Prostaglandin E2 Regulates the Nuclear Receptor NR4A2 in Colorectal Cancer*

Many lines of research implicate cyclooxygenase 2-derived prostaglandins in tumor growth and metastasis. More specifically, we have shown that prostaglandin E2 (PGE2) promotes cell proliferation and invasion through transactivation of the epidermal growth factor receptor, initiates immune evasion through induction of decay accelerating factor, and transactivates peroxisome proliferator-activated receptor δ, leading to increased polyp size and multiplicity. We continue to identify novel PGE2 target genes in colorectal carcinoma cells and report here that an immediate early gene, nuclear factor NR4A2 (Nurr1), is induced by PGE2 that in turn regulates cell death. Originally described as a critical dopaminergic neuron growth factor receptor, NR4A2 expression is rapidly but transiently induced by PGE2 in a cAMP/protein kinase A-dependent manner. NR4A2 binds to the cognate NBRE response element and enhances transcription of a reporter construct in colorectal carcinoma cells. Furthermore, NR4A2 expression is elevated in Apc-/+ mouse adenomas and its levels were further increased following PGE2 treatment. Human colorectal cancers relative to matched normal mucosa showed increased NR4A2 expression. Although not previously described in epithelial tissues, NR4A2 protein localizes to proliferating crypts of Apc-/+ mouse intestine. Finally, functional studies reveal that PGE2-mediated protection from apoptosis is completely inhibited by a dominant-negative NR4A2 construct. Building on previous reports from our group on the peroxisome proliferator-activated receptor family of nuclear receptors, these most recent data suggest that NR4A2, a member of another family of nuclear receptors can stimulate progression of colorectal cancer downstream from cyclooxygenase 2-derived PGE2.

PGE 2 , the most abundant prostaglandin in colorectal cancer (19), can be generated from arachidonic acid by either COX-1 or COX-2. PGE 2 signals by binding to four distinct G protein-coupled cell surface receptors (EP1-EP4): leading to increased calcium flux (EP1), increased (EP2 and -4) or decreased (EP3) cAMP levels all of which modulate downstream networks (20). Our laboratory is actively investigating precisely which genes are regulated by PGE 2 in colorectal cancer and during embryogenesis (21). We now show that expression and activity of NR4A2, a nuclear receptor superfamily member, are induced in colorectal carcinoma cells by PGE 2 .
Transcription factors in the nuclear receptor superfamily regulate gene expression upon ligand binding (22,23). Well known steroid receptors including progestins, estrogens, androgens, glucocorticoids, and mineralcocorticoids comprise the type I group. Conversely, receptors of thyroid hormone, all-trans-retinoic acid, 9-cis-retinoic acid, and vitamin D 3 (VDR) belong to the type II group. The orphan receptors comprise a third class for which physiologic ligands have yet to be identified. The type III group includes, among others, peroxisome proliferator-activated receptors and the NR4A family of receptors.
Based on the expression pattern in the brain, NR4A family members have been strongly implicated in Parkinson disease (31), schizophrenia (32), manic depression (33), and Alzheimer disease (34). NR4A2 is important for dopaminergic neuron function via regulation of tyrosine hydoxylase expression (35). Nurr1 Ϫ/Ϫ mice lack mesencephalic dopaminergic neurons, which are known to degenerate in Parkinson disease (36,37). Preliminary reports suggest a role for this family of receptors in rheumatoid arthritis and cancer through modulation of apoptosis (38). Other functions associated with NR4A2 include regulation of osteocalcin in osteoblasts (39,40), aldosterone synthase in adrenal cortex (41), and aromatase in ovarian granulosa cells (42). With regard to aromatase, our results may help explain another mechanism by which PGE 2 regulates its expression in certain contexts.
We seek a more complete understanding of the role of prostaglandins and downstream targets in epithelial biology, developmental biology, and colorectal carcinogenesis. We have identified key genes that are regulated by PGE 2 in colorectal carcinoma cells. Here we present data suggesting that the nuclear receptor NR4A2 is regulated by PGE 2 in colorectal cancer. This is the first demonstration that prostaglandins regulate this family of transcription factors in neoplastic cells. Ultimately, this novel observation may shed light on the precise role of PGE 2 in colorectal carcinogenesis.

MATERIALS AND METHODS
Reagents-PGE 2 was obtained from Cayman Chemical (Ann Arbor, MI). LY294002 and H-89 were purchased from Calbiochem. Antibodies to NR4A were purchased from Santa Cruz Biotechnology (Santa Cruz, CA) (catalog number SC-990) and R&D Systems (Minneapolis, MN) (catalog number AF2156). ␤-Actin antiserum was obtained from Sigma.
Northern Blotting-Total cellular RNA was isolated from cells by TRI Reagent (Molecular Research Center, Cincinnati, OH) following the manufacturer's protocol. Five g of total RNA was fractionated with a MOPS/formaldehyde-agarose gel and transferred to Hybond N1 membrane (Amersham Biosciences). Following UV cross-linking, the blots were pre-hybridized for 30 min at 42°C in Hybrisol I (Intergen Company, Purchase, NY), hybridized using 32 P-labeled cDNA in the same buffer at 42°C, and subjected to autoradiography. The 0.35-kb NR4A2 (NM_006186) probe was amplified by reverse transcriptase-PCR using primers 5Ј-CACAGGTTGCAATGCGTTCG-3Ј (sense: 1746 -1765) and 5Ј-TCAATTATTGCTGGCGGTGG-3Ј (antisense: 2089 -2105).
Quantitative Real-time PCR-cDNA for each RNA sample was synthesized in 20-l reactions using the SuperScript First Strand synthesis system for reverse transcriptase-PCR (Invitrogen) following the manufacturer's protocol. Primers for PCR were designed using Primer3 software. The real-time PCR contained iQ SYBR Green supermix (Bio-Rad), 50 ng of each primer, and 5 l of 1/1000 diluted reverse transcriptase template in a 25-l reaction volume. Amplification was carried out using the MyiQ Single Color Real-time PCR Detection system (Bio-Rad), with incubation times of 3 min at 95°C, followed by 50 cycles of 95°C for 10 s and 63°C for 30 s. Specificity of the amplification was checked by melt-curve analysis. Relative levels of mRNA expression were calculated according to the ⌬⌬C T method (BMC Biotechnol). Individual expression values were normalized by comparison with ␤-actin mRNA expression. Oligonucleotide sequences used were: human NR4A2 forward, 5Ј-GTCTCAGCTGCTCGACACG-3Ј and reverse, 5Ј-TTTTGCACTGTGCGCTTAAA-3Ј; mouse NR4A2 forward, 5Ј-GCT-TACAGGTCCAACCCAGT-3Ј and reverse, 5Ј-AATGCAGGAGAA-GGCAGAAA-3Ј; ␤-ACTIN forward, 5Ј-AGAAAATCTGGCACCAC-ACC-3Ј and reverse, 5Ј-AGAGGCGTACAGGGATAGCA-3Ј.
Western Blotting-Cells were washed with phosphate-buffered saline and lysed with RIPA buffer (50 mM Tris-Cl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and protease inhibitors from Roche Diagnostics). Protein concentrations were measured using Bio-Rad reagent (Bio-Rad). Proteins were then separated on precast SDS-PAGE gels and electrotransferred onto nitrocellulose membranes. Membranes were blocked in 5% milk in phosphatebuffered saline with 0.1% Tween 20 and incubated with primary antibody overnight at 4°C. The membranes were then treated with horseradish peroxidase-conjugated secondary antibody and developed using an ECL kit (Amersham Biosciences).
Transient Transfection and Luciferase Assay-For luciferase assays, 1.3 ϫ 10 5 LS-174T cells were cultured in a 12-well plate 24 h before transfection. The transfection was carried out using Lipofectamine (Invitrogen) in serum-free media containing 100 ng of reporter plasmid and 5 ng of Renilla luciferase reporter plasmid pRL-SV40 as an internal control following the manufacturer's protocol. This transfection mixture was added to cells and the plates were incubated at 37°C for 24 h. Prostaglandins and other reagents were added after 24 h and incubated for an additional 24 h. Firefly and Renilla luciferase activities were measured using a dual luciferase assay kit (Promega, Madison, WI) and a Luminometer. Firefly luciferase values were normalized to Renilla values.
Immunohistochemistry-Adult Apc Ϫ/ϩ mice were treated with PGE 2 or vehicle for 7 weeks, as previously described (43). Tissue sections from intestine were stained as follows. After CO 2 asphyxiation, the intestine was dissected, washed in phosphate-buffered saline, and fixed in 10% neutral buffered formalin. Paraffin sections (5 m) were dewaxed with xylene and rehydrated; the epitopes were revealed following treatment in a microwave oven. Once endogenous peroxidase activity was quenched, nonspecific immunoglobulins were blocked with normal goat serum (Vector Laboratories, Burlingame, CA) and samples were incubated overnight at 4°C with goat monoclonal antibody against NR4A2 at a dilution of 1:25 (R&D Systems, Minneapolis, MN). Negative controls received no primary antibody. The Vectastain ABC peroxidase system (Vector Laboratories, Burlingame, CA) was used for immunodection following the manufacturer's instructions, and immunolocalization was visualized with the peroxidase substrate 3,3-diaminobenzidine. Samples were counterstained with hematoxylin and mounted. All results were verified by a blinded second independent observer.
Apoptosis Assay-HCT-116 cells (2.5 ϫ 10 5 cells/well of 6-well plate) were transfected with empty vector or a dominant-negative NR4A2 construct. 24 h after transfection the cells were washed with phosphatebuffered saline and replaced with (10%) or without serum containing different concentrations of PGE 2 . FACS analysis was used to measure apoptotic cells using an Annexin V-FITC Apoptosis Detection Kit according to the manufacturer's instructions (R&D System). Briefly, the cells were harvested, washed, and then incubated with annexin V-FITC and propidium iodide followed by FACS analysis. This method allows the identification of different apoptotic cell populations: early apoptotic cells (annexin V-FITC positive), late apoptotic cells (annexin V-FITC and propidium iodide positive), necrotic cells (propidium iodide positive), and viable cells (both negative).
Human Colorectal Tissue Samples-Human colorectal tumor specimens were obtained from surgical resections with Vanderbilt Internal Review Board approval. For each tumor sample, matched adjacent normal mucosa was collected for comparison. All samples were snap frozen and stored in liquid nitrogen until use. Tissue RNA was prepared with TRI Reagent (Molecular Research Center, Cincinnati, OH) as described above.
Statistical Analysis-Each experiment was performed at least three times and data are expressed as the mean Ϯ S.E. Statistical significance was determined by paired Student's t test. p values Ͻ 0.05 were considered statistically significant.

PGE 2 Induces NR4A2 mRNA and Protein Expression in Colorectal
Carcinoma Cells-To identify potential target genes for PGE 2 in colorectal cancer, we performed microarray analysis on LS-174T cells treated with PGE 2 (1 M) at different time points. PGE 2 appears to regulate several transcription factors, including an immediate early gene, NR4A2. In confirmation of the microarray results, Northern blot analysis revealed that PGE 2 rapidly induces NR4A2 expression (Fig. 1A). Upon further investigation, PGE 2 induces NR4A2 in other colon carcinoma cell lines as well, with the greatest response observed in LS-174T and LoVo cells (Fig. 1B). Because NR4A2 induction is most dramatic in LS-174T cells, subsequent experiments focused on this model system. LS-174T cells were originally isolated from a mucinous adenocarcinoma derived from a well differentiated goblet cell lineage.
Quantitative real-time PCR assays and immunoblotting was utilized to determine the kinetics of NR4A2 induction. NR4A2 mRNA induction is strong but transient, with its levels peaking at 1 h (Fig. 1C). At the protein level, NR4A2 expression peaks at 2 h (Fig. 1D) as expected from the temporal pattern seen with changes in RNA expression. PGE 2 induction of NR4A2 is dose-dependent, and the effect is observed at very low levels of PGE 2 , indicating that a receptor mediated process is likely involved (Fig. 1, E and F). PGE 2 -induced NR4A2 Activates NBRE-Like other nuclear receptors, NR4A family members activate target genes through direct inter-action with specific promoter-derived cis-response elements. NR4A2 binds the consensus NBRE site (AAAGGTCA) as a monomer, homodimer, or heterodimer, which is found in the promoter region of genes that are regulated by this nuclear receptor. Thus, we determined whether PGE 2 -induced NR4A2 can bind and activate a 3xNBRE reporter gene construct using transient transfection assays. PGE 2

FIGURE 1. PGE 2 induces NR4A2 mRNA and protein expression in colorectal cancer cells. A, LS-174T cells were cultured in serum-free media for 48 h prior to PGE 2 (1 M) treatment.
Total RNA was isolated following harvest of the cells at the indicated time points. Equal amounts of RNA were loaded and the levels of NR4A2 mRNA were determined by Northern blot. B, cells were cultured in serum-free conditions prior to PGE 2 (1 M) treatment for 4 h. Total RNA was isolated and 5 g were reverse transcribed to cDNA. NR4A2 RNA levels were determined by quantitative real-time PCR. C, the Northern blot time course data were confirmed by real-time PCR as noted above. D, following the isolation of total cellular protein, equal amounts of protein were separated by SDS-PAGE and visualized with NR4A2 antibody. E, LS-174T cells were cultured in serum-free media for 48 h prior to PGE 2 (0.05-10 M) stimulation; cells were harvested after 4 h. The levels of NR4A2 mRNA were determined by real-time PCR analysis as noted above. F, following the isolation of total cellular protein, equal amounts of protein were separated by SDS-PAGE and visualized with NR4A2 antibody. induction of NR4A2 increased luciferase expression by 5-fold via binding the NBRE sites ( Fig. 2A). This experiment evaluated different concentrations of PGE 2 and found that activation of NBRE mirrors the increases we observe in NR4A2 mRNA and protein levels (Fig. 2B).

PGE 2 Regulates NR4A2 Expression and Activity in a cAMP/Protein Kinase A (PKA)-dependent
Manner-Four G protein-coupled receptors mediate PGE 2 action on target cells via distinct second messenger pathways (EP1-EP4). Upon ligand binding these receptors activate several downstream signaling cascades including the epidermal growth factor receptor, phosphatidylinositol 3Ј-kinase/Akt, Src, and PKA. To elucidate the specific mechanism by which PGE 2 induces NR4A2 expression, we employed several inhibitors to identify which signaling pathways mediate PGE 2 regulation of NR4A2. Induction of NR4A2 RNA and protein by PGE 2 is completely blocked by H-89, a selective inhibitor of the cAMP-dependent protein kinase A, PKA (Fig. 3, A and B). These results were further corroborated at a functional level: inhibition of NR4A2 expression by treatment with H-89 also blocks NBRE activation (Fig. 3C). These findings are consistent with the hypothesis that PGE 2 induces NR4A2 expression and activity in a cAMP/PKA-dependent manner.
PGE 2 Can Block Apoptosis Through NR4A2-Cancer cells must evolve to survive in a hostile environment. These hardy clones can modulate gene expression by blocking pathways that promote programmed cell death. We have shown previously that 24-h serum starvation (24 h) of LS-174T cells consistently induces apoptosis by 4 -5-fold (assessed by FACS analysis). Significantly, treatment with PGE 2 can rescue cells from undergoing programmed cell death (43). However, the molecular basis for this observation has not been well elucidated. Here we provide the first data that NR4A2 is an important mediator of this "anti-apoptotic" effect. We transfected cells with the dominant-negative NR4A2 construct and then added increasing amounts of PGE 2 following serum starvation. As shown in Fig. 4, A and B, cells transfected with dominant-negative NR4A2 were not protected from undergoing apoptosis following PGE 2 treatment. These experiments were also replicated with multiple stable clones expressing empty vector or dominant-negative NR4A2 (data not shown).
To further examine the molecular mechanism by which PGE 2 -induced NR4A2 protects cells from apoptosis, we analyzed whole cell lysates for caspase-3. Whereas PGE 2 blocks caspase-3 cleavage upon serum starvation, introduction of dominant-negative NR4A2 abrogates this effect (Fig. 4C). A recent report (44) also indicated that NR4A2 is involved in cell transformation and apoptosis in cervical cancer cells (HeLa), where the authors employed the use of NR4A2 small interfering RNA to inhibit its expression.
NR4A2 Expression Is Increased in Colorectal Cancer-Increased expression of COX-2 has been associated with poor prognosis in several different types of malignancies, including colorectal cancer (45)(46)(47). PGE 2 is the most abundant bioactive lipid in this setting (19). Because PGE 2 was found to induce NR4A2 expression in cultured cells, we sought to determine whether COX-2 expression and NR4A2 expression correlate in vivo. Analysis of 15 week-old Apc Ϫ/ϩ mouse adenomas revealed elevated levels of NR4A2 mRNA and protein, whereas intestinal mucosa with a microscopically normal appearance exhibited little expression (Fig. 6A). To our knowledge, this is the first demonstration that NR4A2 is expressed in intestinal epithelia. Consistent with a role in tumorigenesis, NR4A2 is strongly expressed in the proliferative crypt compartment, where expression is localized to the same region as that of Ki67 (Fig. 5).
To further analyze the role of PGE 2 -induced NR4A2 in vivo, mice were treated with PGE 2 of vehicle for 7 weeks and then intestinal tissues were evaluated for NR4A2 expression levels by immunohistochemistry and quantitative real-time PCR. Immunohistochemistry of tissue sections revealed increased NR4A2 expression in the intestine following PGE 2 treatment (Fig. 6D). Furthermore, quanti- tative PCR analysis indicates that PGE 2 treatment leads to increased NR4A2 transcript levels in colonic mucosa (Fig. 6C). These data support the hypothesis that NR4A2 is induced by PGE 2 in vivo.
Finally, to extend our in vivo studies and assess the clinical relevance of these observations, we examined NR4A2 levels in human colorectal carcinomas. Comparison of 16 paired human colon cancers and matched normal tissue demonstrates increased expression of NR4A2 in malignant tissue relative to normal colonic mucosa (Fig. 6B). These data support the hypothesis that NR4A2 is induced in colorectal cancer, a setting known to involve elevated levels of PGE 2 .

DISCUSSION
The inducible cyclooxygenase isoenzyme, COX-2, is overexpressed in many cancers, with concomitant overproduction of PGE 2 . Some cancers, such as ovarian cancer (48), seem to have increased levels of the COX-1 enzyme but the common denominator in these situations is an overall increase in PGE 2 production. A large body of evidence indicates a 40 -50% reduction in colorectal cancer in individuals taking non-steroidal antiinflammatory drugs regularly. These effects are due, at least in part, to inhibition of the cyclooxygenase enzymes and decreased production of PGE 2 .
Whereas it is known that elevated levels of PGE 2 contribute to carcinogenesis, the downstream effector genes in this pathway are not well characterized. We sought to increase our understanding of the role of PGE 2 in epithelial biology and carcinogenesis by identifying PGE 2 -regulated genes in colorectal cancer.
COX-2-derived PGE 2 regulates critical pro-survival and anti-apoptotic factors that allow for tumor growth. Our group originally identified Bcl-2 as a downstream effector of increased COX-2 expression in rat intestinal epithelial cells (49). In this article, we identify NR4A2, a nuclear hormone receptor belonging to the NGIF family, as a PGE 2regulated gene in colorectal cancer. Originally identified as a critical neurotrophic receptor in the mesencephalon, NR4A2 is implicated in Parkinson disease and other central nervous system pathologic processes. Because of the high expression levels in specific regions of the brain, the majority of NR4A studies examine its role in the central nervous system. Only recently have studies on the function of these receptors in other situations been undertaken. Data presented here provide the first evidence that PGE 2 can drive expression of this neurotrophic nuclear receptor in intestinal epithelial cells.
Using a variety of methods, in this study, we provide the first evidence that PGE 2 can induce NR4A2 in colorectal carcinoma cells via a cAMP/ PKA-dependent pathway. Moreover, evidence from functional studies indicates that NR4A2 is important for PGE 2 -mediated regulation of apoptosis by blocking cleavage of caspase-3. Intestinal epithelium from Apc Ϫ/ϩ mouse adenomas and sporadic colorectal carcinomas exhibit increased NR4A2 expression relative to matched normal mucosa. Finally, treatment of Apc Ϫ/ϩ mice with PGE 2 resulted in increased levels of NR4A2 mRNA and protein.
A wide range of growth factors and cytokines are known to induce NR4A2 gene expression. These include vascular endothelial growth factor, basic fibroblast growth factor, parathyroid hormone, corticotrophin releasing hormone, tumor necrosis factor-␣, and interleukin-1␤ (50 -53). PGE 2 has been shown to regulate NR4A2 in inflamed synovial tissue as well (50). The cAMP/PKA signaling pathway is known to regulate NR4A2 expression in osteoblasts (39,40). The functional role of NR42A receptors in colon cancer has not been examined previously. In this report we demonstrate that PGE 2 induces NR4A2 in colon cancer cells and plays a significant role in mediating the anti-apoptotic effects of PGE 2 .
Future studies will continue to yield greater insight into the effector genes that mediate tumorigenesis downstream of COX-2-derived PGE 2 . A better understanding of this process may reveal new strategies for the treatment and/or prevention of colorectal cancer. Evasion of apoptosis is a critical requirement for tumor progression. Building on previous reports from our group on the peroxisome proliferator-activated receptor family of nuclear receptors, these most recent data suggest a novel mechanism by which COX-2-derived PGE 2 protects carcinoma cells from apoptosis. FIGURE 6. NR4A2 expression is elevated in colorectal cancer. A, total RNA from polyps and normal adjacent tissue of six Apc Ϫ/ϩ mice was isolated and examined by real-time PCR analysis. ***, p Ͻ 0.001. B, total RNA was isolated from 16 human colorectal carcinomas and corresponding normal mucosa. NR4A2 expression was examined by realtime PCR. *, p ϭ 0.045. C, total RNA from polyps and normal adjacent tissue of Apc Ϫ/ϩ mice treated with vehicle or PGE 2 for 7 weeks was isolated and examined by real-time PCR analysis. D, examination of NR4A2 immunoreactivity in colonic tissues of Apc Ϫ/ϩ mice treated with vehicle or PGE 2 for 7 weeks. Tissue sections were stained with goat anti-NR4A2 primary antibody.