Eur. J. Biochem 77, 585-588 (1977)
Structure and Synthesis of a Lipid-Containing Bacteriophage An Endolysin Activity Associated with Bacteriophage PM2 Norihiro TSUKAGOSHI, Rolf SCHAFER, and Richard M. FRANKLIN Department of Structure Biology, Biocenter, University of Basel (Received March 2, 1977)
Endolysin was induced in Pseudomonas BAL-31 infected with bacteriophage PM2 and was also associated with the purified virion. This enzyme required divalent cations for its activity, Ca2' being the most effective cation. Endolysin activity in the virion increased up to three-fold upon disruption and the activity could be localized in the viral nucleocapsid. Thus the enzyme is localized within the virion. After purification of the structural proteins of bacteriophage PM2, only the nucleocapsid protein (111) had endolysin activity.
During maturation of the lipid-containing bacteriophage PM2, no morphologically recognizable viral components are associated with the host cell membrane [l] and late in infection the completed virions are released from the cell simultaneously with cell lysis [1,2]. . . . The present paper describes an endolysin activity which-is induced in the host cell after infection and which is also found associated with the virion. This enzymatic activity may be responsible for lysis of the host cell but may also have another function, as will be discussed.
on cellulose acetate strips (Cellogel from Chemetron, Milan, Italy) in the presence of buffer 3 (Schafer et al., unpublished). The purity of the individual proteins was controlled by gel electrophoresis and then they were assayed for endolysin activity. Isolation of the Viral Nucleocapsid The nucleocapsid of bacteriophage PM2 was prepared as described by Hinnen et al. [3]. Preparation of Radioactive Murein
MATERIALS AND METHODS Cells and Virus Growth Pseudomonas BAL-31 was grown at 25 "C in BAL broth [2]. Growth of the culture was measured by determining the absorbance at 610 nm in a Beckman spectrophotometer. Bacteriophage PM2 was grown on Pseudomonas BAL-31 in BAL broth and purified as described by Hinnen et al. [3]. Fractionation of the Viral Proteins Viral proteins were extracted from the virus with acetic acid [4]. The protein mixture was separated on a 2-m Sephadex G-75 column in the presence of buffer 2 (Schafer et al., unpublished). Three protein peaks were eluted from the column: protein IV, protein 111, and a mixture of proteins I and 11, in that order. The mixture of proteins I and I1 was separated This is paper XXV in the series.
Escherichia coli W 945 T 3282 (diaminopimelatedecarboxylase-negative mutant, given originally to Dr J. Rosenbusch of the Biocenter by Dr U. Henning) was grown at 37 "C, first in tryptone medium supplemented with 50 pg/ml diaminopimelic acid and subsequently in the following synthetic medium : 5.8 g/1 Na2HP04, 3 g/1 KH2P04, 5 g/1 NaCl and 3 g/1 NH4C1, supplemented with 1 % casamino acids, 0.4 % glucose, 1 mM MgS04, 0.1 mM CaC12, 50 pg/ml thymidine, 50 pg/ml thiamin, 50 pg/ml tryptophan, and 20 pg/ml diaminopimelic acid. The culture was diluted 1 : 100 in the same synthetic medium supplemented with 10 pg/ml of diamin~[~H]pimelic acid (300 mCi/mol, Amersham) but without added carrier diaminopimelic acid and grown to the late logarithmic phase of growth. 3H-labeled murein was prepared as described by Henning et al. [6]. The purified murein was suspended in water by a brief sonification and stored at 4 "C. The recovery of radioactivity in the murein fraction was 25 - 30 % of the total radioisotope incorporated into the cells.
586
Endolysin Activity in Bacteriophage PM2
Endolysin Assay
The standard reaction mixtures (50 p1 finalvolume) contained approximately lo5 counts/min of murein, 50 pg bovine serum albumin; samples were incubated at 25 "C for 60 min. The reaction was terminated by adding 1 ml of cold 5 % trichloracetic acid. After 30-min incubation in an ice bath, the tubes were centrifuged at 15000 rev./min for 30 min. Radioactivity was determined on 0.8-ml aliquots of the supernatants [7]. As a control used to measure the background counts, a mixture of [3H]murein and bovine serum albumin (Boehringer, Mannheim, Federal Republic of Germany) was incubated at 25 "C for 60 min, then mixed with 1 ml cold 5 % trichloracetic acid, followed by addition of the samples. Protein Assay
Protein was determined by the method of Lowry et al. [8] using bovine serum albumin as a standard. Buffers
Buffer 1 = 0.5 M NaCl; 10 mM CaClz; 25 mM Tris . HC1, pH 7.5 at 20 "C. Buffer 2 = 8 M guanidine . HCl; 2 % 2-mercaptoethanol; 50 mM phosphate, pH 7.2 at 20 "C. Buffer 3 = 8 M urea; 2 % 2-mercaptoethanol; 0.1 M Tris . HCl, pH 8 at 20 "C. RESULTS Induction of Endolysin in the Infected Culture
Endolysin activity was determined by measuring the hydrolysis of [3H]murein as described in Materials and Methods. [3H]Murein was not hydrolyzed by trypsin or bromelain at a concentration of 10 pg per assay, but by egg white lysozyme (0.05 - 0.1 pg) under the conditions used here. The rate of murein hydrolysis was linearly dependent on time up to at least 60 min. Endolysin activity was not detected at any period of growth of Pseudomonas BAL-31. In infected cultures, endolysin activity was detectable at 30-min post-infection (Fig. 1). After cell lysis, the rate of increase in endolysin activity slowed down. This is consistent with earlier studies which showed a decrease or stoppage of protein synthesis in infected cells at the time of cell lysis [2]. Association of Endolysin Activity with the Virion
Endolysin activity was found in purified virus preparations and remained associated with the virus after column chromatography on Bio-Gel A-1 Sm. This activity increased after freeze/thawing of the virus particles which was accompanied by a decrease in infectivity (Table 1). After three cycles of freeze/
0
5t
1 0
20
40 60 Time (rnin)
80
Fig. 1. Time course of endolysin induction in uninfected and infected cultures. Pseudomonas BAL-31 was grown at 25 "C in BAL broth medium to a cell density of 3 x 10' cells/ml and divided into two cultures. One culture was infected with PM2 at a multiplicity of infection of 10 and the other was used as control. The growth of Pseudomonas BAL-31 and PM2-infected cells is shown in the insert. At appropriate intervals, 2-ml aliquots of the infected (0) and uninfected (0)cultures were quickly frozen in solid COz/methanol and kept at -20 "C until all samples were collected. All samples were then thawed and sonicated for 1 min in an ice bath. 50 p1 of each sample were assayed for the endolysin activity with 50 pg bovine serum albumin and [3H]murein
Table 1. Endolysin activity in virus Purified virus was suspended at a concentration of 1 mg protein/ml of buffer 1. Aliquots were subjected to three freeze/thaw cycles (quick-frozen in solid COz/methanol and quick-thawed in a 25 "C water bath). Endolysin activity in both intact and disrupted virus was assayed as described in Materials and Methods, using 50 pg protein per assay. The activity is expressed as total radioactivity recovered in the acid-soluble fraction Virus prepn
Activity from intact virus (A)
A/B disrupted virus (B)
counts/min
12000 9 200 11 500 15100 8 700
28 200 28 200 27 400 24 300 22600
2.35 3.07 2.38 1.61 2.60
thawing, the titer of infectious virus dropped to 0.1 % of the initial titer. The increase of endolysin activit'j was also observed when intact or freeze/thawdisrupted virus was treated with the nonionic detergents Triton X-100 or Brij 58 (Table 2). Sodium dodecyl sulfate, on the other hand, inhibited the endolysin activity of the disrupted virus (Table 2). The activity inhibited by sodium dodecyl sulfate was partially restored after extensive dialysis to remove
587
N. Tsukagoshi, R. Schdfer, and R. M. Franklin Table 2. Effects of detergents and urea on viral endolysin activity (A) Intact virus in buffer 1 was mixed with Triton X-100 or Brij 58 to a final concentration of 0.5 ”/,. Endolysin activity was determined using 20 pg protein per assay. (B) Disrupted virus was prepared as described in Table 1. Furthermore a second aliquot of intact virus was mixed with an equal volume of 9 M urea made up in buffer 1 and dialyzed against buffer 1 at 4 “C prior to endolysin assay. Activity was assayed using 50 pg protein per assay. (C) Disrupted virus was prepared as described in Table 1. Endolysin activity was determined in the presence or absence of detergents using 40 pg protein per assay. An aliquot of the sample in 0.1 % sodium dodecyl sulfate (NaDodS04) was dialyzed against three changes of buffer 1 prior to assay. The activities are expressed as the total radioactivity in the acid-soluble fraction and (in parentheses) as a percentage of the activity without treatment Virus
Treatment
1
Activity counts/min (%) CaCI2 (mM)
(A) Intact
none + 0.5% Triton X-100 0.5% Brij 58
7 300 16600 14500
(B) Intact
none disrupted by freezing/thawing disrupted with 4.5 M urea
15000
+
(C) Disrupted none + 0.1 %Triton X-100 + 0.1 % Brij 58 + 0.1 % NaDodS04 + 0.1 % NaDodS04 (dialyzed)
24 300 44100 15500 40000 43 000 160 3 200
sodium dodecyl sulfate (Table 2). Furthermore, in the presence of 4.5 M urea, which has been used to isolate the viral nucleoocapsid [3], the endolysin activity associated with the virus increased (Table 2). These results all suggest that endolysin is localized within the virion.
Effects of Divalent Cations on Endolysin Activity Endolysin activity in virus disrupted with urea disappeared when such preparations were dialyzed extensively against water and was restored when Ca2+ was added to the reaction mixture. The effect of Ca2+ on the activity is shown in Fig.2. Endolysin activity was linearly dependent on Ca2 concentration from 0.1 to 1 mM; no activity was detected below 0.1 mM Ca”. Although Ca2+ was the most effective divalent cation examined, Sr2+and Ba” were almost as effective, M g + and Mn2+ ions were less effective, and Zn2 ,Cu2 and Ni2 were inhibitory (Table 3). +
+
+
+
Association of Endolysin Activity with Viral Structural Protein The individual viral proteins were isolated by a combination of column chromatography and Cellogel electrophoresis. Endolysin activity was restored when
Fig. 2. Dependence of endolysin activity on Ca2+ concentration. Virus was prepared as described in Table 3. The reaction mixture was almost the same as that in Table 3 except for the concentration of CaC12; 5 pg protein was used per assay. The activity is expressed as total radioactivity in the acid-soluble fraction. The activity in the absence of CaClz was determined in the presence of 1 mM EDTA
Table 3. Effects of divalent cations on endolysin activity Virus in buffer 1 was dissociated with 4.5 M urea as described in Table 2 B and dialyzed at 4 “C against buffer 1 and then against H20. The reaction mixture contained 10 mM Tris . HCI buffer (pH 7.0 at 20 “C), dissociated virus (10 pg protein), bovine serum albumin (50 pg), [3H]murein and divalent cations in 50 pi final volume. Ions were added as the chloride salt to a final concentration of 1 mM. EDTA was also added to a final concentration of 1 mM in the last test. Activity is expressed as total radioactivity in the acid-soluble fraction and (in parentheses) as a percentage of the activity with Ca2 Metal ion
Activity counts/min (%)
Ca2+ Zn2+ CUZ
+
Ni2 MgZ+ Ba2 Cd2 Sr2+ Mn2 None None + EDTA +
+
+
+
58 200 0 200 600 17 100 40 000 2 600 46 700 18400 2 700 0
guanidine . HCl was removed by dialysis (data not shown). There was no endolysin activity associated with proteins I, I1 or IV. Only the protein I11 fraction contained endolysin activity (Table 4). Although proteins I11 and IV were clearly separated by column chromatography according to size, we could not exclude the possible contamination of the protein I11 fraction with proteins of similar molecular weight.
588
N. Tsukagoshi, R. Schafer, and R. M. Franklin: Endolysin Activity in Bacteriophage PM2
Table 4. Endolysin activity in purifedproteins of bacteriophage PM2 Each fraction was isolated by column chromatography as described in Materials and Methods and dialyzed against 10 mM Tris . HCI buffer (pH 7.4) before the activity was determined. The reaction mixture contained 10 mM Tris . HCI buffer (pH 7.4), 50 pg bovine serum albumin, 1 mM CaCI2, [3H]murein and 4.2 pg of each fraction in 50 p1 final volume. Activity was expressed as total radioactivity in the acid-soluble fraction. The background due to endogenous degradation of murein was between 600 and 700 counts/ min Protein
Start
4
I
1
Endolysin activity in expt 1
2
counts/min
No protein (control) Proteins I and 11 Protein I11 Protein IV
693 595 17302 653
644 623 17693 593
0
o
Therefore we separated this protein fraction according to its net charge at pH 8 on Cellogel in the presence of buffer 3. In this case the endolysin activity comigrated with the band of protein I11 (Fig. 3). DISCUSSION Endolysin, which hydrolyzed the murein fraction prepared from E. coli, was induced in Pseudomonas BAL-31 infected with bacteriophage PM2 and was also associated with the virion. The enzyme required divalent cations for its activity. Ca2+ was the most effective of the various divalent cations examined but and Ba2+. Ca" is recould be replaced by S?' quired for viral replication and this requirement can also be satisfied by Sr2+ or Ba2+ [9]. In addition to those ions, Mg2+ and Mn2+ ions were partially effective replacements of Ca" with respect to endolysin activity. MnZ+ ion, however, appeared not to support viral replication in vivo [9]. The association of endolysin activity with protein 111, the nucleocapsid protein, as well as with the intact nucleocapsid, suggests a possible dual role for this enzyme. It might be involved in the penetration of the nucleocapsid into the host cell as well as in lysis of the cell at the end of a viral growth cycle. Experiments are now in progress to test this hypothesis. Technical assistance was provided by Miss E. Zoumbou and Mrs M. N. Kania. We thank Prof. A. Tsugita for his valuable discussions. This work was supported by grant 3.8530.72 SR from the Schweizerischer Nationalfonds.
l
d
Fractions
-
Fig. 3. Electrophoresis of protein III on Cellogel. Protein I11 was separated by electrophoresis on Cellogel. The solvent/buffer for Cellogel electrophoresis was 8 M urea, 2 % 2-mercaptoethanol, 1 % B-alanine, adjusted with 1 M citric acid to pH 4.5. 500 pg of pure protein 111, eluted from the Sephadex G-75 column, were dialysed against the Cellogel buffer and 80 pg of protein 111 were applied to each cellulose acetate sheet. After electrophoresis at 1000 V for 90 min, the protein band was visualized by cutting off a part of the sheet, staining and destaining [5]. The remainder of the sheet was sliced into 5-mm segments from which protein was recovered by centrifuging at 15000 rev./min for 10 min [5]. The isolated protein fractions were equilibrated against 10 mM Tris . HCI, pH 7.4, by overnight dialysis at 4 "C, and then used for endolysin assay. The endolysin activity was measured over the entire length of the sheet and is plotted below the electropherogram
REFERENCES 1. Dahlberg, J. E. & Franklin, R. M. (1970) Virology, 42, 10731086. 2. Franklin, R. M., Salditt, M. & Silbert, J. A. (1969) Virology, 38, 627 - 640. 3. Hinnen, R., Schafer, R. & Franklin, R. M. (1974) Eur. J . Biochem. 50,l- 14. 4. Hinnen, R., Chassin, R., Schafer, R., Franklin, R. M., Hitz, H. & Schafer, D. (1976) Eur. J. Biochem. 68, 139- 152. . 5. Schafer, R., Hinnen, R. & Franklin, R. M. (1974) Eur. J . Biochem. 50,15 - 27. 6. Hennig, V., Rehn, K., Brdun, V. & Hohn, B. (1972) Eur. J . Biochem. 26,570- 586. 7. Tsukagoshi, N., Petersen, M. M. & Franklin, R. M. (1975) Eur. J. Biochem. 60, 603-613. 8. Lowry, 0. H., Rosebrough, N. F., Farr, A. L. & Randall, R. F. (1951) J . Biol. Chem. 193, 265-275. 9. Snipes, W., Cupp, J., Sands, J. A,, Keith, A. & Davis, A. (1974) Biochim. Biophys, Acta, 339, 311 - 322.
N. Tsukagoshi, R . Schafer, and R. M. Franklin*, Abteilung fur Strukturbiologie, Biozentrum der Universitat Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland
* To whom all correspondence
should be addressed
Structure and Synthesis of a Lipid-Containing Bacteriophage An Endolysin Activity Associated with Bacteriophage PM2 Norihiro TSUKAGOSHI, Rolf SCHAFER, and Richard M. FRANKLIN Department of Structure Biology, Biocenter, University of Basel (Received March 2, 1977)
Endolysin was induced in Pseudomonas BAL-31 infected with bacteriophage PM2 and was also associated with the purified virion. This enzyme required divalent cations for its activity, Ca2' being the most effective cation. Endolysin activity in the virion increased up to three-fold upon disruption and the activity could be localized in the viral nucleocapsid. Thus the enzyme is localized within the virion. After purification of the structural proteins of bacteriophage PM2, only the nucleocapsid protein (111) had endolysin activity.
During maturation of the lipid-containing bacteriophage PM2, no morphologically recognizable viral components are associated with the host cell membrane [l] and late in infection the completed virions are released from the cell simultaneously with cell lysis [1,2]. . . . The present paper describes an endolysin activity which-is induced in the host cell after infection and which is also found associated with the virion. This enzymatic activity may be responsible for lysis of the host cell but may also have another function, as will be discussed.
on cellulose acetate strips (Cellogel from Chemetron, Milan, Italy) in the presence of buffer 3 (Schafer et al., unpublished). The purity of the individual proteins was controlled by gel electrophoresis and then they were assayed for endolysin activity. Isolation of the Viral Nucleocapsid The nucleocapsid of bacteriophage PM2 was prepared as described by Hinnen et al. [3]. Preparation of Radioactive Murein
MATERIALS AND METHODS Cells and Virus Growth Pseudomonas BAL-31 was grown at 25 "C in BAL broth [2]. Growth of the culture was measured by determining the absorbance at 610 nm in a Beckman spectrophotometer. Bacteriophage PM2 was grown on Pseudomonas BAL-31 in BAL broth and purified as described by Hinnen et al. [3]. Fractionation of the Viral Proteins Viral proteins were extracted from the virus with acetic acid [4]. The protein mixture was separated on a 2-m Sephadex G-75 column in the presence of buffer 2 (Schafer et al., unpublished). Three protein peaks were eluted from the column: protein IV, protein 111, and a mixture of proteins I and 11, in that order. The mixture of proteins I and I1 was separated This is paper XXV in the series.
Escherichia coli W 945 T 3282 (diaminopimelatedecarboxylase-negative mutant, given originally to Dr J. Rosenbusch of the Biocenter by Dr U. Henning) was grown at 37 "C, first in tryptone medium supplemented with 50 pg/ml diaminopimelic acid and subsequently in the following synthetic medium : 5.8 g/1 Na2HP04, 3 g/1 KH2P04, 5 g/1 NaCl and 3 g/1 NH4C1, supplemented with 1 % casamino acids, 0.4 % glucose, 1 mM MgS04, 0.1 mM CaC12, 50 pg/ml thymidine, 50 pg/ml thiamin, 50 pg/ml tryptophan, and 20 pg/ml diaminopimelic acid. The culture was diluted 1 : 100 in the same synthetic medium supplemented with 10 pg/ml of diamin~[~H]pimelic acid (300 mCi/mol, Amersham) but without added carrier diaminopimelic acid and grown to the late logarithmic phase of growth. 3H-labeled murein was prepared as described by Henning et al. [6]. The purified murein was suspended in water by a brief sonification and stored at 4 "C. The recovery of radioactivity in the murein fraction was 25 - 30 % of the total radioisotope incorporated into the cells.
586
Endolysin Activity in Bacteriophage PM2
Endolysin Assay
The standard reaction mixtures (50 p1 finalvolume) contained approximately lo5 counts/min of murein, 50 pg bovine serum albumin; samples were incubated at 25 "C for 60 min. The reaction was terminated by adding 1 ml of cold 5 % trichloracetic acid. After 30-min incubation in an ice bath, the tubes were centrifuged at 15000 rev./min for 30 min. Radioactivity was determined on 0.8-ml aliquots of the supernatants [7]. As a control used to measure the background counts, a mixture of [3H]murein and bovine serum albumin (Boehringer, Mannheim, Federal Republic of Germany) was incubated at 25 "C for 60 min, then mixed with 1 ml cold 5 % trichloracetic acid, followed by addition of the samples. Protein Assay
Protein was determined by the method of Lowry et al. [8] using bovine serum albumin as a standard. Buffers
Buffer 1 = 0.5 M NaCl; 10 mM CaClz; 25 mM Tris . HC1, pH 7.5 at 20 "C. Buffer 2 = 8 M guanidine . HCl; 2 % 2-mercaptoethanol; 50 mM phosphate, pH 7.2 at 20 "C. Buffer 3 = 8 M urea; 2 % 2-mercaptoethanol; 0.1 M Tris . HCl, pH 8 at 20 "C. RESULTS Induction of Endolysin in the Infected Culture
Endolysin activity was determined by measuring the hydrolysis of [3H]murein as described in Materials and Methods. [3H]Murein was not hydrolyzed by trypsin or bromelain at a concentration of 10 pg per assay, but by egg white lysozyme (0.05 - 0.1 pg) under the conditions used here. The rate of murein hydrolysis was linearly dependent on time up to at least 60 min. Endolysin activity was not detected at any period of growth of Pseudomonas BAL-31. In infected cultures, endolysin activity was detectable at 30-min post-infection (Fig. 1). After cell lysis, the rate of increase in endolysin activity slowed down. This is consistent with earlier studies which showed a decrease or stoppage of protein synthesis in infected cells at the time of cell lysis [2]. Association of Endolysin Activity with the Virion
Endolysin activity was found in purified virus preparations and remained associated with the virus after column chromatography on Bio-Gel A-1 Sm. This activity increased after freeze/thawing of the virus particles which was accompanied by a decrease in infectivity (Table 1). After three cycles of freeze/
0
5t
1 0
20
40 60 Time (rnin)
80
Fig. 1. Time course of endolysin induction in uninfected and infected cultures. Pseudomonas BAL-31 was grown at 25 "C in BAL broth medium to a cell density of 3 x 10' cells/ml and divided into two cultures. One culture was infected with PM2 at a multiplicity of infection of 10 and the other was used as control. The growth of Pseudomonas BAL-31 and PM2-infected cells is shown in the insert. At appropriate intervals, 2-ml aliquots of the infected (0) and uninfected (0)cultures were quickly frozen in solid COz/methanol and kept at -20 "C until all samples were collected. All samples were then thawed and sonicated for 1 min in an ice bath. 50 p1 of each sample were assayed for the endolysin activity with 50 pg bovine serum albumin and [3H]murein
Table 1. Endolysin activity in virus Purified virus was suspended at a concentration of 1 mg protein/ml of buffer 1. Aliquots were subjected to three freeze/thaw cycles (quick-frozen in solid COz/methanol and quick-thawed in a 25 "C water bath). Endolysin activity in both intact and disrupted virus was assayed as described in Materials and Methods, using 50 pg protein per assay. The activity is expressed as total radioactivity recovered in the acid-soluble fraction Virus prepn
Activity from intact virus (A)
A/B disrupted virus (B)
counts/min
12000 9 200 11 500 15100 8 700
28 200 28 200 27 400 24 300 22600
2.35 3.07 2.38 1.61 2.60
thawing, the titer of infectious virus dropped to 0.1 % of the initial titer. The increase of endolysin activit'j was also observed when intact or freeze/thawdisrupted virus was treated with the nonionic detergents Triton X-100 or Brij 58 (Table 2). Sodium dodecyl sulfate, on the other hand, inhibited the endolysin activity of the disrupted virus (Table 2). The activity inhibited by sodium dodecyl sulfate was partially restored after extensive dialysis to remove
587
N. Tsukagoshi, R. Schdfer, and R. M. Franklin Table 2. Effects of detergents and urea on viral endolysin activity (A) Intact virus in buffer 1 was mixed with Triton X-100 or Brij 58 to a final concentration of 0.5 ”/,. Endolysin activity was determined using 20 pg protein per assay. (B) Disrupted virus was prepared as described in Table 1. Furthermore a second aliquot of intact virus was mixed with an equal volume of 9 M urea made up in buffer 1 and dialyzed against buffer 1 at 4 “C prior to endolysin assay. Activity was assayed using 50 pg protein per assay. (C) Disrupted virus was prepared as described in Table 1. Endolysin activity was determined in the presence or absence of detergents using 40 pg protein per assay. An aliquot of the sample in 0.1 % sodium dodecyl sulfate (NaDodS04) was dialyzed against three changes of buffer 1 prior to assay. The activities are expressed as the total radioactivity in the acid-soluble fraction and (in parentheses) as a percentage of the activity without treatment Virus
Treatment
1
Activity counts/min (%) CaCI2 (mM)
(A) Intact
none + 0.5% Triton X-100 0.5% Brij 58
7 300 16600 14500
(B) Intact
none disrupted by freezing/thawing disrupted with 4.5 M urea
15000
+
(C) Disrupted none + 0.1 %Triton X-100 + 0.1 % Brij 58 + 0.1 % NaDodS04 + 0.1 % NaDodS04 (dialyzed)
24 300 44100 15500 40000 43 000 160 3 200
sodium dodecyl sulfate (Table 2). Furthermore, in the presence of 4.5 M urea, which has been used to isolate the viral nucleoocapsid [3], the endolysin activity associated with the virus increased (Table 2). These results all suggest that endolysin is localized within the virion.
Effects of Divalent Cations on Endolysin Activity Endolysin activity in virus disrupted with urea disappeared when such preparations were dialyzed extensively against water and was restored when Ca2+ was added to the reaction mixture. The effect of Ca2+ on the activity is shown in Fig.2. Endolysin activity was linearly dependent on Ca2 concentration from 0.1 to 1 mM; no activity was detected below 0.1 mM Ca”. Although Ca2+ was the most effective divalent cation examined, Sr2+and Ba” were almost as effective, M g + and Mn2+ ions were less effective, and Zn2 ,Cu2 and Ni2 were inhibitory (Table 3). +
+
+
+
Association of Endolysin Activity with Viral Structural Protein The individual viral proteins were isolated by a combination of column chromatography and Cellogel electrophoresis. Endolysin activity was restored when
Fig. 2. Dependence of endolysin activity on Ca2+ concentration. Virus was prepared as described in Table 3. The reaction mixture was almost the same as that in Table 3 except for the concentration of CaC12; 5 pg protein was used per assay. The activity is expressed as total radioactivity in the acid-soluble fraction. The activity in the absence of CaClz was determined in the presence of 1 mM EDTA
Table 3. Effects of divalent cations on endolysin activity Virus in buffer 1 was dissociated with 4.5 M urea as described in Table 2 B and dialyzed at 4 “C against buffer 1 and then against H20. The reaction mixture contained 10 mM Tris . HCI buffer (pH 7.0 at 20 “C), dissociated virus (10 pg protein), bovine serum albumin (50 pg), [3H]murein and divalent cations in 50 pi final volume. Ions were added as the chloride salt to a final concentration of 1 mM. EDTA was also added to a final concentration of 1 mM in the last test. Activity is expressed as total radioactivity in the acid-soluble fraction and (in parentheses) as a percentage of the activity with Ca2 Metal ion
Activity counts/min (%)
Ca2+ Zn2+ CUZ
+
Ni2 MgZ+ Ba2 Cd2 Sr2+ Mn2 None None + EDTA +
+
+
+
58 200 0 200 600 17 100 40 000 2 600 46 700 18400 2 700 0
guanidine . HCl was removed by dialysis (data not shown). There was no endolysin activity associated with proteins I, I1 or IV. Only the protein I11 fraction contained endolysin activity (Table 4). Although proteins I11 and IV were clearly separated by column chromatography according to size, we could not exclude the possible contamination of the protein I11 fraction with proteins of similar molecular weight.
588
N. Tsukagoshi, R. Schafer, and R. M. Franklin: Endolysin Activity in Bacteriophage PM2
Table 4. Endolysin activity in purifedproteins of bacteriophage PM2 Each fraction was isolated by column chromatography as described in Materials and Methods and dialyzed against 10 mM Tris . HCI buffer (pH 7.4) before the activity was determined. The reaction mixture contained 10 mM Tris . HCI buffer (pH 7.4), 50 pg bovine serum albumin, 1 mM CaCI2, [3H]murein and 4.2 pg of each fraction in 50 p1 final volume. Activity was expressed as total radioactivity in the acid-soluble fraction. The background due to endogenous degradation of murein was between 600 and 700 counts/ min Protein
Start
4
I
1
Endolysin activity in expt 1
2
counts/min
No protein (control) Proteins I and 11 Protein I11 Protein IV
693 595 17302 653
644 623 17693 593
0
o
Therefore we separated this protein fraction according to its net charge at pH 8 on Cellogel in the presence of buffer 3. In this case the endolysin activity comigrated with the band of protein I11 (Fig. 3). DISCUSSION Endolysin, which hydrolyzed the murein fraction prepared from E. coli, was induced in Pseudomonas BAL-31 infected with bacteriophage PM2 and was also associated with the virion. The enzyme required divalent cations for its activity. Ca2+ was the most effective of the various divalent cations examined but and Ba2+. Ca" is recould be replaced by S?' quired for viral replication and this requirement can also be satisfied by Sr2+ or Ba2+ [9]. In addition to those ions, Mg2+ and Mn2+ ions were partially effective replacements of Ca" with respect to endolysin activity. MnZ+ ion, however, appeared not to support viral replication in vivo [9]. The association of endolysin activity with protein 111, the nucleocapsid protein, as well as with the intact nucleocapsid, suggests a possible dual role for this enzyme. It might be involved in the penetration of the nucleocapsid into the host cell as well as in lysis of the cell at the end of a viral growth cycle. Experiments are now in progress to test this hypothesis. Technical assistance was provided by Miss E. Zoumbou and Mrs M. N. Kania. We thank Prof. A. Tsugita for his valuable discussions. This work was supported by grant 3.8530.72 SR from the Schweizerischer Nationalfonds.
l
d
Fractions
-
Fig. 3. Electrophoresis of protein III on Cellogel. Protein I11 was separated by electrophoresis on Cellogel. The solvent/buffer for Cellogel electrophoresis was 8 M urea, 2 % 2-mercaptoethanol, 1 % B-alanine, adjusted with 1 M citric acid to pH 4.5. 500 pg of pure protein 111, eluted from the Sephadex G-75 column, were dialysed against the Cellogel buffer and 80 pg of protein 111 were applied to each cellulose acetate sheet. After electrophoresis at 1000 V for 90 min, the protein band was visualized by cutting off a part of the sheet, staining and destaining [5]. The remainder of the sheet was sliced into 5-mm segments from which protein was recovered by centrifuging at 15000 rev./min for 10 min [5]. The isolated protein fractions were equilibrated against 10 mM Tris . HCI, pH 7.4, by overnight dialysis at 4 "C, and then used for endolysin assay. The endolysin activity was measured over the entire length of the sheet and is plotted below the electropherogram
REFERENCES 1. Dahlberg, J. E. & Franklin, R. M. (1970) Virology, 42, 10731086. 2. Franklin, R. M., Salditt, M. & Silbert, J. A. (1969) Virology, 38, 627 - 640. 3. Hinnen, R., Schafer, R. & Franklin, R. M. (1974) Eur. J . Biochem. 50,l- 14. 4. Hinnen, R., Chassin, R., Schafer, R., Franklin, R. M., Hitz, H. & Schafer, D. (1976) Eur. J. Biochem. 68, 139- 152. . 5. Schafer, R., Hinnen, R. & Franklin, R. M. (1974) Eur. J . Biochem. 50,15 - 27. 6. Hennig, V., Rehn, K., Brdun, V. & Hohn, B. (1972) Eur. J . Biochem. 26,570- 586. 7. Tsukagoshi, N., Petersen, M. M. & Franklin, R. M. (1975) Eur. J. Biochem. 60, 603-613. 8. Lowry, 0. H., Rosebrough, N. F., Farr, A. L. & Randall, R. F. (1951) J . Biol. Chem. 193, 265-275. 9. Snipes, W., Cupp, J., Sands, J. A,, Keith, A. & Davis, A. (1974) Biochim. Biophys, Acta, 339, 311 - 322.
N. Tsukagoshi, R . Schafer, and R. M. Franklin*, Abteilung fur Strukturbiologie, Biozentrum der Universitat Basel, Klingelbergstrasse 70, CH-4056 Basel, Switzerland
* To whom all correspondence
should be addressed
