[CANCER RESEARCH56. 158-163. January 1, 19961 An Engineered Four-Stranded Coiled Coil Substitutes for the Tetramerization Domain of Wild-Type p53 and Alleviates Transdominant Inhibition by Tumor-derived p53 Mutants1

Matthew J. F. Waterman, Jennifer L. F. Waterman, and Thanos D. Halazonetis2

Department of Molecular Oncology, The Wistar Institute fM. J. F. W., J. L F. W., T. D. H.], and Progra,n in Molecular Biology and Genetics fM. J. F. W.] and Department of Pathology and Laboratory Medicine IT. D. HI. University of Pennsylvania, Philadelphia. Pennsylvania 19104

ABSTRACT analysis of mutant GCN4 LZs3 has identified domains with altered subunit stoichiometries (41). One such engineered zipper, which will The tetramerizationdomain of p53 is required for efficient tumor sup hereafter be referred to as the TZ, assembles as a parallel tetramer. pressoractivity.This domain, however,also allows wild-typep53 to hetero This domain has not yet been shown to drive tetramerization of any oligomerize with dominant negative tumor-derived p53 mutants. We ex protein. Furthermore, its ability to form tetramers in vivo has not been plored the feasibility of substituting the native tetramerization domain of characterized. wild-type p53 with an engineered leudne zipper that assembles as a four stranded coiled coil.The engineered zipper drove p53 tetramerization in vitro In an effort to generate a functional p53 protein that is not transdomi and p53 functIon in vivo. Furthermore, It alleviated transdominant inhibition nantly inhibited by tumor-derived p53 mutants, we constructed and by tumor-derivedp53 mutants,implyingthat dominantnegativemutantsact characterized a number of p53 proteins, the oligomerizalion of which is by hetero-oligomeriñngwithwild-type p53. The abffity of the en@neered driven by the tetrameric or by the native GCN4 LZ. Our results indicate zipper to drive tetramerization was critical for p53 function, since p53 dimers, that the engineered TZ, but not the LZ, can drive p53 function in tumor formed by substituting the p53 tetramerization domarn with a native leucine cells and furthermore can alleviate transdominant inhibition by tumor zipper, were weak tumor suppressors. derived p53 mutants. We have thus demonstrated the feasibility of designing a p53 protein with dominant wild-type function.

INTRODUCTION MATERIALS AND METHODS Wild-type p53 is a sequence-specific transcription factor that in duces cell cycle arrest or programmed cell death in response to DNA Recombinant Plasmids. Standard cloning procedures were used, and all damage (1—6).The NH2 terminus of p53 contains a transactivation mutants were generated by PCR-directed mutagenesis (42). Plasmids pGEMhpS3wtB, pGEMhp53A344, and pGEMhp53LZ335Q encode human domain (7, 8), the COOH terminus contains a tetramerization domain wild-type PS3, p53Ala344, and p53LZ335Q, respectively, and have been (9—12), and the central region contains a sequence-specific DNA described (43). Plasmid pGEMhp53LZ343RMKQ encodes a hybrid protein binding domain (13—16).The latter domain is inactivated by point consisting of residues 1—343of human p53 and residues 249—281 of GCN4 mutations in about one-half of all human tumors (17—20). (44). Because the LZ segment in p53LZ335Q refers to residues 253—281of The tumor-derived p53 mutants fail to suppress tumor growth GCN4, for uniformity we consider p53LZ343RMKQ as having a LZ corre (21—25)but in addition act as dominant negatives by inhibiting the sponding to residues 253—281ofGCN4 and a linker consisting of residues function of wild-type p53 (26—29).This latter property may account 249—252 of GCN4, which are Arg-Met-Lys-Gln (RMKQ). Plasmid for their oncogenic activity in primary cells (30—32).Mechanistically pGEMhp53TZ334NR was derived from pGEMhp53wtB by replacing the SstII-Sall fragment with synthetic oligonucleotides encoding the TZ (41) transdominant inhibition of wild-type p53 could be mediated by corresponding to residues 250—281 of GCN4 (44) via an Asn-Arg (NR) sequestration of wild-type p53 into inactive mutant/wild-type het sequence. Plasmid pGEMhp53TZ334NRII352 was derived by inserting erotetramers or by titration of targets required for wild-type p53 sequences encoding an Ile (I), followed by residues 352—393ofhuman p53 function. In favor of the first mechanism, the oncogenic activity of after the last codon in pGEMhp53TZ334NR. tumor-derived p53 mutants requires an intact tetramerization domain, Plasmid pSV2 hp53wtB was prepared by cloning into the SaII-BglII sites of and the isolated tetramerization domain is as transforming as full pSV2 humjun (45) a blunted EcoRI-HindIII p53 insert from pGEMbp53wtB length pS3 mutants (33—35). (43). Plasmids expressing p53 fusion proteins were similarly derived from the Induction of wild-type p53 function in tumor cells leads to growth corresponding pGEM plasmids. A pSV2 plasmid without insert was prepared by ligating pSV2 humjun (45) linearized with Sail and BglII. The tumor arrest or apoptosis (21—22, 24, 36, 37). Thus, introduction of wild-type derived p53 mutation Trp248 was introduced in the context of pGEMhp53wtB p53 into tumor cells could in principle be used for therapeutic pur and pSV2 hp53wtB by site-directed mutagenesis. poses(38—40).Oneobstacleto the effectivenessof such therapy, Plasmids pEp2l/TKseap and pEmdm2lTKseap have one copy of oligonu however, is that about one-half of all human tumors express domi cleotides Ep21 and Emdm2, respectively, cloned in the EcoRV site of pTKseap nant-negative p53 mutants (17—20).Therefore, we wondered whether (46) and express secreted alkaline phosphatase in a p53-responsive manner. it would be possible to engineer a p53 protein with wild-type function The sequences of oligonucleotides Ep21 and Emdm2 are CCC-GAACA that would not be susceptible to transdominant inhibition. TGTCC-CAACA-TGTTG-GGG and GGCT-GGTCA-AGTTG-GGACA One approach to prevent wild-type p53 from interacting with tu CGTCC-GGCGTCGGCTGTCGGAG-GAGCT-A-AGTCC-TGACA-TG mor-derived mutants would be to substitute its native tetramerization TCT-CCAG, respectively. The repeats recognized by p53 are underlined. In Vitro Translation, Subunit Stoichiometry, and DNA Binding. Plas domain with a heterologous oligomerization domain. Recently, an mids of the pGEM series were used to generate in vitro-translated p53 proteins (16, 43, 47). For analysis of subunit stoichiometry, the p53 proteins were Received 8/22/95: accepted 10/25/95. translated in the presence of [35S]methionineand subjected to electrophoresis The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with on native Tris-glycine gels (16). The logarithms of electrophoretic migration I8 U.S.C. Section 1734 solely to indicate this fact. were plotted as a function of the percentage of polyacrylamide by linear I Financial support was provided by Bayer AG. regression analysis (48). 2 To whom requests for reprints should be addressed, at Department of Molecular Oncology, The Wistar Institute, Room 115, 3601 Spruce Street, Philadelphia, PA 19104- 4268. 3 The abbreviations used are: LZ, leucine zipper: TZ, tetrameric zipper. 158

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. DOMINANT WILD-TYPE p53

For analysis of DNA-binding activity, the p53 proteins were incubated with Trans- DNA Tetrame 32P-labeled oligonucleotides and subjected to electrophoresis (16, 43, 47). activation Binding rlzatlon Oligonucleotides BC.V4A.S10 (43), Ep21, and Emdm2 (see above) each @ contain a p53-binding site, whereas oligonucleotide TF3 is a nonspecific DNA p53wt I — I I IIII (47). 90 290 322 355 393 Transcription and Tumor Suppression Assays. Transcriptionalactivity was determined by transfecting in quadruplicate Saos-2 cells with 5 @gof p53 p53TZ334NR expression and 25 @gof reporter plasmids. Alkaline phosphatase activity was 334 determined 48 h later (46). To assay inhibition of transcriptional activities of functional p53 proteins by p53Trp248, Saos-2 cells were cotransfected in p53TZ334NR III ...... -.-..— triplicate with 1 p@gofp53 expression plasmid, 9 @.tgofplasmid expressing the /1352 p53Trp248 mutant, or 9 ,.Lgof pSV2 plasmid without insert and 20 @gofthe Ep2lfl'K-seap reporter plasmid. p53LZ335Q @334352— Tumor suppressor activity was assayed by cotransfecting Saos-2 cells in quadruplicate with 5 @zgofp53 expression plasmid, 1 p@gofpSV7 neo, a 335 plasmid that confers neomycin resistance (45), and 24 @gof pBCl2IPLseap p53LZ343RMKQ I. .... — (46), a carrier plasmid. The transfected cells were selected for G4l8 resistance, and 2 weeks later, the resistant colonies were stained with crystal violet and 343 counted. To assay inhibition of tumor suppressor activities of functional p53 Fig. 1. Constructed p53 hybrids. The hybrid proteins maintain the transactivation and proteins by p53Trp248, Saos-2 cells were cotransfected in triplicate with 2.5 DNA binding (checkered) domains of wild-type (wt) p53 but substitute most of its I.Lg of p53 expression plasmid, 10 @.tgof plasmid expressing p53Trp248 or 10 tetramerization domain (filled) with a zipper (hatched). The names of the hybrid proteins 1.Lg of pSV2 plasmid without insert, I @zg of pSV7 neo, and 16.5 @g of indicate the type of zipper (‘I'ZorLZ), the segment(s) of p53 retained, and the sequence(s) pBCI2/PLseap carrier plasmid. of amino acids (using the single letter code) between the p53 and GCN4 segments. The LZ segment refers to residues 253—281ofGCN4, whereas the 12 corresponds to residues Arrest of cell cycle progression was determined by transfecting Saos-2 cells 250—281of GCN4. The numbering in the p53-zipper hybrid proteins refers to amino acid in duplicate with 30 @gofp53 expression plasmid. Thirty-six h later, the cells positions of wild-type p53. were trypsinized, washed with PBS, and fixed with 0.4% paraformaldehyde at 37°Cfor 10 mm and then with methanol at 4°Cfor 10 mm (49). The cells were then washed with PBS/0.l% Tween 20 (PBS/i') and incubated with antibody pS3LZ343RMKQ, which is also named p53-CC 1—343(58), and DO-l (Oncogene Sciences, Uniondale, NY) in PBS/T with 50% fetal bovine p53LZ335Q (16, 43) assemble as dimers and can serve as controls for serum at 30°C for 3 h. Excess antibody was washed with PBSTF, and the cells were incubated with fluorescein-conjugated anti-mouse F(ab')2 (Boehringer the TZ hybrids (Fig. 1). Mannheim, Indianapolis, IN) diluted 1:100 in PBS,'!' with 50% fetal bovine To examine ifthe engineered GCN4 zipper could drive tetramerization serum at 37°C for 1 h. Excess secondary antibody was washed with PBS/T, of p53, we determined the subunit stoichiometry of pS3TZ334NR by and the cells were incubated with RNase A at 37°Cfor 30 mm and stained with plotting the logarithm of its electrophoretic ml@ration on native gels propidium iodide. The cells were analyzed on a Becton Dickinson FACScan. relative to gel polyacrylamide content. The slope of this curve is propor tional to molecular size (59, 60). As standards we used wild-type p53, RESULTS p53A344, and pS3z@333—393.Wild-typep53 assembles as a tetramer (10—12,16); p53A344 has a single amino acid substitution (Leu 344 to An Engineered Coiled Coil Drives p53 Tetramerization in Vitro. Ala) in the tetramerization domain and assembles as a dimer (43), The tetramerization domain of human p53 maps to residues 322—355 whereas p53L@333—393corresponds to pS3TZ334NR without the TZ and at the p53 COOH terminus (1 1, 12). In an effort to generate p53 is a monomer, because it is mlssh@gmost of the p53 tetramerization proteins that do not rely on the native p53 tetramerization domain for domain (9, 58). On native gels, p53'17.334NR resolved as two species oligomerization, we substituted the p53 COOH-terminal sequences (Fig. 2A). Comparison of the ml@r@tionof these two species relative to with heterologous oligomerization domains derived from the tran the ml@ration of the p53 standards revealed that the major species of scription factor GCN4 (Fig. 1). pS3TZ334NR (slow form) was a tetramer, while the mmnorspecies (fast Oligomerization of GCN4 is mediated by a LZ, a domain charac form) was a dimer (Fig. 2B). terized by heptad repeats of the general sequence a.b.c.d.e.f.g, with Functional Analysis of p53-TZ Hybrids. The ability of hydrophobic residues at positions a and d and polar residues at all p53TZ334NR to bind DNA was examined in an electrophoretic mobility other positions. The native GCN4 LZ has mostly valines at positions shift assay (Fig. 3A). Oligonucleotide Ep21, which contains the p53 site a and leucines at positions d and assembles as a paralleldimer (44, of the p21 gene (61), was recognized efficiently by pS317.334NR. DNA 50—53).Interestingly, a mutant GCN4 zipper (tetrameric zipper) with binding was sequence specific, since it was not inhibited by excess of leucines at positions a and isoleucines at positions d oligomerizes as nonspecific competitor DNA (oligonucleotide TF3). The monomeric a parallel tetrarner in vitro (41). p53A333—393,which is identical to pS3TZ334NR except that it lacks the Since wild-type p53 forms homotetramers (10—12, 16), we rea 17, failedto bindDNA (Fig.3A),highlightingtheimportanceoftheTZ soned that the TZ might substitute for the native p53 tetramerization and confirming previous observations that oligomenzation facilitates p53 domain better than the LZ. Therefore, we generated a protein, DNA binding (16, 62). Like p5317.334NR, p53'fZ334NRII352 also p53TZ334NR, consisting of residues 1—334of human p53, an Asn recognized DNA with high affinity and in a sequence-specific manner. Arg (NR) sequence, and a COOH-terminal tetrameric zipper (Fig. 1). However, efficient DNA-binding activity was dependent on the presence This protein lacks almost the entire native p53 tetramerization domain of antibody PAb421 (Fig. 3A). pS3TZ334NRIL3S2 contains the COOH (1 1, 12). A variant of pS3TZ334NR was also generated by fusing at terminus of wild-type p53 (residues 352—393),which suppresses p53 its COOH terminus residues 352—393of human p53 via an isoleucine DNA binding in vitro, unless masked by antibody PAb421 (43, 47, (I). This variant, p53TZ334NRII352, contains the p53 sequences 55—57).ThisCOOH-terminal region is absentin pS3TZ334NR (Fig. 1). COOH-terminal to the tetramenzation domain, which encode two The DNA-binding activities of pS3TZI34NR and wild-type p53 nuclear localization signals (54) and a region that regulates DNA were subsequently compared using a panel of DNA sites: an artificial binding (47, 55—57).We and others have also previously constructed suboptimal pS3 site present within oligonucleotide BC.V4A.S10 (43); p53 hybrids with the native LZ of GCN4. Two such hybrids, and the p53 sites of the p21 and mdm2 genes present within oligonu 159

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. DOMINANT WILD-TYPE p53 A Protein human cancer, was sufficient to abrogate the tumor suppressor activity of pS3TZ334NR (Fig. 4B). The same substitution also abrogated the None tumor suppressor activity of wild-type p53, as expected (21—25). p53wt I— Interestingly, the dimenc proteins p53LZ343RMKQ and p53LZ335Q were weak tumor suppressors (Fig. 4B). p53wtA333-393 S The differences in transcriptional and tumor suppressor activities be @ p53TZ334NR O • tween the examined p53 proteins could not be attributed to differences in p53A344 S expression. They were all expressed at equivalent levels as determined by @ AA flow cytometry (data not shown, but see Fig. 4C below). OS • To analyze cell cycle arrest, Saos-2 cells were transfected with p53 B expression plasmids and examined by flow cytometry 36 h later. Cells expressing p53 were identified with DO-I, a monoclonal antibody that E recognizes the NH2 terminus of human p53 (56). DNA content was E 1.8 determined by propidium iodide staining and plotted relative to C 0 @ Ntk%NNQNNNNt\NNtk% p53A333393 15 1.5 A PAb Comp. @ a) p53A344-f Protein DNA E p53TZ334N R-f None - a, 1.2 p53A344-s 0 p53wth333.393- p53TZ334NR-s p53TZ334NR - p53wt TF3 0.9 — — — I Ep21 p53TZ334NR - 5 6 7 %T /1352 + Fig. 2. Subunit stoichiometry of pS3TZ334NR. A, migration of 35S-labeledwild-type + TF3 P p53 (0), p53L@333—393(/2s),p53TZ334NR (A, for its slow and fast forms) and p53A344 + Ep21 (C. fortheslowandfastboundariesofitsbroadband)onanativeTris-glycine5.4% I polyacrylamide gel. B, logarithm of electrophoretic migration versus the percentage of B polyacrylamide (%T). Slope values were -0.297 for wild-type p53, -0.278 and -0.198 for 32p DNA ProteIn the slow (s) and fast (f) forms of p53TZ334NR, -0.194 and -0.190 for the slow (s) and fast (f) forms of p53A344, and -0.113 for pS3@333—393. 4 None @@ p53TZ334NR cleotides Ep2 1 and Emdm2, respectively (61 , 63). Both pS3TZ334NR @@(l)p53wt @ and wild-type pS3 recognized these sites with equal efficiency (Fig. p53wt+PAb42I 3B). Binding of wild-type p53 to oligonucleotides BC.V4A.S10 or None Ep21 required the presence of monoclonal antibody PAb421, consist ;; p53TZ334NR $1 ent with previous reports on regulation of p53 DNA binding in vitro I@. p53wt by its COOH terminus (43, 47, 55—57). p53wt+PAb421 a We subsequently examined the DNA-binding activities of the 01 None .c1@ p53-LZ hybrids pS3LZ33SQ and p53LZ343RMKQ. Both recognized .@ p53TZ334NR S E p53wt oligonucleotides Ep21 and Emdm2 less efficiently than wild-type p53 Ui I (Fig. 3C). Since p53LZ335Q and pS3TZ334NR contain essentially p53wt+PAb421 a the same p53 segment fused either to a LZ or to a TZ, we attribute the C lower affinities of the LZ hybrids for DNA to their dimeric subunit 32PDNA Protein stoichiometry. The transcriptional activities of pS3TZ334NR and pS3TZ334NRIL3S2 None

were examined by transient transfection in Saos-2 osteosarcoma cells, ,- p53wt which lack endogenous p53 (21). Both proteins activated transcription @. p53wt+PAb421 • from reporter plasmids containing the p21 or mdm2 p53 sites (Fig. 4A). p53LZ343RMKQ The dimeric pS3-LZ hybrids were less potent transcriptional activators, S p53LZ335Q I especially with the reporter plasmid containing the mdm2 site (Fig. 4A). Transcriptional activity for all p53 proteins examined was sequence None I specific, since none of them activated transcription from a reporter c@ p53wt plasmid that lacked a p53 site (data not shown). .@ p53wt+PAb421 Wild-type p53 inhibits colony formation of tumor cells and arrests S Ui p53LZ343RMKQ their progression through the cell cycle (21—22,24). For the colony p53LZ335Q forming assay, Saos-2 cells were cotransfected with a plasmid direct ing expression of the p53-zipper hybrid and a plasmid conferring G4l8 resistance. The number of G418-resistant colonies is inversely Fig. 3. DNA binding of p53-zipper hybrids. A. binding to 32P-labeled oligonucleotide Ep21. Where indicated, 50-fold excess unlabeled specific (Ep21) or nonspecific (TF3) related to tumor suppressor activity. Both p53TZ334NRII352 and competitor (Comp. ) DNA was added. B and C. binding to 32P-labeled oligonucleotides p53TZ334NR suppressed tumor growth almost as efficiently as wild BC.V4A.S10, Ep21 and Emdm2. Where indicated, 0. 1 ,@gantibody PAb42l was added. The names of the p53 fusion proteins indicate the type of zipper (LZ or TZ), the type p53 (Fig. 4B). A single amino acid substitution, Trp248 (W248), segment(s) of p53 retained, and the residues (using the single-letter code) between the p53 which maps to the p53 DNA-binding domain and is associated with and zipper segments. 160

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. DOMINANT WILD-TYPE p53

A C Protein Transcriptional Activity p53wt

p53TZ334NR p53wt C p53TZ334NR/1352 15 1'C 0 p53LZ343RMKQ C)

z @ p53LZ335Q Ep210Emdm2•0 None

p53W248 — p53TZ334NR/1352p53W248TZ334NR 0 20 40 60 80 100

B Protein Colonies (xlO)

p53TZ334NR

G2

p53LZ343RMKQ Gi

p53- p53+ p53 content 0 15 30 45 60 75 90

Fig. 4. Functional activities of p53-zipper hybrids in Saos-2 cells. A, transcriptional activities from reporter plasmids containing Ep21 or Emdm2 enhancer elements, presented as means + I SE (bars) relative to wild-type p53. B, tumor suppressor activities using the colony-forming assay presented as means + I SE (bars) of G4l8-resistant colonies per plate. C. effect of p53 expression on cell cycle distribution. Expression of p53 (logarithmic scale) is indicated on the horizontal axis. DNA content (linear scale, vertical axis) is indicated by GI for cells in the G1 phase of the cycle and G2 for cells in the G2-M phases. Cells in the S phase have a DNA content intermediate between G1 and G2 cells. Cells transfected with a pSV2 plasmid without insert (None) provide a reference for lack of p53 expression.

p53 content. Cells expressing wild-type p53, pS3TZ334NR, or protein was not straightforward. First there is a paucity of independ pS3TZ334NR/I352 were arrested in G1 (Fig. 4C). In contrast, cells ently folding native tetramerization domains that could be grafted to expressing the tumor-derived mutant p53Trp248 (p53W248) or p53. For this reason, we initially focused on LZ hybrid proteins. p53TZ334NR with the Trp248 substitution were distributed in all However, functional analysis of hybrids, such as pS3LZ33SQ and phases of the cell cycle, as were cells that were not transfected or did pS3LZ343RMKQ, revealed that their DNA binding and transcrip not express significant levels of p53 (Fig. 4C). tional activities were impaired relative to wild-type p53. Furthermore, To determine if the TZ alleviates transdominant inhibition by although p53LZ343RMKQ (referred to as p53-CC 1—343inRef. 58) tumor-derived mutants, we examined the transcriptional and tumor suppresses colony formation of H1299 lung adenocarcinoma cells suppressor activities of pS3-TZ hybrids in the presence of p53Trp248 efficiently (58), we only observed weak tumor suppressor activity (p53W248). Transcriptional activity was assayed with a reporter plas with Saos-2 osteosarcoma cells (Fig. 4B). The discrepancy may mid containing the p53 site of p21; tumor suppressor activity was reflect the sensitivities of the two cell lines to wild-type p53 activity assayed by the colony-forming assay. In contrast to wild-type p53, or the level of p53 expression. Irrespective of the reason, the dimeric neither pS3TZ334NR nor p53TZ334NRI1352 were transdominantly p53 proteins are not as potent as wild-type p53. This conclusion is inhibited by p53Trp248 (Fig. 5). Other dominant-negative, tumor further supported by comparing the DNA-binding activities of these derived p53 mutants also failed to inhibit pS3TZ334NR and proteins in vitro (Fig. 3C). The importance of tetrameric subunit p53TZ334NR/1352 (data not shown). stoichiometry can be rationalized since the p53 sites identified in regulatory sequences of genes contain four specific repeats (61 , 63— DISCUSSION 65). If the affinity of a single DNA-binding domain for a repeat is the Design of a p53 Protein with Dominant Wild-Type Activity. same in the context of p53 dimers and tetramers, then the avidity for Tumor-derived p53 mutants and wild-type p53 form inactive het DNA increases in parallel to the number of recruited DNA-binding erotetramers (26, 29). We, therefore, reasoned that a necessary step in domains. the design of a p53 protein with dominant wild-type activity would be Our inability to generate a pS3-LZ hybrid protein with the desired to substitute the tetramerization domain of wild-type p53 with a properties turned our attention to a mutant GCN4 zipper that contains heterologous oligomerization domain. The design of such a p53 leucines at positions a of the coiled coil and isoleucines at positions d 161

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. DOMINANT WILD-TYPE p53

and assembles as a tetramer in vitro (41). Several properties of this mutant zipper make it appealing as a surrogate tetramerization domain for pS3: (a) it is an independently folding domain, and therefore, it can be grafted to p53; (b) its tetrameric subunit stoichiometry matches the stoichiometry of the native p53 oligomerization domain; (c) its engineered nature decreases the probability that the hybrid p53 pro teins will hetero-oligomerize with cellular proteins; and (d) its topol ogy, a parallel four-stranded coiled coil (41), favors alignment of the p53 DNA-binding domains to DNA (Fig. 6). Certainly an unknown parameter in selecting the TZ was the affinity with which it would drive p53 tetramerization in vitro and more so in vivo. An engineered heterodimerization domain, the design of which was based on the natural heterodimeric Jun and Fos LZs, forms stable dimers in vitro and in cells (66, 67). However, the 17 hasno naturalcounterpart. Remarkably, it drove efficient tetramerization of p53 (Fig. 2). Although we did not quantitate the association constant. it must compare favorably to the constant of the native p53 tetramerization domain, since both domains drove wild-type p53 function with essentially equal efficiency in vitro and in tumor cells (Figs. 3 and 4). Insights into p53 Biology and Prospects for Gene Therapy. The functional properties of the p53-zipper hybrid proteins provide in Fig. 6. Predicted topology of pS3-TZ hybrid proteins. Wild-type p53 (upperpanel) and sights into the biology of p53: (a) they demonstrate the importance of a pS3-TZ hybrid (lower panel) unbound (left) and bound to DNA (right). The p53 tetramerization for efficient p53 function, since the dimeric hybrids tetramerization domains are indicated by their secondary structure elements (rectangles were significantly less active than the tetrameric ones; and (b) they for the (3-strandsand cylinders for the a-helices). The DNA-binding domains are shown as spheres, and the linkers connecting the DNA binding and tetramerization domains are provide a better understanding of the mechanisms by which tumor shown as curved lines. The DNA is shown as a rectangle with arrows representing the derived p53 mutants transform cells. Tumor-derived p53 mutants lack pentanucleotide repeats recognized by p53. tumor suppressor activity (17—25)and also transdominantly inhibit wild-type p53 (26—29). Remarkably, they can even enhance the about one-half of all human tumors express p53 mutants that can tumorigenic potential of cells that do not express endogenous p53 (68, inhibit wild-type p53 (17, 18). Furthermore, it is possible to apply the 69). Substitution of the native tetramerization domain of wild-type paradigm established here with p53 and the engineered TZ to other p53 by the TZ zipper was sufficient to confer dominant tumor suppressor human diseases in which a wild-type protein is sequestered by a activity. We can, therefore, conclude that transdominant inhibition of dominant-negative mutant. wild-type p53 by tumor-derived p53 mutants is mediated solely through hetero-oligomerization and furthermore that any oncogenic activity that tumor-derived p53 mutants may have independent of transdominant ACKNOWLEDGMENTS inhibition must be recessive to wild-type p53 function. We thank Giovanni Rovera, Philip Leder, Guenther Karmann, Phillip Sharp, The pS3-TZ hybrid proteins can be used to study p53 function in Jules Shafer, Frank Rauscher, III, Jill Shenk, and Elena Stavridi for support the background of dominant-negative, tumor-derived p53 mutants. and discussions. Saos-2 cells were kindly provided by David Livingston and However, their greatest significance may relate to the possibility of Steve Friend. using these proteins in a gene therapy approach for treatment of human cancer (38—40).A p53 protein with dominant wild-type func REFERENCES tion would be clearly superior to wild-type p53 for such therapy, since I. Maltzman, W., and Czyzyk, L. UV irradiation stimulates levels of p53 cellular tumor antigen in nontransformed mouse cells. Mol. Cell. Biol., 4: 1689—1694,1984. 2. Kastan, M. B., Onyekwere, 0., Sidransky, D., Vogelstein, B., and Craig, R. W. Participation ofp53 protein in the cellular response to DNA damage. Cancer Res., 51: Protein W248 Activity 6304—6311,1991. C 3. Lowe, S. W., Ruley, H. E., Jacks, T., and Housman, D. E. p53-dependent apoptosis @ p53wt modulates the cytotoxicity of anticancer agents. Cell, 74: 957-967, 1993. 4. Lowe, S. W., Schmitt, E. M., Smith, S. W., Osbome, B. A., and Jacks, T. p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature (Lond.), 362:

.@ p53TZ334NR 847—849,1993. 15 5. Clarke, A. R., Purdue, C. A., Harrison, D. J., Morris, R. 0., Bird, C. C., Hooper, M. L., and Wyllie, A. H. Thymocyte apoptosis induced by p53-dependent and independent @ p53TZ334NR - pathways. Nature (Lond.), 362: 849—852,1993. I@ /1352 + 6. Leonardo, A. D., Linke, S. P., Clarkin, K., and WahI, 0. M. DNA damage triggers a prolonged p53-dependent G@arrest and long-term induction of CipI in normal human

@. p53wt fibroblasts. Genes & Dcv., 8: 2540—2551,1994. + 7. Fields,S.,andJang,S.K. Presenceofapotenttranscriptionactivatingsequenceinthe Eo. p53 protein. Science (Washington DC), 249: 1046—1049, 1990. @@ p53TZ334NR 8. Unger, T., Nau, M. N., Segal, S., and Minna, J. D. p53: a transdominant regulator of I—Cl) /1352 + transcription whose function is ablated by mutations occurring in human cancer. EMBOJ., 11: 1383—1390,1992. 0 25 50 75 100125150 9. Milner, J., Medcalf, E. A., and Cook, A. C. Tumor suppressor p53: analysis of wild-type and mutant p53 complexes. Mol. Cell. Biol.. 11: 12—19,1991. Fig. 5. Transcriptional and tumor suppressor activities of pS3-TZ hybrids in the 10. StUrzbecher, H. W., Brain, R., Addison, C., Rudge, K., Remm, M., Grimaldi, M., presence of p53 tumor-derived mutant Trp248 (W248). The results are presented as Keenan, E., and Jenkins, J. R. A C-terminal a-helix plus basic region motif is the means + 1 SE (bars) relative to activity in the absence of p53Trp248. Tumor suppressor major structural determinant of p53 tetramerization. Oncogene, 7: 1513—1523, activity was assayed in Saos-2 cells. A 2-fold increase in the number of 0418-resistant 1992. colonies in the presence of p53Trp248 represents a 50% reduction in tumor suppressor 11. Sakamoto, H., Lewis, M. S., Kodama, H., Appella, E., and Sakaguchi, K. Specific activity. sequences from the carboxyl terminus of human p53 gene product form anti-parallel 162

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. DOMINANT WILD-TYPE p53

tetramers in solution. Proc. NatI. Acad. Sci. USA, 91: 8974—8978, 1994. adenovirus-mediated transfer of the wild-type p53 gene. Cancer Res., 54: 2287—2291, I2. Wang, P., Reed, M., Wang, Y.. Mayr, 0., Stenger, J. E., Anderson, M. E., Schwedes, 1994. J. F., and Tegtmeyer. P. p53 domains: structure. oligomerization and transformation. 40. Liu, T. J., Zhang, W. W., Taylor, D. L., Roth, J. A., Goepfert, H., and Clayman, 0. L. Mol. Cell. Biol., 14: 5182—5191, 1994. Growth suppression of human head and neck cancer cells by the introduction of a I3. Bargonetti. J., Manfredi, J. J., Chen, X., Marshak. D. R., and Prives, C. A proteolytic wild-type p53 gene via a recombinant adenovirus. Cancer Res., 54: 3662—3667,1994. fragment from the central region of p53 has marked sequence-specific DNA-binding 41. Harbury, P. B., Zhang, T., Kim, P. S., and Alber, T. A switch between two-, three-, activity when generated from wild-type but not from oncogenic mutant p53 protein. and four-stranded coiled coils in GCN4 leucine zipper mutants. Science (Washington Genes & Dcv., 7: 2565—2574, 1993. DC), 262: 1401—1407,1993. 14. Pavletich, N. P., Chambers, K. A., and Pabo, C. 0. The DNA-binding domain of p53 42. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. 0., Smith, J. A., contains the four conserved regions and the major mutation hot spots. Genes & Dcv., and Struhl, K. Current Protocols in Molecular Biology. New York: Greene/Wiley 7: 2556—2564, 1993. Publishing Associates, 1994. 15. Wang. Y., Reed, M.. Wang, P., Stenger, J. E., Mayr, 0., Anderson, M. E., Schwedes, 43. Waterman, J. L. F., Shenk, J. L., and Halazonetis, T. D. The dihedral symmetry of the J. F., and Tegtmeyer. P. p53 domains: identification and characterization of two p53 tetramerization domain mandates a conformational switch upon DNA binding. autonomous DNA-binding regions. Genes & Dcv., 7: 2575—2586, 1993. EMBO J., 14: 512—519,1995. 16. Halazonetis, T. D., and Kandil, A. N. Conformational shifts propagate from the 44. Ellenberger, T. E., Brandl, C. J., Struhl, K., and Harrison, S. C. The GCN4 basic oligomerization domain of p53 to its tetrameric DNA binding domain and restore region leucine zipper binds DNA as a dimer of uninterrupted a helices: crystal DNA binding to select p53 mutants. EMBO 3., 12: 5057—5064, 1993. structure of the protein-DNA complex. Cell, 71: 1223—1237,1992. I7. Hams, C. C. p53: at the crossroads of molecular carcinogenesis and risk assessment. 45. Zhang, K., Chaillet, J. R., Perkins, L. A., Halazonetis, T. D., and Pemmon, N. Science (Washington DC). 262: 1980—1981, 1993. Drosophila homolog of the mammalian jun oncogene is expressed during embryonic 18. Friend, S. p53: a glimpse at the puppet behind the shadow play. Science (Washington development and activates transcription in mammalian cells. Proc. NatI. Acad. Sci. DC), 265: 334—335, 1994. USA, 87: 6281—6285,1990. 19. Bargoneui, J., Friedman, P. N., Kern, S. E., Vogelstein, B., and Prives, C. Wild-type 46. Halazonetis, T. D. An enhancer “core―DNA-bindingand transcriptional activity is but not mutant p53 immunopurified proteins bind to sequences adjacent to the SV4O induced upon transformation of rat embryo fibroblasts. Anticancer Res., 12: 285—292, origin of replication. Cell, 65: 1083—1091,1991. 1992. 20. Kern, S. E., Kinzler, K. W., Bruskin, A., Jarosz, D., Friedman, P., Prives, C., and 47. Halazonetis, T. D., Davis, L. J., and Kandil, A. N. Wild-type p53 adopts a “mutant―- Vogelstein. B. Identification of p53 as a sequence-specific DNA-binding protein. like conformation when bound to DNA. EMBO J., 12: 1021—1028,1993. Science (Washington DC). 252: 1708—1711, 1991. 48. Snedecor, 0. W., and Cochran, W. 0. Statistical Methods, Ed. 6. Ames, IA: Iowa 21. Diller. L., Kassel, J., Nelson, C. E., Gryka, M. A., Litwak, 0., Gebhardt, M., Bressac, State University Press. 1967. B., Ozturk, M., Baker. S. J., Vogelstein, B., and Friend, S. H. p53 functions as a cell 49. Schimenti, K. J., and Jacobberger, J. W. Fixation of mammalian cells for flow cycle control protein in osteosarcomas. Mol. Cell. Biol., 10: 5772—5781, 1990. cytometric evaluation of DNA content and nuclear immunofluorescence. Cytometry, 22. Baker, S. J., Markowitz, S., Fearon, E. R., Willson, J. K. V., and Vogelstein, B. 13: 48—59,1992. Suppression of human colorectal carcinoma cell growth by wild-type p53. Science 50. Landschulz, W. H., Johnson, P. F., and McKnight, S. L. The leucine zipper: a (Washington DC). 249: 912—915,1990. hypothetical structure common to a new class of DNA binding proteins. Science 23. Eliyahu, D., Michalovitz, D., Eliyahu, S., Pinha.si-Kimhi, 0., and Oren, M. Wild-type (Washington DC), 240: 1759—1764, 1988. p53 can inhibit oncogene-mediated focus formation. Proc. Nail. Acad. Sci. USA, 86: 51. Hope, I. A., and Struhl, K. GCN4, a eukaryotic transcriptional activator protein, binds 8763—8767,1989. as a dimer to target DNA. EMBO J., 6: 2781—2784, 1987. 24. Mercer, W. E., Shields, M. T., Amin, M., Sauve, 0. J., Appella, E., Romano, J. W., 52. O'Shea, E. K., Rutkowski, R., and Kim, P. S. Evidence that the leucine zipper is a and Ullrich, S. J. Negative growth regulation in a glioblastoma tumor cell line that coiled coil. Science (Washington DC), 243: 538—542, 1989. conditionally expresses human wild-type p53. Proc. NatI. Acad. Sci. USA, 87: 53. O'Shea, E. K., Klemm, J. D., Kim, P. S., and Alber, T. X-ray stnicture of the GCN4 6166—6170, 1990. leucine zipper, a two-stranded, parallel coiled coil. Science (Washington DC), 254: 25. Finlay. C. A., Hinds, P. W., and Levine, A. J. The p.53 proto-oncogene can act as a 539—544, 1991. suppressor of transformation. Cell, 57: 1083—1093,1989. 26. Martinez,J.,Georgoff.I.,Martinez,J.,andLevine,A.J.Cellularlocalizationandcell 54. Shaulsky, 0., Goldfinger, N., Ben-Ze'ev, A., and Rotter, V. Nuclear accumulation of p53 protein is mediated by several nuclear localization signals and plays a role in cycle regulation by a temperature-sensitive p53 protein. Genes & Dcv., 5: 151—159, tumorigenesis. Mol. Cell. Biol., 10: 6565—6577, 1990. 1991. 27. Bargonetti.J.. Reynisdbttir.I.. Friedman,P.N., andPrives,C. Site-specificbinding 55. Funk, W. D., Pak, D. T., Karas, R. H., Wright, W. E., and Shay, J. W. A transcrip tionally active DNA-binding site for human p53 protein complexes. Mol. Cell. Biol., of wild-type p53 to cellular DNA is inhibited by SV4O T antigen and mutant p53. /2: 2866—2871,1992. Genes & Dcv., 6: 1886—1898, 1992. 28. Kern, S. E.. Pietenpol. J. A., Thiagalingam, S., Seymour. A., Kinzler, K. W., and 56. Hupp, T. R., Meek, D. W., Midgley, C. A., and Lane, D. P. Regulation of the specific Vogelstein. B. Oncogenic forms of p53 inhibit p53-regulated gene expression. DNA binding function of p53. Cell, 71: 875—886,1992. Science (Washington DC), 256: 827—830, 1992. 57. Hupp, T. R., and Lane, D. P. Allosteric activation of latent p53 tetramers. Current 29. Milner, J., and Medcalf, E. A. Cotranslation of activated mutant p53 with wild-type Biol., 4: 865—875, 1994. drives the wild-type p53 protein into the mutant conformation. Cell, 65: 765—774, 58. Pietenpol. J. A., Tokino, T., Thiagalingam, S., El-Deity, W. S., Kinzler, K. W., and 1991. Vogelstein, B. Sequence-specific transcriptional activation is essential for growth 30. Eliyahu. D.. Raz. A., Gross, P.. Givol, D., and Oren, M. Participation of p53 cellular suppression by p53. Proc. Nail. Aced. Sci. USA, 91: 1998—2002, 1994. tumour antigen in transformation of normal embryonic cells. Nature (Lond.), 312: 59. Chrambach, A. The Practice of Quantitative Gel Electrophoresis, pp. 111—128. 646—649,1984. Weinheim, Germany: VCH Verlagsgesellschaft, 1985. 31. Parada, L. F., Land, H., Weinberg. R. A., Wolf, D., and Rotter, V. Cooperation 60. Hedrick, J. L., and Smith, A. J. Size and charge isomer separation and estimation of between gene encoding p53 tumor antigen and ras in cellular transformation. Nature molecular weights of proteins by disc gel electrophoresis. Arch. Biochem. Biophys., (Lond.),312:649—651,1984. 126: 155—164,1968. 32. Jenkins, J. R.. Rudge. K.. and Currie. G. A. Cellular immortalization by a cDNA clone 61. El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., encoding the transformation-associated phosphoprotein p53. Nature (Lond.), 312: Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. WAFI, a potential 651—654,1984. mediator of p53 tumor suppression. Cell, 75: 817—825, 1993. 33. Unger, T., Mietz. J. A., Scheffner, M., Yee, C. L., and Howley, P. M. Functional 62. Hainaut, P., Hall, A., and Milner, J. Analysis of p53 quatemary structure in relation domains of wild-type and mutant p53 proteins involved in transcriptional regulation, to sequence-specific DNA binding. Oncogene, 9: 299—303, 1994. transdominant inhibition, and transformation suppression. Mol. Cell. Biol., 13: 5 186— 63. Wu, X., Bayle, J. H., Olson, D., and Levine, A. J. The p53-mdm2 autoregulatory 5194, 1993. feedback loop. Genes & Dcv., 7: 1126—1132, 1993. 34. Shaulian, E., Zauberman, A., Ginsberg, D., and Oren, M. Identification of a minimal 64. Kastan, M. B., Than, Q., El-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V., transforming domain of p53: negative dominance through abrogation of sequence Plunkett, B. S., Vogelstein, B., and Fornace, A. J. A mammalian cell cycle checkpoint specific DNA binding. Mol. Cell. Biol.. 12: 5581—5592, 1992. pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell, 71: 35. Reed, M.. Wang. Y., Mayr, G., Anderson, M. E., Schwedes, J. F., and Tegtmeyer, P. 587—597,1992. p53 domains: suppression, transformation and transactivation. Gene Expression, 3: 65. Miyashita, T., and Reed, J. C. Tumor suppressor p53 is a direct transcriptional 95—107,1993. activator of the human box gene. Cell, 80: 293—299,1995. 36. Yonish-Rouach, E., Resnitzky, D., Lotem, J., Sachs, L., Kimchi, A., and Oren, M. 66. O'Shea, E. K., Lumb, K. J., and Kim, P. S. Peptide “Velcro―:designof a het Wild-type p53 induces apoptosis of myeloid leukaemic cells that is inhibited by erodimeric coiled coil. Current Biol. 3: 658—667, 1993. interleukin-6. Nature (Lond.), 352: 345—347,1991. 67. Chang, H., Bao, Z., Yao, Y., Tse, A. 0. D., Goyarts, E. C., Madsen, M., Kawasaki, 37. Shaw, P., Bovey, R.. Tardy, S., Sabli. R.. Sordat, B., and Costa, J. Induction of E., Brauer, P. P., Sacchettini, J. C., Nathenson, S. 0., and Reinherz, E. 1. A general apoptosis by wild-type p53 in a human colon tumor-derived cell line. Proc. Natl. method for facilitating heterodimeric pairing between two proteins: application to @ Acad. Sci. USA, 89: 4495—4499, 1992. expression of a and T-cell receptor extracellular segments. Proc. NatI. Acad. Sci. 38. Fujiwara, T., Grimm. E. A., Mukhopadhyay, T., Cai, D. W., Owen-Schaub, L. B., and USA, 91: 11408—11412, 1994. Roth, J. A. A retroviral wild-type p53 expression vector penetrates human lung cancer 68. Dittmer, D., Pati, S., Zambetti, 0., Chu, S., Teresky, A. K., Moore, M., Finlay, C., and spheroids and inhibits growth by inducing apoptosis. Cancer Res., 53: 4129—4133, Levine, A. J. Gain of function mutations in p53. Nat. Genetics, 4: 42—46,1993. I993. 69. Harvey, M., Vogel, H., Morris, D., Bradley, A., Bernstein, A., and Donehower, 1. A. 39. Fujiwara. T., Grimm, E. A., Mukhopadhyay, T.. Zhang, W. W., Owen-Schaub, L. B., A mutant p53 transgene accelerates tumour development in heterozygous but not and Roth, J. A. Induction of chemosensitivity in human lung cancer cells in vivo by nullizygous p53-deficient mice. Nat. Genet., 9: 305—311,1995.

163

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research. An Engineered Four-Stranded Coiled Coil Substitutes for the Tetramerization Domain of Wild-Type p53 and Alleviates Transdominant Inhibition by Tumor-derived p53 Mutants

Matthew J. F. Waterman, Jennifer L. F. Waterman and Thanos D. Halazonetis

Cancer Res 1996;56:158-163.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/1/158

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/56/1/158. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1996 American Association for Cancer Research.