JAB1 interacts with both the progesterone receptor and SRC-1.

JAB1 (Jun activation domain-binding protein-1) has previously been described as a coactivator of AP1 transcription factor. We show here, by yeast and mammalian two-hybrid analyses and by pull-down experiments, that JAB1 also interacts with both the progesterone receptor (PR) and the steroid receptor coactivator 1 (SRC-1) and that it stabilizes PR-SRC-1 complexes. We also show that JAB1 potentiates the activity of a variety of transcription factors known to associate with SRC-1 (nuclear receptors, activator protein-1, and nuclear factor kappaB). This occurs without any modification of PR or SRC-1 concentration. JAB1 is a subunit of a large multiprotein complex that has been called the COP9 signalosome. The latter is present in plant and animal cells and has been shown to be involved in a variety of cellular mechanisms including transcription regulation, cell cycle control, and phosphorylation cascades. We now show that it is also involved in the mechanisms of action of nuclear receptors and of their coactivators.

JAB1 (Jun activation domain-binding protein-1) was initially isolated by a two-hybrid screen (1) using the c-Jun Nterminal activation domain as a bait. It was shown that JAB1 potentiates target gene transcription activation by AP1 proteins (especially c-Jun and JunD) (1). This initial study also pointed to the sequence homology between JAB1 and the yeast protein Pad-1. The latter has been demonstrated to be necessary for the transcriptional activity of Pap-1, the yeast equivalent of c-Jun (1,2). Furthermore it was shown that JAB1 could functionally replace Pad-1 in yeast. Thus JAB1 and Pad-1 appeared to play similar roles in mammalian and yeast cells, respectively. However, recent studies have uncovered the true human homolog of Pad-1, which has been called POH-1. Both Pad-1 and POH-1 are components of the 26 S proteasome (3,4).
On the contrary JAB1 is not found in the proteasome but has very recently been shown to be a subunit of a different multiprotein complex that has been called the signalosome (or COP9 signalosome). This complex contains several proteins with sequence homologies to proteins present in the regulatory 19 S subunit of the 26 S proteasome (5)(6)(7). It has been proposed that the latter and the signalosome have a common evolutionary origin but have diverged to assume different cellular functions.
Recent studies have shown SRC-1, CBP, and nuclear receptors to be present in large multiprotein complexes (27,28). We show here that JAB1 interacts with both nuclear receptors and SRC-1 and activates transcription at the corresponding target genes. There is thus a link between the large regulatory complexes containing nuclear receptor coactivators and the JAB1 containing signalosome.

EXPERIMENTAL PROCEDURES
Construction of Vectors-The yeast (pAS2, pAS2-1, and pACT2) and mammalian (pM, pVP16, and pG5CAT) two-hybrid system vectors were obtained from CLONTECH. The DBD Gal4 -PR 1-930 yeast vector was initially obtained by ligating the rabbit PR cDNA (29) (nucleotides Ϫ3 to 2889 of the cDNA) into the pAS2 vector containing the yeast trp1 gene marker. This vector encodes a DBD Gal4 -HA-PR 1-930 fusion protein.
The DBD Gal4 -PR 897-930 and DBD Gal4 -PR 547-930 vectors were constructed from DBD Gal4 -PR 1-930 vector by enzymatic deletion of DNA fragments encoding amino acids 45-896 or 1-546 of PR, respectively. A human placenta cDNA library, fused to the Gal4 transactivation domain (AD- Gal4 ) and present in the yeast pACT2 vector encoding the Leu2 gene marker was used for the two-hybrid screening. DBD Gal4 -PR 1-930 , DBD Gal4 -PR 547-930 , and DBD Gal4 -PR 897-930 vectors used for two-hybrid experiments in mammalian cells were constructed by subcloning the inserts of pAS2-PR corresponding mutants into the pM vector.
cloned into the mammalian expression vector pSG5 (Stratagene) to create pSG5-JAB1. The same fragment was also subcloned into the same sites of a modified pSG5 vector (pSG5-HA) resulting in the N-terminal in frame fusion of JAB1 with the HA epitope (pSG5-HA-JAB1). Two differences were observed with the JAB1 sequence reported by Karin and co-workers (1). The first was a G/T change in the Val 127 codon which was silent. The second one was a A/G change that transformed the amino acid His 129 into an arginine. It should be noted that Arg 129 is conserved in the sequence of JAB1 homologs pad1 and POH-1 (2,3). The full-length JAB1 cDNA was also subcloned from the pSG5-JAB1 into the yeast pAS2-1 and pACT2 vectors expressing DBD Gal4 -JAB1 and AD Gal4 -JAB1, respectively. The same DNA fragments were introduced into pM and pVP16 vectors resulting in the expression of DBD Gal4 -JAB1 or AD VP16 -JAB1 fusion protein in mammalian cells, respectively. Finally, the full-length JAB1 was expressed in Escherichia coli as an N-terminal GST fusion protein by inserting the full-length cDNA into the pGEX-3X vector (Amersham Pharmacia Biotech) (pGEX-JAB1 vector).
The SRC-1 cDNA was obtained by polymerase chain reaction from the human placenta cDNA library as three overlapping fragments of similar size. The 5Ј upper primer and 3Ј lower primer used were CCC-GAAGATCTGGTGTGAAGTTTTTCAACATGAGTG and AAGCCAGA-TCTAATTTCACATTCCTTTAAAAGTGGTTATT, respectively. The entire open reading-frame of SRC-1 (nucleotides Ϫ18 to 5493) was finally reconstituted into the BglII site of pSG5 vector (pSG5-SRC-1). The full-length cDNA was entirely sequenced on both strands and corresponds to the sequence of the SRC-1 published by Takeshita et al. (15), classified as SRC-1a by Kalkhoven et al. (30). The SRC-1 cDNA was further subcloned from pSG5-SRC-1 into the pSG5-HA vector. This construction (pSG5-HA-SRC-1) resulted in the N-terminal in frame fusion of SRC-1 (nucleotides Ϫ18 to 4348) with the HA epitope. The yeast AD Gal4 -SRC-1 vector was obtained by inserting the SRC-1 cDNA into the pACT2 vector. The construct AD VP16 -SRC-1 for the mammalian two-hybrid system was obtained by cloning the same DNA fragment into the pVP16 vector.
The expression vector encoding the full-length rabbit progesterone receptor (pSG5-rPR) and the reporter plasmid PRE 2 -TATA-CAT have previously been described (31). The reporter plasmid ERE 2 -TATA-CAT was obtained by replacing the BamHI/BglII fragment of the PRE 2 -TATA-CAT vector with the synthetic oligonucleotide containing two EREs and a TATA box GATCCGGTCACAGTGACCAGCTACGGTCA-CAGTGACCGGATCTGAGGTCCACTTCGCTATATATTCCCCA. The plasmids encoding the other nuclear receptors and their respective reporter plasmids have been previously described (31). The reporter plasmids Ϫ73Col-TK-CAT and NFB-CAT and the expression vector for p65 RelA (pCMV-Rel A) have been described (32,33).
Transient Transfections of Mammalian Cells-CV-1 cells or HeLa cells were maintained in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) supplemented with 10% fetal calf serum (Seromed). 3 ϫ 10 5 cells were seeded per well of a six-well multiwell dish in Dulbecco's modified Eagle's medium containing 10% charcoal-stripped serum, and cells were transiently transfected using the calcium phosphate coprecipitation procedure as described previously (38). All transfections were performed with 330 l of calcium-phosphate precipitate/ well containing the reporter gene and expression vectors as indicated for each experiment, and total DNA was normalized to 20 g/ml of precipitate using salmon sperm DNA. The CAT activity was measured with the CAT enzyme-linked immunosorbent assay kit (Roche Molecular Biochemicals). Protein concentrations were determined by using the BCA protein assay (Pierce), and the CAT activity was corrected for protein content.
In Vitro Protein Binding Assays-The pGEX-JAB1 vector was used in the BL21 strain of E. coli to synthesize the GST-JAB1 fusion protein.
The manufacturer's instructions (Amersham Pharmacia Biotech) were followed except that ethylene glycol (10%) (39) and sarkosyl (1%) (40) were introduced in the extract buffer to increase the solubilization of the aggregated protein. After centrifugation, Triton X-100 (2%) was added (buffer A), and the cleared lysate was incubated with GST Sepharose-4B for 1 h at 20°C. The gel beads were then poured into a column, washed first with 20 volumes of buffer A, and then washed with 50 volumes of an inverse gradient buffer made with buffer A and buffer B (20 mM Hepes, 100 mM NaCl, 1 mM EDTA, 0.1% Nonidet P-40, 5 mM dithiothreitol, 10% glycerol, pH 8). The amount of GST-JAB1 present in the affinity gel was quantified by Laemmli-PAGE followed by Coomassie Blue staining. More than 50% of the solubilized fusion protein contained in the crude extract were finally bound to the affinity gel. The gel was stored in aliquots at Ϫ20°C. 35 S-and 3 H-radiolabeled proteins were synthesized by transcription of pSG5 expression vectors and subsequent translation using the TNT T7 coupled reticulocyte lysate system (Promega) as described by the manufacturer. Protein-protein interactions were studied as described (41) using binding buffer (20 mM Tris, 100 mM NaCl, 1 mM EDTA, 0.1% Nonidet P-40, pH 8.0). 5 l of the in vitro-translated lysate were incubated with 10 g of the GST fusion protein immobilized on glutathione-Sepharose in binding buffer in the presence or absence of 1 M R5020. Representative gels were stained with Coomassie Blue before being subjected to autoradiography to ensure that equal amounts of GST fusion proteins were included in each reaction. 35 S radioactivity was quantified using the Instant Imager (Packard).

Interaction between JAB1 and the Progesterone Receptor-
The amino acid sequence 897-930 of the PR contains helix 12 of its ligand-binding domain (LBD). This region has been shown to be of great importance in the transcription activation properties of nuclear receptors (43)(44)(45)(46)(47). We thus searched for proteins interacting with this region using it as a bait in a yeast two-hybrid screen. PR amino acids 897-930 were linked to Gal4 DNA-binding domain. Yeast cells were transformed with this plasmid and with a library of placental cDNAs linked to Gal4 activation domain (Gal4AD). Reporter genes responsive for Gal4 were present encoding a histidine synthesis enzyme (His3) and an adenine synthesis enzyme (Ade 2).
Several cDNA sequences were isolated by their property to allow yeast cells growth in the presence of PR 897-930 and in medium lacking histidine and adenine. They were further studied with the help of a reporter gene encoding ␤-galactosidase. Among these clones two corresponded to a fragment of JAB1 as shown by DNA sequencing. JAB1 was cloned in totality, and it was verified that it interacted with PR amino acids 897-930 in yeast cells (Fig. 1A).
It could be argued that the use of a short amino acid segment of a protein does not relate to the physiological situation because this segment might not be accessible in the full-length protein.
We thus analyzed the interaction of JAB1 with the totality of PR DBD-LBD (amino acids 547-930) using the two-hybrid methodology. PR DBD-LBD by itself exerted a hormone-dependent transactivation; however, the latter was markedly enhanced by the addition of AD Gal4 -JAB1 (Fig. 1B). This interaction between PR and JAB1 was only observed in the presence of hormone.
We then used the in vitro pull-down methodology and the mammalian two-hybrid system to establish that the JAB1-PR interaction was not restricted to yeast. [ 35 S]PR was synthetized in vitro and incubated with either the fusion protein GST-JAB1 or GST alone. Only the former retained a major fraction of the [ 35 S]PR (Fig. 2). There was no clear effect of the hormone on this interaction. A limited (25%) effect of hormone was observed when using high ratios of [ 35 S]PR versus GST-JAB1. Similar discrepancies in hormone dependence between in vivo experiments and in vitro pull-down experiments have been described for several coactivators and corepressors (13,41,48,49). It is possible that PR synthetized in vitro is in the active conformation even in the absence of the hormone (13). The DBD Gal4 -PR 897-930 construct was cotransfected along with the AD VP16 construct into CV-1 cells with the corresponding reporter gene. A low level of transcription was observed. When the AD VP16 -JAB1 (activation domain) construct was cotransfected with the reporter gene, there was a slight increase of the basal level of transcription. However, when DBD Gal4 -PR 897-930 and AD VP16 -JAB1 were cotransfected with the reporter gene, a marked increase in CAT activity was observed, indicating an interaction between PR helix 12 and JAB1.
The experiment was repeated with DBD Gal4 -PR 547-930 corresponding to the DBD-LBD domains of the receptor. A hormonedependent transactivation of the reporter gene was observed with this construct (Fig. 3B). It was enhanced by cotransfection with AD VP16 -JAB1. The latter interaction was hormone-dependent and was not observed in presence of the antagonist RU486.
JAB1 Potentiates Nuclear Receptor-mediated Gene Transactivation-Because JAB1 was shown to interact with PR, it was important to determine whether it modulated the biological activity of PR. We thus cotransfected cells with PR, a PREdriven reporter gene, and increasing amounts of JAB1. As shown in Fig. 4A, JAB1 markedly increased hormone-induced reporter gene transcription. The effect of JAB1 on PR-mediated transcription was strictly hormone-dependent because there was no effect of JAB1 in the absence of hormone nor in the presence of the antagonists RU486 or ZK98299. JAB1 binds to PR LBD. A PR molecule devoid of its LBD is known to exert a constitutive activity (38). As expected, when such a modified PR molecule was cotransfected with JAB1, the latter did not potentiate its activity (Fig. 4B).
The effect of JAB1 on other nuclear receptors was then examined. As shown in Fig. 5, JAB1 potentiated the transactivation properties of most receptors. The strongest activity was exerted on steroid hormone receptors (glucocorticoid, mineralocorticoid, androgenic, and estrogenic receptors). Only limited activity was observed on thyroid hormone, vitamin D, and retinoic acid receptors.
Effect of JAB1 on PR-and GR-mediated Inhibition of AP1 Activity-Glucocorticoid and progesterone receptors are known to inhibit the activity of the AP1 complex (50 -54). Because JAB1 binds to both of these transcription factors, its effect on this repressive activity was examined. As shown in Fig. 6, JAB1 enhanced phorbol ester (TPA) activity on the collagenase promoter. PR and GR incubated with the corresponding hormone inhibited TPA-induced transcription. Cotransfection with JAB1 only partially relieved this inhibition. The experiment was repeated with various concentrations of JAB1 (3-10 g/ml of expression vector) yielding similar results. JAB1 thus does not seem to play a major role in the PR or GR inhibitory effect on AP1 transactivation. JAB1 Interacts with SRC-1-SRC-1 has been shown to be a common coactivator for nuclear receptors and AP1 (8). We thus considered the possibility that the effects of JAB1 on transactivation by both nuclear receptors and AP1 were mediated by SCR-1. Before testing this hypothesis directly, we examined the effect of JAB1 on another transcription factor also known to interact with SRC-1: NFB (55,56). As shown in Fig. 7, co-transfection with JAB1 strongly enhanced NFB activity on a target gene.
The question was thus raised as to whether JAB1 directly contacted SRC-1 or rather that its activity was due to some indirect mechanism. To answer this question, initial experiments were performed with the two-hybrid system in yeast (Fig. 8A). The DBD Gal4 -JAB1 construct alone enhanced reporter gene transcription to some extent. However, cotransfection of yeast cells with AD Gal4 -SRC-1 provoked a markedly stronger transcription activation, demonstrating that an interaction can occur between JAB1 and SRC-1. Similar results were obtained using a mammalian two-hybrid system in CV-1 cells (Fig. 8B).
A pull-down experiment with the fusion protein GST-JAB1 and in vitro translated 35 S-labeled SRC-1 confirmed this interaction (Fig. 9). This method was used to map the site of interaction on SRC-1 using a variety of deletion mutants (Fig. 9B). As shown in Fig. 9 (C and D), a major binding region for JAB1 was localized in the C-terminal part of SRC-1 (amino acids 1139 -1440).
JAB1 Stabilizes the Interaction between PR and SRC-1-Because JAB1 interacts with both PR and SRC-1, it appeared possible that its mechanism of action on PR-mediated transactivation might result from stabilization of PR-SRC-1 complexes. To examine this possibility, we used a mammalian three-hybrid system. Cells were cotransfected with a reporter FIG. 5. JAB1 enhances the hormonedependent transactivation by nuclear receptors. CV-1 cells were cotransfected with expression vectors encoding the different receptors (0.5 g/ ml), the corresponding reporter genes (5 g/ml), and increasing amounts of pSG5-JAB1. The cells were treated with the corresponding hormone (f) or untreated (Ⅺ). Plasmids encoding the human glucocorticoid receptor (pSG-hGR), the human androgen receptor (pSG5-hAR), and the human mineralocorticoid receptor (pSG-MR) were cotransfected with the PRE 2 -TATA-CAT reporter gene in the presence or absence of dexamethasone (10 nM), dihydroxytestosterone (50 nM), and aldosterone (50 nM), respectively. The plasmid encoding the human estrogen receptor (pKSV-hER) was cotransfected with the ERE 2 -TATA-CAT reporter gene in the presence or absence of estradiol (10 nM). The plasmid encoding the mouse thyroid hormone receptor ␣ (pRSV-mTR␣1) was cotransfected with the triiodothyronine-RE-TK-CAT reporter gene in the presence or absence of triiodothyronine (10 nM). The plasmid encoding the human vitamin D receptor (pAd-VDR) was cotransfected with the VDRE-TK-CAT reporter gene (PUT-KAT3) in the presence or absence of 1,25-dihydroxycholecalciferol (100 nM). The plasmids encoding the retinoic acid receptors RAR␤ (pSG-RAR) (0.5 g/ml) and hRXR␣ (0.5 g/ml) were cotransfected with the RARE-TK-CAT reporter gene (⌬MTV-TREp-CAT) in the presence or absence of all-trans-retinoic acid (1 M). The vectors encoding the receptors and the corresponding reporter genes are described either under "Experimental Procedures" or by Guiochon-Mantel et al. (31). CAT activities were normalized (1 ϭ CAT activity in the presence of hormone and in the absence of JAB1). Similar results were obtained in three independent experiments. gene driven by Gal4 upstream activation sequence and a DBD Gal4 -PR 1-930 expression vector (Fig. 10). Under the effect of the hormone, there was enhanced transcription from the reporter gene. If a vector encoding SRC-1-VP16 activation domain was also cotransfected, the interaction between PR and SRC-1 enhanced reporter gene transcription. Finally, addition of JAB1 further increased this transcription, showing a stabilization of PR-SRC-1 complexes.
To confirm this result we studied in vitro the stabilization of PR-SRC-1 complexes by JAB1 (Fig. 11). JAB1 Does Not Regulate PR or SRC-1 Concentration-Many members of the JAB1 family of proteins are present in proteasomes. It has been proposed that the activity of JAB1 is related to a disruption of proteasome activity and to the ensuing increased concentration of transcription regulatory factors (4). We thus examined the possibility that in the above experiment JAB1 was increasing target gene transcription by increasing either the concentration of the corresponding nuclear receptor or of SRC-1. Cells were cotransfected with JAB1 and PR or SRC-1. The cells were either treated by the agonist R5020 or the antagonist RU486 or left untreated. Western blots were performed using either anti-PR or anti-SRC-1 antibodies (Fig. 12).
As described previously (57), treatment by R5020 provoked a decrease of receptor migration (called upshift) that is related to receptor phosphorylation. A similar but less marked effect was observed after RU486 treatment. In all cases, cotransfection with JAB1 did not change receptor concentration (Fig.  12A). SRC-1 concentration was also independent of JAB1 overexpression (Fig. 12B).

FIG. 6. Effect of JAB1 on PR and GR-mediated inhibition of AP1 activity. A, HeLa cells were cotransfected with the reporter plasmid
Ϫ73Col-TK-CAT (2 g/ml), the vector pSG5-PR (0.5 g/ml) and pSG5-JAB1 (5 g/ml). The cells were treated (ϩ) or not (Ϫ) with R5020 (10 nM) and TPA (80 ng/ml). B, identical to A except that no receptor was transfected (GR is present in HeLa cells) and dexamethasone (10 nM) was used instead of R5020. CAT values were normalized (100% ϭ CAT activity in the presence of TPA and in the absence of hormone and JAB1). Similar results were obtained in three separate experiments.

DISCUSSION
JAB1 has been shown to potentiate AP1-mediated transactivation of its target genes (1). The mechanism of this effect remained unclear. It was proposed that JAB1 stabilized AP1-TRE (TPA-response element) complexes. However gel shift experiments failed to demonstrate the presence of DNA complexes containing both AP1 and JAB1. We show here that JAB1 contacts not only AP1 but also SRC-1 and may thus act by stabilizing their interaction. The activity of JAB1 is not restricted to AP1 but also applies to other transcription activation factors acting, at least in part, through SRC-1 and including nuclear receptors and NFB. JAB1 (molecular mass, 37.5 kDa) has very recently been shown to be a subunit of a large (ϳ450 kDa) multiprotein complex called JAB1-containing signalosome (5) or COP9 signalosome (6). The signalosome has been characterized in plants (58) where it was initially called COP9 complex and in animals (5). It is apparently absent from the unicellular eukaryote Saccharomyces cerevisiae. The eight subunits of the signalosome have been cloned. They exhibit homology to components of the regulatory 19 S particle of the proteasome. The latter is composed of a base containing all the ATPases and of a lid (7,59). All the proteins homologous to subunits of the signalosome are present in the lid. Moreover there is a one-to-one homology between subunits of the COP9 signalosome and components of the lid. This observation suggests that both structures have a common evolutionary origin and have diverged to assume different functions. The signalosome has been found to contain a kinase activity that phosphorylates c-Jun, inhibitor B, and p105, the precursor of NFB. Various functions have been described for the different subunits of the signalosome. Subunit 1, also called GPS1, inhibits JNK1 (Jun N-terminal kinase) and represses Jun-dependent promoter activity (60). Subunit 2 has independently been described as TRIP15 (thyroid hormone interacting protein 15) (61). It interacts with the ligand-binding domain of the thyroid hormone receptor and of the retinoic acid receptor RXR. This interaction is inhibited by the addition of ligand. The TRIP15 gene is highly homologous to the Drosophila alien gene, which is expressed in the muscle attachment sites during embryogenesis (62). It has recently been shown that alien is a corepressor of members of the nuclear receptor family (63). Subunit 5 of the signalosome is JAB1, whereas subunit 6 is VIP (human HIV-1 Vpr protein interacting protein), which is involved in the viral protein Vpr-induced cellular differentiation and growth arrest (64). Cells expressing antisense VIP are blocked in the G2-M phase of the cell cycle. Because the description of the signalosome is very recent and minor constituents may also associate with the particle, further description of other functions is highly probable. In the plant Arabidopsis thaliana the COP9 signalosome complex is involved in light-dependent morphogenesis (7,58,65).
Interestingly, whereas the lid of the 19 S proteasome particle and the signalosome have diverged structurally and functionally, other components of the proteasome system have evolved to assume functions often related to transcription regulation in addition to their function in proteolysis. Indeed TRIP1/SUG1, which is a component of the proteasome, interacts with the ligand-binding domain of nuclear receptors and potentiates their transcriptional activity (66,67). MSS1 (mammalian suppressor of sgv1) plays a role in HIV-1 TAT-mediated transcription (68,69). E6-AP (E6 associated protein) is an ubiquitinprotein ligase involved in the Angelman syndrome. It is also a coactivator of the progesterone receptor (70). Ubiquitinylation may be involved in gene transcription regulation because histone-ubiquitin conjugates are concentrated in nucleosomes of transcribed genes (71,72). JAB1 has recently been shown to interact with cyclin-dependent kinase inhibitory protein p27 KIP1 , to provoke its translocation from the nucleus to the cytoplasm and to enhance its degradation (73). JAB1 belongs to a new category of transcription regulators that act by bridging coactivators to receptors. This is also the case for cyclin D1, which binds both estrogen receptor and SRC-1 (74). However, cyclin D1 exerts this effect on both hormone-bound and unliganded estrogen receptor, whereas JAB1 interacts only with hormone-PR complexes.
Nuclear receptors, their coactivators and coregulators are present in large multimeric complexes (27). JAB1 is localized in signalosomes. It is thus possible that both types of multiprotein complexes cross-talk in the cells. Indeed larger structures, associating weakly or transiently with the COP9 signalosome, have been observed (7). It has been shown that the COP9 signalosome contains kinase activities that phosphorylate AP1 and NFB on sites overlapping with those implicated in transcriptional activation. Further work will be necessary to establish whether the signalosome plays a role in the phosphorylation of nuclear receptors and of their coactivators.
Acknowledgments-We are grateful to Dr. P. Legrain (Institut Pasteur Paris, France) for advice and technical help with the yeast twohybrid system. We thank Dr. P. James (University of Wisconsin, Madison, USA) for the kind gift of the yeast strain PJ69 -4A. We also thank Dr. J. Bertoglio (INSERM, Chatenay-Malabry, France) for the reporter FIG. 10. JAB1 stabilizes PR-SRC-1 complexes in vivo. The effect of JAB1 on PR-SRC-1 complexes was analyzed by a three-hybrid system in CV-1 cells. The cells were cotransfected with the pG5CAT reporter plasmid (5 g/ml), the vector encoding AD VP16 -SRC-1 (q) (1 g/ml), the vector encoding DBD-Gal4 fused to the PR (DBD Gal4 -PR 1-930 ) (0.2 g/ ml), along with increasing concentrations of pSG5-JAB1 vector. The cells were incubated with R5020 (10 nM). In control experiments AD VP16 -SRC-1 was replaced by AD VP16 (⅜). The results are the means Ϯ S.E. of three separate experiments.  12. Overexpression of JAB1 does not change the concentration of PR and SRC-1. A, CV-1 cells were transfected with pSG5-rPR (10 g/ml) in the presence or absence of of pSG5-HA-JAB1 (10 g/ml). The cells were treated with either R5020 (10 nM) or RU486 (10 nM). The cells were harvested in Laemmli buffer, and aliquots were electrophoresed as described under "Experimental Procedures." PR was revealed with the monoclonal Let126 antibody. B, CV-1 cells were transfected with pSG5-rPR (5 g/ml) in the presence or absence of pSG5-HA-JAB1 (10 g/ml) and of pSG5-HA-SRC-1 (10 g/ml). The cells were treated or not with R5020 (10 nM). The cells were harvested in RIPA buffer, and aliquots were electrophoresed as described under "Experimental Procedures." SRC-1 was revealed with the monoclonal anti-HA antibody.
plasmid Ϫ73Col-TK-CAT and Dr. F. Bachelerie (Institut Pasteur Paris, France) for the expression plasmid of p65 RelA and its reporter gene NFB-CAT. We thank C. Carreaud for technical assistance and M. Quesne for help in Western blotting. The manuscript was prepared by V. Coquendeau, A.D. Dakhlia and M. Rodrigues.