JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Delerive, P.
Right arrow Articles by Staels, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Delerive, P.
Right arrow Articles by Staels, B.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

J Biol Chem, Vol. 274, Issue 45, 32048-32054, November 5, 1999


Peroxisome Proliferator-activated Receptor alpha  Negatively Regulates the Vascular Inflammatory Gene Response by Negative Cross-talk with Transcription Factors NF-kappa B and AP-1*

Philippe DeleriveDagger §, Karolien De Bosscherparallel , Sandrine Besnard**, Wim Vanden Berghe, Jeffrey M. PetersDagger Dagger , Frank J. GonzalezDagger Dagger , Jean-Charles FruchartDagger , Alain Tedgui**, Guy Haegeman§§, and Bart StaelsDagger ¶¶

From the Dagger  INSERM U325, Département d'Athérosclérose, Institut Pasteur de Lille, 1 rue Pr. Calmette 59019 Lille, and Faculté de Pharmacie, Université de Lille II, 59000 Lille, France, the  Laboratory of Molecular Biology, University of Gent and VIB, K. L. Ledeganckstraat 35, B-9000 Gent, Belgium, the ** INSERM U141, 41 Bd. de la Chapelle, 75745 Paris Cédex 10, France, and the Dagger Dagger  Department of Molecular Carcinogenesis, NCI, National Institutes of Health, Bethesda, Maryland 20892

    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Interleukin-6 (IL-6) is a pleiotropic cytokine, whose plasma levels are elevated in inflammatory diseases such as atherosclerosis. We have previously reported that peroxisome proliferator-activated receptor alpha  (PPARalpha ) ligands (fibrates) lower elevated plasma concentrations of IL-6 in patients with atherosclerosis and inhibit IL-1-stimulated IL-6 secretion by human aortic smooth muscle cells (SMC). Here, we show that aortic explants isolated from PPARalpha -null mice display an exacerbated response to inflammatory stimuli, such as lipopolysaccharide (LPS), as demonstrated by increased IL-6 secretion. Furthermore, fibrate treatment represses IL-6 mRNA levels in LPS-stimulated aortas of PPARalpha wild-type, but not of PPARalpha -null mice, demonstrating a role for PPARalpha in this fibrate action. In human aortic SMC, fibrates inhibit IL-1-induced IL-6 gene expression. Furthermore, activation of PPARalpha represses both c-Jun- and p65-induced transcription of the human IL-6 promoter. Transcriptional interference between PPARalpha and both c-Jun and p65 occurs reciprocally, since c-Jun and p65 also inhibit PPARalpha -mediated activation of a PPAR response element-driven promoter. This transcriptional interference occurs independent of the promoter context as demonstrated by cotransfection experiments using PPARalpha , p65, and c-Jun Gal4 chimeras. Overexpression of the transcriptional coactivator cAMP-responsive element-binding protein-binding protein (CBP) does not relieve PPARalpha -mediated transcriptional repression of p65 and c-Jun. Finally, glutathione S-transferase pull-down experiments demonstrate that PPARalpha physically interacts with c-Jun, p65, and CBP. Altogether these data indicate that fibrates inhibit the vascular inflammatory response via PPARalpha by interfering with the NF-kappa B and AP-1 transactivation capacity involving direct protein-protein interaction with p65 and c-Jun.

    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Atherosclerosis is a complex vascular disease characterized by endothelial injury, monocyte infiltration in the subendothelial space, followed by differentiation into macrophages followed by cholesterol deposition. This further results in foam cell formation, smooth muscle cell (SMC)1 proliferation, and migration from the media to the intima (1). The presence of macrophages, T lymphocytes, as well as numerous cytokines in the atherosclerotic lesion suggests an important immunological component in the pathogenesis of atherosclerosis (1, 2). Interleukin-6 (IL-6), a cytokine, which has been detected in human and rabbit atherosclerotic lesions (3, 4), is secreted by endothelial cells, monocytes/macrophages, and SMC (2). IL-6 controls macrophage and T cell activation, SMC proliferation, and migration and is a major regulator of the acute phase response (5). Even though IL-6 might also possess anti-inflammatory properties (6, 7), this cytokine is considered as a good marker of vascular inflammation.

Peroxisome proliferator-activated receptors (PPARs) belong to the superfamily of nuclear receptors which are ligand-activated transcription factors (8). PPARs regulate gene expression by binding with their heterodimeric partner retinoid X receptor to specific PPAR-response elements (PPREs) (9). Three different PPAR subtypes have been identified: PPARalpha , PPARbeta (NUC-1 or PPARdelta ), and PPARgamma . Fatty acid derivatives and eicosanoids were identified as natural ligands for PPARs (10-14). Furthermore, fibrates are synthetic ligands for PPARalpha (10), which mediates the lipid-lowering activity of these drugs (15). Several indirect observations suggest that fibrates may also exert a direct anti-atherogenic activity at the level of the vascular wall, which occurs independently of their lipid-lowering activity. First, treatment of cholesterol-fed rabbits with the PPARalpha ligand fenofibrate decreases atherosclerotic plaque formation in the thoracic aorta, in the absence of any lowering of plasma lipid levels (16). Second, in a number of intervention trials, such as BECAIT and LOCAT, fibrate treatment slows the progression of coronary atherosclerosis without significantly affecting plasma atherogenic lipoprotein concentrations (17, 18). Finally, Devchand et al. (11) showed that absence of PPARalpha expression in mice prolonged the inflammatory response. We and others (19, 20) reported that fibrates decrease plasma concentrations of inflammatory cytokines, such as IL-6 and tumor necrosis factor alpha , in human patients with angiographically established atherosclerosis and prevent the induction of IL-6 production by IL-1beta in SMC.

Although much is known about gene activation by PPARs acting via PPREs, less information exists about the mechanisms of negative gene modulation by PPARs. Recently, PPARs have been suggested to exert anti-inflammatory activities by antagonizing the AP-1, NF-kappa B, and STAT pathways in macrophages and SMC (19, 21, 22). To address the physiological role of PPARalpha in the regulation of the inflammatory response at the level of the vascular wall, studies were performed using PPARalpha -null mice as a model. Our results demonstrate, using IL-6 secretion as an inflammatory marker, that aortas from PPARalpha -null mice display an exacerbated inflammatory response to LPS. We next carried out experiments to characterize the molecular mechanisms implicated in the down-modulation of IL-6 gene promoter activity by PPARalpha activators.

    MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Cell Culture and Chemical Reagents-- Human aortic SMC (Cascade Biologics, Portland, OR) were cultured in SMC basal medium containing 5% SMC growth supplement (Cascade Biologics). Cells from passages 5 to 8 were used for the experiments. COS-1 cells (ATCC, Manassas, VA) were grown in Dulbecco's modified Eagle's medium supplemented with 2 mM glutamine and 10% (v/v) fetal calf serum (FCS) in a 5% CO2 humidified atmosphere at 37 °C.

Wy-14643 was from Chemsyn, Lenexa, KS; fenofibric acid from Laboratoires Fournier, Dijon, France; IL-1beta from Genzyme, Cambridge, MD; and LPS from Sigma, Saint Quentin, France.

RNA Analysis-- RNA preparation and Northern blot hybridizations were performed as described previously (23). Human IL-6 (24) and 36B4 cDNA fragments were used as probes.

Ex Vivo Studies-- Male PPARalpha -/- (F6 homozygotes; SV/129 genetic background; 10 weeks old) (25) and male PPARalpha +/+ (SV/129 genetic background; 10 weeks old) were sacrificed by pentobarbital injection. The descending aorta was quickly dissected, excised, and cut into two segments, which were cultured in standard medium with or without LPS (10 µg/ml) for 24 h. IL-6 concentrations were measured as described previously (26) and normalized to cellular DNA content determined by a fluorimetric assay (27).

Plasmids-- The pRSV-p65, p(kappa B)3-luc+ (Stratagene), p(AP-1)3-luc+ (Stratagene), pSG5-hPPARalpha , and PPRE-containing reporter plasmids (J6-TK-Luc) were described previously (19). The truncated form of hPPARalpha (pSG5-hPPARalpha Delta LBD) lacking the PPARalpha ligand binding domain was constructed by PCR as described previously.2 The c-Jun and c-Fos expression plasmids under the control of the RSV promoter were provided by Drs. Bakiri and Yaniv (Institut Pasteur, Paris, France). The wild-type human IL-6 promoter construct p1168hu.IL6P-luc+ and the corresponding mutants of the NF-kappa B or AP-1 response elements, as well as the plasmid p(GAL4)250hu.IL6P-luc+, containing two sites for the yeast transcription factor Gal4 in front of the IL-6 promoter TATA box containing minimal promoter, have been described previously (28). The expression plasmids pGal4, pGal4-p65, and pGal4-c-Jun containing the DNA-binding domain of the yeast Gal4 protein either alone or fused to the full-length p65 or c-Jun polypeptides were kindly provided by Dr. Schmitz (German Cancer Research Center, Heidelberg, Germany) (29) and Dr. Kouzarides (Institute of Cancer and Developmental Biology, Cambridge, United Kingdom) (30, 31) respectively. The plasmid pGal5-TK-pGL3 was obtained by inserting five copies of the Gal4 DNA binding site in front of the thymidine kinase promoter of the pTK-pGL3 plasmid. The plasmid pGal4-hPPARalpha was constructed by PCR-amplifying the hPPARalpha DEF domains (aa 166-467) using pSG5-hPPARalpha as template. The resulting PCR product was cloned in pBD-Gal4 (Stratagene, La Jolla, CA) and the chimera subsequently subcloned into the pCDNA3 vector. pGal4p65286-551 (24) was digested with EcoRI/BamHI followed by BsaHI and the insert was subcloned in the EcoRI/AccI sites of pGex5×1 vector (Amersham Pharmacia Biotech) yielding pGexp65286-551. pGexp6512-317, pGexc-Jun1-79, and pGexCBP451-828 were generously provided by Dr. Hay (32), Dr. Karin, and Dr. Goodman, respectively. pGexCBP1-213 and pGexCBP1891-2442 were constructed by PCR amplification using CMVbeta CBP (kind gift of Dr. Eckner) as template and subsequent digestion with BamHI/NotI and XbaI/NotI, respectively, and subcloning into pGex4T2 (Amersham Pharmacia Biotech).

Transient Transfection Assays-- Hek293T cells and COS-1 cells, grown to 50-60% confluence in DMEM supplemented with 10% FCS, were transiently transfected using the calcium phosphate coprecipitation technique with reporter and expression plasmids, as stated in the figure's legend. Phosphoglycerate kinase-beta -galactosidase expression plasmid was cotransfected as a control for transfection efficiency. The total amount of transfected DNA was kept constant by using corresponding empty vector mock DNA. For Hek293T cells, 16 h post-transfection, medium was refreshed and, where necessary, supplemented with Wy-14643 (10 µM) or vehicle (0.1% Me2SO). After 24 h, cells were collected and the luciferase and beta -galactosidase assays were performed as described previously (19). For COS-1 transfection, after 5 h cells were refed with DMEM supplemented with 0.2% FCS and Wy-14643 (10 µM) or vehicle (0.1% Me2SO). 48 h later, the COS-1 cells were collected and also subjected to luciferase and beta -galactosidase assays. All experiments were repeated at least three times.

In Vitro Protein-Protein Interaction Assay (GST Pull-down)-- GST pull-down assays were performed as described elsewhere (33). Briefly, approximately 0.5 µg of GST fusion protein bound to glutathione-Sepharose 4B beads was incubated with 4-8 µl (according to expression efficiency) of in vitro translated [35S]methionine-labeled protein in the presence of 100 µM Wy-14643 dissolved in Me2SO or Me2SO alone in a total volume of 200 µl of incubation buffer (20 mM Tris-HCl, pH 7.4, 100 mM NaCl, 1 mM MgCl2, 10% glycerol, 0.1% Tween 20, 1.5% bovine serum albumin, 1 mM pefabloc, 0.15 IU/ml aprotinin) and rotated at 4 °C. After centrifugation, the beads were washed four times for 15 min with incubation buffer without bovine serum albumin, resuspended in 30 µl of 1 × Laemmli buffer, boiled for 5 min, and centrifuged. The supernatant was loaded on a SDS-polyacrylamide gel electrophoresis gel. After drying, gels were exposed to a phosphorimager (Image Quant) screen.

    RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Aortas from PPARalpha -Null Mice Display an Exacerbated Inflammatory Response to LPS Stimulation and Are Refractory to Fenofibrate Treatment-- In order to provide genetic evidence for a role of PPARalpha in the vascular inflammatory response, basal and LPS-stimulated IL-6 production by aortic segments from PPARalpha -/- and +/+ mice were compared. In the absence of LPS stimulation, basal IL-6 secretion was similar in aortas of both groups of mice (Fig. 1A). LPS stimulation resulted in a significant increase of IL-6 production (approximately 3-fold) in wild-type mice aortas, in agreement with previous observations (26). However, this increase was much greater in aortas isolated from the PPARalpha -/- mice (12-fold, p < 0.03). These observations indicate that PPARalpha deficiency results in an increased vascular inflammatory response, as assessed by enhanced IL-6 secretion. Next, the influence of treatment with the PPARalpha -agonist fenofibrate on the inflammatory response was analyzed in aortas of PPARalpha -/- and +/+ mice. In the absence of LPS stimulation, basal IL-6 mRNA levels were very low (data not shown) and became only detectable after LPS injection. In LPS-injected PPARalpha +/+ mice, treatment with fenofibrate decreased significantly IL-6 mRNA levels in aortas, whereas fenofibrate did not have any effect in PPARalpha -/- mice (Fig. 1B). These data indicate that the anti-inflammatory properties of fenofibrate are PPARalpha -dependent and that PPARalpha controls the vascular inflammatory response at the gene expression level in vivo.


View larger version (14K):
[in this window]
[in a new window]
  		Fig. 1.   Aortas from PPARalpha -null mice display an exacerbated inflammatory response to LPS stimulation and are refractory to fenofibrate treatment. A, aortas from male PPARalpha -/- and +/+ mice (n = 8/group) were isolated, cut into two segments, and exposed to LPS (10 µg/ml) or vehicle for 24 h. Medium was collected and IL-6 concentration measured, as described under "Materials and Methods." Values were normalized to DNA content and expressed as mean ± S.E. Statistical analysis was performed using a two-way analysis of variance (p < 0.05). Statistical significant differences are indicated by different letters. B, male PPARalpha -/- and +/+ mice (n = 6/group) were fed with rodent chow or rodent chow diet supplemented with 0.2% fenofibrate for 2 weeks. At the end of the treatment period, half of the mice of each group received an intraperitoneal injection of LPS (1 mg/kg). The other half received a vehicle (water) injection. After 3 h, aortas from individual mice were isolated and subjected to RNA analysis. IL-6 mRNA levels were measured by Northern blot analysis and normalized to 36B4 mRNA levels. Values (mean ± S.E.) are expressed as a percentage of the untreated control animals. Since IL-6 mRNA levels in the vehicle injected animals are below the detection limit, only results from LPS-injected animals are depicted. Statistical significant differences from controls are indicated by an asterisk (*, p < 0.05).

PPARalpha Activators Inhibit IL-1-induced IL-6 Gene Expression in Human Aortic SMC-- Next, it was determined whether PPARalpha activation inhibits the induction of IL-6 mRNA levels by inflammatory cytokines, such as IL-1beta in human aortic SMC. As expected, stimulation of SMC with IL-1beta resulted in a severalfold increase of IL-6 mRNA (Fig. 2). This induction was, however, inhibited in the presence of Wy-14643. These data indicate that PPARalpha activation inhibits IL-6 gene induction by inflammatory cytokines in vitro.


View larger version (59K):
[in this window]
[in a new window]
  		Fig. 2.   PPARalpha activation inhibits IL-1-induced IL-6 gene expression in human aortic SMC. Human aortic SMC (80% confluence) were incubated for 2 h in standard medium with Wy-14643 (250 µM) or vehicle (Me2SO 0.1%) and subsequently stimulated with IL-1beta (10 ng/ml) during 3 h. IL-6 and 36B4 mRNA levels were measured by Northern blot analysis.

PPARalpha Inhibits IL-6 Gene Transcription by Interfering with the Promoter Transactivation by c-Jun and p65-- Next, it was studied whether PPARalpha interferes with IL-6 gene expression at the transcriptional level. Several regulatory elements such as an AP-1, a C/EBP, an NF-kappa B, and a multiple response element, have been identified in the human IL-6 promoter (34), of which the AP-1 and NF-kappa B response elements have been shown to mediate the IL-6 response to inflammatory stimuli such as IL-1beta (35). To test whether PPARalpha interferes with the transcriptional activation of the IL-6 promoter by the transcription factors AP-1 and/or NF-kappa B, transient cotransfection experiments were performed. Because of the inability to transfect primary cultured human aortic SMC, COS-1 cells were used for these transient transfection experiments. Cotransfection of the p65 NF-kappa B subunit resulted in a strong activation of wild-type IL-6 promoter activity (11-fold) (Fig. 3A). Cotransfection of human PPARalpha alone did not influence basal IL-6 promoter activity. However, the activation of the IL-6 promoter by p65 was significantly (p = 0.007) decreased in the presence of PPARalpha activated by Wy-14643. As expected, p65 cotransfection did not activate a promoter construct mutated in the NF-kappa B site, nor a construct containing only the IL-6 promoter TATA box region in front of the luciferase gene (Fig. 3A). Interestingly, p65 induction of the IL-6 promoter mutated in the AP-1 site was less pronounced as compared with the wild-type promoter, suggesting functional interaction between the NF-kappa B and AP-1 sites. However, similarly as the wild-type, cotransfection of PPARalpha in the presence of its ligand was able to repress p65-mediated induction (p = 0.031) of the IL-6 promoter mutated in the AP-1 site (Fig. 3A).


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3.   PPARalpha inhibits IL-6 gene transcription by interfering with the promoter transactivation by c-Jun and p65. COS cells were transfected with the indicated IL-6 promoter constructs (1 µg) in the presence of hPPARalpha (1 µg), p65 (1 µg) (A), c-Jun and c-Fos (1 µg) (B), or empty expression plasmids. After 5 h, cells were washed and refed with DMEM supplemented with 0.2% FCS in the presence of Wy-14643 (10 µM) when PPARalpha was cotransfected. Statistical analysis was assessed by analysis of variance (p < 0.05). Statistical significant differences between groups were then evaluated by the Student's t test. Statistical significance was assigned when p < 0.05 and is indicated in the text.

Cotransfection of c-Jun and c-Fos also strongly activated the wild-type human IL-6 promoter (Fig. 3B). This induction was reduced by PPARalpha cotransfection (p = 0.042) in the presence of Wy-14643. As expected, the IL-6 promoter mutated on the AP-1 site as well as the minimal IL-6 promoter were not activated by c-Jun/c-Fos. Similarly as for p65 activation on the AP-1 site mutated promoter, the NF-kappa B site-mutated promoter was less inducible by c-Jun/c-Fos than the wild-type promoter, but PPARalpha was able to repress the induction by c-Jun/c-Fos (p = 0.039). However, PPARalpha cotransfection did not result in a significant inhibition of the basal activity of the NF-kappa B site mutated promoter. Taken together, these data indicate that PPARalpha represses IL-6 promoter activation by interfering negatively with the AP-1 and NF-kappa B transcriptional activities.

PPARalpha Represses AP-1 and NF-kappa B Activities Independently of the Promoter Context-- Next, it was investigated whether PPARalpha could interfere with AP-1 and NF-kappa B transactivation independently of the promoter context. Therefore, we analyzed the effect of PPARalpha on the transcriptional activation of a Gal4-dependent reporter, activated by the p65 or c-Jun chimeras (Fig. 4). As a control, PPARalpha did not influence transcriptional activity of the Gal4 DBD alone. Transfection of the chimera containing the NF-kappa B p65 subunit led to a strong transcriptional activation (almost 20-fold) of the reporter construct. This induction was significantly reduced (-60%) by PPARalpha cotransfection in the presence of Wy-14643. Cotransfection of the c-Jun chimera resulted in a less pronounced induction of promoter activity (6-fold), but an almost complete repression (-74%) of c-Jun-mediated transactivation was observed in the presence of cotransfected PPARalpha in the presence of its ligand. These results indicate that PPARalpha interferes negatively with the c-Jun as well as with the NF-kappa B transactivation capacities in a manner independent of the promoter context.


View larger version (12K):
[in this window]
[in a new window]
  		Fig. 4.   PPARalpha represses c-Jun and p65 transcriptional activity in a promoter-independent manner. COS cells were transfected respectively with the p(Gal4)250hu.IL-6P-Luc+ (340 ng) reporter plasmid in the presence of the indicated Gal4 chimera (80 ng) and the human PPARalpha (80 ng) expression plasmids or its corresponding empty vector. Cells were subsequently incubated with Wy-14643 (10 µM). Statistical significant differences (p < 0.05) were evaluated by the Student's t test. Statistical significant effects of PPARalpha cotransfection are indicated by an asterisk.

p65 and c-Jun Reciprocally Repress PPARalpha Transactivation of a PPRE-driven Promoter-- In order to determine whether the transcriptional cross-talk between PPARalpha , NF-kappa B, and AP-1 activities occurs in a reciprocal manner, transfection assays were performed to test the effect of p65 and c-Jun on a PPAR-dependent PPRE-driven promoter. In the absence of cotransfected PPARalpha , the PPRE-driven reporter was slightly activated by addition of Wy-14643 (Fig. 5). As expected, cotransfection of PPARalpha significantly induced the reporter activity in the presence of Wy-14643 (3.5-fold). Cotransfection of increasing amounts of p65 (Fig. 5A) or c-Jun (Fig. 5B) expression vectors led to a dose-dependent inhibition of PPARalpha -induced reporter activity without affecting basal promoter activity. This result indicates the existence of a bidirectional antagonism between PPARalpha , c-Jun, and NF-kappa B activities.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5.   p65 and c-Jun repress PPARalpha transactivation of a PPRE-driven promoter. COS cells were transfected with a reporter construct driven by six copies of the apo AII PPRE (J6-TK-Luc) (10 ng) in the presence of hPPARalpha (30 ng) and increasing amounts (1, 10, 100 ng) of p65 (A) or c-Jun (B) expression plasmids. In the absence of hPPARalpha , 100 ng of p65 (A) or c-Jun (B) were transfected as negative controls. After 5 h, cells were washed and refed with DMEM supplemented with 0.2% FCS in the presence of Wy-14643 (10 µM) or solvent (Me2SO 0.1%).

p65 and c-Jun Functionally Interfere with Different PPARalpha Domains-- In order to delineate which domains of PPARalpha are involved in the transcriptional cross-talk with AP-1 and NF-kappa B, transfection experiments were performed using a chimera containing the PPARalpha C-terminal amino acids (LBD) fused to the Gal4 DBD (amino acids 1-147) in the presence or absence of c-Jun or p65 (Fig. 6, A and B). Neither treatment with Wy-14643 nor cotransfection of p65 and c-Jun exerted a major effect on the reporter activity in the absence of cotransfected Gal4-LBD (Fig. 6, A and B). However, in the presence of the latter, Wy-14643 strongly activated (10-fold) the Gal4-responsive promoter. This induction was significantly repressed by p65 (Fig. 6A), whereas cotransfection of c-Jun had almost no effect (Fig. 6B). When the influence of a recently identified, natural truncated form of PPARalpha , which lacks the entire LBD and only contains amino acids 1-170 of the wild-type PPARalpha , was tested on Gal4-p65 driven transactivation (Fig. 6C), a significant, but less pronounced, repression (-20%) of p65 transactivation was observed when compared with the wild-type form (-60%). By contrast, the truncated form of PPARalpha was able to repress c-Jun transactivation to a similar extent as the wild-type PPARalpha (-60%) (Fig. 6D). Taken together these data indicate that transrepression of NF-kappa B by PPARalpha involves mainly the C-terminal domains of the receptor, whereas transrepression of c-Jun implicates the N terminus.


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 6.   p65 and c-Jun functionally interfere with different PPARalpha domains. A and B, COS cells were transfected with the Gal5-TK-pGL3 reporter plasmid containing five Gal4 response elements, the PPARalpha chimera Gal4-hPPARalpha (0.1 µg), and p65 (A) or c-Jun (B) expression vectors. After 5 h, cells were washed and refed with DMEM supplemented with 0.2% FCS in the presence of Wy-14643 (10 µM) or solvent (Me2SO 0.1%). C and D, COS cells were transfected with the p(Gal4)250hu.IL-6P-Luc+ (340 ng) reporter plasmid in the presence of the Gal4-p65 chimera (80 ng) (C) or the Gal4-c-Jun chimera (80 ng) (D) and the wild-type or LBD-lacking human PPARalpha expression plasmids or its corresponding empty vector (80 ng). After 5 h, cells were washed and refed with DMEM supplemented with 0.2% FCS in the presence of Wy-14643 (10 µM). Statistical significant differences were evaluated by the Student's t test and are indicated by asterisks (*, p < 0.05; **, p < 0.01).

CBP Cotransfection Does Not Affect the Inhibition of p65 and c-Jun Activity by PPARalpha -- Since it was reported that competition for common coactivators could be a mechanism of gene repression by nuclear receptors (36-38), the effect of CBP (a coactivator that has been shown to interact with PPARalpha , c-Jun, and p65 (30, 39, 40)) in the repression of NF-kappa B and AP-1 by PPARalpha was explored. As expected, cotransfection of p65 or c-Jun strongly induced a minimal promoter driven by multiple NF-kappa B (Fig. 7A) or AP-1 (Fig. 7B) response elements, respectively (lane 5). PPARalpha cotransfection in the presence of Wy-14643 resulted in a significant reduction of both p65 and c-Jun reporter transactivation (lane 6). To assess the effect of the co-integrator CBP, increasing amounts of CBP (0.1, 0.2, 0.3 µg) expression vector were cotransfected. Increasing CBP concentrations stepwise enhanced the c-Jun or p65-mediated transcriptional activation (lanes 7, 8, and 9), which could again be repressed in the presence of activated PPARalpha (lanes 10 and 11). These results indicate that the PPARalpha -mediated repression of both p65 and c-Jun transcriptional activities occurs in a CBP-independent manner.


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 7.   CBP cotransfection does not affect the inhibition of p65 and c-Jun activity by PPARalpha . Hek293T cells were transfected with either the p(kappa B)3-luc+ (A) or p(AP-1)3-luc+ reporter construct (80 ng) together with phosphoglycerate kinase-beta -galactosidase (40 ng), hPPARalpha (50 or 100 ng), p65 (20 ng) or c-Jun (40 ng) and CBP (100, 200, or 300 ng) expression vectors. The day after transfection, cells were washed and refed with fresh medium in the presence of 10 µM Wy-14643 for 24 h.

PPARalpha Interacts Physically with p65, c-Jun, and CBP-- To determine whether PPARalpha interacts physically with p65, c-Jun, and CBP, GST pull-down experiments were performed (Fig. 8). Interaction of PPARalpha protein with the p65 Rel homology domain (aa 12-317) could be detected, whereas the C-terminal transactivation domain of p65 (aa 286-551) did not bind to PPARalpha (Fig. 8A). Furthermore, PPARalpha also interacted with the JNK-responsive part of c-Jun (aa 1-79) (Fig. 8B). This interaction occurs via the N-terminal DBD containing part of PPARalpha , since the C-terminal deletion mutant of PPARalpha also binds to c-Jun (Fig. 8C), in agreement with the results from the transfection experiments (Fig. 6D). Finally, in line with a previous report (39), PPARalpha was found to associate with the N-terminal aa 1-213 of CBP. Interestingly, the LBD-lacking PPARalpha variant also interacted strongly with CBP (aa 1-213) (Fig. 8C). Altogether these results indicate that PPARalpha interacts with the N terminus of c-Jun and the Rel homology domain of p65 and both the C- and N-terminal halves of PPARalpha bind to CBP.


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 8.   Different transcription factor and cofactor domains interact with PPARalpha . GST pull-down assays, using the indicated GST fusion proteins and in vitro translated 35S-labeled wild-type PPARalpha (A and B) or PPARalpha Delta LBD (C) proteins, were performed as described under "Experimental Procedures" (arrowheads: 62-kDa wild-type PPARalpha (A and B) and 31-kDa PPARalpha Delta LBD (C) proteins; control, GST beads alone). Incubations with wild-type PPARalpha were performed in the presence of Wy-14643 (100 µM).


    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

Chronic inflammation is a hallmark of atherosclerosis (1, 2, 41). It has therefore been postulated that negatively interfering with the inflammatory response at the level of the vascular wall might lead to a selective inhibition of the atherogenic process. The PPARalpha signaling pathway constitutes a potentially interesting target for anti-inflammatory drug development. Indeed, using PPARalpha -deficient mice, Devchand et al. (11) have demonstrated that PPARalpha plays a role in acute inflammation control. Here, we show that PPARalpha has anti-inflammatory properties at the level of the vascular wall, since aortas from PPARalpha -null mice display an exacerbated inflammatory response to LPS stimulation, as measured by IL-6 production. Furthermore, PPARalpha mediates the anti-inflammatory actions of fibrates, such as fenofibrate, at the level of the vascular wall. This result extends previous reports showing that PPARalpha ligands repress cytokine-induced IL-6 production in SMC (19), inducible nitric-oxide synthase activity in murine macrophages (42), and VCAM-1 expression in endothelial cells (43). The physiological relevance of these observations is further corroborated by the demonstration that fibrates lower plasma levels of inflammatory cytokines such as IL-6, tumor necrosis factor alpha , and interferon gamma  in patients with atherosclerosis (19, 20). Interestingly, not only PPARalpha , but also PPARgamma (22, 44, 45), ligands have been reported to inhibit production of inflammatory cytokines by monocytes/macrophages in vitro. All these studies underline a potential modulatory role of PPAR ligands in the pathogenesis of atherosclerosis. Furthermore, IL-6 production is also inhibited by estrogen receptor (46) and glucocorticoid receptor agonists (24), suggesting that PPARs share anti-inflammatory properties with a number of other nuclear receptors.

PPARalpha ligands exert their activity by negatively regulating IL-1-induced IL-6 gene expression in SMC. Results from mutation analysis demonstrate that PPARalpha represses IL-6 promoter activation by negatively interfering with c-Jun and NF-kappa B transactivation. Similarly, COX-2 repression in SMC by PPARalpha , as well as repression of inducible nitric-oxide synthase gelatinase B, scavenger receptor-A (22), and tumor necrosis factor alpha  expression (44) in murine and human macrophages by PPARgamma have been suggested to be effected by antagonizing the AP-1, STAT, and NF-kappa B pathways (19, 21, 22).

Several molecular mechanisms can be invoked to explain transcriptional negative cross-talk between PPARalpha and other transcription factors such as c-Jun or p65. PPARalpha may compete for binding to identical or overlapping response elements. However, our results show that PPAR activation does not activate basal IL-6 promoter activity, indicating the absence of a functional PPRE. Furthermore, the interference between PPAR and c-Jun or p65 occurs in a promoter-independent manner, since it is observed using Gal4 fusion proteins. Therefore, competition for binding site recognition can be excluded.

In this study, we found that PPARalpha represses p65 as well as c-Jun transactivation of the human IL-6 promoter. Furthermore, our transfection results demonstrate that this interference is reciprocal as described previously for other nuclear receptors such as glucocorticoid receptor (47-51), the retinoic acid receptor (52), the progesterone receptor (52), and the androgen receptor (53).

To assess the hypothesis of a physical interaction between PPARalpha and c-Jun or p65, we performed GST pull-down experiments. Our results indicate that PPARalpha associates with aa 1-79 of c-Jun protein via the N-terminal part of the receptor, since the PPARalpha natural occurring splicing variant lacking the LBD was still able to interact with c-Jun. This result was corroborated by the transfection experiments using the Gal4 fusion proteins showing that PPARalpha represses c-Jun transactivation in a LBD-independent manner. This result is in line with previous works showing that the receptor DBD is required for the interaction between AP-1 and nuclear receptors such as glucocorticoid receptor (47-51), androgen receptor (53), and retinoic acid receptor (52). Our data extend a previous study, which suggested a potential cross-talk between PPARalpha and c-Jun (54). GST pull-down experiments also indicate that PPARalpha interacts weakly with p65 and that this interaction occurs through aa 12-317 of p65. This region contains the Rel homology domain which mediates DNA binding, dimerization, and interaction with Ikappa Balpha . Through this domain, p65 was previously reported to interact with other nuclear receptors such as the glucocorticoid receptor (51). Palvimo et al. (55) also found a weak interaction between p65 and androgen receptor. In addition to the GST pull-down experiments, results from transfection experiments suggest that the cross-talk between PPARalpha and p65 occurs mainly via the LBD of PPARalpha , since the truncated variant was less efficient in NF-kappa B repression. In view of our data, we propose a model of transcriptional cross-talk between PPARalpha and c-Jun or p65, in which PPARalpha represses c-Jun transactivation mainly via its N terminus, whereas p65 transrepression occurs in a LBD-dependent manner.

Since it has been suggested that inhibition of transcriptional activation by nuclear receptors can be effected by competing for limiting amounts of co-activators (36-38), we investigated how CBP might interfere with NF-kappa B and AP-1 activities and their repression by PPARalpha . Cotransfection assays showed that low amounts of CBP are indeed sufficient to increase the activated state of c-Jun or p65, whereas the relative repression by PPARalpha remains unaffected. Furthermore, the three key players involved in PPARalpha -dependent transrepression on CBP-stimulated NF-kappa B and AP-1-dependent reporters are able to interact with each other in vitro. GST pull-down assays confirmed that PPARalpha interacts with the N-terminal part of CBP (aa 1-213), i.e. the nuclear receptor-associating domain, as described previously (39). Furthermore, an additional interacting domain of PPARalpha with CBP was mapped to its N-terminal part. To our knowledge, this is the first demonstration that PPARalpha interacts with CBP via its N-terminal domain. Although so far most coactivators have been shown to interact with the nuclear receptor LBD, the N-terminal part of the thyroid receptor has also been shown to mediate coactivator interaction (56). Finally, and as already stated above, PPARalpha is also able to interact with the DNA-binding domain of p65, as well as with the JNK-responsive part of c-Jun, whereas both proteins have already been described to associate with CBP (40, 57, 58). Hence, the various mutual interactions between the different transcription factors involved and/or CBP as well as their relative abundance may therefore be the critical parameters to determine the actual state of activation and/or repression.

Finally, recent reports (59, 60) demonstrate that nuclear receptors and AP-1 or NF-kappa B can functionally interact by interfering with signaling pathways (such as protein phosphorylation), and this modulates transcription factor activity. Caelles et al. (59) demonstrated that various nuclear receptors block AP-1 activation by interfering with the JNK cascade activation. Since PPARalpha interacts with the JNK phosphorylation-responsive part of c-Jun, our results do not allow us to exclude this aspect in the mechanism of PPARalpha -mediated gene repression.

Apart from being a marker for vascular inflammation, down-regulation of IL-6 may have important (patho)physiological consequences, since this cytokine may be involved in the pathogenesis of atherosclerosis (2). Biswas et al. (61) reported that IL-6 induces monocyte chemotactic protein-1 expression in peripheral blood mononuclear cells and U937 macrophages. Thus, suppression of IL-6 secretion by PPAR ligands may indirectly inhibit the production of potent chemokines involved in monocyte recruitment into the subendothelial space, resulting in less foam cell formation. In conclusion, the results from this study show that, in addition to their lipid-lowering properties, PPARalpha activators may also have beneficial effects in atherosclerosis by inhibiting vascular inflammation.

    ACKNOWLEDGEMENTS

We acknowledge the technical contribution of O. Vidal, B. Derudas, P. Poulain, and K. Van Wesemael.

    FOOTNOTES

* This work was supported by grants of the Institut Pasteur de Lille, INSERM, Comité Français de Coordination des Recherches sur l'Athèrosdèrose et le cholestèrol, Rhône-Poulenc Rorer, Laboratoires Fournier, and the Région Nord-Pas-de-Calais/FEDER.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¶¶ To whom correspondence should be addressed: U.325 INSERM, Dépt. d'Athérosclérose, Institut Pasteur de Lille, 1 rue du Pr. Calmette, 59019 Lille, France. Tel.: 33-3-20-87-73-88; Fax: 33-3-20-87-73-60; E-mail: bart.staels@pasteur-lille.fr.

§ Supported by a grant from the Région Nord-Pas-de-Calais.

parallel Holds a fellowship of the IWT.

§§ Research director with the FWO-Vlaanderen.

2 Gervois, P., Torra, I. P., Chinetti, G., Grötzinger, T., Dubois, G., Fruchart, J. C., Fruchart-Najib, J., Leitersdorf, E., and Staels, B. (1999) Mol. Endocrinol. 13, 1535-1549.

    ABBREVIATIONS

The abbreviations used are: SMC, smooth muscle cells; IL, interleukin; PPAR, peroxisome proliferator-activated receptor; PPRE, PPAR-response element; LPS, lipopolysaccharide; FCS, fetal calf serum; PCR, polymerase chain reaction; aa, amino acids; DMEM, Dulbecco's modified Eagle's medium; GST, glutathione S-transferase; LBD, ligand binding domain; DBD, DNA binding domain; JNK, c-Jun N-terminal kinase; CBP, cAMP-responsive element-binding protein-binding protein.

    REFERENCES
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES

1. Ross, R. (1999) N. Engl. J. Med. 340, 115-126[Free Full Text]
2. Ross, R. (1993) Nature 362, 801-809[CrossRef][Medline] [Order article via Infotrieve]
3. Rus, H. G., Vlaicu, R., and Niculescu, F. (1996) Atherosclerosis 127, 263-271[CrossRef][Medline] [Order article via Infotrieve]
4. Ikeda, U., Ikeda, M., Seino, Y., Takahashi, M., Kano, S., and Shimada, K. (1992) Atherosclerosis 92, 213-218[CrossRef][Medline] [Order article via Infotrieve]
5. Castell, J. V., Gomez-Lechon, M. J., David, M., Andus, T., Geiger, T., Trullenque, R., Fabra, R., and Heinrich, P. C. (1989) FEBS Lett. 242, 237-239[CrossRef][Medline] [Order article via Infotrieve]
6. Xing, Z., Gauldie, J., Cox, G., Baumann, H., Jordana, M., Lei, X. F., and Achong, M. K. (1998) J. Clin. Invest. 101, 311-320[Medline] [Order article via Infotrieve]
7. Tilg, H., Dinarello, C. A., and Mier, J. W. (1997) Immunol. Today 18, 428-432[CrossRef][Medline] [Order article via Infotrieve]
8. Issemann, I., and Green, S. (1990) Nature 347, 645-650[CrossRef][Medline] [Order article via Infotrieve]
9. Tugwood, J. D., Issemann, I., Anderson, R. G., Bundell, K. R., McPheat, W. L., and Green, S. (1992) EMBO J. 11, 433-439[Medline] [Order article via Infotrieve]
10. Forman, B. M., Chen, J., and Evans, R. M. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 4312-4317[Abstract/Free Full Text]
11. Devchand, P. R., Keller, H., Peters, J. M., Vasquez, M., Gonzalez, F. J., and Wahli, W. (1996) Nature 384, 39-43[CrossRef][Medline] [Order article via Infotrieve]
12. Kliewer, S. A., Lenhard, J. M., Wilson, T. M., Patel, I., Morris, D. C., and Lehmann, J. M. (1995) Cell 83, 813-819[CrossRef][Medline] [Order article via Infotrieve]
13. Forman, B. M., Tontonoz, P., Chen, J., Brun, R. P., Spiegelman, B. M., and Evans, R. M. (1995) Cell 83, 803-812[CrossRef][Medline] [Order article via Infotrieve]
14. Nagy, L., Tontonoz, P., Alvarez, J. G. A., Chen, H., and Evans, R. M. (1998) Cell 93, 229-240[CrossRef][Medline] [Order article via Infotrieve]
15. Staels, B., Dallongeville, J., Auwerx, J., Schoonjans, K., Leitersdorf, E., and Fruchart, J. C. (1998) Circulation 98, 2088-2093[Abstract/Free Full Text]
16. Saitoh, K., Mori, T., Kasai, H., Nagayama, T., Tsuchiya, A., and Ohbayashi, S. (1995) Folia Pharmacol. Jpn. 106, 41-50
17. Frick, H., Syvänne, M., Nieminen, S., Kauma, H., Majahalme, S., Virtanen, V., Kesäniemi, A., Pasternack, A., and Taskinen, M. R. (1997) Circulation 96, 2137-2143[Abstract/Free Full Text]
18. Ruotolo, G., Ericsson, C. G., Tettamanti, C., Karpe, F., Grip, L., Svane, B., Nilsson, J., De Faire, U., and Hamsten, A. (1998) J. Am. Coll. Cardiol. 32, 1648-1656[Abstract/Free Full Text]
19. Staels, B., Koenig, W., Habib, A., Merval, R., Lebret, M., Pineda-Torra, I., Delerive, P., Fadel, A., Chinetti, G., Fruchart, J. C., Najib, J., Maclouf, J., and Tedgui, A. (1998) Nature 393, 790-793[CrossRef][Medline] [Order article via Infotrieve]
20. Madej, A., Okopien, B., Kowalski, J., Zielinski, M., Wysocki, J., Szygula, B., Kalina, Z., and Herman, Z. S. (1998) Int. J. Clin. Pharmacol. Ther. 36, 345-349[Medline] [Order article via Infotrieve]
21. Chinetti, G., Griglio, S., Antonucci, M., Torra, I. P., Delerive, P., Majd, Z., Fruchart, J. C., Chapman, J., Najib, J., and Staels, B. (1998) J. Biol. Chem. 273, 25573-25580[Abstract/Free Full Text]
22. Ricote, M., Li, A. C., Willson, T. M., Kelly, C. J., and Glass, C. K. (1998) Nature 391, 79-82[CrossRef][Medline] [Order article via Infotrieve]
23. Staels, B., van Tol, A., Andreu, T., and Auwerx, J. (1992) Arterioscler. Thromb. 12, 286-294[Abstract/Free Full Text]
24. De Bosscher, K., Schmitz, M. L., Vanden Berghe, W., Plaisance, S., Fiers, W., and Haegeman, G. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 13504-13509[Abstract/Free Full Text]
25. Lee, S. S. T., Pineau, T., Drago, J., Lee, E. J., Owens, J. W., Kroetz, D. L., Fernandez-Salguero, P. M., Westphal, H., and Gonzalez, F. J. (1995) Mol. Cell. Biol. 15, 3012-3022[Abstract]
26. Belmin, J., Bernard, C., Corman, B., Merval, R., Esposito, B., and Tedgui, A. (1995) Am. J. Physiol. 268, 2288-2293
27. Hinegardner, R. T. (1971) Anal. Biochem. 39, 197-201[CrossRef][Medline] [Order article via Infotrieve]
28. Vanden Berghe, W., Plaisance, S., Boone, E., De Bosscher, K., Schmitz, M., Fiers, W., and Haegeman, G. (1998) J. Biol. Chem. 273, 3285-3290[Abstract/Free Full Text]
29. Schmitz, M. L., and Bauerle, P. A. (1991) EMBO J. 10, 3805-3817[Medline] [Order article via Infotrieve]
30. Bannister, A. J., Oehler, T., Wilhelm, D., Angel, P., and Kouzarides, T. (1995) Oncogene 11, 2509-2514[Medline] [Order article via Infotrieve]
31. Bannister, A. J., and Kouzarides, T. (1995) EMBO J. 14, 4758-4762[Medline] [Order article via Infotrieve]
32. Jaffray, E., Wood, K. M., and Hay, R. T. (1995) Mol. Cell. Biol. 15, 2166-2172[Abstract]
33. Treuter, E., Albrektsen, T., Johansson, L., Leers, J., and Gustafsson, J. A. (1998) Mol. Endocrinol. 12, 864-881[Abstract/Free Full Text]
34. Akira, S., and Kishimoto, T. (1993) Adv. Immunol. 54, 1-78[Medline] [Order article via Infotrieve]
35. Munoz, C., Pascual-Salcedo, D., Castellanos, M. C., Alfranca, A., Aragones, J., Vara, A., Redondo, J. M., and Landazuri, M. O. (1996) Blood 88, 3482-3490[Abstract/Free Full Text]
36. Kamei, Y., Xu, L., Heinzel, T., Torchia, J., Kurokawa, R., Gloss, B., Lin, S. C., Heyman, R. A., Rose, D. W., Glass, C. K., and Rosenfeld, M. G. (1996) Cell 85, 403-414[CrossRef][Medline] [Order article via Infotrieve]
37. Göttlicher, M., Heck, S., and Herrlich, P. (1998) J. Mol. Med. 76, 480-489[CrossRef][Medline] [Order article via Infotrieve]
38. Sheppard, K. A., Phelps, K. M., Williams, A. J., Thanos, D., Glass, C. K., Rosenfeld, M. G., Gerritsen, M. E., and Collins, T. (1998) J. Biol. Chem. 273, 29291-29294[Abstract/Free Full Text]
39. Dowell, P., Ishmael, J. E., Avram, D., Peterson, V. J., Nevrivy, D. J., and Leid, M. (1997) J. Biol. Chem. 272, 33435-33443[Abstract/Free Full Text]
40. Gerritsen, M. E., Williams, A. J., Neish, A. S., Moore, S., Shi, Y., and Collins, T. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 2927-2932[Abstract/Free Full Text]
41. Mehta, J. L., Saldeen, T. G. P., and Rand, K. (1998) J. Am. Coll. Cardiol. 31, 1217-1225[Abstract/Free Full Text]
42. Colville-Nash, P. R., Qureshi, S. S., Willis, D., and Willoughby, D. A. (1998) J. Immunol. 161, 978-984[Abstract/Free Full Text]
43. Marx, N., Sukhova, G. K., Collins, T., Libby, P., and Plutzky, J. (1999) Circulation 99, 3125-3131[Abstract/Free Full Text]
44. Jiang, C., Ting, A. T., and Seed, B. (1998) Nature 391, 82-86[CrossRef][Medline] [Order article via Infotrieve]
45. Marx, N., Sukhova, G., Murphy, C., Libby, P., and Plutzky, J. (1998) Am. J. Pathol. 153, 1-7[Free Full Text]
46. Ray, A., Prefontaine, K. E., and Ray, P. (1994) J. Biol. Chem. 269, 12940-12946[Abstract/Free Full Text]
47. Schüle, R., Rangarajan, P., Kliewer, S., Ransone, L. J., Bolado, J., Yang, N., Verma, I. M., and Evans, R. M. (1990) Cell 62, 1217-1226[CrossRef][Medline] [Order article via Infotrieve]
48. Yang-Yen, H. F., Chambard, J. C., Sun, Y. L., Smeal, T., Schmidt, T. J., Drouin, J., and Karin, M. (1990) Cell. 62, 1205-1215[CrossRef][Medline] [Order article via Infotrieve]
49. Jonat, C., Rahmsdorf, H. J., Park, K. K., Cato, A. C., Gebel, S., Ponta, H., and Herrlich, P. (1990) Cell 62, 1189-1204[CrossRef][Medline] [Order article via Infotrieve]
50. Ray, A., and Prefontaine, K. E. (1994) Proc. Natl. Acad. Sci. U. S. A.</