the use of the six-host system in the initial isolation and purification protocol. The bacterial host, E. amylovora 110R, was used both in our study and by Schnabel and Jones (18). This particular host bacterium was lysed by all the phages in our collection.

Individual phage isolates grouped together on restriction patterns were isolated from the same site. The eight phages in group 1, for instance, were collected from three sites; the six inhabitants of group 6 were all isolated from the same location. In these cases, it is not unreasonable to make the assumption that phage isolates exhibiting identical restriction patterns and isolated from the same sample are the same phages, isolated multiple times on different bacterial host strains. Phages with identical restriction patterns can be said to be highly related to each other regardless of site of isolation. The groups based on molecular markers will aid in streamlining future research. Based on the above observation, the authors would suggest that in fact phage populations are quite homogeneous in an orchard. In our collection only the phages designated with the location/phage number PEa10 were placed into four out of the six possible restriction groups.

In this study we examined a collection of phages based on restriction endonuclease patterns, phage particle morphology, plaque morphology, and host range. Future research will explore the potential of using phages as biological control agents under field conditions.