| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pregnancy |
Department of Obstetrics and Gynecology (Division of Maternal-Fetal Medicine and Laboratory of Perinatal Research),3
Department of Internal Medicine (Division of Nephrology and Dorothy M. Davis Heart and Lung Research Institute),4 The Ohio State University, College of Medicine and Public Health, Columbus, Ohio 43210
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B) regulatory elements alone, but was evident when a fragment (–891/ +9) of the COX-2 gene 5'-promoter was present. Both agents transiently activated intermediates of multiple signaling pathways potentially involved in the regulation of COX-2 gene expression. The 26 S proteasome inhibitor, MG-132, selectively abrogated IL-1ß-driven NFB activation and COX-2 mRNA expression. Only pharmacologic blockade of the p38 mitogen-activated protein kinase eliminated COX-2 expression following EGF stimulation. We conclude that EGF and IL-1ß appear to signal through different signaling cascades leading to COX-2 gene expression. IL-1ß employs the NFB pathway predominantly, while the spectrum of EGF signaling is broader and includes p38 kinase. The synergism observed between IL-1ß and EGF does not rely on augmented NFB function, but rather, occurs through differential use of independent response elements within the COX-2 promoter.
cytokines, growth factors, parturition, placenta, signal transduction
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
by intrauterine tissues facilitates cervical ripening, fetal membrane rupture, and the development of myometrial contractions [1]. Cyclooxygenase (COX) isoforms catalyze the committed and rate-limiting step in PG production [2]. COX-1 is constitutively present in many tissues and contributes to low-amplitude PG release, whereas transient expression of the inducible COX-2 (also known as prostaglandin synthase 2) isoform is required for high-amplitude PG production [2]. In the context of parturition, the majority of PGs elaborated by intrauterine tissues arise through increased de novo expression of COX-2, which is controlled primarily at the transcriptional level [3].
Although it has been theorized that term parturition results from a localized inflammatory process, current evidence suggests this reaction is modest [4]. While it is true that inflammatory cytokines (interleukin-1 [IL-1], IL-6, IL-8, and tumor necrosis factor [TNF]) increase within amniotic fluid toward term in normal pregnancies [5] and that these factors themselves are capable of stimulating COX-2 expression and PG production [6–8], it is unlikely that these agents work in isolation. Numerous factors capable of increasing COX-2 expression accumulate within the amniotic fluid with advancing gestation [1, 5]. It has been hypothesized that factors originating from the mature fetus (such as epidermal growth factor [EGF] or its close structural relative, transforming growth factor-) may act cooperatively with substances produced by fetal membranes and maternal decidua to control the robust elaboration of PGs, which precedes and accompanies term labor [9–11].
It has previously been demonstrated that EGF potentiates IL-1ß-driven PGE2 production in amnion and amnion-derived cells [11, 12]. To define the mechanism for this observation, it was hypothesized that EGF and IL-1ß might exhibit synergism in regulating COX-2 gene expression. Here, we report on the transcriptional regulation of COX-2 following single-agent or combined challenge with IL-1ß and EGF and examine the signaling mechanisms employed by these factors.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recombinant human EGF and IL-1ß were purchased from R&D Systems (Minneapolis, MN). The antibody detecting glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was purchased from Chemicon International (Temecula, CA). Antibodies against inhibitory factor-B (IB), nuclear factor-kappa B (NFB) subunit p65, IB kinase (IKK)/ß, c-Jun N-terminal kinase (JNK), extracellular-regulated kinase-2 (Erk-2), signal transducer and activator of transcription 3 (STAT3), and STAT5 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies recognizing phospho-IB(Ser32), phospho-IKK(Ser180)/IKKß(Ser181), STAT1, phospho-STAT1 (Tyr701), phospho-STAT3 (Tyr705), and phopho-STAT5 (Tyr694) were obtained from Cell Signaling Technology (Beverly, MA). Antibodies against phospho-Erk-1/Erk-2 (Thr183/Tyr185), phospho-JNK (Thr183/Tyr185), p38 mitogen-activated protein (MAP) kinase, and phospho-p38 MAP kinase (Thr180/Tyr 182) were from Promega (Madison, WI), as were the Dual-Luciferase Reporter Assay System and the Renilla luciferase control expression vector (pRL-SV40). The peptide aldehyde 26 S proteasome inhibitor MG-132, the p38 inhibitor SB-202190, the Janus kinase (JAK) inhibitor AG-490, and the MAP/Erk kinase (MEK) inhibitor PD-98059 were obtained from BIOMOL (Plymouth Meeting, PA). The JNK inhibitor SP-600125 was purchased from Tocris (Ballwin, MO). The 1.8-kilobase (kb) cDNA fragment encoding human COX-2 used as a probe for Northern blotting was a kind gift from Dr. Timothy Hla (University of Connecticut, Farmington, CT). The firefly luciferase reporter plasmid containing a segment (–891/+9) of the human COX-2 promoter (pPGS891LUC) was generously donated by Dr. Lee Ho Wang (University of Texas, Houston, TX) [13]. The PathDetect NFB-firefly luciferase reporter plasmid (pNFB-Luc) was obtained from Stratagene (La Jolla, CA). TRIZOL, Lipofectin, and PLUS reagents were obtained from Invitrogen (Carlsbad, CA). Hybond-N+ nylon and Hybond-C Extra nitrocellulose membranes were from Amersham Biosciences (Piscataway, NJ). Digoxigenin (DIG) Nucleic Acid Detection and DIG-High Prime kits were purchased from Roche Diagnostics (Indianapolis, IN). SuperSignal chemiluminescent substrate was obtained from Pierce Biotechnology (Rockford, IL). ProLong Antifade mounting reagent and Alexa Fluor 594-conjugated goat anti-rabbit antibodies were obtained from Molecular Probes (Eugene, OR). Other reagents were obtained from Sigma (St. Louis, MO)
Cell Culture
Human amnion-derived (WISH) cells [14] were obtained from the American Type Culture Collection (ATCC CCL-25) and maintained in Ham F-12/Dulbecco modified Eagle medium (Invitrogen, Gaithersburg, MD) supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, and 10% newborn calf serum. Cells were grown at 37°C in an atmosphere of 95% air/5% CO2 and used for experiments between the 3rd and 25th passages. All cultures were serum-starved overnight before experiments.
Northern Blot Analysis
Following experiments, total RNA was extracted using TRIZOL (Invitrogen). RNA (20 µg/lane) was fractionated on a 1% (w/v) agarose gel containing 2% (v/v) formaldehyde and transferred to a nylon membrane by downward capillary elution. Membranes were analyzed by hybridization to DIG-containing COX-2 and GAPDH cDNA probes, labeled using the DIG-High Prime kit (Roche). Blots were probed overnight at 42°C in hybridization buffer (7% [w/v] SDS, 50% [v/v] formamide, 325 mM sodium chloride, 32.5 mM sodium citrate, 50 mM sodium phosphate [pH 7.0], 0.1% [w/v] N-lauroylsarcosine, 50 µg/ml sheared salmon sperm DNA, and 2% [w/v] DIG Blocking Reagent [Roche]). Bound probes were identified using the DIG Nucleic Acid Detection kit (Roche). Chemiluminescent signals were detected using the VersaDoc Imaging System and analyzed using Quantity One software (Bio-Rad Laboratories, Hercules, CA).
Immunoblotting
Cellular proteins were extracted and prepared for immunoblotting as previously described [15]. Proteins (30 µg/lane) were resolved by electrophoresis on SDS/10% (w/v) polyacrylamide gels, transferred to nitrocellulose, and probed with antibodies directed against phosphorylated signaling intermediates of the NFB, JAK/STAT, and MAP kinase signal transduction pathways. Conditions for immunoblotting were established by a manufacturer's suggested protocol (Cell Signaling Technology). Immunoreactivity was detected using SuperSignal chemiluminescent substrate (Pierce Biotechnology) and detected using the VersaDoc Imaging System (Bio-Rad Laboratories).
Immunofluorescence
WISH cells were seeded onto flame-sterilized 12-mm2 glass coverslips placed in a 24-well tissue culture plate. Following treatments, cells were fixed for 1 h in 4% (w/v) paraformaldehyde/PBS, made permeable by the addition of 0.2% (v/v) Triton X-100/PBS for 15 min, and blocked in 5% (v/v) horse serum/PBS overnight at 4°C before the addition of antibodies. Monoclonal antibodies directed against p65 (Santa Cruz) were then applied. After stringent washing in PBS, the coverslips were exposed to fluorochrome-conjugated secondary antibodies (Molecular Probes). Following a second series of washes, cells were mounted using the ProLong Antifade kit (Molecular Probes) and visualized using an epifluorescence microscope (Nikon Instruments, Melville, NY).
Transient Transfection and Luciferase Assays
Transient transfections were performed in a 24-well tissue culture plate seeded at 5 x 104 cells per well. For each cotransfection, 0.6 µg of reporter plasmid (pNFB-Luc or pPGS891LUC), 0.1 µg control plasmid (pRL-SV40), 0.6 µl Lipofectin reagent (Invitrogen), and 5 µl PLUS reagent (Invitrogen) were used, according to the manufacturer's instructions. Luciferase assays were performed using the Dual-Luciferase Reporter Assay System (Promega) and a Dynex MLX Microplate Luminometer (Dynex Technologies, Chantilly, VA). Reporter activity was expressed as the ratio of firefly luciferase activity (pNFB-Luc or pPGS891LUC) to Renilla luciferase activity (pRL-SV40).
Statistical Analysis
Statistical analyses were conducted using GraphPad InStat software version 3.06 (GraphPad Software, San Diego, CA), employing one-way ANOVA and the Tukey-Kramer multiple comparisons test. A P value of 0.05 or less was considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The expression of COX-2 mRNA in WISH cells 1 h following challenge with EGF (10 ng/ml), IL-1ß (10 ng/ ml), or both was examined. Relative to untreated cells, incubation with EGF elicited a modest 11-fold increase in COX-2 mRNA expression on average (Fig. 1A). The response to IL-1ß was more robust, with an average 52-fold increase in COX-2 mRNA compared with untreated cells (Fig. 1A). Interestingly, we found that coincubation with EGF and IL-1ß resulted in an average 78-fold increase in COX-2 mRNA. The response to combined challenge was significantly greater than treatment with either factor independently (P < 0.001 vs. EGF, P < 0.05 vs. IL-1ß), suggesting synergy between EGF and IL-1ß in regulating COX-2 gene expression.
|
Blockade of NFB Selectively Attenuates IL-1ß-Induced COX-2 mRNA Expression
Like other proinflammatory cytokines, IL-1ß has been demonstrated to use the NFB cascade for COX-2 gene induction [6, 16, 17]. EGF also has been shown to elicit NFB activation in certain cell types [18, 19]. NFB exists as a dimer of Rel family proteins (prototypically p65/p50 or p50/p50) bound to an inhibitor (IB-) [20]. In the presence of IB-, the net intracellular distribution of the NFB complex is almost entirely cytoplasmic [20]. When activated, IB- is phosphorylated by IB kinases (IKKs) at two N-terminal serinyl residues (Ser32 and Ser36), triggering ubiquitylation and subsequent degradation of this inhibitor by the 26 S proteasome [21]. Through this mechanism, NFB nuclear localization signals are exposed and nuclear migration affords access of this transcription factor to B response elements.
To assess NFB activation, WISH cells in confluent monolayer culture were treated with IL-1ß (10 ng/ml) or EGF (10 ng/ml) and harvested for immunoblot analysis from 0 to 30 min. IL-1ß elicited activation of IKK and IKKß, as evidenced by transient phosphorylation at serinyl residues 180 of IKK and 181 of IKKß (Fig. 1B). This occurred concomitantly with phosphorylation of IB- at serinyl residue 32. The time-dependent loss of immunoreactive IB- between 10 and 30 min followed its site-specific phosphorylation (present at 5 min) and was consistent with proteasome-mediated degradation. By immunofluorescence, nuclear translocation of the NFB subunit p65 was seen 15 min following IL-1ß challenge (Fig. 1C, arrows). In contrast, EGF did not elicit activation of IKK or IKKß, and no phosphorylation or degradation of IB- was discerned (Fig. 1B). Furthermore, we detected no net nuclear translocation of p65 in response to EGF (Fig. 1C).
Use of the 26 S proteasome inhibitor MG-132 (N-acetyl leucinyl-leucinyl-leucinal) prohibits degradation of phosphorylated IB- and prevents NFB-mediated transcriptional activation [21, 22]. When WISH cells were preincubated with MG-132 (30 µM) before IL-1ß challenge, loss of IB- immunoreactivity at 15 min was blocked (Fig. 1D). The presence of nondegraded, phosphorylated IB- in these lanes was demonstrated by probing with a phospho-specific antibody (Fig. 1D). By immunofluorescence, it was confirmed that MG-132 prevented IL-1ß-induced nuclear translocation of the NFB subunit p65 (Fig. 1C). COX-2 mRNA expression induced by IL-1ß was attenuated by 77% in the presence of MG-132 (Fig. 1A). In contrast, MG-132 had no effect on EGF-driven COX-2 expression. Additionally, it was observed that MG-132 reduced COX-2 mRNA expression by only 42% in response to combined EGF/IL-1ß challenge (Fig. 1A). The expression of COX-2 mRNA following combined treatment with IL-1ß, EGF, and MG-132 was consistently greater than that observed in response to EGF alone, suggesting that signaling mechanisms other than NFB may contribute to EGF/IL-1ß cooperativity. Collectively, these results suggest that IL-1ß-mediated COX-2 gene expression is reliant on signaling through NFB, while additional transduction mechanisms are required for COX-2 induction following EGF or combined EGF/IL-1ß treatment.
Multiple Response Elements Are Required for Cooperative Increases in COX-2 Promoter Activity Following Combined EGF/IL-1ß Treatment
Analysis of the 5'-promoter region of the human COX-2 gene has revealed several potential transcription factor binding sequences, including those for NFB, nuclear factor-interleukin-6 (NF-IL6; also known as CCAAT/enhancer binding protein, C/EBP), activator protein-1 (AP-1), cyclic adenosine monophosphate response element binding protein (CREB), and the glucocorticoid receptor (Fig. 2A) [13, 23]. To examine the effects of EGF and IL-1ß on COX-2-promoter activity, cells were transiently transfected with a luciferase reporter plasmid containing a segment (–891/ +9) of the human COX-2 promoter (pPGS891LUC) [13]. This region contains many of the well-characterized transcription factor response elements (including CRE and two B motifs) but does not contain the AP-1 response element located –1587/–1579 base pairs (bp) upstream of the transcriptional start site (Fig. 2A) [23]. In cells transfected with pPGS891LUC, EGF (10 ng/ml) elicited an average 62% increase in reporter activity relative to untreated cells (P < 0.01; Fig. 2B). The response to IL-1ß (10 ng/ml) was only slightly greater, with an 80% increase in reporter activity on average relative to control cells (P < 0.01; Fig. 2B). Combined challenge with IL-1ß and EGF produced a cooperative (153%) increase in luciferase activity relative to untreated cells (P < 0.001; Fig. 2B).
|
To ascertain whether the combined effects of IL-1ß and EGF could be explained by augmented B-dependent transactivation, cells were transiently transfected with a luciferase reporter construct linked with five tandem repeats of the B response element (pNFB-Luc, Stratagene; Fig. 2A). In cells harboring pNFB-Luc, increased reporter activity was observed only in response to IL-1ß (P < 0.01 vs. control; Fig. 2C). This response was not augmented by coincubation with EGF (Fig. 2C). These results suggest that synergism between IL-1ß and EGF does not rely on increased NFB function, but rather, may occur through differential use of independent cis-acting response elements within the COX-2 promoter.
EGF and IL-1ß Transiently Activate Multiple Signal Transduction Pathways
Given that multiple response elements appear necessary for EGF/IL-1ß cooperativity, we next sought supplementary transduction mechanisms whereby these factors may increase COX-2 gene expression. In addition to NFB, IL-1ß has been shown to activate the three major MAP kinase pathways: p38 MAP kinase, extracellular-regulated kinases-1 and -2 (Erk-1/Erk-2), and c-Jun N-terminal kinases (JNK) [24]. Activation of the MAP kinase signaling pathways, particularly JNK and p38, can induce COX-2-promoter activity through recruitment of transcription factors to AP-1 and CRE consensus sequences [25, 26]. The signaling pathways governing COX-2 gene expression in response to EGF in WISH cells have not been clearly established. Based on other model systems, these may include the JAK/ STAT and/or MAP kinase cascades [27–29]. Although response elements to STATs within the human COX-2 5'-promoter have not been described, activated STATs have been shown to modulate the effects of other transcription factors relevant to the control of COX-2 gene activation [30, 31].
WISH cells were treated with IL-1ß (10 ng/ml) or EGF (10 ng/ml) and harvested for immunoblot analysis at time points from 0 to 30 min. Additionally, to examine whether activation of these cascades was altered in response to combination treatment, cells challenged with EGF and IL-1ß individually or together were harvested at a single point (12.5 min postchallenge). Activation of signaling intermediates was assessed through immunodetection of transiently phosphorylated intermediates.
Treatment with EGF, but not IL-1ß, resulted in phosphorylation of STAT1, STAT3, and STAT5 (Fig. 3A). Both EGF and IL-1ß caused transient phosphorylation of p38, JNK isoforms (JNK1 and -2), and Erk-1/Erk-2 (Fig. 3B). Across experiments, phosphorylation of p38 and Erk-1/Erk-2 was consistently greater following EGF treatment relative to stimulation with IL-1ß. Phosphorylation of JNK was more robust in response to IL-1ß at later time points (15 and 30 min), whereas at earlier time points (5 and 10 min), it was greater in response to EGF. Following combined EGF/IL-1ß treatment, there was no discernible potentiation of the amount of phosphorylated signaling intermediates among those examined at 12.5 min (Fig. 3C).
|
Inhibition of p38 MAP Kinase Eliminates EGF-Induced COX-2 mRNA Expression
Activation of Erk-1/Erk-2, JNK, and p38 precedes EGF- and IL-1ß-mediated COX-2 induction. Additionally, activation of STAT isoforms occurs acutely following EGF stimulation. To investigate the relationship of these cascades to COX-2 mRNA expression, the effects of specific pharmacological inhibitors on this process were investigated. Cells were pretreated for 1 h in the presence of 50 µM AG-490 (JAK/STAT pathway inhibitor [32]), 30 µM SB-202190 (p38 MAP kinase inhibitor [33]), 30 µM SP600125 (JNK inhibitor [34]), or 45 µM PD-98059 (MEK/Erk-1/ Erk-2 pathway inhibitor [35]) before challenge with EGF or IL-1ß (both at 10 ng/ml). Control cells received 0.8% dimethyl sulfoxide (DMSO, vehicle for AG-490, SB-202190, and PD-98059) or 0.6% ethanol (ETO, vehicle for SP600125) in the presence or absence of EGF or IL-1ß. Cells were harvested 1 h following challenge and COX-2 mRNA expression was detected by Northern blot analysis.
The expression of COX-2 mRNA in cells stimulated with EGF was blocked by SB-202190 (P < 0.01; Fig. 4A). SP600125 also had a small but discernible effect on EGF-induced COX-2 mRNA accumulation relative to its vehicle control group (P < 0.05; Fig. 4A). The other agents (AG-490 and PD-98059) were without significant effect, even at the relatively high doses used. The effect of SB-202190 on EGF-induced COX-2 mRNA expression was dose dependent (Fig. 4B). Complete blockade of COX-2 expression occurred only at concentrations of 30 µM or greater. At 10 µM, EGF-induced COX-2 expression was reduced by 48% on average, whereas doses 3 µM or below were without effect. IL-1ß-stimulated COX-2 mRNA expression was modestly attenuated by both SB-202190 (average decrease of 26%, P < 0.01) and SP600125 (average decrease of 21%, P < 0.01) relative to their respective vehicle controls (Fig. 4C), whereas the other agents showed no effect. Across experiments, it was noted that the DMSO-treatment group showed discernibly decreased COX-2 induction relative to cells exposed to EtOH, implying a slight vehicle effect (Fig. 4, A and C). By visual inspection through phase-contrast microscopy, there was no evidence of cytotoxicity in any of the treatment groups. These results indicate that p38 MAP kinase activity may contribute most substantially to EGF-driven COX-2 mRNA expression in WISH cells, although a minor role for JNK MAP kinase in this process is suggested. Furthermore, although NFB appears to be the major transduction mechanism through which IL-1ß induces COX-2 expression, these results suggest that downstream effectors of the JNK and p38 MAP kinases might also contribute. These results have been summarized in Figure 5.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Combined EGF/IL-1ß stimulation resulted in summative effects on COX-2 promoter activity, suggesting that cooperative increases in COX-2 mRNA expression are governed, at least in part, by enhanced COX-2 transcription. It is quite likely that IL-1ß-induced COX-2 mRNA accumulation results both from increased transcription as well as posttranscriptional mRNA stabilization [16, 36, 37]. The contribution of increased mRNA stability to COX-2 mRNA expression following combined EGF/IL-1ß treatment was not assessed in the present study. We speculate that increased mRNA stability may be an additional mechanism through which EGF potentiates IL-1ß-induced COX-2 mRNA expression.
In previous studies, it was shown that human COX-2 promoter constructs bearing mutations in one or both of the NFB sites had reduced reporter activity following IL-1ß stimulation in WISH cells [23, 37]. In this report, it was demonstrated that treatment with the 26 S proteasome inhibitor MG-132 blocked NFB activation and significantly attenuated IL-1ß-induced COX-2 mRNA expression in WISH. These results support NFB as a prominent signaling mechanism for IL-1ß-induced COX-2 mRNA expression. However, it was noted that MG-132 failed to completely attenuate IL-1ß-driven COX-2 mRNA expression. Although the residual COX-2 mRNA seen in this case could have been due to incomplete inhibition of the 26 S proteasome, our results suggest that additional transcription factors are likely to contribute to cytokine-induced COX-2 gene expression. Pharmacological inhibition of the both the p38 and JNK signaling pathways modestly abrogated IL-1ß-driven COX-2 expression. This is consistent with previous reports, in which a role for AP-1-binding transcription factors (including those activated by p38 and JNK) in IL-1ß-mediated COX-2 gene expression was suggested [23, 37].
The signaling pathways mediating EGF-induced COX-2 gene expression in WISH cells have not been clearly delineated, although the MAP kinase cascades have been implicated [29]. Our results suggest that EGF-mediated COX-2 mRNA expression in WISH cells does not require activation of NFB. Pharmacological blockade of p38 MAP kinase activity abrogated EGF-stimulated COX-2 gene expression in a dose-dependent fashion. These data are intriguing, particularly in light of the possible role for activated p38 in the process of term parturition in mice has been demonstrated [38]. For these experiments, the potent and selective p38 MAP kinase inhibitor SB-202190 was used. This agent has been employed widely to parse the physiological functions of p38 [33, 39]. In terms of specificity, previous studies have reported that SB-202190 does not inhibit other MAP kinases, such as JNK or Erk-1/Erk-2, even at concentrations as high as 100 µM [33, 39]. However, it is possible that nonspecific effects of this agent could have contributed to our findings. Further studies are needed to corroborate these results. Additionally, it was noted that blockade of JNK isoforms had a limited effect on EGF-mediated COX-2 mRNA expression, suggesting that JNK activation might also contribute to this process. Contrary to a previous report [29], we found no evidence that the MEK/Erk pathway was involved in EGF-dependent COX-2 mRNA expression. It remains unclear what downstream effectors of the p38 and JNK kinases might be important for EGF-mediated COX-2 transcriptional upregulation. We speculate that transcription factors activated by a p38-dependent mechanism (such as ATF-2 or C/EBP) might drive COX-2 transcription through CRE and/or NF-IL6 response elements within the proximal promoter [40, 41]. Additionally, activation of p38 could also contribute to COX-2 mRNA stability [42].
The signaling cascades responsible for EGF/IL-1ß synergism in COX-2 transcription are likely to be more complex than those required for either agent separately. Our current data suggest that COX-2 gene expression in response to combined challenge is only partially dependent on NFB. Furthermore, augmentation of IL-1ß-induced NFB activity by EGF does not appear to be a mechanism through which this synergism occurs. Activation of the COX-2 promoter appears to require differential use of independent response elements located in the region spanning –891/+9 bp relative to the transcriptional start site. This is likely to involve the effectors of multiple signaling pathways (particularly NFB, p38, and JNK), but may also stem from cooperativity among the activated signaling intermediates (e.g., STAT3 and AP-1 transcription factors can interact synergistically, even in the absence of a STAT3 enhancer element [43]), or cross-talk between EGF and IL-1 receptors [44]. Ongoing efforts in our laboratory will be aimed at further investigating the molecular mechanisms through which growth factor/cytokine cooperativity occurs.
We conclude that EGF and IL-1ß appear to signal through different transduction cascades leading to COX-2 gene expression in WISH cells. While IL-1ß employs the NFB pathway predominantly, the spectrum of EGF signaling is broader and includes p38 kinase.
EGF and IL-1ß act cooperatively to increase COX-2 mRNA expression, which is the likely basis for the greater-than-additive effects these agents have on PGE2 production [11, 12]. The synergism observed between IL-1ß and EGF in COX-2 transcriptional activation does not rely on augmented NF-B function, but rather, may occur through differential use of independent response elements within the COX-2 promoter.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Douglas A. Kniss, Laboratory of Perinatal Research, Department of Obstetrics and Gynecology, The Ohio State University, 5th Floor Means Hall, 1654 Upham Drive, Columbus, OH 43210. FAX: 614 293 5728; kniss.1{at}osu.edu
Received: 9 April 2004.
First decision: 28 April 2004.
Accepted: 23 July 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. E Ackerman IV, T. L.S Summerfield, D. D Vandre, J. M Robinson, and D. A Kniss Nuclear Factor-Kappa B Regulates Inducible Prostaglandin E Synthase Expression in Human Amnion Mesenchymal Cells Biol Reprod, January 1, 2008; 78(1): 68 - 76. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Chevillard, A. Derjuga, D. Devost, H. H Zingg, and V. Blank Identification of interleukin-1{beta} regulated genes in uterine smooth muscle cells Reproduction, December 1, 2007; 134(6): 811 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Liu, S.-C. Yang, S. Sharma, J. Luo, X. Cui, K. A. Peebles, M. Huang, M. Sato, R. D. Ramirez, J. W. Shay, et al. EGFR Signaling Is Required for TGF-beta1 Mediated COX-2 Induction in Human Bronchial Epithelial Cells Am. J. Respir. Cell Mol. Biol., November 1, 2007; 37(5): 578 - 588. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mizuno, H. Sotoyama, E. Narita, H. Kawamura, H. Namba, Y. Zheng, T. Eda, and H. Nawa A Cyclooxygenase-2 Inhibitor Ameliorates Behavioral Impairments Induced by Striatal Administration of Epidermal Growth Factor J. Neurosci., September 19, 2007; 27(38): 10116 - 10127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Djavani, O. R. Crasta, J. C. Zapata, Z. Fei, O. Folkerts, B. Sobral, M. Swindells, J. Bryant, H. Davis, C. D. Pauza, et al. Early Blood Profiles of Virus Infection in a Monkey Model for Lassa Fever J. Virol., August 1, 2007; 81(15): 7960 - 7973. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. Ackerman IV, J. M. Robinson, and D. A. Kniss Despite Transcriptional and Functional Coordination, Cyclooxygenase-2 and Microsomal Prostaglandin E Synthase-1 Largely Reside in Distinct Lipid Microdomains in WISH Epithelial Cells J. Histochem. Cytochem., November 1, 2005; 53(11): 1391 - 1401. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. E. Ackerman IV, X. L. Zhang, B. H. Rovin, and D. A. Kniss Modulation of Cytokine-Induced Cyclooxygenase 2 Expression by PPARG Ligands Through NF{kappa}B Signal Disruption in Human WISH and Amnion Cells Biol Reprod, September 1, 2005; 73(3): 527 - 535. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, P < 0.05 vs. IL-1ß; 
, P < 0.01 vs. EGF/ IL-1ß combined (ANOVA). B) Cells were treated with IL-1ß or EGF for 0–30 min and prepared for immunoblot analysis. Activation of NF
