RB18A, Whose Gene Is Localized on Chromosome 17q12-q21.1, Regulates in Vivo p53 Transactivating Activity¹

p53, originally described as an oncogene associated with the large T antigen of SV40(1), was later identified as a tumor suppressor gene in its wt³conformation (2). In normal cells, among its different functions, p53wt is also a transcription factor and transactivates genes by interacting with a p53RE present on the promoter of different genes (3). When mutated, p53 (p53m) shows defects in its transactivating activity (4). One of the open fields on p53 analysis is the identification of factors that interact with p53,thus regulating p53 functions. In this field, we previously identified(5) RB18A, a new gene, whose encoded protein was characterized by an apparent M_r of 205,000. Cloning and sequencing RB18A cDNA led to demonstrating that whereas its encoded 1566-amino-acid product was recognized by three different anti-p53 moAbs (PAb1801, PAb421, and DO1), RB18A did not share any significant nucleotide or amino-acid primary sequence homology with p53. In vitro, RB18A was identified as a DNA-binding protein that could self-oligomerize and bind to p53wt and p53m. The COOH-terminal domain of RB18A, named RB18A C-term carried these properties. In vitro, the interaction of RB18A with p53 increased specific interaction of p53 with its specific DNA consensus sequence. A structural relationship exists between RB18A and coactivators of the transcription system associated with nuclear receptors, as TRIP2 and TRAP220. Indeed, we previously mentioned(5) that a region of 244 bp localized in RB18A between nucleotides 2054 and 2297 presented 100% homology with the partially determined sequence of TRIP2 (6). Whereas TRIP proteins were originally described by their interactions with thyroid hormone receptor and with retinoic X receptor, both of which were proteins belonging to the family of nuclear receptors,TRIP2 function was not determined. In addition, after our work(5), TRAP220 (7), which shared 99% sequence identity within the RB18A coding sequence, with only minor sequence variations (8), was identified as a M_r 220,000 thyroid hormone receptor-associated protein (TRAP). TRAP220 is a member of a very heterogeneous complex that directly interacts with different nuclear receptors, thus acting as a cofactor in the basal transcription machinery (9). TRAP220 was also demonstrated to bind p53wt and p53m (9), as we previously showed for RB18A(5). In addition, DRIP205, a member of the DRIP complex,was also demonstrated to be identical to RB18A (10). However, no in vivo regulatory role of RB18A,TRAP220/DRIP205, or TRIP2 components has been demonstrated on p53 functions. Therefore, we herein analyzed the in vivo role of RB18A on p53 transactivating activity on different physiological promoters as Bax (11), p21Waf (12), and IGF-BP3 (13). Our results clearly demonstrated for the first time that, by forming in vivo heterocomplexes with p53, RB18A regulated the activity of the transcriptional factor p53.

K562 cells (an erythroleukemia cell line) and H1299(pulmonary embryo carcinoma), both p53 null cell lines were maintained in DMEM containing 10% SVF with 100 units/ml penicillin-streptomycin.

All of the expression vectors used were cloned in pcDNA3 (Invitrogen):pCMV-RB18A C-term and pCMV-RB18A coding for the full-length protein were prepared in our laboratory (5); pCMV-p53wt was kindly provided by Dr. Bert Vogelstein (Johns Hopkins University, Baltimore,MD); pCMV-p53m containing p53m in position 175 (Arg→His) or in position 179 (His→Gln), both of which were devoid of transactivating activity, were kindly provided by Dr. Menashe Bar-Eli (M. D. Anderson Cancer Center, University of Texas, Houston, TX).

Vectors used for transactivation experiments were as follows. PG13-CAT with the 13 p53-binding consensus sequences, cloned upstream CAT reporter gene and p21wwp-Luc with p21^Waf1 promoter gene cloned upstream luciferase reporter gene, were kindly provided by Dr. Bert Vogelstein (Johns Hopkins University, Baltimore,MD). pIGF-BP3-Luc, which contained IGF-BP3 promoter with p53-binding consensus sequences was constructed in our laboratory after PCR amplification of IGF-BP3 promoter and inserted in HindIII-KpnI sites of pGL2 vector (Promega). pBax-Luc was kindly provided by Dr. Moshe Oren (Weizmann Institute,Rehovot, Israel). pCMV-β Gal (Clontech) was also used.

K562 cells (10⁷) were transfected by electroporation at 220 V, 1050 μF, in an electroporator (Eurogentec). Cell suspension was then incubated for 48 h at 37°C in 5 ml of RPMI medium containing SVF. H1299 cells (4 × 10⁵) were transfected by calcium phosphate coprecipitation method, after 24-h preincubation in 6-well plates(Falcon) with 4 ml of medium at 37°C. The different plasmid DNAs were added to 450 μl of 10 mm Tris (pH 8)-1 mmEDTA. Total amount of DNA was brought with salmon sperm DNA to 30 μg or 5 μg for CAT or luciferase assay, respectively. For normalization of transfection efficiency, pCMV-β Gal plasmid diluted one-twentieth or one-fourth for CAT or luciferase assay, respectively,was added as an internal control. Transfected cells were then lysed for 30 min at 4°C in TNE buffer [50 mm Tris, 150 mm NaCl, 1 mm EDTA, 1% Triton X-100 (pH 7.6)]. After centrifugation for 15 min at 12,0000 × g, the cell extracts were tested for β-gal activity using reporter lysis buffer (Promega).

For CAT assays, the cell extracts were then incubated for 10 min at 60°C to inactivate the endogenous deacetylase activity. After incubation for 3 h at 37°C, CAT assays either were analyzed on TLC plates (Silica gel 60_F254; Merck, Darmstadt,Germany) after incubating cell extracts with 100 μg of acetyl-CoA and 3 μl of [¹⁴C]chloramphenicol (25 μCi/ml) in 125 μl of 0.25 m Tris (pH 7.9) or were quantified by liquid scintillation counting in a β counter (Beckman) after incubating cell extracts with 25 μg of n-butyryl-CoA and[¹⁴C]chloramphenicol (25 μCi/ml) in 125 μl of H₂0. Analysis of acetylated chloramphenicol,extracted by ethyl acetate, was performed by chromatography on plates that were autoradiographed with Kodak X-OMAT films. Quantification of the n-butyrylated chloramphenicol, extracted with xylene,was performed by mixing the xylene phase with scintillant.

For luciferase assay, cell lysates obtained with TNE buffer were incubated at 4°C with 5 mm ATP (pH 7.0) in 375 μl of Brasier buffer [25 mm Gly-Gly, 15 mmMgSO₄, and 4 mm EGTA (pH 7.8)]. Luciferase activity was immediately measured in a luminometer (Lumat LB9501; Berthold) after injection of 100 μl of 1 mmluciferine for 30 s. Each transactivating experiment described above was performed at least five times. The relative luciferase or CAT activity was calculated as the ratio of the value obtained for each sample to that obtained with the promoter alone.

Total proteins of H1299 or Raji cells were solubilized in 1%NP40, then submitted to immunoprecipitation procedures, as described previously (5), on anti-RB18A moAb, prepared in our laboratory. Briefly, anti-RB18A moAb was covalently bound to protein G/protein A agarose (Oncogene Science). After extensive washes of immunobeads, bound proteins were eluted in sample buffer and were analyzed by 7% SDS-PAGE. Immunoblotting was performed using anti-RB18A or anti-p53 moAb followed with peroxidase-labeled secondary Ab.

Lymphocytes isolated from human blood were cultured in α-MEM supplemented with 10% FCS and phytohemagglutinin at 37°C for 68–72 h. The lymphocyte cultures were treated with bromodeoxyuridine (0.18 mg/ml, Sigma) to synchronize the cell population. The synchronized cells were washed three times with serum-free medium to release the block and were recultured at 37°C for 6 h in α-MEM with thymidine (2.5 μg/ml; Sigma). Cells were harvested, and slides were made (by using standard procedures including hypotonic treatment),fixed, and air-dried.

cDNA probe was biotinylated with dATP using the Life Technologies, Inc. BioNick labeling kit (15°C, 1 h; Ref. 14). The procedure of FISH detection was performed as described previously(14). Briefly, slides were incubated at 55°C for 1 h. After RNase treatment, the slides were denatured in 70% formamide in 2× SSC for 2 min at 70°C and were then dehydrated with ethanol. Probes were denatured at 75°C for 5 min in a hybridization mixture containing 50% formamide and 10% dextran sulfate. Probes were loaded on the denatured chromosomal slides. After overnight hybridization,slides were washed and detected as well as amplified. FISH signals and the DAPI banding pattern were recorded separately by taking photographs, and the assignment of the FISH mapping data with chromosomal bands was achieved by superimposing FISH signals with DAPI banded chromosomes (14).

We studied the in vivo role of RB18A by analyzing its effect on p53 transactivating activity on different promoters that contained p53REs as promoters of Bax (11),p21^Waf1 (12), and IGF-BP3 (13).

We first analyzed whether the full-length RB18A could regulate p53-transactivating activity in transiently transfected H1299 or K562,neither cells expressing p53 protein. Limitations attributable to endogenous RB18A were overcome by using RB18A and p53 expression vectors driven by the strong CMV promoter. We, therefore, analyzed the effect of cotransfection of full-length RB18A, p53wt and the PG13-CAT vector, which contained 13 p53 binding DNA consensus sequences. Indeed,PG13-CAT vector was shown to constitute one of the most helpful elements to study the p53-transactivating activity (15). Forty-eight h after transfection, H1299 cells were lysed and a β-Gal assay (in which β-Gal cDNA was also driven by CMV promoter) was used to normalize transfection efficiency. Preliminary experiments allowed us to determine the optimal p53 amount needed to obtain the maximum CAT activity (data not shown). Then, CAT activity of each assay (Fig. 1) was compared with that obtained in the presence of promoter vector alone (Fig. 1, Lane 1). Whereas RB18A alone had no effect on CAT activation (Fig. 1, Lane 2), p53wt transfected alone enhanced PG13-CAT activity 4-fold (Fig. 1, Lane 3),an increase similar to that described by others (15). Cotransfection of RB18A with p53wt, in a RB18A:p53wt molar ratio of 0.5:1 (Fig. 1, Lane 4) to 1:1 (Fig. 1, Lane 5),enhanced PG13-CAT activity 8- to 25-fold, respectively. In control,p53m/179 in the absence (Fig. 1, Lane 6) or in the presence(Fig. 1, Lane 7) of RB18A (at a 1:1 RB18A:p53m molar ratio)had no effect on PG13 CAT activity. In addition: (a) RB18A COOH-terminal construct, named in our previous publication RB18A-N2 and characterized by amino acid sequence 1234–1566, gave identical results to full-length RB18A (data not shown); (b) another RB18A construct, named previously RB18A-NC1 (amino acid 436-1228) and which did not contain p53-binding site (5), had no effect on PG13 CAT activity (data not shown); and (c) the same results were obtained using K562 instead of H1299 cells (data not shown). All of the data presented above and our previous demonstration that in vitro RB18A interacted through amino acids 1234–1406 of its COOH-terminal domain with p53 (5) strongly supported that RB18A regulated, through its COOH-terminal domain, p53-transactivating activity.

The p53RE (PuPuPuCA/TT/AGPyPyPy) was found on different physiological promoters (3) as Bax (11),p21^Waf1 (Waf1/Cip1/Sdi1; Ref. 12),and IGF-BP3 (13), the genes of which were characterized as downstream genes in the p53 pathway. Thus, regulatory activity of RB18A was measured on p53-transactivating activity on Bax,p21^Waf1, and IGF-BP3 promoters cotransfected in H1299 cells (Fig. 2).

First, when cells were cotransfected with Bax promoter and p53wt cDNA(Fig. 2,A), Bax promoter was activated nine times by p53wt alone (Fig. 2,A, Lane 2). Coexpression of increasing amounts of RB18A (Fig. 2,A, Lanes 5–8) with p53wt increased Bax promoter activity up to a plateau of 27 times, with a 1:1 RB18A:p53wt molar ratio (Fig. 2,A, Lane 8). In addition, RB18A C-term(RB18A-N2), which contained the p53 binding site (5), had the same increasing effect (Fig. 2,A, Lane 10),whereas RB18A-NC1, which did not contain the p53 binding site but retained the DNA binding site (5), used at the same amount, had no regulatory effect (Fig. 2,A, Lane 11). Neither mutated p53/179 (p53m) alone (Fig. 2,A, Lane 3) or in the presence of RB18A (Fig. 2,A, Lane 9) nor RB18A alone (Fig. 2,A, Lane 4) had any effect. Second, cotransfection of p53wt cDNA with p21^Waf1 promoter (Fig. 2,B) induced activation of this promoter 48-fold (Fig. 2,B, Lane 2). When cells were cotransfected with increasing amounts of RB18A (Fig. 2,B, Lanes 5–8) and p53wt cDNAs, p21^Waf1 promoter activity was totally inhibited, in a dose-curve response for a 1:1 RB18A:p53wt molar ratio(Fig. 2,B, Lane 8). In addition, RB18A C-term(RB18A-N2), which contained the p53 binding site (5), had the same inhibiting effect (Fig. 2,B, Lane 10),whereas RB18A-NC1, which did not contain the p53 binding site but retained the DNA binding site (5), used at the same amount, had no regulatory effect (Fig. 2,B, Lane 11). Mutated p53/179 (p53m) alone (Fig. 2,B, Lane 3) or RB18A alone (Fig. 2,B, Lane 4) or in the presence of p53m (Fig. 2,B, Lane 9) had no effect. Third, when cells were cotransfected with IGF-BP3 promoter and p53wt cDNA (Fig. 2,C), IGF-BP3 promoter was activated 34 times (Fig. 2,C, Lane 2). Cotransfection of IGF-BP3 promoter with increasing amounts of RB18A (Fig. 2,C, Lanes 5–8) and p53wt cDNAs induced a total inhibition of IGF-BP3 promoter for a 1:1 RB18A/p53wt molar ratio (Fig. 2,C, Lane 8). In addition, RB18A C-term (RB18A-N2) had the same inhibiting effect (Fig. 2,C, Lane 10), whereas RB18A-NC1 construct (5) used at the same amount had no regulatory effect (Fig. 2,C, Lane 11). RB18A alone(Fig. 2,C, Lane 4) or mutated p53/179 (p53m) in the absence (Fig. 2,C, Lane 3) or in the presence(Fig. 2 C, Lane 9) of RB18A had no effect on IGF-BP3 promoter activity. All of these data demonstrated that RB18A regulated in a dose-dependent effect and through its COOH-terminal domain, the p53-transactivating activity on these physiological promoters. In addition, this regulatory effect (i.e.,increase or decrease of promoter activity) occurred through the p53-binding site of RB18A and was not related to RB18A DNA binding property; indeed, RB18A construct (RB18A-NC1) deleted in its p53 binding site, but retaining its DNA binding site (5), used in the same experimental conditions, did not regulate p53-transactivating activity.

In vivo interaction of RB18A with p53 forms was analyzed. For this purpose, two sets of experiments were performed (Fig. 3). First, cellular components from H1299 cells, cotransfected with RB18A and wt p53 cDNAs were solubilized in 1% NP40, then immunoprecipitated on an anti-RB18A moAb, recently prepared in our laboratory. Among all solubilized components, this anti-RB18A moAb recognized RB18A but not p53 (Fig. 3, Lane 1). Analysis of cellular components immunoprecipitated on anti-RB18A moAb by immunoblotting using either anti-RB18A moAb (Fig. 3, Lane 2) or anti-p53 moAb (Fig. 3, Lane 3) demonstrated that p53wt coprecipitated with RB18A. Second, cellular components from Raji cells, which expressed both RB18A and p53m, were solubilized in 1% NP40, and were then also immunoprecipitated on anti-RB18A moAb. Analysis by immunoblotting using either our anti-RB18A moAb (Fig. 3, Lane 4) or anti-p53 moAb(Fig. 3, Lane 5) demonstrated that in vivo RB18A also interacted with p53m. All of these data clearly demonstrated that,whereas in vivo RB18A interacted with p53wt as well as with p53m, only the interaction of RB18A with p53wt had an effect on the activity of the physiological promoter herein analyzed (Fig. 2).

These data also clearly demonstrated that, whereas RB18A alone had no effect on these promoter activity, by interacting directly through its p53 binding site localized in its COOH-terminal domain with p53wt,RB18A could activate in vivo p53-transactivating activity on Bax promoter and inhibit p53-transactivating activity on p21^Waf1 and IGF-BP3 promoters, in a dose-dependent manner. This difference in RB18A regulatory function on p53-transactivating activity depending on the nature of the physiological promoter was more likely attributable to the property of RB18A to form heterocomplexes with different cellular partners. Indeed,RB18A was structurally identical to two other components identified later than RB18A, i.e., TRAP220 (7, 8) and DRIP205 (10), which belonged to the TRAP (9)or DRIP (10) complexes, and were constituted of at least 12–15 subunits, respectively. TRAP220 (7) and DRIP205(10, 16) interacted with different hormone-activated nuclear receptors, such as thyroid hormone receptor, vitamin D receptor, retinoic acid receptor α, retinoic X receptor α, and PPAR. DRIP205, also named PBP (16), interacted with hormone-activated glucocorticoid receptor (17) or estrogen receptor (18). Thus, RB18A (or TRAP220, DRIP205, PBP) by being a member of multiple-partner complexes and acting as a cofactor of transcriptional machinery may differently modulate different promoters. Additional studies are needed to determine under which hormone stimulation the key regulatory molecule RB18A (or TRAP220,DRIP205, PBP) may switch cell pathways into apoptosis and/or cell cycle arrest.

Using a RB18A cDNA probe of 3.3 kb, FISH mapping analysis proceeded as detailed in Fig. 4 legend. Under the conditions used, the hybridization efficiency was approximately 68% for this probe (among 100 checked mitotic figures,68 of them showed signal on one pair of the chromosomes). Because the DAPI banding was used to identify the specific chromosome, the assignment between signal from probe and the long arm of chromosome 17 was obtained. An example of the mapping results is presented in Fig. 4,A. The detailed position was further determined based on the summary from 10 photos (Fig. 4 B). These data clearly demonstrated that RB18A gene mapped on chromosome 17q12-q21.1. The labeling of two loci on the same chromosome suggested the presence of a family of genes (and the corresponding proteins)sharing sequence homologies. Furthermore, data bank analysis showed that among the genes already mapped on these two loci, some were associated with human cancers. This is the case with BRCA1,the gene for hereditary breast-ovarian cancer (19). Interestingly, BRCA1, which is a tumor suppressor gene, has been also shown, as RB18A, to physically interact with and stimulate p53-transcriptional activity (20). In addition, acute promyelocytic leukemia (21) exhibited a characteristic t(15;17) translocation that fuses the promyelocytic leukemia(MPL) gene on 15q22 to the retinoic acid receptor α(RARA) gene on 17q12-q21.1, the loci where RB18A gene maps. Interestingly, as above mentioned, RB18A/TRAP220/DRIP205 interacts with RAR. The importance of the human RB18A/TRAP220/DRIP205 and murine PBP genes is strongly supported by the recent demonstration that PBP gene null mutation (PBP−/−) in mice is embryonic lethal at E11.5 days, which suggests that PBP is an essential gene for mouse embryogenesis (22). Additional studies are needed to determine the role of RB18A/TRAP220/DRIP205 in human cancers.

We would like to thank See DNA Biotech Inc. (Toronto,Ontario, Canada) for help for FISH mapping and Gérard Drevet for technical assistance.