Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Proliferation and migration of VSMCs in arteries plays an important role in the pathophysiology of atherosclerosis, hypertension, and restenosis after angioplasty. A wide variety of growth factors, cytokines, and hormones have been found to activate these responses in blood vessels (Ross, 1999). A prominent growth factor for VSMCs is ANG II. These effects of ANG II are mediated via the ANG II type 1 (AT1) receptor signaling system (Kim and Iwao, 2000). Multiple signal transduction pathways, including activation of phospholipase C, phospholipase A₂ (PLA₂), tyrosine protein kinases/mitogen-activated protein (MAP) kinase, and PI 3-kinase/p70 S6 kinase signaling pathways are involved in mediating AT1 receptor stimulation of cell responses (Schieffer et al., 1996; Griendling et al., 1997; Kim and Iwao, 2000). In addition, activation of AT1 receptors has been found to increase prostanoid synthesis in smooth muscle cells, which contributes to some cellular responses mediated by ANG II (Shinoda et al., 1997; Dulin et al., 1998; Muthalif et al., 1998).

Recent progress indicates that atherosclerosis has an important inflammatory component; in addition, inflammatory responses may also be involved in other vascular disorders including restenosis after balloon dilatation. Prostaglandins (PGs) are important modulators of inflammatory responses. For example, a variety of hormones, cytokines, and growth factors enhance synthesis of PGs, which serve as autacoids, mediating a variety of responses in the cardiovascular system (Colina-Chourio et al., 2000). PG synthesis is a multistep process, controlled by several important rate-limiting enzymes; an important and extensively studied process involves activation of PLA₂, which catalyzes the release of arachidonic acid (AA) from phospholipids in membranes (Dennis, 1997 and Bingham and Austen, 1999). COX is another rate-limiting enzyme in PG synthesis, catalyzing the conversion of AA to PGG₂ and further to PGH₂. Two isoforms of the COX family have been identified, namely COX-1 (constitutive form) and COX-2, which is inducible by many growth factors and is found at sites of inflammation (Vane and Botting, 1998). ANG II activates PLA₂ in VSMCs (Rao et al., 1994). It has recently been reported that ANG II induces expression of COX-2 in VSMCs (Ohnaka et al., 2000; Young et al., 2000). On the other hand, ANG II appears to decrease expression of COX-2 in the macula densa in kidneys (Wolf et al., 1999).

There is evidence that a group of closely related nuclear receptors called peroxisome proliferator-activated receptors (PPARs), which function as ligand-activated transcription factors (for a review, see Kersten et al., 2000), may be involved in cellular inflammatory responses and in atherosclerosis. Three subtypes of PPAR, , , and  have been identified and cloned. Although PPAR is ubiquitously expressed, specific functions for this receptor are unclear. Physiological and pharmacological ligands of PPAR and PPAR have been recently identified. Selective pharmacological activators of PPAR include WY 14643 and various fibrins (Willson and Wahli, 1997). Selective PPAR activators include the endogenous PPAR receptor ligand 15d-PGJ₂ and synthetic antidiabetic drugs in the thiazolidinedione family. PPAR agonists inhibit inflammatory responses induced by cytokines and peptide growth factors in human aortic VSMCs (Staels et al., 1998). Thiazolidinediones also inhibit rat and human VSMC growth and migration via activation of PPAR receptors (Marx et al., 1998). Negative regulation of VSMC growth and inflammatory responses by PPARs may be related to their capacity to inhibit growth factor expression and blunting of growth-related signaling pathways such as MAP kinase pathways, or inhibition of inflammatory response-related gene expression (Delerive et al., 1999; Marx et al., 1998).

In the course of a series of experiments aimed at finding novel genes induced by ANG II, we have found evidence that COX-2 gene expression is increased by ANG II in VSMCs. The present studies were aimed at elucidating the mechanism by which COX-2 expression was increased and at determining the effects of this increased activity on smooth muscle cell proliferation and migration. We have found that ANG II stimulates transcription of the COX-2 gene and that COX-2 activity itself plays a key role in mediating ANG II-induced mitogenesis and migration of these cells. Furthermore, PPAR activators attenuated the capacity of ANG II to induce COX-2 expression as well as proliferation and migration of these cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. ANG II, competitive ANG II antagonist (A2275), ANG II receptor antagonist (ANG II antipeptide, A4184), AA, pertussis toxin, GF109203X, and NS398 were purchased from Sigma-Aldrich (St. Louis, MO); [-³²P]ATP, [³H]thymidine, and enhanced chemiluminescence Western blotting detection kits were from Amersham Biosciences, Inc. (Piscataway, NJ); and the ANG II AT1 receptor antagonist losartan and the AT2 receptor antagonist PD123319 were obtained from Sigma/RBI (Natick, MA). cPLA₂ inhibitor arachidonyl trifluoromethyl ketone (ATK) and inhibitors for protein kinase C, MAP kinase, and P38 kinase came from Calbiochem (San Diego, CA); cPLA₂ inhibitor methyl arachidonyl fluorophosphonate and nonselective PLA₂ inhibitor 7,7-dimethyl-5,8-eicosadienoic acid were purchased from Cayman Chemical (Ann Arbor, MI). Antibodies against COX-1 and COX-2, and protein A/G plus-agarose were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); COX-1 and COX-2 cDNA probes were purchased from the American Type Culture Collection (Manassas, VA); PPAR agonists were obtained from Calbiochem; cell culture medium and growth factors for human VSMCs were purchased from Clonetics Corp. (San Diego, CA); recombinant human PDGF-BB, cytokine IL-1, IL-6, and newborn bovine serum were obtained from Invitrogen (Carlsbad, CA). All other chemicals were reagent or molecular biology grade and were obtained from standard commercial sources.

Preparation of Cultured Human Aortic VSMCs. Human aortic VSMCs were purchased from Clonetics Corp. Cells were grown in smooth muscle growth medium-2 with 5% fetal bovine serum from Clonetics Corp. at 37°C in a humidified atmosphere of 5% CO₂-95% air as described previously (Hu et al., 1999). The cells were harvested for passaging at confluence with trypsin-EDTA and plated in 100-mm dishes at a density of about 5 × 10⁵, with 80 to 90% confluence being reached about 10 days later. The medium was replaced every 2 days. Cells were generally used for studies at 8 to 10 days after seeding. The cells were treated with agonists or vehicle solution (as control) starting from the longest time point, and the cells were harvested at the same time. Before treatment, the medium was replaced with 0.4% serum-containing medium for 24 h. Cells were stimulated for the indicated times with ANG II or other agonists.

Preliminary Subtractive Suppression Hybridization and Differential Screening to Identify Genes and Gene Clusters. To understand the general effects of ANG II on regulation of gene expression in VSMCs, the mRNA isolated from control or ANG II-treated VSMCs were compared using the techniques of subtractive suppression hybridization and differential screening to identify genes and gene clusters that were activated by ANG II. These experiments were conducted as described in the manufacturer's kits (BD Biosciences Clontech, Palo Alto, CA). A variety of candidate, differentially expressed genes were identified with this technique; one of these was the COX-2 gene. The current studies were designed to confirm this result and investigate the regulation of COX-2 expression in more detail.

Reverse transcriptase-PCR and Northern Blot Analysis. Total RNA and poly(A) (mRNA) were isolated from cells treated with or without ANG II or other agonists for the indicated times as described previously (Hu et al., 1996). The expression levels of COX-1 and COX-2 genes were determined by reverse transcriptase-PCR. Three micrograms of the total RNA were reverse-transcribed with Moloney murine leukemia virus reverse transcriptase (Invitrogen) using oligo(dT) primers. A portion of the cDNA (equivalent to 0.03 µg total RNA/PCR reaction) was used for PCR with Taq polymerase. The PCR reactions were carried out at 95°C initial denaturing temperature for 2 min, followed by 32 cycles of 94°C denaturing temperature for 30 s, 59°C annealing temperature for 30 s, and 72°C extending temperature for 30 s with a final extension at 72°C for 5 min. Primers for human COX-2 were directed at 551-571 and 1081-1060 of the cDNA (GenBank NM-000963), generating a 531-bp PCR product. Primers for COX-1 and -actin (Ambion, Austin, TX) produced 401-bp and 294-bp products, respectively. The PCR products were analyzed on 2% agarose gels and quantitated by using the Fluor-S Imager (Bio-Rad, Hercules, CA). Alternatively, 1 µg of poly(A) mRNA was used for standard Northern blotting. After electrophoresis, mRNA was transferred to a nylon membrane (ICN, Irvine, CA) in 20× SSC by using capillary blotting overnight. Blots were UV-cross-linked, prehybridized (50% formamide, 5× Denhardt's solution, 5× SSC, 0.5% SDS, and 20 mM salmon sperm DNA), and hybridized in the same buffer with a radiolabeled ([-³²P]dCTP) probe for human COX-2. The membranes were washed at 60°C in 1% SDS/2× SSC and autoradiographed with Kodak X-OMAT film at -70°C with an intensifying screen.

Nuclear Run-On Assays and mRNA Stability. To measure the transcription rate of ANG II-induced COX-2 gene expression, nuclear run-on assays were conducted as described previously (Hu and Hoffman, 2000). Active nuclei were isolated from VSMCs treated with or without ANG II or other agonists for the indicated times as described (Hu and Hoffman, 2000). A total of 100 µl of nuclear protein (1 × 10⁷ cells) was used for a transcription reaction using [-³²P]UTP to label nascent RNA transcripts. After hybridization, the filters were washed as described previously. The hybridized RNA was eluted by incubation of the filters with 200 ml of 0.3 M NaOH for 15 min at 65°C followed by the addition of 50 ml of glacial acetic acid and 4 ml of scintillation cocktail. Count ³²P radioactivity provided the relative rates of transcription of the COX gene. The stability of the COX-2 mRNA was measured as described previously (Hu et al., 1996). Briefly, VSMCs was treated with agonist (ANG II) to increase the levels of mRNA. After washing the cells, the transcriptional inhibitor actinomycin D with either ANG II or vehicle was added. Cells were subsequently harvested at the indicated time points and RNA was prepared. Northern blotting was conducted for COX-2 and -actin mRNAs. The blots were scanned by densitometry; the results are expressed as COX-2/-actin ratios.

Electrophoretic Mobility Shift Assays. These assays were done using a double-stranded synthetic consensus CREB (5'-AGAGATTGCCTGACGTCAGAGAGCTAG-3'), SP1 (5'-ATTCGATCGGGGCGGGGGCGAGC-3'), NF-B (5'-AGTGAGGGGACTTTCCCAGGC-3'), and AP2 (5'-GATCGAACTGACCGCCCGCGGCCCGT-3'). DNA was labeled by T4 kinase-catalyzed 5' end-labeling. A typical protein-DNA binding assay was conducted by adding gel shift assay mixture (from Promega's Gel Shift Assay System) with 10 µg of nuclear extract plus [³²P]DNA (10,000-30,000 counts/min corresponding to 0.25-0.5 ng of DNA). The mixture was incubated for 30 min at room temperature. For competition assays or specificity assays, a 100 to 200 times molar excess of nonradioactive DNA or mutant DNA was added to the reaction mixture and incubated at room temperature for 10 min before addition of [³²P]DNA. The reaction was stopped by addition of 2.5 µl of loading buffer. Samples were immediately resolved by using a 4% polyacrylamide gel. The gel was electrophoresed at 140 V for 2.5 h. The specificity of ANG II-induced NF-B binding was determined with a NUSHIFT kit from Geneka Biotechnology Inc. (Montreal, QC, Canada).

Immunoprecipitation and Immunoblot Analysis. Cell crude pellets were prepared as previously described (Hu et al., 1996) and were incubated with lysis buffer (1% Nonidet P-40, 25 mM Hepes (pH 7.5), 50 mM NaCl, 50 mM NaF, 5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 10 µg/ml antipain, aprotinin, and leupeptin) for 30 min on ice. Insoluble material was removed by centrifugation at 12,100g for 20 min. Equal amounts of protein (1 to 2 mg as indicated for individual experiment) were incubated with an appropriate antibody overnight and then further incubated with 25 µl of protein A/G plus-agarose for 2 h at 4°C with constant rotation. For immunodetection, immunoprecipitates were washed four times with lysis buffer and twice with distilled water, and resolved by SDS-PAGE. Resolved proteins were transferred to membrane and detected by an enhanced chemiluminescence system (Amersham Biosciences) with the indicated antibodies and a horseradish peroxidase-conjugated secondary antibody. For immunoblot analysis, 100 µg of protein was generally subjected to SDS-PAGE and transferred to a membrane. Immunodetection was similar to that described above.

ELISA Measurement of PGE₂. VSMCs were incubated in 12-well plates and stimulated for the indicated time. The culture medium was collected and centrifuged at 12,000 rpm for 10 min. ANG II-stimulated production of PGE₂ was measured with an ELISA kit (Cayman Chemical) following the manufacturer's instructions. ANG II-stimulated PGE₂ was normalized by protein concentration (picograms per milligram of protein).

Agonist-Stimulated DNA Synthesis. Near-confluent VSMCs in 24-well plates were made quiescent through incubation of cells in DMEM with 0.1% serum for 48 h. ANG II or other agonists were added to the medium for 20 h. [³H]Thymidine (0.1µCi/well) was added for an additional 4 h, and the incorporation of [³H]thymidine was then determined as described previously (Hu et al., 1998). In cases using inhibitors, these inhibitors were generally added 1 h before addition of agonists.

Agonist-Stimulated Cell Migration. Migration of VSMCs was investigated through in vitro Transwell cell culture chambers using a gelatin-treated polycarbonate membrane with 8-µm pores in 24-well plates (Costar Corp., Cambridge, MA) as described previously (Hu et al., 1998). Preconfluent VSMCs were suspended in DMEM/0.4% FBS to a concentration of 3.0 × 10⁵ cells/ml. For investigating effects of various inhibitors on cell migration, cells were pretreated with inhibitors or vehicle for 30 min at 37°C. DMEM/0.4% FBS (0.6 ml) with or without ANG II (100 nM), IL-1 (100 ng/ml), or PDGF-BB (20 ng/ml) was added to the lower compartment. A 0.1-ml cell suspension (final: 10,000 cells/well, diameter 6.5 mm) was added to the upper compartment with or without inhibitors and cells were then incubated at 37°C (95% air/5% CO₂). After 24 h, the filters were fixed with methanol (10 min at 4°C), followed by counterstaining with hematoxylin. The number of VSMCs per 320× HPF that migrated to the lower surface of the filters was determined microscopically. Four randomly chosen HPFs were counted per filter. Experiments were performed in triplicate.

Data Analysis. Data are presented as mean ± S.E.M., and treatment effects were compared by one-way analysis of variance or Student's paired t test (two-tailed). p < 0.05 was taken as the level of significance.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Angiotensin II Enhancement of COX-2 Expression in VSMCs. Subtractive suppression hybridization and differential screening identified the COX-2 gene as potentially induced by stimulation of VSMCs by ANG II. The first experiments were designed to confirm or refute this observation. Figure 1 illustrates that ANG II caused a time- and concentration-dependent increase in the expression of COX-2 mRNA in cultured human VSMCs. Increased expression of COX-2 mRNA was detected as early as 30 min. In time course experiments this increase at 30 min was significant (2.5 ± 0.4-fold of control, p < 0.01); the maximum change in expression occurred by 3 h (6.5 ± 1.1-fold of control, p < 0.001), increased mRNA values persisted for longer than 24 h in the continued presence of ANG II, and increased expression of COX-2 mRNA was detected at ANG II concentrations as low as 1 nM (Fig. 1A). Similar results were detected by Northern blotting (Fig. 1B). However, ANG II had no significant effect on expression of COX-1 mRNA in these cells. Additionally, ANG II produced similar effects on expression of COX-1 and COX-2 mRNAs in cultured rat VSMCs (data not shown).

View larger version (65K): [in this window] [in a new window]   Fig. 1.   Angiotensin II increases COX-2 mRNA accumulation in VSMCs. Nearly confluent primary cultures of VSMCs were incubated in DMEM containing 0.4% serum for 24 h. Cells were then treated with ANG II (100 nM) for various times or with the indicated concentrations of ANG II for 2 h. PCR for COX-1, COX-2, and -actin genes were done as described under Materials and Methods. The results demonstrate that ANG II increases abundance of the COX-2 mRNA without changing expression of the COX-1 or -actin genes. Data are a representative of four experiments from PCR analysis (panel A). Similar results were obtained using isolation poly(A) (mRNA), and panel B illustrates a representative of three additional Northern blotting experiments.

Angiotensin II-Induced Increase in Expression of COX-2 Involves Transcriptional Activation of the COX-2 Gene. To determine the mechanism(s) for increased accumulation of COX-2 mRNA in these cells, the transcription rate of the COX-2 gene in control and ANG II-treated VSMCs was measured using nuclear run-on assays. ANG II (100 nM) stimulated a time-dependent and marked increase in the transcription rate of the COX-2 gene between 30 min and 1 h of incubation (Fig. 2A). However, the transcription rate of COX-2 gene from VSMCs treated with ANG II from 3 to 24 h was not significantly changed compared with that of control VSMCs. To determine the potential effects of ANG II on the rate of decay of COX-2 mRNA, cells were incubated in the presence and absence of ANG II for 1.5 h; then, the stability of COX-2 mRNA was measured in the presence or absence of ANG II and the transcription inhibitor actinomycin D. ANG II did not significantly change the degradation rate of the COX-2 mRNA; the half-life of decay of COX-2 mRNA was 3.3 ± 0.5 h in ANG II-treated cells versus 2.7 ± 0.6 h in control cells (p > 0.05) (Fig. 2B). Together, these results suggest that ANG II increases the accumulation of COX-2 mRNA by transcriptional activation of the COX-2 gene in human VSMCs. Similar results were obtained in rat VSMCs (data not shown).

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Angiotensin II-induced COX-2 expression via transcriptional activation of the COX-2 gene in VSMCs. Panel A, to measure the transcription rate of the COX-2 gene, quiescent VSMCs were stimulated with ANG II (100 nM) for the indicated times; then, cell nuclear extracts were prepared and nuclear run-on assays were conducted as described under Materials and Methods. Data are mean ± S.E.M of four experiments. Panel B, effect of ANG II on COX-2 mRNA half-life. VSMCs were treated with ANG II (100 nM) for 1.5 h. Actinomycin D (10 mg/ml) with ANG II (100 nM) () or vehicle () was added (t = 0), and cells were harvested and RNA was prepared at the indicated time points. Northern blotting was conducted for COX-2 and -actin mRNAs. The blots were scanned with densitometry; the result was expressed as COX-2/-actin ratios. Data are mean ± S.E.M of three experiments. Panel C, ANG II increases NF-B binding. VSMCs were treated with 100 nM ANG II for the indicated times (left side of panel C). ANG II induced a time-dependent activation of NF-B binding in VSMCs. To determine the specificity of ANG II-induced NF-B binding, VSMCs were treated with 100 nM ANG II for 1 h (right side of panel C). Nuclear extracts were prepared and an electrophoretic mobility shift assay was conducted as described under Materials and Methods. Preincubation of the nuclear extracts with 100-fold wild-type probe completely blocked ANG II-induced increase in NF-B binding. The point mutation of wild-type probe (mutant) resulted in a loss of capability of NF-B probe. Anti-p50 but not anti-p65 NF-B transcription factor antibody induced a super-shift band as indicated; synthetic p50 peptide (1 ng) attenuated the inhibitory effect of anti-p50 antibody. These results indicate that ANG II specifically activates the p50 type of NF-B transcription factor in these cells. Panel D, ANG II does not significantly change binding of transcription factors CREB and SP1 in VSMCs. Data are a representative of four experiments with similar results.

Angiotensin II Increases Expression of COX-2 Protein in VSMCs. The expression of the protein products of the COX-1/2 gene in VSMCs was determined by using specific anti-COX-1 and anti-COX-2 antibodies. ANG II stimulated a significant time- and concentration-dependent increase in expression of COX-2 protein (Fig. 3, A and B). The increase was detectable as early as 30 min, reaching a peak at 3 h after incubation of the cells with ANG II (Fig. 3). The expressed levels of the COX-2 protein remained greater than control values for 24 h (Fig. 3). At concentrations as low as 1 nM, ANG II significantly increased expression of COX-2 protein. The AT1 receptor antagonist losartan and the AT2 receptor antagonist PD123319 were used to determine the role of ANG II receptor subtypes in ANG II-stimulated expression of the COX-2 gene in VSMCs. Pretreatment of VSMCs for 30 min with losartan (1 µM) attenuated the ANG II-mediated increase in expression of COX-2 without affecting the levels of COX-1 (Fig. 4). In contrast, PD123319 (1 µM) had no effect on the ANG II-mediated increases in COX-2 protein accumulation. These data suggest that ANG II increases in COX-2 protein accumulation in VSMCs are mediated by AT1 receptors.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Angiotensin II stimulates time- and concentration-dependent increases in the expression of COX-2 protein in VSMCs. Quiescent VSMCs were treated with ANG II (100 nM) for the indicated times, or with the indicated concentrations of ANG II for 2 h. Cell lysates were then prepared; 100 µg of protein were resolved by 8.5% SDS-polyacrylamide gels. After Western blotting, the membranes were probed using anti-COX-2 or anti-COX-1 antibodies as described under Materials and Methods. Panel A, data are representative of four experiments with similar results. Panel B, data are a summary of four experiments (average ± S.E.M.).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   AT1 receptor and MAP kinase mediate angiotensin II-induced expression of COX-2 in VSMCs. Quiescent VSMCs were pretreated with various inhibitors or antagonists including the ANG II-specific AT1 antagonist losartan (1 µM), the ANG II AT2 receptor antagonist PD1339 (1 µM), pertussis toxin (100 ng/ml), the calcium chelator BAPTA (1 µM), the MAP kinase inhibitor PD98059 (10 µM), the p38 MAP kinase inhibitor SB82302 (10 µM), or the inhibitor of PI 3-kinase LY294,002 (1 µM); the EGF receptor inhibitor AG490 (1 µM) and the protein kinase C inhibitor GF109203X (1 µM) were added 1 h before addition of ANG II (100 nM). Pertussis toxin was added 12 h before treatment. Cell lysates (100 µg of protein) were resolved by 8.5% SDS-PAGE and subjected to Western blotting. The blots were detected by anti-COX-1 or -COX-2 antibodies as described under Materials and Methods. Data presented in panel A are a representative of four experiments. Panel B is a summary of all four experiments (average ± S.E.M.). Comparison to ANG II treatment, , p < 0.05; , p < 0.01.

Angiotensin II Enhances PG Release from VSMCs. Since ANG II enhanced COX-2 expression in VSMCs, we wondered whether the release of PGE₂, a major product of COX-2 in the conversion of AA to PGs, would be increased. Figure 5 illustrates that pretreatment of human VSMCs with ANG II for 6 h led to increases in the release of PGE₂ compared with control cells. ANG II-mediated increased PGE₂ release was attenuated by either the AT1 receptor antagonist losartan or the MAP kinase inhibitor PD98059 but not by the AT2 receptor antagonist PD123319 (Fig. 5), suggesting that ANG II utilizes COX-2 to regulate production of PGs in VSMCs via activation of AT1 receptor pathway.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Angiotensin II stimulates PGE2 production via activation of the COX-2 gene in VSMCs. Quiescent VSMCs were incubated with the indicated concentrations of ANG II or vehicle for 6 h. Various inhibitors including the ANG II AT1 antagonist losartan (1 µM), the AT2 inhibitor PD123319 (1 µM), and the MAP kinase inhibitor PD98059 (10 µM) were added alone or 1 h before the addition of ANG II. The culture medium was then used for ELISA measurement of PGE2 as described under Materials and Methods. The data are mean ± S.E.M of four experiments performed in triplicate. Compared with ANG II treatment, , p < 0.05; , p < 0.01.

Angiotensin II-Induced Increased Expression of COX-2 Does Not Require Activation of PLA₂ in VSMCs. Previous work has suggested that there is a functional relationship between activation of PLA₂ and expression of the COX-2 gene (Tada et al., 1998; Murakami et al., 1999). Two isoforms of PLA₂, namely cytosolic PLA₂ and secretory PLA₂, are expressed in human cells. Several inhibitors of cPLA₂ or nonselective inhibitor of PLA₂ were utilized to determine whether mechanisms downstream of AA release participated in the increased COX-2 expression induced by ANG II. Figure 6 illustrates that the cPLA₂ inhibitor ATK (100 nM to 100 µM) had no effect on ANG II-stimulated expression of the COX-2 protein. Another cPLA₂ inhibitor, methyl arachidonyl fluorophosphonate, and the nonspecific PLA₂ inhibitor 7,7-dimethyl-5,8-eicosadienoic acid also had no effect on ANG II-stimulated COX-2 expression (data not shown). Additionally, AA itself, a major product of PLA₂ activation by ANG II, did not significantly 0increase COX-2 protein expression; furthermore, supplemental AA plus ANG II did not increase COX-2 expression more greatly than ANG II alone (Fig. 6). These results suggest that PLA₂ activity does not contribute to the ANG II-enhanced COX-2 expression.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Angiotensin II-induced increase in expression of COX-2 occurs in the presence of PLA2 inhibition in VSMCs. Quiescent VSMCs were pretreated with the PLA2 inhibitor ATK (100 nM to 100 µM) for 1 h (left side of figure); ANG II stimulation for 2 h increased COX-2 expression in these cells (determined by Western blotting). Arachidonic acid (AA) (10 µM), alone or in combination with ANG II, did not augment induction of COX-2 expression compared with ANG II alone (right side of figure) Data are representative of four experiments.

Inhibition of COX-2 Activity Attenuates Angiotensin II-Induced DNA Synthesis and Migration of VSMCs. ANG II is well known for its capacity to induce VSMC proliferation and migration. We wondered whether these responses to ANG II were dependent on COX-2 activity. In other words, if ANG II stimulation of VSMC proliferation and migration were dependent on COX-2 activity, then the ANG II-mediated increase in COX-2 expression could lead to magnified responses to ANG II over time. The contribution of COX-2-derived prostanoids to VSMC proliferation was evaluated by measuring the effect of the COX-2 inhibitors on ANG II-induced DNA synthesis. An increase in DNA synthesis was observed after exposure of the cells to either ANG II, cytokine IL-1, or PDGF for 24 h (Fig. 7). Preincubation of cells with the COX-2 inhibitor NS-398 (0.1 µM) markedly attenuated the ANG II-, IL-1-, and PDGF-mediated increases in DNA synthesis. This inhibitor did not modify DNA synthesis in the absence of agonist ligands (Fig. 7A).

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Inhibition of COX-2 activity attenuates angiotensin II-induced DNA synthesis and cell migration in VSMCs. Panel A, the quiescent cells were incubated with ANG II (100 nM), IL-1 (100 ng/ml), or PDGF-BB (20 ng/ml) for 20 h. Cells were then incubated with [3H]thymidine (0.1 µCi/dish) for another 4 h. Various inhibitors or antagonists, including the AT1 antagonist losartan (1 µM), the COX-2 inhibitor SNC-389 (1 µM), and the COX-1/2 inhibitor indomethacin (10 µM), were added 1 h before the addition of each agonist. DNA synthesis was measured as [3H]thymidine incorporation into cells as described under Materials and Methods. The data are mean ± S.E.M of three experiments performed in triplicate. Compared with ANG II treatment, , p < 0.05; , p < 0.01. Panel B, 24-h agonist-directed cell migration was performed as described under Materials and Methods. Data are mean ± S.E.M of four experiments performed in triplicate. , p < 0.05 versus ANG II or IL-1 treatment; , p < 0.01 versus ANG II treatment.

PPAR and PPAR Activators Attenuate Angiotensin II-Induced Increase in DNA Synthesis, Cell Migration, and Expression of COX-2. Effects of the PPAR activator WY 14643 and the PPAR activator 15d-PGJ2 on ANG II-, IL-1-, and PDGF-BB-stimulated DNA synthesis and cell migration were determined. Both WY 14643 and 15d-PGJ2 significantly attenuated ANG II-, IL-1-, and PDGF-BB-stimulated increases in DNA synthesis (Fig. 8A). As above, ANG II enhanced cell migration in these experiments to 46.1 ± 3.3 cells/4 HPF in comparison to controls, 19.2 ± 2.4 cells/4 HPF (p < 0.01, n = 6). Similarly, IL-1 and PDGF-BB increased cell migration to 43.8 ± 3.5 cells/4 HPF (p < 0.01, n = 6) and 75.2 ± 8.6 cells/4 HPF (p < 0.01, n = 4), respectively. Neither WY 14643 nor 15d-PGJ2 inhibited the basal migration of VSMCs. However, both these PPAR activators significantly attenuated cell migration that had been stimulated by ANG II (100 nM), IL-1 (100 ng/ml), or PDGF-BB (25 ng/ml) (Fig. 8B).

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Agonist activation DNA synthesis and cell migration are attenuated by PPAR and PPAR activators. Panel A, quiescent cells were incubated with ANG II (100 nM), IL-1 (100 ng/ml), or PDGF-BB (20 ng/ml) for 20 h. Cells were then incubated with [3H]thymidine (0.1 µCi/dish) for another 4 h. Various inhibitors or antagonists including the AT1 antagonist losartan (1 µM), the PPAR activator WY 14354 (1 µM), or the PPAR activator 15d-PGF2 (1 µM) were added 1 h before the addition of ANG II, IL-1, or PDGF-BB. DNA synthesis was measured as [3H]thymidine incorporation into cells. The data are mean ± S.E.M of three experiments performed in triplicate. Compared with ANG II treatment, , p < 0.05; , p < 0.01. Panel B, 24-h agonist-directed cell migration was performed as described under Materials and Methods. For either 1 µM WY 14643 or 1 µM 15d-PGJ2, VSMC migration was inhibited as follows: WY 14643 plus ANG II, 31.4 ± 3.5 cells/4 HPF; 15d-PGJ2 plus ANG II, 23.0 ± 2.4 cells/4 HPF, p < 0.01 versus ANG II alone (43.8 ± 3.5 cells/4 HPF, n = 6). Also, WY 14643 and 15d-PGJ2 inhibited IL-1- and PDGF-BB-induced cell migration 34 ± 3.4 cells/4 HPF and 27.0 ± 3.0 cells/4 HPF for IL-1, and 43.3 ± 7.0 cells/4 HPF and 39.1 ± 5.0 cells/HPF for PDGF-BB, p < 0.05, respectively (panel B). Data are mean ± S.E.M of four experiments performed in triplicate. , p < 0.05 versus ANG II, IL-1, or PDGF-BB treatment; , p < 0.01.

View larger version (52K): [in this window] [in a new window]   Fig. 9.   PPAR and PPAR activators attenuated angiotensin II-induced increase in COX-2 expression. Quiescent VSMCs were pretreated with PPAR activators for 1 h. Cells were then treated with ANG II (100 nM) or vehicle for 2 h in the presence or absence of PPAR activators as indicated (WY 14643, 0.1 µM to 100 µM; 15d-PGJ2, 0.01 µM to 1 µM, respectively). In panels A and B, 100 µg of protein were resolved by 8.5% SDS-PAGE and subjected to Western blotting. The blots were detected by anti-COX-1 or -COX-2 antibody as described under Materials and Methods. Data are a representative of three experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of the current studies demonstrate that ANG II increased expression of the COX-2 gene via AT1 receptors, both at the level of mRNA and protein abundance. ANG II increased transcription of the COX-2 gene which was associated with binding of the transcription factor NF-B to regulatory sequences in the COX-2 gene. Activation of MAP kinase plays a critical role in the induction of COX-2 expression. ANG II enhances DNA synthesis and cell migration in human VSMCs; these responses were blocked by inhibition of COX-2 activity. This suggests that these effects of ANG II on VSMCs may be magnified by increased expression of COX-2 over time. We also found that ANG II-mediated increases in expression of the COX-2 gene, as well as stimulation of DNA synthesis and induction of cell migration, were all attenuated by ligands that activated PPAR or PPAR.

There has been considerable interest over many years in discerning the role of ANG II in vascular biology, not only on account of its capacity to increase blood pressure directly by stimulating smooth muscle contraction, but also as a trophic factor stimulating vascular and cardiac cell growth. Increasing evidence indicates that ANG II-mediated activation of AT1 receptors plays an important role in triggering multiple biological responses in VSMCs. Several signal transduction pathways, including activation of phospholipase C, tyrosine protein kinases, MAP kinases, and PI 3-kinase signaling pathways are involved in mediating AT1 receptor stimulation of cell proliferation, cell migration, and other responses (Kim and Iwao, 2000). In the present study, we found that ANG II activates AT1 rather than AT2 receptors in inducing COX-2 expression. We used a variety of signal transduction inhibitors to explore potential signaling pathways utilized by AT1 receptors in inducing COX-2 expression in VSMCs. We found that ANG II, activating a pertussis toxin-insensitive G protein, induced this response by a pathway critically dependent on MAP kinase activity. On the other hand, we found no evidence that p38 kinase, the EGF receptor, PI 3-kinase, or increases in intracellular calcium concentrations played a role in mediating this response. The results suggest that protein kinase C may play a relatively minor role in inducing this response.

We have found that ANG II increases binding of NF-B to a DNA consensus sequence in gel retardation assays. NF-B has been shown to play a role in regulation of transcription of the COX-2 gene; for example, using decoy oligonucleotides, it was found that oxidative induced damage of cardiac myocytes critically involves NF-B in the increased transcription of the COX-2 gene (Adderley and Fitzgerald, 1999). Also, exposure of vascular smooth muscle cells to polycyclic aromatic hydrocarbons found in tobacco smoke increases NF-B binding to DNA; in addition, stimulation of COX-2 promoter activity by these agents was blocked by a construct containing a mutagenized NF-B site (Yan et al., 2000). In addition, induction of COX-2 expression in endothelial cells by hypoxia involves NF-B. The possible connection between ANG II activation of MAP kinase activity and increased NF-B binding to DNA in human VSMCs requires further exploration.

ANG II increases prostaglandin release from many cell types including VSMCs (Rao et al., 1994), at least in part because ANG II stimulates PLA₂ activity. The potential pathophysiological importance of the observation that ANG II also induces expression of the COX-2 gene became clearer when we found that specific inhibitors of COX-2 enzymatic activity markedly attenuated the capacity of ANG II to stimulate VSMC growth and migration. This suggests that COX-2 and its products, such as thromboxane A₂, play an important role in the ANG II-mediated mitogenic and inflammatory responses in VSMCs. Since ANG II has been found to directly promote vascular remodeling (Kim and Iwao, 2000), the results suggest that specific COX-2 inhibitors might supplement or enhance therapeutic approaches with angiotensin-converting inhibitors or ANG II receptor antagonists in efforts to attenuate these vascular changes in diseases such as hypertension. On the other hand, there is concern that effects of COX-2 inhibitors on nitric oxide synthase in endothelial cells may increase cardiovascular risk.

During the course of our studies, it was reported that ANG II increases COX-2 protein and mRNA expression (Ohnaka et al., 2000). Our results indicate that ANG II increases the abundance of the COX-2 gene by increasing its rate of transcription. Ohnaka et al. (2000) found in their experiments that the increased mRNA abundance was solely due to a decreased rate of mRNA degradation. It is not clear how to account for these differences, although our experiments were largely conducted in human VSMCs, whereas those of Ohnaka et al. (2000) used rat cells. Another possibly significant difference is that we used nuclear run-on assays to measure transcription of the COX-2, whereas Ohnaka and colleagues used upstream regions of the rat COX-2 promoter subcloned into a pGL3 basic luciferase plasmid that was then transfected into smooth muscle cells.

Considerable recent interest has focused on the regulation of the COX-2 gene as it plays a central and very broad role in inflammatory responses in a host of pathophysiological processes. Induction of COX-2 mRNA abundance has been found to be due to increased transcription, increased mRNA stability, or both in a variety of cells stimulated by a number of different inducers (Huang et al., 2000; Subbaramaiah et al., 2000). Several studies have found that MAP extracellular signal-regulated kinase 1/2 or p38 pathways are involved in increasing accumulation of COX-2 (Sheng et al., 1998; Ohnaka et al., 2000; Reddy et al., 2000). We have found in human vascular smooth muscle cells that p38 does not play a role in ANG II activation of COX-2 expression. Consequently, it is difficult to generalize about mechanisms that lead to increased COX-2 protein abundance in different cell types.

PLA₂ is a family of phospholipid-hydrolyzing enzymes that are involved in diverse pathological processes, including inflammatory responses. A functional association of PLA₂ activation and expression of the COX-2 gene has been reported recently (Murakami et al., 1999). Two isoforms of PLA₂, namely cytosolic PLA₂ and secretory PLA₂, are expressed in human cells. Tada et al. (1998) found that activation of secretory PLA₂ up-regulates COX-2 expression in rat serosal mast cells. We have found that none of several inhibitors of PLA₂, nor AA, modified ANG II-stimulated COX-2 expression in human VSMCs. Consequently, it is difficult to generalize about the potential capacity of AA to regulate COX-2 expression.

A major result of our current studies is the finding that activation of PPAR and PPAR inhibited ANG II-stimulated COX-2 expression. The regulation of COX-2 expression by PPARs is complex and appears to be cell-specific. PPARs have been found to increase rather than decrease COX-2 expression in corneal, and in mammary and colonic epithelial cells (Bonazzi et al., 2000). Meade et al. (1999) found that there was a PPAR consensus sequence at about -3900 in the COX-2 promoter. Lefebvre et al. (1998) reported that activation of PPAR did not induce COX-2 in similar colonic cells. PPAR activators also induce COX-2 expression in liver in vivo (Leung and Glauert, 1998). On the other hand, Staels et al. (1998) found that PPAR ligands transcriptionally inhibit interleukin-1-induced expression of COX-2 as a result of repression of NF-B signaling.

PPAR is present in endothelial and smooth muscle cells, monocytes, and monocyte-derived macrophages (Fruchart et al., 1999; Goetze et al., 1999; Law et al., 2000). As a result of a negative transcriptional regulation of the NF-B and AP-1 signaling pathways, PPAR inhibits inducible nitric oxide synthase in macrophages and prevents the IL-1-induced expression of IL-6 and COX-2, as well as thrombin-induced endothelin-1 expression in these cells (Delerive et al., 1999). Interestingly, activation of PPAR in VSMCs has been found to inhibit IL-1-induced NF-B activation, leading to a blunting of increased expression of inducible nitric-oxide synthase in these cells. Goetze et al. (1999) found that inhibition of MAP kinase and activation of PPAR both inhibited VSMC migration; they concluded that PPAR ligands act downstream of the cytoplasmic activation of MAP kinase and appear to exert their effects in the nucleus. Our results provide a potential explanation for these findings. Namely, inhibition of the induction of COX-2 expression occurs after either inhibition of MAP kinase or activation of PPARs; in addition, the finding that induction of COX-2 is required for ANG II-stimulated VSMC migration provides a potential link between these two observations.

Beyond the well known role of ANG II in raising blood pressure in patients with hypertension, its importance in the progression of cardiovascular disease is being increasingly recognized. For example, angiotensin-converting enzyme inhibitors, drugs that attenuate formation of ANG II, have been found to have favorable effects on outcome of patients with coronary artery disease possibly independent of their effects on blood pressure (Kim and Iwao, 2000). The mechanism(s) for these favorable effects are unclear; our results suggest that the induction of COX-2 by ANG II, coupled with the importance of this enzyme in proliferation and migration of VSMCs, suggests that interrupting this pathway could be important in attenuating the progression of atherosclerosis or vascular remodeling. These results, and those showing that PPAR agonists also attenuate potentially deleterious effects of ANG II, warrant further study to evaluate the extent to which stimulation of these pathways by ANG II may be involved in the progression of vascular disease.