Regulation of D2 receptor (D2R) expression is crucial in the function of dopaminergic systems. Because alterations of D2R expression may contribute to the development of different disorders, it is important to elucidate the mechanisms regulating D2R gene transcription. We report the characterization of two putative nuclear factor-κB (NF-κB) motifs, referred to as D2-κB sites, in the human D2R promoter, and demonstrate that they bind NF-κB subunits and stimulate D2R promoter activity. D2-κB sites show different degrees of conservation and specificity, when compared with canonical kB sites. The D2-κB1 site (from −407 to −398) is highly conserved and binds p50/p65 and p50/c-Rel complexes, whereas D2-κB2 (from −513 to −504) is more degenerated and only binds p50/p65 heterodimers. Activation of D2-κB sites in COS-7 cells expressing a luciferase reporter vector containing the D2R promoter resulted in increased transcriptional activity. Site-directed mutagenesis of each D2-κB site differentially modified D2R promoter activity. In particular, mutation of the D2-κB1 motif did not affect D2R promoter response to p50/c-Rel complexes, whereas inactivation of the D2-κB2 site decreased it. Mutations of either D2-κB1 or D2-κB2 sites attenuated the D2R promoter transcriptional efficiency induced by p50/p65 complexes. Thus, D2R transcription mediated by p50/c-Rel is supported mainly by the D2-κB2 site, whereas both sites are necessary to support the full transcriptional activity mediated by p50/p65 complexes. A correlation was found between NF-κB activity and D2R expression in the pituitary and pituitary-derived cells but not in the striatum, suggesting that NF-κB regulation of D2R expression could be a pituitary-specific mechanism.
DOPAMINE (DA) IS the predominant catecholamine neurotransmitter in the central nervous system where it plays a leading role in the regulation of physiological functions such as locomotor activity, cognition, positive reinforcement, and hormone secretion. Dysregulations of dopaminergic transmission have been related to specific pathological conditions such as schizophrenia and Parkinson’s disease.
The effects of DA are mediated by five different receptors that belong to the seven-transmembrane G protein-coupled receptor family and are divided into D1-like (D1 and D5) and D2-like (D2, D3, D4) subtypes on the basis of their structural and functional properties (1). Among these, the D2 receptor (D2R) appears to be particularly relevant because it mediates most of the physiological effects of DA and it is the molecular target of drugs used in the therapy of schizophrenia, Parkinson’s disease, and hyperprolactinemia. It has been shown that the knockout of the D2R gene results in the development of an impaired locomotor parkinsonian-like phenotype (2), in the development of pituitary tumors (3, 4), and in the loss of the rewarding effects of opiates (5), suggesting that alterations of D2R gene expression might contribute to the development of different disorders. Thus, understanding how expression of the D2R-encoding gene is controlled may represent an essential step to develop novel strategies to modulate DA function.
Aberrant expression of the D2R may potentially arise from disturbances at any step in its transcriptional regulatory pathways, including altered activation or expression of specific transcription factors. For example, it has been reported that combined ablation of retinoic acid receptor (RAR) and retinoid X receptor (RXR) genes that transcriptionally control D2R expression (6) results in the loss of striatal D2R and in the appearance of parkinsonian-like neurological symptoms in the null mice (7, 8). These observations highlight the importance of elucidating the transcriptional mechanisms regulating the expression of the D2R gene by identifying pathophysiologically relevant consensus target sequences and the regulators interacting with these sites. Sequence analysis of the D2R promoter revealed that it contains multiple Sp1 binding sites in a GC-rich region that lacks TATA or CAAT boxes (9–13), a feature common to several genes expressed in the central nervous system (14–17) including all of the DA receptor promoters characterized so far (6, 7, 9–13, 18–22). Because this structure is typical of housekeeping genes, one crucial aspect of neuronal gene promoter function is represented by the mechanisms of region-specific and neuron-specific expression. In fact, D2R expression is regulated developmentally and is confined to specific cell populations in the central nervous system and neuroendocrine system (1), suggesting that specific transcription factors must control the pattern of expression of this gene. On this line, various regulatory sequences including Sp1, Sp3, activator protein 1 (AP-1), and retinoic acid responsive element have been identified in the D2R promoter (6, 7, 9–13). Moreover, by using human pituitary tumor cells, where the expression of the D2R is under the control of nerve growth factor (23, 24), we first suggested that a novel regulatory pathway controlling D2R gene expression might involve members of the nuclear factor-κB (NF-κB) family (25). However, the mechanisms of NF-κB-mediated induction of D2R gene transcription are still unknown. In particular, whether a direct binding of NF-κB species to the D2R promoter is involved or other factors are interposed in this interaction remains to be elucidated.
In this study, we report the identification and characterization of two putative NF-κB-responsive elements in the human D2R promoter and demonstrate that these sequences, which we refer to as D2-κB sites, specifically bind NF-κB species and exert an essential positive control on D2R gene transcription.
Materials and Methods
Cell culture
Prolactinoma cells were grown in Ham’s F10 medium supplemented with 2.5% fetal calf serum, 15% horse serum, 4 mm glutamine, and 100 U penicillin-streptomycin. COS-7 cells were cultured in DMEM containing 10% fetal calf serum, 2 mm glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin.
Nuclear extract preparation
Nuclear extracts were obtained from one prolactinoma cell line previously characterized as containing constitutively active NF-κB (25). Cells were rinsed, harvested in ice-cold PBS, and pelleted at 4 C (800 × g, 5 min). Cells were lysed in ice-cold lysis buffer containing 10 mm HEPES, pH 7.9, 10 mm KCl, 1.5 mm MgCl2, 0.5 mm dithiothreitol (DTT), 0.5 mm phenyl-methyl-sulfonyl fluoride, and a complete set of protease inhibitors (Complete Protease Inhibitors; Roche Molecular Biochemicals, Milan, Italy) and centrifuged at 800 × g at 4 C for 5 min. The resulting pellet was resuspended in the lysis buffer added with 0.5% Nonidet P-40, homogenized with a Dounce homogenizer, and centrifuged at 2500 × g for 5 min at 4 C. The resulting pellet, containing the nuclei, was resuspended in ice-cold 20 mm HEPES, pH 7.9, containing 420 mm NaCl, 1.5 mm MgCl2, 0.2 mm EDTA, 25% glycerol, 0.5 mm DTT, 0.5 mm phenyl-methyl-sulfonyl fluoride, and the protease inhibitors, incubated on ice for 20 min, and centrifuged at 14,000 × g for 15 min at 4 C. The supernatant containing the nuclear proteins was stored at −80 C. Protein concentration was determined by the Bio-Rad Bradford assay (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions.
Synthetic DNA oligonucleotides
Oligonucleotides corresponding to the two putative D2-κB sites in the 5′-regulatory region of the human D2R promoter were used. Sequences were as follows: D2-κB1, from −407 to −398 of the D2R promoter (antisense strand), 5′-GCCTGAGGGAGGTGCCTCACAT-3′ and its complementary strand; mutated D2-κB1 (mD2-κB1), 5′-GCCTGActtgGGTGCCTCACAT-3′ and its complementary strand; D2-κB2 from −513 to −504 of the D2R promoter (antisense strand), 5′-GCCTGAGGGGGATTCCTCACAT-3′ and its complementary strand; mutated D2-κB2 (mD2-κB2), 5′-GCCTGAGctcGATgaCTCACAT-3′ and its complementary strand. The specific NF-κB sequence 5′-CAAGTTTAGGGGACTTTCCCGGGCGCCT-3′ from the mouse immunoglobulin k gene and its mutated form 5′-CAAGTTTAGtGtACTTaCCGGGCGCCT-3′ (mNF-κB) and the octamer protein binding site from the IL-2 gene enhancer region (octa) 5′-TATGTGTAATATGTAAAACATTTTGACACC-3′ were also used. The underlined sequences represent the putative NF-κB sites. For gel shift analysis, 200 ng of sense oligonucleotides were end-labeled with [γ-32P]ATP (3000 Ci/mmol; NEN Life Science Products Life Sciences, Milan, Italy) and T4 polynucleotide kinase (Promega, Milan, Italy) for 1 h at 37 C. The labeled oligonucleotides were annealed with their respective complementary strands for 3 min at 90 C, 10 min at 65 C, 10 min at 37 C, and 5 min at room temperature, and the double-stranded oligonucleotides were purified by denaturing 20% polyacrylamide gel electrophoresis.
EMSA assay
EMSA was carried out as previously reported (25). Briefly, DNA binding reactions were initiated by combining 10 μg of nuclear extracts with 100,000 cpm (0.5 ng) of γ-32P-labeled oligonucleotide in 25 μl of 10 mm Tris-HCl, pH 7.5, containing 50 mm NaCl, 1 mm DTT, 1 mm EDTA, 0.5 μg poly(dIdC). In competition experiments, 2.5- and 40-fold molar excess of unlabeled oligonucleotides were added together with 32P-labeled probes. Reactions were carried out for 20 min at room temperature, and protein-DNA complexes were resolved on nondenaturing 5% polyacrylamide gel in Tris/borate/EDTA buffer. Gels were dried and subjected to autoradiography at −80 C by using Kodak Biomax MR films. In supershift experiments, 10 μg of nuclear extracts were incubated for 1 h at 4 C with 25 μg of selected antibodies before addition of the other components of the reaction mixture. Incubation was continued for an additional 20 min at room temperature. The following antibodies were used: a monoclonal anti-p50 antibody (Santa Cruz Biotechnology Inc., Heidelberg, Germany), a monoclonal anti-p65/RelA antibody (Santa Cruz Biotechnology Inc.), a monoclonal anti-p65/RelA antibody (Roche Molecular Biochemicals), a monoclonal anti-cRel antibody (Santa Cruz Biotechnology Inc.); a monoclonal anti-RelB antibody (Santa Cruz Biotechnology Inc.); a monoclonal anti-p52 antibody (Santa Cruz Biotechnology Inc.). Preimmune serum was used as a control of antibody specificity.
Preparation of plasmid constructs
A fragment consisting of 284 bp of the 5′ flanking sequence and 20 bp of the first exon of the human D2R gene has been previously shown to be the minimal promoter driving the transcription of a reporter gene (26). This 304-bp fragment was prepared by a two-step PCR with the sense primer 5′-ACTGGCGAGCAGAGCGTGAGGACCC-3′ and antisense primer 5′-TGCGCGCGTGAGGCTGCCGGTTCGGC-3′ by using the human D2R promoter clone (9) as template (see Fig. 1). The KpnI linker was added to sense and the HindIII linker to antisense primer. The reaction was performed with native Pfu DNA polymerase (Stratagene, Milan, Italy) in the presence of 4% formamide at 98 C for 1 min followed by 35 cycles at 98 C for 20 sec and at 75 C for 5 min. After sequencing, the generated fragment was cloned into the luciferase reporter plasmid pGL3-basic (Promega). The kB/Rel expression plasmids pSG-p50, pSG-p65/RelA, and pSG-cRel have been described previously (27).
Nucleotide sequence of the human D2R gene promoter. The D2R promoter sequence reported by Gandelman et al. (9 ) is shown. Putative response elements for NF-κB, referred to as D2-κB1 and D2-κB2, are in bold and underlined. Primers used to amplify the D2R minimal promoter are in bold and indicated by the arrows.
In vitro mutagenesis
Mutations of the putative D2-κB sites were performed by site-directed mutagenesis combined with overlapping PCR according to Torres et al. (28). In particular, mutations of the D2-κB1 site were obtained by PCR with the mutant sense primer 5′-ATGTGAGGCACCgaagTCAGGC-3′ (primer 1) or its complementary antisense sequence (primer 2) paired with the antisense primer 5′-TGCGCGCGTGAGGCTGCCGGTTCGGC-3′ (primer 3) or the sense primer 5′-ACTGGCGAGCAGAGCGTGAGGACCC-3′ (primer 4), respectively (see Fig. 1), and the pGL3-D2 vector as a template. The reactions were performed with the Pfu DNA polymerase for 35 cycles at 95 C for 30 sec, 60 C for 30 sec, 72 C for 90 sec. The products of these PCR were purified, combined together, and used as template in the overlapping PCR with primers 3 and 4 for 35 cycles at 95 C for 30 sec, 55 C for 30 sec, 72 C for 2 min. The mutation of the D2-κB2 site was obtained with the same method using the mutant oligonucleotide primer 5′-ATGTGAGGAATCgagCTCAGGC-3′ and its complementary sequence paired with primers 3 and 4, respectively, with the pGL3-D2 vector as template. The overlapping PCR was performed with primers 3 and 4 as previously described. pGL3-D2 constructs with mutations of either D2-κB1 or D2-κB2 site were prepared.
Cell transfection and luciferase activity
COS-7 cells (60–80% confluent) were transiently cotransfected for 3 h with 1) pGL3-D2 (1 μg); 2) pGL3-D2 (1 μg) and pSG-cRel (1 μg); 3) pGL3-D2 (1 μg) and pSG-p65/RelA (1 μg); 4) pGL3-D2 (1 μg), pSG-p50 (1 μg), and pSG-cRel (1 μg); or 5) pGL3-D2 (1 μg), pSG-p50 (1 μg), and pSG-p65/RelA (1 μg) using the lipofectamine technique (Lipofectamine Reagent; Invitrogen-Life Technology, Milan, Italy). The total amount of transfected DNA was adjusted to 3 μg with the empty pSG vector. In another group of experiments, pGL3 vectors containing D2-κB1- or D2-κB2-mutated sequences were cotransfected in COS-7 cells with pSG plasmids containing the different NF-κB subunits as follows: 1) pGL3-mD2-κB1 (1 μg), pSG-p50 (1 μg), and pSG-p65/RelA (1 μg); 2) pGL3-mD2-κB1 (1 μg), pSG-p50 (1 μg), and pSG-cRel (1 μg); 3) pGL3-mD2-κB1 (1 μg) and pSG-p50 (2 μg); 4) pGL3-mD2-κB2 (1 μg), pSG-p50 (1 μg), and pSG-cRel (1 μg); 5) pGL3-mD2-κB2 (1 μg), pSG-p50 (1 μg), and pSG-p65/RelA (1 μg); 6) pGL3-mD2-κB2 (1 μg) and pSG-p50 (2 μg). Transfection efficiency throughout the experiments was monitored by cotransfection with a Renilla luciferase expression vector (50 ng). Transfected cells were cultured in the complete medium for 48 h, harvested, lysed, and assayed for luciferase activity by using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer’s instructions. Firefly luciferase activity in each sample was normalized to Renilla luciferase activity.
Results
By means of computer-assisted analysis (29), two putative NF-κB binding sites were identified in the 5′-regulatory region of the human D2R. In particular, according to the D2R promoter sequence published by Gandelman et al. (9), a decameric motif corresponding to 5′-GGGAGGTGCC-3′, designated as D2-κB1, was detected in position −407 to −398 on the antisense strand and one corresponding to 5′-GGGGGATTCC-3′, designated as D2-κB2, was found in position −513 to −504 on the antisense strand (Fig. 1). Analysis of each D2-κB sequence showed that the 5′ pentameric half-site, containing the three subsequent G residues present in all NF-κB binding sites identified so far (30), is well conserved in both D2-κB sites. By contrast, the 3′ pentameric motif is conserved in the D2-κB2 site, but is more degenerated in the D2-κB1 site with G residues in positions 6 and 8, which are rather uncommon in the known NF-κB sites (30). Because these sequences are very similar to the currently known consensus sequences for NF-κB transcription factors (30), we investigated whether they may be binding sites for kB-related proteins. For this purpose, we used nuclear proteins obtained from one prolactinoma cell line, that was previously characterized as expressing the D2R and containing constitutively activated NF-κB proteins (25), and double-stranded 32P-labeled oligonucleotides, comprising the D2-κB sites, in EMSA experiments. The oligonucleotide sequences used are reported in Fig. 2A, and a representative EMSA with 32P-labeled D2-κB1 and D2-κB2 probes is reported in Fig. 2B. Both D2-κB1 and D2-κB2 oligonucleotides (lanes 2 and 4) as well as the canonical NF-κB sequence from the mouse immunoglobulin k-gene (31) (lane 1) were able to form DNA/protein complexes with retarded electrophoretic migration. The binding activity of these oligonucleotides was disrupted by mutations in their core kB-like sequences (Fig. 2, A and B, lanes 3 and 5), suggesting that they form relevant DNA/protein complexes. The D2-κB1 and D2-κB2 sites were characterized in both competition and supershift experiments. As reported in Fig. 3A, the D2-κB1 probe formed a major retarded DNA/protein complex (lane 1) that was specifically competed by both 2.5- and 40-fold molar excess of the same cold oligonucleotide (lanes 2 and 3), but not by its mutated counterpart (lane 4). However, the NF-κB sequence from the mouse immunoglobulin k-gene inhibited the binding of labeled D2-κB1 probe to nuclear proteins with low efficiency. Although a 2.5-fold molar excess of the probe only marginally interfered with D2-κB1/protein interaction (lane 5), a 40-fold molar excess of this probe efficiently competed with D2-κB1 oligonucleotide binding to nuclear proteins (lane 6), suggesting that D2-κB1 may represent a low-specificity site for NF-κB. The mutated version of the NF-κB probe, with three base changes in the kB-like core sequence (lanes 7 and 8), as well as the unrelated oligonucleotide containing an octamer binding site (lane 9), did not show any competing activity.
EMSA analysis of nuclear protein binding to D2-κB1 and D2-κB2 DNA sequences. A, Nucleotide sequences of the NF-κB site of the mouse immunoglobulin k gene, D2-κB1, D2-κB2, and their mutant counterparts (mD2-κB1 and mD2-κB2). The kB motifs are highlighted in bold. The mutated base pairs are in lowercase and underlined. B, EMSA analysis with 10 μg of nuclear proteins obtained from a D2R-expressing human prolactinoma cell line and the γ-32P end-labeled oligonucleotide probes NF-κB (lane 1), D2-κB1 (lane 2), mD2-κB1 (lane 3), D2-κB2 (lane 4), and mD2-κB2 (lane 5). The positions of DNA-protein complexes (solid arrowheads) and of the uncomplexed DNA (open arrowheads) are shown. Data are representative of three independent experiments.
Characterization of the D2-κB1 site. A, DNA binding competition assay. EMSA was carried out with 10 μg of nuclear proteins and 32P-labeled D2-κB1 probe in the absence (lane 1) or in the presence of 2.5- and 40-fold molar excess of cold com-petitors. The following cold oligonucleotides were used: D2-κB1 (lanes 2 and 3), mutated D2-κB1 (lane 4), NF-κB of the mouse immunoglobulin k gene (lanes 5 and 6), mutated NF-κB (lanes 7 and 8), and oligonucleotide for octamer binding protein (octa, lane 9). The positions of DNA-protein complexes (solid arrowheads) are shown. Data are representative of three independent experiments. B, Analysis of the molecular composition of the D2-κB1/protein complex by supershift experiments. Ten micrograms of nuclear extracts were incubated in the presence of 25 μg of antibodies raised against p50 (lane 2), p65/RelA (lane 3), c-Rel (lane 4), and RelB (lane 5) subunits before addition of the 32P-labeled D2-κB1 oligonucleotide and the other components of the reaction mixture, as described in Materials and Methods. The specificity of the antibodies was determined by using the preimmune serum as a control (lane 6). The positions of DNA-protein complexes (solid arrowheads) are shown. Data are representative of three independent experiments.
The NF-κB family of transcription factors is composed of several members, including p50, p52, p65/RelA, c-Rel, and RelB, that form heterodimers which are able to trigger signaling from the cell membrane to the nucleus (31, 32). To identify the NF-κB subunits that bind to the D2-κB1 site, antibodies to p50, p65/RelA, c-Rel, and RelB were tested for their ability to either interfere with DNA-protein binding or supershift DNA-protein complexes. As shown in the representative supershift analysis reported in Fig. 3B, both p50 (lane 2) and p65/RelA antisera (lane 3) inhibited the formation of D2-kB1/protein complexes. The anti-c-Rel antibody apparently interfered with DNA/protein interaction with less efficiency (lane 4), and the anti-RelB antibody did not recognize the D2-κB1/protein complex (lane 5). The specificity of these results was confirmed by incubation with preimmune serum (lane 6). These data suggest that complexes mainly containing the p50, p65/RelA and, to a lesser extent, the c-Rel subunits may interact with the D2-κB1 site. The characterization of the D2-κB2 site is reported in Fig. 4. A representative EMSA is shown in Fig. 4A. D2-κB2/protein complex formation (lane 1) was efficiently and specifically competed by both 2.5-fold and 40-fold molar excess of the same cold oligonucleotide (lanes 2 and 3) and of the canonical NF-κB sequence (lanes 6 and 7), but not by the mutated versions of these probes (lanes 4 and 5 and lanes 8 and 9) and by the unrelated octamer binding sequence (lane 10). A representative supershift analysis is reported in Fig. 4B. The antibodies against p50 (lane 2), p65/RelA (lane 3), and c-Rel (lane 5) interfered with the interaction of the D2-κB2 probe with nuclear proteins. By contrast, the efficiency of anti-RelB (lane 4) and anti-p52 (lane 6) antibodies in inhibiting DNA-protein interaction was similar to that of preimmune serum. Taken together, these observations suggest that complexes containing p50, p65/RelA, and c-Rel subunits may potentially interact with the D2-κB2 site.
Molecular characterization of the D2-κB2 site. A, DNA binding competition assay. EMSA was carried out with 10 μg of nuclear proteins and the 32P-labeled D2-κB2 probe in the absence (lane 1) or in the presence of 2.5- and 40-fold molar excess of cold competitors. The following cold oligonucleotides were used: D2-κB2 (lanes 2 and 3), mutated D2-κB2 (lanes 4 and 5), NF-κB of the mouse immunoglobulin k gene (lanes 6 and 7), mutated NF-κB (lanes 8 and 9), and oligonucleotide for octamer binding protein (lane 10). The positions of DNA-protein complexes (solid arrowheads) are shown. Data are representative of three independent experiments. B, Analysis of the molecular composition of the D2-κB2/protein complex by supershift experiments. Ten micrograms of nuclear extracts were incubated in the presence of 25 μg of antibodies raised against p50 (lane 2), p65/RelA (lane 3), RelB (lane 4), c-Rel (lane 5), and p52 (lane 6) subunits before addition of the 32P-labeled D2-κB2 oligonucleotide and the other components of the reaction mixture, as described in Materials and Methods. The specificity of the antibodies was determined by using the preimmune serum as a control (lane 7). The positions of DNA-protein complexes (solid arrowheads) are shown. Data are representative of three independent experiments.
To define the role of the D2-κB sites in the control of D2R expression, a luciferase reporter vector containing the minimal promoter of the D2R was constructed (26) and transiently transfected into COS-7 cells either alone or in combination with expression vectors containing the kB-related species p50, p65/RelA, and c-Rel (27). As reported in Fig. 5, the transcriptional activity of the D2R promoter was very low in the absence of NF-κB subunits or in the presence of either p65/RelA or c-Rel. However, D2R promoter-driven luciferase expression was significantly increased in COS-7 cells expressing either p65/RelA or c-Rel in combination with the p50 subunit, suggesting that NF-κB is a necessary and sufficient signal to induce D2R gene expression. To further support this view, D2-κB sites in the D2R promoter were mutated in their kB-like core sequence by site-directed mutagenesis. As reported in Fig. 6A, the transcriptional activity of the D2R promoter, coexpressed in COS-7 cells with p50 and c-Rel subunits, was significantly impaired by mutation of the D2-κB2 site, but not by mutation of the D2-κB1 site, an observation that might reflect the low affinity of the D2-κB1 probe for c-Rel, as shown in the supershift analysis reported in Fig. 3B. Similarly, D2R promoter-driven luciferase expression in COS-7 cells expressing the p50 homodimer was significantly decreased by D2-κB2 site mutation, but was not affected by mutation of the D2-κB1 site. On the other hand, D2R promoter activity in COS-7 cells expressing p50 and p65/RelA subunits was significantly impaired by mutations of both D2-κB sites. Taken together, these data support the crucial role of a direct interaction of NF-κB species with specific sequences in the D2R promoter in the regulation of D2R expression.
Transcriptional activation of the D2R promoter by NF-κB subunits. COS-7 cells were cotransfected with the reporter vector pGL3-D2 and the pSG-p50, pSG-p65/RelA, and pSG-c-Rel constructs or the combinations pSG-p50/pSG-p65/RelA and pSG-p50/pSG-c-Rel as described in Materials and Methods. Transfection efficiency in each sample was monitored by cotransfection with the Renilla luciferase expression vector. Forty-eight hours after transfection cells were lysed and assayed for both firefly and Renilla luciferase activities. The firefly luciferase activity in each sample was normalized to Renilla luciferase activity. Each bar represents the mean ± sem of three independent experiments. *, P < 0.001 vs. pGL3-D2, Student′s t test.
Effects of mutations of the D2-κB sites on the transcriptional activity of the D2R promoter. Core sequences of either the D2-κB1 or D2-κB2 site were mutated by site-directed mutagenesis combined with overlapping PCR as described in Materials and Methods. The different constructs were transiently transfected in COS-7 cells together with pSG-p50 and pSG-c-Rel (A) or pSG-p50 and pSG-p65/RelA (B) or pSG-p50 (C) constructs. Transfection efficiency in each sample was controlled by cotransfection with the Renilla luciferase expression vector. Forty-eight hours after transfection, cells were processed for luciferase activity. The firefly luciferase activity in each sample was normalized to Renilla luciferase activity. Data are presented as fold increase of luminescence over the basal value obtained with the pGL3-D2 vector in the absence of NF-κB subunits. Each bar represents the mean ± sem of three independent experiments. **, P < 0.001 vs. pGL3-D2 in the absence of NF-κB subunits; *, P < 0.05 vs. wild-type pGL3-D2, Student’s t test.
We have previously shown that, in pituitary prolactin-secreting tumor cells, the expression of D2R is correlated with NF-κB DNA-binding activity (25). To investigate whether such a correlation may be present also in physiological conditions, various brain regions characterized by different D2R content (1) were analyzed for basal NF-κB activity. As reported in Fig. 7A, areas with the highest D2R expression, namely the striatum (lane 3) and the anterior pituitary (lane 1), showed different NF-κB activity. In particular, whereas in the anterior pituitary NF-κB activity was extremely elevated, in the striatum NF-κB protein binding was similar to that found in other brain areas including the cerebellum (lane 4), where D2R is virtually absent. The NF-kB/DNA binding activity in unstimulated human prolactinoma cell lines (23–25) is reported in Fig. 7B. NF-κB activity was elevated in the cell line that express the D2R (lane 6) and was hardly detectable in the cell line lacking the D2R (lane 7). The functional correlation between NF-κB activity and D2R expression in living cells was previously demonstrated (25). In particular, exposure of a well-characterized prolactinoma cell line that expresses the D2R (23–25) to SN-50, a cell-permeable inhibitor of NF-κB nuclear translocation (33), resulted in D2R loss. Similarly, exposure of prolactinoma cells lacking the D2R to nerve growth factor, that activates NF-κB, resulted in D2R expression (23–25), an effect abolished by SN-50 (25). Taken together, these data suggest that NF-κB regulation of D2R expression may be a mechanism specific for pituitary-derived cells.
EMSA analysis of NF-κB activity in different brain areas and pituitary cells. A, Nuclear proteins from different brain areas were incubated with the γ-32P end-labeled NF-κB oligonucleotide of the mouse immunoglobulin k gene as described in Materials and Methods. Data are representative of three experiments. B, Nuclear proteins from two human prolactinoma cell lines that differ for the presence (D2R+) or the absence (D2R−) of the D2R were incubated with the γ-32P end-labeled NF-κB oligonucleotide. Data are representative of three independent experiments.
Discussion
Precise transcriptional regulation of DA receptor genes is a key element in the physiological function of dopaminergic systems. Among DA receptors, the D2R plays a pivotal role in the regulation of most effects of DA. In fact, it has been shown that deletion of the D2R gene results in an impaired locomotor phenotype resembling Parkinson’s disease (2), and in pituitary tumors in knockout mice (3, 4), suggesting that regulation of D2R gene transcription is crucial in the functioning of dopaminergic systems. Because the D2R possesses a promoter with housekeeping characteristics (6, 10, 11), the expression pattern of this receptor is believed to be temporally and spatially regulated by specific transcription factors. On this line, different regulatory sequences have been identified in the D2R promoter (9–13), and a crucial role of members of RAR and RXR retinoid receptor families in the control of D2R expression in the striatum has been clearly demonstrated (6–8). We recently have suggested that a relevant regulatory pathway controlling D2R gene expression in pituitary cells may involve members of the NF-κB family (25). This family of transcription factors is composed of several distinct DNA-binding subunits which can heterodimerize and homodimerize, thereby forming complexes with distinct cell type distribution, DNA sequence specificity, and transcriptional activity (31, 33, 34).
In this paper, we have identified and characterized two sequences in the 5′-regulatory region of the human D2R gene that are specific binding sites for different NF-κB complexes and are crucial to drive D2R promoter transcriptional activity. In particular, two decameric motifs were identified in the antisense strand at positions −513 to −504 and −407 to −398, with high similarity to the consensus sequence for NF-κB transcription factors (30). These sites, which we referred to as D2-κB, showed different degrees of sequence conservation, when compared with the kB sites identified so far (30). In particular, although the D2-κB2 site was highly conserved, the D2-κB1 site was more degenerated in the 3′ half-site, with two base changes in significant positions. These sequence differences could explain the different specificity of the two D2-κB sites for NF-κB subunits. In the case of the more conserved D2-κB2 site, homologous competition with the unlabeled D2-κB2 sequence and competition with the kB-related site 5′-GGGACTTTCC-3′ found in the mouse immunoglobulin k gene (31) were, in fact, equally effective. By contrast, in the case of the more degenerated D2-κB1 sequence, homologous competition was much more effective than competition with the NF-κB sequence. A significant difference between D2-κB1 and D2-κB2 sites was also evident in supershift experiments, showing that the two sites apparently bind different members of the NF-κB family. Antibodies to p50, p65/RelA, and c-Rel interfered, in fact, with D2-kB2 binding to nuclear proteins, whereas anti-p50 and anti-p65/RelA and, to a lesser extent, anti-c-Rel antibodies interfered with D2-κB1-protein interaction, suggesting that D2-κB2 may interact with the p50/p50 homodimer and with both p50/p65/RelA and p50/c-Rel heterodimers, whereas D2-κB1 only interacts with p50/p50 and p50/p65/RelA complexes. Interestingly these differences were also supported by site-directed mutagenesis of the two D2-κB sites. The minimal D2R promoter inserted into a luciferase reporter vector and transiently transfected into COS-7 cells together with expression vectors containing kB-related species was endowed with transcriptional activity as demonstrated by the fact that the p50/p50 homodimer and both p50/c-Rel and p50/p65/RelA heterodimers, but not p65/RelA and c-Rel homodimers, significantly increased luciferase expression. Site-directed mutagenesis in the core sequences of each D2-κB site differentially modified the transcriptional efficiency of the D2R promoter induced by either p50/p50, p50/c-Rel, or p50/p65/RelA complexes. In particular, mutation of the D2-κB1 motif did not affect the ability of p50/c-Rel complexes to drive D2R promoter-mediated luciferase expression, suggesting that when the D2-κB1 site is inactivated, interaction of p50/c-Rel with D2-κB2 is sufficient to maintain D2R promoter activity. Inactivation of the D2-κB2 site resulted in a remarkable decrease of transcriptional activity of the D2R promoter mediated by p50/c-Rel complexes, suggesting that when the D2-κB2 site is inactivated the p50/c-Rel complex does not support transcription. These observations are in line with the data obtained in supershift experiments showing that D2-κB1 site only marginally interacts with c-Rel-containing complexes and suggest that D2R transcription mediated by c-Rel is entirely supported by the D2-κB2 site. Similar results were obtained in cells expressing the p50/p50 homodimer. In fact, inactivation of the D2-κB2 site, but not of the D2-κB1 site, remarkably decreased D2R promoter response to p50/p50 complexes, suggesting that p50-induced transcription is mediated by the D2-κB2 site. Different results were obtained in cells cotransfected with p50 and p65/RelA subunits. Mutations of either D2-κB1 or D2-κB2 sites attenuated, but did not abolish the transcriptional efficiency of the D2R promoter, suggesting that both sites are necessary to support the full transcriptional activity mediated by p50/p65/RelA complexes.
A delicate balance among several nuclear factors tightly regulates the expression of the D2R gene in specific brain areas. Retinoids and DA receptor regulating factor appear to be required for the full expression of D2R in the striatum. In fact, ablation of RAR-RXR retinoid receptor genes results in D2R loss in the striatum (8), and DA receptor regulating factor activity may be modulated by manipulations of striatal DA transmission-altering D2R expression (12). Our present data strongly suggest that NF-κB represents another critical transcription factor promoting D2R expression. From a functional point of view, this observation is supported by our data obtained in prolactinoma cells showing that D2R expression is correlated with NF-κB activity (25). In particular, NF-κB was constitutively activated in cultured cells obtained from one DA-sensitive prolactinoma that express the D2R (23–25), but not in prolactinoma cells that lack this receptor (23–25). Moreover, modulation of NF-κB activity in living cells resulted in changes in D2R expression. In particular, NF-κB activation in D2R-negative prolactinoma cells resulted in D2R expression, whereas inhibition of NF-κB nuclear translocation in D2R-positive prolactinoma cells resulted in D2R loss (25). Moreover, analysis of different brain areas revealed that basal NF-κB binding activity was remarkably and specifically high in the anterior pituitary compared with the striatum that express similar levels of D2R. Taken together, these observations suggest that direct interaction of NF-κB complexes with D2-κB motifs in the D2R promoter is required for the full expression of the D2R gene in pituitary-derived cells and point to NF-κB as a pituitary-specific transcription factor controlling D2R expression.
It has been reported that D2R stimulation up-regulates NF-κB transcriptional activity in transfected cell systems (35–37). This observation, together with our present data, suggests that an autoregulatory control of D2R expression involving NF-κB may be operated by the D2R itself. Ablation of the D2R gene results in the development of pituitary tumors in knockout mice. Even if our data have been obtained in a pituitary tumor cell model and need to be validated in primary human tumors, they raise the possibility that dysregulations in the expression or function of NF-κB transcription factors, resulting in an aberrant control of D2R gene expression in the pituitary, could contribute to the development of prolactin-secreting tumors.
Acknowledgments
We thank Dr. Karen O’Malley for her gift of the human D2R promoter; Dr. Pierre Jalinot for kindly providing the pSG-p50, pSG-p65/RelA, and pSG-c-Rel expression vectors; and Dr. Laura Bianchi for her gift of the pGL3 basic and pGL3 Renilla vectors and for her helpful suggestions.
This work was supported by grants from Consiglio Nazionale delle Ricerche and Ministero dell’Istruzione, Università e Ricerca (MIUR)—Legge 204, 5/06/1998—Grant FISR-Neurobiotecnologie and by MIUR (Grant FIRB RBAU013CNT) to C.M.
Disclosure Statement: The authors have nothing to disclose.
Abbreviations:
- DA
Dopamine
- D2R
D2 receptor
- DTT
dithiothreitol
- NF-κB
nuclear factor-κB
- RAR
retinoic acid receptor
- RXR
retinoid X receptor.
