1,25-Dihydroxyvitamin D3 Suppresses Renin Gene Transcription by Blocking the Activity of the Cyclic AMP Response Element in the Renin Gene Promoter*

We have shown that 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) down-regulates renin expression. To explore the molecular mechanism, we analyzed the mouse Ren-1c gene promoter by luciferase reporter assays. Deletion analysis revealed two DNA fragments from –2725 to –2647 (distal fragment) and from –117 to +6 (proximal fragment) that are sufficient to mediate the repression. Mutation of the cAMP response element (CRE) in the distal fragment blunted forskolin stimulation as well as 1,25(OH)2D3 inhibition of the transcriptional activity, suggesting the involvement of CRE in 1,25(OH)2D3-induced suppression. EMSA revealed that 1,25(OH)2D3 markedly inhibited nuclear protein binding to the CRE in the promoter. ChIP and GST pull-down assays demonstrated that liganded VDR blocked the binding of CREB to the CRE by directly interacting with CREB with the ligand-binding domain, and the VDR-mediated repression can be rescued by CREB, CBP, or p300 overexpression. These data indicate that 1,25(OH)2D3 suppresses renin gene expression at least in part by blocking the formation of CRE-CREB-CBP complex.

ancestral gene (2). Ren-1 and Ren-2 share 97% amino acid identity (3). The Ren-1 protein is the major source of circulating renin and thus believed to be the major regulator of the reninangiotensin cascade.
Renin biosynthesis is regulated by a complex network of regulatory proteins (4). Transgenic studies have demonstrated that the cis-DNA elements required for the tissue-specific and developmental stage-specific expression as well as for response to a variety of physiological stimuli are located within 5 kb of the 5Ј-flanking region of the murine renin gene (5)(6)(7). In the 4.1 kb 5Ј-flanking region of the Ren-1c gene, a proximal minimal promoter (Ϫ117 to ϩ6) and a potent 242-bp enhancer (Ϫ2866 to Ϫ2625) upstream of the transcriptional start site have been shown to be necessary for a high level expression of the renin gene in As4.1 cells (8). Recent studies have revealed the involvement of multiple regulatory proteins in the regulation of renin expression, which include nuclear receptors LXR ␣ and ␤ (9, 10), Ear2 (11), and RAR/RXR complex (12), and transcription factors CREB/CREM and USF1/USF2 (13,14), HOXD10-PBX1b complex and Est-1 (14,15), Sp1/Sp3 (16), and NF-Y (17).
Cyclic AMP (cAMP) is long known to be a major intracellular signal that stimulates renin production in JG cells. It is well established that cAMP signals through cAMP response elements (CRE) located in target gene promoters, which interacts with members of the ATF/CREB/CREM bZIP transcription factor family in homodimeric or heterodimeric forms. Intracellular cAMP is converted from ATP by adenylate cyclase activated by membrane receptors; cAMP binds to the regulatory subunit of protein kinase A (PKA) to free the catalytic subunit, which enters the nucleus and phosphorylates CREB at serine 133 or CREM at serine 117, resulting in the recruitment of ubiquitous co-activators CBP/p300 to promote gene transcription (18 -20). In fact, a number of CREs have been identified in renin gene promoters that play crucial roles in renin gene transcription. For example, in the human renin gene promoter, a CRE identified at Ϫ219 is bound by CREB and is partially responsible for cAMP induction in chorionic cells (21); another study also shows that renin transcription activation by cAMP is partly mediated by a CREB-dependent mechanism via this CRE in Calu-6 cells (22). In the mouse Ren-1d gene promoter, an overlapping cAMP-response element and a negative response element from Ϫ619 to Ϫ588, known as CNRE, is reported to be bound by LXR␣ and respond to cAMP stimulation (10). In the Ren-1c promoter, a CRE at Ϫ2688 is bound by CREB/CREM and is essential for the high level of basal renin expression in As4.1 cells (13). cAMP has been reported to stimulate renin gene transcription in As4.1 cells (23).
Our previous studies have demonstrated that 1,25-dihydroxyvitamin D 3 (1,25(OH) 2 D 3 ), the hormonal form of vitamin D, is a crucial negative regulator of renin biosynthesis (24). The repressive activity of 1,25(OH) 2 D 3 is mediated by the vitamin D receptor (VDR), a member of the nuclear receptor superfamily (25). Mice lacking the VDR or deficient in vitamin D develop hyperreninemia, high blood pressure, cardiac hypertrophy, and other renal abnormalities (24,26,27). Analysis of the Ren-1c gene promoter reveals that 1,25(OH) 2 D 3 negatively regulates renin gene transcription (24); however, the molecular mechanism underlying this transcriptional repression remains unclear. In the present study we have addressed this important question by molecular approaches. We found that the transrepression of renin gene expression by 1,25(OH) 2 D 3 is mediated by the CRE within the enhancer in the renin gene promoter. Our study identifies the cAMP signaling pathway, a major regulatory pathway involved in renin biosynthesis, as the target of vitamin D in this negative regulation.
Luciferase Reporter Assays-Cells were cultured in 24-well plates to 60 -70% confluency and transiently co-transfected with a luciferase reporter plasmid and the pRL-TK plasmid using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). Six hours after transfection, the cells were treated for 24 -36 h with 1,25(OH) 2 D 3 (2 ϫ 10 Ϫ8 M), forskolin (1 M), 8-bromo-cAMP (1 M), H89 (5 M) or in combinations as indicated in each specific experiment. In some experiments, the cells were cotransfected with expression plasmid for CREB, CBP, or p300 as indicated. Luciferase activity was determined using the Dual-Luciferase Reporter assay system (Promega) and a Luminometer TD-20/20 (Turner Designs). Each transfection was carried out in triplicate and repeated at least three times. Luciferase activity was normalized to the Renilla luciferase activity. In some experiments, pGL3-Basic was included in cell co-transfection for calculation of fold induction. All transfection experiments were repeated at least twice.
Electrophoretic Mobility Shift Assays (EMSA)-As4.1-hVDR cells grown in 10-cm dishes were treated with ethanol or 2 ϫ 10 Ϫ8 M 1,25(OH) 2 D 3 for 24 h. Nuclear extracts were obtained by following an established method (28). Briefly, cells were scraped and suspended in ice-cold hypotonic solution (10 mM HEPES, pH 7.9, 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM dithiothreitol, 0.1 M phenylmethylsulfonyl fluoride) for 10 min on ice and homogenized, and the nuclei were collected by centrifugation. The nuclei were suspended in low salt buffer (20 mM HEPES, pH 7.9, 20% glycerol, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol). KCl concentration was then adjusted to 0.3 M by addition of 4 M KCl, and the nuclear suspension was rocked at 4°C for 30 min. The nuclear extract was collected by centrifugation at 25,000 ϫ g. For EMSA, double-stranded oligodeoxynucleotide probes were end-labeled with [␥-32 P]ATP using polynucleotide kinase and purified with spin columns. Probe A is 5Ј-ATGGTGACCTG-GCTGTACTCTGACCTCTG-3Ј (Underlined are the RAR and RXR binding sites); Probe B is 5Ј-AATCCTCCCAATGACAT-CACTAACCA-3Ј (the CRE is underlined). The labeled probe was incubated with nuclear extracts (10 g) at room temperature for 30 min in the presence of 100 g/ml polyd(I:C), and the mixture was electrophoresed on a 4% non-denatured polyacrylamide gel (28). Specificity of protein-DNA interaction was confirmed by competition with excess unlabeled probes. Incubation with anti-CREB-1 or anti-CREM-1 antibody (Santa Cruz Biotechnology, Santa Cruz, CA) was also used to confirm the presence of the corresponding protein in the DNA-protein complex. The gel was dried and exposed to x-ray film for autoradiography.
Chromatin Immunoprecipitation (ChIP) Assays-ChIP assays were performed using a commercial kit from Upstate Biotechnology (Lake Placid, NY) with some modifications. Briefly, As4.1-hVDR cells were treated with ethanol or 2 ϫ 10 Ϫ8 M 1,25(OH) 2 D 3 . After 24 h, the cells were treated with 1% formaldehyde to cross-link the histones and genomic DNA. The cells were lysed and sonicated to shear the chromatin. The sonicated chromatin was incubated with anti-CREB-1, anti-CREM-1, anti-CBP, or anti-VDR antibody (Santa Cruz Biotechnology) or with non-immune IgG as controls. After an overnight incubation, the chromatin-antibody complex was precipitated with protein-A-agarose beads and digested with proteinase K to remove the proteins. The DNA isolated from the complex was subject to PCR amplification using primers flanking the CRE site in the Ren-1c gene promoter: 5Ј-Ϫ2786 GAACTTGTAGGTCCTGCCCG-3Ј (forward) and 5Ј-Ϫ2567 CAAACTATAGATCAGGCAGG-3Ј (reverse), or primers flanking the VDRE in Cyp24a1 gene promoter reported previously (29): 5Ј-GGTTATCTCCGGGGTG-GAGT-3Ј (forward) and 5Ј-AGTGGCCAATGAGCACGC-3Ј (reverse). The PCR products were run on a 2% agarose gel and stained with ethidium bromide.
Northern Blot-Renin mRNA was determined by Northern blot analysis as described previously (24). Briefly, As4.1 cells were treated with different reagents as indicated for 24 h, and total cellular RNAs were extracted using the TRIzol reagent (Invitrogen). The RNAs were separated on agarose gel and transferred onto nylon membranes. The membranes were hybridized with 32 P-labeled renin cDNA probe. The membranes were then stripped and rehybridized with 36B4 cDNA probe for internal loading control. The mRNA transcripts were detected by autoradiography.
GST Pull-down Assays-pGEX-hVDR was constructed by cloning a PCR-generated cDNA sequence encoding the hinge and ligand-binding domain (LBD) of hVDR (amino acids 110 -427) into pGEX-4T-1 (Amersham Biosciences) as reported previously (30). pGEX-4T-1 and pGEX-hVDR were transformed into Escherichia coli BL21 bacteria and GST and GST-LBD-VDR fusion proteins were purified using glutathione-Sepharose 4B beads (Amersham Biosciences) according to the instruction from the manufacturer, and protein purity was determine by SDS-PAGE, and stained with Coomassie Blue. For the assay, HEK293 cells were transfected with pCMV-CREB using Lipofectamine 2000. After cultured for 24 h, the transfected cells were lysed with phosphate-buffered saline containing 0.5% Triton X-100, 10 mM NaF, 2 mM Na 3 VO 4 , and 2 mM phenylmethylsulfonyl fluoride. For binding, GST or GST-LBD-VDR beads were mixed with 500 g of lysates containing a proteinase inhibitor mixture, and the mixture was rotated at 4°C for 2 h in the absence or presence of 10 Ϫ7 M 1,25(OH) 2 D 3 in the binding buffer containing 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA pH 8.0, and 0.5% Nonidet P-40. The beads were precipitated by centrifugation and washed four times with 0.5 ml of binding buffer. The beads were then suspended in Laemmli buffer, and proteins were resolved by SDS-PAGE (31). After being transfered onto an Immobilon-P membrane, CREB was visualized by Western blot assay using anti-CREB antibody.

Two DNA Fragments in the Ren-1c Promoter Mediate Renin
Transrepression by 1,25(OH) 2 D 3 -Our previous studies have shown that 1,25(OH) 2 D 3 suppresses renin gene transcription (24). To further explore the molecular mechanism of this transrepression, we systematically analyzed the 4.1-kb 5Ј-flanking region of the Ren-1c gene by a sequential deletion strategy. The 4.1-kb Ren-1c promoter contains all cis-elements required for tissue-specific and developmental stage-specific expression of the renin gene, as well as for response to a variety of physiological stimuli (5-7), and we showed that its activity was dramatically suppressed by 1,25(OH) 2 D 3 (24) (Fig. 1A, construct A). A series of deletion constructs were generated and used to transfect As4.1-hVDR cells for luciferase reporter assays. As shown in Fig. 1A, all constructs that contain the previously identified 242-bp enhancer region between Ϫ2866 and Ϫ2625 (8) (i.e. constructs A, B, F, G, and H) maintained a high basal activity, and their activity was markedly suppressed by 1,25(OH) 2 D 3 treatment. In contrast, in constructs that lack this enhancer region (i.e. constructs C, D, and E), the basal activity was decreased drastically as expected, and the activity was barely suppressed by 1,25(OH) 2 D 3 (Fig. 1A).
To narrow down the DNA fragments that mediate the vitamin D transrepression, we subjected construct H, the shortest construct in this series of deletion, to further deletion analyses. As shown in Fig. 1B, once again, luciferase constructs (i.e. pBXB and p5-7) lacking the enhancer region had little activity, while the activity of all enhancer-containing constructs was high and suppressed by 1,25(OH) 2 D 3 . These studies revealed that two DNA fragments in the Ren-1c promoter, a 78-bp fragment from Ϫ2725 to Ϫ2647 that contains a CRE at Ϫ2688 (designated as distal fragment, see Fig. 4A for the sequence) and a 121-bp fragment from Ϫ117 to ϩ6 that is the minimal promoter (designated as proximal fragment), together were sufficient to mediate 1,25(OH) 2 D 3 -induced transrepression (see construct p5-6 in Fig. 1B). When the distal fragment was replaced with SV40 enhancer (pSV40E-117) or three copies of the CRE sequence (p3xCRE-117 in Fig. 1C), the basal activity was dramatically reduced as expected; interestingly, however, in contrast to the pSV40E construct that was made of SV40 enhancer and promoter, 1,25(OH) 2 D 3 was able to inhibit the activity of pSV40E-117 and p3xCRE-117 by 19 and 38%, respectively (Fig.  1C). These data suggest that both the minimal promoter and the CRE are crucial for this negative regulation. We mainly focused on the distal fragment in this study. OCTOBER 12, 2007 • VOLUME 282 • NUMBER 41

JOURNAL OF BIOLOGICAL CHEMISTRY 29823
The CRE Is Required for 1,25(OH) 2 D 3 -induced Transrepression-Cyclic AMP is long known to be a potent stimulator of renin biosynthesis. The CRE within the distal fragment is known to play a critical role in basal renin expression in As4.1 cells (13). To test whether vitamin D suppresses renin expression by targeting the CRE, we determined the effect of 1,25(OH) 2 D 3 using the p5-6 reporter construct, which contained the CRE in the distal fragment ( Fig. 2A). As expected, 1,25(OH) 2 D 3 markedly inhibited the promoter activity in As4.1-hVDR cells (Fig. 2B). Because As4.1 cells did not respond to forskolin (Fig. 2B) because of the constitutively active PKA pathway as reported previously (13), we explored the activity using mesangial cells (MCs) and HEK293 cells. In these cells, the promoter activity was dramatically induced by forskolin, and the induction was markedly attenuated by 1,25(OH) 2 D 3 (Fig. 2, C and D). One difference between these cell lines was that 1,25(OH) 2 D 3 always inhibited the promoter activity much more in As4.1 cells (by 70 -80%) than in MCs or HEK293 cells (by 40 -50%). As expected, forskolin induction was blocked by H89, a PKA inhibitor (Fig. 2E). Therefore, the CRE in the distal fragment can be activated by the PKA pathway and suppressed by vitamin D. Similar vitamin D regulation was observed using p4xCRE-Luc reporter, a universal CRE reporter that contains 4 copies of consensus CRE sequence 5Ј-TGACGTCA-3Ј (data not shown).
To validate that the CRE is directly involved in the repression of renin gene induced by 1,25(OH) 2 D 3 , we mutated the CRE site by site-directed mutagenesis in the p4-6 reporter construct to generate p4-6mCRE (Fig. 2F). We chose p4-6 because of its high basal activity (Fig. 1B), so that the activity of the mutant construct would still be high enough for detection. Indeed, mutation of the CRE site in p4-6 reporter reduced the activity by Ͼ80%. Interestingly, the CRE mutation markedly diminished, but did not completely eliminate, the repressive effect of 1,25(OH) 2 D 3 in As4.1 cells. 1,25(OH) 2 D 3 inhibited p4-6 activity by 80% and p4-6mCRE by 30% (Fig. 2G, compare p4-6 and p4-6mCRE), suggesting that both the CRE and the proximal fragment are critical for vitamin D inhibition of renin expression. Because the cAMP-PKA signaling pathway is constitutively active in As4.1 cells (13), we used MCs to confirm the data. As expected, forskolin dramatically stimulated the activity of p4-6, and 1,25(OH) 2 D 3 suppressed the basal as well as forskolin-induced p4-6 activity in MCs (Fig. 2H). Most interestingly, when p4-6mCRE was used to transfect MCs, neither forskolin stimulation nor vitamin D inhibition was observed (Fig.  2I), confirming that the CRE is absolutely required to mediate the suppression of renin transcription by 1,25(OH) 2 D 3 . Unlike As4.1 cells, mutation of the CRE completely eliminated the vitamin D repression in MCs, suggesting the existence of profound differences in renin gene regulation in As4.1 cells and MCs. In fact, renin mRNA expression in MCs is very low and only detectable by RT-PCR (data not shown).
A previous study identified an unconventional RARE within the 242-bp enhancer that is bound by RAR/RXR heterodimer and plays a critical role in renin gene expression (12). This RARE is composed of two TGACCT half-sites separated by 10 nucleotides and is 16-bp downstream of the CRE (see Fig. 4A for details). To address whether this RARE is involved in the vitamin D-induced renin gene transrepression, we mutated the second half-site of the RARE to generate p4-6mRARE reporter (Fig. 2J), and then determined the mutant reporter activity in response to 1,25(OH) 2 D 3 treatment. A similar mutation has been shown previously to disrupt RAR/RXR binding and reduce renin expression in As4.1 cells (12). Interestingly, although the basal activity of p4-6mRARE was reduced by about 75% compared with that of p4-6 reporter as expected, the RARE mutation has no effect on vitamin D-induced suppres- sion of the reporter activity (Fig. 2K). This result indicates that the RARE is unlikely to mediate the transrepression induced by vitamin D.
We further explored the relationship between the vitamin D and the cAMP-PKA signaling pathways in As4.1 cells by measuring renin mRNA expression under a variety of conditions. As shown in Fig. 3, neither forskolin nor 8-bromo-cAMP markedly up-regulated renin mRNA expression; however, when the PKA pathway was blocked by H89, renin mRNA was markedly reduced. Interestingly, combined treatment with 1,25(OH) 2 D 3 and H89 inhibited renin mRNA expression more than treatment with either of these agents alone (Fig. 3A). Similar findings were observed in luciferase reporter assays with p5-6 transfection (Fig. 3B). This synergistic effect suggests that vitamin D likely counters the cAMP-PKA pathway in the regulation of renin gene expression; however, vitamin D is unlikely to directly target PKA.

1,25(OH) 2 D 3 Inhibits Nuclear Protein Binding to CRE Site-
In the 78-bp distal fragment that is required for vitamin D transrepression, a few cis-DNA elements have been identified that are crucial for the basal renin gene transcription, including the CRE and RARE (12, 13) (Fig. 4A). To explore the molecular mechanism underlying the vitamin D-induced transrepression, we used EMSA to investigate the effect of 1,25(OH) 2 D 3 on the interaction between nuclear proteins and cis-DNA elements. For the EMSA, nuclear extracts isolated from ethanol-or 1,25(OH) 2 D 3 -treated As4.1-hVDR cells were incubated with 32 P-labeled probe A or probe B that covered the RARE or CRE, respectively (Fig. 4A). The results showed that 1,25(OH) 2 D 3 had no effects on the protein-DNA interaction on the RAR/ RXR binding sites (Fig. 4B), consistent with the above observation that mutation of the RARE had no effect on renin transrepression by vitamin D. In contrast, 1,25(OH) 2 D 3 markedly reduced the formation of DNA-protein complex on the CRE site (Fig. 4C). The complex was competed off with an excess amount (50ϫ and 100ϫ) of unlabeled probe B, confirming the   OCTOBER 12, 2007 • VOLUME 282 • NUMBER 41

JOURNAL OF BIOLOGICAL CHEMISTRY 29825
specificity of the DNA-protein interaction. Moreover, inclusion of antibody against CREB or CREM in the incubation reaction either reduced the formation of the complex (for anti-CREB antibody, lanes 4 and 6, Fig. 4D) or created a supershift band (for anti-CREM antibody, lanes 5 and 7,  Fig. 4D) in both ethanol-and 1,25(OH) 2 D 3 -treated nuclear extracts, confirming the presence of CREB and CREM in the complex (13). These observations suggest that 1,25(OH) 2 D 3 directly interferes with protein binding to the CRE, leading to a reduction of renin gene transcription.
Liganded VDR Blocks the Formation of CRE-Protein Complex by Directly Binding to CREB-We further utilized ChIP assays to explore the mechanism underlying the EMSA results. The PCR primers for the ChIP assays were designed to flank the CRE site in the distal fragment of the Ren-1c gene (Fig. 5A). As shown in Fig. 5, in the absence of 1,25(OH) 2 D 3 , the chromatin fragment that contained this CRE was precipitated by antibody against CREB, CREM, or CBP, but not by anti-VDR antibody or non-immune IgG (Fig. 5B), indicating that the CRE was bound by CREB, CREM and CBP, but not by VDR, in the basal state to drive renin transcription. These data confirmed that the cAMP-PKA pathway is indeed constitutively active in As4.1 cells. Remarkably, treatment of the As4.1-hVDR cells with 1,25(OH) 2 D 3 completely blocked the binding of CREB, CREM, and CBP to the CRE site (Fig. 5B), indicating that 1,25(OH) 2 D 3 suppressed renin gene transcription, at least in part, by direct inhibition of CRE-mediated transcriptional activity.
Because VDR was not associated with the CRE regardless of the presence or absence of 1,25(OH) 2 D 3 , VDR was unlikely involved in the formation of the protein-DNA complex on the CRE site. As a positive control for VDR, 1,25(OH) 2 D 3 induced VDR binding to the VDRE in the Cyp24a1 gene promoter in  As4.1 cells (Fig. 5C), consistent with our previous observation that 1,25(OH) 2 D 3 induced vitamin D 24-hydroxylase expression in these cells (32). Moreover, 1,25(OH) 2 D 3 treatment did not seem to markedly change the protein level of CREB, CREM, or RXR or the phosphorylation of CREB (p-CREB) in As4.1-hVDR cells (Fig. 5D).
VDR is known to physically interact with a variety of regulatory proteins such as Smad3, ␤-catenin, and p65 NF-B (33)(34)(35)(36). To explore the possibility that VDR interacts with CREB to block CRE complex formation, we generated and purified GST-LBD-VDR fusion protein that contains the LBD of the VDR protein (Fig. 5E). GST and GST-LBD-VDR were then incubated with cell lysates transfected with CREB. As expected, GST did not pull down CREB regardless the presence or absence of 1,25(OH) 2 D 3 ; interestingly, GST-LBD-VDR interacted with CREB and the interaction was markedly enhanced in the presence of 1,25(OH) 2 D 3 (Fig. 5F). This finding suggests that liganded VDR blocks CREB binding to CRE by directly interacting with CREB.
To confirm that 1,25(OH) 2 D 3 functionally blocks CREB binding to CRE, we compared the effect of 1,25(OH) 2 D 3 on CREB and VP16-CREB activity. VP16-CREB consists of the transcriptional activation domain of the herpes simplex protein VP16 fused to the basic/leucine zipper dimerization and DNAbinding domain of CREB (37). As expected, co-transfection of As4.1-hVDR cells with CREB or VP16-CREB expression plasmid led to significant increase in p4-6 luciferase activity, which was suppressed by 68 and 57%, respectively, in the presence of 1,25(OH) 2 D 3 (Fig. 5G). Thus the magnitude of CREB and VP16-CREB inhibition is essentially the same, suggesting that 1,25(OH) 2 D 3 suppresses renin transcription by blocking CREB DNA binding.
1,25(OH) 2 D 3 Suppression Is Attenuated by CREB, CBP, or p300 Overexpression-We further tested whether an excess amount of CREB can overcome the suppression of CREB bind-ing to CRE by co-transfecting As4.1-hVDR cells with p4-6 reporter and CREB, CBP or p300. As shown in Fig. 7, 1,25(OH) 2 D 3 suppressed the reporter activity by about 80% in the absence of co-transfection, and CREB, CBP, or p300 cotransfection overcame the repression dose-dependently. Interestingly, the suppression was completely reversed by low doses (100 ng) of CREB expression plasmid (Fig. 7A); in the cases of CBP and p300, high doses (500 ng) of plasmid were needed to reverse the repression (Fig. 7B). This observation is consistent with the fact that CREB directly binds to the CRE, whereas CBP and p300 rely on CREB for action. These results further support that liganded VDR suppresses renin gene transcription by blocking CREB DNA binding.

DISCUSSION
Our previous studies have demonstrated that 1,25(OH) 2 D 3 negatively regulates renin expression in JG cells (24). Through deletion analysis of the 5Ј-flanking sequence of the Ren-1c gene, we identified two short DNA fragments in the gene promoter that are sufficient to mediate the transcriptional repression. The proximal fragment is the minimal Ren-1c gene promoter, and the distal fragment is within a 242-bp enhancer previously identified to be important for basal renin expression (8). Most recently, this enhancer was shown to be critical for the control of renin expression in vivo, and mice lacking this enhancer have renin depletion in JG cells and develop low blood pressure (39). Therefore, 1,25(OH) 2 D 3 must target important cis-elements within these fragments in the Ren-1c gene promoter to regulate renin. In the present study we focused on the distal enhancer and identified the CRE as a critical element that mediates the transrepression by vitamin D. 1,25(OH) 2 D 3 regulates gene expression by interaction with the VDR. For positive regulation, liganded VDR heterodimerizes with RXR and binds to the vitamin D response element (VDRE) to stimulate gene transcription. A typical VDRE is composed of two hexanucleotide direct repeats separated by three nucleotides (25). On the other hand, 1,25(OH) 2 D 3 also acts as a negative transcription regulator, but the mechanism of the negative regulation is far more complicated and diverse and only partially understood. For instance, in the case of vitamin D suppression of IL-2 and GM-CSF gene expression, liganded VDR-RXR heterodimer or VDR monomer inhibits the formation of NFAT-1/AP-1 transcriptional complex in the gene promoter (40,41). In the promoter of PTH and PTHrP genes, VDR binds to a negative VDRE (nVDRE), with the involvement of Ku antigen, to suppress gene transcription (42). Vitamin D suppresses Cyp27B1 transcription via interaction with VDIR (43). The VDR-RXR heterodimer can recruit nuclear co-repressors, such as NCoR and Alien, to directly mediate transcriptional repression (44). Moreover, liganded VDR can also directly interact with key regulatory factors, such as Smad3, ␤-catenin, and p65 NF-B (33,34,36), to interfere gene expression.
There are no recognizable conventional VDREs within the two DNA fragments in the Ren-1c promoter. Although the distal fragment contains a few hexanucleotide half-sites (i.e. RAR and RXR binding sites) that may bind to VDR, we did not detect VDR binding to these sites by EMSA or ChIP assays. Moreover, neither mutation of the RARE (Fig. 2K) nor deletion of the entire RAR/RXR binding sites (e.g. construct p4-7 does not contain the RAR/RXR sites, Fig. 1B) had any effects on the transrepression induced by 1,25(OH) 2 D 3 . Therefore, the RAR/RXR sites in the distal fragment are unlikely involved in the VDRmediated transrepression.
Our study provides several lines of evidence demonstrating that the CRE within the distal enhancer plays a key role in mediating the vitamin D transrepression. This CRE (5Ј-TGA-CATCA-3Ј) is one nucleotide different from the canonical consensus CRE (5Ј-TGACGTCA-3Ј), is bound by CREB and CREM in As4.1 cells and is crucial for the basal renin expression (13). Although we could not detect stimulation of CRE activity in As4.1 cells because the cAMP-PKA pathway is constitutively active, using HEK293 and MCs we confirmed that this CRE is functional in response to forskolin stimulation. We showed that 1,25(OH) 2 D 3 inhibits CRE-mediated luciferase activity in both the basal and forskolin-stimulated states, and mutation of this CRE not only abolishes forskolin stimulation but also eliminates 1,25(OH) 2 D 3 suppression of the reporter activity. Interestingly, in As4.1 cells, mutation of the CRE dramatically diminishes but not completely eliminates vitamin D repression, whereas in MCs CRE mutation completely eliminates the transrepression. This observation suggests the existence of other cis-element(s) that is (are) also important for the vitamin D action in a JG cell-specific manner. Further studies are needed to identify the cis-DNA element(s).
Our data strongly suggest that 1,25(OH) 2 D 3 suppresses CREmediated transcriptional activity by blocking CREB binding to the CRE (Fig. 8). Because a combined treatment with 1,25(OH) 2 D 3 and PKA inhibitor H89 additively or synergistically inhibits renin expression, it is unlikely that 1,25(OH) 2 D 3 directly targets PKA. Moreover, 1,25(OH) 2 D 3 does not appear to affect the protein levels of CREB and CREM or the phosphorylation of CREB. ChIP assays demonstrate that CREB, CREM, and CBP are binding to the CRE in the basal state in As4.1 cells, which helps to drive a high level of renin expression. In the presence of 1,25(OH) 2 D 3 , liganded VDR binds to CREB, thus blocks CREB binding to the CRE and disrupts the formation of CRE-CREB-CBP/p300 complex, leading to reduction in renin gene expression  Pol II, RNA polymerase II. (Fig. 8). This model is supported by the data obtained from EMSA, ChIP, and GST pull-down assays as well as by VP16-CREB transfection and CREB overexpression experiments. VDR interacts with CREB by the LBD. However, our data do not exclude other regulatory mechanisms. For example, 1,25(OH) 2 D 3 may affect the CRE activity through regulation of other proteins. The role of CREM is unclear. CREM can function as a transcription activator in response to cAMP. CREM phosphorylated at Ser-117 recruits CBP, and unphosphorylated CREM can recruit ACT in transactivation (18). Isoforms of CREM can act as suppressors of cAMP-induced transcription and modulate the final CRE response (45). Specific knockdown of CREB and CREM by RNA interference may help further delineate the role of CREB and CREM in renin regulation. The fact that liganded VDR suppresses renin transcription by interacting with CREB suggests that CREB is a predominant regulator in renin expression. Whether liganded VDR also interacts with CREM remains to be determined. It is noteworthy that a previous study failed to detect CREB-2 (i.e. CRE-BP1 (46)), ATF-1, ATF-2, or ATF-3 binding to this CRE by EMSA (13).
As the rate-limiting enzyme of the renin-angiotensin cascade, renin biosynthesis is tightly regulated. Among the most important physiological factors that influence renin production and release from JG cells are renal perfusion pressure, tubular sodium chloride load and sympathetic nerve activity (47). It is believed that intracellular cAMP is critically involved in the induction of renin expression stimulated by activation of sympathetic nerve activity (mediated by ␤-adrenergic receptor) or by low sodium chloride concentration in renal distal tubules (mediated by the macula densa cells) (48). Therefore, the cAMPdependent PKA signaling pathway is a major regulatory pathway involved in renin production. In the present study, we demonstrated that 1,25(OH) 2 D 3 down-regulates renin gene transcription by suppressing, at least in part, CRE-mediated transcriptional activity in the renin gene promoter. As CRE is activated by cAMP-PKA signaling, our study identifies a major regulatory pathway as the target in vitamin D inhibition of renin synthesis.