Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
Basic characterization of a lipid-containing bacteriophage specific for plasmids of the P, N, and W compatibility groups1 DAVIDE. BRADLEY~ A N D ELIZABETH L. RUTHEWORD Department o f z o o l o g y , University of Edinburgh, West Mains Road, Edinburgh EH93JT, Scotland Accepted October 1 1, 1974 BRADLEY,D. E., and E. L. RUTHERFORD.1975. Basic characterization of a lipid-containing bacteriophage specific for plasmids of the P, N , and W compatibility groups. Can. J. Microbiol. 21: 152-163. Preliminary studies have shown that bacteriophages PR3 and PR4, originally isolated on Pselrdomonas aeruginosa, resemble the lipid-containing phage PM2 in appearance. Their host range extends intergenerically to species carrying drug-resistance plasmids of the P and N compatibility groups. In this paper, the serological identity of the two isolates is established and it is concluded that they are the same virus, but with some differences in growth characteristics. They contain double-stranded DNA and are probably icosahedra (65 nm) with short (47 nm) noncontractile tails. Their sensitivity to chloroform and low buoyant density in CsCl(1.265 g/ml) indicate that they contain lipid which is probably located in the thickened inner layer of the capsid. A study is made of their adsorption efficiencies to sensitive and resistant bacteria, and it is found that, unlike most sex-specific phages, they adsorb directly to the cell surface and not to sex pili. Their host range is shown to include strains harboring a drug-resistance plasmid of the W compatibility group. BRADLEY.D. E.. et E. L . RUTHERFORD. 1975. Basic characterization of a lioid-containing bacteriophage specific for plasmids of the P, N, and W compatibility groups. Can. J. ~ i robiol. 21: 152-163. Des etudes preliminaires ont montre que les bacteriophages PR3 et PR4, isoles a I'origine chez Pseudomonas aeruginosa, ont I'apparence du phage PM2 qui contient des lipides. Leur gamme d'h8tes possibles s'etend intergeneriquement a d e s especes porteuses de plasmides de rksistance aux antibiotiques, plasmides qui appartiennent aux groupes P et N de compatibilite. Tel que present6 dans ce papier, les deux isolats sont identiques d'apres la serologic. 11 s'agit donc du mOme virus, mais avec certaines variantes dans les caracteres d e croissance. Ils contiennent de I'ADN bicatenaire e t il s'agit probablement d'isocahedres (65 nm) a queues courtes (47 nm) non contractiles. A cause de leur sensibilite au chloroforme et de leur faible densite de flotaison en CsCl(1.265 glml), ces phages doivent contenir un lipide probablement situe dans I'epaississement de la couche interne de la capside. Apres etude d e leur efficacite d'adsorption sur des bacttries sensibles et rtsistantes, on trouve que, contrairement 5. laplupart des phages spkcifiques de sexe, ils s'adsorbent directement a la surface de la cellule e t non aux pili sexuels. Leur gamme d'h8tes comprend des souches porteuses d'une resistance aux antibiotiques plasmidique du groupe de compatibilite W. [Traduit par le journal]
Introduction Several bacteriophages have been isolated on strains of Pseudomonas aeruginosa carrying P group drug-resistance plasmids (9) which can transfer intergenerically. They are the ribonucleic acid (RNA) phage PRRl of Olsen and Thomas (17), the tailed phages PR3 and PR4, and the filamentous phage Pf3 isolated by Stanisich (1 8). Of these, PRRl and Pf3 appear to be specific for the P group plasmids, but PR3 and PR4 are also able to plaque on strains carrying plasmids of the N compatibility group (6) which can also transfer intergenerically (9). Khatoon et al. (1 5 ) have described a filamentous phage, designated IKe, 'Received June 26, 1974. ZPresent address: Faculty of Medicine, Memorial University of Newfoundland, St. John's, Newfoundland, Canada A I C 5S7.
c
which is specific for N group plasmids. The present paper describes the basic characteristics of isolates PR3 and PR4. A preliminary study (6) has shown that PR3 and PR4 appear identical in the electron microscope and have a thickened capsid similar to that of the lipid-containing phage PM2 of Espejo and Canelo (I 1). Unlike PM2, however, PR3 and PR4 are thought to have a short tail. Both isolates, and also Pf3, are believed to adsorb directly to the cell wall of sensitive bacteria, while other "plasmid phages" adsorb to pili. For example, the RNA phage PRRl attaches to pili determined by the plasmid RPl (6). PR3 and PR4 are thus unusual because of thir possible lipid content, host range, and site of adsorption. The present work is concerned with these aspects and other basic characteristics. Pre-
BRADLEY AND RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE
TABLE 1 Details of bacteria used
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
Strain
Plasmid compatibility group
P . aeruginosa PA02605 R P l K/lP04-RPl 18s RPI 18s PU21 S-a PU21 E. coli
Source B. W. Holloway K / l P 0 4 - *X18S R P l T. L. Pitt T. L. Pitt G. A. Jacoby G. A. Jacoby R. H. Olsen Rajul V. Iyer Rajul V. Iyer R. Thompson G. A. Jacoby
*A mutant of strain K with no "normal" polar pili ( S ) , but with R P I pili after mating \ ,
(b).
?This strain carries the plasmid R1822 (16) which has been reidentified a s R P I (Olsen, personal communication).
liminary experiments on their adsorption to bacteria carrying various R factors are also described. The other phages mentioned above are included for comparison.
Materials and Methods Bacteria and Bacteriophages Bacterial strains are described in Table I . The bacteriophages were supplied as follows: P R R l RNA phage (17) by Dr. R. H. Olsen, University of Michigan, U.S.A.; filamentous phageIKe(l5) by Dr. RajulV. Iyer, University of Ottawa, Ottawa, Canada; filamentous phage Pf3 and the phages PR3 and PR4 (18) by Dr. V. Stanisich, Monash University, Australia; and tailed P. aeruginosa pilus phage C5 (7) by Mr. T . L. Pitt, Central Public Health Laboratory, London, England. Cult~rreMedia and Methods All media were made with Oxoid nutrient broth, 2% (w/v) agar being added for plates. Double-agar layer plates contained a final concentration of 0.5% (w/v) agar in the soft top layer. Bacteria were grown at 37' in shake culture or on plates. Preparation of Bacteriophage Suspensions High-titer phage stocks were prepared either by the confluent lysis of bacterial growth (strain PA02605 or K/IP04- R P I ) on large (15-cm diam) double-agar layer plates followed by extraction with broth for 2 h at room temperature, or by the lysis of exponential phase broth cultures of host bacteria. When required, concentration and purification were by alternate high- and low-speed centrifugation (60 000 x g for 2+ h and 7000 x g for 20 min at 4") with the final suspension in broth. Phage susoensions were titrated by standard methods (1). The Spot Test for Bacteriophage Lytic Activity A double-agar layer plate of the bacterium under test was prepared, and a loopful of phage suspension was placed on the surface. After overnight incubation, clearing indicated lytic activity. Anomalous results due t o bac-
teriocin activity were avoided by growing phage stocks on bacterial strains which did not produce bacteriocins active on the strain under test. Serology Anti-PR3 and -PR4 sera were prepared in rabbits. Injections containing about 2 x 109plaque-formingunits (PFU) each were given intravenously on days 0, l , 2 , 5 , 7 , 14, and 18, with bleeding on day 22. Complement was inactivated at 56' for 40 min. The rate of inactivation (K) of a phage by an antiserum was determined by the addition of 0.1 ml of bacteriophage (about 2 x lo7 PFU/ml) to 10 ml of broth containing 0.1 ml of antiserum. The mixture was incubated in a shaker at 37' and 0.1-ml samples were diluted into 10 ml of broth at 30 and 45 min for titration of survivors. K values were calculated for a given phage plus antiserum combination at each time according t o the method of Adams (I), and the average of the two determinations was taken as the final value. N o inactivation of either PR3 or PR4 was found with serum from rabbits which had not been inoculated. Acridine Orange Staining The fluorescent staining method for determining the type and strandedness of bacteriophage nucleic acids was carried out as described in detail elsewhere (3). A PR3 suspension, pretreated with ribonuclease (RNase) (Sigma RNase A, type IA) and deoxyribonuclease (DNase) (Sigma DNase I), in phosphate-buffered saline was dried down as small droplets on glass slides. After fixation (Carnoy's fluid) and staining in 0.01% acridine orange in citric acid and phosphate buffer (pH 3.8) for 5 min, the slides were soaked in 0.15 M Na,HPO, for 15 min and viewed under ultraviolet light (257 nm). The colors obtained were noted together with changes produced by immersion in molybdic or tartaric acids (3). The sensitivitv of the dried droplets to DNase (20 pgjrnl) and R N ~ S ; (100 pg/ml) was also tested (3). Electron Microscopy Bacteriophage particles were mounted for electron microscopy by floating a carbon-coated specimen support
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
154
CAN. J. MICROBII3L. VOL. 21, 1975
grid on the surface of a suspension in broth (prepared as described above) at 37" for as long as possible (1-24 h). Grids were then washed, and negatively stained with neutral 1% sodium phosphotungstate. I n a n alternative method (2) using a different host strain (Escherichia coli 55-3 R199), a loopful of soft agar was removed from the center of a spot test of phage PR4. This was extracted with a small amount of 0.1 M ammonium acetate solution which was mixed with 2% sodium phosphotungstate and dried onto carbon-coated grids. Host bacteria with adsorbed virions were prepared by mixing phage suspension and host cells (log phase) in broth at as high a multiplicity of infection (moi) as conveniently possible (about 20 PFU/bacterium). After gentle shaking of the mixture for 20 min at 37", bacteria were mounted for electron microscopy by centrifuging them directly onto carbon-coated support grids, and negatively staining them. Buoyant Density Determination Bacteriophage suspensions were concentrated at 60 000 x g for 24 h in the ultracentrifuge. Pellets were resuspended in 0.3 ml of neutral 0.1 M ammonium acetate solution and centrifuged again at low speed t o remove debris. A volume of 0.25 ml of phage suspension was added to 4.75 ml of tris(hydroxymethyl)aminomethane (Tris) buffer (pH 8.2) containing 2 g of CsCl in a tube for a Spinco Rotor SW 50.1. After the tube was filled with mineral oil, the samples were centrifuged at 35 000 rpm for 49 h at 13". Six-drop fractions were collected by puncturing the bottoms of the tubes. A volume of 0.05 ml of each fraction was diluted into 0.95 ml of chilled broth. The spot test (sensitivity about 1 x lo3 PFU/ml) was used to determine which fractions contained infectious phage, and these were subsequently titrated. Refractometer measurements were carried out on a series of fractions from each tube to determine their densities according t o the relationship given by Ifft et al. (13). Adsorption of Bacteriophages A sample of exponential phase bacteria from a shake culture was adjusted with broth to about 2 x lo9 cells/ml (estimated by opacity) with broth. Two milliliters was mixed with 2.0 ml of a phage suspension in broth (titer about 2 x lo6 PFUjml) in a conical flask. The mixture was incubated at 37' for 20min with gentle shaking. Bacteria were then removed by two cycles of centrifugation at 7000 x g for 10 min, and the supernatant was titrated for surviving phage. A blank sample, in which the bacterial suspension was replaced with broth, was processed simultaneously. Treatment of Phage PR3 with Organic Solvents The method used was based on that of Espejo and Canelo (11). Aliquots of PR3 suspension in broth were mixed with various proportions of chloroform or ether in sealed medical flat bottles. These were fixed, slightly off the horizontal, to a platform shaking at about two oscillations/s at room temperature. After 4 h, the aqueous phase was separated by centrifugation at 1000 x g for 15 min and titrated for surviving phage. A blank, in which the solvent was replaced with broth, was treated simultaneously. The electron microscopy of PR3 suspensions treated with chloroform was carried out as follows. Equal
volumes of phage suspension and chloroform were shaken together for 4 h at two oscillations/s. After the phases were separated, the nutrient broth suspension of PR3 was centrifuged at 7000 x g for 10 min to remove debris and precipitates. Phage particles were mounted for electron microscopy by floating a carbon-coated grid o n the suspension as described above. A control sample without chloroform was treated simultaneously.
Results Inactivation of Bacteriophages by Antiserum The use of the inactivation or velocity constant K (for a given batch of antiserum with a single phage isolate) provides the best means of ascertaining the serological relatedness of two bacteriophages. K for the heterologous phage is compared with K for the homologous phage, most conveniently by their ratio. This is usually defined as K' and is equal to 1 for the homoor heterologous logous phage, a f r a ~ t i o ~ frelated phages, and 0 for unrelated phages. The comparison for PR3 and PR4 is shown in Table 2 together with the velocity constants for the two batches of antiserum. The two isolates are clearly serologically identical, and are in effect the same virus. Nucleic Acid Type and Strandedness The results of the acridine orange fluorescent staining tests are shown in Table 3, nucleic acid strandedness being indicated by an appropriate prefix (1 = single-stranded, 2 = double-stranded). Phage PR3 is compared with the known 2-DNA coliphage T5. It is clear from the colors obtained, and the sensitivity of PR3 nucleic acid to DNase, that it too contains 2-DNA. Morphology Isolates PR3 and PR4 were identical in appearance and dimensions, as might be expected. Negatively stained preparations gave virions with a varied appearance shown typically in Fig. 1. Both full and empty heads can be sezn, some (arrowed) having a thickened inner layer. TABLE 2 Relative velocity constants (K') for bacteriophages PR3 and PR4 with their antisera Values of K ' for phages Serum (K for homologous phage)
PR3
PR4
BRADLEY AND RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE
155
TABLE 3 Colors obtained with the acridine orange staining of bacteriophage PR3
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
Color obtained after treatment Specimen
Nucleic acid
PR3 T5 Standard Standard Standard Standard
2-DNA 2-DNA 2-RNA I-DNA 1-RNA
?
NaZHPO4
Moly bdic acid
Bright yellowish green Bright green Green Green Red Red
Bright green Bright green Green Green fading Paler green Paler red
'Resistance to a nuclease, -; sensitivity,
Tartaric acid
RNase*
DNase*
-
+ + ++
Orange Reddish orange Orange Red Paler green Paler red
+
-
+.
Tails were only rarely present and often difficult to find, their frequency varying greatly from one specimen grid to another. An example of a PR4 particle with a partly disrupted head, and a tail, is shown in Fig. 2; the distorted thick inner layer is clearly visible. In Fig. 3, there is a slight discontinuity (arrowed) in the capsid of the top particle, suggesting that the tail has fallen off. These small holes were quite frequent. The 9.0-nm-thick inner layer is clearly visible in the lower particle. Both virions have lost their nucleic acid. Figure 5 also shows empty virions, but with tails, the appearance of which suggests that they are disintegrating. It was extremely difficult to find completely intact virions (Fig. 4). The particle illustrated closely resembles coliphage T3 in appearance, the thick inner layer being obscured by nucleic acid. The dimensions for PR3 and PR4 are head size 65 nm and tail length about 47 nm, calibrated with Dow polystyrene latex (264 nm). The tail length is much shorter than previously observed (6). Some tails, as in Fig. 2, were much longer than others and were thought to be in the process of degradation before falling off, despite the fact that the phage suspension was freshly prepared for examination, usually on the same or previous day. The length given refers to those on more or less intact virions. No contractile apparatus could be found on any tails. No phages could be found exhibiting the typical profile associated with an octahedron (4), the more or less regular hexagon suggesting a probable icosahedral form.
Buoyant Density centrifugation profiles of PR3 and PR4 in cesium chloride are shown in Fig. 6. The buoyant densities were calculated as 1.267 g/ml for PR3 and 1.264 g/ml for PR4. A single sharp infectious peak was obtained in each case, and the phage
was visible as an opaque band in the gradient before fractionation.
Host Range of Bacteriophages The lytic activity of PR3 and PR4 on strains of P. aeruginosa and E. coli carrying various plasmids is shown in Table 4. Phages PRRI, Pf3, and IKe are included for comparison. All test strains carrying plasmids of the P, N, or W compatibility groups were lysed by PR3 and PR4, while background strains without plasmids were resistant. The host range of PRRI and Pf3, however, is restricted to RPl, and IKe can only lyse strains carrying RM98 and R199 (N group). A mutant of K/lP04- RPI resistant to PRRI was isolated by streaking soft agar from a spot test of phage PRRI. This proved to be sensitive to phages Pf3, PR3, and PR4 as indicated by spot tests. However, it was unstable and tended to revert to PRR1-sensitivity on subculturing several times. Growth of Infected Bacteria The growth curves for cultures of P. aeruginosa PA02605 infected with PR3 and PR4 at about the same moi are compared in Fig. 7. It can be seen that the onset of lysis occurs after about 130 min for PR3 and 95 min for PR4, a considerable difference which will be discussed below. A curve for strain 18s RPI infected with PR3 at a slightly higher moi shows a relatively small difference in the time to lysis. A knowledge of the time to lysis is required to ensure that adsorption tests are completed before the release of phage progeny. Comparison of Plaque Morphology and Eficiency of Plating for Phages PR3 and PR4 Aliquots of PR3 and PR4 containing about 200 PFU (titrated on strain PA02605) were plated by the double-agar layer method with
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
156
CAN. J. MICROBIOL. VOL. 21, 1975
FIG. 1. Typical appearance of PR3 phage particles. In two of them (arrowed), a thick inner layer is visible. x 115 000. FIG. 2. A broken PR4 particle with a distorted inner layer and a tail. x 300 000. FIG. 3. PR4 particles extracted from a spot test "plaque" on E. coli 35-3 R199, having an appearance identical with phages grown on strains of P. aerrtginosa. A small hole is present in the capsid of the top particle (arrowed). x 300 000. FIG. 4. An intact PR3 particle. The nucleic acid obscures the inner layer of the capsid, and the tail is shorter than that shown in Fig. 2. x 300 000. FIG. 5. PR3 particles with tails in the process of disintegration. Some nucleic acid appears to remain in the heads partly obscuring the inner layer, though one particle (arrowed) has a broken inner layer. x 225 000.
BRADLEY A N D RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE 10 2.9X10
-
10 2.5X10
-
I
l
l
1 I
I
I
I
PR3
I
I
I
I
I
,
-
I
I
I
-
PR4
1.33
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
- 1.31 - 1.29
1
$
1.7~10'~-
-
\
3 LL
a
0
1.27 1
H
i.3xrd01.25
5
I x109
FRACTION NUMBER
FIG.6. Centrifugation profiles of PR3 and PR4 in 40% CCsCI. Buoyant densities were PR3, 1.267 g/rnl, and PR4, 1.264 g/ml. TABLE 4 Lytic activity of bacteriophages Lytic activity of phages* Background strain
P. aert/ginosa PA02605 K/lP0418s 18s PU21 PU21 E. coli CR34 MA50 MA50 J5-3 J53-1
Plasmid
Compatibility RrOUP
PRRl
Pf3
PR3
PR4
P P P
+ + +-
+ + + -
W
-
-
+ + + +-
+ + ++-
+ + -
-
(+) (+)
(+)
RPl RPl RP1 None S-a None RPl RM98 None R199 S-a
*+ indicates clearing, (+) partlal clearing,
-
P N -
N W
-
+ -
-
-
+
-
-
+ +
+
IKe
-
-
-
-
+ +-
-
- no clearing.
various strains of plasmid-carrying bacteria. The plaques obtained were counted and the efficiency of plating (eop) was calculated as a fraction of the number obtained with P. aeruginosa PA02605 (RPI) which was the original isolating host (18). At the same time the morphology of the plaques was noted. The results in Table 5 show that the isolating strain provides the largest number of plaques as expected, and that there is a substantial reduction in eop for all strains of E. coli, probably resulting from restriction. The plaque morphology varies somewhat according to the
host. Plaques of PR3 are generally smaller and more turbid than those of PR4. Adsorption of Bacteriophages The results shown in Table 6 were obtained by titrating surviving phage after incubating mixtures of phage and bacteria (see Materials and Methods). The percentage of phage adsorbed was calculated by subtracting this titer from that of a blank (representing 0% adsorbed). Results obtained within 2% of the blank are given as < 2% phages adsorbed. The values obtained are
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
MINUTES
FIG. 7. Growth curves for bacterial cultures infected with phages PR3 and PR4. A culture of strain PA02605 or 18s RPI (about 5 ml) was grown t o early log phase, and phage suspension (0.2-2.0 ml) was added at 0 min. The cell concentration was taken at intervals (estimated by opacity). The multiplicities of infection were 2.5 PFU/bacterium for PR4 + PA02605, 2.0 PFU/bacterium for PR3 PA02605, and 3.8 PFU/bacterium for PR3 + 18s R P l . The times at which lysis began are indicated by arrows.
+
TABLE 5 Efficiency of plating (eop) and plaque morphology of phages PR3 and P R 4 PR3 Strain P. aeruginosa PA02605 K/IP0418s PU21 E. coli CR34 MA50 353-1
Plasmid
Compatibility group
PR4 -
eoP
RPI RPI RPl S-a
P P P W
1 0.87 0.51 0.36
RPl RM98 S-a
P N W
Basic characterization of a lipid-containing bacteriophage specific for plasmids of the P, N, and W compatibility groups1 DAVIDE. BRADLEY~ A N D ELIZABETH L. RUTHEWORD Department o f z o o l o g y , University of Edinburgh, West Mains Road, Edinburgh EH93JT, Scotland Accepted October 1 1, 1974 BRADLEY,D. E., and E. L. RUTHERFORD.1975. Basic characterization of a lipid-containing bacteriophage specific for plasmids of the P, N , and W compatibility groups. Can. J. Microbiol. 21: 152-163. Preliminary studies have shown that bacteriophages PR3 and PR4, originally isolated on Pselrdomonas aeruginosa, resemble the lipid-containing phage PM2 in appearance. Their host range extends intergenerically to species carrying drug-resistance plasmids of the P and N compatibility groups. In this paper, the serological identity of the two isolates is established and it is concluded that they are the same virus, but with some differences in growth characteristics. They contain double-stranded DNA and are probably icosahedra (65 nm) with short (47 nm) noncontractile tails. Their sensitivity to chloroform and low buoyant density in CsCl(1.265 g/ml) indicate that they contain lipid which is probably located in the thickened inner layer of the capsid. A study is made of their adsorption efficiencies to sensitive and resistant bacteria, and it is found that, unlike most sex-specific phages, they adsorb directly to the cell surface and not to sex pili. Their host range is shown to include strains harboring a drug-resistance plasmid of the W compatibility group. BRADLEY.D. E.. et E. L . RUTHERFORD. 1975. Basic characterization of a lioid-containing bacteriophage specific for plasmids of the P, N, and W compatibility groups. Can. J. ~ i robiol. 21: 152-163. Des etudes preliminaires ont montre que les bacteriophages PR3 et PR4, isoles a I'origine chez Pseudomonas aeruginosa, ont I'apparence du phage PM2 qui contient des lipides. Leur gamme d'h8tes possibles s'etend intergeneriquement a d e s especes porteuses de plasmides de rksistance aux antibiotiques, plasmides qui appartiennent aux groupes P et N de compatibilite. Tel que present6 dans ce papier, les deux isolats sont identiques d'apres la serologic. 11 s'agit donc du mOme virus, mais avec certaines variantes dans les caracteres d e croissance. Ils contiennent de I'ADN bicatenaire e t il s'agit probablement d'isocahedres (65 nm) a queues courtes (47 nm) non contractiles. A cause de leur sensibilite au chloroforme et de leur faible densite de flotaison en CsCl(1.265 glml), ces phages doivent contenir un lipide probablement situe dans I'epaississement de la couche interne de la capside. Apres etude d e leur efficacite d'adsorption sur des bacttries sensibles et rtsistantes, on trouve que, contrairement 5. laplupart des phages spkcifiques de sexe, ils s'adsorbent directement a la surface de la cellule e t non aux pili sexuels. Leur gamme d'h8tes comprend des souches porteuses d'une resistance aux antibiotiques plasmidique du groupe de compatibilite W. [Traduit par le journal]
Introduction Several bacteriophages have been isolated on strains of Pseudomonas aeruginosa carrying P group drug-resistance plasmids (9) which can transfer intergenerically. They are the ribonucleic acid (RNA) phage PRRl of Olsen and Thomas (17), the tailed phages PR3 and PR4, and the filamentous phage Pf3 isolated by Stanisich (1 8). Of these, PRRl and Pf3 appear to be specific for the P group plasmids, but PR3 and PR4 are also able to plaque on strains carrying plasmids of the N compatibility group (6) which can also transfer intergenerically (9). Khatoon et al. (1 5 ) have described a filamentous phage, designated IKe, 'Received June 26, 1974. ZPresent address: Faculty of Medicine, Memorial University of Newfoundland, St. John's, Newfoundland, Canada A I C 5S7.
c
which is specific for N group plasmids. The present paper describes the basic characteristics of isolates PR3 and PR4. A preliminary study (6) has shown that PR3 and PR4 appear identical in the electron microscope and have a thickened capsid similar to that of the lipid-containing phage PM2 of Espejo and Canelo (I 1). Unlike PM2, however, PR3 and PR4 are thought to have a short tail. Both isolates, and also Pf3, are believed to adsorb directly to the cell wall of sensitive bacteria, while other "plasmid phages" adsorb to pili. For example, the RNA phage PRRl attaches to pili determined by the plasmid RPl (6). PR3 and PR4 are thus unusual because of thir possible lipid content, host range, and site of adsorption. The present work is concerned with these aspects and other basic characteristics. Pre-
BRADLEY AND RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE
TABLE 1 Details of bacteria used
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
Strain
Plasmid compatibility group
P . aeruginosa PA02605 R P l K/lP04-RPl 18s RPI 18s PU21 S-a PU21 E. coli
Source B. W. Holloway K / l P 0 4 - *X18S R P l T. L. Pitt T. L. Pitt G. A. Jacoby G. A. Jacoby R. H. Olsen Rajul V. Iyer Rajul V. Iyer R. Thompson G. A. Jacoby
*A mutant of strain K with no "normal" polar pili ( S ) , but with R P I pili after mating \ ,
(b).
?This strain carries the plasmid R1822 (16) which has been reidentified a s R P I (Olsen, personal communication).
liminary experiments on their adsorption to bacteria carrying various R factors are also described. The other phages mentioned above are included for comparison.
Materials and Methods Bacteria and Bacteriophages Bacterial strains are described in Table I . The bacteriophages were supplied as follows: P R R l RNA phage (17) by Dr. R. H. Olsen, University of Michigan, U.S.A.; filamentous phageIKe(l5) by Dr. RajulV. Iyer, University of Ottawa, Ottawa, Canada; filamentous phage Pf3 and the phages PR3 and PR4 (18) by Dr. V. Stanisich, Monash University, Australia; and tailed P. aeruginosa pilus phage C5 (7) by Mr. T . L. Pitt, Central Public Health Laboratory, London, England. Cult~rreMedia and Methods All media were made with Oxoid nutrient broth, 2% (w/v) agar being added for plates. Double-agar layer plates contained a final concentration of 0.5% (w/v) agar in the soft top layer. Bacteria were grown at 37' in shake culture or on plates. Preparation of Bacteriophage Suspensions High-titer phage stocks were prepared either by the confluent lysis of bacterial growth (strain PA02605 or K/IP04- R P I ) on large (15-cm diam) double-agar layer plates followed by extraction with broth for 2 h at room temperature, or by the lysis of exponential phase broth cultures of host bacteria. When required, concentration and purification were by alternate high- and low-speed centrifugation (60 000 x g for 2+ h and 7000 x g for 20 min at 4") with the final suspension in broth. Phage susoensions were titrated by standard methods (1). The Spot Test for Bacteriophage Lytic Activity A double-agar layer plate of the bacterium under test was prepared, and a loopful of phage suspension was placed on the surface. After overnight incubation, clearing indicated lytic activity. Anomalous results due t o bac-
teriocin activity were avoided by growing phage stocks on bacterial strains which did not produce bacteriocins active on the strain under test. Serology Anti-PR3 and -PR4 sera were prepared in rabbits. Injections containing about 2 x 109plaque-formingunits (PFU) each were given intravenously on days 0, l , 2 , 5 , 7 , 14, and 18, with bleeding on day 22. Complement was inactivated at 56' for 40 min. The rate of inactivation (K) of a phage by an antiserum was determined by the addition of 0.1 ml of bacteriophage (about 2 x lo7 PFU/ml) to 10 ml of broth containing 0.1 ml of antiserum. The mixture was incubated in a shaker at 37' and 0.1-ml samples were diluted into 10 ml of broth at 30 and 45 min for titration of survivors. K values were calculated for a given phage plus antiserum combination at each time according t o the method of Adams (I), and the average of the two determinations was taken as the final value. N o inactivation of either PR3 or PR4 was found with serum from rabbits which had not been inoculated. Acridine Orange Staining The fluorescent staining method for determining the type and strandedness of bacteriophage nucleic acids was carried out as described in detail elsewhere (3). A PR3 suspension, pretreated with ribonuclease (RNase) (Sigma RNase A, type IA) and deoxyribonuclease (DNase) (Sigma DNase I), in phosphate-buffered saline was dried down as small droplets on glass slides. After fixation (Carnoy's fluid) and staining in 0.01% acridine orange in citric acid and phosphate buffer (pH 3.8) for 5 min, the slides were soaked in 0.15 M Na,HPO, for 15 min and viewed under ultraviolet light (257 nm). The colors obtained were noted together with changes produced by immersion in molybdic or tartaric acids (3). The sensitivitv of the dried droplets to DNase (20 pgjrnl) and R N ~ S ; (100 pg/ml) was also tested (3). Electron Microscopy Bacteriophage particles were mounted for electron microscopy by floating a carbon-coated specimen support
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
154
CAN. J. MICROBII3L. VOL. 21, 1975
grid on the surface of a suspension in broth (prepared as described above) at 37" for as long as possible (1-24 h). Grids were then washed, and negatively stained with neutral 1% sodium phosphotungstate. I n a n alternative method (2) using a different host strain (Escherichia coli 55-3 R199), a loopful of soft agar was removed from the center of a spot test of phage PR4. This was extracted with a small amount of 0.1 M ammonium acetate solution which was mixed with 2% sodium phosphotungstate and dried onto carbon-coated grids. Host bacteria with adsorbed virions were prepared by mixing phage suspension and host cells (log phase) in broth at as high a multiplicity of infection (moi) as conveniently possible (about 20 PFU/bacterium). After gentle shaking of the mixture for 20 min at 37", bacteria were mounted for electron microscopy by centrifuging them directly onto carbon-coated support grids, and negatively staining them. Buoyant Density Determination Bacteriophage suspensions were concentrated at 60 000 x g for 24 h in the ultracentrifuge. Pellets were resuspended in 0.3 ml of neutral 0.1 M ammonium acetate solution and centrifuged again at low speed t o remove debris. A volume of 0.25 ml of phage suspension was added to 4.75 ml of tris(hydroxymethyl)aminomethane (Tris) buffer (pH 8.2) containing 2 g of CsCl in a tube for a Spinco Rotor SW 50.1. After the tube was filled with mineral oil, the samples were centrifuged at 35 000 rpm for 49 h at 13". Six-drop fractions were collected by puncturing the bottoms of the tubes. A volume of 0.05 ml of each fraction was diluted into 0.95 ml of chilled broth. The spot test (sensitivity about 1 x lo3 PFU/ml) was used to determine which fractions contained infectious phage, and these were subsequently titrated. Refractometer measurements were carried out on a series of fractions from each tube to determine their densities according t o the relationship given by Ifft et al. (13). Adsorption of Bacteriophages A sample of exponential phase bacteria from a shake culture was adjusted with broth to about 2 x lo9 cells/ml (estimated by opacity) with broth. Two milliliters was mixed with 2.0 ml of a phage suspension in broth (titer about 2 x lo6 PFUjml) in a conical flask. The mixture was incubated at 37' for 20min with gentle shaking. Bacteria were then removed by two cycles of centrifugation at 7000 x g for 10 min, and the supernatant was titrated for surviving phage. A blank sample, in which the bacterial suspension was replaced with broth, was processed simultaneously. Treatment of Phage PR3 with Organic Solvents The method used was based on that of Espejo and Canelo (11). Aliquots of PR3 suspension in broth were mixed with various proportions of chloroform or ether in sealed medical flat bottles. These were fixed, slightly off the horizontal, to a platform shaking at about two oscillations/s at room temperature. After 4 h, the aqueous phase was separated by centrifugation at 1000 x g for 15 min and titrated for surviving phage. A blank, in which the solvent was replaced with broth, was treated simultaneously. The electron microscopy of PR3 suspensions treated with chloroform was carried out as follows. Equal
volumes of phage suspension and chloroform were shaken together for 4 h at two oscillations/s. After the phases were separated, the nutrient broth suspension of PR3 was centrifuged at 7000 x g for 10 min to remove debris and precipitates. Phage particles were mounted for electron microscopy by floating a carbon-coated grid o n the suspension as described above. A control sample without chloroform was treated simultaneously.
Results Inactivation of Bacteriophages by Antiserum The use of the inactivation or velocity constant K (for a given batch of antiserum with a single phage isolate) provides the best means of ascertaining the serological relatedness of two bacteriophages. K for the heterologous phage is compared with K for the homologous phage, most conveniently by their ratio. This is usually defined as K' and is equal to 1 for the homoor heterologous logous phage, a f r a ~ t i o ~ frelated phages, and 0 for unrelated phages. The comparison for PR3 and PR4 is shown in Table 2 together with the velocity constants for the two batches of antiserum. The two isolates are clearly serologically identical, and are in effect the same virus. Nucleic Acid Type and Strandedness The results of the acridine orange fluorescent staining tests are shown in Table 3, nucleic acid strandedness being indicated by an appropriate prefix (1 = single-stranded, 2 = double-stranded). Phage PR3 is compared with the known 2-DNA coliphage T5. It is clear from the colors obtained, and the sensitivity of PR3 nucleic acid to DNase, that it too contains 2-DNA. Morphology Isolates PR3 and PR4 were identical in appearance and dimensions, as might be expected. Negatively stained preparations gave virions with a varied appearance shown typically in Fig. 1. Both full and empty heads can be sezn, some (arrowed) having a thickened inner layer. TABLE 2 Relative velocity constants (K') for bacteriophages PR3 and PR4 with their antisera Values of K ' for phages Serum (K for homologous phage)
PR3
PR4
BRADLEY AND RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE
155
TABLE 3 Colors obtained with the acridine orange staining of bacteriophage PR3
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
Color obtained after treatment Specimen
Nucleic acid
PR3 T5 Standard Standard Standard Standard
2-DNA 2-DNA 2-RNA I-DNA 1-RNA
?
NaZHPO4
Moly bdic acid
Bright yellowish green Bright green Green Green Red Red
Bright green Bright green Green Green fading Paler green Paler red
'Resistance to a nuclease, -; sensitivity,
Tartaric acid
RNase*
DNase*
-
+ + ++
Orange Reddish orange Orange Red Paler green Paler red
+
-
+.
Tails were only rarely present and often difficult to find, their frequency varying greatly from one specimen grid to another. An example of a PR4 particle with a partly disrupted head, and a tail, is shown in Fig. 2; the distorted thick inner layer is clearly visible. In Fig. 3, there is a slight discontinuity (arrowed) in the capsid of the top particle, suggesting that the tail has fallen off. These small holes were quite frequent. The 9.0-nm-thick inner layer is clearly visible in the lower particle. Both virions have lost their nucleic acid. Figure 5 also shows empty virions, but with tails, the appearance of which suggests that they are disintegrating. It was extremely difficult to find completely intact virions (Fig. 4). The particle illustrated closely resembles coliphage T3 in appearance, the thick inner layer being obscured by nucleic acid. The dimensions for PR3 and PR4 are head size 65 nm and tail length about 47 nm, calibrated with Dow polystyrene latex (264 nm). The tail length is much shorter than previously observed (6). Some tails, as in Fig. 2, were much longer than others and were thought to be in the process of degradation before falling off, despite the fact that the phage suspension was freshly prepared for examination, usually on the same or previous day. The length given refers to those on more or less intact virions. No contractile apparatus could be found on any tails. No phages could be found exhibiting the typical profile associated with an octahedron (4), the more or less regular hexagon suggesting a probable icosahedral form.
Buoyant Density centrifugation profiles of PR3 and PR4 in cesium chloride are shown in Fig. 6. The buoyant densities were calculated as 1.267 g/ml for PR3 and 1.264 g/ml for PR4. A single sharp infectious peak was obtained in each case, and the phage
was visible as an opaque band in the gradient before fractionation.
Host Range of Bacteriophages The lytic activity of PR3 and PR4 on strains of P. aeruginosa and E. coli carrying various plasmids is shown in Table 4. Phages PRRI, Pf3, and IKe are included for comparison. All test strains carrying plasmids of the P, N, or W compatibility groups were lysed by PR3 and PR4, while background strains without plasmids were resistant. The host range of PRRI and Pf3, however, is restricted to RPl, and IKe can only lyse strains carrying RM98 and R199 (N group). A mutant of K/lP04- RPI resistant to PRRI was isolated by streaking soft agar from a spot test of phage PRRI. This proved to be sensitive to phages Pf3, PR3, and PR4 as indicated by spot tests. However, it was unstable and tended to revert to PRR1-sensitivity on subculturing several times. Growth of Infected Bacteria The growth curves for cultures of P. aeruginosa PA02605 infected with PR3 and PR4 at about the same moi are compared in Fig. 7. It can be seen that the onset of lysis occurs after about 130 min for PR3 and 95 min for PR4, a considerable difference which will be discussed below. A curve for strain 18s RPI infected with PR3 at a slightly higher moi shows a relatively small difference in the time to lysis. A knowledge of the time to lysis is required to ensure that adsorption tests are completed before the release of phage progeny. Comparison of Plaque Morphology and Eficiency of Plating for Phages PR3 and PR4 Aliquots of PR3 and PR4 containing about 200 PFU (titrated on strain PA02605) were plated by the double-agar layer method with
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
156
CAN. J. MICROBIOL. VOL. 21, 1975
FIG. 1. Typical appearance of PR3 phage particles. In two of them (arrowed), a thick inner layer is visible. x 115 000. FIG. 2. A broken PR4 particle with a distorted inner layer and a tail. x 300 000. FIG. 3. PR4 particles extracted from a spot test "plaque" on E. coli 35-3 R199, having an appearance identical with phages grown on strains of P. aerrtginosa. A small hole is present in the capsid of the top particle (arrowed). x 300 000. FIG. 4. An intact PR3 particle. The nucleic acid obscures the inner layer of the capsid, and the tail is shorter than that shown in Fig. 2. x 300 000. FIG. 5. PR3 particles with tails in the process of disintegration. Some nucleic acid appears to remain in the heads partly obscuring the inner layer, though one particle (arrowed) has a broken inner layer. x 225 000.
BRADLEY A N D RUTHERFORD: BASIC CHARACTERIZATION O F BACTERIOPHAGE 10 2.9X10
-
10 2.5X10
-
I
l
l
1 I
I
I
I
PR3
I
I
I
I
I
,
-
I
I
I
-
PR4
1.33
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
- 1.31 - 1.29
1
$
1.7~10'~-
-
\
3 LL
a
0
1.27 1
H
i.3xrd01.25
5
I x109
FRACTION NUMBER
FIG.6. Centrifugation profiles of PR3 and PR4 in 40% CCsCI. Buoyant densities were PR3, 1.267 g/rnl, and PR4, 1.264 g/ml. TABLE 4 Lytic activity of bacteriophages Lytic activity of phages* Background strain
P. aert/ginosa PA02605 K/lP0418s 18s PU21 PU21 E. coli CR34 MA50 MA50 J5-3 J53-1
Plasmid
Compatibility RrOUP
PRRl
Pf3
PR3
PR4
P P P
+ + +-
+ + + -
W
-
-
+ + + +-
+ + ++-
+ + -
-
(+) (+)
(+)
RPl RPl RP1 None S-a None RPl RM98 None R199 S-a
*+ indicates clearing, (+) partlal clearing,
-
P N -
N W
-
+ -
-
-
+
-
-
+ +
+
IKe
-
-
-
-
+ +-
-
- no clearing.
various strains of plasmid-carrying bacteria. The plaques obtained were counted and the efficiency of plating (eop) was calculated as a fraction of the number obtained with P. aeruginosa PA02605 (RPI) which was the original isolating host (18). At the same time the morphology of the plaques was noted. The results in Table 5 show that the isolating strain provides the largest number of plaques as expected, and that there is a substantial reduction in eop for all strains of E. coli, probably resulting from restriction. The plaque morphology varies somewhat according to the
host. Plaques of PR3 are generally smaller and more turbid than those of PR4. Adsorption of Bacteriophages The results shown in Table 6 were obtained by titrating surviving phage after incubating mixtures of phage and bacteria (see Materials and Methods). The percentage of phage adsorbed was calculated by subtracting this titer from that of a blank (representing 0% adsorbed). Results obtained within 2% of the blank are given as < 2% phages adsorbed. The values obtained are
Can. J. Microbiol. Downloaded from www.nrcresearchpress.com by SAVANNAHRIVNATLABBF on 11/23/14 For personal use only.
MINUTES
FIG. 7. Growth curves for bacterial cultures infected with phages PR3 and PR4. A culture of strain PA02605 or 18s RPI (about 5 ml) was grown t o early log phase, and phage suspension (0.2-2.0 ml) was added at 0 min. The cell concentration was taken at intervals (estimated by opacity). The multiplicities of infection were 2.5 PFU/bacterium for PR4 + PA02605, 2.0 PFU/bacterium for PR3 PA02605, and 3.8 PFU/bacterium for PR3 + 18s R P l . The times at which lysis began are indicated by arrows.
+
TABLE 5 Efficiency of plating (eop) and plaque morphology of phages PR3 and P R 4 PR3 Strain P. aeruginosa PA02605 K/IP0418s PU21 E. coli CR34 MA50 353-1
Plasmid
Compatibility group
PR4 -
eoP
RPI RPI RPl S-a
P P P W
1 0.87 0.51 0.36
RPl RM98 S-a
P N W
