in Pseudomonas aeruginosa Strain PAO - PubMed Central Canada

Vol. 23, No. 3 Printed in U.S.A.

JOURNAL OF VIROLoGY, Sept. 1977, p. 461-466 Copyright C 1977 American Society for Microbiology

Identification of the Cell Wall Receptor for Bacteriophage E79 in Pseudomonas aeruginosa Strain PAO KEN JARRELL AND ANDREW M. KROPINSKI* Department of Microbiology and Immunology, Queen's University, Kingston, Ontario, Canada K7L 3N6

Received for publication 25 January 1977

Bacteriophage E79 was shown to interact with the lipopolysaccharide (LPS) of Pseudomonas aeruginosa strain PAO. LPS isolated from an E79-sensitive, smooth strain inactivated the phage, exhibiting a PhlM0 value (concentration of LPS that caused a 50% decrease in the titer of phage during 1 h of incubation at 3700) of 0.04 pmg/ml, whereas the LPS isolated from a rough mutant derived from the wild type showed no neutralizing activity towards E79. EDTA and sodium deoxycholate were demonstrated to abolish the neutralizing capacity of the smooth LPS. One E79 receptor site was shown to be equivalent to 10-"6 g of LPS. Despite the isolation of considerable numbers of bacteriophages for typing Pseudomonas aeruginosa clinical isolates, comparatively few studies on the basic properties of these phages, including their receptor specificities, have been carried out. An exception to this statement are the studies carried out in large part by Bradley and his co-workers, who have shown that the filamentous phage Pf (9), the single-stranded RNA-containing phage PP7 (6), and the Bradley B type (5) phages P04 and P02 (7, 8) adsorb specifically to pili. Recently, phli have been shown to be the receptor for phages C22, M6, PE69, and C5 (10), as well as for the generalized transducing phage F116 (30). Phages 68, PB1 (10), B3, D3, G101, and E79 have been described as cell wall-specific phages (19), but as yet no further studies have been undertaken to isolate the specific component of the wall to which these phages adsorb. The role of somatic receptors in phage adsorption was suggested from the work of Mead and van den Ende (27), who showed that Boivin and ethylene glycol extracts of P. aeruginosa L11 would irreversibly inactivate several of the phages tested. It is of interest to note that these extracts inactivated some phages for which strain L11 was not the host. Bartell and co-workers (2) have reported that P. aeruginosa phage 2 was inactivated by both slime and lipopolysaccharide (LPS), and more recently phage 4PLS-1 was shown to be LPS specific (20). We decided to examine more closely the receptor for phage E79, since it is one of the more significant phages isolated against P. aeruginosa. Since its isolation by Holloway et al. (16), this virulent phage has been used extensively as a counterselecting agent in genetic crosses using the PAO and PAT strains of P. aerugi-

nosa, and two ese loci (sensitivity/resistance to E79) have been mapped (17). We show in this report that the receptor for phage E79 on P. aeruginosa strain PAO is contained in the LPS. MATERIALS AND METHODS Phage and bacterial strains. P. aeruginosa PAO 307, originally obtained from B. W. Holloway and now referred to in this laboratory as P. aeruginosa AK-2, was used in this study. A spontaneous E79resistant strain, designated AK-43, was derived from AK-2 by plating 0.1 ml of an overnight culture of AK-2 with 0.1 ml of a high-titer E79 lysate (ca. 1011 PFU/ml). After overnight incubation, a resistant colony was picked and subjected to several single-colony isolations. Phage E79 was also obtained from B. W. Holloway. Titration of phage. Phage plaque assays were made by the agar overlay technique (1). The top and bottom layers consisted of tryptic soy broth (Difco) containing 0.6 and 1.5% (wt/vol) agar, respectively. Absorption kinetics. The procedure followed was identical to that of Jarrell and Kropinski (20). Isolation of LPS. LPS was extracted from AK-2 cells grown in tryptic soy broth by using the hot phenol-water procedure of Westphal and Jann (36), incorporating the modification of Key et al. (21) to remove contaminating mucopeptide. AK-43 LPS was extracted from tryptic soy broth-grown cells with petroleum ether-chloroform-phenol (PCP) (13). Due to the insolubility of the isolated AK-43 LPS, no further purification by high-speed centrifugation was carried out. AK-2 LPS contained 9.5% (wt/wt) RNA, estimated by absorbency at 260 nm, and 2.5% (wt/wt) protein, determined by the method of Lowry et al. (25), whereas AK-43 LPS contained 4.1% protein and no RNA. De-0-acetylation of LPS. AK-2 LPS (5 mg/ml) was treated with an equal volume of 0.1 M NaOH for

461

462

JARRELL AND KROPINSKI

1 h at 370C (37). After neutralization with 0.1 M HCl, the solution was dialyzed at 40C against distilled water for 24 h and lyophilized. The existence of O-acetyl groups in the LPS was examined using the procedure of Hestrin (15). Isolation of polysaccharide fraction. AK-2 LPS (ca. 10 mg) was hydrolyzed in 1% (vol/vol) acetic acid for 90 min at 1000C in a sealed tube (12). The tube was frozen to promote aggregation and settling of the lipid A. After thawing, the hydrolysate was centrifuged at 10,000 x g for 5 min. The aqueous phase (polysaccharide fraction) was removed, and the precipitate was washed twice with cold distilled water. These washings were combined with the aqueous phase in tared vessels and dried to a constant weight over NaOH in a desiccator. Neutralization studies. The neutralization of phage infectivity was determined using a modification of the procedure of Lindberg (23), as described previously (20). Neutralization experiments were performed using intact LPS, de-O-acetylated AK-2 LPS, and isolated AK-2 polysaccharide. The phage-inactivating capacity of the LPS (Phl50; 3) was expressed as the concentration of LPS (micrograms per milliliter) that caused a 50% decrease in the titer of phage during 1 h of incubation at 370C. Effect of EDTA and DOC on the phage-inactivating capacity of AK-2 LPS. These studies were carried out as previously reported (20), using a final concentration of 5 mM EDTA or sodium deoxycholate (DOC) and a final LPS concentration of 0.5 pig/ ml. Separate controls were included to ensure that the DOC and EDTA did not have any effects on phage E79 viability under the experimental conditions used. Determination of the number of E79 binding sites on AK-2 LPS. The method used was that of Boman and Monner (4), in which 0.1 ml of a hightiter E79 lysate (ca. 5 x 1011 PFU/ml) was incubated for 60 min at 370C and 75 rpm with various concentrations of LPS. The number of phage remaining at each LPS concentration was determined after the 60-min incubation period.

RESULTS P. aeruginosa AK-43 was a spontaneously derived E79-resistant mutant of AK-2. AK-43 colonies were small and nonspreading and possessed a granular consistency typical of LPSdefective (rough) strains (26). In addition, AK43 cells agglutinated in 0.2% (wt/vol) acriflavine and 4% (wt/vol) saline, techniques that have been used to distinguish between smooth and rough strains of P. aeruginosa (22). This was suggestive evidence that E79 is an LPSspecific phage. Adsorption of E79 to intact cells. The adsorption of E79 to AK-2 cells is shown in Fig. 1. The concentration of unadsorbed phage decreased logarithmically with first-order kinetics, and an adsorption rate constant of 1.5 x 10-9 ml/min was calculated. No requirement for

J. VIROL. 100

90

AA~

4~~\-

A~~

A

A

u 0.

\*\

70-

0o 60w 0

0\ I

n

O 50z Z 400.C

jV1 0

1

2

3

4

5

MINUTES

FIG. 1. Adsorption of phage E79 to P. aeruginosa strain AK-2 and its rough derivative, AK-43. Residual phage were determined after chloroform treatment, using AK-2 as the plating host. Symbols: Adsorption to AK-2 (0); AK-43 (A).

divalent cations was shown for phage E79, since the phage efficiency of plating remained unaltered when media containing 0.1 or 1% trisodium citrate were used in place of media supplemented with 1 mM CaCl2 (29). E79 failed to adsorb to the mutant AK-43, indicating that AK43 was a resistant rather than a tolerant mutant (28). Neutralization studies. AK-2 LPS, obtained in 1.63% yield from acetone-dried cells by the phenol-water procedure (36), was a light, white, fluffy powder. In contrast, AK-43 LPS obtained in a 2.8% yield using the PCP method (13) was in the form of hard, white, waterinsoluble aggregates. AK-43 LPS could not be extracted in measurable amounts with the phenol-water technique, and, likewise, AK-2 LPS could not be extracted by the PCP method. Smooth LPS forms are known to be excluded from the PCP extract (13). Thus, the mutant AK-43 is a rough strain derived from the smooth parent AK-2. A variety of techniques has been used for the distinction of bacteriophage receptors on the cell surface, but the classic method is extraction, purification, and analysis of phage-inactivating agents from the susceptible host bacteria. Therefore, the isolated LPS preparation from AK-2 cells was tested for its inactivating capacity towards phage E79. Figure 2 shows the relationship between LPS concentration and the percentage of phage E79 PFU inactivated in the assay system. The concentration of AK-2 LPS needed to inactivate 50% of the phage within 1 h (Phl.,) at 370C was 0.04 ug of LPS per ml. This Phl. value obtained for E79 compares favorably with similar values obtained

100

80-

N

4o0

S 40c1

Oc100

463

PHAGE E79 CELL RECEPTOR

VOL. 23, 1977

Vxx I

1.O

1.0

I

01

01 0.005

pg LPS/ml

FIG. 2. Titration of the phage-inactivating activity of LPS extracted from P. aeruginosa AK-2 (0) and AK-43 (A). AK-2 LPS was treated with 0.1 M NaOH for l h at 37C , followed by neutralization and dialysis. The phage-neutralizing activity of the alkali-treated LPS is shown by the symbol x. The activity of the LPS in the neutralization assay is expressed as the concentration of LPS (micrograms per milliliter) that will inactivate 50% of the PFUs (Phl50) under the experimental conditions. The Phl50 of the AK-2 LPS is 0.04 ug/ml.

recently for other P. aeruginosa LPS-specific phages (Kropinski et al., Can. J. Microbiol., in press) as well as for LPS-specific phages of other genera (14, 23, 37). Even at very high LPS concentrations (>50 ug of LPS per ml), there was still a small fraction of phage that was not neutralized. No inactivation of E79 could be obtained with the rough AK-43 LPS even at a concentration of 100 ug/ml. These results show that the receptor for E79 lies in the LPS of the susceptible host. Since no inactivation of the phage could be obtained with AK-43 LPS, presumably the E79 receptor lies in the polysaccharide portion of the smooth LPS. However, attempts to neutralize the phages by using isolated polysaccharide were not successful (data not shown). This may indicate that the polysaccharide, though containing the receptor site(s), needs the lipid A portion of the LPS to obtain a correct spatial conformation in order to effectively interact with the phage. Several authors have noted the apparent need for O-acetyl groups in phage adsorption (31, 37). When AK-2 LPS was de-O-acetylated and used in a neutralization assay, the results show that the Phl. value for the de-O-acetylated LPS was increased over 200-fold as compared to the native LPS (Fig. 2). Thus, E79 would also appear to require O-acetyl groups in its adsorption step. An alternate explanation is that the treatment used to de-O-acetylate has had an undetermined effect on P. aeruginosa LPS, which it apparently did not have on the Rhizobium LPS used by Zajac et al. (37). This

possibility appears to be valid since no O-acetyl was found in AK-2 LPS with the procedure of Hestrin (15). Effect of surfactants on phage neutralization by LPS. LPS is known to be dissociated, with concomitant loss of biological activity, by a variety of surface-active agents (33). It has been shown by several workers that EDTA and DOC can disrupt LPS and destroy its phageinactivating capacity (18, 20, 34). When LPS, at a concentration of 0.5 ,ug/ml (sufficient to neutralize 90%o of the phage in 1 h), was preincubated with either 5 mM EDTA or 5 mM DOC for 5 min before the addition of E79, the phageinactivating capacity of the LPS was totally destroyed (Fig. 3). Neither reagent, at the concentration used, had any observable harmful effect on the phage over the 1-h incubation period (data not shown). Figure 3 also shows the kinetics of phage neutralization by the LPS. There was a logarithmic decrease in the number of free phage until about 90% of the phage were neutralized, followed by a break in the curve. This break could be due to phenotypic variation in the E79 population, as has been shown to exist with other phages (1). The k value can be calculated using the formula PdP0 = ekct, where P. is the concentration of phage infective at the beginning and P, is that at time t, c is the concentration of LPS, and t is the incubation time in minutes (34). The k value for E79 using 0.5 ,g of AK-2 LPS per ml was 1.5 x 102 ml/min per mg of LPS. This is somewhat lower than the N-

100-

A

A _-_a-

,

x

so

u

ul

0\

OK

e

10-

50

10

20

30

40

50

60

MINUTES

FIG. 3. Effect of surface-active agents on the E79neutralizing activity ofAK-2 LPS: (U) LPS + EDTA (5 mM) + E79; ( x) LPS + DOC (5 mM) + E79; (-) LPS + E79. The concentration of LPS was 0.5 pgl ml. No phage inactivation occurred when E79 was incubated with EDTA or DOC in the absence of LPS or when E79 was incubated alone in tryptic soy broth for I h (data not shown).

464


J. VIROL.

value of 5.03 x 102 ml/min per mg of LPS reported for Salmonella phage E15 (34). Number of E79 binding sites on AK-2 LPS. Using the method of Boman and Monner (4), it was shown (Fig. 4) that there was more neutralization per microgram of LPS when lower concentrations of LPS were used. There is no saturating level reached, up to 100 jig of LPS per ml, in sharp contrast to the result obtained with E. coli LPS and phage W (4), where saturation of sites per microgram occurred at 10 Ag of LPS per ml. It may be that P. aeruginosa LPS aggregates when present in higher concentrations, causing a reduction in the number of available phage binding sites. The 9.8 x 109 phage sites per Ag obtained using a final LPS concentration of 1 ,tg/ml indicates that one phage site is equivalent to 10-16 g of LPS, a value very similar to that obtained for phage W (4). DISCUSSION LPS has long been known to be the receptor for a number of phages in many different genera (24). LPS is known to act as receptor for T3, T4, and T7 in E. coli (35), for T2 and T4 in Shigella dysenteriae (14), for P22 in Salmonella typhimurium (24), and for phage 1P in Rhizobium trifolii (37). However, in P. aeruginosa only phage 2, which is inactivated by slime as well as LPS (2), and phage OPLS-1 (20) have been shown to use LPS as their cell receptor. However, the amount of LPS needed to neutralize 50% of phage 2 and kPLS-1 was greater than 1 /ig, a value much higher than the Phl50 values recorded for other LPS-specific phages. For example, a Phl., value of 0.1 ,g of LPS per ml was obtained inR. trifolii for phage

1P (37), and as little as 0.04 ug of E. coli strain D21e7 LPS neutralized 50% of phage kW (4). In this report, we have shown that E79 was neutralized by LPS isolated from a sensitive host, whereas the LPS isolated from a spontaneous resistant mutant was ineffective in neutralizing the phage. Furthermore, the Phl.,0 value of 0.04 ug of LPS per ml obtained for E79 with AK-2 LPS compared very favorably with the values obtained for LPS-specific phages of other genera. This is the first report of a P. aeruginosa phage having a Phl., value this low. The LPS of the resistant strain, AK-43, was defective and could only be extracted in measurable amounts by the PCP method (13) designed especially for extraction of LPS from rough strains. Chemical analysis of AK-43 LPS showed that it lacked glucose, rhamnose, and fucosamine, which were major constituents of AK-2 LPS. In addition, no high-molecularweight fraction (i.e., side chain) could be demonstrated after gel exclusion chromatography of AK-43 polysaccharide (Jarrell and Kropinski, submitted for publication). Since rough strains are defective in the polysaccharide component of the LPS, it was quite probable that the polysaccharide moiety contains part or all of the receptor site(s) for E79. An attempt was made to isolate the polysaccharide from the wild-type cells and use it to inactivate the phage, but no neutralization of E79 was obtained. It may be that the isolated polysaccharide contains the receptor site, but without the lipid A component it may lack the proper spatial requirements to render the receptor amenable to phage binding. Alternatively, the E79 receptor may in fact be contained in both the polysaccharide and the lipid A, and hence iso6 lated polysaccharide or isolated lipid A would not be able to neutralize the phage. Lipid A is known to form part of the receptor site for q5 in 4 E. coli K-12 (4). A third possibility is that E79, which is known to possess a structure similar to 0 3~ the T-even phages (32), may require specific binding sites for several components. The polysaccharide isolated by the methods used here is c'2 known to be partially degraded and so may not contain, intact, all of the sites required. EDTA and DOC have long been known to I0 10 is 10O 50 2 10 disrupt the LPS, causing a loss of biological pg LPS/ml FIG. 4. Relationship between degree of inactiva- activity: EDTA apparently acts by cheating tion of E79 and concentration of AK-2 LPS. A high- divalent cations needed for LPS integrity (34), titer E79 suspension (ca. 5 101° PFU/ml) was probably Mg2+ in the case ofP. aeruginosa LPS incubated for 60 min at 37°C with the concentrations (11), whereas DOC causes dissociation of P. of LPS indicated. The residual number of plaque- aeruginosa LPS into 12,000- to 16,000-dalton forming particles was determined by the overlay pro- subunits (18). In this report, 5 mM EDTA or 5 cedure, using AK-2 as the plating host. mM DOC totally destroyed all the phage E79-

0

4-

.Q

0

1

c

0

I5

x

PHAGE E79 CELL RECEPTOR

VOL. 23, 1977

neutralizing ability of the smooth LPS. Several authors have reported similar effects, both with P. aeruginosa LPS (20) and with other endotoxins (23, 34). Since E79 does not require divalent cations for adsorption, the EDTA effect is not on adsorption cofactors but on the LPS itself. Neither surfactant had a harmful effect on the E79 plaquing ability under the conditions used. This is in contrast to OPLS-1, which was inactivated to some extent by DOC (20). When high-titer phage preparations (ca. 5 x 1011 PFU/ml) are mixed with LPS so as to saturate the phage with LPS within a short period after mixing, it is then possible to determine the amount of LPS comprising a phage receptor. In contrast to the results of Boman and Monner (4), who obtained a distinct end point at 10 ,.tg of LPS per ml, no end point was obtained with AK-2 LPS over the range of LPS concentrations used. The number of phage neutralized per microgram of LPS decreased as the higher concentrations of LPS were used (Fig. 4), and no leveling off was obtained. This may reflect a greater tendency of P. aeruginosa LPS to aggregate compared with E. coli LPS, thus decreasing the available number of phage receptor sites. If this were true, the aggregation of P. aeruginosa LPS may, in part, account for its lowered activity compared to enterobacterial LPS in some endotoxicity assays. The value of 101-6 g of LPS per receptor site obtained at the lowest concentration of LPS used (1 ,ug of LPS per ml) is very similar to the 2 x 10-16 g of LPS per site obtained for OW in E. coli (4). The knowledge that E79 is an LPS-specific phage can be utilized in obtaining LPS-defective mutants. Several mutants, with varying degrees of roughness, have been already obtained by selection for resistance to E79, and reports on their LPS composition are to be presented elsewhere (Kropinski et al., in preparation). ACKNOWLEDGMENTS This research was supported by grants from the Medical and National Research Councils of Canada, and by an M.R.C. Studentship to one of us (K. J.). LITERATURE CITED 1. Adams, M. H. 1959. Bacteriophages. Interscience Publishers, Inc., New York. 2. Bartell, P. F., T. E. Orr, J. F. Reese, and T. Imaeda. 1971. Interaction of Pseudomonas bacteriophage 2 with the slime polysaccharide and lipopolysaccharide ofPseudomonas aeruginosa strain BI. J. Virol. 8:311317. 3. Beumer, J., and J. Dirkx. 1960. Isolement de substances receptrices des bacteriophages chez les Shigellae. Ann. Inst. Pasteur Paris. 98:910-914. 4. Boman, H. G., and D. A. Monner. 1975. Blocking of bacteriophages 4W and >5 with lipopolysaccharides

465

from Escherichia coli K-12 mutants. J. Bacteriol. 121:465-470. 5. Bradley, D. E. 1955. The fluorescent staining of bacteriophage nucleic acids. J. Gen. Microbiol. 44:383-391. 6. Bradley, D. E. 1966. The structure and infective process of a Pseudomonas aeruginosa bacteriophage containing ribonucleic acid. J. Gen. Microbiol. 45:83-96. 7. Bradley, D. E. 1973. A pilus-dependent Pseudomonas aeruginosa bacteriophage with a long noncontractile tail. Virology 51:489-492. 8. Bradley, D. E. 1973. Basic characterization of a Pseudomonas aeruginosa pilus-dependent bacteriophage with a long noncontractile tail. J. Virol. 12:11391148. 9. Bradley, D. E. 1973. The adsorption of the Pseudomonas aeruginosa filamentous bacteriophage Pf to its host. Can. J. Microbiol. 19:623-631. 10. Bradley, D. E., and T. L. Pitt. 1974. Pilus-dependence of four Pseudomonas aeruginosa bacteriophages with noncontractile tails. J. Gen. Virol. 24:1-15. 11. Brown, M. R. W., and J. Melling. 1969. Loss of sensitivity to EDTA by Pseudomonas aeruginosa grown under conditions of Mg-limitations. J. Gen. Microbiol. 54:439444. 12. Fensom, A. H., and P. M. Meadow. 1970. Evidence for two regions in the polysaccharide moiety of the lipopolysaccharide of Pseudomonas aeruginosa 8602. FEBS Lett. 9:81-84. 13. Galanos, C., 0. Luderitz, and 0. Westphal. 1969. A new method for the extraction of R lipopolysaccharides. Eur. J. Biochem. 9:245-249. 14. Goldham, J., E. Eylan, and B. Kacewman. 1975. Receptors for T2, T4 and T6 phages in Sh. dysenteriae 7 and other serotypes of Sh. dysenteriae. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1 Orig. Reihe A 233:335-341. 15. Hestrin, S. 1949. The reaction of acetylocholine and other carboxylic acid derivatives and its clinical applications. J. Biol. Chem. 180:249-261. 16. Holloway, B. W., J. B. Egan, and M. Monk. 1960. Lysogeny in Pseudomonas aeruginosa. Aust. J. Exp. Biol. 38:321-330. 17. Holloway, B. W., V. Krishnapillai, and V. A. Stanisich. 1971. Pseudomonas genetics. Annu. Rev. Genet. 5:425-446. 18. Ikeda, K., and F. Egami. 1973. Lipopolysaccharide of Pseudomonas aeruginosa with special reference to pyocin R receptor activity. J. Gen. Apple. Microbiol. 19:115-128. 19. Jacoby, G. A. 1974. Properties of R plasmids determining gentamicin resistance by acetylation in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 6:239-252. 20. Jarrell, K., and A. M. B. Kropinski. 1976. The isolation

and characterization of a lipopolysaccharide-specific Pseudomonas aeruginosa bacteriophage. J. Gen. Virol. 33:99-106. 21. Key, B. A., G. W. Gray, and S. G. Wilkinson. 1970. The purification and chemical composition of the lipopolysaccharide of Pseudomonas alcaligenes. Biochem. J. 120:559-566. 22. Kropinski, A. M. B., and J. S. Chadwick. 1976. The pathogenicity of rough strains of Pseudomonas aeruginosa for Galleria mellonella. Can. J. Microbiol. 21:2084-2088. 23. Lindberg, A. A. 1967. Studies of a receptor for Felix 0-1 phages in Salmonella minnesota. J. Gen. Microbiol. 48:225-233. 24. Lindberg, A. A. 1973. Bacteriophage receptors. Annu. Rev. Microbiol. 27:205-241. 25. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

466

J. VIROL.


26. Luderitz, O., A. M. Staub, and 0. Westphal. 1966. Immunochemistry of 0 and R antigens ofSalmonella

27. 28.

29.

30.

31.

32.

and related Enterobacteriaceae. Bacteriol. Rev. 30:192-255. Mead, T. H., and M. van den Ende. 1953. Bacteriophage inhibition by extracts from phage-insensitive bacteria of the genus Pseudomonas. J. Hyg. 51:108-124. Nishikawa, Y., and B. Maruo. 1976. Isolation and characterization of T-even ghost-tolerant mutants of Escherichia coli. J. Bacteriol. 127:40-45. Olsen, R. H., E. S. Metcalf, and J. K. Todd. 1968. Characterization of bacteriophages attacking psychorphilic and mesophilic pseudomonads. J. Virol. 2:357-364. Pemberton, J. M. 1973. F116: a DNA bacteriophage specific for the pili of Pseudomonas aeruginosa strain PAO. Virology 55:558-560. Shaw, D. R. D., and A. N. Chatteijee. 1971. O-acetyl groups as a component of the bacteriophage receptor in Staphylococcus aureus cell walls. J. Bacteriol. 108:584-585. Slayter, H. S., B. W. Holloway, and C. E. Hall. 1964.

33.

34.

The structure ofPseudomonas aeruginosa phages B3, E79 and F116. J. Ultrastruct. Res. 11:274-281. Sultzer, B. M. 1971. Chemical modification of endotoxin and inactivation ofits biological properties, p. 91-126. In S. Kadis, G. Weinbaum, and S. Ail (ed.), Microbial toxins, vol. 5, Bacterial endotoxins. Academic Press Inc., New York. Takeda, K., and H. Uetake. 1973. In vitro interaction between phage and receptor lipopolysaccharide: a novel glycosidase associated with Salmonella phage Virology 52:148-159. Weidel, W. 1958. Bacterial viruses (with particular reference to adsorption/penetration). Annu. Rev. Microbiol. 12:27-48. Westphal, O., and K. Jann. 1965. Bacterial lipopolysaccharides. Extraction with phenol-water and further applications of the procedure, p. 83-91. In R. L. Whistler (ed.), Methods in carbohydrate chemistry, vol. 5. Academic Press Inc., New York. Zajac, E., R. Russa, and Z. Lorkiewicz. 1975. Lipopolysaccharide as the receptor for Rhizobium phage 1P. J. Gen. Microbiol. 90:365-367. E1

35. 36.

37.

.