点击显示 收起
【关键词】 Transcription
The peptide hormone glucagon stimulates hepatic glucose output, and its levels in the blood are elevated in type 2 diabetes mellitus. The nuclear receptor peroxisome proliferator-activated receptor- (PPAR) has essential roles in glucose homeostasis, and thiazolidinedione PPAR agonists are clinically important antidiabetic drugs. As part of their antidiabetic effect, thiazolidinediones such as rosiglitazone have been shown to inhibit glucagon gene transcription through binding to PPAR and inhibition of the transcriptional activity of PAX6 that is required for cell-specific activation of the glucagon gene. However, how thiazolidinediones and PPAR inhibit PAX6 activity at the glucagon promoter remained unknown. After transient transfection of a glucagon promoter-reporter fusion gene into a glucagon-producing pancreatic islet -cell line, ligand-bound PPAR was found in the present study to inhibit glucagon gene transcription also after deletion of its DNA-binding domain. Like PPAR ligands, also retinoid X receptor (RXR) agonists inhibited glucagon gene transcription in a PPAR-dependent manner. In glutathione transferase pull-down assays, the ligand-bound PPAR-RXR heterodimer bound to the transactivation domain of PAX6. This interaction depended on the presence of the ligand and RXR, but it was independent of the PPAR DNA-binding domain. Chromatin immunoprecipitation experiments showed that PPAR is recruited to the PAX6-binding proximal glucagon promoter. Taken together, the results of the present study support a model in which a ligand-bound PPAR-RXR heterodimer physically interacts with promoter-bound PAX6 to inhibit glucagon gene transcription. These data define PAX6 as a novel physical target of PPAR-RXR.
Peroxisome proliferator-activated receptor- (PPAR) belongs to the superfamily of ligand-regulated nuclear hormone receptors (Desvergne and Wahli, 1999). It can be structurally subdivided into an amino-terminal, ligand-independent transactivation domain (AF-1) followed by a DNA-binding domain and a carboxyl-terminal, ligand-binding domain that contains a second, ligand-dependent transactivation surface (AF-2) (Desvergne and Wahli, 1999). PPAR binds as a heterodimer with the 9-cis-retinoic acid (9cis-RA) receptor (RXR) to response elements (PPREs) in target genes to activate transcription. Upon ligand binding and depending on the tissue-specific cofactor environment, the PPAR-RXR heterodimer recruits coactivators to stimulate promoter activity. This recruitment is dependent on ligand-induced allosteric alterations in the AF-2 helical domain (Nolte et al., 1998). In addition to transcriptional stimulation, PPAR has been shown to be also capable of repression of gene transcription (Ricote et al., 1998; Schinner et al., 2002)
PPAR is thought to be involved in a broad-range of cellular functions, including adipocyte differentiation, inflammation, blood pressure regulation, and apoptosis, and in chronic diseases such as obesity, atherosclerosis, metabolic syndrome, and cancer (Kersten et al., 2000). Of particular importance is its role in glucose homeostasis and type 2 diabetes mellitus (Kersten et al., 2000; Semple et al., 2006). Dominant-negative mutations in PPAR are associated with insulin resistance and diabetes mellitus (Semple et al., 2006), suggesting that PPAR agonists may be useful in the treatment of type 2 diabetes mellitus.
Besides naturally existing ligands such as 15-deoxy-12,14-prostaglandin J2, several synthetic ligands of PPAR as a new class of antidiabetic drugs have been synthesized, including the thiazolidinediones rosiglitazone and pioglitazone (Ikeda et al., 1990; Lehmann et al., 1995) In type 2 diabetes mellitus, thiazolidinediones lower blood glucose levels by binding to PPAR, leading to a reduction in hepatic glucose output and a decrease in insulin resistance. PPAR is expressed also in the -cells of the pancreatic islets (Dubois et al., 2000), and we recently identified the glucagon gene as a novel thiazolidinedione target gene (Schinner et al., 2002).
The peptide hormone glucagon is synthesized in the -cell of the endocrine pancreas (Unger and Orci, 1981). Glucagon stimulates glycogenolysis and gluconeogenesis and thereby increases hepatic glucose output. These actions of glucagon are opposite to those of insulin, the peptide hormone from pancreatic islet β-cells. Important in the coordination of the secretion of these two hormones is the direct inhibition by insulin of glucagon synthesis and secretion (Unger and Orci, 1981; Grzeskowiak et al., 2000). In insulin-resistant or -deficient states, glucagon synthesis and secretion become dis-inhibited, leading to hyperglucagonemia that contributes to hyperglycemia in type 2 diabetes mellitus (Unger and Orci, 1981). Thiazolidinediones and PPAR were shown to inhibit glucagon gene transcription and secretion (Schinner et al., 2002). This suggests that inhibition of glucagon gene expression may be among the multiple mechanisms through which thiazolidinediones improve glycemic control in diabetic subjects.
The pancreatic islet cell-specific enhancer sequence sequence motif within the glucagon gene promoter element G1 is required for PPAR responsiveness (Schinner et al., 2002). This sequence motif does not bind PPAR, but rather it binds the transcription factor PAX6 (St-Onge et al., 1997; Grzeskowiak et al., 2000), which is required for -cell-specific activation of the glucagon gene (Grzeskowiak et al., 2000). PAX6 is composed of an amino-terminal paired domain followed by a linker region, a homeodomain, and a carboxyl-terminal transactivation domain. The PAX6 transactivation domain recruits cofactors such as CREB-binding protein/p300, and it mediates the interaction with the general transcription machinery (Hussain and Habener, 1999). When the pancreatic islet cell-specific enhancer sequence motif within G1 was mutated into a GAL4 binding site, the expression of GAL4-PAX6 restored glucagon promoter activity and thiazolidinedione and PPAR responsiveness (Schinner et al., 2002), suggesting that PPAR in a ligand-dependent but DNA binding-independent manner inhibits PAX6 transcriptional activity on glucagon gene transcription. However, how PPAR inhibits PAX6 transcriptional activity in molecular details remained unknown. The results of the present study suggest that thiazolidinediones and PPAR inhibit glucagon gene transcription through a mechanism that involves a direct interaction of PPAR with the PAX6 transactivation domain. In contrast to the glucocorticoid receptor, where transrepression is apparently mediated by glucocorticoid receptor monomers (Reichardt et al., 1998), transrepression of PAX6 seems to be conferred by PPAR-RXR heterodimers. These data define PAX6 as a novel physical target of PPAR-RXR, which may have implications beyond the regulation of glucagon gene transcription and blood glucose control.
Plasmids. The plasmids -350GluLuc, pcDNA3-PPAR (Schinner et al., 2002), pGEX2T (GE Healthcare, Munich, Germany), pGST-PAX6(299-437) (Mikkola et al., 1999), and PPRE-Luc (Schinner et al., 2002) have been described previously. The plasmid pCMV-GF-Ptpz was purchased from Canberra Packard (Dreieich, Germany). The expression vector pcDNA3-RXR was generated by EcoRI digestion of pSG5-RXR (Heinlein et al., 1999) and cloning of the insert into EcoRI-digested pcDNA3. The plasmid pcDNA3-PPAR was used as PCR template for the generation of all described PPAR variants. PCR fragments coding for the described PPAR variants were amplified using primer pairs introducing a 5' KpnI and a 3' XbaI site for cloning into pcDNA3. All constructs were confirmed by sequencing.
In Vitro Transcription/Translation. The TNT T7-coupled reticulocyte lysate system (L4610) was purchased from Promega (Mannheim, Germany). The reactions were carried out as recommended by the manufacturer.
GST Pull-Down Assay. The GST-PAX6(299-437) fusion protein and GST were expressed in Escherichia coli BL 21 and extracted using glutathione-agarose beads (Sigma-Aldrich, Munich, Germany). The proteins were stored bound to agarose beads on ice in phosphate-buffered saline buffer containing 1 mM dithiothreitol and 1 mM PMSF. To equalize amounts of GST-PAX6(299-437) and GST used in GST pull-down assays, the amounts were estimated by Coomassie Blue staining after SDS-PAGE. For the GST-pull down assays, the protein-covered beads were washed with NETN buffer [20 mM Tris/HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, pH 8.0, and 0.5% (v/v) Nonidet-P40] containing 1 mM dithiothreitol and 1 mM PMSF. For the binding step, 30 µl of GST-PAX6(299-437) or GST beads were added to 250 µl of NETN buffer and incubated with 7.5 µl of radioactively labeled in vitro-transcribed/translated protein at 4°C overnight. The beads were washed three times with 500 µl of ice-cold NETN buffer, and after the last washing step the supernatant was carefully removed and the pellet was boiled in 30 µl of 2x Laemmli sample buffer for 10 min. After SDS-PAGE, the radioactively labeled proteins were visualized using a Fuji Bio-Imaging Analyzer IPR 1800 (Fuji Film, Tokyo, Japan). Densitometry of band intensities was done using the program TINA version 2.09g (Santa Cruz Biotechnology, Inc., Heidelberg, Germany).
Cell Culture and Transfection of DNA. The glucagon-producing pancreatic islet -cell line InR1-G9 (Schinner et al., 2002) was grown in RPMI 1640 medium supplemented with 10% fetal calf serum, 100 units/ml penicillin, and 100 µg/ml streptomycin. Cells were trypsinized and transfected in suspension by the DEAE-dextran method (Schinner et al., 2002) with 2 µg of reporter gene plasmids, and, when indicated, 1 µg of expression vector per 6-cm dish. Cotransfections were carried out with a constant amount of DNA, which was maintained by adding pBlueScript (Stratagene, La Jolla, CA). In all experiments, 0.5 µg of cytomegalovirus-green fluorescent protein (GFP) (plasmid pCMV-GFPtpz) per 6-cm dish was cotransfected to check for transfection efficiency. Twenty-four hours after transfection, cells were incubated in RPMI 1640 medium containing 0.5% bovine serum albumin and antibiotics as described above. Cell extracts (Schinner et al., 2002) were prepared 48 h after transfection. The luciferase assay was performed as described previously (Schinner et al., 2002). Green fluorescent protein was measured in the cell extracts using the FluoroCount microplate fluorometer (PerkinElmer Life and Analytical Sciences, Waltham, MA).
Chromatin Immunoprecipitation and Quantitative PCR. Two 16-cm dishes of confluent InR1-G9 cells transiently transfected with PPAR-hemagglutinin were used for one ChIP experiment. For cross-linking, formaldehyde at a final concentration of 1% was added; after 10 min at room temperature, the reaction was stopped by adding glycine at a final concentration of 125 mM. After two washing steps in phosphate-buffered saline buffer, the cells were lysed on ice in 600 µl of lysis buffer [50 mM Tris/HCl, pH 8.1, 10 mM EDTA, 1% SDS (v/v), 1 mM PMSF, 1 mM leupeptin, 1 mM aprotinin, and 1 mM pepstatin] for 10 min. DNA was sonicated using three sets of 15 pulses at medium amplitude. After sonication lysates were pre-cleared for 30 min at 4°C with 50 µl of equilibrated protein A/G agarose (Santa Cruz Biotechnology, Inc.). Ten micrograms of mouse anti-hemagglutinin antibody (Sigma-Aldrich) or IgG was added to 250 µl of the supernatant. The reaction was incubated overnight at 4°C. Then, 60 µl of protein A/G agarose slurry was added, and samples were incubated on a rotor for 2 h at 4°C. The precipitated complexes were washed twice in washing buffer 1 [20 mM Tris/HCl, pH 8.1, 150 mM NaCl, 2 mM EDTA, 0.1% SDS (v/v), and 1% Triton-X (v/v)], once in washing buffer 2 [20 mM Tris/HCl, pH 8.1, 500 mM NaCl, 2 mM EDTA, 0.1% SDS (v/v), and 1% Triton-X (v/v)], once in washing buffer 3 (20 mM Tris/HCl pH 8.1, 250 mM LiCl, 1 mM EDTA, 1% Nonidet P-40 (v/v), 1% deoxycholate), and once in TE buffer. The elution of DNA-protein complexes was performed in 10 mM Tris/HCl, pH 8.1, 1 mM EDTA, and 1% SDS. The crosslinks of DNA and protein were reversed by the addition of NaCl at a final concentration of 370 mM and by overnight incubation at 65°C. After proteinase K treatment, DNA was extracted using a PCR purification kit (QIAGEN, Hilden, Germany). The PCR was performed using a primer pair (hGlu-for 5'-CTGCCCTTT CCATTCCCAAAC-3' and hGlu-rev 5'-TTCTGCACCAGGGTGCCGTGC-3') flanking the PAX6 binding site of the hamster glucagon gene promoter G1 element (annealing temperature 57°C, 25 cycles). Quantitative (real-time) PCR was performed using SYBR Green quantitative PCR core kit (Eurogentec, Seraing, Belgium) following the manufacturer's instructions and using primers 5'-CTG CCC TTT CCA TTC CCA AAC-3' and 5'-TTC TGC ACC AGG GTG CCG TGC-3'. The reaction was carried out in a 7900HT (Applied Biosystems, Weiterstadt, Germany) real-time PCR machine under the following conditions: 95°C for 10 min, 40 cycles of 94 for 15 s, 62°C for 15 s, and 72°C for 1 min. The samples were measured in triplicates, and the average of the three threshold cycles (Ct) was used to calculate the relative amounts of template using the Ct method.
Fig. 1. Inhibition of glucagon gene transcription by PPAR and a PPAR deletion mutant lacking the PPAR DNA-binding domain. A luciferase reporter gene under the control of the rat glucagon gene promoter from -350 to +58 (plasmid -350GluLuc) was transiently transfected into InR1-G9 cells together with different expression vectors coding for PPAR variants: PPAR-(1-475), PPAR wild type; PPAR-(1-475)N, PPAR wild type extended carboxyl terminally by a nuclear localization signal; PPAR-(175-475)N, PPAR variant lacking the AF-1 and DNA-binding domain and carrying carboxyl terminally a nuclear localization signal. Rosiglitazone (30 µM) was added 24 h before harvest. Luciferase activity is expressed as percentage of the mean value of the activity measured in the control (without PPAR transfection, no rosiglitazone treatment). Values are means ± S.E.M. of three independent experiments, each done in duplicate. Top, domain structure of PPAR. The numbers indicate the amino acids.
RNA Isolation, Reverse Transcription-PCR, and mRNA Quantification. RNA was extracted from InR1-G9 cells using TRIzol reagent (Sigma-Aldrich) according to standard procedures. RNA phase was purified using RNeasy kit columns (QIAGEN), and 1 µg of total RNA was transcribed into cDNA using random hexamer primers and SuperScript II reverse transcriptase according to the manufacturer's protocol (Invitrogen, Paisley, UK). The mRNA expression was quantified by iQ SYBR Green Supermix in an iQ-thermocycler (Bio-Rad Laboratories, Munich, Germany). For each reaction, 3 ng of cDNA and the primers 5'-GCCCAAGATTTTGTGCAGTG-3' as sense and 5'-CAATGAATTCCTTTGCTGCC-3' as antisense for glucagon gene amplification and 5'-CTCATGACCACAGTCCATGC-3' as sense and 5'-CGACATGTGAGATCCACGAC-3' as antisense primer in a final concentration of 0.2 pM each for the amplification of the glyceraldehyde-3-phosphate dehydrogenase gene were used. An initial Taq-polymerase activating step at 95°C for 15 min was followed by 30 s at 95°C, 30 s at 58°C, and 30 s at 72°C. After each run, a melting curve analysis was performed to ensure amplification of correct products. The mRNA expression of glucagon was normalized to the mRNA expression of glyceraldehyde-3-phosphate dehydrogenase.
Materials. BMS 649 was kindly provided by Hinrich Gronemeyer (Institut de Génétique et Biologie Moléculaire et Cellulaire, Strasbourg, France). Rosiglitazone was kindly provided by GlaxoSmith-Kline (Welwyn Garden City, Hertfordshire, UK). Luciferin was purchased from Promega, and 9-cis-retinoic acid and GW9662 were from Sigma-Aldrich, respectively.
Fig. 2. The RXR agonists 9-cis-retinoic acid and BMS 649 inhibit glucagon gene transcription in a PPAR-dependent manner. A, scheme of the reporter gene constructs -350GluLuc and PPRE-Luc. Control elements in the 5'-flanking region of the glucagon gene are indicated. B, luciferase reporter gene constructs -350GluLuc (left) and PPRE-Luc (right) were transfected into InR1-G9 cells with or without an expression vector encoding PPAR. Cells were treated with increasing concentrations of 9-cis-retinoic for 24 h before harvest. Luciferase activity is expressed as percentage of the mean value, in each experiment, of the activity measured in the respective untreated controls. Values are means ± S.E.M. of three independent experiments, each done in duplicate. C, luciferase reporter gene constructs -350GluLuc and PPRE-Luc were transfected into InR1-G9 cells with an expression vector encoding PPAR. Cells were treated with rosiglitazone (10 µM) or 9cis-RA (10 µM) for 24 h before harvest. Luciferase activity is expressed aspercentage of the mean value, in each experiment, of the activity measured in the respective untreated controls. Values are means ± S.E.M. of three independent experiments, each done in duplicate. D, constructs -350GluLuc and PPRE-Luc were transfected into InR1-G9 cells together with an expression vector encoding PPAR. Cells were treated with BMS 649 or 9cis-RA for 24 h before harvest. Luciferase activity is expressed as percentage of the mean value, in each experiment, of the activity measured in the respective control. Values are means ± S.E.M. of three experiments, each done in duplicate. *, P < 0.005 using Student's t test. E, luciferase reporter gene construct -350GluLuc was cotransfected with PPAR, and cells were treated with increasing concentrations of the PPAR antagonist GW9662 for 24 h. Luciferase activity is expressed as percentage of the mean value, in each experiment, of the activity measured in the respective untreated controls. Values are means ± S.E.M. of three independent experiments, each done in duplicate.
PPAR Inhibited Glucagon Gene Transcription Independently of Its DNA-Binding Domain. Glucagon gene transcription was studied in the pancreatic islet -cell line InR1-G9. Primary pancreatic islet -cells express high levels of PPAR (Dubois et al., 2000), and PPAR agonists have been shown to inhibit glucagon synthesis and secretion in primary islets (Schinner et al., 2002). In contrast, InR1-G9 cells express low levels of PPAR such that activation of a PPRE-linked reporter gene and inhibition of glucagon gene transcription by PPAR agonists requires transfection of a PPAR expression plasmid (Schinner et al., 2002). Furthermore, in this cell line in the presence of overexpressed PPAR, treatment with 50 µM rosiglitazone decreased the mRNA level of endogenous glucagon by 37 ± 11.4% (P < 0.05; n = 4) (data not shown). This cell line, therefore, allows a direct assessment of the role of PPAR and PPAR variants in glucagon gene transcription. Glucagon gene transcription was studied in InR1-G9 cells using the reporter gene construct -350GluLuc (Schinner et al., 2002), which contains the rat glucagon gene promoter from -350 to +58 fused to the luciferase reporter gene. This glucagon promoter fragment is sufficient to confer tissue-specific gene expression and regulation of gene transcription by cAMP-, calcium-, protein kinase C-, and insulin-induced signaling pathways (Unger and Orci, 1981; Knepel et al., 1990; Schwaninger et al., 1993a,b; Oetjen et al., 1994; Fürstenau et al., 1997; Grzeskowiak et al., 2000). It also confers responsiveness to thiazolidinediones and PPAR (Schinner et al., 2002). To examine which domain of PPAR is involved in the repression by PPAR of glucagon gene transcription, we constructed the PPAR variant PPAR-(175-475)N. This variant lacks the DNA-binding domain, and it carries a carboxyl-terminal extension of eight amino acids representing the nuclear localization sequence of the simian virus 40 large T antigen (PKKKRKVE) (Kalderon et al., 1984). To evaluate the effect of this carboxyl-terminal nuclear localization signal on the inhibition of glucagon gene transcription by thiazolidinediones and PPAR, we fused that sequence also to PPAR full-length [construct PPAR-(1-475)N]. All PPAR variants were cotransfected with -350GluLuc into InR1-G9 cells. After treatment with rosiglitazone, all PPAR variants inhibited glucagon gene transcription by 50 to 70% (Fig. 1). These data show that the ligand-binding domain of PPAR is sufficient to inhibit glucagon gene transcription. These results thus support the previous conclusion that PPAR does not bind to the glucagon promoter to inhibit glucagon gene transcription (Schinner et al., 2002); furthermore, they shown that transrepression by PPAR is also independent of its DNA-binding domain.
RXR Agonists Inhibited Glucagon Gene Transcription in a PPAR-Dependent Manner. PPAR activates gene transcription by binding as a heterodimer with RXR to PPREs in the promoter region of target genes (Desvergne and Wahli, 1999). PPAR-RXR is a permissive heterodimer, the transcriptional activity of which can be activated by both PPAR agonists and RXR agonists (Kliewer et al., 1992). To examine whether RXR is involved in the repression of glucagon gene transcription by thiazolidinediones and PPAR, the effect of RXR agonists on glucagon gene transcription was studied. In the absence of PPAR, the RXR agonist 9-cis-retinoic acid at concentrations up to 1 µM had no effect on glucagon gene transcription (Fig. 2B, left). However, in the presence of PPAR, 9-cis-retinoic acid inhibited glucagon gene transcription (Fig. 2B, left) at concentrations that also activated a PPRE-linked reporter gene (Fig. 2B, right). The activation by 9-cis-retinoic acid of PPRE-Luc in the presence of PPAR but without co-transfection of an RXR expression plasmid indicates that InR1-G9 cells express sufficiently high levels of endogenous RXR, as has been shown previously (Schinner et al., 2002). The maximum inhibition of glucagon gene transcription by the RXR agonist 9-cis-retinoic acid in the presence of PPAR (by approximately 40%) (Fig. 2C, left) was somewhat less than the maximum inhibition by the PPAR agonist rosiglitazone (inhibition by approximately 60%) (Fig. 2C, left). Likewise, the activation of PPRE-Luc by 9-cis-retinoic acid was lower than that by rosiglitazone (Fig. 2C, right). 9-cis-Retinoic acid acts as an agonist at the RXR and retinoic acid receptor (Heyman et al., 1992; Boehm et al., 1994). Although only RXR, but not retinoic acid receptor, heterodimerizes with PPAR (Desvergne and Wahli, 1999), the effect of a specific RXR agonist, BMS 649 (Mukherjee et al., 1997), was studied. Like 9-cis-retinoic acid, also BMS 649 inhibited glucagon gene transcription and it activated a PPRE-linked reporter gene in the presence of PPAR (Fig. 2D). Increasing concentrations of the PPAR antagonist GW9662 (Seargent et al., 2004) stimulated glucagon gene transcription (Fig. 2E). These data show that, in addition to the thiazolidinedione PPAR agonists, RXR agonists also inhibit glucagon gene transcription in a PPAR-dependent manner, consistent with the view that inhibition of glucagon gene transcription is mediated by a PPAR-RXR heterodimer.
A Direct Protein-Protein Interaction between PPAR and the PAX6 Transactivation Domain Depending on the Presence of RXR and the Thiazolidinedione Rosiglitazone Was Revealed by GST Pull-Down Assays. To test whether the inhibition of glucagon gene transcription by thiazolidinediones and PPAR may involve a physical interaction between PPAR and PAX6, the GST pull-down assay was used. The PAX6 transactivation domain was fused to GST, expressed in E. coli, immobilized on glutathione-coated beads, and incubated with PPAR, RXR, or both, that had been radioactively labeled with [35S]methionine by in vitro transcription/translation. After the binding reaction, the agarose beads were washed and retained proteins were analyzed by SDS polyacrylamide gel electrophoresis. Compared with GST, which served as a negative control, the GST-PAX6 transactivation domain retained the cotranscribed/cotranslated PPAR-RXR heterodimer (Fig. 3, A, C, and D), whereas PPAR alone or RXR alone were not sufficiently retained (Fig. 3A). The interaction between the PAX6 transactivation domain and the PPAR-RXR heterodimer was further enhanced by the PPAR ligand rosiglitazone (Fig. 3B). In addition, in the presence of rosiglitazone the interaction depended on RXR because the signal intensity markedly decreased when the PAX6 transactivation domain was incubated with PPAR alone (Fig. 3C, compare first lane with third lane). In addition, the PPAR antagonist GW9662 reduced the interaction between the PAX6 transactivation domain and PPAR-(245-475)-RXR heterodimer (Fig. 3D). In contrast, the ligand-induced interaction did not depend on the PPAR DNA-binding domain, because upon cotranscription/cotranslation, the PPAR-(245-475)-RXR heterodimer bound to the PAX6 transactivation domain (Fig. 3D).
Fig. 3. Protein-protein interaction between PPAR-RXR and the transactivation domain of PAX6 in GST pull-down assays. A, in vitro-transcribed/translated and [35S]methionine-labeled PPAR, RXR, and PPAR/RXR were incubated with GST-PAX6-TAD immobilized on glutathione agarose beads. GST immobilized on glutathione agarose beads was used as a control. The resulting complexes were resolved by SDS-PAGE and visualized by filmless autoradiographic analysis (top). The bands were quantified by densitometry. Optical density is expressed as percentage of the optical density measured in the PPAR-RXR plus GST-PAX6-TAD group. Values are means ± S.E.M. of four independent experiments. PPAR and RXR are of similar size, and they are not clearly separated on the gel used. B, PPAR and RXR were cotranscribed/cotranslated and labeled with [35S]methionine in vitro. PPAR/RXR was incubated with GST-PAX6-TAD immobilized on glutathione agarose beads. GST immobilized on glutathione agarose beads was used as a control. The incubations were performed in the presence or absence of 50 µM rosiglitazone. The resulting complexes were resolved by SDS-PAGE and visualized by filmless autoradiographic analysis (top). The bands were quantified by densitometry. Optical density is expressed as percentage of the optical density measured in the GST-PAX6-TAD group (no rosiglitazone). Values are means ± S.E.M. of five experiments. Lanes 1 and 2 show 5% input. C, in vitro-transcribed/translated and [35S]methionine-labeled PPAR and PPAR/RXR were incubated with GST-PAX6-TAD immobilized on glutathione agarose beads. GST immobilized on glutathione agarose beads was used as a control. The incubations were performed in the presence or absence of 50 µM rosiglitazone. The resulting complexes were resolved by SDS-PAGE and visualized by filmless autoradiographic analysis. The 5% input of PPAR/RXR (lane 1) and PPAR (lanes 2 and 3) is shown. D, PPAR-(245-475), containing only the ligand-binding domain, and RXR were cotranscribed/cotranslated and labeled with [35S]methionine in vitro. PPAR-(245-475)/RXR were incubated in the presence of 50 µM rosiglitazone or 50 µM GW9662 with GST-PAX6-TAD immobilized on glutathione agarose beads and GST-covered beads for control. The resulting complexes were resolved by SDS-PAGE and visualized by filmless autoradiographic analysis. 1, 5% input. Top band, RXR; bottom band, PPAR-(245-475).
Recruitment of PPAR to the Glucagon Gene Promoter Was Revealed by Chromatin Immunoprecipitation. If the PPAR-RXR heterodimer interacts with PAX6 also in vivo, PPAR should be recruited through PAX6 to the glucagon promoter, in spite of the fact that PPAR does not bind to glucagon promoter sequences. This was studied by chromatin immunoprecipitation. After PPAR transfection and rosiglitazone treatment of InR1-G9 cells, chromatin-bound proteins were cross-linked by formaldehyde. After cell lysis and DNA fragmentation by sonication, PPAR-bound DNA was immunoprecipitated and amplified by PCR using a primer pair flanking the PAX6 binding site in the glucagon promoter G1 element. Compared with an immunoprecipitation using unspecific immunoglobulins, serving as a negative control, antibodies against PPAR precipitated a glucagon promoter fragment that includes the G1 PAX6 binding site (Fig. 4A), indicating recruitment of PPAR to the glucagon gene promoter in vivo. To investigate whether rosiglitazone enhanced the recruitment of PPAR to the glucagon gene promoter in vivo, the rat glucagon promoter and the expression for PPAR were cotransfected, cells were treated with rosiglitazone and a ChIP assay followed by quantitative PCR was performed. In the presence of the PPAR agonist, the recruitment of PPAR to the glucagon promoter was 5.8-fold enhanced (Fig. 4B). In addition, rosiglitazone enhanced the recruitment of PPAR to the genomic (hamster) glucagon promoter 5.9- and 1.9-fold, respectively (data not shown), indicating that rosiglitazone stimulated the recruitment of PPAR to the glucagon gene promoter in vivo.
Fig. 4. Recruitment of PPAR to the PAX6-binding proximal glucagon promoter as revealed by chromatin immunoprecipitation. A, InR1-G9 cells were transfected with PPAR and treated with 50 µM rosiglitazone for 24 h. After cell lysis and DNA fragmentation by sonication, PPAR-bound DNA was immunoprecipitated and amplified by PCR using a primer pair flanking the PAX6 binding site in the glucagon promoter G1 element. Chromatin was precipitated with unspecific IgG antibody for control. A positive control for the fragment size is also shown. B, effect of rosiglitazone on the ChIP of PPAR/PAX6 was quantified using quantitative PCR. InR1-G9 cells were transfected with -350GluLuc and PPAR, and they were treated as described above. ChIP signal activity is expressed as percentage of the mean value, in each experiment, of the activity measured in the untreated control. Values are means ± S.D. of two independent experiments.
Thiazolidinediones are a new class of oral antidiabetic drugs. Through binding to PPAR, they lower hepatic glucose output and reduce insulin resistance in type 2 diabetes mellitus, emphasizing the pivotal role of PPAR in blood glucose control (Semple et al., 2006). As part of their antidiabetic effect, thiazolidinediones have been shown to inhibit glucagon gene transcription by inhibiting the transcriptional activity of PAX6 (Schinner et al., 2002), which is required for -cell-specific activation of the glucagon gene (Grzeskowiak et al., 2000). The present study now suggests a mechanism through which thiazolidinediones and PPAR inhibit PAX6 activity and thus glucagon gene transcription in pancreatic islet -cells. The results of the present study are consistent with a model in which a ligand-bound PPAR-RXR heterodimer physically interacts with promoter-bound PAX6 (Fig. 5). This interaction is independent of the PPAR DNA-binding domain. These data thereby define PAX6 as a novel physical target of PPAR.
Fig. 5. Model for the mechanism of rosiglitazone-induced repression of the glucagon gene. See Discussion for details.
It has been shown previously that PPAR inhibits glucagon gene transcription without binding to the glucagon promoter (Schinner et al., 2002). This view is further supported by the finding in the present study that PPAR inhibits glucagon gene transcription also after deletion of its DNA-binding domain. The inhibition by PPAR does, however, seem to depend on heterodimerization with RXR. PPAR and RXR are known to heterodimerize in solution through their ligand-binding domains (Kliewer et al., 1992; Gampe et al., 2000). An intriguing aspect of RXR heterodimers is that some are permissive for activation by RXR ligands, whereas others are not. The PPAR-RXR heterodimer falls into the permissive category of RXR heterodimers (Kliewer et al., 1992). Because of the permissive nature of the PPAR-RXR heterodimer, RXR agonists have many of the same effects as do PPAR agonists, and they have been shown to possess antidiabetic activity in mouse models of type 2 diabetes mellitus (Mukherjee et al., 1997). The asymmetric interactions between the AF-2 of PPAR and helices 7 and 10 of RXR, as revealed by the crystal structure of the PPAR-RXR ligand-binding domains (Gampe et al., 2000), suggest a structural basis for permissiveness. The net effect of these interactions may be the stabilization of the PPAR AF-2 helix in an active conformation even in the absence of a bound PPAR agonist (Gampe et al., 2000). Because in the present study RXR agonists inhibited glucagon gene transcription in a strictly PPAR-dependent manner, the results indicate that a PPAR-RXR heterodimer confers inhibition of glucagon gene transcription. They thereby suggest that also the inhibition by thiazolidinediones is mediated by PPAR-RXR heterodimers.
Several distinct underlying mechanisms for transrepression by nuclear receptors have been described. Nuclear receptors were shown to inhibit activator protein-1-mediated transcription of target genes by competition for limiting amounts of the coactivators CREB-binding protein/p300 (Kamei et al., 1996). Alternatively, nuclear receptors may interfere with signaling pathways. For example, glucocorticoids induce the disassembly of c-Jun NH2-terminal kinase from mitogen-activated protein kinase kinase 7 by promoting its association with the glucocorticoid receptor, leading to the inhibition of c-Jun NH2-terminal kinase-dependent transcription factors and their target genes (Bruna et al., 2003). A third mechanism of transrepression is through direct interaction with transcription factors, as exemplified by the interaction between the glucocorticoid receptor and the p65 subunit of NF-B (Garside et al., 2004). Although the data of the present study do not exclude additional mechanisms, they strongly suggest that the mechanism through which PPAR inhibits PAX6 activity and thus glucagon gene transcription involves a protein-protein interaction with PAX6. The ligand-bound PPAR-RXR heterodimer was found in the GST pull-down assay to bind to the transactivation domain of PAX6. This binding depended on the presence of rosiglitazone and RXR, but it was independent of the PPAR DNA-binding domain. The requirements for the binding of PPAR to PAX6 are thus similar to those for the inhibition of glucagon gene transcription by PPAR (Schinner et al., 2002; this study), suggesting that this binding may underlie the repression of transcription. The chromatin immunoprecipitation assay showed that PPAR becomes recruited to the PAX6-responsive region of the proximal glucagon promoter, although PPAR does not bind to these sequences (Schinner et al., 2002), indicating that PPAR associates with PAX6 also in vivo. Furthermore, that PPAR associates with promoter-bound PAX6. Taken together, the results of the present study thus support a model in which a ligand-bound PPAR-RXR heterodimer physically interacts with promoter-bound PAX6 to inhibit glucagon gene transcription (Fig. 5).
This suggested model contrasts with defined mechanisms of interaction of nuclear receptors with promoter-bound transcription factors. Thus, NF-B activity is inhibited by binding of the p65 subunit of NF-B to the DNA-binding domain of the glucocorticoid receptor (Chung et al., 2000; Nissen and Yamamoto, 2000), whereas PPAR interacts with PAX6 and inhibits glucagon gene transcription independently of its DNA-binding domain. Furthermore, although PPAR acts at PAX6 and the glucagon promoter as a PPAR-RXR heterodimer, transrepression by the glucocorticoid receptor is apparently mediated by glucocorticoid receptor monomers (Reichardt et al., 1998). In fact, transactivation-defective mutants of the glucocorticoid receptor, which cannot dimerize or bind DNA, are fully competent in transrepression (Reichardt et al., 1998). Likewise, PPAR has been shown to bind the p65 subunit of NF-B as a monomer (i.e., in the absence of RXR) and in a ligand-independent manner (Chung et al., 2000). This interaction seems to be stabilized at the mouse inducible nitric-oxide synthase promoter in the RAW264.7 macrophage cell line by multiple cofactors (Pascual et al., 2005). This multiplicity of mechanisms indicates that distinct interaction surfaces are used by nuclear receptors depending on both the specific nuclear receptor, and, for a given receptor, the target transcription factor. The inhibition by the PPAR-RXR heterodimer of transforming growth factor-β1 gene transcription in the L929 fibroblast cell line, in contrast, is not through transrepression but it seems to be indirectly mediated (Lee et al., 2006). PPAR-RXR has been shown to bind a PPRE in the promoter of the gene encoding the 3'-phosphatase phosphatase and tensin homolog deleted from chromosome 10 and to induce phosphatase and tensin homolog deleted from chromosome 10 expression, which through inhibition of phosphoinositide 3-kinase leads to a diminished phosphorylation and thus inactivation of the transcription factor Zf9 that is required for transforming growth factor-β1 gene transcription (Lee et al., 2006).
The present study identifies PAX6 as a novel physical target of PPAR-RXR. This may have implications beyond the regulation of glucagon gene transcription in pancreatic islet cells. PPAR and PAX6 are coexpressed also in pancreatic islet β-cells (Callaerts et al., 1997; St-Onge et al., 1997; Dubois et al., 2000). PAX6 expression in the pancreas is initiated in the pancreatic progenitors, concomitant with the onset of hormone expression, after which the expression persists in all endocrine cells throughout development and in the adult pancreas (Callaerts et al., 1997; St-Onge et al., 1997). The phenotype of PAX6 mutants suggests a specific role for this gene in the differentiation of -cells and for the normal numbers and hormone expression of other endocrine cell types, including β-cells. In contrast, the activation of PPAR has been shown to reduce the proliferation of β-cells (Rosen et al., 2003) in addition to inducing the expression of glucokinase (Kim et al., 2002) and the glucose transporter GLUT2 in β-cells (Kim et al., 2000). An involvement of PAX6 in these actions remains to be examined. A PAX6-PPAR/RXR interaction could have implications also beyond blood glucose control by thiazolidinedione oral antidiabetic drugs in type 2 diabetes mellitus. The transcription factor PAX6 is highly conserved in evolution, and it is considered to be a master gene for the development of the eye in species from Drosophila melanogaster to human (Callaerts et al., 1997). It plays an important role also in the development of the nose and brain (Callaerts et al., 1997). In addition to the pancreas, PAX6 is accordingly expressed in the eye and nasal epithelium. It is also expressed in the developing and adult brain (Callaerts et al., 1997) where its expression exhibits some temporal and spatial overlap with the expression of PPAR (Zhao et al., 2006), raising the possibility of a PPAR-PAX6 interaction also in the brain. Noteworthy, the number of PAX6-expressing neurons in the brain is increased by cerebral ischemia, when the activation of PPAR is known to promote neuroprotection (Zhao et al., 2006).
Acknowledgements
BMS 649 was kindly provided by Hinrich Gronemeyer. We thank Stephan Herzig (Heidelberg, Germany) for advice on the chromatin immunoprecipitation assay.
ABBREVIATIONS: PPAR, peroxisome proliferator-activated receptor; AF, activation function; 9cis-RA, 9-cis-retinoic acid; RXR, retinoid X receptor; PPRE, peroxisome proliferator-activated receptor response element; PCR, polymerase chain reaction; GST, glutathione transferase; PMSF, phenylmethylsulfonyl fluoride; PAGE, polyacrylamide gel electrophoresis; GFP, green fluorescent protein; ChIP, chromatin immunoprecipitation; NF-B, nuclear factor-B; GW9662, 2-chloro-5-nitrobenzanilide; BMS 649, 4-[2-(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)-(1,3)dioxolan-2-yl]benzoic acid.
【参考文献】
Boehm MF, Zhang L, Badea BA, White SK, Mais DE, Berger E, Suto CM, Goldman ME, and Heyman RA (1994) Synthesis and structure-activity relationships of novel retinoid X receptor-selective retinoids. J Med Chem 37: 2930-2941.
Bruna A, Nicolas M, Munoz A, Kyriakis JM, and Caelles C (2003) Glucocorticoid receptor-JNK interaction mediates inhibition of the JNK pathway by glucocorticoids. EMBO J 22: 6035-6044.
Callaerts P, Halder G, and Gehring WJ (1997) PAX-6 in development and evolution. Annu Rev Neurosci 20: 483-532.
Chung SW, Kang BY, Kim SH, Pak YK, Cho D, Trinchieri G, and Kim TS (2000) Oxidized low density lipoprotein inhibits interleukin-12 production in lipopolysac-charide-activated mouse macrophages via direct interactions between peroxisome proliferator-activated receptor-gamma and nuclear factor-kappa B. J Biol Chem 275: 32681-32687.[Abstract/Free Full Text]
Desvergne B and Wahli W (1999) Peroxisome proliferator-activated receptors: nuclear control of metabolism. Endocr Rev 20: 649-688.[Abstract/Free Full Text]
Dubois M, Pattou F, Kerr-Conte J, Gmyr V, Vandewalle B, Desreumaux P, Auwerx J, Schoonjans K, and Lefebvre J (2000) Expression of peroxisome proliferator-activated receptor gamma (PPARgamma) in normal human pancreatic islet cells. Diabetologia 43: 1165-1169.
Fürstenau U, Schwaninger M, Blume R, Kennerknecht I, and Knepel W (1997) Characterization of a novel protein kinase C response element in the glucagon gene. Mol Cell Biol 17: 1805-1816.
Gampe RT Jr., Montana VG, Lambert MH, Miller AB, Bledsoe RK, Milburn MV, Kliewer SA, Willson TM, and Xu HE (2000) Asymmetry in the PPARgamma/RXRalpha crystal structure reveals the molecular basis of heterodimerization among nuclear receptors. Mol Cell 5: 545-555.
Garside H, Stevens A, Farrow S, Normand C, Houle B, Berry A, Maschera B, and Ray D (2004) Glucocorticoid ligands specify different interactions with NF-Bby allosteric effects on the glucocorticoid receptor DNA binding domain. J Biol Chem 279: 50050-50059.[Abstract/Free Full Text]
Grzeskowiak R, Amin J, Oetjen E, and Knepel W (2000) Insulin responsiveness of the glucagon gene conferred by interactions between proximal promoter and more distal enhancer-like elements involving the paired-domain transcription factor Pax6. J Biol Chem 275: 30037-30045.[Abstract/Free Full Text]
Heinlein CA, Ting HJ, Yeh S, and Chang C (1999) Identification of ARA70 as a ligand-enhanced coactivator for the peroxisome proliferator-activated receptor gamma. J Biol Chem 274: 16147-16152.[Abstract/Free Full Text]
Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM, and Thaller C (1992) 9-cis retinoic acid is a high affinity ligand for the retinoid X receptor. Cell 68: 397-406.
Hussain MA and Habener JF (1999) Glucagon gene transcription activation mediated by synergistic interactions of pax-6 and cdx-2 with the p300 co-activator. J Biol Chem 274: 28950-28957.[Abstract/Free Full Text]
Ikeda H, Taketomi S, Sugiyama Y, Shimura Y, Sohada T, Meguro K, and Fujita T (1990) Effects of pioglitazone on glucose and lipid metabolism in normal and insulin resistant animals. Arzneimittelforschung 40: 156-162.
Kalderon D, Roberts BL, Richardson WD, and Smith AE (1984) A short amino acid sequence able to specify nuclear location. Cell 39: 499-509.
Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin SC, Heyman RA, Rose DW, Glass CK, et al. (1996) A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85: 403-414.
Kersten S, Desvergne B, and Wahli W (2000). Roles of PPARs in health and disease. Nature 405: 421-424.
Kim HI, Cha JY, Kim SY, Kim JW, Roh KJ, Seong JK, Lee NT, Choi KY, Kim KS, and Ahn YH (2002) Peroxisomal proliferator-activated receptor-gamma upregulates glucokinase gene expression in beta-cells. Diabetes 51: 676-685.[Abstract/Free Full Text]
Kim HI, Kim JW, Kim SH, Cha JY, Kim KS, and Ahn YH (2000) Identification and functional characterization of the peroxisomal proliferator response element in rat GLUT2 promoter. Diabetes 49: 1517-15124.
Kliewer SA, Umesono K, Mangelsdorf DJ, and Evans RM (1992). Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signalling. Nature 355: 446-449.
Knepel W, Chafitz J, and Habener JF (1990) Transcriptional activation of the rat glucagon gene by the cyclic AMP-responsive element in pancreatic islet cells. Mol Cell Biol 10: 6799-6804.[Abstract/Free Full Text]
Lee SJ, Yang EK, and Kim SG (2006) Peroxisome proliferator-activated receptor-{gamma} and retinoic acid X receptor {alpha} represses the TGFbeta1 gene via PTEN-mediated p70 ribosomal S6 kinase-1 inhibition: role for Zf9 dephosphorylation. Mol Pharmacol 70: 415-425.[Abstract/Free Full Text]
Lehmann JM, Moore LB, Smith-Oliver TA, Wilkison WO, Willson TM, and Kliewer SA (1995) An antidiabetic thiazolidinedione is a high affinity ligand for peroxisome proliferator-activated receptor gamma (PPAR gamma). J Biol Chem 270: 12953-12956.[Abstract/Free Full Text]
Mikkola I, Bruun JA, Bjorkoy G, Holm T, and Johansen T (1999) Phosphorylation of the transactivation domain of Pax6 by extracellular signal-regulated kinase and p38 mitogen-activated protein kinase. J Biol Chem 274: 15115-15126.[Abstract/Free Full Text]
Mukherjee R, Davies PJ, Crombie DL, Bischff ED, Cesario RM, Jow L, Hamann LG, Boehm MF, Mondon CE, Nadzan AM, et al. (1997). Sensitization of diabetic and obese mice to insulin by retinoid X receptor agonists. Nature 386: 407-410.
Nissen RM and Yamamoto KR (2000) The glucocorticoid receptor inhibits NFkappaB by interfering with serine-2 phosphorylation of the RNA polymerase II carboxy-terminal domain. Genes Dev 14: 2314-2329.[Abstract/Free Full Text]
Nolte RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK, and Milburn MV (1998). Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-gamma. Nature 395: 137-143.
Oetjen E, Diedrich T, Eggers A, Eckert B, and Knepel W (1994) Distinct properties of the cAMP-responsive element of the rat insulin I gene. J Biol Chem 269: 27036-27044.[Abstract/Free Full Text]
Pascual G, Fong AL, Ogawa S, Gamliel A, Li AC, Perissi V, Rose DW, Willson TM, Rosenfeld MG, and Glass CK (2005). A SUMOylation-dependent pathway mediates transrepression of inflammatory response genes by PPAR-gamma. Nature 437: 759-763.
Reichardt HM, Kaestner KH, Tuckermann J, Kretz O, Wessely O, Bock R, Gass P, Schmid W, Herrlich P, Angel P, et al. (1998) DNA binding of the glucocorticoid receptor is not essential for survival. Cell 93: 531-541.
Ricote M, Li AC, Willson TM, Kelly CJ, and Glass CK (1998). The peroxisome proliferator-activated receptor-gamma is a negative regulator of macrophage activation. Nature 391: 79-82.
Rosen ED, Kulkarni RN, Sarraf P, Ozcan U, Okada T, Hsu CH, Eisenman D, Magnuson MA, Gonzalez FJ, Kahn CR, et al. (2003) Targeted elimination of peroxisome proliferator-activated receptor gamma in beta cells leads to abnormalities in islet mass without compromising glucose homeostasis. Mol Cell Biol 23: 7222-7229.[Abstract/Free Full Text]
Schinner S, Dellas C, Schroeder M, Heinlein CA, Chang C, Fischer J, and Knepel W (2002) Repression of glucagon gene transcription by peroxisome proliferator-activated receptor gamma through inhibition of Pax6 transcriptional activity. J Biol Chem 277: 1941-1948.[Abstract/Free Full Text]
Schwaninger M, Blume R, Oetjen E, Lux G, and Knepel W (1993a) Inhibition of cAMP-responsive element-mediated gene transcription by cyclosporin A and FK506 after membrane depolarization. J Biol Chem 268: 23111-23115.[Abstract/Free Full Text]
Schwaninger M, Lux G, Blume R, Oetjen E, Hidaka H, and Knepel W (1993b) Membrane depolarization and calcium influx induce glucagon gene transcription in pancreatic islet cells through the cyclic AMP-responsive element. J Biol Chem 268: 5168-5177.[Abstract/Free Full Text]
Seargent JM, Yates EA, and Gill JH (2004) GW9662, a potent antagonist of PPAR-gamma, inhibits growth of breast tumour cells and promotes the anticancer effects of the PPARgamma agonist rosiglitazone, independently of PPARgamma activation. Br J Pharmacol 143: 933-937.
Semple RK, Chatterjee VK, and O'Rahilly S (2006) PPAR gamma and human metabolic disease. J Clin Invest 116: 581-589.
St-Onge L, Sosa-Pineda B, Chowdhury K, Mansouris A, and Gruss P (1997). Pax6 is required for differentiation of glucagon-producing alpha-cells in mouse pancreas. Nature 387: 406-409.
Unger RH and Orci L (1981) Glucagon and the A cell: physiology and pathophysiology (first two parts). N Engl J Med 304: 1518-1524.
Zhao Y, Patzer A, Herdegen T, Gohlke P, and Culman J (2006) Activation of cerebral peroxisome proliferator-activated receptors gamma promotes neuroprotection by attenuation of neuronal cyclooxygenase-2 overexpression after focal cerebral ischemia in rats. FASEB J 20: 1162-1175.[Abstract/Free Full Text]
作者单位：Molecular Pharmacology (R.K., F.F., K.L., M.S., K.R., C.D., E.O., W.K.), Clinical Pharmacology (M.V.T.), and Pediatric Cardiology and Intensive Care Medicine (T.Q.), University of G?ttingen, G?ttingen, Germany