Molecular analysis of different allelic variants of wild-type human p53 FRANCE MOREAU AND GREGMATLASHEWSKI
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
Institute of Parasitology, McGiN University, Macdonald Campus, 21 111 Lakeshore Road, Ste. Anne de Bellevue, Que., Canada H9X 1CO Received February 7, 1992 MOREAU,F., and MATLASHEWSKI, G. 1992. Molecular analysis of different allelic variants of wild-type human p53. Biochem. Cell Biol. 70: 1014-1019. The p53 tumour suppressor gene is intensively studied because mutations in this gene are the most common genetic alteration so far identified in human cancer. Considerable emphasis has thus been placed on characterizing the biological differences between mutant and wild-type p53 protein. This has led to the realization that in cultured cells, mutant p53 behaves like an oncogene, whereas wild-type p53 is a tumour suppressor gene. The p53 protein is also a target for the tumour virus oncogene products SV40 large T, adenovirus ElB, and human papillomavirus type 16 E6, which are all capable of forming complexes to the p53 protein. Although p53 represents an extremely important cellular regulatory molecule which is well conserved, there exists two allelic variants of wild-type human p53 that differ both in primary and confirmational structure. One variant contains an arginine at amino acid 72 (p53Arg), whereas the other form contains a proline at this residue (p53Pro). The possible implications for more than one allelic variant of wild-type human p53 in the general population is unknown. The present study was undertaken to compare some of the biological features of the different wild-type p53 variants. We present data demonstrating that there was a posttranscriptional selection against accumulation of both variants of wild-type human p53 in 3T3-A31 cells, arguing that both forms are proliferation inhibitory in these cells. Both variants of human p53 were stabilized by SV40 large T, but did not displace mouse p53 from SV40 large T. Neither allelic variant of human p53 was able to reduce significantly SV40-mediated anchorage-independent growth of 3T3-A31 cells. Taken together, these data suggest that although there are structurally different variants of wild-type human p53, there is no difference in the biological activity of these molecules at the level of the biological assays performed here. Key words: human p53, large T, transformation, oncogenes, tumour suppressor. MOREAU,F., et MATLASHEWSKI, G. 1992. Molecular analysis of different allelic variants of wild-type human p53. Biochem. Cell Biol. 70 : 1014-1019. Le gkne suppresseur tumoral de la p53 est ktudik de f a ~ o nintensive parce que les mutations dans ce gkne sont I'altQation genetique la plus commune identifike jusqu'ici dans le cancer humain. On a donc mis beaucoup d'emphase sur la caractkrisation des differences biologiques entre la protkine p53 mutante et la p53 de type sauvage. Cela a permis de constater que dans les cellules cultivkes, le gkne de la p53 mutante se comporte comme un oncogkne alors que le gkne de la p53 de type sauvage est un gkne suppresseur tumoral. La protkine p53 est Cgalement une cible pour les produits oncogkniques du virus tumoral SV40 grand T, de l'adknovirus ElB et du papillomavirus humain E6 de type 16 qui sont tous capables de former des complexes avec la protkine p53, Bien que la p53 soit une molecule rkgulatrice cellulaire extrsmement importante et bien conservke, il existe deux variants allkliques de la p53 humaine de type sauvage qui diffkrent tant dans leur structure primaire que dans leur structure conformationnelle. L'un des variants renferme une arginine a la position 72 (p53 Arg) et l'autre, une proline (p53 Pro). Les implications possibles montrant la prksence de plus d'un variant allklique de la p53 humaine de type sauvage dans la population gCnCrale sont inconnues. Nous avons voulu comparer certains des aspects biologiques des diffkrents variants de la p53 de type sauvage. Nous presentons des donnkes dkmontrant qu'il y a selection post-transcriptionnelle contre l'accumulation des deux variants de la p53 humaine de type sauvage dans les cellules 3T3-A31; cela prouve que les deux formes inhibent la proliferation dans ces cellules. Les deux variants de la p53 humaine sont stabilisks par le SV40 grand T, mais ils ne deplacent pas la p53 de souris du SV40 grand T. Ni l'un ni l'autre des variants alltliques de la p53 humaine n'est capable de rkduire de f a ~ o nimportante la croissance indkpendante de l'ancrage induit par le SV40 des cellules 3T3-A31. Dans l'ensemble, ces donnkes montrent que msme si les variants de la p53 humaine de type sauvage ont des structures diffkrentes, il n'existe aucune diffkrence dans l'activitk biologique de ces molkcules au niveau des essais biologiques effectues ici. Mots cl6s : p53 humaine, grand T, transformation, oncogknes, suppresseur tumoral. [Traduit par la rkdaction]
Introduction p53 was first identified in SV40-transformed cells, where it was found to physically complex with the large T oncogene product (Lane and Crawford 1979). Until recently the reason for this complex formation was poorly understood. Subsequently, it was reported that p53 was an oncogene because of its ability to cooperate with ras to transform primary rodent cells (Jenkins et al. 1984; Eliyahu et al. 1984; Parada et al. 1984). However, these original oncogene studies were ABBREVIATIONS: HPV-16, human papillomavirus type 16; SDS-PAGE, sodium dodecyl sulfate - polyacrylamide gel electrophoresis; NP40, Nonidet P-40; LTR, Moloney murine leukemia virus long terminal repeats; LT, large T. Printed in Canada / Imprime au Canada
performed with mutant murine p53 molecules and subsequent experiments revealed that the wild-type p53 was in fact a suppressor of transformation in vitro (reviewed by Levine et al. 1991; Finlay et al. 1989). It has also been revealed that wild-type human p53 was capable of reducing the tumorigenic phenotype of cultured human tumourderived cell lines, further supporting the argument that wildtype p53 is a tumour suppressor molecule (Chen et al. 1990). This view was also consistent with a previous observation showing that Friend virus-induced murine erythroleukemia was commonly associated with disruption of the p53 gene (Mowat et al. 1985). Recent genetic studies have provided further evidence that p53 is a tumour suppressor gene by revealing that mutations
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
MOREAU AND MATLASHEWSKI
in the p53 gene are the most common genetic alteration so far identified in human cancer (reviewed in Levine et al. 1991). For example, 80% of colon carcinomas have cells with no normal p53 alleles, where in the majority of cases one allele is lost through a deletion and the other suffers a point mutation (Baker et al. 1989). In addition, oncogene products from DNA tumour viruses are capable of complexing with tumour suppressor proteins p53 and Rb (reviewed in Levine et al. 1991; Weinberg 1991). For example, SV40 large T (Lane and Crawford 1979), adenovirus E l B (Sarnow et al. 1982), and HPV-16 E6 (Werness et al. 1990) are all capable of forming a physical complex with p53. Likewise, SV40 large T, adenovirus ElA, and HPV-16 E7 are capable of complexing with the Rb protein (reviewed in Levine et al. 1991; Weinberg 1991). One hypothesis for the complex formation with virus oncogene products is that this results in the inactivation of p53 and Rb function, thus releasing the cell from a normal negative control mechanism for proliferation. Taken together, these observations argue that a loss of p53 or Rb function through deletion or complex formation with virus oncogene products represents an important step in the development of a variety of tumours. The biochemical function of wild-type p53 is not clear. However, it has been demonstrated that p53 is capable of binding to a specific DNA sequence (Kern et al. 1991) and can activate transcription (Fields and Jang 1990; Raycroft et al. 1990). It is, therefore, likely that p53 acts to control transcription of specific target genes which may be involved in regulating cell proliferation. It was also revealed that more than one allelic variant of wild-type human p53 exists in the general population and that these alleles code for structurally different p53 proteins (Matlashewski et al. 1987b). One allelic variant codes for a proline at residue 72 (p53Pro) and the other codes for a n arginine at residue 72 (p53Arg). This variation is distinct from mutations occurring in different tumours, since both allelic variants have been identified in normal nontransformed human cells. Clearly this is a nonconservative amino acid difference of which the biological significance is unknown. In the present study, we have examined the expression of both allelic types of wild-type human p53 in murine 3T3-A31 cells in the presence and absence of SV40 large T and HPV-16 E6 and E7. We present evidence that both forms of p53 mRNA can be expressed in these cells, but corresponding protein only accumulated in cells containing SV40 large T. We also present evidence that neither variant of human p53 could influence the level of endogenous mouse p53 or reduce SV40-induced anchorage-independent growth. Taken together, these data argue that the biological activities of the different allelic variants of wild-type human p53 are indistinguishable. Materials and methods Cells, plasmids, and antibodies
Murine 3T3-A3 1 cells used in this study were obtained from the American Type Culture Collection. Cells were routinely maintained in Dulbecco's modified Eagles medium supplemented with 10% fetal calf serum (GIBCO, Burlington). The p53 cDNA containing plasmids (pCDPro and pCDArg) were previously constructed and characterized (Matlashewski et al. 1987b). The plasmid (pHZIP-16) expressing HPV-16 E6 and E7 has also been previously described (Matlashewski et al. 1987~).The plasmid expressing SV40 large T (pBR328-LT) and the anti-SV40 large T monoclonal antibody
1015
PAb-419 (Harlow et al. 1981)were kindly provided by L. Crawford (Imperial Cancer Research Fund, Cambridge, U.K.). The antihuman p53 monoclonal antibody PAb-1801 was developed as previously described (Banks et al. 1986). Anti-mouse p53 monoclonal antibody PAb-248, which does not react with human p53, was kindly provided by S. Benchimol (Ontario Cancer Institute, Toronto). Transfection selection and transformation
3T3-A31 cells were prepared and transfected by the DNA calcium phosphate coprecipitationmethod as previously described (Matlashewski et al. 1987~).Aliquots of DNA - calcium phosphate precipitate containing 10 pg of the relevant plasmid and 1 pg of plasmid pWL-neo were added to 60-mm dishes of subconfluent cells and cells expressing the transfected DNA were selected for in the presence of 500 pg G418/mL. Resistant clones were pooled and RNA and protein assays were performed as soon as enough cells were available. Pooled cells were also seeded within a few days into media containing 0.6% Noble agar and the growth of colonies was monitored over 3 weeks. Protein analysis
Steady-state p53 protein levels were determined by first performing immunoprecipitation of human p53 with polyclonal sera (Matlashewski et al. 1986), followed by SDS-PAGE and Western analysis with PAb-1801 as previously described (Banks et al. 1986). Briefly, cells were lysed in cold NP40 buffer (1% NP40, 150 mM NaCl, 20 mM Tris, pH 8.0) for 30 min on ice. Cell debris was removed by centrifugation and the lysate was precleared with 10% fixed Staphylococntsaureus (GIBCO). Antiserum (5 pL) was then added and left for 12 h at 4°C. The precipitated protein was recovered with 10% fixed Staphylococcus aureus, and the products were resolved by SDS-PAGE and transferred onto nitrocellulose filters. Filters were blocked in 10% fetal calf serum and incubated for 2 h directly in conditioned media from relevant hybridoma cells, and subsequent detection was as previously described (Banks et al. 1986). RNA analysis
Northern blot analysis was carried out as previously described (Descoteaux and Matlashewski 1989). Briefly, total RNA was extracted with RNAzol (TEL-TEST Inc., Friendswood, Tex.), which is a modification of the guanidium-phenol-chloroform method (Chomczynskiand Sacchi 1987). RNA samples (10 pg) were denatured for 1 h at 50°C in the presence glyoxal, and 1 pg of ethidium bromide was added to each sample before electrophoresis in 1% agarose. After electrophoresis, RNA was transferred to Hybond-N membranes (Arnersham, Oakville) The membranes were photographed under UV illumination to confirm equal RNA loading and transfer. Membranes were prehybridized and hybridized at 42°C in the presence of 50% formamide with probes purified from agarose gels and nick translated in the presence of [ 3 2 ~ ](ICN, d ~Toronto). ~ ~ Following hybridization, membranes were washed in 0.5 x SSC (1 x SSC is 0.15 M NaCl and 0.015 M sodium citrate) at 55°C for 30 min and then 0.1 x SSC at 65°C for 15 min. Autoradiography was performed at - 70°C with an intensifying screen using Kodak XAR-5 film (Picker, MontrCal). Results Both allelic variants of wild-type human p53, SV40 large T, and HPV-16 E6 and E7 were expressed individually or in relevant combinations in murine 3T3-A3 1 cells, so that p53-related biological parameters could be studied. The different p53 cDNAs (p53Pr0, p53Arg) and the SV40 large T oncogene were placed under the transcriptional control of the SV40 promoter/enhancer and the E6 and E7 oncogenes from HPV-16 were placed under the transcriptional control of the Moloney murine leukemia virus long terminal repeat
1016
BIOCHEM. CELL
BIOL. VOL. 70, 1992
CGC 72 A r g
CCC
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
72 Pro
LTR
LTR
FIG. 1. Diagram representation of the relevant regions of plasmid constructions which were used for transfections in this study. Open boxes, as indicated, represent genes under investigation. Plasrnid transcription regulatory sequences are in black boxes (SV40 promoter/enhancer) or stippled boxes (LTR, Moloney murine leukemia virus long terminal repeats). pCDPro expressed wild-type p53 with a proline at residue 72 and pCDArg expressed wild-type p53 with an arginine at residue 72. pBR328-LT expressed the SV40 large T oncogene and pHZIP-16 expressed the HPV-16 E6 and E7 oncogenes. control region (LTR) as indicated in Fig. 1. Cells were cotransfected with equimolar amounts of the above constructs, together with plasmid pWL-neo which confers resistance to G418. Cells expressing the transfected plasmids were then selected for in the presence of 500 pg G418/mL and about 100 clones from each transfection were pooled for further analysis. Northern blot analysis was performed to determine whether it was possible to express both allelic variants of wild-type human p53 mRNA in these cells. As shown in Fig. 2A, cells transfected with the different wild-type p53 cDNAs contained detectable levels of human p53 mRNA, whereas control cells transfected with only the pLW-neo plasmid contained no human p53 mRNA. Fig. 2B verified that each lane contained an equal amount of intact RNA. At high stringency hybridization and washing, the human p53 DNA probe did not hybridize to the mouse p53 mRNA (Fig. 2A, lane 1). Cells cotransfected with the p53 cDNAs together with the SV40 large T gene or the HPV-16 E6 and E7 oncogenes also contained human p53 mRNA, but at lower levels than the cells transfected with only the p53 cDNA containing plasmids. In addition, in all cases there were lower levels of p53 mRNA corresponding to the p53Pro form of human p53 as compared with the p53Arg form. We next performed a Western blot analysis with a human specific anti-p53 monoclonal antibody (PAb-1801) to determine the steady-state level of human p53 protein in these transfected cells. This monoclonal antibody does not react against endogenous mouse p53 (Banks et al. 1986). As shown in Fig. 3, only the human p53 cDNA transfected cells which were cotransfected with the SV40 large T oncogene contained detectable levels of human p53 protein. This was
FIG. 2. Northern blot analysis of human p53 mRNA in 3T3-A31 cells transfected with plasmid constructs which express the indicated products (above each lane). (A) Human p53 mRNA levels are shown. (B) An ethidium bromide stain of total RNA on the same filter was probed with labelled p53 cDNA, to verify that each lane contained an equal level of intact RNA (20 pg). LT, large T. presumably due to the stabilization of p53 by the large T molecule. It was possible to distinguish the two forms of human p53, since the p53Pro variant migrated slower than the p53Arg variant in the large-T-containing cells, and this is consistent with previous observations that demonstrated that the two variant wild-type human p53s are structurally different (Matlashewski et al. 1987b). In agreement with the Northern blot data which showed a lower level of p53Pro mRNA than p53Arg mRNA, there was also a slightly lower level of p53Pro protein than p53Arg protein in these cell populations. It is clear from this Western blot that, although the cells transfected with the different human p53 cDNA plasmids by themselves contained the corresponding p53 mRNA, they did not contain human p53 protein. Since these human p53 mRNAs were capable of producing detectable protein in cells containing large T, these data suggest that there was a selection against the accumulation of wild-type human p53 at the posttranscriptional level in cells that did not contain large T. This also reveals that both variants of wild-type human p53 were not tolerated in these cells, arguing that both molecules were inhibitory for cell proliferation. We next investigated the possibility that one or both of the wild-type human p53 variants could influence the level of large T or endogenous mouse p53 in these transfected cells. For this analysis, the steady-state level of mouse p53 and large T was determined by Western blot analysis in control cells and cells that were previously shown in Fig. 2A to contain human p53. Mouse p53 was identified with a mouse-specific monoclonal antibody (PAb-248) which does not cross-react against human p53. Large T was detected with monoclonal antibody PAb-419. As shown in Fig. 4, the levels of large T (A) and mouse p53 (B) were not influenced by the presence of either variant of wild-type
1017
MOREAU AND MATLASHEWSKI
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
LT
LT p53Arg
LT Neo p53Pro
Mouse p53
LT
FIG. 3. Western blot analysis of wild-type human p53 protein in 3T3-A31 cells transfected with plasmid constructs which express the indicated products (above each lane). As shown, the p53Pro variant migrated slower than did the p53Arg variant. The size of the molecular mass markers are as indicated. LT, large T.
human p53. This argues that wild-type human p53 did not affect the ability of large T to stabilize mouse p53. As nontransformed cells are unable to form colonies in soft agar whereas transformed cells are, this is a widely used assay to characterize oncogene-mediated transformation. For example, the large T oncogene induced colony formation in the 3T3-A31 cells used in this study. We were therefore interested in determining whether the different wild-type human p53 variants could influence the ability of large T to stimulate colony formation in these cells, since wild-type p53 molecules are tumour suppressors. Cells expressing large T with and without human p53 mRNA and protein (as shown in Figs. 2A and 3) were seeded in soft agar, and total colonies were counted 2 weeks later. As shown in Table 1, neither variant of wild-type human p53 had a major effect on colony formation induced by the large T oncogene. There did, however, appear to be a slight difference in that the cells containing p53Arg generally formed fewer colonies than cells containing p53Pro. It is unknown whether this small difference is significant. It is clear, however, that neither variant of wild-type p53 had major transformation suppressive activity in this transformation assay.
Discussion The principal observation reported in this study is that the two variants of wild-type human p53 were similar in the biological activities tested here. This included the selection against their accumulation in 3T3-A31 cells at the posttranscriptional level, their inability to compete with mouse p53 for stabilization with large T, and their inability to inhibit SV40-mediated transformation of 3T3-A31 cells to anchorage-independent growth in soft agar. These data suggest that the different variant wild-type human p53 molecules are biologically indistinguishable. Although this conclusion was somewhat expected, it was nevertheless important to address this issue experimentally.
LT LT p53Arg p53Pro Neo p53Arg p53Pro
FIG. 4. Western blot analysis of SV40 large T and mouse p53 in transfected 3T3-A31 cells. Each lane contained lysates from cells expressing the indicated plasmid product. The level of SV40 large T (A) and mouse p53 (B) are shown. The size of the molecular mass markers are as indicated. LT, large T. TABLE1. Effect of wild-type human p53 on colony formation of 3T3-A31 cells by SV40 large T Soft agar coloniesb.c Plasmid product(s)" Neo p53Pro p53Arg Large T Large T Large T - -
+ p53Pro
+ p53Arg
1
2
7 13 3 964 1118 -876
16 14 2 801 1088 665
'All cells were cotransfected with pWL-neo. *colonies were counted 2 weeks after plating. %ach set of numbers represents the numbers of colonies in one of two duplicate plates from the same transfection.
Cells transfected with plasmids containing the different allelic forms of wild-type p53 contained higher levels of human p53 mRNA than did cells cotransfected with the same p53 plasmids together with the viral oncogenes. However, analysis of the human p53 proteins revealed that only cells cotransfected with the SV40 large T gene contained wildtype human p53. As expected, cells cotransfected with the HPV-16 E6 and E7 oncogenes contained no detectable p53, since the E6 protein has been reported to destabilize p53 (Scheffner et al. 1990). These data suggest that in the cells transfected with the human p53 plasmids alone, there was a selection against both allelic variants of p53 at the posttranscriptional level. Alternatively, the levels of human p53 were too low to detect. We feel this second alternative is less likely, since these cells contained high levels of translatable p53 mRNA. This is in contrast to what has been observed in human tumours where wild-type p53 is almost
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
1018
BIOCHEM. CELL BIOL. VOL. 70,
universally selected against before transcription, either through a point mutation of the gene or deletion of one or both alleles (Levine et al. 1991). This difference may have been due to the different types of selection placed on the cells; in this study viable cells wereselected within 2 weeks, whereas tumour cell development which is a multistage process can take several years. Based on the observations in this study, it may be of interest t o examine tumours with wild-type p53 genes t o determine whether a similar posttranscriptional selection against p53 can occur in such instances. Taken together, these data are consistent with the view that increased levels of either allelic variant of wildtype human p53 is not compatible with continued cell proliferation. These data are similar to a recent report showing that wildtype human p53 protein was not tolerated in human transformed cells lacking wild-type p53 (Johnson et al. 1991). I n contrast however, it has also been reported that it was possible t o express detectable levels of both allelic variants of human p53 in a human osteosarcoma cell line which lacked endogenous p53, where it was also shown that this resulted in cells with a reduction in tumorigenicity in nude mice (Chen et al. 1990). It is not known why there is variability in the ability t o express wild-type p53 in transformed cell lines, but this may be due t o the degree of cell transformation. For example, cells that are more aggressively tumorigenic may be more capable of tolerating wild-type p53 than are cells that are only weakly tumorigenic. SV40 large T was capabie of stabilizing both human and mouse p53 in the transfected cells and the presence of either allelic variant of wild-type human p53 did not alter the steady-state level of mouse p53 or large T. These data suggest that the p53 molecules were not in competition with mouse p53 for binding t o large T. These data also indicate that mouse p53 was wild-type in these cells, since it has been reported that mutant mouse p53 from transformed cells d o not bind large T (Levine et al. 1991). It was possible t o stimulate soft agar colony formation efficiently in 3T3-A31 cells transfected with the SV40 large T oncogene. We then compared each allelic variant of human wild-type p53 for their ability to inhibit or impair SV40-mediated colony formation. We observed that neither allelic variant of p53 had a significant impact on colony formation. This was despite the fact that some of the largeT-containing cells did contain high levels of wild-type p53 protein when they were seeded in soft agar. These data are consistent with a previous report which showed that wildtype p53Arg had no effect on HPV-16 E7 oncogene induced colony formation in NIH-3T3 cells (Crook et a[. 1991). It was, however, intriguing that there were consistently fewer colonies in cells transfected with the p53Arg allele. It is not known whether this is significant or whether this was due t o the lower level of p53Pro than p53Arg in these cell populations as shown in Fig. 3. It is also interesting that wild-type human 53 could not impair viral oncogene induced growth in soft agar, since previous studies have shown that wild-type p53 (murine or human) could inhibit viral oncogene induced transformation of primary rodent cells (Finlay et al. 1989; Eliyahu et al. 1989; Crook et al. 1991). This indicates that the mechanism(s) in which DNA tumour virus oncogenes transform primary rodent cells differs from
1992
that in which they induce colony formation in established rodent cells. p53 appears t o be capable of impairing the former mechanism, but not the latter. In conclusion, the data presented within show that at the level of the biological assays performed in this study, the variant allelic forms of wild-type p53 have similar, if not identical, biological activities. This is despite the fact that the primary and tertiary structure of these molecules differ. The major unresolved question is why are there two allelic variant forms of such a n important human regulatory molecule. It may be worth considering a n epidemiologic study to determine whether one allelic variant of p53 is more closely associated with any particular neoplastic disease. Acknowledgments This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and from the National Cancer Institute of Canada. Research at the Institute of Parasitology is supported by NSERC and Fonds pour la formation de chercheurs et l'aide a la recherche du Quebec. Baker, S., Fearon, E.R., Nigro. S., Hamilton, A.C., Preisinger, J.M., Jessup, J.M., van Tuinen, D.H., Ledbetter, D.F., Baker, D., Nakamura, Y., White, R., and Vogelstein, B. 1989. Chromosome 17 deletions and p53 gene mutations in colorectal carcinomas. Science (Washington, D.C.), 244: 217-221. Banks, L., Matlashewski, G., and Crawford, L. 1986. Isolation of human p53 specific monoclonal antibodies and their use in the studies of human p53 expression. Eur. J. Biochem. 159: 529-534. Chen, P.-L., Chen, Y., Bookstein, R., and Lee, W.-H. 1990. Genetic mechanisms of tumor suppression by the human p53 gene. Science (Washington, D.C.), 250: 1576-1580. Chomczynski, P., and Sacchi, N. 1987. Single step method of RNA isolation by acid guanidinium thiocynate - phenol - chloroform extraction. Anal. Biochem. 162: 156-161. Crook, T., Fisher, C., and Vousden, K.H. 1991. Modulation of immortalizing properties of human papillomavirus type 16 E7 by p53 expression. J. Virol. 65: 505-510. Descoteaux, A., and Matlashewski, G. 1989. c-fos and tumour necrosis factor gene expression in Leishmania donovani-infected cells. Mol. Cell. Biol. 9: 5223-5227. Eliyahu, D., Raz, A., Gruss, P., Givol, D., and Oren, D. 1984. Participation of p53 cellular tumour antigen in transformation of normal embryonic cells. Nature (London), 312: 646-649. Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinhasi-Kimhi, O . , and Oren, M. 1989. Wild-type p53 can inhibit oncogene mediated focus formation. Proc. Natl. Acad. Sci. U.S.A. 86: 8763-8767. Fields, S., and Jang, S.K. 1990. Presence of a potent transcription activating sequence in the p53 protein. Science (Washington, D.C.), 249: 1046-1049. Finlay, C., Hinds, P.W., and Levine, A.J. 1989. The p53 protooncogene can act as a suppressor of transformation. Cell, 57: 1083-1093. Harlow, E., Crawford, L., Pim, D., and Williamson, N. 1981. Monoclonal antibodies specific for simian virus 40 tumour antigens. J. Virol. 39: 861-869. Jenkins, J., Rudge, K., and Currie, G.A. 1984. Cellular immortalization by a cDNA clone encoding the transformationassociated phosphoprotein p53. Nature (London), 312: 651-654. Johnson, P., Gray, D., Mowat, M., and Benchimol, S. 1991. Expression of wild-type p53 is not compatible with continued growth of p53-negative tumor cells. Mol. Cell. Biol. 11: 1-11. Kern, S.E., Kinzler, K.W., Bruskin, A., Jarosz, D., Friedman, P., Prives, C., and Vogelstein, B. 1991. Identification of p53
I
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
MOREAU AND MATLASHEWSKI
as a sequence-specific DNA-binding protein. Science (Washington, D.C.), 252: 1708-1711. Lane, D., and Crawford, L. 1979. T antigen is bound to a host protein in SV40-transformed cells. Nature (London), 278: 261-263. Levine, A.J., Momand, J., and Finlay, C.A. 1991. The p53 tumour suppressor gene. Nature (London), 351: 453-456. Matlashewski, G., Banks, L., Pim, D., and Crawford, L. 1986. Analysis of human p53 proteins and mRNA levels in normal and transformed cells. Eur. J. Biochem. 154: 665-672. Matlashewski, G., Schneider, J., Banks, L., Jones, N., Murray, A., and Crawford. L. 1987a. Human papillomavirus type 16 DNA cooperates with activated ras in transforming primary cells. EMBO J. 6: 1741-1746. Matlashewski, G..Tuck, S., Pim, D., Lamb, P., Schneider, J., and Crawford, L. 1987b. Primary structure polymorphism at amino acid residue 72 of human p53. Mol. Cell. Biol. 7: 961-963. Mowat, M., Cheng, N., Kirnura, N., Bernstein, A., and Benchimol, S. 1985. Inactivation of the p53 gene in erythroleukemic cells transformed by Friend virus. Nature (London), 314: 633-636.
Parada, L., Land, H., Weinberg, R.A., Wolf, D., and Rotter, V. 1984. Cooperation between gene encoding p53 tumour antigen and ras in cell transformation. Nature (London), 312: 649-651. Raycroft, L., Wu, H., and Lozano, G. 1990. Transcriptional activation by wild-type but not transforming mutants of the p53 anti-oncogene. Science (Washington, D.C.), 249: 1049-105 1. Sarnow, P., Ho, Y.S., Williams, J., and Levine, A.J. 1982. Adenovirus Elb-58kd tumour antigen and SV40 large tumour antigen are physically associated with the same 54 kd cellular protein in transformed cells. Cell, 28: 387-394. Scheffner, M., Werness, B.A., Huibregtse, J., Levine, A.J., and Howley, P. 1990. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell, 63: 1129-1 136. Weinberg, R.A. 1991. Tumor suppressor genes. Science (Washington, D.C.), 254: 1138-1 146. Werness, B., Levine, A.J., and Howley, P.M. 1990. Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science (Washington, D.C.), 248: 76-79.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
Institute of Parasitology, McGiN University, Macdonald Campus, 21 111 Lakeshore Road, Ste. Anne de Bellevue, Que., Canada H9X 1CO Received February 7, 1992 MOREAU,F., and MATLASHEWSKI, G. 1992. Molecular analysis of different allelic variants of wild-type human p53. Biochem. Cell Biol. 70: 1014-1019. The p53 tumour suppressor gene is intensively studied because mutations in this gene are the most common genetic alteration so far identified in human cancer. Considerable emphasis has thus been placed on characterizing the biological differences between mutant and wild-type p53 protein. This has led to the realization that in cultured cells, mutant p53 behaves like an oncogene, whereas wild-type p53 is a tumour suppressor gene. The p53 protein is also a target for the tumour virus oncogene products SV40 large T, adenovirus ElB, and human papillomavirus type 16 E6, which are all capable of forming complexes to the p53 protein. Although p53 represents an extremely important cellular regulatory molecule which is well conserved, there exists two allelic variants of wild-type human p53 that differ both in primary and confirmational structure. One variant contains an arginine at amino acid 72 (p53Arg), whereas the other form contains a proline at this residue (p53Pro). The possible implications for more than one allelic variant of wild-type human p53 in the general population is unknown. The present study was undertaken to compare some of the biological features of the different wild-type p53 variants. We present data demonstrating that there was a posttranscriptional selection against accumulation of both variants of wild-type human p53 in 3T3-A31 cells, arguing that both forms are proliferation inhibitory in these cells. Both variants of human p53 were stabilized by SV40 large T, but did not displace mouse p53 from SV40 large T. Neither allelic variant of human p53 was able to reduce significantly SV40-mediated anchorage-independent growth of 3T3-A31 cells. Taken together, these data suggest that although there are structurally different variants of wild-type human p53, there is no difference in the biological activity of these molecules at the level of the biological assays performed here. Key words: human p53, large T, transformation, oncogenes, tumour suppressor. MOREAU,F., et MATLASHEWSKI, G. 1992. Molecular analysis of different allelic variants of wild-type human p53. Biochem. Cell Biol. 70 : 1014-1019. Le gkne suppresseur tumoral de la p53 est ktudik de f a ~ o nintensive parce que les mutations dans ce gkne sont I'altQation genetique la plus commune identifike jusqu'ici dans le cancer humain. On a donc mis beaucoup d'emphase sur la caractkrisation des differences biologiques entre la protkine p53 mutante et la p53 de type sauvage. Cela a permis de constater que dans les cellules cultivkes, le gkne de la p53 mutante se comporte comme un oncogkne alors que le gkne de la p53 de type sauvage est un gkne suppresseur tumoral. La protkine p53 est Cgalement une cible pour les produits oncogkniques du virus tumoral SV40 grand T, de l'adknovirus ElB et du papillomavirus humain E6 de type 16 qui sont tous capables de former des complexes avec la protkine p53, Bien que la p53 soit une molecule rkgulatrice cellulaire extrsmement importante et bien conservke, il existe deux variants allkliques de la p53 humaine de type sauvage qui diffkrent tant dans leur structure primaire que dans leur structure conformationnelle. L'un des variants renferme une arginine a la position 72 (p53 Arg) et l'autre, une proline (p53 Pro). Les implications possibles montrant la prksence de plus d'un variant allklique de la p53 humaine de type sauvage dans la population gCnCrale sont inconnues. Nous avons voulu comparer certains des aspects biologiques des diffkrents variants de la p53 de type sauvage. Nous presentons des donnkes dkmontrant qu'il y a selection post-transcriptionnelle contre l'accumulation des deux variants de la p53 humaine de type sauvage dans les cellules 3T3-A31; cela prouve que les deux formes inhibent la proliferation dans ces cellules. Les deux variants de la p53 humaine sont stabilisks par le SV40 grand T, mais ils ne deplacent pas la p53 de souris du SV40 grand T. Ni l'un ni l'autre des variants alltliques de la p53 humaine n'est capable de rkduire de f a ~ o nimportante la croissance indkpendante de l'ancrage induit par le SV40 des cellules 3T3-A31. Dans l'ensemble, ces donnkes montrent que msme si les variants de la p53 humaine de type sauvage ont des structures diffkrentes, il n'existe aucune diffkrence dans l'activitk biologique de ces molkcules au niveau des essais biologiques effectues ici. Mots cl6s : p53 humaine, grand T, transformation, oncogknes, suppresseur tumoral. [Traduit par la rkdaction]
Introduction p53 was first identified in SV40-transformed cells, where it was found to physically complex with the large T oncogene product (Lane and Crawford 1979). Until recently the reason for this complex formation was poorly understood. Subsequently, it was reported that p53 was an oncogene because of its ability to cooperate with ras to transform primary rodent cells (Jenkins et al. 1984; Eliyahu et al. 1984; Parada et al. 1984). However, these original oncogene studies were ABBREVIATIONS: HPV-16, human papillomavirus type 16; SDS-PAGE, sodium dodecyl sulfate - polyacrylamide gel electrophoresis; NP40, Nonidet P-40; LTR, Moloney murine leukemia virus long terminal repeats; LT, large T. Printed in Canada / Imprime au Canada
performed with mutant murine p53 molecules and subsequent experiments revealed that the wild-type p53 was in fact a suppressor of transformation in vitro (reviewed by Levine et al. 1991; Finlay et al. 1989). It has also been revealed that wild-type human p53 was capable of reducing the tumorigenic phenotype of cultured human tumourderived cell lines, further supporting the argument that wildtype p53 is a tumour suppressor molecule (Chen et al. 1990). This view was also consistent with a previous observation showing that Friend virus-induced murine erythroleukemia was commonly associated with disruption of the p53 gene (Mowat et al. 1985). Recent genetic studies have provided further evidence that p53 is a tumour suppressor gene by revealing that mutations
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
MOREAU AND MATLASHEWSKI
in the p53 gene are the most common genetic alteration so far identified in human cancer (reviewed in Levine et al. 1991). For example, 80% of colon carcinomas have cells with no normal p53 alleles, where in the majority of cases one allele is lost through a deletion and the other suffers a point mutation (Baker et al. 1989). In addition, oncogene products from DNA tumour viruses are capable of complexing with tumour suppressor proteins p53 and Rb (reviewed in Levine et al. 1991; Weinberg 1991). For example, SV40 large T (Lane and Crawford 1979), adenovirus E l B (Sarnow et al. 1982), and HPV-16 E6 (Werness et al. 1990) are all capable of forming a physical complex with p53. Likewise, SV40 large T, adenovirus ElA, and HPV-16 E7 are capable of complexing with the Rb protein (reviewed in Levine et al. 1991; Weinberg 1991). One hypothesis for the complex formation with virus oncogene products is that this results in the inactivation of p53 and Rb function, thus releasing the cell from a normal negative control mechanism for proliferation. Taken together, these observations argue that a loss of p53 or Rb function through deletion or complex formation with virus oncogene products represents an important step in the development of a variety of tumours. The biochemical function of wild-type p53 is not clear. However, it has been demonstrated that p53 is capable of binding to a specific DNA sequence (Kern et al. 1991) and can activate transcription (Fields and Jang 1990; Raycroft et al. 1990). It is, therefore, likely that p53 acts to control transcription of specific target genes which may be involved in regulating cell proliferation. It was also revealed that more than one allelic variant of wild-type human p53 exists in the general population and that these alleles code for structurally different p53 proteins (Matlashewski et al. 1987b). One allelic variant codes for a proline at residue 72 (p53Pro) and the other codes for a n arginine at residue 72 (p53Arg). This variation is distinct from mutations occurring in different tumours, since both allelic variants have been identified in normal nontransformed human cells. Clearly this is a nonconservative amino acid difference of which the biological significance is unknown. In the present study, we have examined the expression of both allelic types of wild-type human p53 in murine 3T3-A31 cells in the presence and absence of SV40 large T and HPV-16 E6 and E7. We present evidence that both forms of p53 mRNA can be expressed in these cells, but corresponding protein only accumulated in cells containing SV40 large T. We also present evidence that neither variant of human p53 could influence the level of endogenous mouse p53 or reduce SV40-induced anchorage-independent growth. Taken together, these data argue that the biological activities of the different allelic variants of wild-type human p53 are indistinguishable. Materials and methods Cells, plasmids, and antibodies
Murine 3T3-A3 1 cells used in this study were obtained from the American Type Culture Collection. Cells were routinely maintained in Dulbecco's modified Eagles medium supplemented with 10% fetal calf serum (GIBCO, Burlington). The p53 cDNA containing plasmids (pCDPro and pCDArg) were previously constructed and characterized (Matlashewski et al. 1987b). The plasmid (pHZIP-16) expressing HPV-16 E6 and E7 has also been previously described (Matlashewski et al. 1987~).The plasmid expressing SV40 large T (pBR328-LT) and the anti-SV40 large T monoclonal antibody
1015
PAb-419 (Harlow et al. 1981)were kindly provided by L. Crawford (Imperial Cancer Research Fund, Cambridge, U.K.). The antihuman p53 monoclonal antibody PAb-1801 was developed as previously described (Banks et al. 1986). Anti-mouse p53 monoclonal antibody PAb-248, which does not react with human p53, was kindly provided by S. Benchimol (Ontario Cancer Institute, Toronto). Transfection selection and transformation
3T3-A31 cells were prepared and transfected by the DNA calcium phosphate coprecipitationmethod as previously described (Matlashewski et al. 1987~).Aliquots of DNA - calcium phosphate precipitate containing 10 pg of the relevant plasmid and 1 pg of plasmid pWL-neo were added to 60-mm dishes of subconfluent cells and cells expressing the transfected DNA were selected for in the presence of 500 pg G418/mL. Resistant clones were pooled and RNA and protein assays were performed as soon as enough cells were available. Pooled cells were also seeded within a few days into media containing 0.6% Noble agar and the growth of colonies was monitored over 3 weeks. Protein analysis
Steady-state p53 protein levels were determined by first performing immunoprecipitation of human p53 with polyclonal sera (Matlashewski et al. 1986), followed by SDS-PAGE and Western analysis with PAb-1801 as previously described (Banks et al. 1986). Briefly, cells were lysed in cold NP40 buffer (1% NP40, 150 mM NaCl, 20 mM Tris, pH 8.0) for 30 min on ice. Cell debris was removed by centrifugation and the lysate was precleared with 10% fixed Staphylococntsaureus (GIBCO). Antiserum (5 pL) was then added and left for 12 h at 4°C. The precipitated protein was recovered with 10% fixed Staphylococcus aureus, and the products were resolved by SDS-PAGE and transferred onto nitrocellulose filters. Filters were blocked in 10% fetal calf serum and incubated for 2 h directly in conditioned media from relevant hybridoma cells, and subsequent detection was as previously described (Banks et al. 1986). RNA analysis
Northern blot analysis was carried out as previously described (Descoteaux and Matlashewski 1989). Briefly, total RNA was extracted with RNAzol (TEL-TEST Inc., Friendswood, Tex.), which is a modification of the guanidium-phenol-chloroform method (Chomczynskiand Sacchi 1987). RNA samples (10 pg) were denatured for 1 h at 50°C in the presence glyoxal, and 1 pg of ethidium bromide was added to each sample before electrophoresis in 1% agarose. After electrophoresis, RNA was transferred to Hybond-N membranes (Arnersham, Oakville) The membranes were photographed under UV illumination to confirm equal RNA loading and transfer. Membranes were prehybridized and hybridized at 42°C in the presence of 50% formamide with probes purified from agarose gels and nick translated in the presence of [ 3 2 ~ ](ICN, d ~Toronto). ~ ~ Following hybridization, membranes were washed in 0.5 x SSC (1 x SSC is 0.15 M NaCl and 0.015 M sodium citrate) at 55°C for 30 min and then 0.1 x SSC at 65°C for 15 min. Autoradiography was performed at - 70°C with an intensifying screen using Kodak XAR-5 film (Picker, MontrCal). Results Both allelic variants of wild-type human p53, SV40 large T, and HPV-16 E6 and E7 were expressed individually or in relevant combinations in murine 3T3-A3 1 cells, so that p53-related biological parameters could be studied. The different p53 cDNAs (p53Pr0, p53Arg) and the SV40 large T oncogene were placed under the transcriptional control of the SV40 promoter/enhancer and the E6 and E7 oncogenes from HPV-16 were placed under the transcriptional control of the Moloney murine leukemia virus long terminal repeat
1016
BIOCHEM. CELL
BIOL. VOL. 70, 1992
CGC 72 A r g
CCC
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
72 Pro
LTR
LTR
FIG. 1. Diagram representation of the relevant regions of plasmid constructions which were used for transfections in this study. Open boxes, as indicated, represent genes under investigation. Plasrnid transcription regulatory sequences are in black boxes (SV40 promoter/enhancer) or stippled boxes (LTR, Moloney murine leukemia virus long terminal repeats). pCDPro expressed wild-type p53 with a proline at residue 72 and pCDArg expressed wild-type p53 with an arginine at residue 72. pBR328-LT expressed the SV40 large T oncogene and pHZIP-16 expressed the HPV-16 E6 and E7 oncogenes. control region (LTR) as indicated in Fig. 1. Cells were cotransfected with equimolar amounts of the above constructs, together with plasmid pWL-neo which confers resistance to G418. Cells expressing the transfected plasmids were then selected for in the presence of 500 pg G418/mL and about 100 clones from each transfection were pooled for further analysis. Northern blot analysis was performed to determine whether it was possible to express both allelic variants of wild-type human p53 mRNA in these cells. As shown in Fig. 2A, cells transfected with the different wild-type p53 cDNAs contained detectable levels of human p53 mRNA, whereas control cells transfected with only the pLW-neo plasmid contained no human p53 mRNA. Fig. 2B verified that each lane contained an equal amount of intact RNA. At high stringency hybridization and washing, the human p53 DNA probe did not hybridize to the mouse p53 mRNA (Fig. 2A, lane 1). Cells cotransfected with the p53 cDNAs together with the SV40 large T gene or the HPV-16 E6 and E7 oncogenes also contained human p53 mRNA, but at lower levels than the cells transfected with only the p53 cDNA containing plasmids. In addition, in all cases there were lower levels of p53 mRNA corresponding to the p53Pro form of human p53 as compared with the p53Arg form. We next performed a Western blot analysis with a human specific anti-p53 monoclonal antibody (PAb-1801) to determine the steady-state level of human p53 protein in these transfected cells. This monoclonal antibody does not react against endogenous mouse p53 (Banks et al. 1986). As shown in Fig. 3, only the human p53 cDNA transfected cells which were cotransfected with the SV40 large T oncogene contained detectable levels of human p53 protein. This was
FIG. 2. Northern blot analysis of human p53 mRNA in 3T3-A31 cells transfected with plasmid constructs which express the indicated products (above each lane). (A) Human p53 mRNA levels are shown. (B) An ethidium bromide stain of total RNA on the same filter was probed with labelled p53 cDNA, to verify that each lane contained an equal level of intact RNA (20 pg). LT, large T. presumably due to the stabilization of p53 by the large T molecule. It was possible to distinguish the two forms of human p53, since the p53Pro variant migrated slower than the p53Arg variant in the large-T-containing cells, and this is consistent with previous observations that demonstrated that the two variant wild-type human p53s are structurally different (Matlashewski et al. 1987b). In agreement with the Northern blot data which showed a lower level of p53Pro mRNA than p53Arg mRNA, there was also a slightly lower level of p53Pro protein than p53Arg protein in these cell populations. It is clear from this Western blot that, although the cells transfected with the different human p53 cDNA plasmids by themselves contained the corresponding p53 mRNA, they did not contain human p53 protein. Since these human p53 mRNAs were capable of producing detectable protein in cells containing large T, these data suggest that there was a selection against the accumulation of wild-type human p53 at the posttranscriptional level in cells that did not contain large T. This also reveals that both variants of wild-type human p53 were not tolerated in these cells, arguing that both molecules were inhibitory for cell proliferation. We next investigated the possibility that one or both of the wild-type human p53 variants could influence the level of large T or endogenous mouse p53 in these transfected cells. For this analysis, the steady-state level of mouse p53 and large T was determined by Western blot analysis in control cells and cells that were previously shown in Fig. 2A to contain human p53. Mouse p53 was identified with a mouse-specific monoclonal antibody (PAb-248) which does not cross-react against human p53. Large T was detected with monoclonal antibody PAb-419. As shown in Fig. 4, the levels of large T (A) and mouse p53 (B) were not influenced by the presence of either variant of wild-type
1017
MOREAU AND MATLASHEWSKI
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
LT
LT p53Arg
LT Neo p53Pro
Mouse p53
LT
FIG. 3. Western blot analysis of wild-type human p53 protein in 3T3-A31 cells transfected with plasmid constructs which express the indicated products (above each lane). As shown, the p53Pro variant migrated slower than did the p53Arg variant. The size of the molecular mass markers are as indicated. LT, large T.
human p53. This argues that wild-type human p53 did not affect the ability of large T to stabilize mouse p53. As nontransformed cells are unable to form colonies in soft agar whereas transformed cells are, this is a widely used assay to characterize oncogene-mediated transformation. For example, the large T oncogene induced colony formation in the 3T3-A31 cells used in this study. We were therefore interested in determining whether the different wild-type human p53 variants could influence the ability of large T to stimulate colony formation in these cells, since wild-type p53 molecules are tumour suppressors. Cells expressing large T with and without human p53 mRNA and protein (as shown in Figs. 2A and 3) were seeded in soft agar, and total colonies were counted 2 weeks later. As shown in Table 1, neither variant of wild-type human p53 had a major effect on colony formation induced by the large T oncogene. There did, however, appear to be a slight difference in that the cells containing p53Arg generally formed fewer colonies than cells containing p53Pro. It is unknown whether this small difference is significant. It is clear, however, that neither variant of wild-type p53 had major transformation suppressive activity in this transformation assay.
Discussion The principal observation reported in this study is that the two variants of wild-type human p53 were similar in the biological activities tested here. This included the selection against their accumulation in 3T3-A31 cells at the posttranscriptional level, their inability to compete with mouse p53 for stabilization with large T, and their inability to inhibit SV40-mediated transformation of 3T3-A31 cells to anchorage-independent growth in soft agar. These data suggest that the different variant wild-type human p53 molecules are biologically indistinguishable. Although this conclusion was somewhat expected, it was nevertheless important to address this issue experimentally.
LT LT p53Arg p53Pro Neo p53Arg p53Pro
FIG. 4. Western blot analysis of SV40 large T and mouse p53 in transfected 3T3-A31 cells. Each lane contained lysates from cells expressing the indicated plasmid product. The level of SV40 large T (A) and mouse p53 (B) are shown. The size of the molecular mass markers are as indicated. LT, large T. TABLE1. Effect of wild-type human p53 on colony formation of 3T3-A31 cells by SV40 large T Soft agar coloniesb.c Plasmid product(s)" Neo p53Pro p53Arg Large T Large T Large T - -
+ p53Pro
+ p53Arg
1
2
7 13 3 964 1118 -876
16 14 2 801 1088 665
'All cells were cotransfected with pWL-neo. *colonies were counted 2 weeks after plating. %ach set of numbers represents the numbers of colonies in one of two duplicate plates from the same transfection.
Cells transfected with plasmids containing the different allelic forms of wild-type p53 contained higher levels of human p53 mRNA than did cells cotransfected with the same p53 plasmids together with the viral oncogenes. However, analysis of the human p53 proteins revealed that only cells cotransfected with the SV40 large T gene contained wildtype human p53. As expected, cells cotransfected with the HPV-16 E6 and E7 oncogenes contained no detectable p53, since the E6 protein has been reported to destabilize p53 (Scheffner et al. 1990). These data suggest that in the cells transfected with the human p53 plasmids alone, there was a selection against both allelic variants of p53 at the posttranscriptional level. Alternatively, the levels of human p53 were too low to detect. We feel this second alternative is less likely, since these cells contained high levels of translatable p53 mRNA. This is in contrast to what has been observed in human tumours where wild-type p53 is almost
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
1018
BIOCHEM. CELL BIOL. VOL. 70,
universally selected against before transcription, either through a point mutation of the gene or deletion of one or both alleles (Levine et al. 1991). This difference may have been due to the different types of selection placed on the cells; in this study viable cells wereselected within 2 weeks, whereas tumour cell development which is a multistage process can take several years. Based on the observations in this study, it may be of interest t o examine tumours with wild-type p53 genes t o determine whether a similar posttranscriptional selection against p53 can occur in such instances. Taken together, these data are consistent with the view that increased levels of either allelic variant of wildtype human p53 is not compatible with continued cell proliferation. These data are similar to a recent report showing that wildtype human p53 protein was not tolerated in human transformed cells lacking wild-type p53 (Johnson et al. 1991). I n contrast however, it has also been reported that it was possible t o express detectable levels of both allelic variants of human p53 in a human osteosarcoma cell line which lacked endogenous p53, where it was also shown that this resulted in cells with a reduction in tumorigenicity in nude mice (Chen et al. 1990). It is not known why there is variability in the ability t o express wild-type p53 in transformed cell lines, but this may be due t o the degree of cell transformation. For example, cells that are more aggressively tumorigenic may be more capable of tolerating wild-type p53 than are cells that are only weakly tumorigenic. SV40 large T was capabie of stabilizing both human and mouse p53 in the transfected cells and the presence of either allelic variant of wild-type human p53 did not alter the steady-state level of mouse p53 or large T. These data suggest that the p53 molecules were not in competition with mouse p53 for binding t o large T. These data also indicate that mouse p53 was wild-type in these cells, since it has been reported that mutant mouse p53 from transformed cells d o not bind large T (Levine et al. 1991). It was possible t o stimulate soft agar colony formation efficiently in 3T3-A31 cells transfected with the SV40 large T oncogene. We then compared each allelic variant of human wild-type p53 for their ability to inhibit or impair SV40-mediated colony formation. We observed that neither allelic variant of p53 had a significant impact on colony formation. This was despite the fact that some of the largeT-containing cells did contain high levels of wild-type p53 protein when they were seeded in soft agar. These data are consistent with a previous report which showed that wildtype p53Arg had no effect on HPV-16 E7 oncogene induced colony formation in NIH-3T3 cells (Crook et a[. 1991). It was, however, intriguing that there were consistently fewer colonies in cells transfected with the p53Arg allele. It is not known whether this is significant or whether this was due t o the lower level of p53Pro than p53Arg in these cell populations as shown in Fig. 3. It is also interesting that wild-type human 53 could not impair viral oncogene induced growth in soft agar, since previous studies have shown that wild-type p53 (murine or human) could inhibit viral oncogene induced transformation of primary rodent cells (Finlay et al. 1989; Eliyahu et al. 1989; Crook et al. 1991). This indicates that the mechanism(s) in which DNA tumour virus oncogenes transform primary rodent cells differs from
1992
that in which they induce colony formation in established rodent cells. p53 appears t o be capable of impairing the former mechanism, but not the latter. In conclusion, the data presented within show that at the level of the biological assays performed in this study, the variant allelic forms of wild-type p53 have similar, if not identical, biological activities. This is despite the fact that the primary and tertiary structure of these molecules differ. The major unresolved question is why are there two allelic variant forms of such a n important human regulatory molecule. It may be worth considering a n epidemiologic study to determine whether one allelic variant of p53 is more closely associated with any particular neoplastic disease. Acknowledgments This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and from the National Cancer Institute of Canada. Research at the Institute of Parasitology is supported by NSERC and Fonds pour la formation de chercheurs et l'aide a la recherche du Quebec. Baker, S., Fearon, E.R., Nigro. S., Hamilton, A.C., Preisinger, J.M., Jessup, J.M., van Tuinen, D.H., Ledbetter, D.F., Baker, D., Nakamura, Y., White, R., and Vogelstein, B. 1989. Chromosome 17 deletions and p53 gene mutations in colorectal carcinomas. Science (Washington, D.C.), 244: 217-221. Banks, L., Matlashewski, G., and Crawford, L. 1986. Isolation of human p53 specific monoclonal antibodies and their use in the studies of human p53 expression. Eur. J. Biochem. 159: 529-534. Chen, P.-L., Chen, Y., Bookstein, R., and Lee, W.-H. 1990. Genetic mechanisms of tumor suppression by the human p53 gene. Science (Washington, D.C.), 250: 1576-1580. Chomczynski, P., and Sacchi, N. 1987. Single step method of RNA isolation by acid guanidinium thiocynate - phenol - chloroform extraction. Anal. Biochem. 162: 156-161. Crook, T., Fisher, C., and Vousden, K.H. 1991. Modulation of immortalizing properties of human papillomavirus type 16 E7 by p53 expression. J. Virol. 65: 505-510. Descoteaux, A., and Matlashewski, G. 1989. c-fos and tumour necrosis factor gene expression in Leishmania donovani-infected cells. Mol. Cell. Biol. 9: 5223-5227. Eliyahu, D., Raz, A., Gruss, P., Givol, D., and Oren, D. 1984. Participation of p53 cellular tumour antigen in transformation of normal embryonic cells. Nature (London), 312: 646-649. Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinhasi-Kimhi, O . , and Oren, M. 1989. Wild-type p53 can inhibit oncogene mediated focus formation. Proc. Natl. Acad. Sci. U.S.A. 86: 8763-8767. Fields, S., and Jang, S.K. 1990. Presence of a potent transcription activating sequence in the p53 protein. Science (Washington, D.C.), 249: 1046-1049. Finlay, C., Hinds, P.W., and Levine, A.J. 1989. The p53 protooncogene can act as a suppressor of transformation. Cell, 57: 1083-1093. Harlow, E., Crawford, L., Pim, D., and Williamson, N. 1981. Monoclonal antibodies specific for simian virus 40 tumour antigens. J. Virol. 39: 861-869. Jenkins, J., Rudge, K., and Currie, G.A. 1984. Cellular immortalization by a cDNA clone encoding the transformationassociated phosphoprotein p53. Nature (London), 312: 651-654. Johnson, P., Gray, D., Mowat, M., and Benchimol, S. 1991. Expression of wild-type p53 is not compatible with continued growth of p53-negative tumor cells. Mol. Cell. Biol. 11: 1-11. Kern, S.E., Kinzler, K.W., Bruskin, A., Jarosz, D., Friedman, P., Prives, C., and Vogelstein, B. 1991. Identification of p53
I
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by YORK UNIV on 11/10/14 For personal use only.
MOREAU AND MATLASHEWSKI
as a sequence-specific DNA-binding protein. Science (Washington, D.C.), 252: 1708-1711. Lane, D., and Crawford, L. 1979. T antigen is bound to a host protein in SV40-transformed cells. Nature (London), 278: 261-263. Levine, A.J., Momand, J., and Finlay, C.A. 1991. The p53 tumour suppressor gene. Nature (London), 351: 453-456. Matlashewski, G., Banks, L., Pim, D., and Crawford, L. 1986. Analysis of human p53 proteins and mRNA levels in normal and transformed cells. Eur. J. Biochem. 154: 665-672. Matlashewski, G., Schneider, J., Banks, L., Jones, N., Murray, A., and Crawford. L. 1987a. Human papillomavirus type 16 DNA cooperates with activated ras in transforming primary cells. EMBO J. 6: 1741-1746. Matlashewski, G..Tuck, S., Pim, D., Lamb, P., Schneider, J., and Crawford, L. 1987b. Primary structure polymorphism at amino acid residue 72 of human p53. Mol. Cell. Biol. 7: 961-963. Mowat, M., Cheng, N., Kirnura, N., Bernstein, A., and Benchimol, S. 1985. Inactivation of the p53 gene in erythroleukemic cells transformed by Friend virus. Nature (London), 314: 633-636.
Parada, L., Land, H., Weinberg, R.A., Wolf, D., and Rotter, V. 1984. Cooperation between gene encoding p53 tumour antigen and ras in cell transformation. Nature (London), 312: 649-651. Raycroft, L., Wu, H., and Lozano, G. 1990. Transcriptional activation by wild-type but not transforming mutants of the p53 anti-oncogene. Science (Washington, D.C.), 249: 1049-105 1. Sarnow, P., Ho, Y.S., Williams, J., and Levine, A.J. 1982. Adenovirus Elb-58kd tumour antigen and SV40 large tumour antigen are physically associated with the same 54 kd cellular protein in transformed cells. Cell, 28: 387-394. Scheffner, M., Werness, B.A., Huibregtse, J., Levine, A.J., and Howley, P. 1990. The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell, 63: 1129-1 136. Weinberg, R.A. 1991. Tumor suppressor genes. Science (Washington, D.C.), 254: 1138-1 146. Werness, B., Levine, A.J., and Howley, P.M. 1990. Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science (Washington, D.C.), 248: 76-79.
