Functional Interaction of NF-Y and Sp1 Is Required for Type A Natriuretic Peptide Receptor Gene Transcription*

The vasorelaxant and anti-mitogenic activities of the atrial and brain natriuretic peptides depend upon their binding to the type A natriuretic peptide receptor (NPR-A) expressed on the surface of vascular cells. In-tervention strategies aimed at controlling NPR-A expression are limited by the paucity of studies in this area. Here we identify a sequence CCAAT between 2 141 and 2 137 of the NPR-A promoter that, when mutated, reduces promoter activity by 90% in rat aortic smooth muscle (RASM) cells. Protein/DNA cross-linking and immunoperturbation of electrophoretically shifted complexes formed between RASM nuclear extracts and an oligonucleotide surrounding the CCAAT sequence indicates that the heterotrimeric transcription factor NF-Y binds specifically to the wild-type, but not mutated, CCAAT element. Cotransfection of a dominant negative mutant of the NF-YA subunit results in a concentration-dependent decrease in the activity of the NPR-A promoter in RASM cells confirming that endogenous NF-Y is an activator of the promoter. Mutation of the CCAAT element, in conjunction with mutation of all three Sp1 sites previously shown to be involved in NPR-A promoter regulation, virtually eliminates NPR-A promoter activity in RASM cells. Coexpression of all three NF-Y subunits together with Sp1 in Drosophila cells deficient in these factors indicates that NF-Y and Sp1 act synergistically to reconstitute NPR-A promoter activity. A direct physical association between NF-Y

The natriuretic peptides are a family of vasoactive hormones that play an important role in the regulation of blood pressure and cardiovascular homeostasis (1). Atrial natriuretic peptide and brain natriuretic peptide are both produced predominantly in the heart and circulate in plasma. Their natriuretic, diuretic, and vasorelaxant activities are mediated through the type A natriuretic peptide receptor (NPR-A) 1 (2) (also known as gua-nylyl cyclase A) present on the surface of vascular smooth muscle and other cells. Recent studies by two independent groups (3,4) showed that complete absence of NPR-A in mice leads to hypertension, cardiac hypertrophy, and sudden death, indicating a critical role for NPR-A in the regulation of cardiovascular homeostasis.
The molecular regulation of NPR-A gene transcription is only poorly understood. The rat NPR-A gene has been cloned and sequenced. Sequence analysis identified very few sequence elements for known transcription factors. However, three putative Sp1 consensus binding sites (positioned between Ϫ341 and Ϫ51) and a CCAAT motif (positioned at Ϫ137) were present upstream from the promoter of the NPR-A gene (5). Our earlier studies (6) demonstrated a critical role for the Sp1 family of transcription factors in regulating NPR-A gene transcription in rat aortic smooth muscle (RASM) cells, but the role of the CCAAT sequence remains unknown.
The CCAAT box is present in a number of eukaryotic promoters (7)(8)(9)(10) and has been demonstrated to be important for the transcription of many of those genes (7)(8)(9)(10). In relatively simple, TATA-less promoters, which, like the NPR-A promoter, contain only one or two additional cis-acting elements, the CCAAT box is absolutely required for regulating gene transcription (10 -12). In contrast, the CCAAT box is somewhat less critical for TATA-containing promoters (10 -12). Typically, the CCAAT element is found as a single copy in the forward or reverse orientation immediately upstream of the transcription start site. In the TATA-less NPR-A promoter, the CCAAT sequence is positioned in the reverse orientation between Ϫ137 and Ϫ141 upstream of the transcription start site. In the present study, we show that the CCAAT box is important for NPR-A gene transcription. We have identified the nuclear proteins that associate with the CCAAT box as the heterotrimeric NF-Y complex, and we have demonstrated that functional, and possibly physical, interaction of NF-Y with Sp1 is essential for optimal transcription of the NPR-A promoter in vascular smooth muscle cells.

EXPERIMENTAL PROCEDURES
Materials-Rabbit polyclonal antibodies directed against CCAAT/ enhancer-binding protein (C/EBP) ␣, C/EBP␤, and C/EBP␦, and goat polyclonal antibody directed against Sp1 were a kind gift from S. McKnight (University of Texas, Southwestern Medical Center, Dallas, TX). Mouse monoclonal antibody directed against nuclear factor-Y (NF-Y) A was purchased from PharMingen (San Diego, CA). Rabbit polyclonal antibody directed against NF-YB was a gift from M. Roberto (University of Milan, Milan, Italy). Poly(dI-dC), glutathione-Sepharose 4B, and T7 Sequenase were purchased from Amersham Pharmacia Biotech. Schneider cell medium was obtained from Life Technologies, Inc. All oligonucleotides were synthesized by Cruachem, Inc. Other reagents were obtained through standard commercial suppliers. * This work was supported by Grant HL45637 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ Plasmid Construction and Site-directed Mutagenesis-The construction of Ϫ387 rat NPR-A luciferase has been described previously (6). The NF-YA expression plasmids pNF-YA (wild-type) and pNF-YA29 (dominant negative form) were provided by R. Mantovani (University of Milan, Milan, Italy). pPacSp1 and Copia ␤-galactosidase were obtained from R. Tjian (University of California, Berkeley, CA). pPacNF-YA, pPacNF-YB, and pPacNF-YC were provided by T. F. Osborne (University of California, Irvine, CA). pcDNA3/Sp1 was from K.-S. Chang (University of Texas, M. D. Anderson Cancer Center, Houston, TX). Full-length and mutant GST-Sp1 in a pGex2TKMSC expression vector were provided by E. Wintersberger (Universitä t Wien, Vienna, Austria). pCite-CBF-A, pCite-CBF-B, and pCite-CBF-C were from B. de Crombrugghe (M. D. Anderson Cancer Center). GST-CBF-C (NF-YC) in the pGEX-4T-3 vector was provided by S. N. Maity (M. D. Anderson Cancer Center). GST-CBF-A (NF-YB) and GST-CBF-B (NF-YA) were constructed by PCR amplification of the corresponding coding sequences from pCite-CBF-A and pCite-CBF-B, respectively. The fulllength CBF-A cDNA was generated with oligonucleotides containing BamHI (sense) and XhoI (antisense) restriction sites at their 5Ј termini and the full-length CBF-B cDNA generated with oligonucleotides containing EcoRI (sense) and XhoI (antisense) at their 5Ј termini. PCR products were cut with the appropriate restriction enzymes and inserted into the pGEX-4T-3 vector.
Site-directed mutagenesis of the CCAAT site was carried out with the QuickChange kit from Stratagene (La Jolla, CA). In brief, 10-50 ng of Ϫ387 NPR-A luciferase, 125 ng of two complementary mutagenic primers (sense, 5Ј-GTTAAAGAGTCAGGgcgttTTCCCCCGGCTCTC-3Ј; antisense, 5Ј-GAG-AGCCGGGGGAAaacgcCCTGACTCTTTAAC-3Ј); mutagenized bases are indicated by lowercase letters), 0.2 mM dNTPs, and 2.5 units of Pfu DNA polymerase were mixed in the PCR reaction buffer. PCR was carried out for 18 cycles using 30 s denaturation at 95°C, 1 min annealing at 55°C, and 2 min/kilobase extension at 68°C. After PCR, 1 l of DpnI was added to the reaction to cut the parental DNA template, and 5 l of this digest was used for transformation. Several candidate colonies were screened by sequencing, and positive colonies were chosen for large scale DNA preparation. Mutation of three consensus Sp1 sites in the NPR-A gene promoter has been described previously (6).
Cell Culture-Embryonic RASM cells (passage 19) were kindly provided by H. Ives (University of California, San Francisco, CA). Cells were cultured at 37°C in a 5% CO 2 humidified incubator in Dulbecco's Modified Eagle's-H21 medium containing 10% fetal bovine serum, 100 units/ml penicillin, 100 g/ml streptomycin, and 2% (v/v) broth, tryptose phosphate. Drosophila Schneider cells (SL-2) were obtained from the Cell Culture Facility at the University of California (San Francisco, CA). Cells were cultured in Schneider's medium containing 10% fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin at 25°C.
Transfection, Luciferase, and ␤-Galactosidase Assays-RASM cells were transiently transfected with 10 g of Ϫ387 rat NPR-A luciferase or the relevant promoter mutant and 1-5 g of pNF-YA or pNF-YA29 by electroporation (Gene-Pulser, Bio-Rad) at 250 mV and 960 F. For Drosophila Schneider cells, 5 g of Ϫ387 NPR-A luciferase or the relevant promoter mutant and 2 g of Copia ␤-galactosidase were cotransfected along with increasing amounts (1-10 g) of pPacSp1, and/or plasmids encoding NF-Y subunits (pPacNF-YA, pPacNF-YB, and pPacNF-YC), alone or in combination, by electroporation at 180 mV and 960 microfarads. After transfection, cells were plated on six-well plastic plates and cultured for 48 h. Cells were harvested and lysed in 100 l of cell culture lysis reagent (Promega, WI). Protein concentration of each cell extract was measured using Coomassie protein reagent (Pierce). Cell lysates were processed (20 g of protein/sample) and assayed for luciferase as described previously. Measurements of ␤-galactosidase activity were made using the Galacto-Light Plus kit from Tropix, Inc. (Bedford, MA). Preparation of Nuclear Extracts-Cells were harvested and lysed by the addition of 0.5 ml of lysis buffer (containing 10 mM HEPES, pH 7.9, 1.5 mM MgCl 2 , 10 mM KCl, 0.5% Nonidet P-40, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, and 1 g/ml aprotinin) on ice for 10 min. Lysates were centrifuged; the pelleted nuclei were resuspended in buffer B (containing 20 mM HEPES, pH 7.9, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 25% glycerol, and the above protease inhibitors) and kept on ice for 30 min. Nuclei were centrifuged at 12,000 rpm for 15 min, and the supernatant extracts were saved. Extracts were stored at Ϫ80°C prior to use.
UV Cross-linking Analysis-Purified 32 P-end-labeled, doublestranded oligonucleotide spanning the CCAAT site in the NPR-A promoter or a control oligonucleotide encoding a C/EBP binding site was incubated with 10 g of RASM nuclear extract in binding reaction buffer (25 mM HEPES, pH 7.5, 50 mM KCl, 1 mM dithiothreitol, 10 M ZnSO 4 , 0.2 mg/ml bovine serum albumin, 10% glycerol, and 0.1% Nonidet P-40) containing 0.5 g of poly(dI-dC) at room temperature for 30 min. Each reaction mixture was then pipetted onto parafilm and subjected to irradiation for 10 -20 min at a distance of 5 mm using an ultraviolet lamp with 254-nm emission. Samples were resolved by electrophoresis on 10% denaturing polyacrylamide gel and exposed to an x-ray film for autoradiography.
Electrophoretic Mobility Shift Assay (EMSA)-The CCAAT oligonucleotides used for EMSAs were as follows: wild-type, 5Ј-GTTAAAGAG-TCAGGATTGGTTCCCCCGGCTCTC-3Ј; mutant, 5Ј-GTTAAAGAGTC-AGGgcgttTTCCCCCGGCTCTC-3Ј. Only coding strand sequence is provided; mutagenized bases are identified by lowercase letters, and the CCAAT sequence is underlined. The sequence (coding strand) of the C/EBP binding oligonucleotide is: 5Ј-TAGCTGAGATCTTGCGTAACC-ATTGCCCA-3Ј. Nuclear extracts (10 g) were incubated in binding reaction buffer containing 0.5 g of poly(dI-dC) at room temperature for 10 min. Purified 32 P-end-labeled, double-stranded oligonucleotide was added for an additional 10 min in a total volume of 20 l. For competition experiments, a 1-100-fold molar excess of unlabeled doublestranded oligonucleotide was added to the binding reaction. For immunoperturbation experiments, nuclear extracts were incubated on ice for 1 h with 2 g of polyclonal antibody against C/EBP␣, C/EBP␤, C/EBP␦, NF-YA, NF-YB, DBP, TEF, Sp1, or Sp3 prior to the addition of labeled probe. Independent studies with the anti-C/EBP␣ (23) and C/EBP␤ (data not shown) antibodies demonstrated that each was capable of disrupting (C/EBP␣) or supershifting (C/EBP␤) the relevant DNA-protein complex in the mobility shift assay. All samples were resolved on 5% nondenaturing polyacrylamide gels. Gels were dried and exposed to film for autoradiography.
GST Pull-down Assay-35 S-Labeled Sp1 protein was synthesized in vitro using the TnT SP6 quick-coupled transcription/translation system from Promega (Madison, WI) according to the manufacturer's instructions. GST fusion protein expression vectors including pGEX-CBF-A, pGEX-CBF-B, pGEX-CBF-C, and pGEX-Sp1 were transformed into the BL-21 strain of Escherichia coli (Stratagene, La Jolla, CA), expanded in suspension culture and induced (3 h) with 1 mM isopropyl ␤-D-thiogalactopyranoside. Cells were pelleted, sonicated in TST buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 0.05% Tween 20, and centrifuged. The resultant supernatant was then added to 300 l of glutathione-Sepharose beads, mixed on a rotating wheel at 4°C for 1 h, and centrifuged. The pellet containing the bound GST fusion protein was washed three times with TST buffer, then resuspended in 300 l of protein binding buffer containing 20 mM HEPES (pH 7.9), 150 mM KCl, 25 mM MgCl 2 , 10% glycerol, 1 mM dithiothreitol, 0.1% Triton X-100, and 0.1% Nonidet P-40. Bound protein was quantitated using the Coomassie protein assay reagent (Pierce).
Ten g of each GST-CBF subunit bound to glutathione-Sepharose beads was incubated with 4 l of in vitro translated 35 S-Sp1 in 150 l of protein binding buffer at 4°C for 1 h. The reaction contents were then precipitated by centrifugation. The precipitate was washed three times with 500 l of protein binding buffer, resuspended in 15 l of SDS sample buffer, and loaded on a 10% denaturing polyacrylamide gel. The gel was dried and exposed to x-ray film prior to autoradiography.
Coimmunoprecipitation-One mg of RASM nuclear extract was mixed with 2 g of either anti-Sp1 antibody or anti-Rel A (p65) antibody in 200 l of protein binding buffer. As a control, 200 g of RASM nuclear extract was mixed with either 10 g of GST alone or GST-Sp1 bound to glutathione-Sepharose beads. After incubation at 4°C for 2 h, the reaction mixtures were pelleted, and the precipitates were washed three times with protein binding buffer, resuspended, and boiled with SDS sample buffer. The protein was resolved on 10% SDS-PAGE gel, transferred to nitrocellulose membrane, and immunoblotted with anti-NF-YA antibody. The immunoprecipitated protein signal was detected using the ECL Western blot detection system (Amersham Pharmacia Biotech).
Statistic Analysis-Data were evaluated by one-way analysis of variance using Newman-Keul's test for significance.

RESULTS
Three Sp1 consensus sites and one CCAAT site have been identified by DNA sequence analysis in the proximal 5Ј-flanking sequence of NPR-A gene (5,6). All of these regulatory elements are present within a segment of the NPR-A gene, extending from 387 base pairs upstream to the transcription start site, which we showed previously to direct the optimal level of NPR-A promoter activity in RASM cells (6). The relative location and specific mutations introduced at each of these sites are presented in Fig. 1A. To determine the relative contribution of each of these putative regulatory elements to NPR-A gene transcription, mutations were introduced into each site within the context of the Ϫ387 NPR-A luciferase reporter and transfected into RASM cells. Our previous studies showed that both Sp1 and Sp3 bind to each of the three Sp1 consensus sites in the NPR-A promoter and that mutation of all three Sp1 sites in concert reduced Ϫ387 NPR-A luciferase reporter activity in RASM cells to 10% that of the native promoter. In the present study, we showed that mutation of the CCAAT site resulted in a similar ϳ90% reduction in activity whereas mutation of the CCAAT and Sp1 sites in combination virtually eliminated NPR-A promoter activity (Fig. 1B). Thus, a CCAAT-binding transcription factor is a dominant activator of the NPR-A promoter.
A number of transcriptional activators that bind to the CCAAT motif have been described. One or more of these may participate in CCAAT-dependent, NPR-A promoter activity. These factors include C/EBP (13)(14)(15)(16), NF-Y (17)(18)(19), mouse y-box protein 1 (MSY-1) (20), and CCAAT binding transcription factor/nuclear factor-1 (CTF/NF-1) (19,21). We performed UV cross-linking and electrophoretic mobility shift assays (EMSA) to identify the protein(s) that interact with the CCAAT motif in the proximal NPR-A promoter. UV cross-linking analysis showed at least two protein complexes in RASM nuclear extracts, with molecular mass values of ϳ117 and ϳ78 kDa, that associated with a labeled oligonucleotide spanning the CCAAT motif in the NPR-A promoter, but not with an otherwise identical oligonucleotide containing the mutations that block NPR-A promoter activity (Fig. 2). We also cross-linked the same RASM extracts with an oligonucleotide that binds to FIG. 2. UV cross-link analysis of proteins present in RASM nuclear extracts that bind to the CCAAT site in the rat NPR-A promoter. Ten micrograms of RASM nuclear extract was incubated with 32 P-labeled, double-stranded CCAAT or CCAAT mutant oligonucleotide (see "Experimental Procedures" for description) or a control 32 P-labeled, double-stranded oligonucleotide encoding a known C/EBP binding site. The reaction mixture was irradiated by UV at 254 nm for 10 -20 min, then separated on a 10% denaturing polyacrylamide gel and exposed to x-ray film.
C/EBPs in rat liver tissue and pituitary progenitor GHFT1-5 cells (22,23). Three nuclear proteins of ϳ42, ϳ48, and ϳ52 kDa were identified. Although a low intensity ϳ42-kDa band was seen in the RASM extracts, the ϳ48and ϳ52-kDa bands were clearly distinguishable from those interacting with the NPR-A CCAAT sequence (Fig. 2). This suggests that C/EBP proteins are present in the RASM nuclear extract but, for the most part, do not bind to the CCAAT motif in the NPR-A promoter.
EMSA of DNA-binding proteins present in RASM extracts was carried out. A single slowly migrating band was observed which was effectively competed by unlabeled wild-type (WT) oligonucleotide but not by its mutated counterpart (Fig. 3A).

FIG. 3. EMSA analysis of the interaction of RASM nuclear extracts with the CCAAT site in the rat NPR-A promoter.
Ten g of RASM nuclear extract was incubated with 32 P-labeled, double-stranded NPR-A CCAAT oligonucleotide or 32 P-labeled, double-stranded C/EBP-binding oligonucleotide and subjected to EMSA. Panel A, competition of RASM nuclear protein interaction with 32 P-labeled CCAAT probe by increasing concentrations (1-100-fold excess) of unlabeled double-stranded NPR-A CCAAT oligonucleotide. Panel B, competition of RASM nuclear protein interaction with 32 P-labeled NPR-A CCAAT probe by increasing concentrations (1-100-fold excess) of unlabeled, double-stranded C/EBP-binding oligonucleotide. Panel C, identification of CCAAT binding complexes by EMSA. Ten g of RASM nuclear extracts was preincubated on ice for 1 h with 2 g of antibody directed against C/EBP␣, C/EBP␤, C/EBP␦, NF-YA, NF-YB, DBP, TEF, Sp1, Sp3, or with preimmune serum before the addition of labeled NPR-A CCAAT probe. The position of the CCAAT oligonucleotide/NF-Y complex is indicated by the arrow.
EMSA with the C/EBP-specific oligonucleotide demonstrated a faster-migrating protein complex supporting our contention (see above) that C/EBP does not participate in formation of the complex identified on the NPR-A CCAAT element (Fig. 3B). This conclusion drew further support from the failure of the C/EBP oligonucleotide to compete with the native complex on the NPR-A promoter fragment (Fig. 3B). The 117-kDa size of the UV cross-linked protein is similar to the size predicted for the CCAAT-binding, heterotrimeric NF-Y complex, raising the possibility that this transcription factor is involved in regulation of the NPR-A gene. To test this hypothesis, we employed specific antibodies directed against subunits of the NF-Y protein complex. Immunoperturbation studies showed that the slowly migrating band in the EMSA was supershifted or disrupted by antibody directed against the A or B subunits of NF-Y, but was not recognized by antibody directed against C/EBP␣, C/EBP␤, C/EBP␦, DBP, or TEF (Fig. 3C). Collectively, these findings support the hypothesis that the RASM nuclear protein that binds to the CCAAT box of the NPR-A promoter is NF-Y.
NF-Y (also called CBF or CP1) is a ubiquitous transcription factor that binds to CCAAT motifs in the proximal promoters of a large number of mammalian genes (7)(8)(9)(10)(11)(12)(17)(18)(19). NF-Y/CBF consists of three subunits, A, B, and C, all of which are required for DNA binding. To confirm that the CCAAT box is essential for NPR-A gene expression, we examined the effect of forced overexpression of NF-Y or an NF-Y dominant negative mutant on NPR-A promoter activity. Overexpression of wild-type NF-YA failed to activate the NPR-A luciferase reporter (Fig. 4); however, cotransfection of a dominant negative mutant of NF-YA (pNF-YA29) resulted in a dose-dependent reduction of Ϫ387 NPR-A promoter activity in RASM cells (Fig. 4). The mutant NPR-A reporter lacking the CCAAT element was unaffected by pNF-YA29. The failure of NF-YA to activate the NPR-A promoter might reflect the fact that the ubiquitously expressed NF-Y protein, or more specifically NF-YA, is not limiting for NPR-A gene transcription in RASM cells. To confirm that NF-Y is an activator of the NPR-A promoter, we introduced the NPR-A luciferase reporter into Schneider cells, a Drosophila cell line that does not express appreciable levels of a number of mammalian transcription factors, including Sp1 and NF-Y. As shown in Fig. 5A, overexpression of the three NF-Y subunits (A, B, and C) led to a concentration-dependent increase in NPR-A-driven reporter activity in Schneider cells.
Cooperative interactions among transcription factors have been shown to be important for the regulation of a number of gene promoters. Indeed, simultaneous mutation of all of the Sp1 sites and the CCAAT motif led to a reduction in NPR-A promoter activity to background levels (Fig. 1B). To determine whether Sp1 cooperates with NF-Y at the NPR-A promoter, Ϫ387 NPR-A luciferase was cotransfected into Schneider cells along with Drosophila expression vectors encoding Sp1 and the three NF-Y subunits. Whereas expression of either Sp1 or NF-Y activated the NPR-A promoter to a moderate degree, simultaneous expression of both Sp1 and NF-Y together resulted in a dramatic increase in promoter activity (Fig. 5A). This increase exceeded that seen with either transcription factor alone, implying that Sp1 and NF-Y act cooperatively to drive NPR-A gene transcription.
Transcription factors may cooperate at a promoter by separately contacting different rate-limiting targets, by mutually contacting a single rate-limiting target or by interacting with each other to form a novel activity. We performed a GST pulldown assay to probe the potential interaction of Sp1 and NF-Y in vitro. As shown in Fig. 5B, [ 35 S]methionine-labeled Sp1 was able to bind GST-NF-YA and NF-YC, but not GST alone or NF-YB, suggesting that Sp1 interacts physically with the NF-YA and NF-YC subunits. For the converse experiment, nuclear extracts from RASM cells were incubated with GST-Sp1 or GST alone, and the washed beads were electrophoresed, blotted onto filters, and then probed with anti-NF-YA antibody. This demonstrated that endogenous NF-YA in RASM cells also interacted with Sp1 (Fig. 5C). Finally, we conducted coimmunoprecipitation studies to show that Sp1 and NF-YA interact in the context of the intact cell. Immunoprecipitation of Sp1 from the same extracts followed by Western blot analysis for NF-YA revealed a low intensity but specific band, which migrated in a position the size of NF-YA (Fig. 5C). As a control, NF-YA was not coimmunoprecipitated when anti-Rel A antibody was used in place of anti-Sp1 antibody (Fig. 5C). Thus, NF-Y and Sp1 interact both in vitro and in the context of the intact cell. Finally, we asked whether this interaction could be demonstrated to have functional sequelae in the cell. To address this, we transfected NPR-A promoter-driven luciferase constructs, with mutations at either the Sp1 or NF-Y binding sites, into Drosophila Schneider cells, alone or in combination with Sp1 or NF-Y expression vectors. As shown in Fig. 5D, the CCAAT mutant promoter displayed a robust response to cotransfected Sp1 but no response to NF-Y. However, when Sp1 and NF-Y were cotransfected together, there was a small but statistically significant increment in promoter activity. We interpret this increment as reflective of the ability of Sp1 to recruit NF-Y into the regulatory complex through protein-protein interaction. Similar findings were observed with the triple Sp1 mutant promoter. Cotransfection with NF-Y effected a significant increase in promoter activity while Sp1 was completely ineffective. Once again, overexpression of NF-Y and Sp1 together led to an increment in promoter activity that exceeded that seen with either transcription factor alone. DISCUSSION In this study, we demonstrated that a CCAAT box in the proximal 5Ј flanking sequence of the NPR-A gene is essential for transcriptional activity. Mutation of the CCAAT motif spanning Ϫ142 to Ϫ138 relative to the transcriptional start site resulted in 90% reduction in NPR-A promoter activity. A number of nuclear transcription factors have been shown to bind to the CCAAT box, including C/EBP, NF-Y, MSY-1, and CTF/ NF-1 (13)(14)(15)(16)(17)(18)(19)(20)(21). Only NF-Y requires all five base pairs for binding (7). UV cross-linking, EMSA competition, and immunoperturbation studies demonstrated that the trans-acting factor NF-Y, but not C/EBP, bound to the CCAAT site in the proximal NPR-A promoter. Moreover, forced expression of a dominant negative mutant of NF-YA resulted in a dramatic reduction in NPR-A promoter activity. Together, these data suggest that NF-Y, when bound to the CCAAT element, plays a critical role in the regulation of NPR-A gene transcription.
NF-Y is a ubiquitous heterotrimeric transcription factor, also referred as CP1 or CBF, that consists of NF-YA, NF-YB, and NF-YC subunits with molecular mass values of 42, 36, and 40 kDa, respectively (24 -27). NF-YB and NF-YC contain a conserved histone-fold motif, which forms a dimer that interacts with NF-YA (26,27). All three NF-Y subunits are required for binding to the CCAAT motif (26,27). Our cross-linking studies demonstrated that at least two complexes with molecular masses of 78 and 117 kDa specifically cross-linked to the CCAAT sequence of the NPR-A promoter. These are very close to the predicted molecular mass of NF-YB/NF-YC dimers and the heterotrimeric complex of all three NF-Y subunits, respec-

FIG. 5. Cooperative interaction of NF-Y and Sp1 is required for NPR-A gene transcription.
Panel A, overexpression of NF-Y and Sp1 synergistically activates NPR-A promoter activity. Drosophila Schneider cells were transiently cotransfected with 5 g of Ϫ387 NPR-A-LUC, 2 g of Copia ␤-galactosidase, increasing concentrations (1, 5, or 10 g) of pPacSp1, and a mixture of equal amounts of pPacNF-YA, pPacNF-YB, and pPacNF-YC. After 48 h of culture, cells were lysed and assayed for luciferase and ␤-galactosidase activity. Luciferase measurements were normalized for ␤-galactosidase activity within a given sample. The data represent the mean Ϯ S.D. for five experiments performed in duplicate or triplicate. *, p Ͻ 0.01 versus control, Sp1, or NF-Y alone. Panel B, in vitro interaction of Sp1 and NF-Y. Ten g of of each GST-NF-Y subunit bound to glutathione-Sepharose beads was incubated with 4 l of in vitro translated 35 S-Sp1 in 150 l of protein binding buffer at 4°C for 1 h. Reactions were then precipitated by centrifugation. The precipitate was washed, resuspended in SDS sample buffer, and resolved on a 10% denaturing polyacrylamide gel. The gel was dried and exposed to film for autoradiography. The result shown is representative of two independent experiments. Panel C, NF-Y associates with Sp1 in vivo. One mg of RASM nuclear extract was incubated with anti-Sp1 or anti-Rel A antibody in protein binding buffer at 4°C for 2 h. As a control, 200 g of RASM nuclear extract was incubated with 10 g of either GST alone or GST-Sp1 bound to glutathione-Sepharose beads in protein binding buffer at 4°C for 2 h. The reactions were pelleted, and the precipitates were washed, resolved on 10% SDS-PAGE gel, transferred to the nitrocellulose membrane, and immunoblotted with anti-NF-YA antibody. The protein signal was detected using a Western blotting kit. The result presented is representative of three independent experiments. Panel D, Drosophila Schneider cells were transiently cotransfected with 5 g of Ϫ387 NPR-A CCAAT mutant or the triple Sp1 mutant, 5 g of pPacSp1, and a mixture of equal amounts of pPacNF-YA, pPacNF-YB, and pPacNF-YC. After 48 h of culture, cells were lysed and assayed for luciferase activity. The data represent the mean Ϯ S.D. from four experiments. #, p Ͻ 0.05 versus Sp1 (CCAAT mutant); *, p Ͻ 0.01 versus NF-Y (triple Sp1 mutant). tively. Functional analysis has shown that NF-Y is crucial for transcriptional activation and reinitiation on genes like NPR-A, which lack a TATA box (10 -12, 28). Interestingly, NF-Y has been shown to recognize CCAAT boxes in several genes involved in growth regulation, including murine ribonucleotide reductase R2 (29), murine E2F-1 (30), cyclin B1 (31), cdc2 (32), cyclin A (32), the protein phosphatase cdc25C (32), and human thymidine kinase (33). Recent studies also demonstrated that stable expression of a dominant negative mutant of NF-Y in mouse fibroblast cells resulted in retardation of cell growth and inhibition of transcription of various cellular genes (34), implying that NF-Y may be crucial for cell cycle progression. The liganded NPR-A has been demonstrated to function in a growth suppressant mode in vascular smooth muscle, endothelial, and cardiac fibroblast cells in culture (35)(36)(37), implying that transcriptional control by NF-Y may be more than fortuitous.
Our earlier studies documented the importance of Sp1 to NPR-A gene transcription (6). We have shown here that NF-Y does not work in isolation, but rather operates in conjunction with Sp1 to regulate NPR-A gene transcription. Mutation of the CCAAT and Sp1 sites in combination completely eliminated NPR-A promoter activity, whereas overexpression of both Sp1 and NF-Y in Drosophila cells led to amplification of NPR-A promoter activity beyond that seen with either transcription factor alone. Furthermore, we demonstrated a physical interaction between these two transcription factors in vitro and in vivo that may play a role in their cooperative functional activity. Gel shift studies in which both Sp1 and NF-Y were examined, either alone or in combination, did not suggest that their functional cooperativity was due to cooperative binding of these factors to the NPR-A promoter, at least in vitro (data not shown). However, Wright et al. (38) have demonstrated that the half-life of either NF-Y or Sp1 binding is dramatically increased when both transcription factors are bound to the proximal promoter of the major histocompatibility complex class II-associated invariant chain (Ii) gene. We cannot exclude the possibility that similar interactions between Sp1 and NF-Y might stabilize their binding on the NPR-A promoter. In addition, Sp1 and NF-Y can bind individually to separate loci on the p300 coactivator (39 -41). It is possible that the functional interaction of NF-Y and Sp1 could involve synergistic regulation of this important transcriptional coactivator.
Taken together, the studies presented here demonstrate that the bulk of NPR-A gene expression in RASM cells depends upon functional, and possibly physical, interactions between two transcription factors, NF-Y and Sp1. Since regulation of blood pressure depends, in part, upon the correct expression of NPR-A in relevant target tissues, the intricate molecular processes controlling NF-Y and Sp1 expression and activity are likely to be important components in the regulatory machinery governing cardiovascular homeostasis at the organismal level.