THE JOURNAL OF BKX.OCICAL CHEMISTRY Vol. 251, No. 10, Issue of May 25, pp. 3185-3190, 1976

Prmted m U.S.A.

Characterization of a Virion Protein Kinase as a Virus-specified Enzyme*

(Received for publication, August 28, 1975)

HARVEY SILBERSTEIN AND J. THOMAS AUGUST

From the Department of Molecular Biology, Diuision of Biological Sciences, Albert Einstein College of

Medicine, Bronx, New York 10461

Antibodies which completely inhibited the enzymatic activity of the protein kinase associated with

virions of frog virus were obtained by immunization of rabbits with the purified enzyme. This inhibition

provided a specific probe for the frog virus protein kinase, since this antiserum had no inhibitory effect on

a variety of other protein kinases, including the activity of uninfected cells, or the protein kinase

associated with vesicular stomatitis virus or vaccinia viruscultivated in the same cell line as frog virus.

The frog virus protein kinase was characterized as a virus-specified protein on the basis of the following

observations: (a) the virion protein kinase was antigenically distinct from essentially all of the protein

kinase expressed in uninfected cells; (6) following infection by frog virus more than a 15-fold increase was

detected in the specific activity of intracellular protein kinase and most of this activity was antigenically

related to the virion enzyme; (c) when frog virus was grown in cells derived from widely different species,

the antigenic and biochemical specificities of the virion protein kinase remained identical; and (cl)

screening of cells infected with different temperature-sensitive mutants of frog virus indicated that

certain viral mutants failed to synthesize this protein kinase when cultivated at the nonpermissive

temperature.

Virions of a variety of enveloped animal viruses contain a

protein kinase activity (EC 2.7.1.37; ATP:protein phospho-

transferase) which catalyzes the phosphorylation of specific

virion polypeptides as well as several nonviral proteins (for a

review see Ref. 1). In the accompanying manuscript (I) we

described the purification and biochemical characterization of

a protein kinase associated with virions of frog virus 3. In the

present study we present immunochemical evidence indicating

that the frog virus protein kinase is a viral gene product. As

such, this protein kinase may play an important role in the

replication of frog virus.

EXPERIMENTAL PROCEDURE

Materials

All chemicals were reagent or enzyme grade. Materials, including the Sephadex G-200 fraction of frog virus protein kinase, were obtained as described (1). Heat-inactivated calf serum, Medium 199, and phosphate-buffered saline (2) were purchased from the Grand Island Biological Co.; Freund incomplete and Freund complete adjuvant from Beckton, Dickinson and Co.; chymotrypsinogen from the Sigma Chemical Co.

Methods

Growth of Cells-Minnow cells, ATCC CCL 42, were grown as described (1). BHK 21/13 cells (4), ATCC CCL 10, were grown at 37” as

* This study, conducted under Contract CP-33311 within the Virus Cancer Program of the National Cancer Institute, National Institutes of Health, was supported in part by Grant GM 11301 and 5T5-GM 1674 from the National Institutes of Health. A preliminary report of these studies was presented at the Annual Meeting of the American Society for Microbiology, 1975.

monolayer cultures as described previously (5), except that heat-inac- tivated calf serum replaced fetal calf serum. Cultures of chick embryo fibroblasts were prepared from 12.day-old embryos as described by Vogt (6) and grown at 37” as monolayers in Medium 199 containing 10% calf serum, 10% tryptose phosphate broth, and organic buffers as described for minnow cells (1). Secondary cultures were used to support the growth of frog virus.

Growth and Purification of Frog Virus-The frog virus 3 isolate of polyhedral cytoplasmic deoxyvirus (7), ATCC VR-567, was cultivated and titered in minnow cells, and the cell-associated virus purified by velocity sedimentation as described (1, 5). The virus was further purified by isopycnic banding on linear gradients of 40 to 80% (w/v) sucro.se in Tris/NaCl/EDTA’ in the Spinco SW 25.2 rotor (20,000 rpm; 18 hours). The major band of virus, located about two-thirds of the way down the tube, was harvested by puncturing the bottom of the cen- trifuge tube, diluted with 4 volumes of Tris/NaCl/EDTA, pelleted by centrifugation (100,000 x g; 45 min), suspended in Tris/NaCl/EDTA plus 10% glycerol by brief ultrasonic vibration, and stored at -20“. The overall procedure produced virus with a specific infectivity of 2 to 5 x lo* plaque-forming unit&g of protein, with a yield of 20 to 30% of the initial plaque-forming units.

Stocks of frog virus were prepared in BHK cells, and secondary chick embryo fibroblasts as described for minnow cells (5). The virus was serially passaged a minimum of two times in each of these cells lines. Frog virus was purified from infected BHK and chick cells as described above.

The isolation of several of the temperature-sensitive mutants of frog virus 3 has been described (8). Minnow cells infected with various temperature-sensitive mutants of frog virus were generously supplied by Allan Granoff and Rakesh Goorha, St. Jude’s Children’s Hospital. Briefly, duplicate roller bottle cultures containing 4 x lOa minnow cells

‘Tris/NaCl/EDTA is 20 rnM Tris-HCl, pH 7.6, 100 rn~ NaCl, and 1 rn~ EDTA.

3185

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3186 Virus-specified Protein Kinase

were infected with 1 plaque-forming unit/cell of wild type or mutant frog virus. After 45 min for virus adsorption, the inoculum was replaced with fresh medium, and the incubation continued for 18 hours at 25” (permissive temperature) or 30” (nonpermissive temperature). In- fected cells were collected with the aid of a rubber spatula, concen- trated by centrifugation, and washed twice by gentle suspension in phosphate-buffered saline, pH 7.2, and centrifugation. Cell pellets were frozen at 20” and shipped to this laboratory on dry ice.

Growth and Purification of Vesicular Stomatitis Virus-Vesicular stomatitis virus (Indiana strain), grown in HeLa cells, was obtained from Paul Atkinson of this institution. The virus was passaged at 37” in monolayers of BHK 21/13 cells by infection at a multiplicity of 6 plaque-forming units/cell. Eighteen hours after infection the virus containing growth medium from identical cultures was pooled and cellular debris removed by centrifugation (2000 x g; 5 min). Vesicular stomatitis virus was then purified essentially as described by Huang et al. (9). All steps were performed at 4”. Briefly, virus was collected by centrifugation in the Spinco type 19 rotor (18,500 rpm; 2.5 hours), suspended in Tris/NaCl/EDTA to 2 mg of protein/ml by gentle ultrasonic vibration, and centrifuged on linear gradients of 10 to 50% (w/v) sucrose in Tris/NaCl/EDTA in the Spinco SW 25.2 rotor (25,000 rpm; 90 min). The band of infectious B particles (9) was collected, diluted with 4 volumes of Tris/NaCl/EDTA, pelleted by centrifugation (100,000 x g; 90 min), suspended in Tris/NaCl/EDTA plus 10% glycerol, and stored at -20”. The purity of the virus was monitored by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate as shown in Fig. 4.

Growth and Purification of Vuccinia Virus-Vaccinia virus (WR strain), grown in HeLa cells, was obtained from Joseph Kates, State University of New York at Stony Brook. The virus was passaged at 37” in monolayers of BHK 21/13 cells by infection at a multiplicity of 10 plaque-forming units/cell. Forty hours after infection when most of the cells showed a marked cytopathic effect, cells were harvested with the aid of a rubber spatula and concentrated by centrifugation (4000 x g; 10 min). The cell-associated virus was then purified as follows. All steps were at 4”. Infected cells were suspended at lOa cells/ml in Tris/NaCl/EDTA plus 10% glycerol by sonic oscillation with a Branson model W 185 probe sonifier (100 watts; 1 min), centrifuged (2ooO x g; 10 min), and the supernatant containing released virus collected. The pellet was suspended in one-third the original volume of Tris/NaCl/ EDTA plus 10% glycerol by sonication as above, and centrifugation repeated. The supernatant containing additional virus was pooled with the first supernatant, and vaccinia virus was then pelleted by centrifugation through a cushion of 36% (w/v) sucrose in Tris/NaCl/ EDTA and purified by velocity sedimentation on linear gradients of 25 to 40% (w/v) sucrose in Tris/NaCl/EDTA, following the procedures outlined by Joklik (10). The pooled virus from the sucrose gradients was diluted 4.fold with Tris/NaCl/EDTA, pelleted by centrifugation (100,000 x g; 45 min), suspended in Tris/NaCl/EDTA plus 10% glycerol, and stored at -20”. The purity of the virus was monitored by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate as shown in Fig. 4.

Preparation of Cell Extracts-Uninfected or frog virus-infected minnow cells were harvested with the aid of a rubber spatula, concentrated by centrifugation (2000 x g; 10 min), and washed by gentle suspension in cold Tris/NaCI/EDTA and centrifugation. Con- trol studies indicated that this method of harvesting the cells resulted in a minimal loss of cellular protein kinase, as with each sample less than 5% of the total activity was recovered in the supernatants. Protein kinase was solubilized by suspending the cells at 108/ml in Tris/NaCl/ EDTA plus 10% glycerol by sonic oscillation with a Branson model W 185 probe sonifier (100 watts; 1 min; 4”). Duplicate aliquots were removed to determine the total cellular protein content, after which KC1 and Triton X-100 were added to final concentrations of 1.0 M and 1.5% (v/v), respectively. After centrifugation (100,000 x g; 1 hour), the supernatants were collected and the pellets suspended by sonication as above in one-half the original volume of a solution of 10% glycerol, 1.0 M KCl, 1.5% Triton X-100, and 20 mM Tris-HCl, pH 8.0. The suspended pellets were centrifuged as above, the supernatants col- lected and pooled with the first supernatant, and dialyzed for 15 hours in 2 liters of 10% glycerol, 500 rnM KCI, 0.3% Triton X-100, and 20 rnM Tris-HCl, pH 8.0 (100,000 x g soluble fraction). This procedure solubilized greater than 95% of the total protein kinase activity of cells.

Preparation of Rabbit Anti-Frog Virus Protein Kinase Serum-Rab- bit anti-frog virus protein kinase serum (immune serum) was obtained from animals injected with the Sephadex G-200 fraction of the frog

virus protein kinase. For primary injections, 12 Kg of protein kinase were emulsified with an equal volume of Freund complete adjuvant and injected into two footpads and multiple intradermal sites. Booster injections of 8 pg of protein kinase emulsified in an equal volume of Freund incomplete adjuvant were given intramuscularly at 3-week intervals. Rabbits were bled through an ear vein prior to booster injections. The antibody fraction of rabbit serum was precipitated at 4” by addition of ammonium sulfate to 38% saturation (21.3 g/100 ml), collected by centrifugation (20,000 x g; 20 min), suspended in one-fifth the original volume of Tris/NaCl/EDTA and dialyzed overnight in 2 liters of Tris/NaCl/EDTA. Aliquots of this fraction were stored at - 20” and thawed prior to use.

Measurement of Protein-The protein content of purified viruses and cell extracts was determined by a modification of the Lowry method (ll), using bovine serum albumin as a standard. The protein concentration of the antibody fraction of rabbit serum was calculated from absorption measurements assuming an extinction of 1.4 A,.,,,/ mg of protein. The protein concentration of the Sephadex G-200 fraction of protein kinase was calculated from the specific radioactivity as described (1).

RESULTS

Antibody-mediated Inhibition of Frog Virus Protein Kinase

-Antibodies against purified frog virus protein kinase were

prepared as described under “Methods.” Incubation of the

purified protein kinase with increasing quantities of the

antibody fraction from an immune serum sample achieved

nearly 100% inhibition of enzymatic activity, while equivalent quantities of a similar fraction from pre-immune serum had no

100 200 0 15 30 45 p9 ANTISODY MIN OF INC”SATTION AT 37’

FIG. 1 (left). Inhibition of protein kinase by anti-frog virus protein kinase antibody. The l-y1 aliquots (7 ng) of the Sephadex G-200 fraction of frog virus protein kinase were added to reaction mixtures of 40 ~1 containing 2 mg/ml of bovine serum albumin, 20 rnM Tris-HCl, pH 8.0, 100 mM NaCl, and the indicated quantity of immune or pre-immune antibody. After incubation at 37” for 30 min, protein kinase activity of each sample was measured by the addition of 0.20 ml of a reaction mixture containing 25 mM Tris-HCl, pH 8.0, 20 rnM MgCl,, 5 mg/ml of casein, and 0.5 rnM [Y~*P]ATP (19.1 cpm/pmol) and the incubation continued at 30” for 10 min. Reactions were terminated by the addition of 10 ~1 of 2% Triton X-100 followed immediately by 0.50 ml of cold 12% trichloroacetic acid containing 30 mM sodium pyrophosphate, and processed for “zP incorporation as described for the protein kinase assay (1). The control value, representing the incorpora- tion in the absence of added antibody, was 1010 pmol/lO min. Values for each point indicate the mean of duplicate samples. O-O, immune; O---O, pre-immune.

FIG. 2 (right). Kinetics of protein kinase inactivation by antibody. The 1.~1 aliquots (7 ng) of the Sephadex G-200 fraction of frog virus protein kinase were added to reaction mixtures of 40 ~1, as described in Fig. 1, containing immune or pre-immune antibody as indicated below. After incubation at 37” for the time period indicated on the figure, protein kinase activity of each sample was measured as described in Fig. 1. Control samples containing 215 pg of pre-immune antibody were run for each time point indicated. Minimal thermal inactivation of the enzyme occurred during the incubation at 37”, as the actual incorpora- tion for the controls varied between 1220 and 1100 pmol for the 0- and 45-min time points, respectively. Values for each point indicate the mean of duplicate samples. O-O, 57 fig of immune antibody fraction; 04,114 pg of immune antibody fraction; A-A, 228 pg of immune antibody fraction.

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Virus-specified Protein Kinase 3187

inhibitory effect (Fig. 1). A study of the kinetics of antibody-

mediated inhibition of the protein kinase indicated that

maximum inhibition was achieved after a 30-min preincuba-

tion of antibody with the enzyme (Fig. 2). As indicated by the

levels of incorporation for the controls in this experiment (Fig.

2) the protein kinase demonstrated less than a 10% loss of

activity as a result of thermal inactivation during the preincu-

bation, due to the addition of a stabilizing protein (bovine

serum albumin) to the reaction mixture. Addition of as much

as 0.2% Triton X-100 and/or 50 mM KC1 (final concentrations)

to the preincubation reaction did not significantly affect the

final extent of enzyme inhibition. This result was important for

studies reported below, where small amounts of Triton X-100

and KC1 were introduced into this reaction as components of

the enzyme solution. All experiments reported in this study

were carried out using a single immune serum sample, al- though additional bleedings from each of two immunized

rabbits contained antibodies of approximately equivalent in-

hibitory titer.

Measurement of Protein Kinase in Uninfected and Frog

Virus-infected Cells-A previous characterization of the puri-

fied frog virus protein kinase indicated that the combination of

casein and ATP achieved the maximal levels of enzyme

activity in vitro (1). Therefore, these substrates were employed

to examine the specific activity of protein kinase in crude

extracts from uninfected and frog virus-infected minnow cells

(Table I). Although there was a background of activity in

uninfected cells, more than a 15fold increase in the specific

activity of protein kinase was observed in infected cultures. In

control studies, this elevated level of enzyme activity was

readily apparent in an extract prepared from a mixture of both

infected and uninfected cells, indicating that there was no

selective inhibition of protein kinase activity by uninfected

cells which might have interfered in these calculations. Cells

harvested very early after infection (30 min) demonstrated

levels of protein kinase similar to those of uninfected cultures

indicating that the elevated levels of activity observed later in

the infection were not due to the protein kinase of input virus. Immunochemical Characterization of Protein Kinase from

Uninfected and Frog Virus-infected Cells-The nature of the

protein kinase activity of uninfected and infected cells was

investigated using the antibodies prepared against the virion

enzyme (Fig. 3). Antiserum had no inhibitory affect on the

protein kinase from uninfected cells, demonstrating that the

TABLE I

Protein kinase actiuity in infected and uninfected cells

Cells infected with 6 plaque-forming units/cell of frog virus were incubated as indicated at 28” before being harvested. Extracts were prepared as described under “Methods” from samples consisting of approximately 4 x lOa uninfected or frog virus-infected minnow cells.

Aliquots of the dialyzed 100,000 x g soluble fractions were assayed for protein kinase activity as described (l), using 5 mg/ml casein and 0.50 mM [r-“‘PI ATP (48 cpm/pmol).

Cells T incorporation

nmullminlmg cell protein

Uninfected 0.6 Infected (20 hours) 10.0 Mixture” 4.5 Infected (30 min) 0.7

a Sample consisted of 2 x IO8 uninfected cells and 2 x 1O8 frog virus-infected cells (20 hours).

host cell and virion enzymes are antigenically distinct. In

contrast, most of the protein kinase from infected cells was

inhibited by these antibodies. In control studies, the purified

virion protein kinase was completely inhibited by antibody

when mixed with the uninfected cell extract, indicating that

uninfected cells did not contain any component which might

block antibody-mediated inhibition of protein kinase. The

above results indicated that the major protein kinase activities

expressed in uninfected cells are distinct from the frog virus

activity. However, on the basis of these studies alone, it was not possible to distinguish if the high level of activity of

infected cells resulted from the derepression or perhaps a

modification of a host cell-specified protein kinase or resulted

from the synthesis of a virus-specified kinase.

Comparison of Protein Kinase of Frog Virus Grown in

Different Cells-One approach to determine whether the virion

protein kinase is virus or host cell-specified is to compare the

immunochemical and biochemical properties of the protein

kinases of frog virus grown in phylogenetically distinct cell

lines. Moreover, it would also be informative to compare the

frog virus activity with the kinase associated with genetically

unrelated viruses grown in the same cell line as frog virus. Frog

virus has a very broad host range in vitro (7), and for these

experiments was cultivated in cells originating from minnow,

chicken, and hamster species. The virus was passaged a minimum of two times in each of these hosts to eliminate any

cross-contamination of virion protein components. Vaccinia

and vesicular stomatitis viruses, each of which contain a

protein kinase activity (3, 5, 12-15), were grown in hamster

cells. Viruses were purified as described under “Methods” and

their polypeptide composition examined by polyacrylamide gel

electrophoresis in the presence of sodium dodecyl sulfate in

I I / I 100 200 300

pg ANTIBODY

Fro. 3. Effect of anti-frog virus protein kinase antibody on the protein kinase activity of infected and uninfected cells. The protein kinase from infected or uninfected minnow cells represented the 100,000 x g soluble fractions prepared as described in Table I. The 3.~1 aliquots of the 20.hour infected cell extract, 10-J aliquots of the uninfected cell extract, or IO-p1 aliquots of the uninfected cell extract plus 1 ~1 (7 ng) of the Sephadex G-100 fraction of frog virus protein kinase were added to reaction mixtures of 40 ~1, as described in Fig. 1, containing the indicated quantities of anti-frog virus protein kinase antibody. After incubation at 37” for 30 min, the protein kinase activity of each sample was measured as described in Fig. 1. Control values, representing the incorporation obtained in the presence of 215 yg of pre-immune antibody, were 1,640 pmol/lO min for the infected cell extract, 04; 224 pmol/lO min for the uninfected cell extract, 0-O; 1,430 pmol/lO min for the mixture, A-A. Values for each point indicate the mean of duplicate samples.

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Virus-specified Protein Kinase

order to verify the authenticity of the preparations (Fig. 4). As described previously (5, 16), highly purified frog virus consists of at least 20 distinct size classes of polypeptides, characterized by the major capsid protein of apparent molecular weight 48,000. It was noteworthy that the polypeptide composition of frog virus grown in different cell lines is remarkably similar (Fig. 4, B to 0). As described by Sarov and Joklik (17), purified vaccinia virus is composed of approximately 30 classes of polypeptides, characterized by the two major capsid polypep- tides of apparent molecular weights of 58,000 and 63,000 which appear as a fused, dense band in this experiment (Fig. 4E). As reviewed by Wagner et al. (18), purified vesicular stomatitis virus consists of five major classes of polypeptides of apparent molecular weights 200,000, 65,000, 48,000,45,000, and 30,000 in the present study (Fig. 4F). Due to their similar size, and the large amount of the 48,000 protein relative to the 45,000 protein, these components were poorly resolved in this electro- phoretic system.

The antibodies prepared against the purified frog virus protein kinase were employed as a probe to compare the antigenic relatedness of the protein kinases of these different viruses (Fig. 5). The activities from frog virus grown in different hosts appeared antigenically related, as they dis- played similar inhibition curves when titrated with antibody. In contrast, the protein kinases of vaccinia or vesicular stomatitis viruses were not affected by these antibodies even though they were cultivated in the same host cells as the frog virus. For this experiment the amount of protein kinase from each virus preparation was selected so that the assay would certainly have had the required sensitivity to detect an inhibition of the vaccinia or vesicular stomatitis virus enzymes because as indicated by the total incorporation for the controls (Fig. 5), a lesser quantity of enzyme activity was measured from these latter viruses as compared to the activity from frog virus. Their failure to be inhibited indicated that they were antigenically distinct from the frog virus protein kinase. The observation that the antigenic determinants of frog virus

17,800 -

A BCD E F

FIG. 4. Polyacrylamide gel electrophoresis of purified viruses. Virus samples, grown and purified as described under “Methods,” were analyzed by polyacrylamide slab gel electrophoresis in the presence of 0.1% sodium dodecyl sulfate as described (1). A, standard proteins; phosphorylase a (9 Kg; molecular weight 92,500), bovine serum albumin (3 pg; molecular weight 67,000), ovalbumin (7 pg; molecular weight 45,000), DNase I (5 pg; molecular weight 31,000), chymotryp- sinogen (5 pg; molecular weight 25,500). myoglobin (5 pg; molecular weight 17,800). B, frog virus grown in minnow cells (70 yg of protein). C, frog virus grown in hamster cells (84 fig of protein). D, frog virus grown in chick embryo fibroblasts (70 pg of protein). E, vaccinia virus grown in hamster cells (80 pg of protein). F, vesicular stomatitis virus grown in hamster cells (50 fig of protein).

protein kinase are independent of the host used to grow the virus provides strong evidence for the virus-specific nature of this activity.

The biochemical relatedness of the protein kinases of frog virus grown in different hosts was studied by activity measure- ments employing a variety of protein substrates (Table II). The reaction conditions selected, as well as the substrate concentra- tions tested, were those found to be optimal for the purified virion enzyme (1). Consistent with the conclusion derived from immunochemical studies, the protein kinase from frog virus grown in different hosts demonstrated the identical relative activity throughout the entire range of substrates tested.

Induction of Protein Kinase in Cells Infected with Condi- tional Lethal Mutants of Frog Virus-The virus-specified nature of the protein kinase suggested that viral mutants might demonstrate an altered expression of this activity. Thusly, cells infected with various temperature-sensitive mu- tants of frog virus were screened for the synthesis of the viral protein kinase (Table III). Cultures grown at either 25” (permissive temperature) or 30” (nonpermissive temperature) were harvested late in the infection and employed as the enzyme source. Protein kinase assays were done at the nonper- missive temperature. Cells infected with wild type virus, mutants 150, 1912, 3674, and 9467, grown either at 25’ or 30”, or mutants 2436 and 6642, grown at 25”, demonstrated the characteristic lo- to 20-fold higher specific activity of protein

-I 100 200

pg ANTIBODY

FIG. 5. Effect of anti-frog virus protein kinase antibody on several virion protein kinases. Virus samples, grown and purified as described under “Methods,” were disrupted by adding Triton X-100 and KC1 to final concentrations of 1.5% (v/v) and 1.0 M, respectively, and a 100,000 x g soluble fraction prepared as described for frog virus grown in minnow cells (1). This procedure solubilized 90% of the total protein kinase of frog virus and vesicular stomatitis virus, and approximately 50% of the total protein kinase activity of vaccinia virus (measured with 5 mg/ml of casein and 0.5 mM ATP as substrates). The 2-~1 aliquots of the 100,000 x g soluble fractions were added to reaction mixtures of 40 pl, as described in Fig. 1, containing anti-frog virus protein kinase antibody, as indicated. After incubation at 37” for 30 min, protein kinase activity was measured as described in Fig. 1, except that the added reaction mixture contained, in addition, 5 mM dithiothreitol. Control values, representing the incorporation in the presence of 215 pg of pre-immune antibody, were 986 pmol/lO min for frog virus grown in minnow cells, O--O; 822 pmol/lO min for frog virus grown in hamster cells, 0-O; 790 pmoVl0 min for frog virus grown in chicken cells, A-A; 90 pmol/lO min for vaccinia virus grown in hamster cells, A-A; and 75 pmol/lO min for vesicular stomatitis virus grown in hamster cells, m---H. Values for each point indicate the mean of duplicate samples.

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Virus-specified Protein Kinase 3189

TABLE II DISCUSSION

Substrates ofprotein kimse from frog oirusgrown in drfferent host cells

Aliquots of the 100,000 x g soluble fraction from frog virus grown in the indicated host cells, prepared as described in Fig. 5, were assayed for protein kinase activity with various phosphate acceptor

proteins and [r-32P]ATP as substrates, ab described in Table II from the accompanying manuscript (1).

A major question concerning the virion-associated protein

kinase of enveloped animal cell viruses is whether the enzyme

is a host cell or virus-specified component. The experiments

described herein lend strong support to the hypothesis that in

the case of frog virus the virion protein kinase is virus-speci-

fied. The virion enzyme was not detectable by immunological

methods in uninfected cells and appeared as the major protein kinase activity in cells following frog virus infection. Moreover,

the kinases of frog virus grown in phylogenetically distinct cell

lines appeared closely related by immunochemical techniques

and had the identical biochemical substrate specificity, while

the protein kinases of several other enveloped viruses grown in

the same cell line as frog virus were immunologically unrelated

to the frog virus activity. For the frog virus enzyme to be host

cell-specified, these findings would require that (a) this activ- ity be only a very minor component of the total protein kinase

activity normally expressed in uninfected cells, (6) the enzyme

be specifically derepressed following infection by frog virus, (c)

the enzyme be specifically packaged in virions of frog virus and not in the virions of other enveloped viruses grown in the same

cell line, and (d) these protein kinase activities from cells of

widely different species have similar substrate specificity and

nearly complete antigenic cross-reactivity, despite the observa-

tion that numerous other cyclic adenosine 3’:5’-monophos-

phate-independent protein kinases (including the activity

associated with vaccinia or vesicular stomatitis viruses grown

in the same cell line as frog virus, the protein kinase activity

from uninfected minnow cells, a cyclic nucleotide-independent

protein kinase from a hamster insulinoma,’ or the protein

kinases of Rauscher murine leukemia virus or Simian sarcoma

virus3) demonstrated no detectable cross-reactivity with the

frog virus enzyme.

Casein Phosvitin Histone mixture Arginine-rich histone Lysine-rich histone Protamine sulfate

pmdlminl~l enq me

68.0 (lOO)a 56.7 (100) 54.6 (100)

49.0 (72) 42.0 (74) 39.3 (72)

32.6 (48) 30.1 (53) 25.8 (47)

32.5 (48) 30.6 (54) 27.7 (51) 10.2 (15) 6.7 (12) 6.3 (12)

14.3 (21) 9.9 (17) 7.4 (14)

“Values in parentheses, reported as percentages, indicate the enzyme activity standardized to the incorporation with casein as substrate.

TABLE III

Synthesis of viral-specified protein kinase by temperature-sensitive mutants of frog virus

Extracts were prepared as described under “Methods” from the frozen pellets (approximately 4 x lOa cells each) of uninfected, wild type, or mutant frog virus-infected minnow cells. Aliquots of the dialyzed 100,000 x g soluble fraction were assayed for protein kinase activity as described (l), using 5 mg/ml of casein and 0.5 rnM [y-32P]ATP (42 cpm/pmol).

Uninfected Infected with wild type Infected with TS 150 Infected with TS 1912 Infected with TS 3674 Infected with TS 9467 Infected with TS 2436 Infected with TS 6642

nmollmmlmg cellprotein

0.4 (0)” 0.4 (0)

9.9 (90) 8.7 (87) 4.1 7.1

9.4 5.6

9.1 4.2

4.2 4.0

5.7 (82) 0.7 (48) 15.7 (92) 0.9 (54)

“Several of the extracts were titrated with anti-frog virus protein kinase serum as described in Fig. 1, and the maximal inhibition of protein kinase activity, reported as a percentage of the control, is given in parentheses.

kinase as compared to uninfected cells. Notable exceptions

were cultures infected with mutants 2436 and 6642, grown at

300, whose level of protein kinase was nearly equal to the

activity of uninfected cultures. When extracts from cells

infected with these latter two mutants were titrated with

anti-frog protein kinase antibodies, approximately 50% of the

protein kinase activity appeared to be virus-specified (Table

III). It was not clear whether this small amount of viral protein

kinase was due to the enzyme activity carried in by the virus

used to initiate the infection, or represented a reduced net

synthesis of the viral enzyme. The finding of viral mutants

which demonstrated a defect in the synthesis of the virion

protein kinase further substantiated the virus-specific nature

of the enzyme. However, it is not known at present whether

either of these viruses actually contain a mutation in the virion

protein kinase gene.

An additional possibility to be considered was that one or

several frog virus gene products might direct the modification

of a host cell protein kinase, in which case the antibodies

prepared by immunization with the virion enzyme might be

directed specifically against the modified portions of the

enzyme molecules. Previous studies have established that after

infection by frog virus, host cell protein synthesis is rapidly

diminished (19, 20). Assuming that this inhibition affects

cellular proteins nonspecifically, as a host cell-specified activ-

ity the synthesis of the virion protein kinase would occur

predominantly prior to virus infection. However, direct mea-

surement of the specific radioactivity of highly purified protein

kinase from frog virus grown in the presence of 3H-labeled

amino-acids indicated that the virion activity was synthesized

during the virus growth cycle (1). It therefore seems highly

unlikely that the virion protein kinase is a modified host cell

activity, although additional studies are required to prove this.

The protein kinase associated with vesicular stomatitis virus

has been suggested to be host cell-specified, as reduced enzyme

activity was observed in virions obtained from cells treated

prior to the infection with actinomycin D and cycloheximide

(3). We wish to emphasize that in lieu of the complexity of

most animal viruses, it is possible that the protein kinases of

different viruses may have markedly distinct properties and

we do not take the findings with frog virus to imply that other

virion protein kinases are virus-specified. Clearly, however, the

fact that with some viruses this activity is virus-coded suggests

‘U. Schubart, unpublished observations. ‘H. Silberstein, unpublished observations.

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3190 Virus-specified Protein Kinase

that it may play an important role in viral replication whether it be of viral or host origin.

2. Kuo, J. E., and Greengard, P. (1969) Proc. Natl. Acad. Sci. U. 5’. A. 64, 1349-1355

The biological role of protein phosphorylation in animal virus systems is presently not understood. As a virus-specified activity, the frog virus enzyme bears a resemblance to a protein kinase specified by the genome of T, bacteriophage which is induced in Escherichia coli shortly following infection (21). By labeling with 3zP, during infection, the T,-induced activity was shown to catalyze the phosphorylation of at least 12 polypep- tides, some of which are thought to be host cell ribosomal proteins (21). In the case of frog virus, the physiological substrates are as yet unknown; however, the functioning of this activity could be implicated at any step in the infection, including virus penetration and uncoating, nucleic acid repli- cation, and virus assembly and maturation. Mutants of the frog virus should prove useful in establishing the function of this enzyme.

3. Imblum, R. L., and Wagner, R. R. (1974) J. Viral. 13, 113-124 4. Stoker, M., and MacPherson, I. (1964) Nature 203, 1355-1357 5. Silberstein, H., and August, J. T. (1973) J. Viral. 12, 511-522 6. Vogt, P. K. (1969) in Fukdamental Techniques in Vir&ogy (Habel,

K., and Salzman, N. P., eds) pp. 198-211, Academic Press, New York

7. Granoff, A., Came, P. E., and Greeze, D. C. (1966) Virology 29, 133-148

8. Naegele, R. F., and Granoff, A. (1971) Virology 44, 286-295 9. Huang, A. S., Greenawalt, J. W., and Wagner, R. R. (1966)

Virology 30, 161-172 10. Joklik, W. K. (1962) Biochim. Biophys. Acta 61, 290-301 11. Hartree, E. F. (1972) Anal. Biochem. 48, 422-427 12. Strand, M., and August, J. T. (1971) Nature New Biol. 233,

137-140 13. Paoletti, E., and Moss, B. (1972) J. Viral. 10.417-424 14. Downer, D. N., Rogers, H. W., and Randall, C. C. (1973) Virology

52, 13-21

Acknoruledgments-We would like to thank Mette Strand for helpful discussions; Mark Rosenberg and Jimmy Jones for assistance in handling and immunization of rabbits; and Drs.

R. L. Soffer and P. Silverman for critically reading the manuscript.

REFERENCES

1. Silberstein, H., and August, J. T. (1976) J. Biol. Chem. 251, 3176-3184

15. Moyer, S. A., and Summers, D. F. (1974) J. Viral. 13,455-465 16. Tan, K. B., and McAuslan, B. R. (1971) Virology 45, 200-207 17. Sarov, I., and Joklik, W. K. (1972) Virology 50, 579-592 18. Wagner, R. R., Prevec, L., Brown, F., Summers, D. F., Sokol, F.,

and MacLeod, R. (1972) J. Viral. 10, 12281230 19. Maes, R., and Granoff, A. (1967) Virology 33, 491-502 20. Goorha, R., and Granoff, A. (1974) Virology 60, 237-259 21. Rahmsdorf, H. J., Pai, S. H., Ponta, H., Herrlich, P., Roskoski, R.,

Jr., Schweiger, M., and Studier, F. W. (1974) Proc. N&l. Acad. Sci. U. S. A. 71, 586-589

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from

H Silberstein and J T AugustCharacterization of a virion protein kinase as a virus-specified enzyme.

1976, 251:3185-3190.J. Biol. Chem. 

  http://www.jbc.org/content/251/10/3185Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/251/10/3185.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 7, 2018

http://ww

w.jbc.org/

Dow

nloaded from