Abstract

Bacteriophage PM4, PM5 and PM6 were isolated on different mesophilic Aeromonas strains. These bacteriophage use the flagellum as their primary bacterial receptor since purified flagella from these strains are able to inactivate these bacteriophages, independently, and the phage-resistant mutants are aflagellate and nonmotile. Furthermore, we showed that these bacteriophage may be useful to initiate the serotyping of mesophilic Aeromonas for the H-antigen (flagellum).

1 Introduction

Bacteriophage that attack motile strains, either Gram-positive or Gram-negative bacteria, are probably widespread [1] but only a few have been isolated. The best studied are the bacteriophage χ1 of Salmonella[2] and PBS1 of Bacillus subtilis[3]. We previously described a similar phage (PM3) on Aeromonas hydrophila TF7 [4]. Motile Aeromonas sp. are ubiquitous in the aquatic environment [5] and are also considered normal inhabitants of the intestinal tract of fish [6]. Mesophilic Aeromonas sp. are opportunistic, as well as primary pathogens of a variety of aquatic and terrestrial animals including humans [7]. The clinical manifestations of Aeromonas sp. infections range from gastroenteritis to soft tissue infections, septicemia, and meningitis [8, 9]. Besides that motility is an important characteristic of these group of bacteria [5], only serotyping on the basis of the O-antigen lipopolysaccharide (somatic antigen) has been described [10].

In this paper, we report the isolation and characterization of three different bacteriophage infecting several mesophilic Aeromonas sp. strains. The receptor of these phages is the flagellum, a fact that allowed us to initiate the serogrouping of mesophilic Aeromonas sp. for the H-antigen (flagellum).

2 Materials and methods

2.1 Bacteria, bacteriophages and media

A. hydrophila AH-44 and A. veronii biovar sobria AH-41 and AH-42 are strains of serogroup O:11 with an S-layer and have been previously described [11]. Other mesophilic Aeromonas sp. used (from all the O-serogroups) are from our laboratory collection or were kindly provided by T. Shimada (National Institute of Health, Tokyo). Tryptone-soya broth (TSB) was used for bacterial growth and phage propagation. TSB was supplemented with 1.5% agar (w/v) (TSA) or with 0.6% agar (TSA soft agar). For titration and inactivation assays, phage suspensions were diluted in phage buffer [12].

Spontaneous bacteriophage-resistant mutants were isolated by spreading a mixture containing ca. bacteria (108 colony forming units) and 109 phage plaque-forming units (PFU) on TSA. After 48 h at 30°C, colonies of phage-resistant mutants were picked, purified by streaking, and cross-streaked against the same bacteriophage to confirm resistance.

2.2 Bacteriophage isolation

Sewage samples were centrifuged and the supernatant was enriched at 30°C with an exponentially growing culture of A. hydrophila AH-44, or A. veronii biovar sobria AH-41 and AH-42, independently. After overnight incubation, bacteria were removed by centrifugation and filtration, and the supernatant was plated on the same strain using the double agar layer method of Adams [13]. Plaques which formed on the plates were stabbed with a needle and eluted with a small volume of phage buffer. Each phage suspension was serially propagated twice on the same strain.

2.3 General phage techniques

The methods of Adams [13] were used. The incubation temperature was 30°C, and the plates were incubated for 24 h. The bacteriophage host range was assayed by spot test.

2.4 Bacteriophage inactivation experiments

Bacteriophage (103 PFU) were incubated for 20 min at 30°C with either 107 bacterial cells, 200 μg of deoxycholate (DOC)-solubilized outer membrane (OM) [14] (see below) or 100 μg of purified flagella. Chloroform (3–4 drops) was added, mixed for 60 s, and the mixture was centrifuged at 12 000×g for 10 min at 4°C. The supernatants were assayed for phage activity directly on strains A. hydrophila AH-44, and A. veronii biovar sobria AH-41 and AH-42, independently.

2.5 Cell surface isolation and analyses

Cell envelopes were prepared by French pressure cell lysis at 16 000 Pa of whole cells followed by the removal of unbroken cells at 10 000×g for 10 min and by sedimentation of the membrane fraction at 100 000×g for 2 h. Cytoplasmic membranes were solubilized twice with sodium lauryl sarcosinate, and the OM fraction was sedimented twice at 100 000×g for 2 h. OM proteins were solubilized in 1% DOC-2 mM EDTA [14]. Lipopolysaccharide (LPS) was purified by the method of Westphal and Jann [15] as modified by Osborn [16]. Purification of flagella was performed as previously described [4]. Proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) by a modification [17] of the Laemmli procedure [18]. Protein gels were routinely stained with Coomassie blue or silver nitrate, and protein concentration was determined by the Lowry procedure with bovine serum albumin as the standard.

2.6 Electron microscopy

Techniques for visualizing stained phage, whole cells, purified flagella and phage-flagella mixtures were as previously described [4].

2.7 Motility

For the observation of motility, mesophilic Aeromonas sp. strains were grown overnight at 30°C, and the methodology used was previously described [4].

3 Results and discussion

Mesophilic Aeromonas sp. strains from serogroup O:11 showed an S-layer around the cell covering up the O-antigen LPS and the OM proteins [19]. For this reason the unique additional structures more exposed in the surface than the S-layer are the flagella. Due to this fact, bacteriophage isolated on these strains (serogroup O:11) may use the flagella as primary bacterial receptors, as on the isolation of bacteriophage PM3 on A. hydrophila TF7 (O:11). Following this approach we decided to use different strains of serogroup O:11 (S-layer+) in order to search for these bacteriophage.

3.1 Bacteriophage isolation and characterization

Bacteriophage PM4, PM5 and PM6 were isolated from sewage samples obtained from several sources on A. hydrophila AH-44, and A. veronii biovar sobria AH-41 and AH-42, respectively. Phage PM4, PM5 and PM6 gave turbid plaques, of approximately 1.5 mm, at both 30 and 37°C on their respective host strains. PM4 is able to multiply on A. hydrophila AH-44 but not on A. veronii biovar sobria AH-41 or AH-42. A similar situation is found for bacteriophage PM5 or PM6, which are able to replicate in their host strains but not in the host of the other phages. None of them is able to replicate on A. hydrophila TF7, the host strain of phage PM3. PM3 is also unable to replicate in strains AH-44, AH-41 or AH-42. All the bacteriophage (PM4, PM5 and PM6) are able to replicate in different mesophilic Aeromonas strains independently of the species or the O-serotype. We could not find any single strain able to host at the same time two of these bacteriophage, including PM3. None of these bacteriophage is able to replicate either in A. salmonicida or in Vibro anguillarum (serotypes O1 and O2), V. ordalii, V. vulnificus or V. parahaemolyticus. All of the different Enterobacteriaceae tested (more than 40 strains belonging to six different genera) were resistant to these bacteriophage (data not shown). Bacteriophage PM4, PM5 and PM6 showed similar morphologies consisting of a polyhedric head and a long noncontractile tail with a tapering end with some tail fibers that may be easily lost on phage purification (Fig. 1). According to their morphological characteristics, all these phage may be tentatively classified as Siphoviridae, probably morphotype B1, according to Ackermann [20]. All three bacteriophage incorporated [3H]TdR into their nucleic acid, which are degraded by DNases, i.e. exonuclease type III, but not by RNases. Therefore, they are DNA phages. However, all the restriction enzymes tested (PstI, SalI, HindIII, EcoRI, BamHI, BglII, KpnI and AvaI) were unable to cleave PM4, PM5 and PM6 bacteriophage DNAs. Cleavage of these DNAs with nuclease S1 (Fig. 2), as well as the lack of increase in the OD260 after heat or alkali denaturation, seems to indicate that these bacteriophage contain single-stranded DNA or at least that they have single-stranded gaps in their DNA. Because all the Siphoviridae phages described so far contain double-stranded DNA, the taxonomic location of bacteriophage PM4, PM5 and PM6 remains uncertain. Work is in progress, together with H.W. Ackermann, to classify these bacteriophage.

1

Electron micrographs of purified bacteriophages PM4 (1) and PM5 (2) negatively stained with 1% phosphotungstic acid. Bacteriophage PM5 showed a collar whereas this structure is not present in bacteriophage PM4. Bar represents 100 nm.

1

Electron micrographs of purified bacteriophages PM4 (1) and PM5 (2) negatively stained with 1% phosphotungstic acid. Bacteriophage PM5 showed a collar whereas this structure is not present in bacteriophage PM4. Bar represents 100 nm.

2

Nuclease S1 digestion [22] of PM4 bacteriophage DNA treated with the same amount of enzyme (1 unit). Lanes: 0, MW standard (λHindIII-digested DNA); 1, 3 and 5, PM4 bacteriophage DNA incubated without enzyme at 37°C for 5, 15 and 30 min, respectively; 2, 4 and 6, the same DNA treated with nuclease S1 incubated at 37°C for 5, 15 and 30 min, respectively. Similar results were obtained for nuclease S1 digestion of PM5 and PM6 bacteriophage DNAs (not shown).

2

Nuclease S1 digestion [22] of PM4 bacteriophage DNA treated with the same amount of enzyme (1 unit). Lanes: 0, MW standard (λHindIII-digested DNA); 1, 3 and 5, PM4 bacteriophage DNA incubated without enzyme at 37°C for 5, 15 and 30 min, respectively; 2, 4 and 6, the same DNA treated with nuclease S1 incubated at 37°C for 5, 15 and 30 min, respectively. Similar results were obtained for nuclease S1 digestion of PM5 and PM6 bacteriophage DNAs (not shown).

3.2 PM4, PM5 and PM6 bacteriophages surface receptor

Bacteriophage PM4, PM5 and PM6 adsorbed readily (more than 90% of phage inactivation) to purified flagella from strains AH-44, AH-41 and AH-42, respectively, whereas no phage adsorption (less than 5%) was found with OM proteins or purified LPS from the same strains. Purified flagella from A. hydrophila AH-44 are able to inactivate bacteriophage PM4, but not phage PM3, PM5 and PM6. A similar situation is observed for the purified flagella of the other host strains and bacteriophage studied. Furthermore, these bacteriophage absorbed readily to A. hydrophila AH-44, A. veronii biovar sobria AH-41 and AH-42 cells grown under motile conditions (shaking at 200 rpm at 30°C), but were unable to adsorb to the same cells grown under restrictive motile conditions (static growth at 20°C). Mutants resistant to bacteriophage PM4, PM5 or PM6, obtained independently, showed similar LPS and OM proteins as their respective wild-type strains. However, all these phage-resistant mutants were nonmotile whereas the wild-type strains were motile. Furthermore, all these phage-resistant mutants were aflagellate when observed in negatively stained preparations at the electron microscope in all the different growth conditions tested (30–37°C, static or shaking conditions). All these results prompted us to conclude that the primary bacterial receptor for these bacteriophage is the flagellum of the different strains. Furthermore, mutants lacking flagella and nonmotile are useful tools for pathogenic studies about the role of flagella and motility, as we reported previously [21].

Together with bacteriophage PM3, bacteriophage PM4, PM5 and PM6 could be very useful to initiate the serotyping of mesophilic Aeromonas strains by their H-antigen (flagellum) on the basis of the phagotype. Table 1 shows the results of a preliminary study with 201 mesophilic Aeromonas sp. strains distributed as follows: 102 A. hydrophila, 34 A. veronii sobria, 12 A. veronii veronii, 31 A. caviae, 9 A. schubertii, 7 A. trota, 4 A. jandaei and 2 A. eucrenophila. As can be observed, more than 40% of the strains are sensitive to one of these bacteriophage, with the exceptions of A. jandaei and A. eucrenophila (no strains sensitive to phage, but the number of these strains tested is very low). In some cases, like in A. veronii veronii, the percentage of sensitive strains to these bacteriophage reaches 66%. Although approximately only 50% of the most frequent mesophilic Aeromonas sp. strains are sensitive to one of these bacteriophage, the results are promising for establishing a phagotype that correlates with the flagellum type (H-antigen).

1

Percentage of sensitivity of different mesophilic Aeromonas sp. strains to bacteriophages PM3, PM4, PM5 and PM6

Phage Number of Aeromonas strains sensitive to the bacteriophagea 
 hydrophila sobria veronii caviae schubertii trota 
 n = 102 n = 34 n = 12 n = 31 n = 9 n = 7 
PM3 12 (11) 6 (17) 2 (16) 3 (9) 1 (11) 0 (0) 
PM4 16 (15) 5 (14) 3 (25) 8 (25) 1 (11) 3 (42) 
PM5  5 (4)  4 (11) 1 (8)  0 (0) 1 (11) 0 (0) 
PM6  8 (7)  2 (5) 2 (16)  4 (12) 1 (11) 1 (14) 
None 61 (59) 17 (50) 4 (33) 16 (51) 5 (55) 3 (42) 
Phage Number of Aeromonas strains sensitive to the bacteriophagea 
 hydrophila sobria veronii caviae schubertii trota 
 n = 102 n = 34 n = 12 n = 31 n = 9 n = 7 
PM3 12 (11) 6 (17) 2 (16) 3 (9) 1 (11) 0 (0) 
PM4 16 (15) 5 (14) 3 (25) 8 (25) 1 (11) 3 (42) 
PM5  5 (4)  4 (11) 1 (8)  0 (0) 1 (11) 0 (0) 
PM6  8 (7)  2 (5) 2 (16)  4 (12) 1 (11) 1 (14) 
None 61 (59) 17 (50) 4 (33) 16 (51) 5 (55) 3 (42) 

aPercentage of sensitive strains in parentheses, n is total number of strains.

None of the four A. jandaei or the two A. eucrenophila strains tested was sensitive to these phage.

1

Percentage of sensitivity of different mesophilic Aeromonas sp. strains to bacteriophages PM3, PM4, PM5 and PM6

Phage Number of Aeromonas strains sensitive to the bacteriophagea 
 hydrophila sobria veronii caviae schubertii trota 
 n = 102 n = 34 n = 12 n = 31 n = 9 n = 7 
PM3 12 (11) 6 (17) 2 (16) 3 (9) 1 (11) 0 (0) 
PM4 16 (15) 5 (14) 3 (25) 8 (25) 1 (11) 3 (42) 
PM5  5 (4)  4 (11) 1 (8)  0 (0) 1 (11) 0 (0) 
PM6  8 (7)  2 (5) 2 (16)  4 (12) 1 (11) 1 (14) 
None 61 (59) 17 (50) 4 (33) 16 (51) 5 (55) 3 (42) 
Phage Number of Aeromonas strains sensitive to the bacteriophagea 
 hydrophila sobria veronii caviae schubertii trota 
 n = 102 n = 34 n = 12 n = 31 n = 9 n = 7 
PM3 12 (11) 6 (17) 2 (16) 3 (9) 1 (11) 0 (0) 
PM4 16 (15) 5 (14) 3 (25) 8 (25) 1 (11) 3 (42) 
PM5  5 (4)  4 (11) 1 (8)  0 (0) 1 (11) 0 (0) 
PM6  8 (7)  2 (5) 2 (16)  4 (12) 1 (11) 1 (14) 
None 61 (59) 17 (50) 4 (33) 16 (51) 5 (55) 3 (42) 

aPercentage of sensitive strains in parentheses, n is total number of strains.

None of the four A. jandaei or the two A. eucrenophila strains tested was sensitive to these phage.

Acknowledgements

This work was supported by Grant PB94-0906 from DGICYT (Ministerio de Educación y Ciencia) and Grant 1995SGR00414 from the Generalitat de Catalunya. X.R. and A.A. are F.P.I. fellowships from the Ministerio de Educación y Cultura, and M.M.N. from the Generalitat de Catalunya. We also thank Maite Polo for her technical assistance and Toshio Shimada for providing the mesophilic Aeromonas strains.

References

1
Edwards
S.
Meynell
G.G.
(
1968
)
The widespread occurrence of enteric flagellar phages
.
J. Gen. Virol.
 
2
,
443
444
.
2
Iino
T.
Mitani
M.
(
1966
)
Flagellar-shape mutants in Salmonella
.
J. Gen. Microbiol.
 
44
,
27
40
.
3
Joys
T.M.
(
1965
)
Correlation between susceptibility to bacteriophage PBS1 and motility in Bacillus subtilis
.
J. Bacteriol.
 
90
,
1575
1577
.
4
Merino
S.
Camprubıć
S.
Tomás
J.M.
(
1990
)
Isolation and characterization of bacteriophage PM3 from Aeromonas hydrophila the bacterial receptor for which is the monopolar flagellum
.
FEMS Microbiol. Lett
 .
69
,
277
282
.
5
Neilson
A.H.
(
1978
)
The occurrence of Aeromonas in activated sludge: isolation of Aeromonas sobria and its possible confusion with E. coli
.
J. Appl. Bacteriol.
 
44
,
259
264
.
6
Trust
T.J.
Sparrow
R.A.H.
(
1974
)
The bactericidal flora in the alimentary tract of fresh-water salmonid fish
.
Can. J. Microbiol.
 
20
,
1219
1228
.
7
Howard
S.P.
Buckley
J.T.
(
1985
)
Phospholipids and lipopolysaccharide of Aeromonas hydrophila
.
J. Bacteriol.
 
161
,
463
465
.
8
Freij
B.J.
(
1984
)
Aeromonas: biology of the organism and diseases in children
.
Pediatr. Infect. Dis.
 
3
,
164
175
.
9
Ljungh
A.
Wadstrom
T.
(
1983
)
Toxins of Vibrio parahaemolyticus and Aeromonas hydrophila
.
J. Toxicol. Toxin Rev.
 
1
,
257
307
.
10
Sakazaki
R.
Shimada
T.
(
1984
)
O-serogrouping scheme for mesophilic Aeromonas strains
.
Jpn. J. Med. Sci. Biol.
 
37
,
247
255
11
Janda
J.M.
Oshiro
L.S.
Abbott
S.L.
Duffey
P.S.
(
1987
)
Virulence markers of mesophilic aeromonads: association of the autoagglutination phenomenon with mouse pathogenicity and the presence of a peripheral cell-associated layer
.
Infect. Immun.
 
55
,
3070
3077
.
12
Clowes
R.C.
Hayes
W.
(
1968
)
Experiments in Microbial Genetics
 .
Blackwell Scientific
,
Oxford
.
13
Adams
M.H.
(
1959
)
Bacteriophages
 .
Interscience
,
New York
.
14
Randall-Hazelbauer
L.
Schwartz
M.
(
1973
)
Isolation of the bacteriophage lambda receptor from Escherichia coli
.
J. Bacteriol.
 
116
,
1436
1446
.
15
Westphal
O.
Jann
K.
(
1965
)
Bacterial lipopolysaccharides: extraction with phenol-water and further applications of the procedure
.
Methods Carbohydr. Chem.
 
5
,
161
164
.
16
Osborn
M.J.
(
1966
).
Preparation of lipopolysaccharide from mutant strains of Salmonella
.
Methods Enzymol.
 
8
,
161
164
.
17
Ames
G.F.L.
Spudich
E.N.
Nikaido
H.
(
1974
)
Protein composition of the outer-membrane of Salmonella typhimurium: effect of lipopolysaccharide mutations
.
J. Bacteriol.
 
117
,
406
416
.
18
Laemmli
U.K.
(
1970
)
Cleavage of structural proteins during the assembly of bacteriophage T4
.
Nature
 
227
,
680
685
.
19
Merino
S.
Rubires
X.
Aguilar
A.
Albertıć
S.
Hernández-Allés
S.
Benedıć
V.J.
Tomás
J.M.
(
1996
)
Mesophilic Aeromonas sp. serogroup O:11 resistance to complement-mediated killing. Infect
.
Immun
 .
64
,
5302
5309
.
20
Ackermann
H.W.
DuBow
M.S.
Gershman
M.
Karska-Wysocki
B.
Kasatiya
S.S.
Loessner
M.J.
Mamet-Bratley
M.D.
Regué
M.
(
1997
)
Taxonomic changes in tailed phages of enterobacteria
.
Arch. Virol.
 
142
,
1381
1390
.
21
Merino
S.
Rubires
X.
Aguilar
A.
Tomás
J.M.
(
1997
)
The role of flagella and motility in the adherence and invasion to fish cell lines by Aeromonas hydrophila serogroup O:34 strains
.
FEMS Microbiol. Lett.
 
151
,
213
217
.
22
Maniatis
T.
Fritsch
E.F.
Sambrook
J.
(
1982
)
Molecular Cloning: A Laboratory Manual
 .
Cold Spring Harbor Laboratory
,
Cold Spring Harbor, NY
.