|Home | About | Journals | Submit | Contact Us | Français|
Prostaglandin-E2 (PGE2) is a hormone derived from the metabolism of arachidonic acid whose functions include regulation of platelet aggregation, fever and smooth muscle contraction/relaxation. PGE2 mediates its physiological and pathophysiological effects through its binding to four G-protein coupled receptor subtypes, named EP1, EP2, EP3 and EP4. The EP3 prostanoid receptor is unique in that it has multiple isoforms generated by alternative mRNA splicing. These splice variants display differences in tissue expression, constitutive activity and regulation of signaling molecules. To date there are few reports identifying differential activities of EP3 receptor isoforms and their effects on gene regulation. We generated HEK cell lines expressing either the human EP3-Ia, EP3-II or EP3-III isoforms. Using immunoblot analysis we found that nM concentrations of PGE2 strongly stimulated the phosphorylation of ERK 1/2 by the EP3-II and EP3-III isoforms; whereas, ERK 1/2 phosphorylation by the EP3-Ia isoform was minimal and only occurred at μM concentrations of PGE2. Furthermore, the mechanisms of the PGE2 mediated phosphorylation of ERK 1/2 by the EP3-II and EP3-III isoforms were different. Thus, PGE2 stimulation of ERK 1/2 phosphorylation by the EP3-III isoform involves activation of a Gαi/PI3K/PKC/Src and EGFR-dependent pathway; while for the EP3-II isoform it involves activation of a Gαi/Src and EGFR-dependent pathway. These differences result in unique differences in the regulation of reporter plasmid activity for the downstream effectors ELK1 and AP-1 by the EP3-II and EP3-III prostanoid receptor isoforms.
Prostaglandin E2 (PGE2) is an autocrine and paracrine hormone derived from the metabolism of arachidonic acid via the activities of cyclooxygenase. PGE2 mediates its physiological and patho-physiological effects through its four G-protein coupled receptor subtypes EP1–4. Four distinct genes encode the EP receptor subtypes and each receptor couples uniquely to G-proteins. In humans, the EP1 receptor couples to Gq, EP2 to Gs, EP3 to Gαi, Gαs, Gαq and Gα12/13 and EP4 couples to Gαs and Gαi [1–4]. The EP3 receptor is unique among the EP receptor subtypes, in that there are multiple isoforms generated through alternative mRNA splicing in the carboxyl tail of the EP3 receptor gene. Thus far 10 mRNA splice variants of the human EP3 receptor have been identified [4–8]. Previous studies have shown that the differences in the carboxyl tail impart differences in constitutive activity, G-protein coupling and agonist induced internalization [5, 9, 10]. Evidence of unique signal transduction pathways and regulation of gene expression among individual receptor isoforms has also been demonstrated in both published and unpublished studies [5, 11, 12].
Physiologically, the EP3 receptor is important in a number of functions including the febrile response, gastro-duodenal bicarbonate secretion, vasoconstriction of the pulmonary arteries, growth inhibition in keratinocytes and inhibition of aromatase activity in breast fibroblasts [13–18]. Its expression has mostly been observed in clusters of multiple isoforms rather than singly. In human uterus, mRNAs for EP3-V and EP3-VI receptor isoforms have been detected, whereas, in primary keratinocytes EP3A1, EP3C and EP3D splice variants are expressed [19, 20]. The EP3A1, EP3C and EP3D splice variants correspond to EP3-Ia, EP3-II and EP3-IV respectively, according to the Kotani nomenclature .
Studies of the activities of EP3 receptors have uncovered a number of discrepancies especially related to its effects on proliferation and differentiation. An inhibitory role of EP3 signaling on proliferation has been described using the EP3 receptor agonist GR 63799X. GR 63799X inhibits the proliferation of keratinocytes and HCA-7 colon cancer cells and induces 3T6 fibroblast growth arrest [13, 19, 22, 23]. In contrast, EP3 receptor activation has also been reported to induce the proliferation of hepatocytes, endometrial stromal cells and A549-adenocarcinoma cells [20, 24, 25]. The divergent data regarding the influences of EP3 receptor activation on proliferation indicate that EP3 receptors may have different effects based on the tissue type in which they are expressed. However, few studies on EP3 receptor activity have identified the specific receptor isoforms expressed, therefore, the divergent effects attributed to EP3 receptor activation may actually be due to differences in isoform expression.
Mitogen activated protein kinase (MAPK) pathway activation is often associated with survival signals and increased cellular proliferation. The p44/p42 MAPKs, also known as ERK 1/2, are serine/threonine kinases activated by extracellular stimuli, typically growth factors, and involve the sequential activation of Ras, Raf and MEK. ERK 1/2 activities include regulation of cell proliferation, differentiation, survival and apoptosis. Upstream kinases modulating the activation status of ERK 1/2 include protein kinase C (PKC), phosphoinositide-3 kinase (PI3K), protein kinase A (PKA) and transactivation by receptor tyrosine kinases such as the epidermal growth factor receptor (EGFR) [26–28].
Previously, we described the activation of the ERK 1/2 pathway by the EP3-Ia receptor expressed in COS-7 cells . Here we characterize the differential mechanisms of ERK 1/2 pathway activation by three human EP3 receptor isoforms stably expressed in human embryonic kidney cells (HEK). We also show that these differences in ERK 1/2 signaling mechanisms result in differences in the activation of both ELK1 and AP-1 transcription. ELK1 is an Ets-related transcription factor activated by MAPK that binds to the serum response element of a variety genes that regulate growth, including c-fos. AP-1 is a transcription factor complex made up of heterodimers of members Jun (c-jun, JunB and JunD) and Fos (c-fos, fosB, Fra-1 and Fra-2) family of proteins. Like ELK1, AP-1 is regulated by MAPK phosphorylation and binds to serum response element (SRE) to mediate the transcription of proteins important for cell growth and survival [3, 29–31]. Our present studies suggest a mechanism by which EP3 receptor mediated signaling may be fine-tuned by regulation of isoform expression. Differences in the expression patterns of EP3 receptor isoforms may help clarify the disparate data in the literature.
Dulbecco’s Modified Eagle’s Medium (DMEM), DMEM without phenol red, OptiMEM, G418 and gentamicin sulfate were from Mediatech (Herndon, VA). Fetal bovine serum (FBS), hygromycin B, pCEP4 vector and HEK-EBNA cells were from Invitrogen (Carlsbad, CA). QIAQUICK gel purification kit was from Qiagen (Venlo, The Netherlands). T4 DNA ligase was from New England Biolabs (Ipswich, Massachusetts). Fugene 6 transfection reagent was from Roche Diagnostics (Indianapolis, IA). Horseradish peroxidase conjugated anti-rabbit IgG antibody, Dual Luciferase Assay and Promega CellTiter 96® AQueous Non-Radioactive Cell Proliferation Assay kits were from Promega (Madison, WI). Horseradish peroxidase-conjugated anti-mouse IgG and nitrocellulose membranes were from Bio-Rad Laboratories (Hercules, CA). Antibodies against ERK 1/2 and phospho-ERK 1/2 were from Cell Signaling Technology (Waltham, MA). AP-1 and ELK1Trans-Reporting System were from Stratagene (La Jolla, CA). PGE2 and sulprostone were from Cayman Chemical Company (Ann Arbor, MI). Enhanced chemiluminescence substrate was from Pierce (Rockford, IL). PD98059, phorbol-12-myristate-13-acetate (PMA), forskolin, AG1478, bisindolylmaleimide I, pertussis toxin and PP2 were from Calbiochem (San Diego, CA). [3H]PGE2 and [3H]cAMP were from Amersham/GE Healthcare (Buckinghamshire, England). Protein kinase A, 3-isobutyl-1-methylxanthine (IBMX) and cAMP were from Sigma (St. Louis, MO).
The EP3 receptor isoforms were amplified from the pBC plasmid constructs previously generated by polymerase chain reaction (PCR) . The following primers were used: a common EP3 sense primer includes nucleotides 1–23 (bold) and contains a HindIII site (italics): 5′-ATTATTAAGCTTATGAAGGAGACCCGGGGCTACGG-3′. All EP3 anti-sense primers contain an Xho-1 site (italics). EP3-Ia anti-sense primer represented nucleotides 1154–1173 (bold) 5′-CCGCCGCTCGAGTTATCTTTCCAAATGGTCGC-3′. EP3-II anti-sense primer represented nucleotides 1143–1167 (bold) 5′-CCGCCGCTCGAGTCATGCTTCTGTCTGTATTATTTC-3′. EP3-III anti-sense primer represented nucleotides 1077–1098 (bold). 5′-CCGCCGCTCGAGTTAATTTCCCCAAAATTCCTCC-3′. PCR products were electrophoresed on 1.5% agarose gel, excised and purified using QIAQUICK gel purification kit. Following purification, the individual EP3 receptor PCR products were sequenced at the University of Arizona Genomic Analysis and Technology Core facility. After sequence confirmation, the individual EP3 receptor fragments and the pCEP4 expression vector were digested using the restriction enzymes Xho1 and Hind III. EP3 receptor fragments were then ligated into the pCEP4 vector using T4 DNA ligase. The plasmids generated were named as follows: pCEP4/EP3-Ia, pCEP4/EP3-II and pCEP4/EP3-III.
HEK-EBNA cells were grown in Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% fetal bovine serum (FBS), 250 μg/ml G418 and 100 μg/ml gentamicin sulfate and incubated at 37°C and 5% CO2. Transfected cells HEK-EBNA cells were grown in the same media as above, but with 200 μg/ml Hygromycin B. One million HEK cells were seeded in 10cm plates and the following day transfected with 10 μg pCEP4/EP3-Ia, pCEP4/EP3-II and pCEP4/EP3-III plasmid DNA using Fugene 6 transfection reagent. One day after transfection, cells were selected for pCEP4/EP3 receptor expression using 200 μg/ml hygromycin B. Approximately one week after hygromycin selection, 3–5 clones for each receptor isoform were expanded and examined for functional analysis. Whole cell radioligand binding, cAMP assay and inositol phosphate assays were used to select a clone for each EP3 receptor isoform.
Competition binding experiments were conducted as previously described by Regan et al. using HEK-EBNA cells stably expressing either the pCEP4 plasmid alone, pCEP4/EP3-Ia, pCEP4/EP3-II or pCEP4/EP3-III . Assays were conducted by incubating whole cells for one hour at room temperature in a final concentration of 2.5 nM [3H]PGE2.
HEK 293 cells stably expressing pCEP4 plasmid alone, pCEP4/EP3-Ia, pCEP4/EP3-II and pCEP4/EP3-III were seeded into 6 well plates at a density of one million cells per well in normal culture media and incubated overnight at 37°C, 5% CO2. In experiments using pertussis toxin, cells were dosed with 100 ng/ml PTX 8 hrs after seeding cells and 16 hrs before experiments were conducted. The following day cells were rinsed twice with phosphate buffered saline (PBS) then incubated with 2 mls Opti-MEM reduced serum media containing 0.1 mg/ml of the phosphodiesterase inhibitor, IBMX, for 15 mins at 37°C, 5% CO2. After IBMX treatment cells were incubated with DMSO or sulprostone for 1 hour followed by treatment with 3 μM forskolin at 37°C, 5% CO2. Following drug treatment, media was aspirated and 720 μl cold Tris/EDTA buffer (TE) (50 mM Tris/HCl, 4 mM EDTA, pH 7.5) was added to each well. Samples were scraped and cell lysates were transferred into labeled screw cap microcentrifuge tubes and boiled for 8 mins. Samples were then placed on ice for 5 mins then tubes were centrifuged 12,000 rpm for 2 mins. A standard curve was prepared to determine the concentration of cAMP in each sample. The standard curve reactions were prepared as follows in microcentrifuge tubes; 50 μl of 0.125-64 picomoles of cAMP diluted in TE or TE alone, 50 μl of 0.9 pmol [3H]cAMP diluted in TE, 100 μl of 0.06 mg/ml protein kinase A diluted in 1% BSA or 1% BSA alone. Experimental sample reactions were prepared similarly except 50 μl of cell lysates was added instead of untritiated cAMP. Reactions were incubated for two hrs at 4°C. Following reaction incubation, 100 μl of 2.5 mg/ml activated charcoal in 2% BSA was added to each microcentrifuge tube. Samples were vortexed, then centrifuged at room temperature for 1 minute, and 200 μl of supernatant was transferred into labeled scintillation vials containing 9 ml of Safety-Solve scintillation cocktail. Vials of samples were vortexed then radioactivity was detected using a Beckman LS100 scintillation counter.
Forty-eight hrs prior to experiments ~0.5 × 106 HEK cells expressing EP3-Ia, EP3-II and EP3-III were seeded into 10 cm plates and incubated at 37°C and 5% CO2. On the day of the experiment, cells were treated with either vehicle (DMSO) or PGE2 or phorbol-12-myristate-13-acetate for 10 mins. Following treatment cells were lysed on ice with 300 μl RIPA buffer containing 150 mM NaCl, 50 mM Tris-HCl (pH 7.5), 5 mM EDTA, 1% NP-40, 0.5% sodium deoxycholate, 10 mM sodium fluoride, 10 mM disodium pyrophospate, 0.1% SDS, 1 mM phenylmethylsulphonyl fluoride (PMSF), 1 mM sodium orthovanadate, 10 mg/ml leupeptin and 10 μg/ml aprotonin. Samples were sonicated on ice then centrifuged at 15,000 g for 15 mins. Fifty μg of sample lysates were electrophoresed on 7.5% SDS-polyacrylamide gels and transferred to nitrocellulose membranes. Membranes were blocked for one hour in 5% nonfat milk in tris buffered saline and 0.1% Tween 20 (TBS-T) at room temperature. Membranes were then incubated at 4°C overnight with anti-phospho-ERK 1/2 primary antibody diluted in 5% non-fat milk in TBS-T. Following incubation in primary antibody, membranes were washed three times with TBS-T, then incubated with anti-rabbit secondary antibody conjugated with horseradish peroxidase in 5% nonfat milk TBS-T for one hour at room temperature. After washing membrane three times, immunoreactivity was detected using SuperSignal enhanced chemiluminescence. Membranes were stripped using 2% SDS, 62.5 mM Tris (pH 7.6) and 100 mM β-mercaptoethanol for 30 mins at 55°C, then re-probed with an anti-ERK 1/2 antibody to determine total protein loading.
HEK-EBNA cells stably expressing EP3-Ia, EP3-II and EP3-III isoforms were seeded at ~0.35 × 106 cells in triplicate in 6-well plates. Twenty-four hrs later cells were transfected with 50 ng ELK1 hybrid fusion protein containing the DNA binding domain of the yeast transcription factor Gal4 plasmid, 1 μg Gal4-luciferase plasmid and 12.5 ng of renilla (pRL-TK) plasmid. For AP-1 experiments, 1 μg AP-1 luciferase reporter plasmid and 10 ng of pRL-TK plasmid were used. Cells were pre-treated with either vehicle or MAPK inhibitors for 30 mins followed by treatment with 50 nM PGE2 for 16–24 hrs. Cells were then lysed with 200 μl Passive Lysis Buffer® (Promega, Madison WI), centrifuged at 15,000g and supernatant transferred to fresh tubes. Samples were analyzed on a Turner Design TD-20 Dual-Luciferase luminometer. In AP-1 and ELK1 experiments, renilla luciferase activities for each individual transfection were used to normalize firefly luciferase activity.
All data are presented as the mean ± SEM. Statistical analyses were performed using one-way ANOVA with Bonferroni post hoc test using GraphPad Prism software.
Radioligand binding experiments with [3H]PGE2 were undertaken to characterize the parameters of PGE2 binding to the human EP3 receptor isoforms stably expressed in HEK cells. Cells were prepared and incubated with [3H]PGE2 and either vehicle (DMSO) or final concentrations of 1 nM, 10 nM, 100 nM, 1 μM, or 10 μM nonradioactive PGE2. Figure 1A shows the results for binding to cells expressing the EP3-Ia, EP3-II and EP3-III receptor isoforms and for cells stably transfected with the empty vector. The IC50 values for EP3-Ia, EP3-II and EP3-III receptor isoforms were 23 nM, 69 nM and 29 nM, respectively. The maximal binding (Bmax) of [3H]PGE2 to cells expressing EP3-Ia, EP3-II and EP3-III were 857 fmol/mg protein, 1169 fmol/mg protein and 1087 fmol/mg protein, respectively. There was no specific binding of [3H]PGE2 to cells stably transfected with the vector alone.
The EP3 receptor is often characterized as Gαi coupled and inhibits the activation of the enzyme, adenylyl cyclase, and reduces intracellular cAMP levels. To assess Gαi coupling potential following activation of the EP3 receptor isoforms, cAMP assays were completed as described in Methods. Following pre-treatment with the EP3 agonist, sulprostone, activation of all three isoforms inhibits forskolin stimulated cAMP production. Figure 1B shows that in all EP3 receptor isoform cell lines forskolin treatment increases cAMP levels above those observed with vehicle treatment alone. Pre-treatment with sulprostone inhibits forskolin stimulated cAMP production in all three cells lines, but the inhibition is greatest in the EP3-Ia and EP3-III expressing cells. Inhibition of Gαi with pertussis toxin pre-treatment reverses the effects of sulprostone on forskolin stimulated cAMP production. In EP3-II expressing cells, pertussis toxin pre-treatment increased cAMP levels above the levels observed in forskolin treated cells. Sulprostone treatment, alone, had no effect on cAMP production.
Immunoblot assays were utilized to assess possible differences in the phosphorylation of ERK 1/2 among the cells expressing the EP3 receptor isoforms. Phorbol-12-myristate 13 acetate (PMA) induces PKC activation and was used as a positive control to ensure all cell lines were capable of inducing ERK 1/2 phosphorylation. As shown in Figure 2, treatment of cells expressing the EP3-II isoform with PGE2 resulted in robust phosphorylation of ERK 1/2 at all concentrations tested. For cells expressing the EP3-III isoform, treatment with PGE2 clearly stimulated phosphorylation of ERK 1/2, but at lower levels than those observed for the EP3-II isoform. For cells expressing the EP3-Ia isoform, there was no significant PGE2 induced phosphorylation of ERK 1/2, especially as compared with cells expressing the EP3-II or EP3-III isoforms. Additionally, it should be noted that while the autoradiographs for the EP3-II and EP3-III isoforms were exposed for similar lengths of time, the autoradiographs for the EP3-Ia isoform had to be exposed 5–10 times longer in order to see anything on the film. Thus, at physiological concentrations of PGE2, in the range of 1–100 nM, the EP3-II and EP3-III receptor isoforms clearly signal through the activation of ERK 1/2; whereas, the EP3-Ia isoform does not. Further experiments, therefore, utilized 50 nM PGE2 to determine the mechanism of this activation and focussed solely on the EP3-II and EP3-III isoforms.
To examine the pathways responsible for PGE2 mediated phosphorylation of ERK 1/2 by the EP3-II and EP3-III receptor isoforms, we utilized several well characterized pharmacological inhibitors of signaling pathways that could potentially be activated by the EP3 receptor isoforms. Figure 3 shows the results of immunoblot assays for the PGE2 mediated phosphorylation of ERK 1/2 following the pre-treatment of cells expressing the EP3-II and EP3-III isoforms with the various pathway inhibitors. The data include both a representative autoradiograph and the results of densitometry with the data normalized to PGE2 stimulated cells that were not pre-treated with inhibitors. As can be seen, pre-treatment of cells with either the PKC inhibitor, BIM I, or the PI3K inhibitor, wortmannin, strongly decreased ERK 1/2 phosphorylation induced by the EP3-III isoform, but not by the EP3-II isoform. Inhibition of Gαi by pre-treatment of the cells with pertussis toxin (PTX) completely abrogated PGE2 stimulated ERK 1/2 phosphorylation mediated by both the EP3-II and EP3-III isoforms. These data indicate that both isoforms couple to Gαi, but that the EP3-II isoform does not utilize the PKC and PI3K pathways to induce ERK 1/2 phosphorylation following treatment with PGE2 in this model system. Pre-treatment of the cells with genistein, a nonspecific inhibitor of tyrosine kinases had little effect on the PGE2 stimulated phosphorylation of ERK 1/2; whereas, selective inhibition of the Src tyrosine kinase with PP2 completely blocked PGE2 stimulated ERK 1/2 phosphorylation by both receptor isoforms. Selective inhibition of the EGFR tyrosine kinase with AG1478 partially blocked PGE2 stimulated ERK 1/2 phosphorylation mediated by the EP3-II and EP3-III isoforms. These data show that the EP3-II and EP3-III receptor isoforms did not differ with respect to the potential involvement of receptor tyrosine kinases in PGE2 stimulated phosphorylation of ERK 1/2, but there was an absolute requirement for the activation of Scr by both isoforms. Pre-treatment with the PKA inhibitor, H-89, did not result in any significant inhibition of PGE2 stimulated ERK 1/2 phosphorylation in cells expressing either the EP3-II or EP3-III isoforms (data not shown).
In Figure 4 luciferase reporter assays were used to investigate whether differences in the mechanism of the PGE2 stimulation of ERK 1/2 phosphorylation by the EP3-II and EP3-III isoforms translated into differences in the activation of the downstream targets ELK1 and AP-1. For these experiments the same pharmacological inhibitors used in Figure 3 were again employed to determine which pathways contributed to activation of ELK1 and AP-1 reporter gene activity. As shown in Panel A of Figure 4, treatment of cells with 50 nM PGE2 resulted in a significant stimulation of both ELK1 and AP-1 luciferase activity in cells expressing the EP3-II and EP3-III isoforms, but not in cells expressing the EP3-Ia isoform. These results agree nicely with the PGE2 stimulated phosphorylation of ERK 1/2 by the three isoforms as shown in Figure 2; thus, in the range of 10–100 nM, PGE2 stimulated ERK 1/2 phosphorylation in cells expressing the EP3-II and EP3-III isoforms, but not in cells expressing the EP3-Ia isoform.
As shown in Panel B of Figure 4, pre-treatment of cells with the inhibitors PD98059, AG1478, PP2 and pertussis toxin, significantly reduced PGE2 stimulated ELK1 luciferase reporter activity in EP3-II and EP3-III isoform expressing cells. ELK1 luciferase activity in cells expressing the EP3-III isoform was also significantly inhibited by pre-treatment with wortmannin and BIM I. For both EP3-II and EP3-III expressing cells, ELK1 luciferase reporter activity was MAPK dependent and mirrored the results obtained in Figure 3 for the PGE2 mediated induction of ERK 1/2 phosphorylation by these isoforms.
Again, as shown in Figure 4B, pre-treatment of cells with PD98059, BIM I, PP2 and pertussis toxin significantly decreased AP-1 luciferase reporter activity in EP3-II and EP3-III expressing cells. Although the PGE2 stimulation of AP-1 luciferase activity clearly involved the activation of MAPK signaling, since it was inhibited by pre-treatment with PD98059, there were several unexpected findings. First, the PKC inhibitor, BIM I, failed to reduce PGE2 mediated ERK 1/2 phosphorylation in EP3-II expressing cells (Figure 3), but nevertheless significantly inhibited AP-1 lucifierase reporter activity. Also, wortmannin pre-treatment failed to inhibit EP3-III activation of AP-1 luciferase reporter activity, whereas, it reduced PGE2 dependent ERK 1/2 phosphorylation by 50% (Figure 3). Finally, AG1478 pre-treatment partially inhibited ERK 1/2 phosphorylation in both EP3-II and EP3-III expresssing cells, but had no effect on PGE2 stimulated AP-1 luciferase activity in these cells.
Several laboratories have undertaken the task of elucidating the functional differences between the EP3 receptor isoforms. Because the EP3 receptor isoforms are usually found expressed in multiple combinations, rather than singly, it has been nearly impossible to identify the signaling properties of the individual isoforms without the use of recombinant expression systems. Here we show that in HEK cells stably expressing either the human EP3-Ia, EP3-II or EP3-III isoforms, there are distinct differences in the PGE2 mediated activation of ERK 1/2 phosphorylation. In addition, these differences in the induction of ERK 1/2 phosphorylation appear to translate into differences in the activation of gene expression by these isoforms, as evidenced by their stimulation of AP-1 and ELK1 luciferase reporter gene activity. For example, in cells expressing the EP3-Ia isoform, the induction of ERK 1/2 phosphorylation by PGE2 was negligible as compared with cells expressing the EP3-II and EP3-III isoforms. Likewise, there was no significant PGE2 mediated stimulation of AP-1 or ELK1 reporter gene activity in cells expressing the EP3-Ia isoform; whereas, AP-1 and ELK1 reporter gene activity was markedly stimulated by PGE2 in cells expressing the EP3-II and EP3-III isoforms. These signaling differences were not due to significant differences in receptor number as the Bmax values for the three cell lines were very similar (857 fmol/mg protein, 1169 fmol/mg protein and 1087 fmol/mg protein for cells expressing the EP3-Ia, EP3-II and EP3-III, respectively).
In addition to signaling differences between the EP3-Ia isoform versus the EP3-II and EP3-III isoforms, the mechansims underlying the PGE2 mediated induction of ERK 1/2 phosphorylation by the EP3-II and EP3-III isoforms also differed. Thus, both the EP3-II and EP3-III isoforms utilize a pathway involving a pertussis toxin-sensitive activation of MAPK signaling, which is dependent on the activation of Src and EGFR. However, the induction of ERK 1/2 phosphorylation by the EP3-III isoform is additionally dependent upon the activation of PKC and PI3K signaling, whereas, this is not the case for the EP3-II isoform.
The involvement of Gαi, PKC, Src and EGFR demonstrated here for the induction of ERK 1/2 phosphorylation by the EP3-III isoform has previously been shown for the induction of ERK 1/2 phosphorylation by the thromboxane-A2 (TxA2) prostanoid receptor . However, unlike the TxA2 receptor, the phosphorylation of ERK 1/2 mediated by the EP3-III receptor is also dependent on the activation of PI3K signaling. In contrast, the induction of ERK 1/2 phosphorylation by the EP3-II isoform is more similar that observed for the lysophosphatidic acid receptor [33, 34]. Thus, both receptors utilize a mechanism involving the activation of Gαi, Src and EGFR that is independent of the activation of PKC and PI3K signaling.
We have found that the stimulation of ELK1 and AP-1 luciferase reporter gene activity by both the EP3-II and EP3-III receptor isoforms is dependent upon the activation of MAPK signaling. Thus, pre-treatment with the MEK inhibitor, PD98059, significantly decreased the stimulation of ELK1 and AP-1 luciferase activity by PGE2. As expected, many of the same signaling pathway inhibitors that decreased the PGE2 mediated phosphorylation of ERK 1/2 also inhibited PGE2 stimulated AP-1 and ELK1 luciferase activity in cells expressing the EP3-II and EP3-III isoforms. The EP3-II receptor isoform was found to activate ERK 1/2 and subsequently ELK1 via a Gαi-Src/EGFR dependent pathway, but its activation of AP-1 was found to involve a Gαi-Src-PKC dependent pathway. On the other hand, the EP3-III isoform was found to activate ERK 1/2 and ELK1 by a Gαi-Src/EGFR-PKC-PI3K dependent pathway and its activation of AP-1 was essentially by the same mechanism, except for the involvement of PI3K. The inability of the EP3-Ia receptor isoform to stimulate AP-1 and ELK1 luciferase activity is not surprising given its failure to stimulate the phosphorylation of ERK 1/2 at low concentrations of PGE2. Even at higher concentrations of PGE2, stimulation of AP1 and ELK1 would be unlikely since it has been found that AP-1 components, such as FRA-1, require elevated and sustained levels of ERK 1/2 activity for successful activation of transcription [35, 36].
In conclusion, the present studies have explored the mechanisms of the PGE2 mediated phosphorylation of ERK 1/2 by the EP3-Ia, EP3-II and EP3-III receptor isoforms. Our findings contribute to a better understanding of the signaling differences between the EP3 receptor isoforms that underlie their involvement in the regulation physiological systems. Of particular significance is the potential of these receptor isoforms to differentially regulate gene transcriptional activity.
This work was supported by the National Institutes of Health (grant EY11291) and Allergan Inc.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.