Regulation of renin expression by the orphan nuclear receptors Nr2f2 and Nr2f6

Abstract

Understanding the transcriptional mechanisms of renin expression is key to understanding the regulation of the renin-angiotensin system. We previously identified the nuclear receptors RAR/RXR and Nr2f6 (EAR2) as positive and negative transcriptional regulators of renin expression, respectively (Liu X, Huang X, Sigmund CD. Circ Res 92: 1033–1040, 2003). Both mediate their effects through a hormone response element (HRE) within the renin enhancer. Here, we determined whether another nuclear receptor, Nr2f2 (Coup-TFII, Arp-1), identified in a screen of proteins that bind the HRE, also regulates renin expression. Luciferase assays indicate that Nr2f2 negatively regulates the renin promoter more potently than Nr2f6. Gel-shift and chromatin immunoprecipitation (ChIP) indicate that Nr2f2 and Nr2f6 can bind directly to the renin enhancer through the HRE. Surprisingly, baseline expression of endogenous renin was not effected when Nr2f2 was knocked down in As4.1 cells, whereas knockdown of Nr2f6 increased renin expression twofold. Interestingly, however, knockdown of Nr2f2 augmented the induction of renin expression caused by retinoic acid. These data indicate that both Nr2f6 and Nr2f2 can negatively regulate the renin promoter, under baseline conditions and in response to physiological queues, respectively. Therefore, Nr2f2 may require an initiating signal that results in a change at the chromatin level or activation of another transcription factor to exert its effects. We conclude that both Nr2f2 and Nr2f6 negatively regulate renin promoter activity, but may do so by divergent mechanisms.

because of the importance of renin in the rate-limiting step of the RAS, its transcriptional regulation has been the target of many studies. Evidence from experiments using As4.1 cells has identified two regions within the Ren-1^c gene that are critical for its controlled expression. A proximal promoter element (PPE; −197 to −50 bp) and enhancer (−2866 to −2625 bp) work in conjunction to direct renin expression. Inclusion of the enhancer in reporter constructs results in a ∼50-fold increase in promoter activity in an orientation- and position-independent manner (29). The mouse renin (mREN) enhancer shares 71% homology with a human renin (hREN) enhancer that lies ∼11 kb upstream of the transcription start site. There are three binding sites at the 3′-end of the enhancer that share 100% homology between the mouse and human genome. Mutation of any one of these sites results in a dramatic decrease in enhancer activity (27, 31). The three sites consist of a cAMP response element (CRE), an E-box, and one half-site of a hormone response element (HRE) (32). The third site is of particular interest because it is part of a TGACCT direct repeat that makes up a larger HRE which partially accounts for the increased enhancer activity observed in the mouse (15). The HRE can bind the retinoic acid receptor (RAR), retinoid X receptor (RXR), Nr2f6 (EAR2), peroxisome proliferator-activated receptor (PPARγ), and vitamin D receptor (VDR) (8, 22, 23, 31). However, VDR has been reported to exert its effects indirectly by binding CREB and inhibiting its transactivation through the CRE (38). We have identified the HRE to be of critical importance for enhancer activity, yet all of the factors that can bind and regulate renin expression have yet to be identified (23, 31).

Our group and others have extended in vitro analyses of the renin enhancer in vivo using transgenic and knockout models. Elimination of the enhancer in a human renin gene transgenic mouse resulted in a significant decrease in baseline renin expression (39). However, this did not influence cell- or tissue-specific expression and physiological responses to angiotensin II. In a BAC transgenic mouse in which the renin gene was replaced with the green fluorescent protein gene, deletion of the 3′-end of the enhancer including the three conserved elements led to a decrease in green fluorescent protein (GFP) expression as measured by ELISA (11). These results were confirmed in a mouse model where the enhancer at the endogenous locus was deleted (24). Interestingly, the latter two models exhibited muted responses to stimuli that normally induce renin expression, suggesting that the enhancer may play a role in modulating the regulation of renin in response to physiological cues.

Despite the identification of at least four nuclear receptors that can bind to the HRE and regulate renin expression, none of them have the large impact on renin expression that would be expected based on luciferase assays comparing wild-type (WT) and mutant constructs lacking the HRE. Additionally, As4.1 cell nuclear extracts result in four HRE shift complexes in electrophoretic mobility shift assays that cannot be entirely accounted for by RAR, RXR, and Nr2f6 (23, 31). Using the same yeast one-hybrid screen that identified Nr2f6, we identified another orphan nuclear receptor Nr2f2 (Arp-1, Coup-TFII). This receptor is a member of the same group of nuclear receptor subfamily 2 and is thus closely related to Nr2f6. Because Nr2f6 is a negative regulator of the renin promoter, we hypothesized that Nr2f2 would act in a similar manner.

METHODS

Luciferase Assays

As4.1 cells (ATCC CRL2193) were split and transfected the following day using Fugene 6 (Roche). One microgram of m4.1kP-luc (30) and 2 ng phRL-TK (Promega) were included in all transfections. Each well of a six-well plate was transfected with 1 μg cDNA or short hairpin (sh) RNA expression plasmid corresponding to the specific nuclear receptor. Forty-eight hours posttransfection, cells were lysed and extracts were analyzed using Promega's Dual-Luciferase Assay System.

Orphan Nuclear Receptor Knockdown

Adenoviruses expressing the same shRNAs to GFP, Nr2f2, or Nr2f6 used in luciferase experiments were constructed and tested as previously described (9, 37). Twenty-four hours after being split into six-well dishes, As4.1 cells (60% confluent) were infected using an MOI of 100. Adenovirus and polybrene (5 μg/ml; Millipore) were mixed in serum-free DMEM and added to duplicate wells for each shRNA. After a 6-h incubation, cells were washed and complete DMEM (10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin) was added. Forty-eight hours later, total RNA was isolated using a kit (Purelink RNA Mini Kit, Invitrogen), and protein was extracted using RIPA buffer (50 mM Tris·Cl, pH 7.4, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS).

After RNA isolation, 1 μg was reverse transcribed using 200 U of Superscript III (Invitrogen) in a 20-μl reaction. Reactions were incubated at 50°C for 30 min, 55°C for 15 min, 60°C for 15 min, and 70°C for 15 min to inactivate the reaction. The cDNA was diluted 1:20, and gene expression was measured using Taqman Gene Expression Master Mix (Applied Biosystems) and probes. Nr2f2 and Nr2f6 probes were from Applied Biosystems as listed on the Nuclear Receptor Signaling Atlas website (www.nursa.org; Table 1). The renin (Mm02342888_gH) and β-actin probe (4352933E) were from Applied Biosystems. Data were analyzed using the 2^−ΔΔCt method to calculate fold-changes relative to shGFP samples. Assay PCR efficiency was determined to be 90–105% using a 7-log serial dilution series of the cDNA samples.

Table 1.

Probes and primers

Taqman Probes
Gene	Forward	Reverse	Probe
Nr2f2	GCATGAGACGGGAAGCTGTAC	CGTTGGTCAGGGCAAACTG	AGGCATCCTGCCTCT
Nr2f6	GAGGGCTGCAAGAGTTTCTTC	TCCGGTGGTGCTGATCAA	CCGGTCCAACCGTGAC
EMSA probes
WT	GATCTGGTGACCTGGCTGTACTCTGACCTCTCAGAT
μb	GATCTGGTGACCTGGCTGTACTCTTTCCTCTCAGAT
μc	GATCTGGTTTCCTGGCTGTACTCTGACCTCTCAGAT
μbc	GATCTGGTTTCCTGGCTGTACTCTTTCCTCTCAGAT
DAPA probes
WT	/5BioTEG/CAAAACTGCAGGATGGTGACCTGGCTGTACTCTGACCTCTCAGAT
μbc	/5BioTEG/CAAAACTGCAGGATGGTTTCCTGGCTGTACTCTTTCCTCTCAGAT

ChIP Primers
Target	Forward	Reverse
HRE	TTGGACCCTCTCCATTCCTTCACG	ATGCGCTATCACAACCAGCCACTC
NG	ACAGAAGGAGGTCGGAAGAC	ACAGAAGGAGGTCGGAAGAC
SOD1	ACATCCTAGCATCCTGTCTTAAC	GCCCTGCTACACTCTCCTAC

WT, wild-type; DAPA, DNA affinity precipitation assay; ChIP, chromatin immunoprecipitation; HRE, hormone response element.

Protein extracts were quantified, and 10 μg of each were mixed with SDS sample buffer. Samples were loaded and run on 10% SDS-PAGE gels, transferred to polyvinylidene difluoride (Millipore), and probed for renin (sc-22671, Santa Cruz Biotechnology), Nr2f2 (PP-H7147–00, Perseus Proteomics), and actin (ab1801, Abcam). Horseradish peroxidase-conjugated secondary antibodies included the ECL anti-mouse IgG (NA931, GE Healthcare) for Nr2f2 blots and the ECL anti-Rabbit IgG (NA-934, GE Healthcare) for actin blots.

All-trans retinoic acid (RA) treatment (10 μM; Sigma) or vehicle (DMSO) was added to As4.1 cells cultured in DMEM with 10% charcoal treated FBS 24 h after adenovirus infection. Cells were treated for 20 h, and fresh media plus RA or vehicle was added a second time and incubated for an additional 4 h. Following incubation, total RNA was extracted and RT-qPCR was performed as described above. Data was analyzed using the 2^−ΔΔCt method to calculate fold-changes relative to vehicle-treated samples for each shRNA.

EMSA and Supershift Assay

EMSAs were carried out using double-stranded DNA probes corresponding to the HRE designed with 5′-GATC overhangs and labeled using [α-³²P]dATP (Table 1). In vitro translated proteins were generated using the TNT Quick Coupled Transcription/Translation System (Promega). Parallel reactions to assess protein production were run in which proteins were labeled using [³⁵S]methionine.

Probes were incubated at room temperature for 30 min with 1 μl of unlabeled in vitro translated protein or 6 μg of As4.1 nuclear extract in Tris binding buffer (10 mM Tris·Cl, pH 7.4, 1 mM EDTA, pH 8.0, 60 mM KCl, 10 mM DTT, 0.1% Triton X-100, 4% glycerol) with 1 μg poly[d(I-C)]. Binding reactions were loaded onto 5% native polyacrylamide gels and run for 2 h in 0.5× TBE. Gels were dried, exposed to phospho-screens overnight, and scanned using a Molecular Dynamics Storm 840 phosphoimager. Supershift analysis was performed by adding 1 μg of the appropriate antibody after the initial incubation period for 15 min on ice before electrophoresis.

DNA Affinity Purification Assay

DNA affinity purification assays were carried out with slight modifications as described by Butter et al. (5) using two biotin-TEG 5′-labeled double-stranded DNA probes (Table 1). Nuclear extracts from As4.1 cells (40 μg) were mixed with 80 pmol of double-stranded probe in the same binding buffer as that used in EMSAs with protease and phosphatase inhibitors (Roche), plus 4 μg poly[d(I-C)] (Roche) for a total binding reaction of 40 μl. Nuclear extract and probe were incubated on ice for 30 min followed by addition of 50 μl of streptavidin-conjugated Dynabeads MyOne C1 (Invitrogen). Following 90-min incubation at 4°C while rotating, beads were collected using a DynaMag-2 magnet (Invitrogen) and washed three times with binding buffer. Beads were subsequently boiled, collected, and the extracts were loaded onto a 10% SDS-PAGE gel. Western blots were probed for Nr2f2 and Nr2f6 (ab65012, Abcam).

Chromatin Immunoprecipitation

As4.1 cells in a 15-cm dish were fixed for 8 min with 1% formaldehyde and quenched with 0.125 M glycine. Subsequently, cells were washed twice with PBS, collected by scraping and centrifugation, then lysed with 3 ml of lysis buffer (0.15 M NaCl, 0.01 M HEPES, pH 7.4, 0.0015 M MgCl₂, 0.01 M KCl, 0.5% NP-40, 0.0005 M DTT). Nuclei were then collected and resuspended in nuclear lysis buffer (0.05 M Tris, pH 8.0, 0.01 M EDTA, 1% SDS). Nuclei were diluted with 2 vol of chromatin immunoprecipitation (ChIP) dilution buffer (0.15 M NaCl, 0.0167 M Tris, pH 7.5, 0.0033 M EDTA, 1% Triton X-100, 0.1% SDS, 0.5% Na-Doc) and subjected to sonication using a model 250 Branson Scientific Sonic Dismembrator at an amplitude of 30% for 18–20 cycles of a 5-s pulse with 25 s between each pulse. Chromatin (500 μg) was then subjected to immunoprecipitation using 10 μg of Nr2f2 or Nr2f6 antibody bound to protein G magnetic beads (Invitrogen). As a negative control, chromatin was also precipitated with 1 μg of mouse IgG (sc-2025, Santa Cruz Biotechnology) or rabbit IgG (sc-2027, Santa Cruz Biotechnology). Precipitated chromatin was eluted from the beads and crosslinks were reversed overnight at 65°C. Chromatin was treated with RNase A, proteinase K, and the DNA was column purified (PCR Purification kit, Qiagen). Purified DNA was PCR amplified using primers targeting the renin enhancer region, the promoter region, or a region 10 kb upstream of the enhancer as a negative control (Table 1).

RESULTS

The HRE was first shown to be a response element that RAR/RXR could bind and mediate the induction of renin expression by RA (31). We now recognize the HRE as an element that can bind several other nuclear receptors that can regulate the renin promoter in both a positive and negative manner. Previously, we identified Nr2f6 as an HRE binding protein using the HRE as the bait in a yeast one-hybrid screen (23). Nr2f6 was further shown to act as a negative regulator of the renin promoter through its interaction with the HRE in the enhancer. However, Nr2f6 can only account for one of four complexes that could bind with the HRE. Nr2f2 was also identified as an HRE binding protein by yeast one-hybrid. The studies performed herein are to establish the role of Nr2f2 in regulating activity of the renin promoter and its influence on endogenous renin expression.

To test whether Nr2f2 can regulate the mouse renin promoter, we cotransfected cDNA expression vectors for Nr2f2 and Nr2f6 (as a control) with m4.1kP-luc reporter vector into As4.1 cells. The m4.1kP-luc vector contains 4.1 kb of the 5′-upstream sequence of the mouse Ren1 gene driving expression of firefly luciferase. Overexpression of Nr2f2 led to a 70% reduction in promoter activity compared with the empty vector control (pcDNA3.1) (Fig. 1A). Nr2f6 overexpression resulted in an ∼44% reduction. We next performed the complementary experiment where shRNAs targeting endogenous Nr2f2 and Nr2f6 were cotransfected with m4.1kP-luc. Trial experiments in COS-7 cells cotransfecting shRNA-expressing constructs with Nr2f2 or Nr2f6 expression vectors revealed at least 80% knockdown of the target gene without cross reactivity with the other (data not shown). Knockdown of Nr2f2 or Nr2f6 each increased renin promoter activity by more than twofold (Fig. 1B). Together, these results indicate that Nr2f2 (and Nr2f6) negatively regulates activity of the mouse renin promoter.

Fig. 1.

Renin promoter activity when orphan nuclear receptors Nr2f2 and Nr2f6 are overexpressed or knocked down. A: Nr2f2- or Nr2f6 cDNA-expressing plasmids cotransfected with 4.1kP-luc. B: plasmids expressing short hairpin (sh) RNAs to Nr2f2 or Nr2f6 cotransfected with 4.1kP-luc. Shown is the ratio of firefly luciferase (FF; 4.1kP-luc) to Renilla luciferase (RL; phRL-TK) plotted (n = 5, *P < 0.05, 1-way repeated measures ANOVA).

Immunofluorescence revealed that both nuclear receptors are primarily localized to the nucleus in As4.1 cells, although a smaller proportion of Nr2f2 was also present in the cytoplasm (Fig. 2). Recently, gene expression in a variety of kidney cell types has been profiled, including native juxtaglomerular (JG) cells isolated by virtue of a eYFP reporter gene inserted into the renin locus (4). This analysis revealed that both Nr2f2 and Nr2f6 are expressed in native JG cells (www.gudmap.org and Gomez AR, personal communication), although their level of expression is not enriched relative to other cell types in the kidney. Thus, like many ubiquitously expressed transcription factors, in particular nuclear receptors, Nr2f2 and Nr2f6 are primarily localized to the nucleus of As4.1 cells, and their expression is retained in native JG cells.

Fig. 2.

Localization of Nr2f2 and Nr2f6 in As4.1 cells. A: Nr2f2 was labeled using mouse monoclonal anti-Nr2f2 antibody followed by staining with goat anti-mouse Alexa 488-conjugated secondary antibody. B: Nr2f6 was labeled using rabbit anti-Nr2f6 antibody followed by staining with goat anti-rabbit Alexa 488-conjugated secondary antibody. Nuclei were stained with TOPRO3.

We next used EMSAs to determine whether Nr2f2 could directly bind to the HRE. We combined equivalent volumes of in vitro translated Nr2f2 and Nr2f6 with the HRE probe (Fig. 3A). Two specific shift complexes are formed that correspond to those seen when Nr2f2 and Nr2f6 are incubated with the HRE probe alone (Fig. 3B). Each complex is effectively competed away by a 100-fold excess WT cold probe (Fig. 3B). That competition is blunted when either the half-site (μb or μc) or both (μbc) are mutated. These results signify that both half-sites are necessary for binding to the HRE, and therefore Nr2f2 and Nr2f6 bind as dimers. This is consistent with the proposed functional DNA binding form of Nr2f2 (7, 20). However, our data do not eliminate the possibility that Nr2f2 and Nr2f6 can bind as monomers, as some competition is maintained with an intact b site and mutated c site (μc competitor). It is possible that binding to the HRE could be primarily dependent on a single intact b site, with Nr2f2 and Nr2f6 each binding as monomers. It has been shown that Nr2f2 and Nr2f6 can form DNA binding heterodimers (2). The top complex was entirely supershifted by the addition of a Nr2f2 antibody, whereas the bottom complex was shifted by a Nr2f6 antibody (Fig. 3B). Neither antibody supershifted both bands, strongly suggesting that Nr2f2 and Nr2f6 do not form heterodimers when binding to the HRE, at least in vitro.

Fig. 3.

Analysis of Nr2f2 and Nr2f6 direct binding to the hormone response element (HRE). A: probe sequences used for EMSA analysis. The HRE repeat motif is in bold, while mutated bases are in lower case. B: EMSA analysis using equivalent volumes of in vitro transcribed/translated Nr2f2 (F2), Nr2f6 (F6), or both (B) added to the same binding reaction with a ³²P-labeled dsDNA HRE probe. Probes used for competition include the unmutated wild-type (WT) probe and probes with mutated half-sites (μb or μc) or with both mutated (μbc). Supershifts were done by adding Creb1 (C1), Nr2f2 (F2), or Nr2f6 (F6) antibody to binding reactions. Creb1 was used as a negative control. C: EMSA using As4.1 cell nuclear extracts (NE). The 2 panels represent separate experiments.

Our gel-shift analysis suggests that the negative regulation of the renin promoter might be mediated by a direct binding of Nr2f2 and Nr2f6 to the HRE. To verify that Nr2f2 is one of the proteins from As4.1 cells that form a complex with the HRE, we first performed the same gel-shift analysis with nuclear extracts. As observed previously, four shift complexes (a, b, c, d) are effectively competed away by excess WT probe (Fig. 3C) (23). Complexes b and c are still competed away when only one of the half-sites is mutated (μb or μc) but is lost with mutation of both half-sites (μbc). In contrast, mutant half-site probes are less effective competitors for unidentified complexes a and d. We hypothesized that complexes b and c correspond to Nr2f2 and Nr2f6, respectively, based on their similarity to the shift complexes formed by in vitro translated Nr2f2 and Nr2f6. In support of complex c being Nr2f6, the addition of Nr2f6 antibody results in the supershift of complex c (Fig. 3C). Surprisingly, none of the complexes was supershifted or reduced by addition of Nr2f2 antibody, making it difficult to definitively identify complex b as Nr2f2. However, it is difficult to predictably resolve complex b using nuclear extracts and the HRE probe in EMSA assays (compare lane 4 with lane 9 in Fig. 3C). Because of this difficulty, we performed a DNA affinity precipitation assay (DAPA) as an alternative to determine whether endogenous Nr2f2 in nuclear extracts binds to the HRE. DAPA is an attractive alternative to EMSA because it involves a two-step process which increases specificity and sensitivity. Proteins which bind to the HRE are first affinity purified and then are identified by Western blotting using specific antisera. Comparing WT to mutant HRE provides an assay of specificity. DAPA also provides an opportunity to monitor what flows through (FT) the affinity beads, what washes off under nonstringent conditions, and finally what elutes under denaturing conditions. Both Nr2f2 and Nr2f6 show clear enrichment for binding to the WT DAPA probe compared with a mutant probe, where both half-sites (elements b and c) are mutated (Fig. 4). The selective pull down of Nr2f2 and Nr2f6 with the WT probe shows that their binding is strongly dependent on intact TGACCT repeats and provides further support that both Nr2f2 and Nr2f6 bind to the HRE. Most importantly, ChIP analysis was next used to validate this conclusion by showing that both nuclear receptors bind to the HRE in its native chromatin context (Fig. 5). There is clear enrichment for the binding of Nr2f2 or Nr2f6 at the enhancer region when specific antibodies were used to immunoprecipitate As4.1 chromatin. There was no signal when control IgG antisera was used, and the specificity of the enhancer signal was validated when the same chromatin was probed at a region ∼10 kb upstream of the enhancer. Therefore, Nr2f2 and Nr2f6 bind to the enhancer under baseline conditions in As4.1 cells.

Fig. 4.

DNA affinity precipitation assays (DAPA). Shown are Western blots from DAPA using WT or mutant biotin-TEG-labeled HRE probe and As4.1 cell nuclear extracts. The same nuclear extract input from As4.1 cells (In) was incubated with a biotin-labeled WT or mutant (M) HRE probe. Probes were precipitated with streptavidin-labeled beads, washed, and eluted. Input, unbound (flows through; FT), washed, and eluted fractions were run on the same Western blots and probed for Nr2f2 or Nr2f6. Size markers are in kDa.

Fig. 5.

Binding of Nr2f2 and Nr2f6 to the endogenous renin enhancer. Chromatin immunoprecipitation (ChIP) analysis was performed using increasing amounts of Nr2f2 or Nr2f6 antibody. Immunoprecipitated DNA was PCR amplified using primers targeting the HRE or a negative control region (NG) 10 kb upstream of the renin enhancer. PCR for each primer set were run on the same 2% agarose gel. Brightness and contrast were changed uniformly across all gels.

Data from multiple assays (luciferase assays, EMSA, DAPA, and ChIP) strongly suggest that Nr2f2 acts as a negative regulator of renin promoter activity through its binding to the HRE of the renin enhancer. We aimed to substantiate that role further by determining the response of the endogenous renin gene to Nr2f2 knockdown. To ensure a high-efficiency knockdown, we used the same shRNAs used in luciferase assays but incorporated them into adenoviral vectors. As4.1 cells were incubated with AdshNr2f2, AdshNr2f6, or a control virus expressing an shRNA directed against GFP. Expression of Nr2f2, Nr2f6, and renin mRNA was measured using qPCR. We observed robust knockdown of endogenous Nr2f2 at both the mRNA and protein levels in response to AdshNr2f2 (Fig. 6A). Despite some sequence similarity, knockdown of Nr2f2 was specific as there was no effect on expression of Nr2f6 mRNA or protein (Fig. 6B). Even with efficient knockdown, there was no effect on baseline renin expression (Fig. 7A). Nr2f6 knockdown was also efficient and specific (Fig. 6) and resulted in a 1.8-fold increase in endogenous renin mRNA (Fig. 7A). These results suggest that although Nr2f2 can negatively modulate renin promoter activity in transiently transfected cells, loss of Nr2f2 is insufficient to alter expression of the renin gene when in its native chromatin environment under baseline conditions. This led us to ask whether the knockdown of Nr2f2 results in a state where the response to an inducer of renin expression might be enhanced. When Nr2f2 expression was knocked down in As4.1 cells, the induction of renin expression was significantly enhanced in response to RA (Fig. 7B). The response to RA was greater than that seen in both shGFP- and shNr2f6-expressing cells. This indicates that although baseline renin expression is unaffected by Nr2f2, it can modulate the response to a positive stimulus such as RA.

Fig. 6.

Orphan nuclear receptor expression following Nr2f2 or Nr2f6 knockdown in As4.1 cells. A: Nr2f2 mRNA and protein levels in As4.1 cells expressing shRNA to Nr2f2 or Nr2f6 relative to renin mRNA in shGFP-expressing cells. GFP, green fluorescent protein. B: Nr2f6 mRNA and protein levels in As4.1 cells expressing shRNA to Nr2f2 or Nr2f6 relative to renin mRNA in shGFP-expressing cells. Expression levels of mRNA were determined by RT-qPCR (n = 6, *P < 0.05, 1-way repeated measures ANOVA). Western blots for Nr2f2 and Nr2f6 were performed using protein extracts from shGFP-, shNr2f2-, and shNr2f6-expressing As4.1 cells. Size markers are in kDa..

Fig. 7.

Renin expression following Nr2f2 or Nr2f6 knockdown in As4.1 cells. A: renin mRNA levels in As4.1 cells expressing shRNA to Nr2f2 or Nr2f6 relative to renin mRNA in shGFP-expressing cells (n = 6, *P < 0.05, 1-way repeated measures ANOVA). B: fold-change in renin expression induced by retinoic acid relative to vehicle (DMSO). Control cells were infected with shGFP-, shNr2f2-, or shNr2f6-expressing adenovirus and stimulated with 10 μM all trans-retinoic acid or DMSO for 24 h. Fold-change is relative to the DMSO-treated control for each shRNA (n = 4, *P < 0.05 vs. control. #P < 0.05 vs. shNr2f6, 1-way repeated measures ANOVA).

A recent study utilized binding site conservation, microarrays, and ChIP to identify Nr2f1 binding sites across the entire genome (26). The consensus binding site sequence identified in this study resembles the HRE found in the renin enhancer, and Nr2f1 is a closely related subfamily member of Nr2f2 and Nr2f6. Consequently, we utilized target genes identified in that study to determine whether Nr2f2 or Nr2f6 knockdown was sufficient to effect transcription of other potential target genes in As4.1 cells. None of the genes tested (Sod1, Crabp1, or Foxo3a) were differentially expressed upon knockdown of Nr2f2 (data not shown). However, expression of superoxide dismutase 1 (SOD1) was increased by Nr2f6 knockdown (Fig. 8). There was also a positive ChIP signal for Nr2f6 in the region identified previously as an Nr2f1 binding site. These results indicate that Nr2f6 knockdown is sufficient to effect the transcription of other genes with similar HREs. However, there is some level of selectivity as the expression of some genes containing this sequence were not changed.

Fig. 8.

Regulation of superoxide dismutase 1 (SOD1) by Nr2f6. The expression of SOD1 was determined by qPCR using RNA samples from As4.1 cells expressing shGFP, shNr2f2, or shNr2f6. Inset: ChIP analysis was performed using Nr2f6 antibody. Immunoprecipitated DNA was PCR amplified using primers targeting a region of the SOD1 promoter (SOD1). Primer sequences used were described previously (26).

DISCUSSION

Despite the identification of a number of nuclear receptors that can regulate the renin enhancer, when ablated, none of them causes the same robust change that results when the HRE is mutated. Furthermore, we have yet to identify all of the proteins in As4.1 cells that are able to bind to the HRE. Previously, we identified the orphan nuclear receptor Nr2f6 (EAR2) as a negative regulator of the renin gene. The same yeast one-hybrid screen that identified Nr2f6 indicated that another subfamily member, Nr2f2 (Coup-TFII), could bind to the renin enhancer HRE. Therefore, we proceeded to characterize the role of Nr2f2 in regulating renin expression.

As a first line of study, we utilized transient transfections of a vector containing the mouse renin promoter driving expression of luciferase cotransfected with overexpression or shRNA plasmids. Luciferase assays indicate that Nr2f2 and Nr2f6 act as repressors of renin promoter activity. That repression appears to be mediated by direct binding to the HRE as indicated by EMSA, DAPA, and ChIP experiments. Luciferase and DNA-binding experiments taken together suggest that the repressor activity of Nr2f2 and Nr2f6 are mediated through the HRE. However, Nr2f2 and Nr2f6 appear to have divergent actions on endogenous renin expression as it was increased after Nr2f6 knockdown but was unchanged by knockdown of Nr2f2. The opposite was true with RA treatment. Whereas Nr2f6 knockdown had very little effect, Nr2f2 knockdown resulted in a significant enhancement of RA-induced renin expression. It suggests that Nr2f2 knockdown has a negative effect on endogenous renin promoter activity when the RAR is activated. Thus Nr2f2 and Nr2f6 may regulate renin expression in response to different signals. A role for Nr2f2 in RA signaling is well established (21, 34, 35). It appears that Nr2f2 can carry out an inhibition of RA signaling though either active repression or transrepression of RAR or RXR. The fact that both Nr2f2 and RAR bind to the HRE suggests that Nr2f2 mediates its effects on RA signaling through active repression. Nevertheless, a transrepression mechanism where Nr2f2 directly binds to RAR (or some other factor) cannot be ruled out. It is interesting to note that Nr2f2 can interact with a protein called Rere that can bind to both p300 and HDACs, proteins involved in remodeling chromatin (35, 36, 40). Microarray analysis suggests that Rere is expressed in As4.1 cells (GSE14243) (14). Further experimentation is warranted to examine how Nr2f2 exerts is repressive effects.

Examination of the crystal structure of Nr2f2 suggests that it is an autorepressed conformation but promotes transcription activity in multiple cell lines (20). Furthermore, mutation of sites responsible for cofactor binding, dimerization, and ligand binding reduces transcription activity. These results suggest that there may be a ligand for Nr2f2. Therefore, Nr2f2 could potentially act as both an inhibitor of RA-induced renin expression and activator in response to another ligand. It is possible that Nr2f2 is regulated posttranslationally, but to date no modifications have been identified. Konoshita et al. (18, 19) have shown that Nr2f2 negatively regulates the human renin gene through a sequence in the promoter. Furthermore, genetic data in humans and Dahl salt-sensitive rats show an association of Nr2f2 with hypertension (3, 10, 17, 33). Given these data, the modulation of Nr2f2 activity might play an important role in the development of hypertension through the control of renin expression.

Our data support the role of Nr2f6 as a negative regulator of renin expression through binding to the HRE. We replicated previous luciferase experiments and showed that its knockdown modestly increases endogenous renin expression. Inducing renin gene expression in As4.1 cells has been shown to be difficult. For instance, stimulation of renin expression in As4.1 cells via activation of the cAMP pathway by forskolin requires phosphodiesterase inhibition. That primary JG cell cultures respond to the same stimulus (16) suggests that cAMP signaling and renin expression may be near maximal in As4.1 cells. Consequently, inhibition of the cAMP-PKA pathway might be required to see the full impact of Nr2f6 knockdown on renin expression.

The mechanisms underlying Nr2f6 regulation and the genes which it regulates are poorly understood. A recent study showed that Nr2f6 binding is inhibited when phosphorylated by PKC (13). Although increases in calcium, which can activate PKC, play an important role in negatively regulating renin expression, the phosphorylation of Nr2f6 in As4.1 cells would prevent its binding and negative regulation of the renin promoter. In that same study, in vitro kinase assays revealed a phosphorylation of Nr2f6 by PKA. A mechanism by which PKA phosphorylation of Nr2f6 leads to a decrease in its binding to the HRE and activation of renin expression is an attractive one. In fact, we have reported a robust inhibition of endogenous renin expression when Nr2f6 is overexpressed (23). In that situation, Nr2f6 levels may be high enough to provide an unphosphorylated pool capable of binding the HRE and repressing the renin promoter.

In the process of validating knockdown of our nuclear receptors, we identified a gene (Sod1) that was upregulated by Nr2f6 knockdown. Sod1 converts superoxide radicals to oxygen and H₂O₂. We previously reported that hydrogen peroxide (H₂O₂) negatively regulates renin expression (14). Thus the upregulation of Sod1 could lead to an increase in H₂O₂, which might in turn act to decrease renin expression. However, we observed increased renin expression during Nr2f6 knockdown, suggesting that a Sod1-mediated increase in H₂O₂, if present, was not the primary mechanism underlying the Nr2f6-mediated change in renin expression.

Our experiments have revealed that the orphan receptors Nr2f2 and Nr2f6 play a role in the negative regulation of renin promoter activity. We initially hypothesized that HRE binding proteins would be positive regulators given its potency in activating promoter activity both in vitro and in vivo. Moreover, mutation of the HRE decreases, not increases renin promoter activity. The RAR exerts a positive effect, but it is not very robust and may therefore play a minor role. However, we show here that decreasing the activity of another HRE binding protein can result in a more robust response. There is at least one binding complex from EMSA analysis that remains unidentified, and other transcription factors not detected by EMSA could be binding to the HRE. Future experiments will utilize our DAPA protocol combined with mass spectrometry analysis to identify this protein. This will provide for a high-throughput, unbiased approach for identifying HRE binding proteins from As4.1 cells. This has been used previously to identify binding proteins for other motifs (5, 25).

Perspectives

Nuclear receptors are a diverse family of transcription factors that regulate many physiological processes including development, metabolism, vascular function, circadian rhythm, and reproduction. Many are activated by ligands that allow them to respond to physiological changes in the body and modify transcriptional programs. The nuclear receptor superfamily is an attractive pharmacological target because receptor-selective, cell-type selective, activity selective, as well as partial, full, and inverse agonists have been developed (12). Some nuclear receptors are identified as orphans because they lack a known endogenous ligand. These have become increasingly interesting given the synthetic ligands that have been discovered for orphan receptors such as PPARγ. It serves as a good example of activity selective ligands for a nuclear receptor. In the case of PPARγ, the ligand MRL24 shows moderate affects on PPARγ transcriptional activity, but strong inhibition of its phosphorylation (1, 6, 28). These two pathways result in different mechanisms of gene regulation (6, 28). Because of their pharmacological and therapeutic potential, orphan nuclear receptors may provide novel targets for the treatment of hypertension.

GRANTS

This work was supported through National Institutes of Health Grants HL084207, HL048058, and HL061446. The authors also gratefully acknowledge the generous research support of the Roy J. Carver Trust.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: E.T.W. and C.D.S. provided conception and design of research; E.T.W. and X.L. performed experiments; E.T.W., X.L., and C.D.S. analyzed data; E.T.W., X.L., and C.D.S. interpreted results of experiments; E.T.W. and C.D.S. prepared figures; E.T.W. and C.D.S. drafted manuscript; E.T.W., X.L., and C.D.S. edited and revised manuscript; E.T.W., X.L., and C.D.S. approved final version of manuscript.