Inactivation of Receptors for Bacteriophage T5 ... - Journal of Virology

Vol. 24, No. 1 Printed in U.S.A.

JOURNAL OF VIROLOGY, Oct. 1977, p. 419-421 Copyright 0 1977 American Society for Microbiology

Inactivation of Receptors for Bacteriophage T5 During Infection of Escherichia coli B GLENDA B. DUNN AND DONNA H. DUCKWORTH* Department of Immunology and Medical Microbiology, J. Hills Miller Health Center, Gainesville, Florida 32610

Received for publication 6 June 1977

During infection of Escherichia coli by bacteriophage T5, the cell surface receptors for the phage were inactivated so that they could not be isolated from the infected cells. A mutant ofT5 that could only inject 8% of the T5 DNA did not cause the inactivation. In their pioneering studies in the 1950s on bacteriophage adsorption, Weidel and Kellenberger observed that lysates of T5 bacteriophage-infected cells contained structures that were morphologically identical to T5 receptors but did not bind any of the T5 in the lysate. Receptors isolated from uninfected cells, which caused inactivation of the phage, could be easily seen attached to the tips of T5 tails. From this they concluded that the phage may have a mechanism to inactivate the receptors before or at the moment of lysis (8). Teleologically, of course, this makes very good sense, because if active receptors were released from the lysing cells, there would be the possibility that they could inactivate many of the newly synthesized phage. In spite of this postulated receptor-inactivating mechanism, it is commonly thought that lysates of T5 must be purified by CsCl density gradient centrifugation to prevent the phage from being inactivated by the free receptors in the lysate (D. J. McCorquodale, personal communication). We, however, have kept lysates of T5 in their completely native state, without even removing the bacterial debris, for months and have found virtually no loss of PFU. Inactivation would be expected if the phage could attach to free receptors or fragments of receptor-containing walls. Because of the suggestion of Weidel and Kellenberger and our observations regarding the stability of unpurified lysates of T5, we have attempted to isolate T5 receptors from T5-infected cells. We have found that late in the infectious cycle of T5, the receptors are inactivated and cannot be isolated from the infected cells. T5 receptors can easily be removed from Escherichia coli B with dilute alkali (2, 8, 9). The receptors thus isolated appear to be spherical particles approximately 31 nm in diameter

(8) and are composed of a complex of protein, lipid, and carbohydrate (8, 9). Ifthese are added to T5 phage, the phage attach by the tips of their tails to the spherical receptor and are inactivated either by virtue of extruding their DNA (8) or because they can no longer attach to whole cells (10). Thus, the receptor activity of a preparation can easily be measured by phage inactivation. Conditions can be adjusted so that the inactivation of phage by receptors isolated in this way is a first-order reaction; we find the rate to be linearly related to the concentration of receptors used. Figure 1 shows the inactivation of T5 by receptors isolated from uninfected E. coli B and from E. coli B infected for 15 and 30 min with T5H23c. This mutant phage contains an amber mutation in an early gene that prevents the production of any progeny phage (4) and allows testing of receptors isolated from infected cells for inactivation of wild-type T5 without the problem of interference by any newly synthesized phage. By 30 min postinfection it was no longer possible to isolate active receptors from the T5H23c-infected cells. Since we could still detect some receptor activity in a 20-fold dilution of receptors isolated in the same way from uninfected cells, it appeared that at least 95% of the receptors from the infected cells were lost. Addition of [3H]thymidine-labeled T5 to whole cells, followed by filtration at various times, to measure the rate of adsorption confirmed that the receptor activity of infected cells was greatly diminished and also negated the possibility that we could not isolate receptors from the infected cells because they had somehow become more tightly bound. We found that the rate of adsorption of the labeled phage to E. coli B that had been infected for 30 min with T5H23c was reduced by 80% as compared with uninfected E. coli B. Infection ofE. coli F, a more commonly used host for T5 (7), with 419

420

J. VIROL.

NOTES

20.20

CLO\

0o

0

20

30

40

50

60

Time of incubaon (min.)

FIG. 1. Inactivation of bacteriophage T5 by receptors isolated from infected and uninfected cells. Receptors were isolated from 500 ml ofE. coli B grown

in M9 (5) medium containing 0.05% yeast extract, 0.5% glucose, and 5 x 10-4 M CaCl2 to 7 x 108 cells per ml. If infection is to take place, phage are added at a multiplicity of 10 phagelbacterium and the cells are incubated for the indicated times. The infection is stopped by adding 50 fg of chloramphenicol per ml of culture and chilling on ice. The cells are centrifuged for 15 min at 5,000 rpm (3,000 x g), washed once with M9, and suspended in 4 ml of distilled water at room temperature. Two milliliters of 0.1 M NaOH is added dropwise with stirring. C02 is bubbled through the suspension until the pH is between 6 and 7. The cells and debris are removed by centrifugation at 16,000 rpm (30,000 x g) for 1 h. The supernatant from this centrifugation contains the T5 receptors; NaN3 is added to it to give a final concentration of 0.03 M. Receptors are assayed by mixing 0.5 ml of receptor suspension with 0.5 ml of a phage suspension containing 105 T5 (wild type) phage/ml. This mixture is incubated at 370C in an incubator shaker. Samples are removed at 0, 15, 30, and 60 min and assayed for the number of T5 PFU by plating (using the agar overlay method [1]) on E. coli B. Symbols: (0) Receptors isolated from uninfected cells; (*) receptors isolated from cells infected for 15 min with T5H23c; (0) receptors isolated from cells infected for 30 min with T5H23c; (A) receptors isolated from cells infected for 30 min with T5am231. PI P0 is the number ofPFU found at the indicated time divided by the number of PFU at zero time.

T5H23c for 30 min caused the rate of adsorption of labeled phage to these cells to be reduced by 90%. If phage were added to cells pretreated with 100 mg of chloramphenicol per ml of culture, the receptors were not inactivated. This was shown both by isolation of active receptors from cells infected for 30 min with T5H23c in the presence of chloramphenicol and with the labeled-phage assay (data not shown). Fully active receptors could be isolated after infection for 30 min by T5am231 (Fig. 1). This phage contains an amber mutation in the A2 gene and can, hence, inject only 8% of its DNA (6, 7). This shows that the receptors are not inactivated by attachment of the incoming

phage and that the inactivation requires, at least, the whole DNA molecule be injected. T5 107a, which has an amber mutation in gene D5 and cannot synthesize any DNA (3) or late proteins (McCorquodale, personal communication) when it infects E. coli B, caused inactivation of 80% of the receptors on the cells it infected, as judged by the number of receptors that could be isolated at 30 min postinfection (data not shown). It is thus unlikely that the receptors were being bound by free phage tails that might be made in excess. If T5H23c-infected cells were mixed with an equal volume of uninfected cells and receptors were isolated from this mixture, the activity of the receptor preparation was intermediate between the activity of the receptors isolated from each culture alone. It appears, therefore, that the inactivation of the receptors does not proceed through an extracellular intermediate and suggests that something is synthesized in the infected cells that either changes the structure of or binds to the T5 receptors, making them unavailable for binding the newly synthesized T5. This hypothetical substance could either irreversibly bind to the receptors, preventing T5 adsorption and inactivation, or reversibly bind, competing with T5 for the receptors and effectively lowering the adsorption and inactivation rate. Studies to differentiate between these possibilities are in progress. It is also possible that the receptors are released into the growth medium after infection and are not inactivated at all. We investigated this possibility by looking for phage inactivation by the medium in which the cells were grown. Although we did find some inactivation of T5 by growth medium removed from infected cells, we found the same amount of inactivation produced by medium removed from uninfected cells. It appears, therefore, that both normal and T5-infected cells release something into the medium that can inactivate T5 but that infected cells do not release more of this substance than uninfected cells. Furthermore, the amount of this substance was less than 5% of the phage-inactivating ability of whole cells. We do not feel, therefore, that the infection causes the release of the phage receptors. We do not know at this time whether the inactivating substance found in the growth medium is the T5 receptor or some other substance. This, as well as the mechanism of inactivation of the T5 receptors, is currently under investigation, using electron microscopy and biochemical and kinetic studies. We are indebted to D. J. McCorquodale for sending us the mutants.

NOTES

VOL. 24, 1977 This work was supported by grant BG16296 from the

National Science Foundation and by Public Health Service grant 7R01 A12056 from the National Institute of Allergy and Infectious Disease. D.H.D. is a recipient of Public Health Service development award 1 K04 A10054-01 from the Research Career Program of the National Institute of Allergy and Infectious Disease. 1. 2.

3.

4.

LITERATURE CITED Adams, M. H. 1959. Bacteriophages. Interscience Publishers, Inc., New York. Braun, V., K. Schaller, and H. Wolff. 1973. A common receptor for phage T5 and colicin M in the outer membrane of E. coli B. Biochim. Biophys. Acta 323:87-97. Chinnadurai, G., and D. J. McCorquodale. 1974. Dual role of gene D5 in the development of bacteriophage T5. Nature 247:554-657. Hendrickson, H. E., and D. J. McCorquodale. 1972. Genetic and physiological studies of bacteriophage T5. III. Patterns of deoxyribonucleic acid synthesis induced by mutants of T5 and identification of genes influencing the appearance of phage-induced dihy-

5. 6. 7.

8. 9.

10.

421

drofolate reductase and deoxyribonuclease. J. Virol. 9:981-989. Herriott, R. M., and J. L. Barlow. 1952. Preparation, purification, and properties of E. coli virus T2. J. Gen. Physiol. 36:17-28. Lanni, Y. T. 1968. First-step-transfer deoxyribonucleic acid of bacteriophage T5. Bacteriol. Rev. 32:227-242. McCorquodale, D. J. 1975. The T-odd bacteriophages. Crit. Rev. Microbiol. 4:101-159. Weidel, W., and E. Kellenberger. 1955. The E. coli B receptor for the phage T5. II. Electron microscopic studies. Biochim. Biophys. Acta 17:1-9. Weidel, W., G. Koch, and K. Bobosch. 1954. Sober die Receptor Substanz fur den Phagen T5. I. Extraktion und Reindarstellung aus E. coli B. Physikalische, chemische und funktionelle Charakterisierung. Z. Naturforsch. 9b:573-579. Zarnitz, M. L., and W. Weidel. 1963. Uber die Rezeptorsubstanz fur den Phagen T5. VI. Die Thermodynamik der Kontaktbildung zwischen Phage und Rezeptor sowie deren mogliche Bedeutung als morphogenetischer Modellmechanismus. Z. Naturforsch. 18b:276-280.