EXPERIMENTAL

CELL

RESEARCH

199,

lo-18

(19%)

Cooperation of ~53 and Polyoma Virus Middle T Antigen in the Transformation of Primary Rat Embryo Fibroblasts ELLEN Department

REIHSAUS,’

of Biochemistry,

STEFAN KBAISS,~ ANGELIKA

BARNEKOW,*

AND MATHIAS

University of Urn, P.O. Box 4066, D 7900 Urn, Germany; and *Department University of Miinster, Badestrasse 9, D 4400 Miinster, Germany

the activated tion. 0 1992

Cell transformation in uiuo seems to be a multistep process. In in vitro studies certain combinations of two oncogenes, a cytoplasmic gene product together with a nuclear gene product, are sufficient to transform primary rodent cells. Polyoma virus large T antigen can immortalize and, in cooperation with polyoma virus middle T antigen, transform primary cells. On the other hand mutant mouse ~63 can also immortalize and, in cooperation with an activated Ha-ms oncogene, transform primary cells. In the present study we analyzed whether mutant p53 can replace polyoma virus large T antigen in a cell transformation assay with polyoma virus middle T antigen. Transfection of mutant p53 alone resulted in a cell line which had retained the actin cable network, grew poorly in medium with low concentration of serum, and failed to grow in semisolid agar. Cotransfection of mutant p53 together with polyoma virus middle T led to cells which grew in medium containing low serum concentration, grew well in semisolid agar, and displayed an altered morphology with the tendency to overgrow the normal monolayer. By these criteria these cells were considered fully transformed. The rate of p53 synthesis was similar in both cell lines. However, only ~63 from the transformed cell line turned out to be stable. Cells transformed by mutant pIi3 and polyoma virus middle T expressed nearly the same amount of the c-src-encoded ~~60”” protein as cells transformed by the same p53 and cotransfected activated Ha-ras oncogene. However, only the polyoma virus middle T/PBS-transformed cells exhibited an elevated level of ppGO’+“- specific tyrosine kinase activity. Thus, despite different mechanisms leading to cell transformation, mutant p53 can replace polyoma virus large T antigen and polyoma virus middle T can replace

oncogene

in

cell

Tumorbiology,

transforma-

Inc.

A broad range of viral proteins encoded by small DNA tumor viruses are able to confer tumorigenic or transformed properties to primary or established cells. Among the most investigated virally encoded proteins which are known to possess oncogenic potential are the T antigens of the papova virus family. Those of the SV40 virus promote initiation and maintenance of transformation to nonpermissive cells. The immortalizing and the transforming capacity is located on a single virally encoded polypeptide, the SV40 large T antigen [l], which seems to be a multifunctional protein. Studies with polyoma virus (Py) strongly support the idea of a cooperation between at least two different virally encoded proteins in cell transformation, namely Py large T antigen (LT) and Py middle T antigen (mT), which are both required for cell transformation [2]. PyLT can immortalize primary cells but fails to transform such cells [3] whereas PymT is able to transform established cells but not primary cells [2}. However, coexpression of large and middle T antigen resulted in full transformation of primary cells. The manner in which these DNA tumor viruses transform cells in culture has been the subject of intense studies for several years. Some of these proteins have been shown to stimulate the transcription of viral and cellular genes. Furthermore, it was demonstrated that the T antigens interact with critical cellular regulatory proteins. Thus, SV40 large T antigen binds to and appears to inactivate ~53 and the retinoblastoma gene product, the Rb protein, which are both growth suppressor proteins [4,5]. In contrast, PyLT binds Rb protein but not ~53, which may be relevant to the additional requirement for PymT in polyoma virus cell transformation. p53 has been known as a cellular protein involved in cell transformation since 1979 when it was found in elevated levels in SV40 and chemically trans10

Inc. reserved.

Press,

of Experimental

INTRODUCTION

’ Present address: Cetus Corporation, 1400 Fifty-Third Street, Emeryville, CA. * Present address: Boehringer Mannheim, D-8132 Tutzing, Bahnhofstrasse 8-15. 3 To whom correspondence and reprint requests should be addressed.

0014-4827192 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

Ha-ras Academic

MONTENARH~

p53 AND

POLYOMA

MIDDLE

T ANTIGEN

formed cells [6,7]. However, the nature of ~53 as a cellular oncogene on its own was established when it was shown by DNA transfection assays that ~53 acts to immortalize primary cells in culture [8] and can complement an activated ras-oncogene to achieve complete transformation of primary cells [8-lo]. p53 in these experiments thus acted in a manner similar to some other nuclear oncogenes such as myc or PyLT. It was recently shown that the ~53 genomic and cDNA clones that were used for these experiments were all mutant forms of this gene and that the wild-type ~53 gene failed to have these biological activities [ll, 121. Thus, mutant ~53 has to be considered as an oncogene and wild-type ~53 as a growth suppressor gene [13,14]. In a model system for cooperation between nuclear oncogenes and cytoplasmic oncogenes it was demonstrated that the nuclear myc protein cooperates with the cytoplasmic membrane associated ras protein to transform primary rat embryo fibroblasts [15]. Similar experiments were described for the nuclear mutant ~53 in cooperation with an activated ras oncogene [16]. The immortalizing PyLT is also located at the nucleus whereas the transforming PymT is cytoplasmic membrane-associated [17]. In the present investigation we wanted to analyze whether mutant ~53 might be able to replace polyoma large T antigen in a transformation assay using rat embryo fibroblasts and a cotransfected PymT. We found that overexpression of p53 in combination with expression of PymT alters the growth properties and morphology of the cotransfected cells. Overexpression of ~53 alone was found to be sufficient to immortalize but insufficient to transform primary cells, whereas transfection of PymT alone did not result in transformation. MATERIALS

AND

METHODS

Plasmids. Plasmid LTRp53cG9 was kindly provided by Dr. Moshe Oren, The Weizmann Institute of Science (Rehovot, Israel) [18]. This plasmid directed the overproduction of mutant mouse p53 which possesses a change at amino acid 135 from alanine to valine. The plasmid pSV2Neo contains the resistance gene to neomycin (G418) under the control of an SV40 promoter. PyMTl was originally described by Zhu et al. [19] and kindly provided by Dr. Cesare Vesco (Rome, Italy). Transfection protocol. Transfection of whole plasmid DNA was performed by the calcium phosphate method [20] combined with a glycerol shock [21]. Cells were cultured in fresh Dulbecco’s modified Eagle’s minimal essential medium (DMEM) 4 h before transfection. Calcium-phosphate-plasmid DNA precipitates were added to the cultures (10 pg DNA/GO-mm dish) and incubated for 6 h in 5% CO, at 37°C. Cells were washed with HBS buffer (140 mM NaCl, 25 mit4 Hepes, 0.7 n&f Na2HP0,, pH 7.1) and exposed to 15% glycerol in HBS buffer and finally fed with new DMEM containing 10% fetal calf serum. Forty-eight hours later, cells were fed with fresh medium containing 10% fetal calf serum and 400 pglml of G418 (Sigma, Munich, Germany). DMEM containing G418 was changed every 3 days and the cells were kept exposed to G418 until they were all resistant to this selection condition. Growing cells were trypsinized and passaged

IN TRANSFORMATION

11

in DMEM containing 5% fetal calf serum. After several passages cell were trypsinized, harvested, and grown in semisolid agar (0.5% agar in DMEM) containing 5% fetal calf serum. Cells picked from growing soft agar colonies were trypsinized, replated in 60-mm dishes, and propagated further. Cells which did not grow in semisolid agar were diluted to a concentration of one single cell per 100 microliters medium and plated in 96-well plates. When cells reached confluence they were expanded into 60-mm dishes and propagated further. Individual cultures of newly established cell lines were examined for the expression of p53 protein or polyoma middle T antigen by immunofluorescence and immunoprecipitation. Cell culture. Clone 6 is a transformed cell line, established from rat embryo fibroblasts by cotransfection of an activated Ha-ro.s gene and pLTRp53cG9 [16]. Ret2 are primary rat embryo fibroblasts at passage 2. The c-src-transfected 3T3 cells [22] were a gift from D. Shalloway (Pennsylvania State University, PA). All cells were grown at 37°C in a 5% CO, incubator in lo-cm dishes in DMEM containing 5% fetal calf serum (RPmp, Ref2p53, clone 6) or 10% fetal calf serum (Ret??, c-src-transfected 3T3 ceils). Antibodies. The monoclonal antibody 327 directed against PPGOC-‘~ [23] was obtained from J. Brugge (University of Pennsylvania, PA). For the immunoprecipitation of p53 monoclonal antibodies, PAb246 and PAb248 [24] or PAb421[25] was used. The monoclonal antibodies were purified from hybridoma supernatants by protein A-Sepharose chromatography. Saturating amounts of monoclonal antibody were used to ensure complete immunoprecipitation of p53 or pp60c-8m. Monoclonal antibody olPyMT7 specific for PymT [26,27] was kindly donated by Dr. B. Griffin (London, UK). Growth rate. Growth curves and population doubling time of cells cultivated in DMEM containing 0.5% fetal calf serum were obtained by trypsinizing cells from 6-cm dishes and counting the cells in a Neubauer chamber daily for 7 days. Indirect immunofiuorescence microscopy. For indirect immunofluorescence studies, trypsinized cells at different passages were seeded on g-mm coverslips and grown in culture for 48 h. Cells were fixed with ethanol for 20 min at -20°C and washed with phosphate-buffered saline (PBS). Cells were incubated with TRITC-labeled phalloidin (0.05 mg/ml) for 40 min in the dark. Cells were extensively washed with PBS and mounted on glass slides. Immunostained cells were examined under a Zeiss microscope (Axioscope) equipped with phase contrast, fluorescein, and rhodamine optics. Radiolabeling and extraction of cells. Subconfluent monolayers of cells in lo-cm dishes (4 X lo6 cells) were washed three times with methionine-free DMEM and then labeled with 110 X 10’ Bq of [?S]methionine for 2 h. For the pulse-chase experiments cells were labeled for 30 min. After labeling, cells were washed twice with DMEM containing 5% fetal calf serum and nonlabeled methionine in excess and then further cultured in this medium for various periods (chase) or extracted immediately (pulse). For extraction, cells were rinsed with ice-cold PBS, scraped off the plates, pelleted at 400 g for 2 min, and lysed with 0.4 ml extraction buffer (0.5% Nonidet P-40 (NP-40), 100 mM Tris-HCl, pH 9.0,O.l M NaCl) per 4 X 10s cells for 30 min on ice. To protect proteins from proteolysis, 1% Trasylol (Bayer AG, Leverkusen, Germany) and phenylmethylsulfonylfluoride (final concentration, 0.25 mg/ml) were added to the extraction buffer. After pelleting for 2 min at 800 g, the supernatants were centrifuged for 30 min at 105,000 g in a Type 50 Beckman rotor at 4°C. Aliquots of the supernatants were taken immediately for determination of protein content by the method of Lowry et al. [28]. For immunoprecipitation equal amounts of total protein from the pulse and chase values were employed. Zmmunoprecipitation, sodium dodecyl sulfate (SDS)-polyacrylumide gel electrophoresis. The cell extracts were precleared overnight at 4’C with 100 pl of a 10% suspension of heat-inactivated formaldehyde-fixed Staphylococcus aureus (Cowan I) as described [29]. S. aureus was pelleted and the supernatants were further incubated for 2 h

12

REIHSAUS

at 4°C with PAb246 and S. aureus followed by PAb248 and S. aureu.s to precipitate ~53. Immunoprecipitation with the first and the second antibody in a sequential immunoprecipitation was repeated up to three times to ensure a quantitative precipitation. For the precipitation of PymT we used the polyoma virus middle T antigen specific antibody otPyMT7 [27]. pp60’~“” was immunoprecipitated from the cell extract using 1 pl MAb327 and a rabbit anti-mouse antibody. After addition of a 10% suspension of S. aureuS the immunocomplexes were washed and the in oitro phosphorylation was carried out as described previously [30]. Immune complexes were washed three times with 50 mA4 Tris-HCl, pH 8.0,500 mJt4 LiCl, 1 mM DTT, 1 m&f EDTA, 1% Trasylol, twice with 50 m&f Tris-HCl, pH 7.4, 0.15 M NaCI, 5 m&f EDTA containing 1% NP-40, and then finally with 50 mM NH,HCO,. Immune complexes were eluted with 200 ~1 elution buffer (50 mh4 NH,HCO,, 1% SDS, 5% P-mercaptoethanol) for 45 min at 4”C, lyophilized, and dissolved in 20 ~1 of sample buffer (65 mM Tris-HCl, pH 6.8, 5% P-mercaptoethanol, 10% glycerol, 0.01% bromophenol blue) by boiling for 10 min. After centrifugation for 1 min at 10,000 rpm in an Eppendorf centrifuge, the samples were loaded onto lo-cm discontinuous 10% SDS-polyacrylamide gels and run at room temperature at a constant current of 18 mA as described [31]. After electrophoresis, gels were prepared for fluorography on Kodak X-AR films as previously described [32]. To quantitate the amount of labeled proteins at the different time points of the pulsechase experiments and to further calculate the half-life of ~53, the protein bands on the exposed films were analyzed by densitometric tracing. The pulse-labeling time point was taken to be zero. The halflife was taken to be the time when 50% of the highest amount of immunoprecipitated protein label was left. Western blot analysis. Immunoprecipitates were analyzed on a 10% SDS-polyacrylamide gel and transferred electrophoretically (60 mA, 45 min) to an immobilon (Millipore, Germany) transfer membrane prewashed with 0.3 M Tris-HCI, 20% methanol, then with 25 mM Tris-HCl, 20% methanol, and finally with 25 mA4 Tris-HCI, 20% methanol, 40 mM 6-amino-N-hexanoic acid. Transfer membranes were blocked in PBS, 0.5% Tween 20 for 20 min at 40°C and incubated with MAb327 (diluted 1:2000 with PBS, 0.25% Tween 20) or with PBS, 0.25% Tween 20 in the absence of antibodies for at least 20 h at 4°C followed by incubation with biotinylated anti-mouse antibody (diluted 1:lOOO in PBS, 0.25% Tween 20) for 60 min at 25°C. Alkaline phosphatase-streptavidin diluted 1:lOOO in buffer 1 (0.1 M Tris-HCl, 0.1 M NaCl, 2.5 M MgCl,, 2% NaN,, pH 7.4) was added. After incubation for 20 min at 25°C the membranes were washed in buffer 1 and subsequently in buffer 2 (1 M Tris-HCl, pH 9.4, 0.1 M NaCl, 2.5 M MgCl,) and the blots developed by the addition of 0.5% 5-bromo-4-chloro-3-indolylphosphate and 0.03% 4-nitroblue tetrazolium chloride (Serva, Heidelberg, Germany) for 3-4 min at 40°C.

RESULTS

Transformation and p53

of Rat Embryo Fibroblasts

by PymT

In order to analyze whether mutant ~53 can cooperate with PymT in cell transformation, primary rat embryo fibroblasts not older than passage 2 or 3 were transfected with plasmid pLTRp53cG9 which encodes mutant p53 with an exchange of amino acid 135 alanine to valine [16], plasmid PyMTl which harbors the polyoma virus middle T antigen coding sequences [19] or a combination of both. All transfection experiments were performed in the presence of plasmid pSV2Neo as a selectable marker. After G418 selection, only cells trans-

ET AL.

fected with pLTRp53cG9 or a combination of pLTRp53cG9 and PyMTl, but not cells transfected only with PyMTl, proved to be viable and were further propagated. Permanently growing cells transfected with pLTRp53cG9 or a combination of pLTRp53cG9 and PyMTl were regarded as immortal and termed Ref2p53 or RPmp, respectively. The morphology of the newly established cell lines Ref2p53 and RPmp is shown in Fig. 1. Ref2p53 were morphologically unaltered when compared to their parental REF cells whereas RPmp cells showed a different morphology. RPmp cells were spindle shaped and exhibited a more refractile morphology. Moreover, RPmp cells lost contact inhibition and were able to form foci overgrowing a cell monolayer (data not shown). Apparently, by morphological criteria RPmp cells exhibited a transformed phenotype. To support these observations we analyzed RPmp cells for other transformation criteria. Primary and nontransformed cells display a well-organized actin cable network. Staining for actin cables in RPmp and Ref2p53 cells was performed using fluorescent-labeled phalloidin [33]. Figure 2 clearly demonstrates that Ref2p53 exhibited a well-defined actin cable structure. In contrast RPmp cells failed to express a comparable actin network, indicating that RPmp cells expressed a transformed phenotype and Ref2p53 cells displayed characteristics of a normal phenotype.

Growth Properties of RPmp and Ref2p53 Cells In order to characterize both cell lines in detail we wanted to define the growth characteristics of Ref2p53 cells and RPmp cells. To this end Ref2p53 and RPmp cells were cultivated in medium containing a low serum concentration (0.5% fetal calf serum) and subsequently the growth rate was determined. We found that RPmp cells grew much more rapidly than Ref2p53 cells (Fig. 3). Ref2p53 nearly stopped growing or had only a very low growth rate. Thus, RPmp cells seemed to be highly independent of serum factors whereas Ref2p53 cells required relatively high concentrations of fetal calf serum for normal growth. In this respect, RPmp cells behaved like transformed cells. Next, we tested the ability of Ref2p53 and RPmp cells to grow in semisolid medium. Both cell lines were seeded in semisolid agar and monitored for their growth properties. As shown in Fig. 4, only RPmp cells formed colonies in soft agar whereas Ref2p53 cells lacked this property completely. Cells from different foci and also from foci of repeated experiments were picked and expanded into continuously growing cell lines. In up to five different clones tested, morphology growth properties and the properties of p53 turned out to be very similar if not identical. Furthermore, all individual clones (seven clones) of Ref2p53 established after dilution to a concentration of one sin-

p53 AND

POLYOMA

MIDDLE

T ANTIGEN

RPmp

REF2p53

FIG.

1.

Morphology

of Ref2p53 and RPmp cells. Cells were photographed

gle cell per culture well and subsequent expansion to mass cultures turned out to have the same properties with respect to morphology, growth properties, and protein expression. In summary, these experiments have shown that cells cotransfected with a plasmid directing the synthesis of mutant p53 together with a plasmid coding for PymT gave rise to cells which have to be classified as fully transformed cells, whereas cells which were obtained by transfection with the p53 encoding plasmid alone were not transformed as demonstrated by these various criteria. Thus, we have to conclude that mutant p53 can complement PymT in transformation of primary cells.

Ref2p53

13

IN TRANSFORMATION

at subconfluence.

Magnification

X200.

Expression of p53 and PymT in Ref2p53 and RPmp Cells In order to analyze Ref2p53 and RPmp cells for the expression of p53 protein and the polyoma virus middle T antigen, 3 X lo6 cells of each cell line were radiolabeled with [35S]methionine. Cells were extracted and p53 was immunoprecipitated from the cell extract using the p53 specific monoclonal antibodies PAb246 and PAb248. Both monoclonal antibodies specifically recognize mouse p53 [24]. PAb246 was used because this monoclonal antibody is known to precipitate p53 in a wild-type conformation whereas PAb248 recognized both wild-type and mutant p53 [ 12,341. As shown in Fig.

RPmp

FIG. 2. Actin cable staining in Ref2p53 and RPmp cells. Cells were seeded on coverslips, pled phalloidin. Actin-containing fibers are visualized.

fixed with ethanol, and stained with TRITC-cou-

14

REIHSAUS

the cell extract of RPmp cells using the polyoma middle T antigen specific monoclonal antibody cuPyMT7 [27]. The immunoprecipitate was analyzed on a 10% SDSpolyacrylamide gel. As shown in Fig. 6, monoclonal antibody aPyMT7 precipitated PymT from the cell extract of RPmp cells. Thus, RPmp cells expressed both proteins PymT and ~53. Furthermore, in both cell lines ~53 was found in two different immunologically defined subclasses.

6

0 ) 0

ET AL.

1 2

3

4

5

6

7

days Growth curves of Ret’2p53 and RPmp cells. Cells were grown for 7 days on 6-cm culture dishes. Cell numbers were determined after trypsination and counting in a Neubauer chamber.

5 both cell lines expressed p53 in a PAb246+ and a PAb246- form. The PAb246+ form is a minor form compared to the amount precipitated by PAb248. Since both antibodies are mouse specific and do not cross-react with rat p53 these results demonstrate that mouse p53 was indeed expressed in the transfected rat cells. It was demonstrated earlier that mouse ~53 can complex with rat ~53 [16] and we cannot exclude that both proteins may also form complexes in Ref2p53 and in RPmp cells. However, under our experimental conditions we were unable to separate mouse and rat ~53 on the SDSpolyacrylamide gels. For the detection of PymT, RPmp cells were labeled with [35S]methionine and extracted after 2 h labeling. PymT was immunoprecipitated from

Ref 2 p53

Metabolic Stability of ~53 is Correlated with the Phenotype of the Cell Line Usually there is a marked difference between the levels of ~53 in normal and those in transformed cells which can largely be explained by an increased metabolic stability of the ~53 gene product in transformed cells [35-371. As shown in Fig. 5, both Ref2p53 and RPmp cells express high amounts of mutant mouse ~53. These properties enabled us to analyze the metabolic stability of the same mutant ~53 in the nontransformed but immortalized cell line Ref2p53 and in the transformed RPmp cell line. To this end Ref2p53 and RPmp cells were pulse-labeled with [35S]methionine for 30 min and chased with unlabeled methionine for various times. Cells were extracted and equal amounts of total protein in the cell extract were incubated with the ~53 specific monoclonal antibody PAb421. Immunoprecipitated proteins were analyzed on a 10% SDS-polyacrylamide gel. After exposure on an X-ray film, protein bands were scanned with a laser densitometer. The highest amount of radioactivity was taken as 100% and the radioactivity determined for the various chase pe-

RPmp

FIG. 4. Growth of Ref2p53 and RPmp in semisolid agar. Cells were seeded in semisolid agar and cultured day with fresh fetal calf serum. Growing colonies were photographed at Day 12. Magnification X200.

for 12 days by feeding every third

p53 AND

A

POLYOMA

MIDDLE

T ANTIGEN

Ref2p53

IN TRANSFORMATION

B

Nab

15

RPmp

92.5-

-p53m+r

69

-

45

-

-p53m+r

FIG. 6. Metabolic labeling of p53 from Ref2p53 (A) and RPmp (B) cells. Cells (3 X 10’) of each line were labeled with [?S]methionine for 2 h. p53 was immunoprecipitated from the cell extracts by using monoclonal antibody PAb246(a). The supernatants of this immunoprecipitation were then incubated with PAb248(b) to precipitate the remaining ~53. Immunoprecipitates were analyzed on a 10% SDS-polyacrylamide gel followed by fluorography. The positions of the molecular weight markers phosphorylase a (92,500 Da [92.5]), bovine serum albumin (69,000 Da [69]) and ovalbumin (46,000 Da [46]) are shown at the left. N, hamster control serum; p53m+r, mouse and rat ~53.

sumption RPmp cells were extracted and ~~60”.“” was immunoprecipitated with the pp60”-“” specific monoclonal antibody MAb327 [23]. After washing, one half of the immunoprecipitate was incubated with [T-~~P]ATP and the in vitro phosphorylation reaction was carried out as described previously [30]. Proteins were eluted from the immune complex and further analyzed on an SDS-polyacrylamide gel. The other half of the immunoprecipitate was directly analyzed on an SDS-polyacrylamide gel and ~~60”.“” protein was detected by Western blot analysis. Lane 1 in Fig. 8A shows the autophosphorylated ~~60’.“” which was obtained after the in reaction. Lane 1 of Fig. 8B shows vitro phosphorylation the corresponding Western blot analysis. In order to compare this pp60”-“” kinase activity from RPmp cells with that from another transformed cell line which was RPmp Cells Are Characterized by an Elevated pp60C-S’c obtained by a cotransfection experiment of the same Kinase Activity mutant p53 with an activated ras oncogene, namely PymT is known to increase the activity of ~~60’~“” clone 6 cells [16], we repeated the same experiment as kinase and the stimulation of this kinase activity is described above with clone 6 cells. As shown in lane 2 strictly correlated with cell transformation [38]. One (Fig. 8B), by Western blot analysis in clone 6 cells we would assume that RPmp cells should exhibit an elefound amounts of pp60”+” protein which were comparavated pp60”-“” kinase activity. In order to test this asble to those shown for RPmp cells before. However, we were unable to detect a significant kinase activity associated with this pp60”-“” protein (lane 2, Fig. 8A). Thus, RPmp although both transformed cell lines contain comparaNC ble amounts of ~~60’.“” protein only the p53/PymTtransformed RPmp cell line displayed a considerable pp60”-“” specific tyrosine kinase activity. In order to evaluate the relative amount of ~~60’.“” protein in these two cell lines we precipitated ~~60’.“” protein from 3T3 -pYmT cells which were transfected with a c-src construct [ 221. As shown in lane 3 (Fig. 8B) the amounts of ~~60’.“” in both RPmp and clone 6 cells are considerable lower than the amount of ~~60”.“” molecules detected in c-srcFIG. 6. Metabolic labeling of PymT from RPmp cells. Cells (3 x 10’) were labeled with [?S]methionine for 2 h and PymT was precipitransfected 3T3 cells. riods was calculated in relation to this 100% value. The half-life was determined as the time point where 50% of the highest amount of radioactivity was left. As shown in Fig. 7A, p53 from the immortalized Ref2p53 cells proved to be very unstable with a half-life of approximately 20 min. In contrast the same p53 in the transformed RPmp cell line turned out to be considerably more stable with a half-life of about 200 min. Thus, mutant mouse p53 from Ref2p53 cells showed a stability which is typical for nontransformed cells whereas the same p53 from RPmp cells exhibited a stability typical for transformed cells. Moreover, these results demonstrate that the stability of p53 is dependent on the phenotype of the cell and independent of mutant or wildtype ~53.

l

tated from the cell extract by using monoclonal antibody cuPymT7 and S. aureus. The immunoprecipitate was analyzed on a 10% SDSpolyacrylamide gel followed by fluorography. Molecular weight markers are the same as described in the legend to Fig. 5. N, hamster control serum.

DISCUSSION

Cell transformation and tumor formation seem to be the result of a multistep process. These processes can

16

REIHSAUS

ET AL.

*

10 8 , 20

5 1

40 Chase

10 Chase

60

lf timdhl

. P53

FIG. ‘7. Metabolic stability of p53 from Ref2p53 (A) and RPmp cells (B). For each time point 3 X lo6 cells were labeled with [?S]methionine for 30 min. Cells were chased for 20,40, and 90 min (Reflp53) or 2,5, and 15 h (RPmp). Cell extracts with equal amounts of total protein were immunoprecipitated with monoclonal antibody PAb421. After analysis of the proteins on a 10% SDS-polyacrylamide gel and fluorography the relative amounts of the radioactively labeled proteins were determined by densitometry.

include chromosomal translocations, gene amplification and point mutations of “gain of function” oncogenes, and, in several cases, the loss of growth suppressor genes. Experiments with primary rodent cells indicate that the dysregulated expression of two or more independent oncogenes is required for cell transformation in vitro. There are a number of experimental systems in which cooperation between oncogenes can be analyzed. The general observation was that certain com-

A

B

FIG. 8. Phosphorylation of ppGO’-““(A) and quantitative Western blot analysis of endogenous ppGO’-““(B). ppGO’-‘” was immunoprecipitated from the cell extract of RPmp(1) or clone 6 (2) cells by MAb327 and the in uitro phosphorylation was performed with [y32P]ATP (A). Cell extracts from RPmp(l), clone 6 (2), and c-srctransfected 3T3 cells (3) were separated on a 8% polyacrylamide gel and the proteins electrophoretically transferred onto immobilon membranes. Proteins were detected using MAb327 followed by incubation with biotinylated anti-mouse antibody. The blot was developed using alkaline phosphatase-strepavidin as described recently

[ssl (B).

binations of oncogenes could transform primary rodent cells in culture whereas individually the same oncogenes were inactive [ 39,401. Very early evidence for oncogene cooperation came from experiments carried out with individual early oncogenes of the polyoma virus such as PymT, PyLT, and Pyst, and combinations of these genes [2]. Early genetic analysis implied that polyoma virus DNA codes for so-called early proteins that are involved in transformation [41]. Further analysis showed that the Py large T antigen, which is a nuclear protein, appears to be responsible for cellular immortalization and the Py middle T antigen for transformation. The role of Py small T antigen is less clear, but it seems to confer certain mitogenic activity to cells in the course of a cell transformation. Py large T antigen allows cells to grow in low concentrations of serum [2]. Further studies showed that also cellular oncogenes could, similarly to the two polyoma virus oncogenes, cooperate in cell transformation [ 15,421. A variety of different combinations were tested for their ability to cooperate in cell transformation and the results of these experiments led to the concept that nuclear oncogene products can cooperate best with cytoplasmic membrane bound oncogenes [39]. The nuclear class of oncogenes includes myc, mutant ~53, fos, jun, adenovirus Ela, and polyoma virus large T antigen whereas the cytoplasmic category includes MS, src, erb B, cytoplasmic SV40 large T antigen, and polyoma virus middle T antigen [39, 43, 441. The generality of this model is doubtful since there are examples where two nuclear oncoproteins [45] or two cytoplasmic oncoproteins [46] can successfully cooperate in cell transformation.

p53 AND

POLYOMA

MIDDLE

T ANTIGEN

It became clear, in particular from the studies with the transforming proteins of DNA tumor viruses, that a variety of the viral oncogenes transform or immortalize by an interaction with cellular regulatory proteins. Two such cellular proteins which have been identified are wild-type ~53 and the retinoblastoma Rb protein [5,47]. Both proteins belong to the class of growth suppressor proteins. They both bind to the SV40 large T antigen. The transforming protein Ela of adenovirus binds Rb whereas the adenovirus Elb protein binds ~53. The E7 proteins of human papilloma virus HPV16 and HPV18 bind Rb, and the E6 proteins bind to ~53. In contrast, polyoma virus large T antigen binds Rb protein whereas none of the polyoma viral oncoproteins bind to ~53. Thus, one has to assume an alternative route in cell transformation by polyoma virus in comparison to the other DNA tumor viruses. Since there is a striking similarity with respect to oncogenic potential between mutant ~53 and PyLT we analyzed whether p53 could substitute for PyLT in a typical cotransfection experiment. For these experiments we used the ~53 expression vector pLTRp53cG9 which codes for a p53 protein that possesses a change at amino acid 135 from alanine to valine and is known to be activated for cell transformation [48]. This construct was used previously in cotransfection experiments with an activated Ha-ras oncogene to transform primary rat embryo fibroblasts [16]. RPmp cells which were obtained after a cotransfection experiment of mutant ~53 with a PymT construct fulfil several criteria of fully transformed cells. Besides a transformed morphology, these cells had lost their actin cables and they grew in medium with low serum concentration and in semisolid agar. In contrast, Ref2p53, which were selected after transfection with mutant ~53 alone, displayed the typical behavior of immortalized cells, similar to the cells obtained earlier using this or another ~53 mutant [37, 49 1. Both cell lines showed a similar rate of synthesis for ~53 since they expressed comparable amounts of ~53 after metabolic labeling with a radioactive amino acid. However, there is a striking difference in the stability of ~53 in these cell lines. ~53 from the immortalized cell line Ref2p53 proved to be quite unstable with a half-life of less than 20 min. This instability is normally observed for wild-type ~53 in nontransformed cells [35,36, 501 although very recently it has been demonstrated that ~53 protein stability seems to be independent of the genetic lesions, but to correlate with the phenotype of the cells [37]. ~53 in the transformed RPmp cells was quite stable with a half-life of 200 min. This observation is in agreement with earlier results where the half-life of the same mutant p53 protein in transformed cells was found ranging between 150 and 420 min [36]. Together with the present data this makes the assumption that a specific factor might be involved in the stabilization of ~53 in transformed cells reasonable. Although we trans-

IN TRANSFORMATION

17

fected a mutant ~53 into the rat embryo fibroblasts we could detect two different immunologically defined variants of p53. Hinds et al. [ 121 provided evidence that the epitope for PAb246 is missing in mutants of the ~53 protein and it is now assumed that PAb246 might specifically recognize wild-type ~53 [34]. Our present results demonstrate that mutant ~53 from a nontransformed and from a transformed cell can also express the PAb246 epitope. PymT is the transforming protein of polyoma virus. This protein interacts with cellular membranes and has an associated tyrosine-specific protein kinase activity [51-531. The oncogenic activity of PymT is dependent on the ability to form a stable complex with the cellular kinase ~~60”.““. ~~60’.“” present in the complex has an enhanced kinase activity compared with that of the free uncomplexed form [54, 551. As shown in the present study, only RPmp cells exhibited a significant pp60”-“” kinase activity although the amount of ~~60”~“” protein in RPmp cells is comparable to the amount of ~~60”~“” found in p53lrus-transformed cells. Thus, an increase of the c-src kinase activity is a specific feature of polyoma virus middle T transformants. These data support the idea of quite different mechanisms operative in cell transformation induced by the combination of two different oncogenes. p53/ras-transformed cells are an example to show that an elevated pp60”+” kinase activity is not absolutely necessary for cell transformation. Our results show that polyoma virus middle T can substitute an activated Ha-ras gene and mutant ~53 can substitute for polyoma virus large T in cell transformation, although very different mechanisms and cellular pathways were activated. PyLT binds and inactivates Rb, a growth suppressor. When PyLT is replaced by another oncogene, either Rb or ~53 have to be inactivated by an alternative mechanism. Since mutant ~53 is able to complex endogenous wild-type ~53 this complex formation may essentially account for cell transformation in this system. We thank E. Ossendorf, U. Wend, and R. Korber for excellent technical assistance, J. Brugge for the monoclonal antibody MAb327, D. Shalloway for the c-src-transfected NIH3T3 cells, M. Oren for the plasmid pLTRp53cG9 and clone 6 cells, C. Vesco for pPyMT, and B. Griffin for monoclonal antibody cuPyMT7 directed against PymT. We thank Alison Gatrill for careful reading of the manuscript and helpful suggestions. This research was supported by grants from Deutsche Forschungsgemeinschaft (Ba 876/1-l) to A.B. and (SFB322, Al, and Fonds der Chemischen Industrie) to M.M.

REFERENCES 1. 2. 3.

Tooze, J. (1981) Cold Spring Harbor Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Rassoulzadegan, M., Cowie, A., Carr, A., Glaichenhaus, N., Kamen, R., and Cuzin, F. (1982) Nature 300, 713-718. Cherington, V., Morgan, B., Spiegelman, B. M., and Roberts, T. M. (1986) Proc. Natl. Acad. Sci. USA 83, 4307-4311.

18 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

15.

REIHSAUS Levine, A. J. (1990) Virology 177, 419-426. Marshall, C. J. (1991) Cell 64,313-326. Lane, D. P., and Crawford, L. (1979) Nature 278, 261-263. Linzer, D. I. H., and Levine, A. J. (1979) Cell 17,43-52. Jenkins, J. R., Rudge, K., and Currie, G. A. (1984) Nature 312, 651-654. Eliyahu, D., Raz, A., Gruss, P., and Oren, M. (1984) Nature 312, 646-649. Parada, L. F., Land, H., Weinberg, Nature 312,649-651.

R. A., and Rotter,

Kessler, S. W. (1975) J. Immunol. 115, 1617-1624. Barnekow, A., and Gessler, M. (1986) EMBO J. 5, 701-705. Montenarh, M., and Henning, R. (1983) J. Viral. 45, 531-538. Bonner, W. M., and Laskey, R. A. (1974) Eur. J. Biochem. 46, 83-88.

33.

Wulf, E., Deboben, A., Bautz, F. A., Faulstich, H., and Wieland, Th. (1979) Proc. Natl. Acad. Sci. USA 9, 4498-4502. Werness, B. A., Levine, A. J., and Howley, P. M. (1990) Science 248, 76-79.

34. 35.

Baker, S. J., Fearon, E. R., Nigro, J. M., Hamilton, S. R., Preisinger, A. C., Jessup, J. M., vanTuinen, P., Ledbetter, D. H., Barker, D. F., Nakamura, Y., White, R., and Vogelstein, B. (1989) Science 244,217-221. Land, H., Parada, L. F., and Weinberg, R. A. (1983) Nature 304, 596-602.

37. 38. 39. 40. 41. 42. 43.

Pinhasi-Kimhi, (1986) Nature

17.

Dilworth, S. M., Hansson, H.-A., Darnfors, C., Bjursell, G., Streuli, C. H., and Griffin, B. E. (1986) EMBO J. 5.491-499.

44.

18.

Michalowitz, 6,3531-3536.

D., and Oren, M. (1986) Mol. Cell. Biol.

45.

19.

Zhu, Z., Veldman, G. M., Cowie, A., Carr, A., Schaffhausen, B., and Kamen, R. (1984) J. Virol. 51, 170-180. Graham, F. L., and van der Eb, A. J. (1973) Virology 52, 456467. Frost, E., and Williams, J. (1978) Virology 91, 39-50.

46.

20. 21. 22. 23.

A., and Oren, M.

36.

16.

D., Eliyahu,

D., Ben-Zeev,

29. 30. 31. 32.

V. (1984)

Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinhasi-Kimhi, O., and Oren, M. (1989) Proc. Natl. Acad. Sci. USA 86, 8763-8767. Hinds, P., Finlay, C., and Levine, A. J. (1989) J. Virol. 63,739746. Finlay, C. A., Hinds, P. W., and Levine, A. J. (1989) Cell 57, 1083-1093.

O., Michalovitz, 320, 182-184.

ET AL.

Shalloway, D., Coussens, P. M., and Yaciuk, P. (1984) Proc. Natl. Acad. Sci. USA 81,7071-7075. Lipsich, L. A., Lewis, A. J., and Brugge, J. S. (1983) J. Virol. 48, 552-560.

24.

Yewdell, J. W., Gannon, J. V., and Lane, D. P. (1986) J. Virol. 59,444-452.

25.

Harlow, E., Crawford, L. V., Pim, D. C., and Williamson, (1981) J. Viral. 39, 861-869. Dilworth, S. M. (1982) EMBO J. 1,1319-1328.

26. 27.

Dilworth, S. M., and Griffin, USA 79,1059-1063.

28.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, (1951) J. Biol. Chem. 193,265-275.

47. 48. 49. 50.

l, lOl-110. 51. 52.

N. M. 53. 54.

B. E. (1982) Proc. N&l. Acad. Sci.

Received July 18, 1991 Revised version received October 21, 1991

R. J.

Halevy, O., Hall, A., and Oren, M. (1989) Mol. Cell. Biol. 9,33853392. Reihsaus, E., Kohler, M., Kraiss, S., Oren, M., and Montenarh, M. (1990) Oncogene 5, 137-145. Kraiss, S., Spiess, S., Reihsaus, E., and Montenarh, M. (1991) Exp. Cell Res. 192, 157-164. Nemeth, S. P., Fox, L. G., and DeMarco, M. (1989) Mol. Cell. Biol. 9, 1109-1119. Weinberg, R. A. (1985) Science 230, 770-776. Ruley, H. E. (1990) Cancer Cells 2, 258-268. Griffin, B. E., and Dilworth, S. M. (1983) Adu. Cancer Res. 39, 183-268. Ruley, H. E. (1983) Nature 304,602-606. Fischer-Fantuzzi, L., and Vesco, C. (1987) Oncogene Res. 1,229242. Michalowitz, D., Fischer-Fantuzzi, L., Vesco, C., Pipas, J. M., and Oren, M. (1987) J. Virol. 61, 2648-2654. Ruppert, J. M., Vogelstein, B., and Kinzler, K. W. (1991) Mol. Cell. Biol. 11, 1724-1728. Reed, J. C., Haldar, S., Croce, C. M., and Cuddy, M. P. (1990) Mol. Cell. Biol. 10, 4370-4373. Levine, A. J., and Momand, J. (1990) Biochim. Biophys. Acta 1032,119-136. Finlay, C. A., Hinds, P. W., Tan, T.-H., Eliyahu, D., Oren, M., and Levine, A. J. (1988) Mol. Cell. Biol. 8, 531-539. Jenkins, J. R., Rudge, K., Redmond, S., and Wade-Evans, A. (1984) Nucleic Acids Res. 12, 5609-5626. Oren, M., Maltzman, W., and Levine, A. J. (1981) Mol. Cell. Biol.

55. 56.

Smith, A. E., and Ely, B. K. (1983) Adu. Viral Oncology 3,3-30. Courtneidge, S. A., and Smith, A. E. (1983) Nature 303, 435439. Courtneidge, S. A., and Smith, A. E. (1984) EMBO J. 3,585-591. Bolen, J. B., Thiele, C. J., Israel, M. A., Yonemoto, W., Lipsich, L. A., and Brugge, J. B. (1984) Cell 38, 767-777. Courtneidge, S. A. (1985) EMBO J. 4, 1471-1477. Barnekow, A., Jahn, R., and Schartl, M. (1990) Oncogene 5, 1019-1024.