Originally published In Press as doi:10.1074/jbc.M702228200 on April 20, 2007
J. Biol. Chem., Vol. 282, Issue 24, 17517-17529, June 15, 2007
Expression of the Rat Sterol Regulatory Element-binding Protein-1c Gene in Response to Insulin Is Mediated by Increased Transactivating Capacity of Specificity Protein 1 (Sp1)*
Xiong Deng
1,
Chandrahasa Yellaturu
,
Lauren Cagen
,
Henry G. Wilcox,
Edwards A. Park,
Rajendra Raghow2, and
Marshall B. Elam
From the
Medical and Research Service, Department of Veterans Affairs Medical Center, Memphis, Tennessee 38104 and the Departments of Pharmacology and Medicine, University of Tennessee Health Science Center, Memphis, Tennessee 38163
Received for publication, March 14, 2007
, and in revised form, April 19, 2007.
 |
ABSTRACT
|
|---|
The induction of genes involved in lipid biosynthesis by insulin is mediated in part by the sterol regulatory element-binding protein-1c (SREBP-1c). SREBP-1c is directly regulated by insulin by transcriptional and post-transcriptional mechanisms. Previously, we have demonstrated that the insulin-responsive cis-acting unit of the rat SREBP-1c promoter is composed of several elements that include a sterol regulatory element, two liver X receptor elements, and a number of conserved GC boxes. Here we systematically dissected the role of these GC boxes and report that five bona fide Sp1-binding elements of the SREBP-1c promoter determine its basal and insulin-induced activation. Luciferase expression driven by the rat SREBP-1c promoter was accelerated by ectopic expression of Sp1, and insulin further enhanced the transactivation potential of Sp1. Introduction of a small interfering RNA against Sp1 reduced both basal and insulin-induced activation of the SREBP-1c promoter. We also found that Sp1 interacted with both SREBP-1c and LXR
proteins and that insulin promoted these interactions. Chromatin immunoprecipitation studies revealed that insulin facilitated the recruitment of the steroid receptor coactivator-1 to the SREBP-1c promoter. These studies identify a novel mechanism by which maximal activation of the rat SREBP-1c gene expression by insulin is mediated by Sp1 and its enhanced ability to interact with other transcriptional regulatory proteins.
|
INTRODUCTION
|
|---|
Insulin is a potent inducer of the enzymes catalyzing de novo synthesis of fatty acids in liver and adipose tissue. This enhanced hepatic synthesis of fatty acids contributes to overproduction of very low density lipoprotein and hypertriglyceridemia in insulin-resistant states, including obesity and type II diabetes. The induction of lipogenic enzymes by insulin is mediated by the transcription factor SREBP-1c,3 which is itself regulated via a combination of transcriptional and post-transcriptional mechanisms (1–3). The insulin-mediated increase of de novo lipogenesis is opposed by the counter-regulatory hormone glucagon or other agents that increase intracellular cyclic AMP (4, 5).
We have investigated the molecular mechanisms by which insulin and glucagon regulate rat SREBP-1c gene expression. We have demonstrated previously that there are multiple insulin-response elements in the SREBP-1c promoter, including a sterol regulatory element (SRE) complex (SRE, E-box, NF-Y, and Sp1-binding sites) in the proximal region of the promoter and two LXR-response elements (LXREs) located upstream. The SRE complex and LXREs are highly conserved in the promoters of mouse and rat (1, 2). It has been proposed that these two cis-acting modules are specifically activated by insulin to express SREBP-1c protein and thereby provide a "feed forward" amplification of the response (1, 6). This rapid induction of SREBP-1c as insulin levels increase leads to sustained up-regulation of SREBP-1c in hyperinsulinemic states (7). Insulin treatment increases the transactivating capacity of LXR, and in the mouse gene the activation of the SREBP-1c promoter by insulin has been postulated to be primarily mediated by the two LXREs in the mouse gene (2, 8). However, we have demonstrated that even after both LXREs were rendered inactive, the rat SREBP-1c promoter retained a significant response to insulin (2). Based on these studies, we postulated that the regulation of the SREBP-1c promoter by insulin involves combinatorial interactions of multiple cis-acting elements. In addition to the SRE complex and LXREs, the SREBP-1c promoter also contains multiple GC-rich regions that may bind to members of the Sp1 family of transcription factors.
Sp1 is an important transcription factor for the expression of both ubiquitous and tissue-specific genes (9). In addition to its role in maintaining basal transcription, Sp1 appears to selectively mediate transcriptional activation of some genes by hormonal and nutritional stimuli (10). The presence of Sp1-binding motifs adjacent to an SRE has been observed in the promoters of a number of SREBP-responsive genes that include cytosolic acetyl-coenzyme A synthetase (11) and the low density lipoprotein receptor (12). Sp1-binding sites are also found in the promoters of many insulin-responsive genes, suggesting the involvement of Sp1 in the insulin response (10). However, the current understanding of the mechanism by which insulin regulates the transactivational capacity of Sp1 is limited and based on indirect evidence (10). Examination of the rat and mouse SREBP-1c promoters reveals the presence of multiple potential Sp1 sites located in close proximity to LXR- and SREBP-1c-binding motifs (1, 6). However, the exact role of Sp1 in mediating the response of the SREBP-1c promoter to insulin is unknown. Therefore, we undertook a systematic analysis of the role of Sp1 in the response of the SREBP-1c promoter to insulin. We demonstrate that the five functioning Sp1-binding sites in the rat SREBP-1c promoter regulate both basal and insulin-stimulated activity. We present evidence to indicate that insulin increases the transactivating capacity of Sp1 and promotes its interaction with other transcription factors and coactivators to optimally activate the SREBP-1c promoter.
|
MATERIALS AND METHODS
|
|---|
Cell Culture—Hepatocytes were obtained from livers of male Sprague-Dawley rats (
300 g; Harlan Laboratories, Indianapolis, IN) by collagenase perfusion as described previously (13). Cells were suspended in RPMI medium (Invitrogen) containing 5% fetal bovine serum (Sigma), 10 mM glucose, 1 µM dexamethasone, and 100 nM insulin. Each 60-mm diameter culture dish, coated with rat tail collagen (Collaborative Biochemical Products, Bedford, MA), was seeded with 3 x 106 cells. After 4 h, nonadherent cells were removed, and adherent cells were incubated overnight in RPMI medium without serum or hormones. Cells were transfected with reporter or expression plasmids as described below, and incubation was continued for a further 24 h in fresh medium supplemented with 0.75% delipidated bovine serum albumin, 20 mM glucose, and 100 nM dexamethasone, and with addition of insulin (100 nM), dibutyryl cyclic AMP (Bt2cAMP, 100 µM), or with the LXR agonist, T0901317 (10 µM) as described below. Human embryonic kidney (HEK 293) cells were purchased from Invitrogen, grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, and maintained at 37 °C in a humidified chamber supplied with 5% CO2. Rat hepatoma cells McArdle 7777 (MCA) were obtained from the ATCC (Manassas, VA) and were maintained in the same manner as for HEK 293.
Plasmids—The construction of the full-length and truncated vectors of the rat SREBP-1c promoter-luciferase construct (pSREBP-luc) has been described in detail (1, 2). Mutations of the Sp1-binding sites in pSREBP(–1516/+40)-luc were conducted by site-directed mutagenesis using the QuikChange kit (Stratagene, La Jolla, CA). The sequence of each Sp1 site was mutated as depicted in Table 1 to abolish binding of Sp1, and the mutations were confirmed by DNA sequencing. A truncated SREBP-1c luciferase reporter consisting of the two LXREs and flanking Sp1-binding sites of a 97-bp proximal SREBP-1c promoter corresponding to nucleotides –157 to –253 was synthesized by PCR using primers 5'-cactaagcttcaggggctgggacggcagt-3' (forward) and 5'-cactggatcctgaatggggccggggttact-3' (reverse) and cloned into the pTKLUC vector (ATCC).
View this table:
[in this window]
[in a new window]
|
TABLE 1
Oligonucleotides for EMSA of Sp1
Double-stranded oligonucleotides from five putative Sp1 sites in rat SREBP-1c promoter were tested for DNA binding by EMSA. The Sp1 motifs are in boldface. Each oligonucleotide contains an XbaI overhang to facilitate radiolabeling of the sequence by [32P]dCTP. The GC-box sequence represents a consensus oligonucleotide for Sp1 binding and is used for positive control in EMSA.
|
|
The pCMV-Sp1 plasmid was a gift from Dr. Guntrum Suske (Institut fur Molekularbiologie und Tumorforschung, Philipps-Universitaet Marburg, Marburg, Germany). The synthetic Sp1 reporter construct pGAG6 and control construct pGAM were provided by Dr. J. E. Kudlow (University of Alabama at Birmingham). Plasmid pCMV-hLXR was a gift from Dr. David J. Mangelsdorf (Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas). pTarget-hLXR was constructed by ligating the insert from pCMV-hLXR into pTarget empty vector (Promega, Madison, WI). The expression vectors for CBP (pCBP) and SRC-1 (pSRC-1) have been described previously (14). To obtain the SREBP-1c expression plasmid, the cDNA copy of the rat SREBP-1c gene was first made using the ProtoScript First Strand cDNA synthesis kit (New England Biolabs). The fragments of mRNA encoding the first 403 amino acids in the N terminus of SREBP-1c were then amplified by PCR. The primers used in PCR were 5'-cctgaattccgatggattgcacatttgaagacatgc-3' (forward) and 5'-attaggtacctcaggtttcatgccctccataga-3' (reverse). The PCR-amplified fragments were ligated into the pEGFP-C3 expression vector (Clontech).
The Gal4 luciferase reporter vector driven by four Gal4 DNA-binding sites has been described previously (14). The Gal4-Sp1 construct was made by ligating the Sp1 cDNA into the expression vector pM (Clontech) the cDNA encoding amino acids 93–592 of the transactivation domain of the human Sp1 protein. The cDNA fragment of Sp1 was obtained by PCR following cDNA synthesis using forward (5'-cactgaattcacaggtgagcttgacctc-3') and reverse (5'-cactaagctttgggctgttttctcctt-3') primers. The in-frame fusion of the cDNA fragment with Gal4 DNA-binding domain was confirmed by sequencing.
Small Interference RNA for Sp1 and LXR—Duplexes of the small interference RNA for Sp1 (RNAi-Sp1) and LXR (RNAi-LXR) were synthesized by IDT (Coralville, IA). The sequences of RNAi-Sp1 are 5'-uugagucacccaaugagaatt-3' (forward) and 5'-ttaacucaguggguuacucuu-3' (reverse), respectively, targeting both rat (GI: 6981569) and human (GI: 38272900) Sp1 coding region 213–231 relative to the translation start site; the sequences of RNAi-LXR are 5'-acagcucugcccagaacaatt-3' (forward) and 5'-uuguucugggcagagcugutt-3' (reverse), targeting the coding region of the rat LXR (GI: 555751) 630–648 relative to the translation start site. The single-stranded siRNAs were dissolved in the duplex buffer (100 mM potassium acetate, 30 mM HEPES, pH 7.5) and annealed by incubating at 90 °C for 2 min followed by cooling to room temperature.
Preparation of Nuclear Extracts—Nuclear extracts from rat hepatocytes and HEK 293 cells were prepared as outlined previously (2). In brief, cells were washed and scraped in ice-cold phosphate-buffered saline and resuspended in cell lysis buffer consisting of 20 mM HEPES, pH 7.9, 20% glycerol, 0.1% Triton X-100, 10 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 1 mM EDTA, 1 mM DTT, 1.0 µg/ml pepstatin A, 1 µg/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, and 1/100 volume phosphatase inhibitor mixture (Sigma). Cells were incubated in the lysis buffer for 15 min at 4 °C on a rocking platform, and crude nuclear pellets (5 min at 500 x g) were resuspended in extraction buffer containing 20 mM HEPES, pH 7.9, 20% glycerol, 420 mM NaCl, 10 mM MgCl2, 1 mM EGTA, 1 mM EDTA, 1 mM DTT, and protease and phosphatase inhibitors. The nuclei were incubated on a rocking platform for 1 h at 4 °C and then centrifuged at 15000 x g for 30 min; nuclear extracts (supernatant) were stored at –80 °C until analysis.
Immunoprecipitation—For the immunoprecipitation of ectopically expressed transcription factors, HEK 293 cells were transfected with expression vectors for LXR, retinoid X receptor , and SREBP-1c. Forty eight h after transfection, the nuclear extracts were prepared as described above. For immunoprecipitation of endogenous LXR, SREBP-1, and Sp1, nuclear extracts were isolated from hepatocytes with or without treatment of insulin (100 nM), Bt2cAMP (100 µM), or both. For all immunoprecipitation assays, 500 µg of nuclear extracts were precleared with 20 µl of protein A/G plus agarose beads for 2 h and incubated with 5 µg of antibodies specific for LXR (Santa Cruz Biotechnology), SREBP-1 (BD Biosciences), Sp1 (Upstate), or normal IgG for 4–12 h. Following that, 30 µl of protein A/G plus agarose beads were added to the reaction and incubated for an additional 2 h. The beads were washed with 1 ml of cell lysis buffer five times, resuspended in 20 µl of 3x SDS-PAGE loading buffer, and denatured by heating at 95 °C for 5 min. The bound proteins were resolved in 7.5% SDS-PAGE followed by Western blotting analysis.
Western Blot Analysis—Total nuclear extracts (100 µg) or immunoprecipitated proteins in 20-µl aliquots were size-fractionated by SDS-PAGE and transferred onto nitrocellulose membranes. After transfer, the membrane was blocked in 5% nonfat powdered milk for 1 h followed by incubation with primary antibodies at room temperature for 1 h. Afterward, the blotted membranes were washed five times with 0.1% Tween 20 Tris-buffered saline (TBST) and incubated for 1 h with horseradish peroxidase-conjugated secondary antibody (Pierce) (1:5000 dilution). The membranes were washed five times with TBST, and the immunoreactive proteins were visualized by the SuperSignal West Femto maximum sensitivity substrate (34095, Pierce).
Quantity One (Bio-Rad) software was used to assess the band densities and quantify the protein levels. The primary antibodies used for immunoblot analyses were LXR (RLD-1) (SC-1206, Santa Cruz Biotechnology), SREBP-1 (557036, BD Biosciences), Sp1 (07-645, Upstate), SRC-1 (SC-8995, Santa Cruz Biotechnology), and p300 (SC-585, Santa Cruz Biotechnology).
Chromatin Immunoprecipitation (ChIP) Assays—ChIP assays were performed following the method described previously (2) with minor modifications. Hepatocytes were incubated in 10% formaldehyde for 10 min at room temperature, and the cross-linked cells were washed in ice-cold phosphate-buffered saline and lysed in SDS lysis buffer. After that, chromatin in cell lysate was sheared to fragments of an average size of 500 bp by sonication. The samples were pre-cleared with salmon sperm DNA-protein A-agarose slurry and then incubated overnight with 10 µg of anti-Sp1 antibodies at 4 °C. The agarose beads were pelleted and washed in low salt, high salt, LiCl, and TE buffer. The adsorbed chromatin was eluted with 1% SDS, 100 mM NaHCO3, and protein cross-links were reversed by heating at 65 °C for 6 h. Protein was removed with proteinase K, and DNA was recovered by chloroform/methanol extraction and ethanol precipitation. The recovered DNA was used as a template for PCR (30 cycles) to amplify a 260-bp segment of the rat SREBP-1c promoter encompassing nucleotides –267 and –8. The primers used were 5'-tggttgcctgtgcggcag-3' (forward) and 5'-tcaggccccgccaggctttaa-3' (reverse). Amplified products were resolved by electrophoresis on 2% agarose gel and visualized by ethidium bromide.
Electrophoretic Mobility Shift Assay (EMSA)—Double-stranded oligonucleotides corresponding to the five putative Sp1-binding sites in the proximal SREBP-1c promoter were designed with XbaI overhangs (Table 1) and labeled with [-32 P]dCTP using the Klenow fragment of DNA polymerase. Radiolabeled probes (30,000 dpm), and rat liver nuclear extract (5 µl) derived from fed rats were combined in a buffer containing 10 mM Tris/HCl, pH 7.5, 1 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 1.0 mg of poly(dI-dC)-poly(dI-dC), and 5% glycerol in a total volume of 20 µl. In supershift experiments, the nuclear extracts were incubated with 1 µg of anti-Sp1 antibody (Upstate) at 4 °C for 1 h before EMSA. Binding was carried out at room temperature (22 °C) for 20 min, and the protein-DNA complexes were then resolved in a nondenaturing 5% acrylamide gel in 20 mM Tris, 190 mM glycine, and 0.1 mM EDTA. After electrophoresis (4 °C for 90 min at 180 V), the gels were dried, and the radioactive bands were visualized by autoradiography as outlined previously (2).
|
RESULTS
|
|---|
Activation of SREBP-1c Promoter by Insulin Requires Sp1—To determine the effect of ectopic expression of Sp1 on insulin stimulation of the SREBP-1c promoter, we transfected rat hepatocytes with pSREBP-1c(–1516/+40)-luc and pCMV-Sp1 or pCMV vector and then incubated for 24 h in the absence or presence of insulin, Bt2cAMP, or both agents. Ectopic expression of Sp1 resulted in a 3-fold increase in the SREBP-1c promoter activity, and in the absence of exogenous expression of Sp1, insulin treatment enhanced SREBP-1c promoter activity by 5-fold (Fig. 1A). However, when hepatocytes coexpressing pSREBP-1c(–1516/+40)-luc and pCMV-Sp1 were treated with insulin, there was a 14-fold enhancement in the expression of luciferase (Fig. 1A). These findings indicate that the rat SREBP-1c promoter responds to exogenous coexpression of Sp1 and insulin synergizes this response. Consistent with our previous findings that Bt2cAMP inhibits the insulin induction of the SREBP-1c promoter (2, 3), the ability of insulin to augment both basal and pCMV-Sp1-stimulated promoter activity was attenuated by Bt2cAMP (Fig. 1A). We directly assessed if transfection of rat hepatocytes with pCMV-Sp1 led to increased expression of Sp1. Whole cell lysates from control and transfected hepatocytes were collected 24 h after transfection, subjected to SDS-PAGE, and analyzed by Western blotting. Parallel transfection of hepatocytes and McArdle 777 (MCA) cells with pCMV-Sp1 followed by assessment of Sp1 expression by Western analysis were done to compare Sp1 expression in two cell types with different transfection efficiencies (data not shown). There was a small but significant (23%) increase in the immunoreactive Sp1 in the transfected hepatocytes; in contrast, there was a 3-fold increase in the Sp1 levels in MCA cells (Fig. 1B). These data demonstrate that the exogenous overexpression of Sp1 affects both basal and insulin-mediated activation of the rat SREBP-1c promoter. We should note that although there was only moderate increase in Sp1 as detected by Western blotting, coexpression of Sp1 led to a 2-fold increase in the basal activity and a 5-fold increase in the insulin response of the SREBP-1c promoter.
View larger version (26K):
[in this window]
[in a new window]
|
FIGURE 1.
Activation of the SREBP-1c promoter by Sp1 and insulin. A, hepatocytes (3 x 106 cells) were cotransfected with pSREBP-1c (–1516/+40)-luc, and pCMV-Sp1 or pCMV vectors were incubated for 24 h with either insulin (100 nM), Bt2cAMP (db-cAMP) (100 µM), or both as indicated. Cells were cotransfected with Renilla luciferase expression plasmid pRL-CMV to normalize transfection efficiencies. Luciferase activity from hepatocytes that had been cotransfected with pSREBP-1c(–1516/+40)-luc, and the empty vector without insulin treatment was set as 1. Transfections were conducted in triplicate in four independent preparations of hepatocytes. Data are provided as mean ± S.E. B, rat hepatocytes or McA-RH7777 cells (MCA) were transfected with 1 µg of pCMV-Sp1 plasmid or empty pCMV vector in 60-mm plates, and nuclear extracts were analyzed for Sp1 expression by Western blot analysis as described under "Materials and Methods." The bands was quantified using the Quantity One software to compare the protein expression levels. Western analysis was repeated three times, and a representative blot is shown.
|
|
To corroborate our observation on the effect of exogenous expression of Sp1 on the SREBP-1c promoter and its modulation by insulin, we knocked down the Sp1 protein using small interference RNA specific for Sp1 (RNAi-Sp1). Rat hepatocytes were transfected with RNAi-Sp1 or cotransfected with pCMV-Sp1 and SREBP1c-luc. Following overnight incubation, the hepatocytes were treated with insulin, and the luciferase activity driven by the SREBP-1c reporter was quantified 24 h later. As shown in Fig. 2A, insulin enhanced the basal activity of the SREBP-1c promoter, and this effect was attenuated by RNAi-Sp1. The greatest stimulatory effect of insulin on the activity of the SREBP-1c promoter was achieved when Sp1 was overexpressed, and this effect was completely abolished by RNAi-Sp1 (Fig. 2A). These results demonstrate that RNAi-Sp1 inhibits insulin-induced luciferase activity in hepatocytes containing endogenous Sp1 and almost completely abolishes the synergistic effect of insulin and ectopically expressed Sp1.
To confirm the effect of RNAi-Sp1 on Sp1 levels, we transfected 50 nM of RNAi-Sp1 into either MCA cells or rat hepatocytes. After overnight incubation, the cells were fed with Dulbecco's modified Eagle's medium supplemented with 10% FBS for 24 h. Whole cell lysates were prepared from cells that had been transfected with or without RNAi-Sp1. Proteins were separated by SDS-PAGE and analyzed by Western blotting. The results indicated that RNAi-Sp1 suppressed Sp1 expression in both MCA cells and rat hepatocytes, reducing Sp1 levels by 66% (Fig. 2B) and 35% (Fig. 2C), respectively. The effect of RNAi-Sp1 on the biosynthesis of Sp1 protein was specific for Sp1 because steady-state expression of Sp3, histone H1, and 
-actin were not affected by RNAi-Sp1 (Fig. 2B).
Identification and Functional Assessment of Putative Sp1-binding Sites—The major insulin-responsive elements of the SREBP-1c promoter apparently reside in the proximal 400 bp; this minimal insulin-responsive SREBP-1c promoter contains an SRE complex (SRE, E-box, and nuclear factor-Y (NF-Y)-binding site) and two LXREs that are located 5'-upstream, and the overall location of these cis-acting elements is conserved between the rat and mouse promoter (1, 6). Examination of the proximal portion of the rat SREBP-1c promoter reveals five GC-rich regions that may be putative Sp1-binding sites (Fig. 3). Two potential Sp1-binding sites (Sp1(site 1) and Sp1 (site 2)) are located either within or immediately adjacent to the SRE complex, and two other Sp1 motifs (Sp1 (site 4) and Sp1 (site 5)) flank the two LXREs. A fifth GC box (Sp1 (site 3)) resides between the SRE complex and LXREs (Fig. 3). We systematically determined whether the putative Sp1-binding sites of the SREBP-1c promoter were needed for its basal and/or insulinmediated activation. We first tested whether rat liver nuclear extracts (RLNE) could bind to the five putative Sp1 sites. In parallel experiments, we assessed the effects of mutations of the five Sp1 sites individually on SREBP-1c promoter activity. As shown in Fig. 4A, radioactively labeled double-stranded oligonucleotides encompassing each of the five Sp1-binding sites (Table 1) individually bound to a protein(s) in RLNE. The addition of anti-Sp1 antibody to RLNE greatly diminished the intensity of the major protein-DNA complex. Based on these EMSAs, we concluded that the rat SREBP-1c promoter contains five bona fide Sp1 motifs (Fig. 4A). The Sp1-DNA complex formation could also be abolished if 50-fold excess of cold Sp1 motif-containing double-stranded oligonucleotides was included in the RLNE (data not shown). Although we did not carry out the EMSA shown in Fig. 4A under rigorous quantitative conditions, we noted that under identical binding conditions the five Sp1 motifs fell into two categories. Thus, Sp1 site 1, Sp1 site 2, and Sp1 site 3 formed more intense protein-DNA complexes that were specifically inhibited by antibodies to Sp1, and the binding of Sp1 site 4 and Sp1 site 5 DNA to RLNE yield lower intensity DNA-protein complexes (Fig. 4A).
To determine the contribution of the Sp1-binding sites to SREBP-1c promoter activation, each Sp1 site was individually mutated (Table 2) in the pSREBP-1c(–1516/+40)-luc construct to abolish Sp1 binding (data not shown). Hepatocytes were transfected with DNA vectors in which the luciferase gene was driven by wild-type SREBP-1c promoter or promoters containing mutations in one of the five Sp1-binding sites. Luciferase expression was quantified to determine the basal promoter activity as well as its modulation by insulin (Fig. 4B). Compared with the wild-type SREBP-1c promoter, basal activity of the SREBP-1c promoters containing mutations in Sp1 site 1 or Sp1 site 2 was reduced to 55 and 61%, respectively (Fig. 4B). Unexpectedly, mutation of the Sp1 site 3 motif, located between the SRE complex and first LXRE, resulted in a 2-fold increase in basal promoter activity compared with WT-pSREBP-1c. Mutations in Sp1 site 4 and Sp1 site 5 motifs that had the lowest apparent binding in EMSA did not significantly affect the basal activity of the SREBP-1c promoter. In contrast, although mutations of the Sp1 site 1, Sp1 site 3, and Sp1 site 4 motifs blunted the induction of the SREBP-1c promoter by insulin (Fig. 4B), the Sp1 site 1 mutant was most effective in this regard. These data indicate that each of the five Sp1-binding sites uniquely contributes to the SREBP-1c promoter by affecting its basal activity and/or the response of SREBP-1c promoter to insulin.
View this table:
[in this window]
[in a new window]
|
TABLE 2
Plasmids containing mutant sequences of the five Sp1 motifs in the full-length rat SREBP-1c promoter
Site-specific mutagenesis was carried out as detailed under "Materials and Methods." The mutated and wild type-Sp1 motifs are shown. These pSREBP-1c(-1517/+40)-luc constructs, each containing a single mutated Sp1 site (sites 1-5) were tested following transient transfection in primary hepatocyte cultures for basal activity and response to treatment with insulin (100 nM) or the LXR agonist T0901317 (10 µM).
|
|
View larger version (40K):
[in this window]
[in a new window]
|
FIGURE 3.
Nucleotide sequence of the minimal insulin-responsive SREBP-1c promoter. The sequence of the SREBP-1c promoter (–360/+40) is shown. The location of the nucleotide sequence motifs that recognize various transcription factors have been underlined. Five Sp1-binding sites are numbered to indicate their location with respect to the SRE complex and LXREs.
|
|
Abundance of Sp1 Protein and the Binding of Sp1 to SREBP-1c Promoter—Having demonstrated that Sp1 sites uniquely influence both the basal and insulin-stimulated activity of the SREBP-1c promoter, we explored a number of potential mechanisms by which insulin might regulate Sp1 to enhance SREBP-1c transcription. We first determined whether insulin increased the abundance of nuclear Sp1. We measured Sp1 protein content of nuclear extracts from rat hepatocytes following 4 to 12 h of treatment with insulin by Western blotting. As illustrated in Fig. 5A, we found no significant changes in the levels of immunoreactive Sp1 following treatment with either insulin or cAMP. Longer treatment (24 h) did not increase nuclear Sp1 levels either (data not shown). These results indicate that in primary rat hepatocyte cultures insulin does not increase the nuclear abundance of Sp1.
View larger version (49K):
[in this window]
[in a new window]
|
FIGURE 4.
Sp1-binding motifs in SREBP-1c promoter interact with nuclear proteins. A, double-stranded oligonucleotides encompassing the five Sp1 sites (detailed in Table 1) were synthesized, radiolabeled with [32P]dCTP, and tested for DNA binding. Probes corresponding to wild-type sequence of Sp1-binding sites 1–5 of the SREBP-1c promoter and a canonical GC box (GC) as a control were mixed with RLNE. An empty lane (E) where no RLNE was added to the GC-box probe was used as a negative control for DNA binding. The identity of the DNA-protein complexes was confirmed in RLNE incubated with antibody to Sp1 (anti-Sp1). B, basal activity and response to insulin of (WT) pSREBP-1c(–1516/+40)-luc and pSREBP-1c(–1516/+40)-luc with mutated Sp1 sites were determined in primary hepatocytes. The schematic (inset) illustrates the location of DNA-binding motifs of transcription factors in the SREBP-1c promoter. Hepatocytes were transfected with WT or mutant (SPMut1–5) promoter constructs, and luciferase activity was measured in cell lysates harvested 24 h following treatment with 100 nM insulin or no addition (Control). Luciferase activity was corrected for differences in transfection efficiency of Renilla luciferase. Transfections were conducted in triplicate in four independent preparations of hepatocytes. Data are provided as means ± S.E.
|
|
We next examined the possibility that insulin might stimulate SREBP-1c transcription by enhancing the binding of Sp1 to the SREBP-1c promoter. ChIP assays were performed to test whether the in vivo binding of Sp1 to the SREBP-1c promoter would be altered by insulin. Chromatin-associated proteins were cross-linked to DNA by formaldehyde from primary hepatocyte cultures following 24 h of treatment with or without insulin. DNA was recovered following immunoprecipitation with anti-Sp1 antibody and was used as a template for PCR to amplify a 260-bp segment of the rat SREBP-1c promoter corresponding to nucleotides –267 and –8. Fig. 5B shows that a SREBP-1c-specific DNA fragment was amplified from chromatin precipitated with anti-Sp1 antibody, demonstrating the in vivo binding of Sp1 to the proximal SREBP-1c promoter. However, we did not detect a change in the amount of PCR product amplified from chromatin prepared from insulin-treated hepatocytes. Our data indicate that enhanced binding of Sp1 to the proximal portion of SREBP-1c promoter was not a mechanism for activation of the promoter by insulin.
To examine the effect of insulin treatment on binding of Sp1 and LXR to the SREBP-1c promoter in vitro, we carried out EMSA using nuclear extracts prepared from hepatocytes treated for 12 h with or without insulin or cAMP. A double-stranded oligonucleotide probe corresponding to LXRE-2 and Sp1 (5) (Fig. 2) sites from the SREBP-1c promoter (5'-ctagaaggggctgggacggcagtgaccgccagtaacccct-3') was labeled with [32P]dCTP and used for EMSA following the procedure described previously (2). As shown in Fig. 5C, two DNA-protein complexes were detected. The intensity of the upper band was greatly reduced by anti-Sp1 antibody, and the formation of the lower band was prevented by anti-LXR antibody; both bands were eliminated if nuclear extracts were incubated for 1 h with anti-Sp1 and anti-LXR antibodies together. These results suggest that Sp1 and LXR present in the rat liver nuclear extracts bind to their cognate DNA sequence in vitro, but treatment with insulin or cAMP did not enhance binding of either Sp1 or LXR to this portion of the promoter.
View larger version (49K):
[in this window]
[in a new window]
|
FIGURE 5.
Nuclear abundance and DNA binding of Sp1 are not affected by insulin. A, 100 µg of nuclear extracts from hepatocytes treated with or without 100 nM insulin or 100 µM Bt2cAMP (db-cAMP) for 4–12 h were subjected to SDS-PAGE and analyzed by Western blotting using anti-Sp1 antibody as outlined under "Materials and Methods." Results are representative of at least three separate experiments. B, Sp1 binds to the SREBP-1c promoter in vivo. Rat hepatocytes were incubated for 24 h in medium with or without insulin (100 nM). Cross-linked chromatin was precipitated with IgG or antibody specific for Sp1 as described under "Materials and Methods." Immunoprecipitated DNA was used as template to amplify a 260-bp segment of the proximal SREBP-1c promoter (–8 to –267). No-IP, no immunoprecipitation. C, EMSA was conducted with 32-P-labeled DNA probe containing the LXRE-2 and Sp1 site 5 from the SREBP-1c promoter. Nuclear extracts were prepared from rat hepatocytes treated with insulin or cAMP for 12 h. Antibodies to Sp1 (anti-Sp1) or LXR (anti-LXR) were added to the binding reaction. Lane E, empty lane.
|
|
Effects of Insulin on Transactivational Activity of Sp1—Because the results of ChIP and EMSA did not support the hypothesis that insulin activates the SREBP-1c promoter by increasing the abundance or DNA-binding affinity of Sp1, we designed experimental strategies to determine whether insulin increases SREBP-1c transcription by increasing the transactivating potency of Sp1. We first tested the ability of insulin to enhance the response of a promoter construct composed of six synthetic Sp1-response elements (pGAG6-luc) to ectopically expressed Sp1 in transfected hepatocytes. Ectopic expression of Sp1 activated the pGAG6-luc 3-fold, and the joint action of ectopically expressed Sp1 and insulin led to a 5-fold additional increase in the activity of pGAG6-luc (Fig. 6A). It is noteworthy that Bt2cAMP alone did not affect the ability of Sp1 to activate the pGAG-6 promoter; in contrast, Bt2cAMP treatment markedly attenuated the effect of insulin on pGAG6-luc (Fig. 6A).
View larger version (21K):
[in this window]
[in a new window]
|
FIGURE 6.
Insulin increases transactivating capacity of recombinant Sp1. A, the reporter plasmid pGAG6-luc contains the TK promoter driven by six Sp1-binding sites as illustrated in the schematic (inset). Hepatocytes were cotransfected with pGAG6-luc and pCMV-Sp1 or an empty expression vector and treated with insulin (100 nM) and/or Bt2cAMP (db-cAMP) (100 µM). Cell lysates were prepared for luciferase assay 24 h following treatments. Luciferase activity of the reporter gene from cells without ectopic expression of Sp1 and insulin treatment is set as 1. Transfections were conducted in triplicate in four independent preparations of hepatocytes. Data are provided as mean ± S.E. B, reporter plasmid pGal4-luc is composed of four Gal4-binding sites fused to a TK promoter as illustrated in the schematic (inset). Hepatocytes were cotransfected with pGal4-luc and a vector expressing a chimeric protein containing the Gal4 DNA-binding sequence with the transcriptional activation domain of human Sp1 (pGal4-Sp1) or the empty Gal4 expression vector. Luciferase activity was quantified in hepatocyte extracts following 24 h of treatment with insulin (100 nM) or Bt2cAMP (100 µM). Data are presented as means ± S.E. from 12 to 18 determinations from 4 to 6 independent experiments.
|
|
Because insulin increases the ability of Sp1 to activate a synthetic hexameric Sp1-response element, we next tested whether insulin specifically affects the transactivation domain of Sp1. A plasmid (pM-Sp1) expressing a chimeric Sp1 protein was constructed in which the transactivation domain of Sp1 (amino acids 93–592) was fused to the DNA-binding domain of the yeast transcription factor Gal4. Hepatocytes were cotransfected with a Gal4 luciferase reporter and a DNA vector (pGal4-Sp1) that ectopically expresses the Gal4-Sp1 fusion protein. As shown in Fig. 6B, the activity of the Gal4-driven luciferase reporter was increased (3-fold) in hepatocytes transfected with the Gal4-Sp1 expression vector. Although insulin alone had no detectable effect on Gal4-luciferase activity as expected, it augmented the ability of the Gal4-Sp1 fusion protein to activate the reporter gene expression, resulting in twice as much activation of the Gal4 luciferase gene as that by Gal4-Sp1 alone (Fig. 6B). Because insulin would not be expected to increase the affinity of the Gal4 DNA-binding domain of the Gal4-Sp1 chimera for its cognate DNA sequence, we conclude that insulin increases the intrinsic transactivating capacity of Sp1.
Functional Cooperativity between Sp1 and LXR—The presence of multiple Sp1 sites in close proximity to the SRE complex and LXREs and the observation of functional cooperativity between insulin and Sp1 to activate the SREBP-1c promoter raised the possibility that insulin increases SREBP-1c transcription by enhancing functional interaction between Sp1 and SREBP-1c and/or LXR. Therefore, we examined the effect of Sp1 and LXR expression on the activity of a TK-luciferase reporter construct driven by a (–253 to –157 bp) truncated SREBP-1c promoter containing the two LXREs flanked by the Sp1 sites 4 and 5. The results of the luciferase assays showed that expression of Sp1 increased the activity of the luciferase reporter construct 3.6-fold, whereas expression of LXR increased the activity of the truncated reporter 12.7-fold (Fig. 7A). Exogenous expression of both Sp1 and LXR led to a 26.5-fold increase in the promoter activity, and insulin treatment further enhanced this effect (Fig. 7A). These observations suggest that the SREBP-1c promoter is activated in a cooperative fashion by Sp1 and LXR and that insulin further enhances this activity. In contrast, ectopic expression of Sp1 and SREBP-1c did not augment the effect of SREBP-1c on the expression of luciferase reporter construct driven by a 149-bp (–109/+40) proximal SREBP-1c promoter containing the SRE, Sp1 sites 1–3 (data not shown).
Because Sp1 sites 4 and 5 in the truncated promoter (–253 to –157) contain only the Sp1-binding sites of Sp1 (sites 4 and 5) (Fig. 4A), we next examined the effect of Sp1 and LXR on the activity of a luciferase reporter construct driven by the full-length, 1.5-kb SREBP-1c promoter (–1516/+40) that contains all the five Sp1 sites as well as the two LXREs. The results of luciferase assays indicated that Sp1 or LXR increased the promoter activity by 2.4- and 5.1-fold, respectively, and coexpression of both Sp1 and LXR led to a 7.5-fold increase in the promoter activity. Insulin augmented this effect further. When both Sp1 and LXR were coexpressed, insulin treatment led to a 32-fold increase in the promoter activity (Fig. 7B).
To better define the roles of Sp1 and LXR in the activation of the SREBP-1c promoter, we investigated the impact of RNAi-Sp1 and RNAi-LXR on the response of the 1.5-kb SREBP-1c promoter to ectopically expressed Sp1 and LXR. RNAi-LXR and RNAi-Sp1 significantly reduced the response of the promoter to exogenous Sp1 and LXR and attenuated the additional response to insulin (Fig. 7C). Furthermore, we determined the effect of suppressing endogenous Sp1 and LXR on the response of the SREBP-1c promoter to insulin. Inhibition of endogenous LXR and Sp1 slightly reduced the basal activity of the promoter but significantly blunted the response of promoter to insulin (Fig. 7D). These results revealed that both LXR and Sp1 are required for a full insulin stimulation of the SREBP-1c promoter. To confirm that inhibition of Sp1 and LXR by RNAi-Sp1 and RNAi-LXR contributed to the reduction in the responses of the SREBP-1c promoter to insulin, whole cell lysates were prepared from hepatocytes transfected with or without RNAi-LXR and RNAi-Sp1. Proteins were resolved by SDS-PAGE and analyzed by Western blotting. The results verified that RNAi-LXR and RNAi-Sp1 specifically inhibit the expression of LXR and Sp1 (Fig. 7E), suggesting that reduced Sp1 and LXR levels may be a direct cause of a decrease in the activity of the SREBP-1c promoter.
Physical Association of Sp1 with Other Transcription Factors and Coactivators—Because insulin augmented the functional cooperation between Sp1 and LXR and enhanced the ability of Sp1 to activate the SREBP-1c promoter (Fig. 7, A and B), we hypothesized that insulin may promote the association of Sp1 with LXR. Previous studies have shown that SREBP-1a physically interacts with a number of transcription factors including Sp1 (15). However, whether LXR and Sp1 interact physically has not been examined. We first tested whether we could detect SREBP-1c or LXR in HEK 293 cells. HEK 293 cells were transfected with expression plasmids for SREBP-1c or LXR. Nuclear extracts were subjected to SDS-PAGE followed by Western blot analysis using antibodies specific for SREBP-1 and LXR. The results of Western blot analysis showed that SREBP-1 and LXR could be detected in the nuclear extracts from cells transfected with SREBP-1c and LXR expressing plasmids, respectively. Although the antibody to SREBP-1 does not differentiate between isoforms 1a and 1c (data not shown), we detected an immunoreactive polypeptide band only in the nuclear extracts from cells transfected with the SREBP-1c expression plasmid; apparently the endogenous SREBP-1a or -1c was undetectable under these conditions (Fig. 8A). To test for physical association of the endogenous Sp1 with either LXR or SREBP-1c or both, we immunoprecipitated SREBP-1 and LXR from the nuclear extracts of HEK 293 cells ectopically expressing LXR or SREBP-1c. Immunoprecipitated proteins were size-fractionated and subjected to sequential Western blot analysis with antibodies against Sp1. As shown in Fig. 8B, Sp1 could be detected in the immunoprecipitates of SREBP-1 and LXR-specific antibodies. Therefore, the results of coimmunoprecipitation indicated that both SREBP-1c and LXR were able to physically associate with Sp1 (Fig. 8B).
Although these results show that Sp1 can associate with ectopically expressed SREBP-1 and LXR proteins, it is not known whether these interactions occur under physiological conditions. Therefore, we tested whether Sp1 interacts with the endogenous SREBP-1c or LXR in rat hepatocytes. Because the steroid receptor coactivator-1 (SRC-1) and cAMP-response element-binding protein (CBP/p300) are known to associate with LXR (16) and Sp1 (17), we investigated the possibility that insulin increases association of Sp1 with these coactivators. We first determined the effect of insulin treatment on the abundance of these factors by Western blotting of nuclear extracts from control and insulin or Bt2cAMP-treated hepatocytes. Insulin increased the nuclear content of SREBP-1 5-fold and SRC-1 1.5-fold. The effect of insulin on the nuclear content of SRC-1 and SREBP-1 was effectively reversed by Bt2cAMP (Fig. 9A). In contrast, the steady-state levels of total immunoreactive Sp1, LXR, and p300 remained unchanged regardless of insulin or Bt2cAMP treatment (Fig. 9A).
View larger version (37K):
[in this window]
[in a new window]
|
FIGURE 7.
Functional cooperativity between Sp1 and LXR. A, primary hepatocytes were transfected with a truncated 97-bp SREBP-1c TK-luciferase reporter construct encompassing the two LXREs and flanking Sp1 sites (depicted in schematic on top) corresponding to nucleotides –253 to –157 of the SREBP-1c promoter sequence shown in Fig. 3. Hepatocytes were cotransfected with pTarget-Sp1 (Sp1) and pTarget-LXR (LXR), separately or in combination, or empty expression vector pTarget (vector) as indicated. Luciferase (Luc) activity in control and insulin-treated (100 nM) cells was assayed after 24 h of incubation. Transfections were conducted in triplicate, and the data represent mean ± S.E. from four independent experiment. B, primary hepatocytes were transfected with the full-length 1.5-kb SREBP-1c luciferase reporter construct (depicted in schematic on top) along with pTarget-Sp1 (Sp1) and pTarget-LXR (LXR), separately or in combination, or with an empty vector. Treatment and data analysis were conducted in the same manner as described in A. C, full-length 1.5-bp SREBP-1c promoter construct was cotransfected for 24 h with both pTarget-Sp1 and pTarget-LXR, and with either siRNA for LXR (RNAi-LXR) or siRNA for Sp1 (RNAi-Sp1). Luciferase activity in control and insulin-treated (100 nM) cells was assayed after 24 h of incubation. Transfections were conducted in triplicate, and the data represent the mean ± S.E. from four independent experiment. D, full-length 1.5-bp SREBP-1c promoter construct was cotransfected into primary hepatocytes with RNAi-LXR or RNAi-Sp1 for 24 h. Luciferase activity in control and insulin-treated (100 nM) cells was assayed after 24 h of incubation. Transfections were conducted in triplicate, and the data represent mean ± S.E. from four independent experiment. E, primary rat hepatocytes were transfected with RNAi-Sp1 or RNAi-LXR with or without pTarget-Sp1 or pTarget-LXR. Whole cell lysates were prepared for Western blot analysis using antibodies to LXR or Sp1. Blots using anti--actin antibodies to assess equal protein loading are included. A representative blot of three independent experiments is shown.
|
|
To determine whether insulin increases the association of Sp1 with these transcription factors and or coactivators, we next immunoprecipitated the nuclear proteins from insulin- or Bt2cAMP-treated hepatocytes with antibodies specific for SREBP-1, LXR, or p300 and probed the immunoprecipitated proteins for the presence of other transcription factors and coactivators. As shown in Fig. 9B, immunoprecipitated SREBP-1c-specific complex contained Sp1, and the association between SREBP-1c and Sp1 was greatly enhanced in insulin-treated cells. Insulin also dramatically increased the association of Sp1 with p300 and SRC-1, and the effects of insulin in promoting the association of Sp1 with SREBP-1c, p300, and SRC-1 were inhibited by Bt2cAMP (Fig. 9B).
Insulin Enhances Interaction between Coactivators SRC-1 and CBP to Increase the Transcriptional Activation of SREBP-1c—A number of nuclear receptors, including LXRs, utilize SRC-1 as a coactivator (16). Similarly, CBP and a close homolog p300, also associate with numerous transcription factors and coactivators (18). Based on the observation that insulin increases the association of Sp1 with p300 and SRC-1, we hypothesized that insulin could activate the SREBP-1c promoter by promoting the association of Sp1 with SRC-1 and/or CBP. Therefore, we determined whether the rat SREBP-1c was responsive to ectopic expression of SRC-1 and CBP and, if so, whether insulin treatment would enhance the response. Plasmids expressing SRC-1 and CBP were transfected individually or simultaneously into rat hepatocytes, along with a 97-bp (–253/–157) SREBP-1c promoter-driven TK-luciferase construct (used in Fig. 7A). When hepatocytes were transfected with plasmids expressing CBP or SRC-1 individually, luciferase activity driven by the SREBP-1c promoter was similar to that seen in cells transfected with empty expression vectors. When SRC-1 and CBP were coexpressed, the activity of the SREBP-1c promoter was slightly increased. Similarly, in hepatocytes ectopically expressing SRC-1 or CBP individually, insulin treatment only modestly increased the luciferase expression. In contrast, in cells transfected with plasmids expressing CBP and SRC-1 together, insulin treatment resulted in a robust 10-fold induction of luciferase activity (Fig. 10A). These results suggest that SRC-1 and CBP enhance the insulin induction of SREBP-1c gene expression.
View larger version (33K):
[in this window]
[in a new window]
|
FIGURE 9.
Effects of insulin on expression of endogenous transcription factors and coactivators and their interactions. A, hepatocytes were plated in Dulbecco's modified Eagle's medium overnight and then treated with or without insulin (100 nM) or cAMP (100 µM) or both for 12 h. Nuclear extracts were isolated from control and treated hepatocytes. One hundred micrograms of nuclear extracts were resolved by SDS-PAGE followed by Western blotting analysis. Blots are representative of three comparable experiments. B, 500 µg of nuclear extracts were subjected to immunoprecipitation (IP) with normal IgG or the indicated specific antibodies. Proteins were eluted from protein A/G plus agarose beads, separated by SDS-PAGE, and analyzed by Western blotting (WB). Asterisks denote the IgG heavy chain band. The above results were representative of three independent experiments.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
FIGURE 10.
Insulin induces activation of the SREBP-1c promoter by SRC-1 and CBP. A, rat hepatocytes were transfected with a vector containing –253 to –157 bp of the SREBP-1c promoter driven by TK-luc. An empty vector or plasmids that expressed SRC-1 or CBP were cotransfected with the reporter plasmid. After 24 h of treatment with 100 nM insulin, cell extracts were prepared for luciferase assays. Transfections were conducted in triplicate in four independent preparations of hepatocytes. Data are provided as mean ± S.E. B, hepatocytes were cross-linked with 1% formaldehyde 4 or 12 h after addition of insulin, and ChIP assays were conducted to identify protein complexes associated with the SREBP-1c promoter. In vivo association of SRC-1, CBP, Sp1 and LXR with the SREBP-1c promoter was demonstrated by the amplification of SREBP-1c-specific DNA fragments from chromatin complexes precipitated by antibodies for SRC-1, CBP, Sp1, and LXR, respectively, as detailed under "Materials and Methods."
|
|
Having demonstrated that SRC-1 and CBP potentiate the insulin stimulation of SREBP-1c, we conducted ChIP assays to further investigate whether insulin enhances the in vivo association of CBP and/or SRC-1 with the SREBP-1c promoter. Hepatocytes were cross-linked with 1% formaldehyde at 4 and 12 h after insulin addition. Chromatin from untreated or insulin-treated hepatocytes was analyzed by ChIP (Fig. 10B). A 260-bp DNA fragment encompassing all the five Sp1 sites and the two LXREs of the SREBP-1c promoter-proximal chromatin was amplified from chromatin precipitated with antibodies for SRC-1, CBP, Sp1, and LXR, respectively. Insulin increased the association of SRC-1 with the SREBP-1c promoter at 4 and 12 h; the ChIP data are consistent with the observations of increased nuclear SRC-1 levels and increased association of SRC-1 with Sp1 (Fig. 9B). The increased association of SRC-1 with the promoter may enhance the effect of Sp1 and LXR, which may in turn enhance interactions among other transcription factors such as Sp1 and SREBP-1c and lead to an increase in the promoter activity. Even though the association of CBP with the promoter was not increased, it is still possible that insulin enhances interaction between LXR and other transcription factor(s) for which CBP serves as a coactivator. It should be noted that the ChIP assay data presented here were from samples taken at 4 and 12 h after insulin treatment. Because the association and dissociation of transcription factors and coactivators with the promoter are under constant dynamic change, our data cannot exclude the possibility that the association of CBP with the SREBP-1c promoter is increased at a point we did not capture. But based on the current observations, our results suggest that insulin enhances interactions of the coactivator SRC-1 with transcription factors and increases its recruitment to the SREBP-1c promoter.
|
DISCUSSION
|
|---|
Expression of SREBP-1c, a key regulator of genes that encode lipogenic enzymes, is induced by insulin. We reported previously that a minimal insulin-responsive promoter of the rat SREBP-1c gene has binding sites for several nuclear proteins, including LXR, NF-Y, and SREBP-1c itself (1). Here we report that interspersed among these sequence motifs are five bona fide Sp1 sites that are also essential for full induction of the rat SREBP-1c promoter by insulin in rat hepatocytes. Although neither the nuclear content of Sp1 or DNA binding ability was significantly altered by insulin, the transactivational potency of Sp1 was significantly enhanced by insulin. Insulin also increases the interactions between Sp1 and other transcription factors and coactivators, and elicits interaction between SRC-1 and CBP to optimally activate the rat SREBP-1c promoter. These findings strongly support the hypothesis that induction of SREBP-1c biosynthesis in response to insulin is mediated at least in part by promoting the assembly of a transcriptosome that in addition to SREBP-1c includes such components as Sp1, LXR, SRC-1, and CBP/p300.
It is becoming increasingly evident that Sp1 plays promoter-specific roles in mediating transcriptional effects of hormonal and nutritional signals (10). The promoters of the low density lipoprotein receptor (19), acetyl-CoA carboxylase (20), and fatty-acid synthase (21, 22) genes share a common motif of adjacent binding sites for SREBP and Sp1. Because SREBP-1a and SREBP-1c alone are relatively weak activators of these gene promoters, Sp1 has been postulated to enhance the action of SREBP (23). Previous studies have also demonstrated that Sp1 interacts with SREBP-1a, NF-Y, and a distinct coactivator complex called ARC (19, 24–27) that consists of 16 or more subunits (28, 29). The rat farnesyl-diphosphate synthase gene is activated by synergistic interaction between NF-Y and SREBP1 (29), and cooperativity among NF-Y, Sp1, and SREBP-1 is thought to regulate the promoter of the fatty-acid synthase and S14 genes (22, 30). Here, we report the novel finding that Sp1 physically and functionally interacts with LXR in the context of the rat SREBP-1c promoter. Although we do not directly address the possibility of interaction between Sp1 and NF-Y in the present study, we have previously demonstrated that mutation of the NF-Y-binding site in the proximal SREBP-1c promoter significantly attenuates its response to insulin (3). The potential mechanistic interactions between Sp1 and NF-Y in the SREBP-1c promoter require further study.
We and others have demonstrated that the SRE complex of the SREBP-1c promoter provides feed forward activation that is primarily mediated by SREBP-1c itself (1, 6). In the current studies we sought to determine the role of Sp1 in the activation of the SREBP-1c promoter via the SRE complex, and we show that mutation of the proximal Sp1-binding site diminishes both basal and insulin-stimulated activity of the full-length promoter. Although we could demonstrate physical interaction between Sp1 and SREBP-1, consistent with published data (22–29), exogenous coexpression of these two transcription factors did not synergistically activate the truncated SREBP-1c promoter consisting of only the SRE complex (data not shown). These findings suggest that although both Sp1 and SREBP-1c are important for full activation of the SREBP-1c promoter, they exert their influence independently. Alternatively, the inability to observe synergistic activation of the truncated SREBP-1c promoter may reflect the requirement of additional sequence motifs for full response. On the other hand, we have presented strong evidence for both physical interaction and functional cooperativity between Sp1 and LXR. The SREBP-1c promoter contains two functional LXREs (1, 31) and responds to ectopic expression of LXR, LXR, retinoid X receptors, and their respective ligands (2, 8). We have determined that the binding sites for Sp1 that flank the two LXREs in the SREBP-1c promoter are required for its full response to insulin. We have also shown that Sp1 and LXR functionally cooperate to activate the SREBP-1c promoter.
Although direct association of Sp1 with the cognate DNA was unaffected by insulin in vitro and in vivo, insulin treatment significantly enhanced Sp1-mediated activation of two synthetic promoter constructs GAG6-luciferase and Gal4-luciferase. The enhanced transactivating capacity of Sp1 could result from either post-translational modification of Sp1 or its ability to recruit or activate other transcription factors and/or coactivators. The increased interaction of the coactivators SRC-1 and CBP and concomitant enhancement of transactivating capacity of Sp1 in response to insulin corroborate such a mechanism. Another mechanism underlying the increased transactivating capacity of Sp1 may result from post-translational modification. Sp1 contains consensus phosphorylation sites for calmodulin kinases, casein kinases 1 and 2, protein kinases A and C, and mitogen-activated kinases Erk1 and -2 (10) and may be regulated either positively or negatively, depending on the site of phosphorylation and the target promoter (32–34). Recently, Majumdar et al. (35) have demonstrated that insulin promotes phosphorylation of Sp1 in a rat hepatoma cell line, H411E. We were unable to detect quantitative differences in phosphorylation of Sp1 after insulin treatment by Western blot analysis using anti-phosphoserine antibodies (data not shown). Similarly, it has been postulated that nuclear accumulation of phosphorylated or O-GlcNAc-modified Sp1 may be modulated by increasing nuclear translocation of Sp1 (36). But we did not observe increased nuclear levels of Sp1 in rat hepatocytes following insulin treatment. The increased nuclear localization of Sp1 in response to insulin in H4IIE cells and insulin-treated streptozotocin-induced diabetic rats may reflect a more complete depletion of Sp1 following severe insulin deprivation (35–37). Goldberg et al. (38) have postulated that basal levels of O-GlcNAcylation of Sp1 may be sufficient for the nuclear translocation of Sp1. We speculate that O-GlcNAcylation of Sp1 may increase the transactivating capacity of Sp1 by increasing its interactions with other transcription factors or cofactors as observed for STAT5 and CBP/p300 (39). This mechanism may account for the increased association of SREBP-1c, SRC-1, and p300 with Sp1 demonstrated in the present study.
Based on these findings we favor a mechanistic scenario in which insulin enhances SREBP-1c gene expression in part via enhanced association of Sp1 with a multiprotein complex that contains other transcription factors (e.g. LXR, NF-Y, and SREBP-1c) and coactivators (SRC-1 and p300). We posit that enhancing the transactivating capacity of Sp1 is a key consequence of insulin signaling, but the molecular details of the underlying mechanism remain to be elucidated.
|
FOOTNOTES
|
|---|
* This work was supported in part by grants from the Office of Research and Development, Department of Veterans Affairs (to M. B. E.), from the American Heart Association (Southeast Affiliate) (to M. B. E.), the University of Tennessee Vascular Biology Center of Excellence, and by Grant DK-059368 from the National Institutes of Health (to E. A. P.). 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. 
This paper is dedicated to the memory of our wonderful colleague, Dr. Lauren Cagen, who recently passed away.
2 Senior Research Career Scientist for the Department of Veterans Affairs.
1 Recipient of a postdoctoral fellowship award from the American Heart Association, Southeast Affiliate. To whom correspondence should be addressed: Medical and Research Service, Department of Veterans Affairs Medical Center, 1030 Jefferson Ave., Memphis, TN 38104. Tel.: 901-448-5825; Fax: 901-448-7206; E-mail: xdeng{at}utmem.edu.
3 The abbreviations used are: SREBP-1c, sterol regulatory element-binding protein-1c; SRE, sterol regulatory element; LXR, liver X receptor; LXRE, liver X receptor element; ChIP, chromatin immunoprecipitation; Bt2cAMP, dibutyryl cyclic AMP; RNAi, RNA interference; siRNA, small interfering RNA; DTT, dithiothreitol; RLNE, rat liver nuclear extract; WT, wild type; TK, thymidine kinase; EMSA, electrophoretic mobility shift assay; CBP, cAMP-response element-binding protein-binding protein.
|
ACKNOWLEDGMENTS
|
|---|
We thank the following individuals who contributed essential reagents for the conduct of these studies. The plasmid pCMV-Sp1 was a gift from Dr. Guntrum Suske (Institut fur Molekularbiologie and Tumorforschung, Philipps-Universitaet Marburg, Marburg, Germany). The synthetic Sp1 reporter construct pGAGC6(194) and control construct pGAM(191) adM were provided by Dr. J. E. Kudlow (University of Alabama at Birmingham, Birmingham). Plasmid pCMV-hLXR was a gift from Dr. David J. Mangelsdorf (Department of Pharmacology, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas). We thank Poonam Kumar for expert technical assistance.
|
REFERENCES
|
|---|
- Deng, X., Cagen, L. M., Wilcox, H. G., Park, E. A., Raghow, R., and Elam, M. B. (2002) Biochem. Biophys. Res. Commun. 290, 256–262[CrossRef][Medline]
[Order article via Infotrieve]
- Cagen, L. M., Deng, X., Wilcox, H. G., Park, E. A., Raghow, R., and Elam, M. B. (2005) Biochem. J. 385, 207–216[CrossRef][Medline]
[Order article via Infotrieve]
- Yellaturu, C. R., Deng, X., Cagen, L. M., Wilcox, H. G., Park, E. A., Raghow, R., and Elam, M. B. (2005) Biochem. Biophys. Res. Commun. 332, 174–180[CrossRef][Medline]
[Order article via Infotrieve]
- Beynen, A. C., Vaartjes, W. J., and Geelen, M. J. (1979) Diabetes 28, 828–835
TOP
ABSTRACT
INTRODUCTION
