Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Microtubule-interfering Agents Stimulate the Transcription of Cyclooxygenase-2

Microtubule-interfering Agents Stimulate the Transcription of Cyclooxygenase-2 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 275, No. 20, Issue of May 19, pp. 14838 –14845, 2000 © 2000 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Microtubule-interfering Agents Stimulate the Transcription of Cyclooxygenase-2 EVIDENCE FOR INVOLVEMENT OF ERK1/2 AND p38 MITOGEN-ACTIVATED PROTEIN KINASE PATHWAYS* Received for publication, January 18, 2000, and in revised form, February 16, 2000 Kotha Subbaramaiah‡§, Janice C. Hart‡, Larry Norton , and Andrew J. Dannenberg‡ From the ‡Department of Medicine, New York Presbyterian Hospital-Cornell and Strang Cancer Prevention Center and Breast Cancer Medicine Service, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 possibly cancer (10 –12). For example, the expression of COX-2 We investigated whether microtubule-interfering agents (MIAs: taxol, colchicine, nocodazole, vinblastine, is increased in inflamed tissues such as rheumatoid synovium vincristine, 17-b-estradiol, 2-methoxyestradiol) altered (13), and selective COX-2 inhibitors are useful for the treat- cyclooxygenase-2 (COX-2) expression in human mam- ment of arthritis (11). COX-2 is also overexpressed in trans- mary epithelial cells. MIAs enhanced prostaglandin E 2 formed cells (8, 14, 15) and in tumors (16 –19). Mice engineered synthesis and increased levels of COX-2 protein and to be COX-2-deficient are protected against developing intesti- mRNA. Nuclear run-off assays revealed increased rates nal tumors (20) and skin papillomas (21). Additionally, selec- of COX-2 transcription after treatment with MIAs. Cal- tive COX-2 inhibitors prevent the formation of intestinal (20, phostin C, an inhibitor of protein kinase C, blocked the 22) and skin tumors (23) in experimental animals and suppress induction of COX-2 by MIAs. The stimulation of COX-2 the growth of transplantable tumors (24, 25). With this in promoter activity by MIAs was inhibited by overex- mind, it is reasonable to postulate that compounds that induce pressing dominant negative forms of Rho and Raf-1. COX-2 could predispose to cancer or inflammation. MIAs stimulated ERK, JNK, and p38 mitogen-activated Microtubule-interfering agents (MIAs) are widely used for protein kinases (MAPK); pharmacological inhibitors of the treatment of cancer. The anti-cancer properties of MIAs MAPK kinase and p38 MAPK blocked the induction of have been attributed in part to interference with microtubule COX-2 by MIAs. Overexpressing dominant negative forms of ERK1 or p38 MAPK inhibited MIA-mediated assembly, impairment of mitosis, and changes in cytoskeleton activation of the COX-2 promoter. MIAs stimulated the (26). There is growing evidence, however, that MIAs have mul- binding of the activator protein-1 transcription factor tiple cellular effects. For example, MIAs stimulate mitogen- complex to the cyclic AMP response element in the activated protein kinases (MAPKs) and gene expression (27– COX-2 promoter. A dominant negative form of c-Jun 29). Taxol, a novel anti-cancer drug and MIA, up-regulates inhibited the activation of the COX-2 promoter by MIAs. COX-2 and synthesis of PGE (30, 31). Colchicine and vinblas- Additionally, cytochalasin D, an agent that inhibits ac- tine, two other MIAs, stimulate PGE production (32, 33). tin polymerization, stimulated COX-2 transcription by These findings raise the possibility that COX-2 is a down- the same signaling pathway as MIAs. Thus, microtubule- stream target of MIAs and possibly other compounds that or actin-interfering agents stimulated MAPK signaling affect the cytoskeleton. and activator protein-1 activity. This led, in turn, to In this study, we show that MIAs stimulate the expression of induction of COX-2 gene expression via the cyclic AMP the COX-2 gene in human mammary epithelial cells. These response element site in the COX-2 promoter. effects were mediated via extracellular-regulated kinases (ERK1/2) and p38 MAPKs. Possibly, MIA-mediated induction of COX-2 will decrease the efficacy of these compounds as Two isoforms of cyclooxygenase (COX), designated COX-1 anti-cancer agents or explain, in part, the toxicity of these and COX-2 catalyze the synthesis of prostaglandins (PGs) from drugs. arachidonic acid. COX-1 is constitutively expressed in most tissues and appears to be responsible for various physiological EXPERIMENTAL PROCEDURES functions (1, 2). In contrast, COX-2 is not detectable in most Materials—Minimum Eagle’s medium and LipofectAMINE were normal tissues but is induced by oncogenes, growth factors, from Life Technologies, Inc. Keratinocyte basal medium (KBM) was cytokines, and tumor promoters (3–9). from Clonetics Corp. (San Diego, CA). Sodium arachidonate, epidermal COX-2 is an important target for treating arthritis, pain, and growth factor, 17-b-estradiol, 2-methoxyestradiol, hydrocortisone, and o-nitrophenyl-b-D-galactopyranoside were from Sigma. Calphostin C, * This work was supported in part by S/G 2P01 CA68425 and the taxol, colchicine, nocodazole, vinblastine, vincristine, and cytochalasin James E. Olson Memorial Fund. The costs of publication of this article D were from Biomol Research Laboratories, Inc. (Plymouth Meeting, were defrayed in part by the payment of page charges. This article must PA). PD 98059 (29-amino-39-methoxyflavone) and SB 202190 (4-(4-flu- therefore be hereby marked “advertisement” in accordance with 18 orophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole) were from U.S.C. Section 1734 solely to indicate this fact. Calbiochem. Enzyme immunoassay reagents for PGE assays were § To whom correspondence should be addressed: New York Presby- 32 32 32 from Cayman Co. (Ann Arbor, MI). [ P]ATP, [ P]CTP, and [ P]UTP terian Hospital-Cornell, Div. of Gastroenterology and Hepatology, Rm. were from NEN Life Science Products. Random-priming kits were from F-203, 1300 York Ave., New York, NY 10021. Tel.: 212-746-4402; Fax: Roche Molecular Biochemicals. Nitrocellulose membranes were from 212-746-4885; E-mail: [email protected]. 1 Schleicher & Schuell. Reagents for the luciferase assay were from The abbreviations used are: COX, cyclooxygenase; AP-1, activator Analytical Luminescence (San Diego, CA). The 18 S rRNA cDNA was protein-1; CRE, cyclic AMP response element; DN, dominant negative; from Ambion, Inc. (Austin, TX). Goat polyclonal anti-human COX-2 was ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to kinase; KBM, keratinocyte basal medium; MAPK, mitogen-activated phosphosphorylated forms of ERK1/2, c-Jun, and p38 were from New protein kinase; MIA, microtubule-interfering agent; PGs, prostaglan- dins; PGE England Biolabs Inc. (Beverly, MA). Western blotting detection re- , prostaglandin E . 2 2 14838 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Microtubule-interfering Agents Induce COX-2 14839 FIG.2. COX-2 protein is induced by MIAs or cytochalasin D. Cells were treated with MIAs or cytochalasin D for 4.5 h. Cellular lysate protein (25 mg/lane) was loaded onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently transferred onto nitrocellulose. Im- munoblots were probed with antibody specific for COX-2. A, lysate protein was from cells treated with vehicle (lane 1) or taxol (0.01, 0.1, 1, 5, 10 mM; lanes 2– 6). B, lysate protein was from cells treated with vehicle (lane 1) or colchicine (1, 5, 10 mM; lanes 2– 4). C, lysate protein was from cells treated with vehicle (lane 1) or vincristine (0.01, 0.1, 1, 10 mM; lanes 2–5). D, lysate protein was from cells treated with vehicle (lane 1)or17-b-estradiol (0.5, 1, 5, 10 mM; lanes 2–5). E, lysate protein was from cells treated with vehicle (lane 1) or 2-methoxyestradiol (0.025, 0.050, 0.1, 1, 5 mM; lanes 2– 6). F, lysate protein was from cells treated with vehicle (lane 1) or cytochalasin D (1, 5, 10 mM; lanes 2– 4). PGE was normalized to protein concentrations. Western Blotting—Cell lysates were prepared by treating cells with lysis buffer (150 mM NaCl, 100 mM Tris (pH 8.0), 1% Tween 20, 50 mM diethyldithiocarbamate, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluo- ride, 10 mg/ml aprotinin, 10 mg/ml trypsin inhibitor, and 10 mg/ml leupeptin). Lysates were sonicated for 20 s on ice and centrifuged at 10,000 3 g for 10 min to sediment the particulate material. The protein concentration of the supernatant was measured by the method of Lowry et al. (35). SDS-polyacrylamide gel electrophoresis was performed un- der reducing conditions on 10% polyacrylamide gels as described by Laemmli (36). The resolved proteins were transferred onto nitrocellu- lose sheets as detailed by Towbin et al. (37). The nitrocellulose mem- FIG.1. PGE production is increased by treatment with MIAs. brane was then incubated with primary antisera. Secondary antibody to 184B5/HER cells were treated with 0 –10 mM taxol (A), vinblastine (B), IgG conjugated to horseradish peroxidase was used. The blots were or colchicine (C) for 4.5 h. The medium was then replaced with fresh probed with the ECL Western blot detection system according to the medium containing 10 mM sodium arachidonate. 30 min later, the manufacturer’s instructions. medium was collected to determine the synthesis of PGE . Production of Northern Blotting—Total cellular RNA was isolated from cell mono- PGE was determined by enzyme immunoassay. Columns, means; bars, layers using an RNA isolation kit from Qiagen Inc. 10 mg of total S.D.; n 5 6. *, p , 0.01 compared with untreated cells. cellular RNA per lane were electrophoresed in a formaldehyde-contain- ing 1.2% agarose gel and transferred to nylon-supported membranes. agents (ECL) were from Amersham Pharmacia Biotech. Plasmid DNA After baking, membranes were prehybridized overnight in a solution was prepared using a kit from Promega Corp. (Madison, WI). containing 50% formamide, 53 sodium chloride/sodium phosphate/ Tissue Culture—The 184B5/HER cell line has been described previ- EDTA buffer (SSPE), 53 Denhardt’s solution, 0.1% SDS, and 100 mg/ml ously (34). Cells were maintained in minimum Eagle’s medium-KBM single-stranded salmon sperm DNA and then hybridized for 12 h at mixed in a ratio of 1:1 (basal medium) containing epidermal growth 42 °C with radiolabeled cDNA probes for human COX-2 cDNA and 18 S factor (10 ng/ml), hydrocortisone (0.5 mg/ml), transferrin (10 mg/ml), rRNA. After hybridization, membranes were washed twice for 20 min at gentamicin (5 mg/ml), and insulin (10 mg/ml) (growth medium). Cells room temperature in 23 SSPE, 0.1% SDS, twice for 20 min in the same were grown to 60% confluence, trypsinized with 0.05% trypsin-2 mM solution at 55 °C, and twice for 20 min in 0.13 SSPE, 0.1% SDS at EDTA, and plated for experimental use. In all experiments, 184B5/HER 55 °C. Washed membranes were then subjected to autoradiography. cells were grown in basal medium for 24 h before treatment. Treat- COX-2 and 18 S rRNA probes were labeled with [ P]CTP by random ments with vehicle (0.1% Me SO) or MIAs were always carried out in priming. basal medium. Nuclear Run-off Assay—2.5 3 10 cells were plated in four T150 PGE Production—5 3 10 cells/well were plated in 6-well dishes and dishes for each condition. Cells were grown in growth medium until grown to 60% confluence in growth medium. Levels of PGE released by approximately 60% confluent. Nuclei were isolated and stored in liquid the cells were measured by enzyme immunoassay (8). Production of nitrogen. For the transcription assay, nuclei (1.0 3 10 ) were thawed 14840 Microtubule-interfering Agents Induce COX-2 FIG.5. Rho is important for MIA-mediated induction of COX-2. Cells were transfected with 0.9 mg of a human COX-2 promoter con- struct ligated to luciferase (2327/159) and 0.2 mgofpSVbgal. Bars labeled Rho DN represent cells that received 0.9 mg of expression vector for dominant negative Rho. The total amount of DNA in each reaction was kept constant at 2 mg by using the corresponding empty expression vectors. Cells were treated with vehicle (Control), 10 mM taxol, or 10 mM cytochalasin D for 8 h. Luciferase activity represents data that have been normalized to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 0.01 compared with taxol or cytochalasin D. FIG.3. Microtubule- or actin-interfering agents induce COX-2 mRNA. Total RNA was isolated from cells that were treated with MIAs for 3 h. In each panel, lane 1 represents vehicle. Lanes 2–5 represent 1, 2.5, 5, and 10 mM taxol (A), colchicine (B), vinblastine (C), nocodazole (D), and cytochalasin D (E). 10 mg of RNA was added to each lane. Blots were hybridized with probes that recognized COX-2 mRNA and 18 S mRNA. FIG.4. COX-2 transcription is stimulated by MIAs and cy- tochalasin D. Cells were treated with vehicle (lane 1)or10 mM taxol (lane 2), cytochalasin D (lane 3), or colchicine (lane 4) for 2 h. Nuclear FIG.6. Induction of COX-2 by MIAs is mediated by protein run-offs were performed as described under “Experimental Proce- kinase C and Raf-1. A, cells were treated with vehicle (lane 1), 10 mM dures.” The COX-2 and 18 S rRNA cDNAs were immobilized onto taxol (lane 2), 10 mM taxol plus 1 mM calphostin C (lane 3), or 10 mM taxol nitrocellulose membranes and hybridized with labeled nascent RNA plus 2 mM calphostin C (lane 4) for 4.5 h. B, cells were treated with transcripts. vehicle (lane 1), 10 mM colchicine (lane 2), 10 mM colchicine plus 1 mM calphostin C (lane 3), or 10 mM colchicine plus 2 mM calphostin C (lane 4). In panels A and B, cellular lysate protein (25 mg/lane) was loaded and incubated in reaction buffer (10 mM Tris (pH 8), 5 mM MgCl , and onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently 0.3 M KCl) containing 100 mCi of uridine 59-[ P]triphosphate and 1 mM transferred onto nitrocellulose. Immunoblots were probed with anti- unlabeled nucleotides. After 30 min, labeled nascent RNA transcripts body specific for COX-2. C, cells were transfected with 0.9 mgofa were isolated. The human COX-2 and 18 S rRNA cDNAs were immo- human COX-2 promoter construct ligated to luciferase (2327/159) and bilized onto nitrocellulose and prehybridized overnight in hybridization 0.2 mgofpSVbgal. Bars labeled Raf-1 DN represent cells that received buffer. Hybridization was carried out at 42 °C for 24 h using equal 0.9 mg of expression vector for dominant negative Raf-1. The total cpm/ml labeled nascent RNA transcripts for each treatment group. The amount of DNA in each reaction was kept constant at 2 mg by using membranes were washed twice with 23 SSC (buffer for 1 h at 55 °C and corresponding empty expression vectors. Cells were treated with 10 mM then treated with 10 mg/ml RNase A in 23 SSC at 37 °C for 30 min, taxol (open bar), colchicine (black bar), or cytochalasin D (stippled bar) dried, and autoradiographed. for 8 h. Luciferase activity represents data that have been normalized Plasmids—The COX-2 promoter constructs (21432/159, 2327/159, to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 2220/159, 2124/159, 252/159, KBM, ILM, CRM) were a generous gift 0.01 compared with MIA or cytochalasin D. Microtubule-interfering Agents Induce COX-2 14841 of Dr. Tadashi Tanabe (National Cardiovascular Center Research In- TX). The DN expression vectors for JNK and p38 were generously stitute, Osaka, Japan) (6). The human COX-2 cDNA was generously provided by Dr. Roger Davis (University of Massachusetts, Worcester, provided by Dr. Stephen M. Prescott (University of Utah, Salt Lake MA). A c-Jun DN expression vector was a gift of Dr. Tom Curran (St. City, UT). The dominant negative (DN) expression vector for Rho was a Jude Children’s Research Hospital, Memphis, TN). pSV-bgal was ob- gift of Dr. Alan Hall (University College, London, UK). The expression tained from Promega Corp. vector for kinase-deficient Raf-1 was obtained from Dr. Ulf Rapp (Uni- Oligonucleotides—CRE-decoy and control oligonucleotides were versity of Wurzburg, Germany). The ERK1 DN expression vector was phosphorothioate oligonucleotides. Their sequences were as follows: obtained from Dr. Melanie Cobb (Southwestern Medical Center, Dallas, 24-mer CRE palindrome, 59-TGACGTCATGACGTCATGACGTCA-39; 24-mer CRE mismatch control, 59-TGTGGTCATGTGGTCATGTG- GTCA-39; and 24-mer mutant-sequence palindrome, 59-CTAGCTAGC- TAGCTAGCTAGCTAG-39. In addition, the following oligonucleotides containing the CRE of the COX-2 promoter were synthesized: 59-AAA- CAGTCATTTCGTCACATGGGCTTG-39 (sense), 59-CAAGCCCATGTG- ACGAAATGACTGTTT-39 (antisense). These oligonucleotides were syn- thesized by Genosys Biotechnologies, Inc. (The Woodlands, TX). Transient Transfection Assays—184B5/HER cells were seeded at a density of 5 3 10 cells/well in 6-well dishes and grown to 50 – 60% mg of plasmid DNA were introduced into confluence. For each well, 2 cells using 8 mg of LipofectAMINE as per the manufacturer’s instruc- tions. After7hof incubation, the medium was replaced with basal medium. The activities of luciferase and b-galactosidase were measured in cellular extract as described previously (38). Statistics—Comparisons between groups were made by the Student’s t test. A difference between groups of p , 0.05 was considered significant. FIG.7. Microtubule- or actin-interfering agents induce the ac- tivities of ERK1/2, p38, and JNK MAPK. A, cells were treated with RESULTS vehicle (lane 2)or10 mM colchicine (lane 3), taxol (lane 4), cytochalasin D(lane 5), nocodazole (lane 6) for 10 min. Lane 1, phospho-ERK1/2 Microtubule-interfering Agents Induce COX-2—The possibil- standard. B, cells were treated with vehicle (lane 1)or10 mM taxol (lane ity that MIAs could stimulate PG synthesis was investigated in 2), cytochalasin D (lane 3), vinblastine (lane 4), vincristine (lane 5), 184B5/HER human mammary epithelial cells. This HER-2/ colchicine (lane 6), nocodazole (lane 7) for 10 min. Lane 8, phospho-p38 neu-overexpressing mammary epithelial cell line was used be- standard. C, cells were treated with vehicle (lane 2)or10 mM taxol (lane 3), cytochalasin D (lane 4), colchicine (lane 5), nocodazole (lane 6), cause MIAs such as taxol are used to treat women with HER- vincristine (lane 7) for 15 min. Lane 1, phospho-c-Jun standard. Cellu- 2/neu-positive breast cancer. The data in Fig. 1 show that taxol, lar protein (100 mg/lane) was loaded onto a 10% SDS-polyacrylamide vinblastine, and colchicine caused dose-dependent increases in gel, electrophoresed, and subsequently transferred to nitrocellulose. PGE production. Nocodazole, vincristine, and 2-methoxyestra- Immunoblots were probed with antibodies to phosphorylated forms of 2 ERK1/2 (A), p38 (B), and c-Jun (C). diol also stimulated PG synthesis (data not shown). To deter- FIG.8. ERK1/2 and p38 MAPKs are important for MIA-mediated induction of COX-2. A, lysate protein was from cells treated with vehicle (lane 1), 10 mM taxol (lane 2), 10 mM taxol with 10 or 20 mM PD 98059 (lanes 3 and 4), 10 mM taxol with 5 or 10 mM SB 202190 (lanes 5 and 6). B, lysate protein was from cells treated with vehicle (lanes 1 and 6), colchicine (1 mM, lanes 2 and 7), colchicine (1 mM) with PD 98059 (10, 20, 30 mM, lanes 3–5), or colchicine (1 mM) with SB 202190 (1, 2.5, 5, 10 mM, lanes 8 –11) for 4.5 h. Lane 12 represents an ovine Cox-2 standard. Cellular lysate protein (25 mg/lane) was loaded onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently transferred onto nitrocellulose. Immuno- blots were probed with an antibody specific for COX-2. C, cells were transfected with 0.9 mg of a human COX-2 promoter construct ligated to luciferase (2327/159) and 0.2 mgofpSVbgal. Bars labeled ERK1 DN, JNK DN, p38 DN represent cells that received 0.9 mg of expression vector for ERK1 DN, JNK DN, and p38 DN, respectively. The total amount of DNA in each reaction was kept constant at 2 mg by using corresponding empty expression vectors. Cells were treated with vehicle or taxol (10 mM)for8h. Luciferase activity represents data that have been normalized to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 0.01 compared with taxol. 14842 Microtubule-interfering Agents Induce COX-2 mine whether these changes in PGE synthesis were related to differences in amounts of COX-2, Western blotting of cell lysate protein was carried out. Fig. 2 shows that multiple MIAs in- cluding taxol, colchicine, and vincristine induced COX-2 pro- tein. In contrast, COX-2 was not induced by 10-deacetylbacca- tin III and b-lumicolchicine, structural analogues of taxol and colchicine that do not interfere with microtubules (data not shown). Because of evidence (39) that estrogen and its deriva- tives interfere with microtubules, the effects of 17b-estradiol and 2-methoxyestradiol were determined. Treatment with 17b- estradiol or 2-methoxyestradiol induced COX-2 protein (Figs. 2, D and E). COX-2 was up-regulated at concentrations as low as 25 nM 2-methoxyestradiol (Fig. 2E). The extent of MIA-medi- ated induction of COX-2 was comparable with that observed following treatment with PMA (data not shown). To further elucidate the mechanism responsible for MIA- mediated induction of COX-2, we determined steady-state lev- els of COX-2 mRNA by Northern blotting (Fig. 3). Treatment with MIAs markedly increased amounts of COX-2 mRNA. Nu- clear run-offs were performed to determine if MIAs stimulated COX-2 transcription. As shown in Fig. 4, higher rates of syn- thesis of nascent COX-2 mRNA were detected after treatment with MIAs (taxol, colchicine). Defining the Signaling Mechanism by Which Microtubule- interfering Agents Induce COX-2 Expression—Rho is important for regulating both the cytoskeleton and COX-2. Hence, we determined whether a dominant negative form of Rho blocked MIA-mediated stimulation of COX-2 promoter activity. Taxol- mediated induction of COX-2 promoter activity was blocked by DN Rho (Fig. 5). Similar effects were observed with other MIAs (data not shown). By contrast, DN Ras did not inhibit MIA- FIG.9. Localization of region of COX-2 promoter that mediates mediated stimulation of COX-2 promoter activity (data not the effects of MIAs and cytochalasin D. A, schematic of human COX-2 promoter. B, 184B5/HER cells were transfected with 1.8 mgofa shown). Rho can activate protein kinase C, a known regulator series of human COX-2 promoter deletion constructs ligated to lucifer- of COX-2 transcription (40). Therefore, we determined whether ase (21432/159, 2327/159, 2220/159, 2124/159, 252/159), and 0.2 calphostin C, an inhibitor of protein kinase C, could block mg pSVbgal. C, 184B5/HER cells were transfected with 1.8 mg of a series MIA-mediated induction of COX-2. Calphostin C inhibited the of human COX-2 promoter-luciferase constructs (2327/159; KBM; ILM; CRM) and 0.2 mg pSVbgal. KBM represents the 2327/159 COX-2 induction of COX-2 protein by taxol (Fig. 6A) and colchicine promoter construct in which the NFkB site was mutagenized; ILM (Fig. 6B). represents the 2327/159 COX-2 promoter construct in which the NF- Protein kinase C can activate Raf-1 which, in turn, regulates IL6 site was mutagenized; CRM refers to the 2327/159 COX-2 pro- MAPK kinase activity (41). To determine the role of Raf-1 in moter construct in which the CRE was mutagenized. After transfection, cells were treated with vehicle or 10 mM taxol, colchicine, cytochalasin mediating the effects of MIAs on COX-2, transient transfec- D. Reporter activities were measured in cellular extract 7 h later. tions were performed. As shown in Fig. 6C, MIA-mediated Luciferase activity represents data that have been normalized with stimulation of COX-2 promoter activity was blocked by kinase- b-galactosidase. Columns, means; bars, SD; n 5 6. deficient Raf-1. Additionally, MIAs have been reported to in- duce MAPK activity. Treatment with MIAs activated ERK1/2, next attempted to define the region of the COX-2 promoter that p38, and JNK MAPKs (Fig. 7). Subsequently, experiments responded to MIAs. This was accomplished using a series of were done to demonstrate that MIA-mediated increases in COX-2 promoter deletion constructs. As shown in Fig. 9B, basal MAPK activity were linked to the induction of COX-2. In the COX-2 promoter activity was highest when the 21432/159 first experiment, we utilized PD 98059, a specific inhibitor of promoter construct was used. As the COX-2 promoter was MAPK kinase activity that prevents activation of ERK1 and shortened, lower basal activities were detected. Thus, the 252/ ERK2 (42). PD 98059 inhibited the induction of COX-2 by MIAs 159 COX-2 promoter construct was about 30% as active as the (Fig. 8). Similarly, SB 202190, a selective inhibitor of p38 full-length 21432/159 COX-2 promoter construct. Treatment MAPK activity (43), blocked MIA-mediated induction of COX-2 with MIAs markedly induced COX-2 promoter activity with all protein (Fig. 8). To further investigate the importance of promoter-deletion constructs except the 252/159 promoter MAPKs in mediating the effects of MIAs, a series of transient construct. A CRE is present between nucleotides 259 and 253, transfections was performed (Fig. 8C). The induction of COX-2 suggesting that this element was responsible for mediating the promoter activity by taxol was blocked by transiently overex- effects of MIAs. To test this idea, transient transfections were pressing dominant negatives for ERK1 or p38 MAPK. Similar performed using COX-2 promoter constructs in which specific results were obtained with colchicine (data not shown). enhancer elements including the CRE were mutagenized. As The Cyclic AMP Response Element and AP-1 Mediate the shown in Fig. 9C, mutagenizing the CRE site had multiple Induction of COX-2 by Microtubule-interfering Agents—Tran- effects including a decrease in basal promoter activity and a sient transfections were performed with human COX-2 promot- loss of responsiveness to MIAs. By contrast, mutagenizing the er-luciferase constructs (Fig. 9A) to identify the region of the NFkB and NF-IL-6 sites had little effect on COX-2 promoter COX-2 promoter that was responsible for MIA-mediated induc- activity. To confirm the importance of the CRE for mediating tion of COX-2. Treatment with MIAs (10 mM) led to about a the induction of COX-2 by MIAs, a separate series of transient 3-fold increase in COX-2 promoter (21432/159) activity. We transfections were performed. We examined the effects of a Microtubule-interfering Agents Induce COX-2 14843 FIG. 10. Increased binding of c-Jun, c-Fos, and ATF-2 to the CRE of the COX-2 promoter is detected in MIA-treated cells. A, 184B5 cells were transfected with 0.9 mg of a human COX-2 promoter construct ligated to luciferase (2327/159) (Control)or COX-2 promoter plus decoy CRE (0.4 mg) or COX-2 promoter plus mismatch CRE (0.4 mg) or COX-2 promoter plus mutant CRE (0.4 mg). All cells received 0.2 mgofpSVbgal. The total amount of DNA in each reaction was kept constant at 2 mg by using empty vector. Cells were treated with 10 mM taxol. Reporter activities were measured in cellular extract 8 h later. Luciferase activity represents data that have been normalized with b-galactosidase. Columns, means; bars, S.D.; n 5 6. B,in lanes 1–7,5 mg of nuclear protein from 184B5/HER cells was incubated with a P-labeled oligonucleotide containing the CRE of COX-2. Lane 1, vehicle-treated cells; lane 2, cells treated with 10 mM taxol for 30 min; lanes 3 and 4 represent nuclear extract from taxol-treated cells incubated with antibodies to c-Jun and ATF-2, respectively; lanes 5–7, cells treated with 10 mM vinblastine, cytochalasin D, and colchicine for 30 min. C,5 mg of nuclear protein from 184B5/HER cells was incubated with a P-labeled oligonucleotide containing the CRE of COX-2. Lane 1, cells treated with 10 mM taxol for 30 min; lanes 2 and 3, nuclear extract from taxol-treated cells incubated with 1 ml and 2 mlof antibody to c-Fos, respectively. In B and C, the protein DNA complex that formed was separated on a 4% polyacrylamide gel. CRE-decoy oligonucleotide on MIA-mediated stimulation of addition to the actin network, cytoplasmic microtubules repre- COX-2 promoter activity. The CRE-decoy oligonucleotide effec- sent another major element in the cytoskeleton that have been tively inhibited taxol-mediated activation of the COX-2 pro- implicated in intracellular signaling (26). The current results moter (Fig. 10A). In contrast, neither a CRE mismatch control show that agents that interfere with microtubules or actin oligonucleotide nor a nonsense-sequence palindrome blocked polymerization induce COX-2 gene expression and PG synthe- MIA-mediated induction of COX-2 promoter activity. sis in human mammary epithelial cells. Prior studies have Electrophoretic mobility shift assays were performed to iden- shown that cell transformation and Wnt signaling alter the tify the transcription factor that mediated the induction of cytoskeleton and induce COX-2 (8, 44, 45). Possibly, COX-2 will COX-2 by MIAs. MIAs caused increased binding to the CRE be induced by a range of biological processes which affect the site of the COX-2 promoter (Fig. 10B). The DNA binding com- cytoskeleton. plex induced by taxol was removed with antibodies to c-Jun, Rho GTPases control the organization of the actin cytoskel- ATF-2 (Fig. 10B) or c-Fos (Fig. 10C). In contrast, antibodies to eton (46). Additionally, COX-2 can be induced by a Rho-depend- NFkB p65 or NF-IL6 did not affect binding to the CRE (data ent signaling pathway (47). It is noteworthy, therefore, that a not shown). To further evaluate the importance of AP-1 for dominant negative form of Rho blocked both MIA- and cytocha- mediating the induction of COX-2 by MIAs, transient transfec- lasin D-mediated activation of the COX-2 promoter. As noted tions were performed. MIAs stimulated AP-1 promoter activity above, Rho can activate protein kinase C (40), which in turn is (Fig. 11A). Moreover, a dominant negative form of c-Jun inhib- a known regulator of COX-2 gene expression (48). The induc- ited the induction of COX-2 promoter activity by MIAs (Fig. tion of COX-2 by microtubule- or actin-interfering agents was 11B). Taken together, these results indicate that c-Jun is im- blocked by calphostin C, an inhibitor of protein kinase C activ- portant for mediating the induction of COX-2 by MIAs. ity. Although MIAs have been reported to activate gene expres- Cytochalasin D, an Inhibitor of Actin Polymerization, Stim- sion via the Ras pathway (29), to the best of our knowledge this ulates COX-2 Transcription—The induction of COX-2 by MIAs is the first time that the protein kinase C pathway has been could be mediated by effects on the cytoskeleton. Hence, we implicated. MIAs activate Raf-1 (27), a downstream target of also determined whether cytochalasin D, an inhibitor of actin protein kinase C. The induction of COX-2 promoter activity by polymerization, induced COX-2. Cytochalasin D induced microtubule- or actin-interfering agents was inhibited by ki- COX-2 protein (Fig. 2F) and COX-2 mRNA (Fig. 3E) by stim- nase-deficient Raf-1. This is consistent with prior evidence that ulating COX-2 transcription (Fig. 4). These inductive effects COX-2 is a Raf-1-dependent gene (48). were blocked by calphostin C (data not shown). Treatment with Previously, MIAs were found to induce p38, JNK, and ERK cytochalasin D increased the activity of ERK1/2 and p38 MAPK MAPK activities (27–29, 31). The expression of COX-2 can be (Fig. 7). Furthermore, dominant negative forms of ERK1 and affected by changes in MAPK activity (48 –52). Several lines of p38 MAPK blocked cytochalasin D-mediated activation of the evidence suggest that MIAs and cytochalasin D induce COX-2 COX-2 promoter (data not shown). Similar to the MIAs, the by activating ERK and p38 MAPKs. Treatment with MIAs or inductive effects of cytochalasin D localized to the CRE of the cytochalasin D stimulated the activities of ERK and p38 COX-2 promoter (Fig. 9). Moreover, c-Jun was important for MAPK; inhibitors of MAPK kinase and p38 MAPK blocked the both cytochalasin D- and MIA-mediated induction of COX-2 induction of COX-2 by microtubule- or actin-interfering agents. promoter activity (Fig. 11). Furthermore, overexpression of dominant negatives for ERK1 DISCUSSION or p38 suppressed the induction of COX-2 promoter activity by MIAs or cytochalasin D. Interestingly, the induction of JNK There is growing evidence that the cytoskeleton is involved in the propagation of signals that alter gene expression. In activity by MIAs or cytochalasin D did not appear to be neces- 14844 Microtubule-interfering Agents Induce COX-2 of AP-1 was established because MIA- or cytochalasin D-medi- ated induction of COX-2 promoter activity was suppressed by dominant negative c-Jun (Fig. 11B). This finding is consistent with previous reports that MIAs can stimulate AP-1 activity (29). The results are also consistent with the findings of Xie and Herschman (49, 50), who showed that, in response to expres- sion of v-src or treatment with platelet-derived growth factor, c-Jun induced murine Cox-2 via the CRE site. Tumor-promot- ing phorbol esters, transforming growth factor a, and ceramide also activate COX-2 transcription via the CRE site in the hu- man COX-2 promoter (48, 52, 54). Pharmacologic concentrations of estrogen inhibit cell prolif- eration in estrogen receptor-positive and -negative human breast cancer cell lines (39). This effect has been attributed to estradiol-induced microtubule disruption (39, 55). Hence, it was relevant to determine whether 17b-estradiol or 2-me- thoxyestradiol, a metabolite of estradiol, induced COX-2 in transformed human mammary epithelial cells. Both com- pounds induced COX-2; COX-2 was induced at concentrations as low as 25 nM 2-methoxyestradiol. Although the physiologic significance of this finding is uncertain, 30 nM 2-methoxyestro- gens has been detected in the serum of pregnant females (56). The products of COX-2 activity, i.e. PGs, stimulate cell pro- liferation (57), inhibit immune surveillance (58), increase the invasiveness of malignant cells (59), and enhance the produc- tion of vascular endothelial growth factor (60). Possibly, MIA- mediated induction of COX-2 will decrease the efficacy of taxol, vincristine, or vinblastine as anti-cancer agents. It will be worthwhile, therefore, to evaluate whether the addition of a selective COX-2 inhibitor can increase the antitumor activity of MIAs such as taxol. COX-2-derived PGs also contribute to inflammation and pain (11). Taxol has been reported to cause arthralgias and myalgias in up to 75% of patients (61, 62). The symptoms of joint and muscle discomfort may last for 3–5 days following completion of taxol therapy. The results of this study FIG. 11. MIAs and cytochalasin D activate the COX-2 promoter via c-Jun. A, cells were cotransfected with 1.8 mg of the collagenase raise the possibility that the toxicity (e.g. myalgias, arthral- promoter (which contains an AP-1 site) and 0.2 mgofpSVbgal. Twenty- gias) of drugs such as taxol could be mediated in part by four h after transfection, cells were treated with 0 –10 mM taxol, colchi- COX-2-derived PGs. Additional studies are warranted to deter- cine, or cytochalasin D. Reporter activities were measured in cellular mine whether COX-2 inhibitors can decrease the side effects of extract 8 h later. B, cells were cotransfected with a human COX-2 promoter construct ligated to luciferase (2327/159) and 0.2 mgof MIAs. pSVbgal. Bars labeled c-Jun DN represent cells that received 0.9 mgof expression vector for c-Jun DN. The total amount of DNA in each REFERENCES reaction was kept constant at 2 mg by using corresponding empty 1. Smith, W. L., Garavito, R. M., and DeWitt, D. L. (1996) J. Biol. Chem. 271, expression vectors. Twenty-four h after transfection, cells were treated 33157–33160 with vehicle or 10 mM taxol, colchicine, or cytochalasin D for 8 h. In 2. Smith, W. L., and DeWitt, D. L. (1995) Semin. Nephrol. 15, 179 –194 panels A and B, luciferase activity represents data that have been 3. Kujubu, D. A., Fletcher, B. S., Varnum, B. C., Lim, R. W., and Herschman, H. R. (1991) J. Biol. Chem. 266, 12866 –12872 normalized to b-galactosidase activity. Columns, means; bars, SD. n 5 4. Jones, D. A., Carlton, D. P., McIntyre, T. M., Zimmerman, G. A., and Prescott, 6. *, p , 0.01 compared with compound. S. M. (1993) J. Biol. Chem. 268, 9049 –9054 5. DuBois, R. N., Awad, J., Morrow, J., Roberts, L. J., and Bishop, P. R. (1994) J. Clin. Invest. 93, 493– 498 sary for induction of COX-2. For example, taxol-mediated stim- 6. Inoue, H., Yokoyama, C., Hara, S., Tone, Y., and Tanabe, T. (1995) J. Biol. Chem. 270, 24965–24971 ulation of COX-2 promoter activity was not inhibited by a 7. Simmons, D. L., Levy, D. B., Yannoni, Y., and Erikson, R. L. (1989) Proc. Natl. dominant negative form of JNK. This suggests that phospho- Acad. Sci. U. S. A. 86, 1178 –1182 rylation of c-Jun is not rate-limiting for MIA-mediated induc- 8. Subbaramaiah, K., Telang, N., Ramonetti, J. T., Araki, R., DeVito, B., Weksler, B. B., and Dannenberg, A. J. (1996) Cancer Res. 56, 4424 – 4429 tion of COX-2. This finding is potentially explained by the 9. Mestre, J. R., Subbaramaiah, K., Sacks, P. G., Schantz, S. P., Tanabe, T., presence of high levels of both unphosphorylated and phospho- Inoue, H., and Dannenberg, A. J. (1997) Cancer Res. 57, 2890 –2895 10. Anderson, G. D., Hauser, S. D., McGarity, K. L., Bremer, M. E., Isakson, P. C., rylated c-Jun in 184B5/HER cells under basal conditions. and Gregory, S. A. (1996) J. Clin. Invest. 97, 2672–2679 The AP-1 transcription factor complex consists of a collection 11. Lipsky, P. E., and Isakson, P. C. (1997) J. Rheumatol. 24, (Suppl. 49), 9 –14 of dimers of members of the Jun, Fos, and ATF/cAMP-response 12. Dannenberg, A. J., and Zakim, D. (1999) Semin. Oncol. 26, 499 –504 13. Crofford, L. J., Wilder, R. L., Ristimaki, A. P., Sano, H., Remmers, E. F., Epps, element-binding protein bZip families. MAPKs regulate AP-1 H. R., and Hla, T. (1994) J. Clin. Invest. 93, 1095–1101 activity by both increasing the abundance of AP-1 components 14. Kutchera, W., Jones, D. A., Matsunami, N., Groden, J., McIntyre, T. M., Zimmerman, G. A., White, R. L., and Prescott, S. M. (1996) Proc. Natl. Acad. and stimulating their activity (53). ERK1/2 MAPK regulates Sci. U. S. A. 93, 4816 – 4820 the transcription of c-Fos. p38 MAPK phosphorylates and ac- 15. Sheng, G. G., Shao, J., Sheng, H., Hooton, E. B., Isakson, P. C., Morrow, J. D., tivates ATF-2 (53). The current results suggest that the AP-1 Coffey, R. J., DuBois, R. N., and Beauchamp, R. D. (1997) Gastroenterology 113, 1883–1891 transcription factor complex mediates the induction of COX-2 16. Kargman, S. L., O’Neil, G. P., Vickers, P. J., Evans, J. F., Mancini, J. A., and by microtubule- or actin-interfering agents. Treatment with Jothy, S. (1995) Cancer Res. 55, 2556 –2559 17. Ristimaki, A., Honkanen, N., Jankala, H., Sipponen, P., and Harkonen, M. MIAs or cytochalasin D augmented binding to the CRE of the (1997) Cancer Res. 57, 1276 –1280 COX-2 promoter (Fig. 10); c-Jun, c-Fos, and ATF-2 were iden- 18. Parett, M. L., Harris, R. E., Joarder, F. S., Ross, M. S., Clausen, K. P., and tified in the DNA binding complex. The functional importance Robertson, F. M. (1997) Int. J. Oncol. 10, 503–507 Microtubule-interfering Agents Induce COX-2 14845 19. Chan, G., Boyle, J. O., Yang, E. K., Zhang, F., Sacks, P. G., Shah, J. P., 41. Reunanen, N., Westermarck, J., Hakkinen, L., Holmstrom, T. H., Elo, I., Edelstein, D., Soslow, R. A., Koki, A. T., Woerner, B. M., Masferrer, J. L., Eriksson, J. E., and Kahari, V.-M. (1998) J. Biol. Chem. 273, 5137–5145 and Dannenberg, A. J. (1999) Cancer Res. 59, 991–994 42. Dudley, D. T., Pang, L., Decker, S. T., Bridges, A. J., and Saltiel, A. R. (1995) 20. Oshima, M., Dinchuk, J. E., Kargman, S. L., Oshima, H., Hancock, B., Kwong, Proc. Natl. Acad. Sci. U. S. A. 92, 7686 –7689 E., Trzaskos, J. M., Evans, J. F., and Taketo, M. M. (1996) Cell 87, 803– 809 43. Jiang, Y., Chen, C., Li, Z., Guo, W., Gegner, J. A., Lin, S., and Han, J. (1996) 21. Tiano, H., Chulada, P., Spalding, J., Lee, C., Loftin, C., Mahler, J., Morham, S., J. Biol. Chem. 271, 17920 –17926 and Langenbach, R. (1997) Proc. Am. Assoc. Cancer Res. 38, 1727 (abstr.) 44. Gunderson, G. G., and Cook, T. A. (1999) Curr. Opin. Cell Biol. 11, 81–94 22. Kawamori, T., Rao, C. V., Seibert, K., and Reddy, B. S. (1998) Cancer Res. 58, 45. Howe, L. R., Subbaramaiah, K., Chung, W. J., Dannenberg, A. J., and Brown, 409 – 412 A. M. C. (1999) Cancer Res. 59, 1572–1577 23. Fischer, S. M., Lo, H-H., Gordon, G. B., Seibert, K., Kelloff, G., Lubet, R. A., 46. Mackay, D. J. G., and Hall, A. (1998) J. Biol. Chem. 273, 20685–20688 and Conti, C. C. (1999) Mol. Carcinog. 25, 231–240 47. Slice, L. W., Walsh, J. H., and Rozengurt, E. (1999) J. Biol. Chem. 274, 24. Sheng, H., Shao, J., Kirkland, S. C., Isakson, P., Coffey, R. J., Morrow, J., 27562–27566 Beauchamp, R. D., and DuBois, R. N. (1997) J. Clin. Invest. 99, 2254 –2259 48. Subbaramaiah, K., Chung, W. J., and Dannenberg, A. J. (1998) J. Biol. Chem. 25. Sawaoka, H., Kawano, S., Tsuji, S., Tsujii, M., Gunawan, E. S., Takei, Y., 273, 32943–32949 Nagano, K., and Hori, M. (1998) Am. J. Physiol. 274, G1061-G1067 49. Xie, W., and Herschman, H. R. (1995) J. Biol. Chem. 270, 27622–27628 26. Jordan, M. A., and Wilson, L. (1998) Curr. Opin. Cell Biol. 10, 123–130 50. Xie, W., and Herschman, H. R. (1996) J. Biol. Chem. 271, 31742–31748 27. Schmid-Alliana, A., Menou, L., Manie, S., Schmid-Antomarchi, H., Millet, 51. Guan, Z., Buckman, S. Y., Pentland, A. P., Templeton, D. J., and Morrison, M-A., Giuriato, S., Ferrua, B., and Rossi, B. (1998) J. Biol. Chem. 273, A. R. (1998) J. Biol. Chem. 273, 12901–12908 3394 –3400 52. Matsuura, H., Sakaue, M., Subbaramaiah, K., Kamitani, H., Eling, T. E., 28. Lee, L.-F., Li, G., Templeton, D. J., and Ting, J. P.-Y. (1998) J. Biol. Chem. 273, Dannenberg, A. J., Tanabe, T., Inoue, H., Arata, J., and Jetten, A. M. (1999) 28253–28260 J. Biol. Chem. 274, 29138 –29148 29. Wang, T.-H., Wang, H.-S., Ichijo, H., Giannakakou, P., Foster, J. S., Fojo, T., 53. Whitmarsh, A. J., and Davis, R. J. (1996) J. Mol. Med. 74, 589 – 607 and Wimalasena, J. (1998) J. Biol. Chem. 273, 4928 – 4936 54. Subbaramaiah, K., Chung, W. J., Michaluart, P., Telang, N., Tanabe, T., 30. Moos, P. J., and Fitzpatrick, F. A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, Inoue, H., Jang, M., Pezzuto, J. M., and Dannenberg, A. J. (1998) J. Biol. 3896 –3901 Chem. 273, 21875–21882 31. Moos, P. J., Muskardin, D. T., and Fitzpatrick, F. A. (1999) J. Immunol.162, 55. D’Amato, R. J., Lin, C. M., Flynn, E., Folkman, J., and Hamel, E. (1994) Proc. 467– 473 Natl. Acad. Sci. U. S. A. 91, 3964 –3968 32. Gemsa, D., Kramer, W., Brenner, M., Till, G., and Resch, K. (1980) J. Immu- 56. Berg, D., Sonsallla, R., and Kuss, E. (1983) Acta Endocrinol. 103, 282–288 nol.124, 376 –380 57. Sheng, H., Shao, J., Morrow, J. D., Beauchamp, R. D., and DuBois, R. N. (1998) 33. Yeh, C-K., and Rodan, G. A. (1987) Biochim. Biophys. Acta 927, 315–323 Cancer Res. 58, 362–366 34. Zhai, Y-F., Beittenmiller, H., Wang, B., Gould, M. N., Oakley, C., Esselman, 58. Goodwin, J. S., and Ceuppens, J. (1983) J. Clin. Immunol. 3, 295–315 W. J., and Welsch, C. W. (1993) Cancer Res. 53, 2272–2278 59. Tsujii, M., Kawano, S., and DuBois, R. N. (1997) Proc. Natl. Acad. Sci. U. S. A. 35. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. 94, 3336 –3340 Chem. 193, 265–275 60. Tsujii, M., Kawano, S., Tsuji, S., Sawaoka, H., Hori, M., and DuBois, R. N. 36. Laemmli, U. K. (1970) Nature 227, 680 – 685 (1998) Cell 93, 705–716 37. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 61. Holmes, F. A., Walters, R. S., Theriault, R. L., Forman, A. D., Newton, L. K., 76, 4350 – 4354 Raber, M. N., Buzdar, A. U., Frye, D. K., and Hortobagyi, G. N. (1991) 38. Mestre, J. R., Subbaramaiah, K., Sacks, P. G., Schantz, S. P., Tanabe, T., J. Natl. Cancer Inst. 83, 1797–1805 Inoue, H., and Dannenberg, A. J. (1997) Cancer Res. 57, 1081–1085 39. Aizu-Yokata, E., Ichinoseki, K., and Sato, Y. (1994) Carcinogenesis 15, 62. Eisenhauer, E. A., Huinink, W. W. T. B., Swenerton, K. D., Gianni, L., Myles, J., van der Burg, M. E. L., Kerr, I., Vermorken, J. B., Buser, K., Colombo, 1875–1879 40. Chang, J. H., Pratt, J. C., Sawasdikosol, S., Kapeller, R., and Burakoff, S. J. N., Bacon, M., Santabarbara, P., Onetto, N., Winograd, B., and Canetta, R. (1998) Mol. Cell. Biol. 18, 4986 – 4993 (1994) J. Clin. Oncol. 12, 1654 –2666 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Microtubule-interfering Agents Stimulate the Transcription of Cyclooxygenase-2

Subbaramaiah, Kotha; Hart, Janice C.; Norton, Larry; Dannenberg, Andrew J.
Journal of Biological Chemistry – May 1, 2000
Download PDF
8 pages

Loading next page...
 
/lp/unpaywall/microtubule-interfering-agents-stimulate-the-transcription-of-7yuGk0HySm

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.275.20.14838
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 275, No. 20, Issue of May 19, pp. 14838 –14845, 2000 © 2000 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Microtubule-interfering Agents Stimulate the Transcription of Cyclooxygenase-2 EVIDENCE FOR INVOLVEMENT OF ERK1/2 AND p38 MITOGEN-ACTIVATED PROTEIN KINASE PATHWAYS* Received for publication, January 18, 2000, and in revised form, February 16, 2000 Kotha Subbaramaiah‡§, Janice C. Hart‡, Larry Norton , and Andrew J. Dannenberg‡ From the ‡Department of Medicine, New York Presbyterian Hospital-Cornell and Strang Cancer Prevention Center and Breast Cancer Medicine Service, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 possibly cancer (10 –12). For example, the expression of COX-2 We investigated whether microtubule-interfering agents (MIAs: taxol, colchicine, nocodazole, vinblastine, is increased in inflamed tissues such as rheumatoid synovium vincristine, 17-b-estradiol, 2-methoxyestradiol) altered (13), and selective COX-2 inhibitors are useful for the treat- cyclooxygenase-2 (COX-2) expression in human mam- ment of arthritis (11). COX-2 is also overexpressed in trans- mary epithelial cells. MIAs enhanced prostaglandin E 2 formed cells (8, 14, 15) and in tumors (16 –19). Mice engineered synthesis and increased levels of COX-2 protein and to be COX-2-deficient are protected against developing intesti- mRNA. Nuclear run-off assays revealed increased rates nal tumors (20) and skin papillomas (21). Additionally, selec- of COX-2 transcription after treatment with MIAs. Cal- tive COX-2 inhibitors prevent the formation of intestinal (20, phostin C, an inhibitor of protein kinase C, blocked the 22) and skin tumors (23) in experimental animals and suppress induction of COX-2 by MIAs. The stimulation of COX-2 the growth of transplantable tumors (24, 25). With this in promoter activity by MIAs was inhibited by overex- mind, it is reasonable to postulate that compounds that induce pressing dominant negative forms of Rho and Raf-1. COX-2 could predispose to cancer or inflammation. MIAs stimulated ERK, JNK, and p38 mitogen-activated Microtubule-interfering agents (MIAs) are widely used for protein kinases (MAPK); pharmacological inhibitors of the treatment of cancer. The anti-cancer properties of MIAs MAPK kinase and p38 MAPK blocked the induction of have been attributed in part to interference with microtubule COX-2 by MIAs. Overexpressing dominant negative forms of ERK1 or p38 MAPK inhibited MIA-mediated assembly, impairment of mitosis, and changes in cytoskeleton activation of the COX-2 promoter. MIAs stimulated the (26). There is growing evidence, however, that MIAs have mul- binding of the activator protein-1 transcription factor tiple cellular effects. For example, MIAs stimulate mitogen- complex to the cyclic AMP response element in the activated protein kinases (MAPKs) and gene expression (27– COX-2 promoter. A dominant negative form of c-Jun 29). Taxol, a novel anti-cancer drug and MIA, up-regulates inhibited the activation of the COX-2 promoter by MIAs. COX-2 and synthesis of PGE (30, 31). Colchicine and vinblas- Additionally, cytochalasin D, an agent that inhibits ac- tine, two other MIAs, stimulate PGE production (32, 33). tin polymerization, stimulated COX-2 transcription by These findings raise the possibility that COX-2 is a down- the same signaling pathway as MIAs. Thus, microtubule- stream target of MIAs and possibly other compounds that or actin-interfering agents stimulated MAPK signaling affect the cytoskeleton. and activator protein-1 activity. This led, in turn, to In this study, we show that MIAs stimulate the expression of induction of COX-2 gene expression via the cyclic AMP the COX-2 gene in human mammary epithelial cells. These response element site in the COX-2 promoter. effects were mediated via extracellular-regulated kinases (ERK1/2) and p38 MAPKs. Possibly, MIA-mediated induction of COX-2 will decrease the efficacy of these compounds as Two isoforms of cyclooxygenase (COX), designated COX-1 anti-cancer agents or explain, in part, the toxicity of these and COX-2 catalyze the synthesis of prostaglandins (PGs) from drugs. arachidonic acid. COX-1 is constitutively expressed in most tissues and appears to be responsible for various physiological EXPERIMENTAL PROCEDURES functions (1, 2). In contrast, COX-2 is not detectable in most Materials—Minimum Eagle’s medium and LipofectAMINE were normal tissues but is induced by oncogenes, growth factors, from Life Technologies, Inc. Keratinocyte basal medium (KBM) was cytokines, and tumor promoters (3–9). from Clonetics Corp. (San Diego, CA). Sodium arachidonate, epidermal COX-2 is an important target for treating arthritis, pain, and growth factor, 17-b-estradiol, 2-methoxyestradiol, hydrocortisone, and o-nitrophenyl-b-D-galactopyranoside were from Sigma. Calphostin C, * This work was supported in part by S/G 2P01 CA68425 and the taxol, colchicine, nocodazole, vinblastine, vincristine, and cytochalasin James E. Olson Memorial Fund. The costs of publication of this article D were from Biomol Research Laboratories, Inc. (Plymouth Meeting, were defrayed in part by the payment of page charges. This article must PA). PD 98059 (29-amino-39-methoxyflavone) and SB 202190 (4-(4-flu- therefore be hereby marked “advertisement” in accordance with 18 orophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole) were from U.S.C. Section 1734 solely to indicate this fact. Calbiochem. Enzyme immunoassay reagents for PGE assays were § To whom correspondence should be addressed: New York Presby- 32 32 32 from Cayman Co. (Ann Arbor, MI). [ P]ATP, [ P]CTP, and [ P]UTP terian Hospital-Cornell, Div. of Gastroenterology and Hepatology, Rm. were from NEN Life Science Products. Random-priming kits were from F-203, 1300 York Ave., New York, NY 10021. Tel.: 212-746-4402; Fax: Roche Molecular Biochemicals. Nitrocellulose membranes were from 212-746-4885; E-mail: [email protected]. 1 Schleicher & Schuell. Reagents for the luciferase assay were from The abbreviations used are: COX, cyclooxygenase; AP-1, activator Analytical Luminescence (San Diego, CA). The 18 S rRNA cDNA was protein-1; CRE, cyclic AMP response element; DN, dominant negative; from Ambion, Inc. (Austin, TX). Goat polyclonal anti-human COX-2 was ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to kinase; KBM, keratinocyte basal medium; MAPK, mitogen-activated phosphosphorylated forms of ERK1/2, c-Jun, and p38 were from New protein kinase; MIA, microtubule-interfering agent; PGs, prostaglan- dins; PGE England Biolabs Inc. (Beverly, MA). Western blotting detection re- , prostaglandin E . 2 2 14838 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Microtubule-interfering Agents Induce COX-2 14839 FIG.2. COX-2 protein is induced by MIAs or cytochalasin D. Cells were treated with MIAs or cytochalasin D for 4.5 h. Cellular lysate protein (25 mg/lane) was loaded onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently transferred onto nitrocellulose. Im- munoblots were probed with antibody specific for COX-2. A, lysate protein was from cells treated with vehicle (lane 1) or taxol (0.01, 0.1, 1, 5, 10 mM; lanes 2– 6). B, lysate protein was from cells treated with vehicle (lane 1) or colchicine (1, 5, 10 mM; lanes 2– 4). C, lysate protein was from cells treated with vehicle (lane 1) or vincristine (0.01, 0.1, 1, 10 mM; lanes 2–5). D, lysate protein was from cells treated with vehicle (lane 1)or17-b-estradiol (0.5, 1, 5, 10 mM; lanes 2–5). E, lysate protein was from cells treated with vehicle (lane 1) or 2-methoxyestradiol (0.025, 0.050, 0.1, 1, 5 mM; lanes 2– 6). F, lysate protein was from cells treated with vehicle (lane 1) or cytochalasin D (1, 5, 10 mM; lanes 2– 4). PGE was normalized to protein concentrations. Western Blotting—Cell lysates were prepared by treating cells with lysis buffer (150 mM NaCl, 100 mM Tris (pH 8.0), 1% Tween 20, 50 mM diethyldithiocarbamate, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluo- ride, 10 mg/ml aprotinin, 10 mg/ml trypsin inhibitor, and 10 mg/ml leupeptin). Lysates were sonicated for 20 s on ice and centrifuged at 10,000 3 g for 10 min to sediment the particulate material. The protein concentration of the supernatant was measured by the method of Lowry et al. (35). SDS-polyacrylamide gel electrophoresis was performed un- der reducing conditions on 10% polyacrylamide gels as described by Laemmli (36). The resolved proteins were transferred onto nitrocellu- lose sheets as detailed by Towbin et al. (37). The nitrocellulose mem- FIG.1. PGE production is increased by treatment with MIAs. brane was then incubated with primary antisera. Secondary antibody to 184B5/HER cells were treated with 0 –10 mM taxol (A), vinblastine (B), IgG conjugated to horseradish peroxidase was used. The blots were or colchicine (C) for 4.5 h. The medium was then replaced with fresh probed with the ECL Western blot detection system according to the medium containing 10 mM sodium arachidonate. 30 min later, the manufacturer’s instructions. medium was collected to determine the synthesis of PGE . Production of Northern Blotting—Total cellular RNA was isolated from cell mono- PGE was determined by enzyme immunoassay. Columns, means; bars, layers using an RNA isolation kit from Qiagen Inc. 10 mg of total S.D.; n 5 6. *, p , 0.01 compared with untreated cells. cellular RNA per lane were electrophoresed in a formaldehyde-contain- ing 1.2% agarose gel and transferred to nylon-supported membranes. agents (ECL) were from Amersham Pharmacia Biotech. Plasmid DNA After baking, membranes were prehybridized overnight in a solution was prepared using a kit from Promega Corp. (Madison, WI). containing 50% formamide, 53 sodium chloride/sodium phosphate/ Tissue Culture—The 184B5/HER cell line has been described previ- EDTA buffer (SSPE), 53 Denhardt’s solution, 0.1% SDS, and 100 mg/ml ously (34). Cells were maintained in minimum Eagle’s medium-KBM single-stranded salmon sperm DNA and then hybridized for 12 h at mixed in a ratio of 1:1 (basal medium) containing epidermal growth 42 °C with radiolabeled cDNA probes for human COX-2 cDNA and 18 S factor (10 ng/ml), hydrocortisone (0.5 mg/ml), transferrin (10 mg/ml), rRNA. After hybridization, membranes were washed twice for 20 min at gentamicin (5 mg/ml), and insulin (10 mg/ml) (growth medium). Cells room temperature in 23 SSPE, 0.1% SDS, twice for 20 min in the same were grown to 60% confluence, trypsinized with 0.05% trypsin-2 mM solution at 55 °C, and twice for 20 min in 0.13 SSPE, 0.1% SDS at EDTA, and plated for experimental use. In all experiments, 184B5/HER 55 °C. Washed membranes were then subjected to autoradiography. cells were grown in basal medium for 24 h before treatment. Treat- COX-2 and 18 S rRNA probes were labeled with [ P]CTP by random ments with vehicle (0.1% Me SO) or MIAs were always carried out in priming. basal medium. Nuclear Run-off Assay—2.5 3 10 cells were plated in four T150 PGE Production—5 3 10 cells/well were plated in 6-well dishes and dishes for each condition. Cells were grown in growth medium until grown to 60% confluence in growth medium. Levels of PGE released by approximately 60% confluent. Nuclei were isolated and stored in liquid the cells were measured by enzyme immunoassay (8). Production of nitrogen. For the transcription assay, nuclei (1.0 3 10 ) were thawed 14840 Microtubule-interfering Agents Induce COX-2 FIG.5. Rho is important for MIA-mediated induction of COX-2. Cells were transfected with 0.9 mg of a human COX-2 promoter con- struct ligated to luciferase (2327/159) and 0.2 mgofpSVbgal. Bars labeled Rho DN represent cells that received 0.9 mg of expression vector for dominant negative Rho. The total amount of DNA in each reaction was kept constant at 2 mg by using the corresponding empty expression vectors. Cells were treated with vehicle (Control), 10 mM taxol, or 10 mM cytochalasin D for 8 h. Luciferase activity represents data that have been normalized to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 0.01 compared with taxol or cytochalasin D. FIG.3. Microtubule- or actin-interfering agents induce COX-2 mRNA. Total RNA was isolated from cells that were treated with MIAs for 3 h. In each panel, lane 1 represents vehicle. Lanes 2–5 represent 1, 2.5, 5, and 10 mM taxol (A), colchicine (B), vinblastine (C), nocodazole (D), and cytochalasin D (E). 10 mg of RNA was added to each lane. Blots were hybridized with probes that recognized COX-2 mRNA and 18 S mRNA. FIG.4. COX-2 transcription is stimulated by MIAs and cy- tochalasin D. Cells were treated with vehicle (lane 1)or10 mM taxol (lane 2), cytochalasin D (lane 3), or colchicine (lane 4) for 2 h. Nuclear FIG.6. Induction of COX-2 by MIAs is mediated by protein run-offs were performed as described under “Experimental Proce- kinase C and Raf-1. A, cells were treated with vehicle (lane 1), 10 mM dures.” The COX-2 and 18 S rRNA cDNAs were immobilized onto taxol (lane 2), 10 mM taxol plus 1 mM calphostin C (lane 3), or 10 mM taxol nitrocellulose membranes and hybridized with labeled nascent RNA plus 2 mM calphostin C (lane 4) for 4.5 h. B, cells were treated with transcripts. vehicle (lane 1), 10 mM colchicine (lane 2), 10 mM colchicine plus 1 mM calphostin C (lane 3), or 10 mM colchicine plus 2 mM calphostin C (lane 4). In panels A and B, cellular lysate protein (25 mg/lane) was loaded and incubated in reaction buffer (10 mM Tris (pH 8), 5 mM MgCl , and onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently 0.3 M KCl) containing 100 mCi of uridine 59-[ P]triphosphate and 1 mM transferred onto nitrocellulose. Immunoblots were probed with anti- unlabeled nucleotides. After 30 min, labeled nascent RNA transcripts body specific for COX-2. C, cells were transfected with 0.9 mgofa were isolated. The human COX-2 and 18 S rRNA cDNAs were immo- human COX-2 promoter construct ligated to luciferase (2327/159) and bilized onto nitrocellulose and prehybridized overnight in hybridization 0.2 mgofpSVbgal. Bars labeled Raf-1 DN represent cells that received buffer. Hybridization was carried out at 42 °C for 24 h using equal 0.9 mg of expression vector for dominant negative Raf-1. The total cpm/ml labeled nascent RNA transcripts for each treatment group. The amount of DNA in each reaction was kept constant at 2 mg by using membranes were washed twice with 23 SSC (buffer for 1 h at 55 °C and corresponding empty expression vectors. Cells were treated with 10 mM then treated with 10 mg/ml RNase A in 23 SSC at 37 °C for 30 min, taxol (open bar), colchicine (black bar), or cytochalasin D (stippled bar) dried, and autoradiographed. for 8 h. Luciferase activity represents data that have been normalized Plasmids—The COX-2 promoter constructs (21432/159, 2327/159, to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 2220/159, 2124/159, 252/159, KBM, ILM, CRM) were a generous gift 0.01 compared with MIA or cytochalasin D. Microtubule-interfering Agents Induce COX-2 14841 of Dr. Tadashi Tanabe (National Cardiovascular Center Research In- TX). The DN expression vectors for JNK and p38 were generously stitute, Osaka, Japan) (6). The human COX-2 cDNA was generously provided by Dr. Roger Davis (University of Massachusetts, Worcester, provided by Dr. Stephen M. Prescott (University of Utah, Salt Lake MA). A c-Jun DN expression vector was a gift of Dr. Tom Curran (St. City, UT). The dominant negative (DN) expression vector for Rho was a Jude Children’s Research Hospital, Memphis, TN). pSV-bgal was ob- gift of Dr. Alan Hall (University College, London, UK). The expression tained from Promega Corp. vector for kinase-deficient Raf-1 was obtained from Dr. Ulf Rapp (Uni- Oligonucleotides—CRE-decoy and control oligonucleotides were versity of Wurzburg, Germany). The ERK1 DN expression vector was phosphorothioate oligonucleotides. Their sequences were as follows: obtained from Dr. Melanie Cobb (Southwestern Medical Center, Dallas, 24-mer CRE palindrome, 59-TGACGTCATGACGTCATGACGTCA-39; 24-mer CRE mismatch control, 59-TGTGGTCATGTGGTCATGTG- GTCA-39; and 24-mer mutant-sequence palindrome, 59-CTAGCTAGC- TAGCTAGCTAGCTAG-39. In addition, the following oligonucleotides containing the CRE of the COX-2 promoter were synthesized: 59-AAA- CAGTCATTTCGTCACATGGGCTTG-39 (sense), 59-CAAGCCCATGTG- ACGAAATGACTGTTT-39 (antisense). These oligonucleotides were syn- thesized by Genosys Biotechnologies, Inc. (The Woodlands, TX). Transient Transfection Assays—184B5/HER cells were seeded at a density of 5 3 10 cells/well in 6-well dishes and grown to 50 – 60% mg of plasmid DNA were introduced into confluence. For each well, 2 cells using 8 mg of LipofectAMINE as per the manufacturer’s instruc- tions. After7hof incubation, the medium was replaced with basal medium. The activities of luciferase and b-galactosidase were measured in cellular extract as described previously (38). Statistics—Comparisons between groups were made by the Student’s t test. A difference between groups of p , 0.05 was considered significant. FIG.7. Microtubule- or actin-interfering agents induce the ac- tivities of ERK1/2, p38, and JNK MAPK. A, cells were treated with RESULTS vehicle (lane 2)or10 mM colchicine (lane 3), taxol (lane 4), cytochalasin D(lane 5), nocodazole (lane 6) for 10 min. Lane 1, phospho-ERK1/2 Microtubule-interfering Agents Induce COX-2—The possibil- standard. B, cells were treated with vehicle (lane 1)or10 mM taxol (lane ity that MIAs could stimulate PG synthesis was investigated in 2), cytochalasin D (lane 3), vinblastine (lane 4), vincristine (lane 5), 184B5/HER human mammary epithelial cells. This HER-2/ colchicine (lane 6), nocodazole (lane 7) for 10 min. Lane 8, phospho-p38 neu-overexpressing mammary epithelial cell line was used be- standard. C, cells were treated with vehicle (lane 2)or10 mM taxol (lane 3), cytochalasin D (lane 4), colchicine (lane 5), nocodazole (lane 6), cause MIAs such as taxol are used to treat women with HER- vincristine (lane 7) for 15 min. Lane 1, phospho-c-Jun standard. Cellu- 2/neu-positive breast cancer. The data in Fig. 1 show that taxol, lar protein (100 mg/lane) was loaded onto a 10% SDS-polyacrylamide vinblastine, and colchicine caused dose-dependent increases in gel, electrophoresed, and subsequently transferred to nitrocellulose. PGE production. Nocodazole, vincristine, and 2-methoxyestra- Immunoblots were probed with antibodies to phosphorylated forms of 2 ERK1/2 (A), p38 (B), and c-Jun (C). diol also stimulated PG synthesis (data not shown). To deter- FIG.8. ERK1/2 and p38 MAPKs are important for MIA-mediated induction of COX-2. A, lysate protein was from cells treated with vehicle (lane 1), 10 mM taxol (lane 2), 10 mM taxol with 10 or 20 mM PD 98059 (lanes 3 and 4), 10 mM taxol with 5 or 10 mM SB 202190 (lanes 5 and 6). B, lysate protein was from cells treated with vehicle (lanes 1 and 6), colchicine (1 mM, lanes 2 and 7), colchicine (1 mM) with PD 98059 (10, 20, 30 mM, lanes 3–5), or colchicine (1 mM) with SB 202190 (1, 2.5, 5, 10 mM, lanes 8 –11) for 4.5 h. Lane 12 represents an ovine Cox-2 standard. Cellular lysate protein (25 mg/lane) was loaded onto a 10% SDS-polyacrylamide gel, electrophoresed, and subsequently transferred onto nitrocellulose. Immuno- blots were probed with an antibody specific for COX-2. C, cells were transfected with 0.9 mg of a human COX-2 promoter construct ligated to luciferase (2327/159) and 0.2 mgofpSVbgal. Bars labeled ERK1 DN, JNK DN, p38 DN represent cells that received 0.9 mg of expression vector for ERK1 DN, JNK DN, and p38 DN, respectively. The total amount of DNA in each reaction was kept constant at 2 mg by using corresponding empty expression vectors. Cells were treated with vehicle or taxol (10 mM)for8h. Luciferase activity represents data that have been normalized to b-galactosidase activity. Columns, means; bars, S.D.; n 5 6; *, p , 0.01 compared with taxol. 14842 Microtubule-interfering Agents Induce COX-2 mine whether these changes in PGE synthesis were related to differences in amounts of COX-2, Western blotting of cell lysate protein was carried out. Fig. 2 shows that multiple MIAs in- cluding taxol, colchicine, and vincristine induced COX-2 pro- tein. In contrast, COX-2 was not induced by 10-deacetylbacca- tin III and b-lumicolchicine, structural analogues of taxol and colchicine that do not interfere with microtubules (data not shown). Because of evidence (39) that estrogen and its deriva- tives interfere with microtubules, the effects of 17b-estradiol and 2-methoxyestradiol were determined. Treatment with 17b- estradiol or 2-methoxyestradiol induced COX-2 protein (Figs. 2, D and E). COX-2 was up-regulated at concentrations as low as 25 nM 2-methoxyestradiol (Fig. 2E). The extent of MIA-medi- ated induction of COX-2 was comparable with that observed following treatment with PMA (data not shown). To further elucidate the mechanism responsible for MIA- mediated induction of COX-2, we determined steady-state lev- els of COX-2 mRNA by Northern blotting (Fig. 3). Treatment with MIAs markedly increased amounts of COX-2 mRNA. Nu- clear run-offs were performed to determine if MIAs stimulated COX-2 transcription. As shown in Fig. 4, higher rates of syn- thesis of nascent COX-2 mRNA were detected after treatment with MIAs (taxol, colchicine). Defining the Signaling Mechanism by Which Microtubule- interfering Agents Induce COX-2 Expression—Rho is important for regulating both the cytoskeleton and COX-2. Hence, we determined whether a dominant negative form of Rho blocked MIA-mediated stimulation of COX-2 promoter activity. Taxol- mediated induction of COX-2 promoter activity was blocked by DN Rho (Fig. 5). Similar effects were observed with other MIAs (data not shown). By contrast, DN Ras did not inhibit MIA- FIG.9. Localization of region of COX-2 promoter that mediates mediated stimulation of COX-2 promoter activity (data not the effects of MIAs and cytochalasin D. A, schematic of human COX-2 promoter. B, 184B5/HER cells were transfected with 1.8 mgofa shown). Rho can activate protein kinase C, a known regulator series of human COX-2 promoter deletion constructs ligated to lucifer- of COX-2 transcription (40). Therefore, we determined whether ase (21432/159, 2327/159, 2220/159, 2124/159, 252/159), and 0.2 calphostin C, an inhibitor of protein kinase C, could block mg pSVbgal. C, 184B5/HER cells were transfected with 1.8 mg of a series MIA-mediated induction of COX-2. Calphostin C inhibited the of human COX-2 promoter-luciferase constructs (2327/159; KBM; ILM; CRM) and 0.2 mg pSVbgal. KBM represents the 2327/159 COX-2 induction of COX-2 protein by taxol (Fig. 6A) and colchicine promoter construct in which the NFkB site was mutagenized; ILM (Fig. 6B). represents the 2327/159 COX-2 promoter construct in which the NF- Protein kinase C can activate Raf-1 which, in turn, regulates IL6 site was mutagenized; CRM refers to the 2327/159 COX-2 pro- MAPK kinase activity (41). To determine the role of Raf-1 in moter construct in which the CRE was mutagenized. After transfection, cells were treated with vehicle or 10 mM taxol, colchicine, cytochalasin mediating the effects of MIAs on COX-2, transient transfec- D. Reporter activities were measured in cellular extract 7 h later. tions were performed. As shown in Fig. 6C, MIA-mediated Luciferase activity represents data that have been normalized with stimulation of COX-2 promoter activity was blocked by kinase- b-galactosidase. Columns, means; bars, SD; n 5 6. deficient Raf-1. Additionally, MIAs have been reported to in- duce MAPK activity. Treatment with MIAs activated ERK1/2, next attempted to define the region of the COX-2 promoter that p38, and JNK MAPKs (Fig. 7). Subsequently, experiments responded to MIAs. This was accomplished using a series of were done to demonstrate that MIA-mediated increases in COX-2 promoter deletion constructs. As shown in Fig. 9B, basal MAPK activity were linked to the induction of COX-2. In the COX-2 promoter activity was highest when the 21432/159 first experiment, we utilized PD 98059, a specific inhibitor of promoter construct was used. As the COX-2 promoter was MAPK kinase activity that prevents activation of ERK1 and shortened, lower basal activities were detected. Thus, the 252/ ERK2 (42). PD 98059 inhibited the induction of COX-2 by MIAs 159 COX-2 promoter construct was about 30% as active as the (Fig. 8). Similarly, SB 202190, a selective inhibitor of p38 full-length 21432/159 COX-2 promoter construct. Treatment MAPK activity (43), blocked MIA-mediated induction of COX-2 with MIAs markedly induced COX-2 promoter activity with all protein (Fig. 8). To further investigate the importance of promoter-deletion constructs except the 252/159 promoter MAPKs in mediating the effects of MIAs, a series of transient construct. A CRE is present between nucleotides 259 and 253, transfections was performed (Fig. 8C). The induction of COX-2 suggesting that this element was responsible for mediating the promoter activity by taxol was blocked by transiently overex- effects of MIAs. To test this idea, transient transfections were pressing dominant negatives for ERK1 or p38 MAPK. Similar performed using COX-2 promoter constructs in which specific results were obtained with colchicine (data not shown). enhancer elements including the CRE were mutagenized. As The Cyclic AMP Response Element and AP-1 Mediate the shown in Fig. 9C, mutagenizing the CRE site had multiple Induction of COX-2 by Microtubule-interfering Agents—Tran- effects including a decrease in basal promoter activity and a sient transfections were performed with human COX-2 promot- loss of responsiveness to MIAs. By contrast, mutagenizing the er-luciferase constructs (Fig. 9A) to identify the region of the NFkB and NF-IL-6 sites had little effect on COX-2 promoter COX-2 promoter that was responsible for MIA-mediated induc- activity. To confirm the importance of the CRE for mediating tion of COX-2. Treatment with MIAs (10 mM) led to about a the induction of COX-2 by MIAs, a separate series of transient 3-fold increase in COX-2 promoter (21432/159) activity. We transfections were performed. We examined the effects of a Microtubule-interfering Agents Induce COX-2 14843 FIG. 10. Increased binding of c-Jun, c-Fos, and ATF-2 to the CRE of the COX-2 promoter is detected in MIA-treated cells. A, 184B5 cells were transfected with 0.9 mg of a human COX-2 promoter construct ligated to luciferase (2327/159) (Control)or COX-2 promoter plus decoy CRE (0.4 mg) or COX-2 promoter plus mismatch CRE (0.4 mg) or COX-2 promoter plus mutant CRE (0.4 mg). All cells received 0.2 mgofpSVbgal. The total amount of DNA in each reaction was kept constant at 2 mg by using empty vector. Cells were treated with 10 mM taxol. Reporter activities were measured in cellular extract 8 h later. Luciferase activity represents data that have been normalized with b-galactosidase. Columns, means; bars, S.D.; n 5 6. B,in lanes 1–7,5 mg of nuclear protein from 184B5/HER cells was incubated with a P-labeled oligonucleotide containing the CRE of COX-2. Lane 1, vehicle-treated cells; lane 2, cells treated with 10 mM taxol for 30 min; lanes 3 and 4 represent nuclear extract from taxol-treated cells incubated with antibodies to c-Jun and ATF-2, respectively; lanes 5–7, cells treated with 10 mM vinblastine, cytochalasin D, and colchicine for 30 min. C,5 mg of nuclear protein from 184B5/HER cells was incubated with a P-labeled oligonucleotide containing the CRE of COX-2. Lane 1, cells treated with 10 mM taxol for 30 min; lanes 2 and 3, nuclear extract from taxol-treated cells incubated with 1 ml and 2 mlof antibody to c-Fos, respectively. In B and C, the protein DNA complex that formed was separated on a 4% polyacrylamide gel. CRE-decoy oligonucleotide on MIA-mediated stimulation of addition to the actin network, cytoplasmic microtubules repre- COX-2 promoter activity. The CRE-decoy oligonucleotide effec- sent another major element in the cytoskeleton that have been tively inhibited taxol-mediated activation of the COX-2 pro- implicated in intracellular signaling (26). The current results moter (Fig. 10A). In contrast, neither a CRE mismatch control show that agents that interfere with microtubules or actin oligonucleotide nor a nonsense-sequence palindrome blocked polymerization induce COX-2 gene expression and PG synthe- MIA-mediated induction of COX-2 promoter activity. sis in human mammary epithelial cells. Prior studies have Electrophoretic mobility shift assays were performed to iden- shown that cell transformation and Wnt signaling alter the tify the transcription factor that mediated the induction of cytoskeleton and induce COX-2 (8, 44, 45). Possibly, COX-2 will COX-2 by MIAs. MIAs caused increased binding to the CRE be induced by a range of biological processes which affect the site of the COX-2 promoter (Fig. 10B). The DNA binding com- cytoskeleton. plex induced by taxol was removed with antibodies to c-Jun, Rho GTPases control the organization of the actin cytoskel- ATF-2 (Fig. 10B) or c-Fos (Fig. 10C). In contrast, antibodies to eton (46). Additionally, COX-2 can be induced by a Rho-depend- NFkB p65 or NF-IL6 did not affect binding to the CRE (data ent signaling pathway (47). It is noteworthy, therefore, that a not shown). To further evaluate the importance of AP-1 for dominant negative form of Rho blocked both MIA- and cytocha- mediating the induction of COX-2 by MIAs, transient transfec- lasin D-mediated activation of the COX-2 promoter. As noted tions were performed. MIAs stimulated AP-1 promoter activity above, Rho can activate protein kinase C (40), which in turn is (Fig. 11A). Moreover, a dominant negative form of c-Jun inhib- a known regulator of COX-2 gene expression (48). The induc- ited the induction of COX-2 promoter activity by MIAs (Fig. tion of COX-2 by microtubule- or actin-interfering agents was 11B). Taken together, these results indicate that c-Jun is im- blocked by calphostin C, an inhibitor of protein kinase C activ- portant for mediating the induction of COX-2 by MIAs. ity. Although MIAs have been reported to activate gene expres- Cytochalasin D, an Inhibitor of Actin Polymerization, Stim- sion via the Ras pathway (29), to the best of our knowledge this ulates COX-2 Transcription—The induction of COX-2 by MIAs is the first time that the protein kinase C pathway has been could be mediated by effects on the cytoskeleton. Hence, we implicated. MIAs activate Raf-1 (27), a downstream target of also determined whether cytochalasin D, an inhibitor of actin protein kinase C. The induction of COX-2 promoter activity by polymerization, induced COX-2. Cytochalasin D induced microtubule- or actin-interfering agents was inhibited by ki- COX-2 protein (Fig. 2F) and COX-2 mRNA (Fig. 3E) by stim- nase-deficient Raf-1. This is consistent with prior evidence that ulating COX-2 transcription (Fig. 4). These inductive effects COX-2 is a Raf-1-dependent gene (48). were blocked by calphostin C (data not shown). Treatment with Previously, MIAs were found to induce p38, JNK, and ERK cytochalasin D increased the activity of ERK1/2 and p38 MAPK MAPK activities (27–29, 31). The expression of COX-2 can be (Fig. 7). Furthermore, dominant negative forms of ERK1 and affected by changes in MAPK activity (48 –52). Several lines of p38 MAPK blocked cytochalasin D-mediated activation of the evidence suggest that MIAs and cytochalasin D induce COX-2 COX-2 promoter (data not shown). Similar to the MIAs, the by activating ERK and p38 MAPKs. Treatment with MIAs or inductive effects of cytochalasin D localized to the CRE of the cytochalasin D stimulated the activities of ERK and p38 COX-2 promoter (Fig. 9). Moreover, c-Jun was important for MAPK; inhibitors of MAPK kinase and p38 MAPK blocked the both cytochalasin D- and MIA-mediated induction of COX-2 induction of COX-2 by microtubule- or actin-interfering agents. promoter activity (Fig. 11). Furthermore, overexpression of dominant negatives for ERK1 DISCUSSION or p38 suppressed the induction of COX-2 promoter activity by MIAs or cytochalasin D. Interestingly, the induction of JNK There is growing evidence that the cytoskeleton is involved in the propagation of signals that alter gene expression. In activity by MIAs or cytochalasin D did not appear to be neces- 14844 Microtubule-interfering Agents Induce COX-2 of AP-1 was established because MIA- or cytochalasin D-medi- ated induction of COX-2 promoter activity was suppressed by dominant negative c-Jun (Fig. 11B). This finding is consistent with previous reports that MIAs can stimulate AP-1 activity (29). The results are also consistent with the findings of Xie and Herschman (49, 50), who showed that, in response to expres- sion of v-src or treatment with platelet-derived growth factor, c-Jun induced murine Cox-2 via the CRE site. Tumor-promot- ing phorbol esters, transforming growth factor a, and ceramide also activate COX-2 transcription via the CRE site in the hu- man COX-2 promoter (48, 52, 54). Pharmacologic concentrations of estrogen inhibit cell prolif- eration in estrogen receptor-positive and -negative human breast cancer cell lines (39). This effect has been attributed to estradiol-induced microtubule disruption (39, 55). Hence, it was relevant to determine whether 17b-estradiol or 2-me- thoxyestradiol, a metabolite of estradiol, induced COX-2 in transformed human mammary epithelial cells. Both com- pounds induced COX-2; COX-2 was induced at concentrations as low as 25 nM 2-methoxyestradiol. Although the physiologic significance of this finding is uncertain, 30 nM 2-methoxyestro- gens has been detected in the serum of pregnant females (56). The products of COX-2 activity, i.e. PGs, stimulate cell pro- liferation (57), inhibit immune surveillance (58), increase the invasiveness of malignant cells (59), and enhance the produc- tion of vascular endothelial growth factor (60). Possibly, MIA- mediated induction of COX-2 will decrease the efficacy of taxol, vincristine, or vinblastine as anti-cancer agents. It will be worthwhile, therefore, to evaluate whether the addition of a selective COX-2 inhibitor can increase the antitumor activity of MIAs such as taxol. COX-2-derived PGs also contribute to inflammation and pain (11). Taxol has been reported to cause arthralgias and myalgias in up to 75% of patients (61, 62). The symptoms of joint and muscle discomfort may last for 3–5 days following completion of taxol therapy. The results of this study FIG. 11. MIAs and cytochalasin D activate the COX-2 promoter via c-Jun. A, cells were cotransfected with 1.8 mg of the collagenase raise the possibility that the toxicity (e.g. myalgias, arthral- promoter (which contains an AP-1 site) and 0.2 mgofpSVbgal. Twenty- gias) of drugs such as taxol could be mediated in part by four h after transfection, cells were treated with 0 –10 mM taxol, colchi- COX-2-derived PGs. Additional studies are warranted to deter- cine, or cytochalasin D. Reporter activities were measured in cellular mine whether COX-2 inhibitors can decrease the side effects of extract 8 h later. B, cells were cotransfected with a human COX-2 promoter construct ligated to luciferase (2327/159) and 0.2 mgof MIAs. pSVbgal. Bars labeled c-Jun DN represent cells that received 0.9 mgof expression vector for c-Jun DN. The total amount of DNA in each REFERENCES reaction was kept constant at 2 mg by using corresponding empty 1. Smith, W. L., Garavito, R. M., and DeWitt, D. L. (1996) J. Biol. Chem. 271, expression vectors. Twenty-four h after transfection, cells were treated 33157–33160 with vehicle or 10 mM taxol, colchicine, or cytochalasin D for 8 h. In 2. Smith, W. L., and DeWitt, D. L. (1995) Semin. Nephrol. 15, 179 –194 panels A and B, luciferase activity represents data that have been 3. Kujubu, D. A., Fletcher, B. S., Varnum, B. C., Lim, R. W., and Herschman, H. R. (1991) J. Biol. Chem. 266, 12866 –12872 normalized to b-galactosidase activity. Columns, means; bars, SD. n 5 4. Jones, D. A., Carlton, D. P., McIntyre, T. M., Zimmerman, G. A., and Prescott, 6. *, p , 0.01 compared with compound. S. M. (1993) J. Biol. Chem. 268, 9049 –9054 5. DuBois, R. N., Awad, J., Morrow, J., Roberts, L. J., and Bishop, P. R. (1994) J. Clin. Invest. 93, 493– 498 sary for induction of COX-2. For example, taxol-mediated stim- 6. Inoue, H., Yokoyama, C., Hara, S., Tone, Y., and Tanabe, T. (1995) J. Biol. Chem. 270, 24965–24971 ulation of COX-2 promoter activity was not inhibited by a 7. Simmons, D. L., Levy, D. B., Yannoni, Y., and Erikson, R. L. (1989) Proc. Natl. dominant negative form of JNK. This suggests that phospho- Acad. Sci. U. S. A. 86, 1178 –1182 rylation of c-Jun is not rate-limiting for MIA-mediated induc- 8. Subbaramaiah, K., Telang, N., Ramonetti, J. T., Araki, R., DeVito, B., Weksler, B. B., and Dannenberg, A. J. (1996) Cancer Res. 56, 4424 – 4429 tion of COX-2. This finding is potentially explained by the 9. Mestre, J. R., Subbaramaiah, K., Sacks, P. G., Schantz, S. P., Tanabe, T., presence of high levels of both unphosphorylated and phospho- Inoue, H., and Dannenberg, A. J. (1997) Cancer Res. 57, 2890 –2895 10. Anderson, G. D., Hauser, S. D., McGarity, K. L., Bremer, M. E., Isakson, P. C., rylated c-Jun in 184B5/HER cells under basal conditions. and Gregory, S. A. (1996) J. Clin. Invest. 97, 2672–2679 The AP-1 transcription factor complex consists of a collection 11. Lipsky, P. E., and Isakson, P. C. (1997) J. Rheumatol. 24, (Suppl. 49), 9 –14 of dimers of members of the Jun, Fos, and ATF/cAMP-response 12. Dannenberg, A. J., and Zakim, D. (1999) Semin. Oncol. 26, 499 –504 13. Crofford, L. J., Wilder, R. L., Ristimaki, A. P., Sano, H., Remmers, E. F., Epps, element-binding protein bZip families. MAPKs regulate AP-1 H. R., and Hla, T. (1994) J. Clin. Invest. 93, 1095–1101 activity by both increasing the abundance of AP-1 components 14. Kutchera, W., Jones, D. A., Matsunami, N., Groden, J., McIntyre, T. M., Zimmerman, G. A., White, R. L., and Prescott, S. M. (1996) Proc. Natl. Acad. and stimulating their activity (53). ERK1/2 MAPK regulates Sci. U. S. A. 93, 4816 – 4820 the transcription of c-Fos. p38 MAPK phosphorylates and ac- 15. Sheng, G. G., Shao, J., Sheng, H., Hooton, E. B., Isakson, P. C., Morrow, J. D., tivates ATF-2 (53). The current results suggest that the AP-1 Coffey, R. J., DuBois, R. N., and Beauchamp, R. D. (1997) Gastroenterology 113, 1883–1891 transcription factor complex mediates the induction of COX-2 16. Kargman, S. L., O’Neil, G. P., Vickers, P. J., Evans, J. F., Mancini, J. A., and by microtubule- or actin-interfering agents. Treatment with Jothy, S. (1995) Cancer Res. 55, 2556 –2559 17. Ristimaki, A., Honkanen, N., Jankala, H., Sipponen, P., and Harkonen, M. MIAs or cytochalasin D augmented binding to the CRE of the (1997) Cancer Res. 57, 1276 –1280 COX-2 promoter (Fig. 10); c-Jun, c-Fos, and ATF-2 were iden- 18. Parett, M. L., Harris, R. E., Joarder, F. S., Ross, M. S., Clausen, K. P., and tified in the DNA binding complex. The functional importance Robertson, F. M. (1997) Int. J. Oncol. 10, 503–507 Microtubule-interfering Agents Induce COX-2 14845 19. Chan, G., Boyle, J. O., Yang, E. K., Zhang, F., Sacks, P. G., Shah, J. P., 41. Reunanen, N., Westermarck, J., Hakkinen, L., Holmstrom, T. H., Elo, I., Edelstein, D., Soslow, R. A., Koki, A. T., Woerner, B. M., Masferrer, J. L., Eriksson, J. E., and Kahari, V.-M. (1998) J. Biol. Chem. 273, 5137–5145 and Dannenberg, A. J. (1999) Cancer Res. 59, 991–994 42. Dudley, D. T., Pang, L., Decker, S. T., Bridges, A. J., and Saltiel, A. R. (1995) 20. Oshima, M., Dinchuk, J. E., Kargman, S. L., Oshima, H., Hancock, B., Kwong, Proc. Natl. Acad. Sci. U. S. A. 92, 7686 –7689 E., Trzaskos, J. M., Evans, J. F., and Taketo, M. M. (1996) Cell 87, 803– 809 43. Jiang, Y., Chen, C., Li, Z., Guo, W., Gegner, J. A., Lin, S., and Han, J. (1996) 21. Tiano, H., Chulada, P., Spalding, J., Lee, C., Loftin, C., Mahler, J., Morham, S., J. Biol. Chem. 271, 17920 –17926 and Langenbach, R. (1997) Proc. Am. Assoc. Cancer Res. 38, 1727 (abstr.) 44. Gunderson, G. G., and Cook, T. A. (1999) Curr. Opin. Cell Biol. 11, 81–94 22. Kawamori, T., Rao, C. V., Seibert, K., and Reddy, B. S. (1998) Cancer Res. 58, 45. Howe, L. R., Subbaramaiah, K., Chung, W. J., Dannenberg, A. J., and Brown, 409 – 412 A. M. C. (1999) Cancer Res. 59, 1572–1577 23. Fischer, S. M., Lo, H-H., Gordon, G. B., Seibert, K., Kelloff, G., Lubet, R. A., 46. Mackay, D. J. G., and Hall, A. (1998) J. Biol. Chem. 273, 20685–20688 and Conti, C. C. (1999) Mol. Carcinog. 25, 231–240 47. Slice, L. W., Walsh, J. H., and Rozengurt, E. (1999) J. Biol. Chem. 274, 24. Sheng, H., Shao, J., Kirkland, S. C., Isakson, P., Coffey, R. J., Morrow, J., 27562–27566 Beauchamp, R. D., and DuBois, R. N. (1997) J. Clin. Invest. 99, 2254 –2259 48. Subbaramaiah, K., Chung, W. J., and Dannenberg, A. J. (1998) J. Biol. Chem. 25. Sawaoka, H., Kawano, S., Tsuji, S., Tsujii, M., Gunawan, E. S., Takei, Y., 273, 32943–32949 Nagano, K., and Hori, M. (1998) Am. J. Physiol. 274, G1061-G1067 49. Xie, W., and Herschman, H. R. (1995) J. Biol. Chem. 270, 27622–27628 26. Jordan, M. A., and Wilson, L. (1998) Curr. Opin. Cell Biol. 10, 123–130 50. Xie, W., and Herschman, H. R. (1996) J. Biol. Chem. 271, 31742–31748 27. Schmid-Alliana, A., Menou, L., Manie, S., Schmid-Antomarchi, H., Millet, 51. Guan, Z., Buckman, S. Y., Pentland, A. P., Templeton, D. J., and Morrison, M-A., Giuriato, S., Ferrua, B., and Rossi, B. (1998) J. Biol. Chem. 273, A. R. (1998) J. Biol. Chem. 273, 12901–12908 3394 –3400 52. Matsuura, H., Sakaue, M., Subbaramaiah, K., Kamitani, H., Eling, T. E., 28. Lee, L.-F., Li, G., Templeton, D. J., and Ting, J. P.-Y. (1998) J. Biol. Chem. 273, Dannenberg, A. J., Tanabe, T., Inoue, H., Arata, J., and Jetten, A. M. (1999) 28253–28260 J. Biol. Chem. 274, 29138 –29148 29. Wang, T.-H., Wang, H.-S., Ichijo, H., Giannakakou, P., Foster, J. S., Fojo, T., 53. Whitmarsh, A. J., and Davis, R. J. (1996) J. Mol. Med. 74, 589 – 607 and Wimalasena, J. (1998) J. Biol. Chem. 273, 4928 – 4936 54. Subbaramaiah, K., Chung, W. J., Michaluart, P., Telang, N., Tanabe, T., 30. Moos, P. J., and Fitzpatrick, F. A. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, Inoue, H., Jang, M., Pezzuto, J. M., and Dannenberg, A. J. (1998) J. Biol. 3896 –3901 Chem. 273, 21875–21882 31. Moos, P. J., Muskardin, D. T., and Fitzpatrick, F. A. (1999) J. Immunol.162, 55. D’Amato, R. J., Lin, C. M., Flynn, E., Folkman, J., and Hamel, E. (1994) Proc. 467– 473 Natl. Acad. Sci. U. S. A. 91, 3964 –3968 32. Gemsa, D., Kramer, W., Brenner, M., Till, G., and Resch, K. (1980) J. Immu- 56. Berg, D., Sonsallla, R., and Kuss, E. (1983) Acta Endocrinol. 103, 282–288 nol.124, 376 –380 57. Sheng, H., Shao, J., Morrow, J. D., Beauchamp, R. D., and DuBois, R. N. (1998) 33. Yeh, C-K., and Rodan, G. A. (1987) Biochim. Biophys. Acta 927, 315–323 Cancer Res. 58, 362–366 34. Zhai, Y-F., Beittenmiller, H., Wang, B., Gould, M. N., Oakley, C., Esselman, 58. Goodwin, J. S., and Ceuppens, J. (1983) J. Clin. Immunol. 3, 295–315 W. J., and Welsch, C. W. (1993) Cancer Res. 53, 2272–2278 59. Tsujii, M., Kawano, S., and DuBois, R. N. (1997) Proc. Natl. Acad. Sci. U. S. A. 35. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. 94, 3336 –3340 Chem. 193, 265–275 60. Tsujii, M., Kawano, S., Tsuji, S., Sawaoka, H., Hori, M., and DuBois, R. N. 36. Laemmli, U. K. (1970) Nature 227, 680 – 685 (1998) Cell 93, 705–716 37. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. U. S. A. 61. Holmes, F. A., Walters, R. S., Theriault, R. L., Forman, A. D., Newton, L. K., 76, 4350 – 4354 Raber, M. N., Buzdar, A. U., Frye, D. K., and Hortobagyi, G. N. (1991) 38. Mestre, J. R., Subbaramaiah, K., Sacks, P. G., Schantz, S. P., Tanabe, T., J. Natl. Cancer Inst. 83, 1797–1805 Inoue, H., and Dannenberg, A. J. (1997) Cancer Res. 57, 1081–1085 39. Aizu-Yokata, E., Ichinoseki, K., and Sato, Y. (1994) Carcinogenesis 15, 62. Eisenhauer, E. A., Huinink, W. W. T. B., Swenerton, K. D., Gianni, L., Myles, J., van der Burg, M. E. L., Kerr, I., Vermorken, J. B., Buser, K., Colombo, 1875–1879 40. Chang, J. H., Pratt, J. C., Sawasdikosol, S., Kapeller, R., and Burakoff, S. J. N., Bacon, M., Santabarbara, P., Onetto, N., Winograd, B., and Canetta, R. (1998) Mol. Cell. Biol. 18, 4986 – 4993 (1994) J. Clin. Oncol. 12, 1654 –2666

Journal

Journal of Biological Chemistry – Unpaywall

Published: May 1, 2000

There are no references for this article.