Induction of Cell Growth Regulatory Genes by P53

CANCER RESEARCH 56. 5384-5390. December 1. 996] Induction of Cell Growth Regulatory Genes by p53

Stephen L. Madden,1 Elizabeth A. Galella, Deborah Riley, Arthur H. Bertelsen, and Gary A. Beaudry

Department of Molecular and Cellular Biology. PharmaGenics. Inc., Allendale. New Jersey 07401

ABSTRACT kinase inhibitor by directly interacting with cyclin-dependent kinase proteins during progression of the cell cycle (20). Overexpressiom of Transcriptionally regulated growth-response genes play a pivotal role @2jWAF@@'I via transfection suppresses cell growth in a wide array in the determination of the fate of a cell.p53 is known to transcriptionally of cancer cell lines. In contrast, MDM2 functions by negatively regulate genes important in regulating cell growth potential. Using differ ential reverse transcription-PCR analysis of rat embryo fibroblast cells regulating p53 activity, and overexpression of MDM2 leads to un containing a temperature-sensitive p53 allele, we were able to isolate controlled cell growth and tumorigenesis (8). However, p53-induced several transcripts up-regulated specifically in cells harboring functional transcriptional responses appear ill-conserved, suggesting that cell PS3 protein. Two of these genes, SM2Oand microsomal epoxide hydrolase type-, genetic-, and species-dependent factors may contribute to p53 (mElt), are previously described genes. Two previously uncharacterized responsiveness; for example, both @2JwAFl/cIPIand MDM2 are tran cDNAs, cell growth regulatory (CGR) genes CGRJJ and CGRJ9, were scriptionally activated in a p53-dependent manner in rodent cells isolated. The predicted amino acid sequence of these novel proteins con harboring a temperature-sensitive p53 allele, but only MDM2 is tamknownmotifs;EF-handdomains(CGR11)anda ring-fingerdomaln induced in an analogous human system (30). In addition, activation of (CGRJ9),suggestiveoffunction. CGRJJand CGRJ9appearto beprimary @2jWAF1@@'I has been shown to be both p53-dependent and -mdc response genes expressed to moderate levels in functional p53 cells. Both pendent, suggesting alternate pathways for regulation of its expression CGRIJ and CGRJ9 are able to inhibit the growth of several cell lines. (31, 32). Furthermore, it appears that the levels of apoptotic regulatory gene products, such as BCL2 or E1B, are influential in determining if INTRODUCTION a cell will undergo p53-dependent growth arrest or apoptosis (9, 33). These results suggest that a complex regulatory cascade, using nu Significant advances have been made recently toward the identifi merous effector gene products, determines cell growth potential. The cation of numerous potential growth effectors within cells. Although complexities of cell growth regulation underscores the need to further previous analyses have evaluated the phenotypic effects of cellular define the relevant molecular determinants. environment, growth factors and stress, it is only recently that the The well-characterized murine temperature-sensitive p53 mutation underlying molecular events responsible for perturbing cell growth (VAL13S) has been used to further characterize potential transcrip are beginning to be defined. In a large number of cases, the cause of tional responses of wild-type p53 protein. Growth of rat embryo a particular cellular growth response can be attributed to transcrip fibroblast cells (REF-l 12) transformed with activated RAS and p.53- tional artenuation (1, 2). Indeed, the recent gene discoveries directly VAL135 is temperature sensitive ( 10, 34). Growth of REF- 112 cells relevant to cell cycle regulation have been based, to a significant at 38Â°Cmaintains the p53 protein in a nonfunctional conformation degree, on differential gene expression analyses (2â€”4). complexed with HSP7O in the cytoplasm (10, 35). When the cells are The p53 gene remains the most frequenfly mutated gene in human shifted to 32Â°C,the p53 protein adopts a functional, wild type-like cancers, undoubtedly reflecting an important regulatory function for conformation, migrates into the nucleus, and regulates the transcrip this gene in controlling cell growth (reviewed in Ref. 5). p53 function tion of target genes (10). Under these latter conditions, the cells were can be attenuated by interaction with viral or cellular proteins (6â€”9) initially described as being growth-arrested predominately in the G1 and cytoplasmic sequestration (10, 11), ultimately leading to alter phase of the cell cycle (34). However, we have observed that a ations in cell growth potential. Although p53 is thought to exert significant population of cells undergo apoptosis when maintained at growth-regulatory functions in response to DNA damage and by 32Â°C.DNA laddering indicative of apoptotic cells was observed as directly inhibiting DNA replication ( 12, 13) or inducing apoptosis (9), early as 6 h after shift of cells to the permissive temperature (data not a large body of data indicates that the primary function of p53 is to shown).2 Because previous studies have suggested that p53-induced transcriptionally regulate downstream effector genes. p53 contains a growth arrest and apoptosis might be in distinct and divergent path potent transcriptional activation domain and is able to bind DNA in a ways (12, 31, 36, 37), both apoptotic-specific as well as growth sequence-specific manner (14â€”16), allowing for both transcriptional arrest-specific genes might be elevated in this system upon shift of activation and repression of target genes (17, 18). p53-mediated p53 to a functional conformation. Utilizing differential display anal transcriptional repression has also been proposed to occur via p53 ysis of temperature-sensitive, p53-induced (32Â°C), and uninduced protein interaction with general transcription factors (19). (38Â°C)REF-l 12 cells, we have been able to demonstrate several Identification of the p53 transcriptionally responsive genes genes the transcript levels of which are significantly elevated in the p2IW@)l (2, 20), MDM2 (1, 8), GADD45 (21, 22), H/C (23), presence of functional p53. Two of these genes are previously unde cyclin G (24, 25), and BAX (26â€”29),the products of which have been scribed and encode proteins containing known functional motifs. We suggested to have direct effects on cell growth, emphasizes the pivotal demonstrate here that these genes have growth-suppressive potential role of pS3 in modulating expression levels of growth-response genes. in several cell lines examined. To date, two of the best characterized direct effectors of cell growth that are transcriptionally regulated by p53 include @2jWAFI/CIPI(2, MATERIALS AND METHODS 20) and MDM2 (I, 8). @2jWAFI/CIPlfunctions as a cyclin-dependent Cell Culture. Rat embryo fibroblast cells REF-I 12 (p53-VAL135) and Received 5/3 1/96; accepted 10/2/96. REF-132 [p53-PHE132; kindly provided by B. Vogelstein (Johns Hopkins The costs of publication of this article were defrayed in part by the payment of page University, Baltimore, MD) and M. Orem (Weizmanm Institute, Rehovot, charges. This article must therefore be hereby marked advertisement in accordance with Israel)1,were grown in DMEM containing 10%fetal bovine serum in 5% CO2 18 U.S.C. Section 1734 solely to indicate this fact. @ To whom requests for reprints should be addressed, at Department of Molecular and @ Cellular Biology, PharmaGenics, Inc., 4 Pearl Court, Allendale, NJ 07401. 2 Thou, personal communication. 5384

a b c 0 1 2 4 6 810 32 38 3238 @NlIWAF1 â€” iii cycin

Fig. I. Characterization of REF-112 RNA. a, kinetics of @2JwAFI/cIP1induction. Total RNA from REF-l 12 cells maintained at 32Â°Cfor the indicated times (h) was probed with ratp21WAFI@Cll'IcDNA. b, cyclin G RT-PCR. Total RNA harvested from REF-l 12 cells maintained at 38Â°C,or 32Â°Cfor 8 h, was reverse-transcribed with an anchored olido(dT) primer. PCR was performed with this primer and a primer specific for the 3' end of rat cyclin G. The arrowhead shows an â€”300-bp,differentiallyexpressed rat cyclimG RT-PCR product. c, Northern analysis of total RNA from REF-112 cells. The cyclin G RT-PCR cDNA described in b was used as a probe for RNA isolated from 38Â°C-or 32Â°C-maintained REF-i 12 cells.

at either 38Â°Cor32Â°C.Cellswere split and seeded 48 h before any temperature performed in a thermocycler for 40 cycles of: 94Â°Cfor 3 s, 40Â°Cfor 2 mm, and shifts. Temperature shifts were performed by simple transfer of subconfluent 72Â°Cfor 30 s. Reactions were terminated by the addition of 38% formamide, flasks to preequilibrated incubators without media changes. For transfections, 8 mr@iEDTA, 0.02% bromphenol blue, and xylene cylanol. Samples were 4 X iO@cells(T98G and SW480 cells) or 0.8 X l0@cells (SKOV3-IP1) were heated for 2 mm at 70Â°Cand run on a 6% denaturing polyacrylamide gel. PCR seeded in six-well dishes. Transfections were performed using lipofectin (Life products were purified after drying the gel onto Whatmann 3MM paper and Technologies, Inc.) as described by the manufacturer. Briefly, 40 @.dofreduced exposure to film. Excised bands were resuspended in 120p.1of H20 for 10mm serum medium (Optimem; Life Technologies, Inc.) was added to 2 @.tgofDNA at room temperature, followed by 15 mm of boiling. Debris was pelleted by (pCEP4 derivatives; InVitrogen). A 50-pd mixture containing 10 @doflipo centrifugation, and 10 p.1 of 3 M sodium acetate, 5 p.1 of 10 mg/mI glycogen, fectin and 40 ,.d of Optimem was added to the DNA mixture. After a 15-ruin and 400 p.1of ethanol were added to 100 p.1of the eluted DNA. DNA was room temperature incubation, 1 ml of Optimem was added, and the mixture allowed to precipitate overnight at â€”20Â°C.TheDNA was pelleted, amplified was overlayed onto Optimem-washed cells. Cells were allowed to incubate for using the same primers used for the original RT-PCR, gel purified, and cloned 5 h in a 37Â°C/5% CO2 incubator, following which the transfection mixture was into pCRH (InVitrogen). RT-PCR amplification of specific clones was replaced with normal growth medium. After 44 h, cells were split into selection achieved using the following primer pairs: CGRJJ, dT12AT/5'-TACAAC medium containing hygromycin (0.25 mg/mi). After 12â€”14days,colonies GAGG-3'; SM2O,dT12GG/5'-GATCATAGCG-3'; CGRJ9, dT12CG/5'-GAT were stained with 2% methylene blue in 50% ethanol and counted. Only CATAGCC-3'; and mEH, dT12CA/5'-GATCATGGTC-3'. colonies containing >50 cells were scored. 5' RACE. RACE wasperformedwith theMarathonkit (Clonetech)using Plasmid Constructs. pCEP4 derivatives were constructed as follows: REF-1l2 RNA harvested from 32Â°C-maintainedcells as described by the rCGR19 and rCGR11 were cloned as KpnL/XhoIand NotI/BamHI fragments, manufacturer. Oligonucleotides specific for the 3' ends of rat CGRJJ and respectively, from the original, full-length 5' RACE3clone contained in pCRII CGRJ9 were used in the procedure. (InVitrogen). Full-length rSM2O and rmEH were cloned by PCR amplification Northern Analysis. Northern analysis was performed using 10â€”20p.gof as an XhoL/BamHI and KpnLIXhoI fragment, respectively. Full-length hCGR1 1 total RNA electrophoresed on 1.2% formaldehyde gels. Blotting and probing was PCR amplified from an original pBluescript-SK clone (rescued from the was essentially as described (38). Probes were gel-purified cDNA fragments AZAP-expresshuman brain library) as a KpnI/SalI fragment and cloned into cr-32P-labeled by random priming (Boehringer Mannheim Biochemicals). Rat the KpnIJBamHIsites of pCEP4. hCGR11AEF was derived by overlap PCR tissue blots were obtained from Clonetech and probed, stripped, and reprobed amplification, resulting in a deletion ofboth CGR11 EF-hand domains. pCEP4 as recommended by the manufacturer. Rat @2JwAF1/CIPI cDNA was kindly clones for hp53 and hp53(as) were derived from pC53-SN3 (provided by B. provided by B. Vogelstein. Vogelstein). pCEP-WAF1 was obtained from B. Vogelstein and has been DNA Sequencing. Sequencing of all cDNA clones was performedby described previously (2). manual, Sanger dideoxy sequencing (United States Biochemical Corp.) using RNA Isolation. Total RNA was isolated by direct lysis in RNazol B primers that allowed for sequence determination of both strands along the (Tel-Test, Inc.) as described by the manufacturer. Poly(A)@RNA was isolated entire length of the cDNAs. Multiple clones were sequenced for both rat and from total RNA preparations using a MessageMaker kit (Life Technologies, human CGRJJ and CGRJ9 cDNAs. Inc.) as described by the manufacturer. Database Searching. All homology searches (FASTA), alignments Oligonucleotides. Twelve anchored oligo(dT) primers [dT12(A,C,G)- (BES1'FIT), and structural features (MOTIFS) were performed using The (A,C,G,T) combinations] and 25 random lO-mers (3) were used for RT-PCR Genetics Computer Group (Madison, WI) software programs. Final searches reactions. For cyclin G RT-PCR reactions, dT12GCand 5'-TCTTCACTGC-3' were performed using GenBank version 94.0. primer pairs were used to amplify a â€”300-bp fragment. RT-PCR Reactions. Reversetranscriptionreactionswereperformedusing RESULTS 200 ng of total RNA in 5 mM MgCl2/lO mrvtTris, pH 8.3, 10 mM KC1, 20 p.M deoxynucleotide triphosphates, 20 units RNase inhibitor (Perkin-Elmer), 50 Characterization of REF-112 RNA. To initially characterize total p.M T12NN, and 50 units Moloney murine leukemia virus reverse transcriptase RNA prepared from REF-l12 cells grown at both 38Â°Cand 32Â°C,we (Perkin-Elmer) by heating samples (without reverse transcriptase) to 65Â°Cfor evaluated the expression of genes known to be regulated by p53. 5 mm, and then placing the reaction at 37Â°Cfor5 ruin. Reverse transcriptions Northern analysis of @2JwAF1/CIP1showeda rapid induction of tran were allowed to proceed for 55 mm at 37Â°C.Reversetranscriptase reactions scription, with relatively high levels of @2JwAF1/CIP1 @4Aobserved were inactivated by incubating for 5 rain at 95Â°C.PCRreactions were per by 8 h after shifting from 38Â°C(Fig. la). Furthermore, we have formed with 2 p.1 of the reverse transcriptase reaction in 1.5 mrsi MgCl2, 10 mrvi Tris (pH 8.3), 10 mMKC1,4 p.Mdeoxynucleotide triphosphates, 2.5 units Taq observed a corresponding induction of apoptosis within 8â€”10h after polymerase (Perkin-Elmer), 10 p.Ci of [a-35S]dATP (12.5 p.Ci/p.l; DuPont shifting to 32Â°C(data not shown). Based upon these observations, we NEN), 50 p.M T12NN, and 0.2 p.M random lO-mer. PCR reactions were chose an 8-h induction time for the differential transcript analyses presented here. In addition, rat cyclin G Northern analysis showed

3 The abbreviations used are: RACE, rapid amplification of cDNA ends; RT, reverse high levels of transcript at the growth-inhibitory temperature of 32Â°C transcription. but little or no such transcript in cells growing exponentially at 38Â°C 5385

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1996 American Association for Cancer Research. CELL GROWrH REGULATORY GENES a C 1234 3238 1234 3238 Fig. 2. Differentially expressed genes from wild-type p53-containing REF-I 12 cells. RT-PCR and Northern analysis from 38Â°Cand 32Â°C(8 h) @ REF-I 12 cells representing CGRII (a). CGRI9 -@â€” (b), SM2O (c), and mEH (d). Specific RT-PCR primers and amplification conditions are described in â€œMaterialsandMethods.' Arrows. differentially @ui@ii@rw-j expressed cDNA that was excised, reamplified. 2 and used as a probe for the accompanying Northern analysis. Lanes I and 3. 32Â°C-maintained REF-I 12 RNA harvested after an 8-h incubation. Lanes 3 b d and 4, 38Â°CREF.I 12 RNA. Duplicate samples are independently isolated total RNA preparations. -Ã¸@g:@_@ a

(Fig. Ic). Both cyclim G and @2JwAFI/cIP1appeared to be induced to partial clone 19 cDNA as a probe revealed a more ubiquitous expres similar levels (data not shown; see below). sion pattern, with a 1.4-kb transcript showing the highest levels of RT-PCR reactions were performed on total RNA isolated from both expression in rat testis. Because neither of these genes have been growth conditions with primers specific for the 3' end of rat cyclin 0 described previously, we have chosen to name them cell growth to characterize the RNA preparation prior to identifying novel p53- regulatory genes CGRJJ and CGRJ9. Both CGRJJ and CGRJ9 also regulated genes. As anticipated, a differentially expressed band of 300 showed significant transcriptional induction at the permissive temper bp was detected in the 32Â°C-induced RNA but was absent in the RNA ature in baby rat kidney cells transformed with temperature-sensitive from uninduced cells (Fig. lb). Excision of the cyclim 0 RT-PCR p53 and E1A (apoptotic cells) and cells transformed with temperature band and subsequent cloning and Northern analysis (Fig. Ic) con sensitive PS3, E1A, and E1B (growth-arrested cells; data not shown). firmed the identity of this band as rat cyclin 0 (data not shown). Kinetics of inductions for CGRJJ and CGRJ9 transcripts following Differential Analysis of REF-112 RNA. To identify other a shift in temperature from 38Â°Cto 32Â°Cin REF-l 12 cells was growth-regulated genes, we performed RT-PCR reactions using 12 performed to aid in determining whether the transcriptional induction â€œanchoredâ€•oligo(dT)primers in conjunction with 25 â€œrandomâ€•10- of these genes was an early event in the cascade leading to perturba mers (see â€œMaterialsand Methodsâ€•). RT-PCR reactions were per tion of growth (Fig. 3b). Induction kinetics for CGRI9 closely parallel formed on duplicate, independently isolated total RNA from REF-l 12 those for @2JwAFI/CIPl(compare Figs. la and 3b). A slightly slower cells maintained at 38Â°Cor 32Â°C.Primer pairs that would amplify induction rate was observed for CGRJJ (Fig. 3b), and a higher basal either rat @2JWAFI/CIPIorMDM2 were omitted. (uninduced) transcript level existed for CGRJ9 than for CGRJJ (Fig. A total of 35 differentially expressed RT-PCR products were cho 2). Expression levels of CGRJJ and CGRJ9 could also be indirectly sen for further analysis based on apparent induction of product at the assessed relative to @2JwAFI/ctP1and cyclin 0 in this REF-I 12 induced temperature of 32Â°C.Although RT-PCR reactions were per system. Probe hybridization to an induced REF-l 12 cDNA library formed on duplicate RNA samples, all reactions containing potentially yielded cyclin 0 levels 2-fold higher than @2jWAFI/CIPl,with @ interesting transcripts were repeated in full to further avoid potential being expressed at â€”0.1% of the total mRNA in the cell artifacts that are a common problem in differential display analyses (3, and cyclin 0 being expressed at â€”0.2% (data not shown). Serial 39). Subsequent Northern analyses ruled out all but four of these analysis of gene expression (46) with induced REF-l 12 cells, which RT-PCR products as being differentially expressed genes. None of permits a more quantitative assessment of message abundance, also these four genes showed any transcriptional induction in RNA from yielded cyclin 0 levels at â€”0.2%.@Using cyclin G as the standard for control cells harboring a non-temperature-sensitive p53 mutation quantitation, we conclude that CGRJJ is expressed at 0.05%, and [REF-l32 cells (PHE132)] grown at 32Â°C,suggesting that the tran CGRI9 is expressed at 0.01% of the total message in p53-induced scriptional induction observed was not due to the shift in temperature REF-112 cells. (data not shown). Two of the induced genes, SM2O (40, 41) and cDNA and Deduced Protein Sequences for Cell Growth Regu microsomal epoxide hydrolase (mEH; Refs. 42â€”45),have been de latory Genes. Potentially full-length cDNAs for CGRJI and CGRI9 scribed previously. Although correlative gene expression analyses were obtained from REF- 112 RNA by 5' RACE (47) and by hybrid have described these genes as being differentially expressed depend ization to a human fetal brain cDNA library. Deduced protein se ing on various growth conditions, p53-dependent transcription of quences for rat and human clones are shown in Fig. 4. Rat and human these genes has not been proposed previously. CGRJ I cDNAs obtained are 1209 and 1113 bp in length, respectively, Differential RT-PCR products from the two novel cDNAs, desig and have an open reading frame encoding protein products of 272 and mated 11 and 19, as well as SM2O and mEH are shown in Fig. 2. 301 amino acids (Fig. 4a). No in-frame stop codons were observed for Northern analysis for each of these partial cDNAs confirmed that either cDNA upstream of the putative start site (data not shown). The these transcripts were differentially expressed in cells grown at 38Â°C use of GeneTrapper technology (Life Technologies, Inc.) did not yield or 32Â°C(Fig. 2). Partial cDNA fragments 11 and 19 were used as cDNAs with longer 5' extensions for the human clone. Optimal 5' probes for expression analysis from various rat tissues (Fig. 3a). A alignment between the rat and human CGR1 1 proteins truncates seven restricted expression pattern was observed using the partial 11 cDNA amino acids (MSRWLMQ) from the first rat CGRJ 1 ATG, disallow probe, revealing a I .3-kb transcript present predominantly in whole

brain and kidney with limited expression in heart, lung, liver, and 4 S. L Madden, E. A. Galella, J. Thou, A. H. Bertelsem, and G. A. Beaudry. SAGE skeletal muscle, and no expression in spleen and testis. Using the transcript profiles for p53-dependent growth regulation, submitted for publication. 5386

a c@ b

@ bo@H L@ 0 1 2 4 6 8 10

@ . , CGR11

â€¢. â€¢@@@CGR19 @ @.ua

...... Io.. actin

Fig. 3. Expression profiles for CGRII and CGRI9. a. rat multiple tissue Northern blot probed with 32P-labeled cDNA fragments specific for CGRJI, CGRJ9, and @3-actin.b, induction kinetics of CGRJI and CGRI9. REF-I 12 cells were maintained at 32Â°Cforthe indicated times (h). Total RNA was isolated, electrophoresed, blotted, and probed with 32P-labeledcDNA specific for rat CGRIJ or CGRI9.

Table I Inhibition of colonyformation b.vcell growth regulatory genes domain at the NH2 terminus of the protein (Fig. 44). No accompany Colony numberâ€•(%of controlb) ing B-box domain was observed in CGRI9, as has been observed in some ring-finger-containing proteins. Interestingly, we have also iso IP1T98GhWAF1Clomec SW480SKOV lated a full-length CGRJ9 cDNA from REF-I 12 cells containing a 8.6hp53 29.0 Â±3.434.5 Â±6.820.6 Â± 2.3hp53(as) 0.2 Â±0.25.1 Â±1.45.7 Â± l30-bp insert within the middle of the cDNA (data not shown). This 29rCGRI 92.7Â±3762.4 Â±20151.5 Â± insert contains 5' GT and 3' AG ends, consistent with exon-intron IIrSM2O1 18.2 Â±8.831.5 Â±9.325.8 Â± boundaries (48). This suggests that either some of the CGRJ9 tran 3.2rmEH 4.4 Â±3.616.4 Â±8.31 1.4 Â± 3.2rCGRI9 73.6 Â±6.121.9 Â±7.35.2 Â± script in the cells is incompletely processed, or that an alternate 7.6hCGR1 74.8 Â±7.738.3 Â±1713.3 Â± transcript containing a retained intron is produced. The latter CGRI9 IIhCGR1I ND8.3 Â±0.157.6 Â± 2.6a taEF ND1 10.0 Â±5.098.8 Â± would produce an in-frame termination codon within the putative intron, resulting in a protein chimera retaining the first 140 amino Experiments were performed in duplicate. balone.C @EP4 vector acids of CGRI9. RT-PCR analysis from 32Â°C-induced REF-l 12 cells h, human; r, rat. ND, not determined. reveals two bands of equal intensity, consistent with the stable gem eration of two CGRJ9 transcripts. Additional analysis of protein products encoded by these two transcripts is required to determine ing for a definitive designation of the rat ATG start codon. The rat and whether this differentially processed transcript is significant. human CGR1 1 proteins are 65% identical, whereas two highly con Growth Suppressive Function of Cell Growth Regulatory served putative Ca2tbinding EF-hand motifs (amino acids 82â€”94and Genes. To begin to assess the potential growth-suppressive properties 127â€”139in the human protein; overlined in Fig. 4b) share nearly of the differentially expressed genes isolated in this study, we evalu 100% identity. Additionally, four clustered 17-amino acid repeats ated growth inhibition via stable transfection in a colony inhibition exist within the COOH-terminal portion of only the human CGR11 assay. All genes examined were expressed from the cytomegalovirus protein [consensus; PGPRGEAEGQAEA(KIR)GDA], suggesting a promoter on episomally maintained plasmids (pCEP4), and colony structure resembling four a-helical domains interrupted by distinct formation was scored 2â€”3 weeks after transfection. As controls, turns. In vitro production of rat CGR1 I and CGR19 proteins yielded @2JWAFh/@ , p53 and p53 antisense constructs were also trans products of â€”37and 34 kDa, respectively (data not shown). Further fected. We chose three cell lines, all harboring different p53 alleles for more, the CGR1 1 protein is very acidic, with a net negative charge of the analysis of growth-suppressive function: the SW480 colon carci â€”29and a p1 predicted to be 4.21. noma cell line contains two point mutations, H1S273 and SER3O9; an The rat and human CGR19 proteins are 85% identical at the amino ovarian carcinoma line, SKOV3 IPl, which is p53-null; and a glio acid level, with consistent homology throughout the entire length of blastoma cell line, T98G, containing a single point mutation, the 332-amino acid proteins (Fig. 4c). Although no in-frame stop MET237, in p53. Rat mEH, SM2O, CGRJJ, and CGRJ9 all exhibited codons are found upstream from the putative initiation codon, the rat some growth-suppressive effects but to varying degrees, depending on and human protein homologues diverge immediately upstream of the the cells analyzed (summarized in Table 1). p53 consistently showed start codon, further suggesting that the designated ATG is indeed the the most potent growth-inhibitory effect. @2JwAF1/cIPIexhibited be start codon for CGR19 (data not shown). The predicted protein tween 65 and 80% inhibition, in agreement with previous results (2). sequence of CGR19 suggests limited homology to known proteins Only CGRJ 1 and SM2O showed marked growth inhibition in SW480 with the exception of a putative zinc-binding C3HC4 ring-finger cells (80 and 95%, respectively), but all clones tested showed growth 5387

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1996 American Association for Cancer Research. CELL GROWTh REGULATORY GENES a human 1 ) J@PLTMT@JLILLLLPTGQWKDGVTRPDSEVQHQLLPNPFQPGQEQLGLLQSYLKGLGRT rat (1)____*******____L - A-L-P-A-Q--TT - P---RR-RD - ERM I (62 )EVQLEHLSREQVLLYLFALHDYDQSGQLDGLELLSMLTAALAPGAANSPTTNPVILIVDKV (55)-QMN F--N HF_*I I' (123 )LETQDLNGDGLMTPAELINFPGVALRHVEPGEPLAPSPQEPQAVGRQSLLAKSPLRQETQE (115) D E_PXR***A_S_P_AL____PA_S_P___N___QS___Q

(184 )APGPREEAKGQVEARRESLDPVQEPGGQAEADGDVPGPRGEAEGQAEAKGDAPGPRGEAEG

I @ac s S (216):__SE ED_ER:V_SXD_EG___D__A____TQ__P_VV_A_SI:L (272)

b Dx(DNS)x(DENSTG)(DNQGHRX)x(LIVMC) (DENQSTAG)xx(DE) (LIVMFYW)

Fig. 4. Deduced amino acid sequences of rat and human CGRI I and I. DY. CGRI9. a, deduced amino acid sequence of human (top) and rat F N (bottom) CGRJJ. Dashes represent identity between the human and rat sequences. Asterisks represent gaps present within the rat CGRI I sequence. The overline represents the two putative EF-hand domains. b, II. DL. CGRI I EF-hand domains I and II aligned with the consensus EF-hand D sequence. Rat amino acids differing from the human sequence are shown on the bottom line. c, deduced amino acid sequence of human C (top) and rat (bottom) CGRI9. The overlined sequence encodes a potential zinc-binding, ring-finger domain. Dashes indicate identity human ( 1 ) MAAVFLVTLYEYSPLFYIAVVFTCFIVTTGLVLGWFGWDVPVILRNSEEPQFSTRVFKKQM between the human and rat sequences. d. CGRJ9 ring-finger domain rat AL A aligned with the consensus ring-finger sequence.5 (62) RQVKNPFGLEITNPSSASIT1@GITLTTDCLEDSLLTCYWGCSVQKLYEALQXHVYCFRIST

E----D--HR H SS-GV----P

(184) DREIYDIISMVSVIHIPDRTYXLSCRILYQYLLLAQGQFHDLKQLFMSANNNFTPSNNSSS K----P IY SA---RDQ-P

(245 )EEKNTDRSLLEKVGLSESEVEPSEENSKDCVVCQNGTVNWVLLPCRHTCLCDGCVKYFQQC ADGSVEH----A---AGA--D-V--S G

(306 )PMCRQFVQESFALCSQKEQDXDKPETL(332) G---A---ILE-S d CxxC-x(9-27)-Cx(1-3)Bxx-CxxCx(3-47)CxxC Consensus @ IIII II CVVC--x(11)--Cx( 1 )HTCLCDGCz( 6 )CPMC CGR19

inhibition in SKOVIP1 andT98G cells (60â€”95%inhibition), demon lationship critical to growth inhibition. A similar analysis is ongoing strating potential conservation of CGRJJ and CGRJ9 function in with respect to the ring-finger domain within CGRJ9. Site-specific human cells. mutagenesis to identify the critical residues within these domains may Although the above results are suggestive of a growth-inhibitory yield additional information on the function of these domains with function for the tested cDNAs, it remains possible that overexpression regard to cell growth suppression. of these proteins yields the inhibitory effects observed due to non specific toxicity. To address this possibility, we constructed an EF DISCUSSION hand deletion mutant within the human CGRI I cDNA and assessed this protein for growth-inhibitory potential. Previous results with We have describedthe isolation of potential growth-regulatory EF-hand-containing proteins suggest that a deletion of one of two genes from rat and human cells based on a differential transcript EF-hand motifs results in a functional protein with respect to Ca2t analysis from rat cells temperature-sensitive for p53 function. This binding potential (49). Therefore, we deleted both of the EF-hand analysis revealed the induction of four genes in cells harboring func domains within human CGRJJ for this analysis (CGRl1@EF, amino tional p53. Two of the genes isolated, SM2O (40, 41) and mEH acids 82â€”139).Unlike intact CGRJJ, EF-hand-deleted CGRIJ cDNA (42â€”45), are previously described genes, although no correlation transfectants were unable to inhibit colony formation in SKOV3 IPl between p53 status and transcriptional induction of these genes had or T98G cells (Table 1). Human wild-type CGRJI cDNA suppressed been established. SM2O was described as a platelet-derived growth cell growth to 8% (SKOV3 IP1) and 58% (T98G) relative to the tested factor-A-induced transcript in vascular smooth muscle cells (40); the mutant. Thus, it likely that, at least with respect to CGRJJ, the function of SM2O has not yet been elucidated, although expression has growth-suppressive effects observed reflects a structure-function re been demonstrated in the arterial wall within smooth muscle (41). In contrast, mEH is known to be involved in the catalytic detoxification 5 GemBank accession numbers: U66468, U66469, U66470, U6647l. of xenobiotics, including metabolizing reactive epoxides (42â€”45). 5388

Although no known correlation exists between mEH and wild-type rently assessing other apoptotic and growth arrest-susceptible systems PS3' induction of p53 activity through epoxides may aid in maintain for induction of these genes. In any event, transcriptional induction of ing genomic stability in the presence of toxins. these genes may have important implications for discerning the com CGRJJ Northern analysis showed the highest level of induction of plex regulatory cascade leading to growth perturbation. the four genes, followed by mEH, CGRJ9, and SM2O.REF-l 12 serial Further work will focus on determining if clones CGR1 I and analysis of gene expression has confirmed the relative abundance of CGR19 are direct targets of p.53 regulation or are downstream effec CGRIJ and CGRJ9 as being 0.05 and 0.01%, respectively, ofthe total tors relevant in determining the growth status of a particular cell. It mRNA in the wild-type p53-harboring cells.4 Both CGRJJ and remains to be determined if the induction of these clones is conserved CGRJ9 amino acid sequences exhibit features described previously in in other growth-regulatory systems, both p53 dependent and mdc proteins directly involved in regulating cell growth. CGRJI contains pendent. Finally, work is ongoing in the hopes of defining any two putative Ca2@-binding EF-hand domains. These domains have potential protein-interactive partners for these clones in the hopes that been described in numerous proteins including calmodulin, 5100 defining these partners will aid in the elucidation of their function. proteins, troponin C, and CDC31 in yeast. More recently, an EF-hand containing protein, ALG-2, was identified as being essential for FAS-induced apoptosis in T cells (49). Because calcium is a major ACKNOWLEDGMENTS second messenger within cells, the prevalence of calcium-binding We thank B. Vogelstein and Shibim Zhou (Johns Hopkins University, domains within numerous cellular proteins reflects the importance of Baltimore, MD) for cDNA clones, sequence information, and helpful discus calcium-mediated regulation of cell growth. sions and 0. Coppola and A. Brown (PharmaGemics, Inc.) for help with cell The CGR1 1 protein is a very acidic protein exhibiting a net nega culture and DNA transfectioms. tive charge of â€”29and a p1 predicted to be 4.21. Interestingly, the CGRJJ homologues derived from rat and human are highly conserved only in the region of the EF-hand domains, although both rat and REFERENCES human homologues are highly acidic. Indeed, a 17-amino acid clus I. Barak, Y., Gottlieb, E., Juven-Gershom, T., and Oren, M. Regulation of mdm2 tered repeat region is present within the CGRJJ COOH terminus but expression by p53: alternative promoters produce transcripts with monidentical trans lation potential. Genes Dcv., 8: 1739â€”1749, 1994. is absent from the rat homologue. It is unclear if structural features 2. el-Deity, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Tremt, J. M., aside from the EF-hand domain are maintained within the CGRJI Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. WAF1. a potential homologues. Both rat and human CGRJJ homologues displayed mediator of p53 tumor suppression. Cell, 75: 817â€”825, 1993. 3. Bauer, D., Muller, H., Reich, J., Riedel, H., Ahrenkiel, V., Warthoe, P., and Strauss, growth-inhibitory potential at least as potent as @2JWAF1/CIP1@By M. Identification of differentially expressed mRNA species by an improved display contrast, EF-hand deletion mutants of CGRJ 1 were unable to inhibit technique (DDRT-PCR). Nucleic Acids Rca., 21: 4272â€”4280, 1993. 4. Liang, P., and Pardee, A. B. Differential display of eukaryotic messenger RNA by growth in the colony-inhibition assay, suggesting that the EF-hand means of the polymerase chain reaction. Science (Washington DC), 257: 967â€”971, domain within this protein is essential for the observed growth 1992. inhibitory activity. CGRJJ expression in rat tissues was limited to, but 5. Ko, L. J., and Prives, C. p53: puzzle and paradigm. Genes Dcv., 10: 1054â€”1072, 1996. abundant in, rat brain and kidney, suggesting a more defined and 6. Maheswaran, S., Englert, C., Bennett, P., Heinrich, G., and Haber, D. A. The Wil tissue-specific role for the CGR1 1 protein than for the more ubiqui gene product stabilizes p53 and inhibits p53-mediated apoptosis. Genes Dcv., 9: tously expressed CGRJ9 transcript. 2143â€”2156,1995. 7. Mietz, J. A., Unger, T., Huibregtse, J. M., and Howley, P. M. The transcriptional The CGRJ9 cDNAs isolated from rats and humans are highly transactivatiomfunction ofwild-type p53 is inhibited by SV4Olarge T-antigen and by conserved, exhibiting 85% amino acid identity. Structural similar HPV-16 E6 oncoproteim.EMBO J., 11: 5013â€”5020,1992. 8. Olimer, J. D., Pietenpol, J. A., Thiagalingam, S., Gyuns. J., Kimzler, K. W., and ities of CGRJ9 to other previously described genes is limited to a Vogelstein, B. OmcoproteinMDM2 conceals the activation domain of tumour sup putative zinc-binding ring-finger domain at the very COOH termi pressor p53. Nature (Lond.), 362: 857â€”860, 1993. nus of the protein. In contrast to some other ring-finger-containing 9. White, E. Life, death, and the pursuit of apoptosis. Genes Dcv., 10: 1â€”15,1996. 10. Martinez, J., Georgoff, I., and Levine, A. J. Cellular localization and cell cycle proteins, CGR19 lacks a ring-finger-associated B-box domain (re regulation by a temperature-sensitive p53 protein. Genes Dcv., 5: 151â€”159,1991. viewed in Ref. 50). Interestingly, a correlation between ring-finger I 1. Gannon, J. V., and Lane, D. P. Protein synthesis required to anchor a mutant p53 protein which is temperature-sensitive for nuclear transport. Nature (Land.), 349: positioning within a given protein and potential functional impli 802â€”806,1991. cations has been established. Generally, ring-finger proteins either 12. Soddu, S., Blandino, G., Scardigli, R., Martinelli, R., Rizzo, M. G., Crescenzi, M., harbor the ring-finger domain at the NH2 or COOH terminus of the and Sacchi, A. Wild-type p53 induces diverse effects in 32D cells expressing different oncogenes. Mol. Cell. Biol., 16: 487â€”495, 1996. protein. Proteins harboring this domain at the COOH terminus 13. Cox, L. S., Hupp, T., Midgley, C. A., and Lane, D. P. A direct effect of activated have been linked to vacuolar/peroxisomal biogenesis and include human p53 on nuclear DNA replication. EMBO J., 14: 2099â€”2105.1995. the Zellweger's syndrome-associated ring-finger protein PAF1 14. el-Deiry, W. S., Kern, S. E., Pietenpol, J. A., Kinzler, K. W., and Vogelstein, B. Definition of a consensus binding site for p53. Nat. Genet., I: 45â€”49, 1992. (50). Ring-finger-containing proteins have also been implicated in 15. Kern, S. E., Kimzler, K. W., Bruskin, A., Jarosz, D., Friedman, P., Prives, C., and regulating development, oncogenesis, and vacuolar/peroxisomal Vogelstein, B. Identification of p53 as a sequence-specific DNA-binding protein. biogenesis (50). It remains unclear whether ring-finger domains Science (Washington DC), 252: 1708â€”1711, 1991. 16. Zambetti, G. P., Bargometti, J., Walker, K., Prives, C., and Levine, A. J. Wild-type are functional with respect to nucleic acid binding, protein-protein p53 mediates positive regulation of gene expression through a specific DNA sequence interactions, or protein-lipid interactions. Currently, we are eval element. Genes Dcv., 6: 1143â€”1152,1992. 17. Farmer, G., Bargoneui, J., Thu. H., Friedman, P., Prywes, R., and Prives, C. Wild uating the functional significance of the CGRJ9 ring-finger domain type p53 activates transcription in vitro. Nature (Lond.), 358: 83â€”86,1992. with respect to growth inhibition. 18. Ginsberg, D., Mechta, F., Yaniv, M., and Oren, M. Wild-type p53 can down-modulate Kinetics of induction of these genes varied, although induction the activity of various promoters. Proc. Nati. Acad. Sci. USA, 88: 9979â€”9983,1991. 19. Mack, D. H., Vartikar, J., Pipas, J. M., and Laimins, L. A. Specific repression of patterns were similar to the previously characterized @2JWAF1@@ TATA-mediated but not initiator-mediated transcription by wild-type p53. Nature transcript. This suggests that the induction of these genes is a rela (Lond.), 363: 281â€”283,1993. 20. Harper, J. W., Adami, G. R., Wei, N., Keyomarsi, K., and Elledge, S. J. The p21 tively early event in the cascade leading to growth inhibition. This Cdk-interactimgprotein Cipl is a potent inhibitor of 01 cyclin-dependent kinases. analysis has not definitively addressed the question of whether these Cell, 75: 805â€”816,1993. genes are direct transcriptional targets of p53 or are consequences of 21. Kastan, M. B., Than, Q., el-Deity, W. S., Carrier, F., Jacks, T., Walsh, W. V., Plumkeu,B.S., Vogelstein,B.,and Fornace,A. J., Jr. A mammaliancellcycle reduced cell growth. Similarly, it remains unclear if induction of these checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. genes is specific for apoptotic or growth-arrested cells. We are cur Cell, 7!: 587â€”597,1992. 5389

22. Sun, X., Shirnizu, H., and Yamamoto, K. Identification of a novel p53 promoter 36. Canman, C. E., Gilmer, T. M., Coutts, S. B., and Kastam, M. B. Growth factor modulation element involved in gemotoxicstress-inducible p53 gene expression. Mol. Cell. Biol., of p53-mediatedgrowth arrest versus apoptosis.Genes Dcv., 9: 600â€”611,1995. 15: 4489â€”4496, 1995. 37. Strasser, A., Harris, A. W., Jacks, T., and Cory, S. DNA damage can induce apoptosis 23. Wales, M. M., Bid, M. A., el Deiry, W., Nelkin, B. D., Issa, J. P., Cavenee, W. K., in proliferating lymphoid cells via p53-independent mechanisms inhibitable by Bcl-2. Kuerbitz, S. J., and Baylin, S. B. p53 activates expression of H/C-I, a new candidate Cell,79:329â€”339,1994. tumour suppressor gene on l7pl3.3. Nat. Med., 1: 570â€”577, 1995. 38. Sambrook, J., Fritsh, E. F., and Maniatis, T. Molecular Cloning: A Laboratory 24. Okamoto, K., and Beach, D. Cyclin G is a transcriptional target of the p53 tumor Manual, Ed. 2. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989. suppressor protein. EMBO J., 13: 4816â€”4822,1994. 39. Liang, P., Averboukh, L., and Pardee, A. B. Distribution and cloning of eukaryotic 25. Zauberman, A., Lupo, A., and Oren, M. Identification of p53 target genes through mRNAs by means of differential display: refinements and optimization. Nucleic immune selection of genomic DNA: the cyclin G gene contains two distinct p53 Acids Res., 21: 3269â€”3275, 1993. binding sites. Oncogene. 10: 2361â€”2366, 1995. 40. Wax, S. D., Rosenfield, C. L., and Taubman, M. B. Identification of a novel growth 26. Miyashita, T., Krajewski, S., Krajewska, M., Wang, H. G., Lin, H. K., Liebermann, factor-responsive gene in vascular smooth muscle cells. J. Biol. Chem., 269: 13041â€” D. A., Hoffman, B., and Reed, J. C. Tumor suppressor p53 is a regulator of bcl-2 and 13047,1994. box gene expression in vitro and in vivo. Oncogene. 9: 1799â€”1805.1994. 41. Wax, S. D., Tsao, L., Lieb, M. E., Fallom.J. T., and Taubman, M. B. SM-20 is a novel 27. Miyashita, T., and Reed, J. C. Tumor suppressor p53 is a direct transcriptional 40-kd protein whose expression in the arterial wall is restricted to smooth muscle. activator of the human box gene. Cell, 80: 293â€”299,1995. Lab. Invest., 74: 797â€”808, 1996. 28. Selvakumaran, M., Lin, H. K.. Miyashita. T., Wang, H. G., Krajewski, S., Reed, J. C., 42. Porter, T. D., Beck, T. W., and Kasper, C. B. Complementary DNA and amino acid Hoffman, B., and Liebermann, D. Immediate early up-regulation of bax expression by sequence of rat liver microsomal, xenobiotic epoxide hydrolase. Arch. Biochem. p53 but not TGF 3: a paradigm for distinct apoptotic pathways. Oncogene, 9: Biophys.,248:121â€”129,1986. 1791â€”1798,1994. 43. Falany, C. N., McQuiddy, P., and Kasper, C. B. Structure and organization of the 29. Than, Q., Fan. S., Bae, I., Guillouf, C., Liebermann, D. A., O'Connor, P. M., and microsomal xenobiotic epoxide hydrolase gene [published erratum appears in J. Biol. Fornace, A. J., Jr. Induction of bax by genotoxic stress in human cells correlates with Chem., 262: 10412, 1987]. J. Biol Chem., 262: 5924â€”5930,1987. normal p53 status and apoptosis [published erratum appears in Oncogene, 10: 1259, 44. Homscha,W., Oesch, F., and Friedberg, T. Tissue-specific expression and differential 19951. Oncogene, 9: 3743â€”3751, 1994. inducibility of several microsomal epoxide hydrolase mRNAs which are formed by 30. Hirano, Y., Yamato, K., and Tsuchida, N. A temperature sensitive mutant of the alternative splicing. Arch. Biochem. Biophys., 287: 380â€”385,1991. human p53, Vall38, arrests rat cell growth without induced expression of cipl/wafl/ 45. Kondo, S., Chou, Y., Gertson, P. N., Yokoyama, K., and Itakura, K. A bipartite sdil after temperature shift-down. Oncogene. 10: 1879â€”1885, 1995. suppressor conjunction of two distinct factor-binding sites is essential for down 31. Johnson, M., Dimitrov, D., Vojta, P. J., Barrett, J. C., Noda, A., Pereira-Smith, 0. M., regulation in rat epoxide hydrolase gene expression. Mol. Cell. Biol., 12: 4496â€”4502, and Smith, J. R. Evidence for a p53-independent pathway for upregulation of 1992. SDII/CIP1/WAFI/p2l RNA in human cells. Mol. Carcinog., 11: 59â€”64, 1994. 46. Velculescu, V. E., Zhang, L., Vogelstein, B., and Kimzler, K. W. Serial analysis of 32. Kobayashi, T., Consoli, U., Andreeff, M., Shiku, H., Deisseroth, A. B., and Zhang, W. gene expression. Science (Washington DC), 270: 484â€”487, 1995. Activation of p21WAFI/Cipl expression by a temperature-sensitive mutant of human 47. Chenchik, A., Moqadam, F., and Siebert, P. Marathon cDNA amplification: a new p53 does not lead to apoptosis. Oncogene, 11: 231 1â€”2316,1995. method for cloning full-length cDNAs. CLONTECHniques, X: 5â€”8.1995. 33. Yew. P. R., Liu, X., and Berk, A. J. Adenovirus EIB oncoprotein tethers a transcrip 48. Breathnach, R., and Chambon, P. Organization and expression of eucaryotic split tional repression domain to p53. Genes Dcv., 8: 190â€”202,1994. genes coding for proteins. Annu. Rev. Biochem., 50: 349â€”383,1981. 34. Michalovit.z, D., Halevy, 0., and Oren, M. Conditional inhibition of transformation and 49. Vito, P., Lacana, E., and D'Adamio, L. Interfering with apoptosis: Ca(2+)-binding of cell proliferationby a temperatare-sensitivemutantof p53. Cell, 62: 671â€”680,1990. protein ALG-2 and Alzheimer's disease gene ALG-3. Science (Washington DC), 271: 35. Halnaut, P., and Milner, J. Interaction of heat-shock protein 70 with p53 translated in 521â€”525,1996. vitro: evidence for interaction with dimeric p53 and for a role in the regulation of p53 50. Freemont, P. S. The RING finger: a novel protein sequence motif related to the zinc conformation. EMBO J.. 11: 3513â€”3520,1992. finger. Ann. NY Aced. Sci., 684: 174â€”192,1993.

5390

Stephen L. Madden, Elizabeth A. Galella, Deborah Riley, et al.

Cancer Res 1996;56:5384-5390.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/23/5384

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/56/23/5384. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.