Activating Transcription Factor 3 and Early Growth Response 1 Are the Novel Targets of LY294002 in a Phosphatidylinositol 3-Kinase–Independent Pathway

Research Article

Kiyoshi Yamaguchi,1 Seong-Ho Lee,1 Jong-Sik Kim,3 Jay Wimalasena,2 Shigetaka Kitajima,4 and Seung Joon Baek1

Departments of 1Pathobiology and 2Obstetrics and Gynecology, University of Tennessee, Knoxville, Tennessee; 3Department of Biological Science, Andong National University, Andong, Kyungpook, South Korea; and 4Department of Biochemical Genetics, Medical Research Institute, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan

Abstract Indeed, activated PI3K signaling followed by phosphorylation of LY294002, a phosphatidylinositol 3-kinase (PI3K) inhibitor, Akt leads to progression of colon adenocarcinoma tumorigenesis has been widely used to study the function of PI3K in cellular (2). Inhibition of this pathway results in the suppression of cell responses. Based on its inhibitory effect on PI3K, LY294002 proliferation and/or the induction of apoptosis in several types has been shown to exert antitumorigenic effect in vivo and of cancer cells. PI3K inhibitors LY294002 and wortmannin have in vitro. Here, we report that LY294002 alters early growth been widely used to study the role of PI3K in cellular responses. response 1 (EGR-1) phosphorylation and subsequently enhan- LY294002 shows antitumorigenic activity in human colon cancer ces activating transcription factor 3 (ATF3) expression cells in vivo (3) and the inhibition of PI3K by LY294002 can cause independently of PI3K inhibition. This pathway may be, in induction of apoptosis and cell growth arrest in vitro (3, 4). part, responsible for the antitumorigenic effect of LY294002 in LY294002 treatment induces several proapoptotic genes, including human colorectal cancer cells. ATF3 expression was increased nonsteroidal anti-inflammatory drug-activated gene (NAG-1), in by LY294002, followed by the induction of apoptosis in several HCT-116 cells (4) and wortmannin treatment induces tumor colorectal cancer cell lines. This is consistent with results necrosis factor–related apoptosis-inducing ligand expression in showing that the down-regulation of the ATF3 gene by small HT-29, HCT-116, and Caco-2 colon cancer cells (5). In addition, interfering RNA suppressed LY294002-induced apoptosis in LY294002 enhances sodium butyrate–induced apoptosis and HCT-116 cells. On the other hand, ATF3 expression was not decreases KM20 cell viability (6). Thus, therapies using LY294002 affected by another PI3K inhibitor, wortmannin, as well as may provide promising new alternatives for the treatment of phosphatase and tensin homologue or dominant-negative Akt colorectal cancer. overexpression. We also found that LY294002 increases ATF3 Transcription factors play a key role in controlling cell pro- promoter activity and the transactivation is partly mediated liferation, cell cycle progression, and apoptosis and also have been by a GC-rich sequence located in the promoter. EGR-1 binds to known as potential targets for therapeutic drugs (7). Activating the ATF3 promoter as assessed by gel shift assay. Furthermore, transcription factor 3 (ATF3) encodes a member of the ATF/cAMP- phosphorylated EGR-1 was highly increased in LY294002- responsive element binding protein (CREB) family of transcription treated cells, indicating that EGR-1 phosphorylation induced factors (8, 9). ATF3 has been known as a stress-inducible gene and by LY294002 may facilitate ATF3 transactivation. Our data induction of the ATF3 expression correlates with cellular damage. suggest that EGR-1 acts as a mediator in LY294002-induced For instance, ATF3 expression is regulated by a variety of signals, ATF3 expression via a PI3K-independent pathway. ATF3 and including cytokines, genotoxic agents, and compounds which EGR-1 may provide a novel explanation for the antitumori- induce cell death (10). Although the physiologic functions of genic properties of LY294002 in human colorectal cancer cells. ATF3 are not well understood, there is some evidence that ATF3 (Cancer Res 2006; 66(4): 2376-84) functions as a proapoptotic gene in several cancer cells. ATF3 is involved in the stress kinase–mediated signal transduction lead- Introduction ing to caspase protease activation (11). Hartman et al. (12) showed that ATF3 mediates stress-induced apoptosis in h cells. Overex- Phosphatidylinositol 3-kinase (PI3K) signaling is a well-charac- pression of ATF3 enhances the apoptotic effect of curcumin, a terized pathway known to promote cell growth and proliferation polyphenol derived from the turmeric plant, in MDA-1986 cells and suppresses apoptosis in many types of cancer. Akt, a crucial (13). Epicatechin gallate, one of the major catechins in green tea, downstream target of PI3K, is responsible for many of the and indole-3-carbinol induce proapoptotic gene NAG-1 expression biological consequences of PI3K activation (1). A number of mediated by ATF3 in human colorectal cancer cells (14, 15). Topo- studies have shown that deregulated/inappropriate activation of isomerase inhibitor, etoposide- and camptothecin-induced apo- the PI3K/Akt pathway, achieved through the numerous genetic and ptosis, and caspase activity are accelerated by ATF3 overexpression epigenetic alterations, contributes substantially to tumorigenesis. in human epitheloid carcinoma HeLa-S3 cells (16). These studies show that ATF3 transcription factor can mediate cytotoxic agent– induced apoptosis in several model systems. Requests for reprints: Seung Joon Baek, Department of Pathobiology, College of Early growth response 1 (EGR-1, also known as NGFI-A, Zif268, Veterinary Medicine, University of Tennessee, 2407 River Drive, Knoxville, TN 37996. krox24, and TIS8) is another transcription factor family that is Phone: 865-974-8216; Fax: 865-974-5616; E-mail: [email protected]. I2006 American Association for Cancer Research. regulated by diverse stimuli. Expression of EGR-1 and its role in doi:10.1158/0008-5472.CAN-05-1987 cancer are complicated and EGR-1 may either inhibit or stimulate

Cancer Res 2006; 66: (4). February 15, 2006 2376 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2006 American Association for Cancer Research. ATF3 Mediates LY294002-Induced Apoptosis growth depending on the cellular context (17). It has been shown normalized to the pRL-null luciferase activity using a dual luciferase assay kit (Promega). that activation of the prostaglandin E2 (PGE2) receptor (EP4) by Small interfering RNA transfections. ATF3 small interfering RNA PGE2 induces EGR-1 expression and in turn leads to the expression of tumorigenic gene products such as cyclin D1 and PGE synthase (siRNA) was purchased from Santa Cruz Biotechnology. HCT-116 cells were 2 transfected with the ATF3 siRNA at a concentration of 100 nmol/L or with (18). EGR-1 can also induce metastasis-related factors in vitro, such negative control siRNA using TransIT-TKO transfection reagent (Mirus, as the vascular endothelial growth factor receptor Flt-1 and matrix Madison, WI). After transfection for 24 hours, the medium was replaced metalloproteinase (19, 20), indicating that EGR-1 may act as a with serum-free medium containing vehicle or LY294002. The cells were master protein in directing invasion and metastasis during cancer incubated for 24 hours and total protein was subjected to Western blot progression. On the other hand, a number of reports also indicate analysis. that EGR-1 acts as a tumor suppressor gene. EGR-1 is down- DNA microarray experiment. Five micrograms of total RNA were used regulated in several types of neoplasia as well as in an array of as a template for cDNA synthesis and the cDNA was labeled with biotin- tumor cell lines. EGR-1 induces cell growth arrest and apoptosis dUTP using AmpoLabeling-LPR kit (SuperArray Bioscience Co., Frederick, (21–24) and is an important factor involved in neuronal apoptosis MD). The cDNA probe was applied to prehybridized Human cyclic AMP 2+ (25). EGR-1 is induced very early in the apoptotic process where it (cAMP)/Ca Signaling PathwayFinder Gene Array (HS-028) and Human mediates the activation of downstream regulators such as p53 (24). Signal Transduction in Cancer Gene Array (HS-044) membranes. The hybridization was done at 60jC for 12 hours. After washing, the membranes EGR-1 may also activate phosphatase and tensin homologue were blocked and treated with alkaline phosphatase–conjugated streptavi- (PTEN) tumor suppressor gene during UV irradiation (26) and din and finally exposed to CDP-Star alkaline phosphatase chemilumines- suppresses the growth of transformed cells both in soft agar and in cent substrate. The membrane was exposed to X-ray film. The intensity of athymic nude mice (27). Whereas these results indicate that EGR-1 the spots was compared using glyceraldehyde-3-phosphate dehydrogenase plays a significant role in growth suppression, the consequences (GAPDH) as a positive control. of altered EGR-1 expression may be different, depending on cell Adenoviral infections. HCT-116 cells were infected with viral vectors context. This variety may be dependent on expression of other as previously described (32). Briefly, LacZ, PTEN, and dominant-negative members of the EGR-1 binding proteins or on the EGR-1 target Akt (dnAkt) were transduced at 30, 90, and 30 multiplicity of infection, gene usage in different cells. Therefore, EGR-1 may regulate many respectively. After infection for 1 hour, the cells were grown for 24 hours and genes involved in divergent functions. In this report, we have found total cell lysates were harvested for Western blot analysis. Western blot analysis. Cells were washed once with PBS and lysed that ATF3 is a novel target gene of EGR-1 and mediates LY294002- with radioimmunoprecipitation assay (RIPA) buffer (1Â PBS, 1% NP40, 0.5% induced apoptosis in human colorectal cancer cells. These effects sodium deoxycholate, 0.1% SDS) supplemented with protease (1 Ag/mL are caused by a PI3K/Akt–independent pathway. aprotinin, 1 Ag/mL leupeptin, and 1 mmol/L phenylmethylsulfonyl fluoride)

and phosphatase inhibitors (1 mmol/L Na3VO4 and 1 mmol/L NaF). The soluble protein concentrations were determined by bicinchoninic acid Materials and Methods protein assay kit (Pierce Biotechnology, Rockford, IL). Proteins were Cell lines and reagents. Human colorectal cancer cells (HCT-116, separated by SDS-PAGE and transferred onto nitrocellulose membrane SW480, Caco-2, and HT-29 cells) were purchased from American Type (Osmonics, Minnetonka MN). The membranes were incubated first with Culture Collection (Manassas, VA). Inhibitors or antibodies and their primary antibody at 4jC overnight and then with horseradish peroxidase– sources were as follows: LY294002 was purchased from Promega (Madison, conjugated secondary antibody for 1 hour. The signal was detected by WI); wortmannin was from MP Biomedicals (Eschwege, Germany); 1L-6- enhanced chemiluminescence system (Amersham Biosciences, Piscataway, hydroxymethyl-chiro-inositol 2(R)-2-O-methyl-3-O-octadecylcarbonate was NJ) and quantified by Scion Image Software (Scion Corp. Frederick, MD). from Calbiochem (San Diego, CA); anti-ATF3, EGR-1, and actin antibodies Electrophoretic mobility shift assay. Nuclear extracts were prepared were from Santa Cruz Biotechnology (Santa Cruz, CA); anti–poly(ADP- using Nuclear Extract kit (Active Motif, Carlsbad, CA). For the electropho- ribose) polymerase, Akt, and phospho-glycogen synthase kinase 3h anti- retic mobility shift assay (EMSA), double-stranded oligonucleotides bodies were from Cell Signaling Technology (Beverly, MA); and anti-PTEN corresponding to the ATF3 GC-rich sequence (sense, 5V-cgaccgcgccgccagcc- antibody was from NeoMarkers (Fremont, CA). gaccgcgccgccagc-3V;antisense,5V-gctggcggcgcggtcggctggcggcgcggtcg-3V) Construction of plasmids. The luciferase constructs containing the ATF3 were synthesized and end-labeled with [g-32P]ATP by T4 polynucleotide promoter were previously described (28). The putative interleukin-6 response kinase (Promega). Assays were done by incubating 20 Ag of nuclear extracts element-binding protein (IL-6 RE-BP) binding site or GC-rich sequence (GC in binding buffer (Promega) containing 2 Â 105 cpm of labeled probe for box) was deleted using the QuickChange II site-directed mutagenesis kit 20 minutes at room temperature. To ensure the specific binding of EGR-1, (Stratagene, Cedar Creek, TX). To delete these sites on the À84/+34 ATF3 the recombinant EGR-1 proteins, generated by in vitro translation (TNT promoter region, the following primers were used: DIL-6 RE-BP sense, Quick Coupled Transcription/Translation Systems, Promega) or obtained 5V-gtaagcttgcaacacggagtaaacgaccgc-3V; DIL-6 RE-BP antisense, 5V-actccgtgttg- from Sigma (St. Louis, MO), were incubated with the labeled probe and caagcttacttagatcgca-3V; DGC box sense, 5V-gagtaaacgaagcctgagggctataaaagg-3V; 10- or 50-fold molar excesses of unlabeled oligonucleotide. Samples were DGC box antisense, 5V-ccctcaggcttcgtttactccgtgttgcca-3V. The deletions were then electrophoresed on 6% nondenaturing polyacrylamide gels with 0.25Â confirmed by DNA sequencing. Plasmid for ATF3 overexpression experiment, Tris borate/EDTA and gels were dried and subjected to autoradiography. pCG-ATF3, was kindly provided by Dr. Tsonwin Hai (Ohio State University, Immunoprecipitation of EGR-1. HCT-116 cells overexpressed with Columbus, OH). The full-length EGR-1, Sp1, and NGFI-A-binding protein EGR-1 were grown to subconfluency on a 10-cm dish and then treated with (NAB1) cDNA in the pcDNA3 expression vector and the pEBS14 luc construct LY294002 for 24 hours. After washing with ice-cold PBS, the cells were were previously described (29–31). scraped with RIPA buffer containing protease and phosphatase inhibitors Transient transfection and luciferase reporter assays. HCT-116 cells and mixed at 4jC for 15 minutes. The cell suspension was centrifuged and were plated in 12-well plates at 105 per well and were grown for 16 hours. protein concentration was measured for adjusting protein amount. After Plasmid mixtures containing ATF3 promoter linked to luciferase and preclearing with protein A/G PLUS-agarose (Santa Cruz Biotechnology), the pRL-null (Promega) and/or indicated expression vectors were transfected supernatant was incubated with 2 Ag of EGR-1 antibody and protein A/G by LipofectAMINE (Invitrogen, Carlsbad, CA) according to the protocol of PLUS-agarose overnight at 4jC with rotation. After centrifugation, the the manufacturer. After transfection, the medium was replaced with serum- immunoprecipitate was washed four times with PBS and RIPA buffer. free medium and the indicated agent was added. Cells were harvested The pellet was resuspended in sample buffer and subjected to 4% to in luciferase lysis buffer and luciferase activity was determined and 12% gradient SDS-PAGE (Invitrogen) followed by Western blotting. www.aacrjournals.org 2377 Cancer Res 2006; 66: (4). February 15, 2006

Anti–phosphoserine/threonine antibody (BD Biosciences, San Jose, CA) was Results used at a dilution of 1:2,000. Soft agar cloning assay. HCT-116 cells were resuspended at 3,000 cells in LY294002 has antitumorigenic activity and induces ATF3 2 mL of 0.4% agar in McCoy’s 5A medium and plated on top of 1 mL of 0.8% expression. 2-(4-Morpholinyl)-8-phenyl-4H-1-benzopyran-4-one agar in six-well plates. Plates were incubated for 3 weeks at 37jC. Cell colonies (LY294002) has been shown to have antitumorigenic activity in were visualized by staining with p-iodonitrotetrazolium violet (Sigma). human colorectal cancer both in vivo and in vitro (3). We have also Detection of apoptotic cells. The DNA contents for vehicle- and reported that LY294002 causes cell growth arrest and induction of LY294002-treated HCT-116 and HT-29 cells were determined by fluores- apoptosis in HCT-116 cells (4). To confirm the effect of LY294002 cence-activated cell sorting (FACS) as previously described (4). For detection on antitumorigenesis, we did a soft agar cloning assay. As shown of apoptosis by ATF3 overexpression, HCT-116 cells were stained with FITC- in Fig. 1A, a significant reduction in the number of colonies was labeled Annexin V and propidium iodide using Annexin V-FITC apoptosis observed with LY294002 treatment compared with vehicle treat- detection kit (BD Biosciences PharMingen, San Diego, CA) according to the ment, suggesting that LY294002 inhibits an anchorage-independent Â 5 instruction of the manufacturer. Briefly, HCT-116 cells were plated at 3 10 proliferation of HCT-116 cells. Based on our observations and per well in six-well plates and transfected with pCG-ATF3 by FuGENE 6 previous reports, LY294002 has an antitumorigenic activity and (Roche, Indianapolis, IN) for 24 hours. Subsequently, the cells were grown for affects Ca2+ signaling (3, 4, 33). We conducted a gene expres- 24 hours and stained with Annexin V-FITC and propidium iodide. A total of 10,000 cells were examined by flow cytometry using a Beckman Coulter Epixs sion profiling study to identify novel mediators of LY294002 using XL equipped with ADC and ModFit LT software. Cells were gated on side DNA microarray membranes designed for signal transduction in 2+ scatter and forward scatter to exclude debris. Doublets were eliminated using cancer and the cAMP/Ca signaling pathway. Enhanced expres- peak versus integral analysis. Annexin V–positive/propidium iodide–positive sion of several mRNAs was detected in LY294002-treated HCT-116 and Annexin V–positive/propidium iodide–negative cell populations were cells; ATF3, arginine vasopressin, proliferating cell nuclear antigen, determined as apoptotic cells from the total gated cells. and tyrosine hydroxylase with the Human cAMP/Ca2+ Signaling

Figure 1. Antitumorigenic compound LY294002 induces ATF3 expression in several colorectal cancer cell lines. A, HCT-116 cells were grown in 0.4% soft agar for 3 weeks and stained with p-iodonitrotetrazolium. The colonies were counted from multiple, randomly selected fields. Columns, mean from three replicate experiments; bars, SD. B, HCT-116 cells were incubated for 24 hours with LY294002 at the concentration of 50 Amol/L or vehicle. After isolation of RNA, membrane microarray analysis was done using the GEArray Q Series Human cAMP/Ca2+ Signaling PathwayFinder Gene Array. Levels of gene transcripts were normalized to the levels of GAPDH. C to E, HCT-116, SW480, Caco-2, or HT-29 cells were treated with vehicle or LY294002 (10-50 Amol/L). At the indicated times, the cell lysates were harvested for Western blot analysis. Equal loading was confirmed by determining actin immunoreactivity. F, apoptosis was analyzed by FACS using HCT-116 and HT-29 cells treated with LY294002 (50 Amol/L) for 24 hours. Columns, mean fold increase over apoptotic percentage of vehicle-treated cells from three replicate experiments; bars, SD. *, P < 0.05; ***, P < 0.001, versus vehicle treatment (Student’s t test).

Cancer Res 2006; 66: (4). February 15, 2006 2378 www.aacrjournals.org

PathwayFinder Gene Array membrane (Fig. 1B); and h-casein and polymerase expression by 4-fold whereas treatment of ATF3 siRNA– p53 with the Human Signal Transduction in Cancer Gene Array transfected cells with LY294002 increased cleaved poly(ADP-ribose) membrane (data not shown). We focused on ATF3 transcription polymerase expression by 2-fold compared with vehicle treatment. factor because it is one of the genes most strongly induced by These results suggest a linkage between ATF3 expression and LY294002 in HCT-116 cells. The induction of ATF3 transcripts LY294002-induced apoptosis. was confirmed by reverse transcription-PCR, showing more than Evidence for PI3K-independent pathway in LY294002- 2-fold induction (data not shown). LY294002-induced ATF3 expres- induced ATF3 expression. Inhibition of PI3K activity by sion was also confirmed by Western blot analysis. Induction of LY294002 and Akt inactivation is likely responsible for the ATF3 expression was observed at 30 Amol/L LY294002 (Fig. 1C) antitumorigenic and apoptotic action of this compound. However, and was detected in a time-dependent manner (Fig. 1D). ATF3 LY294002 affects signaling pathways other than PI3K (33, 36, 37). To induction by LY294002 (50 Amol/L) was also seen in SW480 and investigate the molecular mechanism by which LY294002 induces Caco-2 cells but not in HT-29 cells (Fig. 1E). Consistent with ATF3 expression in a PI3K-independent manner, we treated HCT- Western blot analysis, LY294002-induced apoptosis in HT-29 cells 116 cells with the PI3K inhibitor wortmannin. As shown in Fig. 3A, was very weak compared with that in HCT-116 cells (Fig. 1F) wortmannin did not affect ATF3 expression, implying indepen- although LY294002 caused G1 cell cycle arrest in HT-29 cells dence of PI3K inhibition in LY294002-induced ATF3 expression. (vehicle: 61.2 F 1.8% versus LY294002: 83.6 F 0.7%; P < 0.001). A significant inhibition of PI3K activity by LY294002 and These results indicate that ATF3 expression may play an important wortmannin was confirmed by dephosphorylation of glycogen role in LY294002-induced apoptosis in several types of human synthase kinase 3h, which is a downstream target of the PI3K/Akt colorectal cancer cells. pathway. In addition, a specific Akt inhibitor, 1L-6-hydroxymethyl- ATF3 contributes to LY294002-induced apoptosis in HCT- chiro-inositol 2(R)-2-O-methyl-3-O-octadecylcarbonate, was used 116 cells. Although there is ample evidence that induction of to assess the contribution of Akt in LY294002-induced ATF3 ATF3 expression correlates with cellular damage and/or apoptosis expression. Treatment with Akt inhibitor did not increase ATF3 (11–14, 16), ATF3 also has antiapoptotic effects in cardiac expression (Fig. 3B). To obtain further evidence, PTEN, which myocytes (34) and superior nerve ganglion neurons (35). These antagonizes PI3K activity (38), or dnAkt was overexpressed using studies indicate that the role of ATF3 may depend on cell context. adenoviral vectors. Consistently, neither PTEN nor dnAkt affected We used Annexin V staining to assess whether the overexpression ATF3 expression compared with LacZ adenovirus control infection of ATF3 results in the induction of apoptosis in human colorectal (Fig. 3C), supporting the conclusion that LY294002 causes ATF3 cancer cells. HCT-116 cells were transfected with ATF3 expression induction in a PI3K/Akt–independent pathway. vector and apoptosis was analyzed. Increased ATF3 expression and EGR-1 mediates LY294002-induced ATF3 expression at the a significant increase in cells that are undergoing apoptosis were transcriptional level. To investigate the effects of LY294002 on observed in ATF3 overexpressed HCT-116 cells (Fig. 2A). To assess transcriptional activity of ATF3, ATF3 promoter constructs contai- the involvement of ATF3 in LY294002-induced apoptosis, we did ning À1850/+34 (pATF3À1850/+34) or À84/+34 (pATF3À84/+34) knockdown of endogenous ATF3 gene by RNA interference and were transfected into HCT-116 cells and luciferase activity was analyzed apoptosis-related protein poly(ADP-ribose) polymerase measured. As shown in Fig. 4A, LY294002 transactivated ATF3 expression. As shown in Fig. 2B, treatment of control siRNA– promoter (pATF3À84/+34) in a concentration-dependent manner, transfected cells with LY294002 increased cleaved poly(ADP-ribose) up to 50 Amol/L. Moreover, a significant increase of reporter

Figure 2. ATF3 in part mediates LY294002- induced apoptosis. A, HCT-116 cells were transfected with pcDNA3 or pCG-ATF3. The cells were grown for 24 hours and subsequently stained with Annexin V-FITC and propidium iodide and analyzed by flow cytometry. Columns, mean from three independent transfections; bars, SD. Significance for the apoptotic cell populations after transfection with pCG-ATF3 was calculated by Student’s t test; **, P < 0.01, versus pcDNA3-transfected cells. B, HCT-116 cells were transfected with control or ATF3 siRNA, followed by LY294002 (50 Amol/L) treatment for 24 hours. The cell lysates were harvested for Western blot analysis. Apoptosis was assessed by cleaved poly(ADP-ribose) polymerase expression. Columns, mean from three independent transfections; bars, SE. ***, P < 0.001, versus control siRNA–transfected cells in the presence of LY294002 (Paired t test).

www.aacrjournals.org 2379 Cancer Res 2006; 66: (4). February 15, 2006

ATF3 promoter at the basal level (data not shown). However, LY294002 treatment of the EGR-1-overexpressing cells enhanced the ATF3 promoter activity compared with vector-transfected cells (16.5-fold versus 6.7-fold, respectively). The dramatic increase of luciferase activity in LY294002-treated EGR-1-overexpressing cells was markedly reduced by deletion of GC-rich sequence on pATF3À84/+34 (Fig. 4D), suggesting that GC-rich sequence is required, at least in part, to mediate EGR-1 action on ATF3 promoter. To confirm the functional influence of EGR-1 on the ATF3 promoter, expression vectors encoding EGR-1 and/or NAB1 cDNAs were transfected with the reporter construct pATF3À84/ +34 into HCT-116 cells. NAB1 has been known to block the biological activity of EGR-1 (39, 40). As shown in Fig. 4E, NAB1 alone did not affect ATF3 promoter activity; however, NAB1 markedly down-regulated the ATF3 promoter activation induced by EGR-1 in the presence of LY294002. Taken together, these data suggest that EGR-1 contributes to LY294002-induced ATF3 transactivation in human colorectal cancer cells. Finally, we have examined whether LY294002 enhances a reporter gene containing four copies of EGR-1 binding sites. As shown in Fig. 4F, LY294002 increases a luciferase activity in pEBS14 luc–transfected cells. This indicates that LY294002 may not only induce ATF3 promoter but also may induce other promoters containing EGR-1 binding sites. EGR-1 binds to ATF3 promoter. To determine whether EGR-1 can interact with the ATF3 promoter, EMSA was done using the Figure 3. ATF3 induction by LY294002 is independent of PI3K/Akt pathway. A, HCT-116 cells were treated with vehicle, wortmannin (0.5 and 1 Amol/L), or oligonucleotides containing two copies of GC-rich sequence on the LY294002 (50 Amol/L) for 24 hours. B, HCT-116 cells were treated with vehicle, ATF3 promoter region. Nuclear extracts were purified from either 1L-6-hydroxymethyl-chiro-inositol 2(R)-2-O-methyl-3-O-octadecylcarbonate, a selective Akt inhibitor (10 Amol/L), or LY294002 (50 Amol/L) for 24 hours. vehicle- or LY294002-treated HCT-116 cells and used for EMSA. C, adenoviral vectors were used to overexpress PTEN, dnAkt, and LacZ First, to identify whether EGR-1 binds to the ATF3 promoter, (control). Twenty-four hours after treatment with the indicated inhibitor or in vitro synthesized EGR-1 protein was incubated with the probe. infection, the cell lysates were harvested to perform Western blot analysis. As shown in Fig. 5A, EGR-1 can interact with the ATF3 promoter and complex formation was diminished by the addition of 10- or activity in both pATF3À1850/+34 and pATF3À84/+34 constructs 50-fold molar excess of cold oligonucleotides. Second, to evaluate was observed in LY294002-treated cells but not in Akt inhibitor– the effect of LY294002 treatment on the binding affinity of EGR-1 treated cells (Fig. 4B), consistent with previous data showing that for its DNA recognition element, nuclear extracts from vehicle- ATF3 is induced by LY294002 but not by Akt inhibitor (Fig. 3B). or LY294002-treated cells were used. The intensities of the Dramatic activation by LY294002 was observed in both ATF3 corresponding EGR-1-probe complex were not different between promoter constructs, indicating that there are positive regulators vehicle- and LY294002-treated cells (Fig. 5A, right) although the within À84/+34 bp of ATF3 promoter. The promoter region within protein complexes were competing out with addition of cold À84 bp was further investigated to find any specific cis-acting oligonucleotide (Fig. 5A, middle). This suggests that LY294002 elements conferring LY294002-induced ATF3 expression. Two may not enhance the EGR-1 binding affinity to the ATF3 different putative binding sites, IL-6 RE-BP and GC-rich sequence, promoter. It has been reported that conformational changes were identified in the À84/+34 promoter region using Transcrip- such as phosphorylation lead to transactivation of the target tion Element Search System (http://www.cbil.upenn.edu/tess/). To gene without increasing its DNA binding affinity (41, 42). To ana- determine the functional role for these sites in LY294002-induced lyze the phosphorylation status of EGR-1 in LY294002-treated ATF3 expression, we generated ATF3 promoter internal deletion HCT-116 cells, EGR-1-overexpressing cells were treated with vehi- clones and the reporter activity in the presence of LY294002 was cle or LY294002 and immunoprecipitated with EGR-1 antibody, compared with the wild-type construct. As shown in Fig. 4C, the followed by Western blotting with antibody against phosphoser- transfection of constructs with deleted putative GC-rich sequence ine/threonine or EGR-1. Initially, the same transfection efficiency results in a significant reduction in the LY294002-induced promoter of EGR-1 was confirmed by Western blot analysis when the cells activity compared with wild-type pATF3À84/+34 construct where- were overexpressed with EGR-1 (data not shown). As shown in as deletion of the putative IL-6 RE-BP site did not affect the Fig. 5B, highly phosphorylated serine/threonine residues on EGR- promoter activity induced by LY294002. Because transcription 1 were detected in the LY294002-treated HCT-116 cells compared factors Sp1 and EGR-1 bind to GC-rich sequence and regulate gene with vehicle-treated cells. These results suggest that EGR-1 expression, we speculate that Sp1 and/or EGR-1 may be involved in phosphorylation in LY294002-treated cells at serine/threonine LY294002-induced ATF3. Sp1 or EGR-1 expression vector was residues may be important for EGR-1-mediated ATF3 expression. cotransfected along with the pATF3À84/+34 reporter vector and Consistent with EMSA data, there is no induction of EGR-1 luciferase activity was measured. Overexpression of EGR-1 in the expression observed in LY294002-treated cells (Fig. 5C). There- absence of LY294002 did not affect ATF3 promoter activity whereas fore, once EGR-1 binds to DNA, the EGR-1 phosphorylation Sp1 overexpression showed f3.1-fold increase compared with may facilitate transcriptional activation of ATF3 in the presence control (pcDNA3), suggesting that Sp1 expression up-regulates of LY294002.

Cancer Res 2006; 66: (4). February 15, 2006 2380 www.aacrjournals.org

Discussion compounds. A synthetic compound, LY294002, was designed as a LY294002 has been widely used as a specific PI3K inhibitor and PI3K inhibitor based on the flavonoid quercetin (43). Wortmannin recognized to have antitumorigenic and proapoptotic properties in was originally isolated from Penicillium wortmannii and subse- several types of cancer cell lines based on the inhibitory effect of quently identified as a specific PI3K inhibitor (44). Unlike wort- PI3K. In this study, we report that ATF3 is up-regulated by mannin, LY294002 has been shown to affect other targets. For LY294002 (Fig. 1) and mediates apoptosis induced by LY294002 in instance, LY294002, but not wortmannin, causes a decrease of peak 2+ human colorectal cancer cells (Fig. 2). Further, LY294002-induced Ca responses to serotonin in rat airway smooth muscle cells (33). ATF3 expression is partially mediated by EGR-1 transcription LY294002 directly binds to the estrogen receptor and subse- factor at the ATF3 promoter region (Fig. 4). Interestingly, ATF3 quently inhibits 17h-estradiol-induced transcriptional activity of expression induced by LY294002 is independent of the PI3K an estrogen-sensitive reporter gene (37). LY294002 inhibits mono- pathway because another PI3K inhibitor (wortmannin) and an Akt cyte chemoattractant protein-1 expression in a PI3K inhibition– inhibitor do not enhance ATF3 expression (Fig. 3) nor do biological independent manner (36). These data support our results, showing inhibitors of the PI3K/Akt pathway. Although both LY294002 and that LY294002 has additional targets besides PI3K in human wortmannin inhibit PI3K activity, they are structurally distinct colorectal cancer cells. The physiologic function of ATF3 is likely

Figure 4. Overexpression of EGR-1 augments LY294002-induced ATF3 promoter activity. A, pATF3À84/+34 (0.5 Ag) was transiently transfected with 0.05 Agof pRL-null vector into HCT-116 cells. Subsequently, the cells were treated with vehicle or LY294002 (10-50 Amol/L). B, pATF3À1850/+34 or pATF3À84/+34 (0.5 Ag each) was transfected and the cells were treated with vehicle, Akt inhibitor (10 Amol/L), or LY294002 (50 Amol/L) for 24 hours. C, pATF3À84/+34 or its internal deletion clones (DIL-6 RE-BP and DGC box; 0.5 Ag each) were transfected and the cells were treated with vehicle or LY294002 (50 Amol/L) for 24 hours. D, pATF3À84/+34 or pATF3À84/+34DGC box (0.2 Ag each) was transfected with pcDNA3 (0.3 Ag), Sp1, or EGR-1 (0.15 Ag each) into HCT-116 cells and the cells were treated with vehicle or LY294002 (50 Amol/L) for 24 hours. E, pATF3À84/+34 (0.2 Ag) were transfected with pcDNA3 (0.3 Ag), EGR-1, or NAB1 (0.15 Ag each) into HCT-116 cells and the cells were treated with vehicle or LY294002 (50 Amol/L) for 24 hours. F, the pEBS14 luc construct was transfected with pRL-null vector into HCT-116 cells. The cells were incubated for 24 hours with vehicle or LY294002 (50 Amol/L). The promoter activities were measured by luciferase activity. Transfection efficiency for luciferase activity was normalized to the Renilla luciferase (pRL-null vector) activity. Columns, mean from three to six independent transfections; bars, SD. Fold induction over vehicle treatment per each transfection. www.aacrjournals.org 2381 Cancer Res 2006; 66: (4). February 15, 2006

promoter region resulted in a significant reduction in LY294002- induced ATF3 promoter activity (Fig. 4C). However, the result of deletion of GC-rich sequence in the ATF3 promoter also implies involvement of additional factors in response to LY294002 although we have shown that EGR-1 plays an important role in LY294002- activated ATF3 promoter. IL-6 RE-BP is not involved in LY294002- induced ATF3 but other cis-acting elements, including activator protein 2a and GATA-1 within À84 bp of ATF3 promoter, have not been considered in this report. Therefore, further experimentation is required to exclude the possibility that other cis-acting elements may be involved in the LY294002-induced ATF3 expression. It has been reported that functional interplay between Sp1 and EGR-1 contributes to either activation or suppression of their target genes because GC-rich sequence overlaps the DNA sequence for binding of Sp1 and EGR-1 (30, 45–47). Indeed, EGR-1 overexpression enhances the LY294002-induced promoter activity (16.5-fold increase) whereas Sp1 expression prevents the induction (2.6-fold increase; Fig. 4D). In addition, cotransfection with both EGR-1 and Sp1 suppressed the promoting effect of EGR-1 on ATF3 promoter (data not show). Thus, Sp1 transcription factor negatively regulates ATF3 promoter in the presence of LY294002 although Sp1 expression can activate the promoter at the basal levels (data not shown). This is consistent with a previous report that Sp1 plays a pivotal role in the basal transcription machinery (29). EGR-1 contains several serine/threonine phosphorylation sites in its sequences. There is conflicting evidence about a role for phosphorylation of EGR-1 in influencing its DNA binding and transcriptional activities (48, 49). In this report, we have determined the EGR-1 phosphorylation status and alteration of its DNA binding affinity by LY294002 because LY294002 does not change EGR-1 protein levels (Fig. 5C). Highly phosphorylated serine/threonine residues of EGR-1 were observed in the LY294002-treated HCT-116 Figure 5. Possible role of EGR-1 phosphorylation for LY294002-induced ATF3 expression. A, gel shift assays were done with in vitro translated (IVT) EGR-1 cells compared with vehicle-treated cells (Fig. 5B). However, protein or nuclear extracts (NE) from either vehicle (V) or LY294002(L)-treated LY294002 treatment did not change the EGR-1 binding affinity to HCT-116 cells for 24 hours. The nuclear extracts (20 Ag) were incubated with a 32P-labeled double-stranded oligonucleotide corresponding to the GC-rich sequence of ATF3 promoter. Competitions were done in the presence of 10- and 50-fold molar excesses of nonradiolabeled probe. The binding reactions were resolved by 6% nondenaturing acrylamide electrophoresis. B, HCT-116 cells were transfected with EGR-1 expression vector and grown to subconfluency. Twenty-four hours after incubation with either vehicle or LY294002 (50 Amol/L), the cell lysates were harvested. The cell extracts were immunoprecipitated (IP) with antibody against EGR-1 and immunoblotted (IB) with antibodies against phosphoserine/threonine and EGR-1. C, HCT-116 cells were treated with LY294002 (50 Amol/L). At the indicated time points, the cell lysates were prepared for Western blot analysis. HCT-116 cell lysates treated with 30 Amol/L of sulindac sulfide were used as a positive control (C; ref. 54). dependent on cellular context (10, 12, 13, 16, 34, 35). We have also suggested that ATF3 plays an important role in mediating apoptotic signals in colorectal cancer cells (14). Along with previous reports, we found that LY294002 does not cause a functional induction of apoptosis in HT-29 cells where ATF3 expression is not induced by LY294002 (Fig. 1F). Furthermore, knockdown of ATF3 gene by siRNA suppressed LY294002-induced apoptosis compared with vehicle (Fig. 2B). These results indicate that ATF3 expression mediates, at least in part, LY294002-induced apoptosis in colorectal cancer. To clarify the molecular mechanism by which LY294002 induces

ATF3 expression, the promoter activity of ATF3 was assessed by Figure 6. Schematic diagram of ATF3 up-regulation by LY294002. In addition to luciferase assay in response to LY294002. We found that LY294002 the inhibitory effect on PI3K, LY294002 up-regulates ATF3 in human colorectal transactivates ATF3 and the proximal promoter region spanning cancer cells. Unlike LY294002, wortmannin does not induce ATF3 expression. EGR-1 in part mediates LY294002-induced ATF3 transactivation. ATF3 and positions À84 to +34 located LY294002 response elements (Fig. 4B). EGR-1 are identified as novel target genes for LY294002 and play an important Interestingly, deletion of GC-rich sequence on the proximal role for the antitumorigenic effect of LY294002 in colorectal cancer cells.

Cancer Res 2006; 66: (4). February 15, 2006 2382 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2006 American Association for Cancer Research. ATF3 Mediates LY294002-Induced Apoptosis the ATF3 promoter as assessed by EMSA (Fig. 5A). Phosphorylation range kinase inhibitor) did (data not shown). It remains to be has been known to create a high-affinity binding site for other elucidated what kinase was activated by LY294002, leading to the proteins and may allow the high-affinity binding of EGR-1 to the phosphorylation of EGR-1. coactivator. In fact, the transcription factor CREB is activated by In conclusion, we have shown that transcription factor ATF3 protein kinase A phosphorylation, and interaction of phosphory- expression is increased by LY294002 in a PI3K-independent lated CREB with coactivators such as CREB-binding protein and manner. Our results suggest that EGR-1 plays an important role p300 modulates cAMP-regulated gene expression (50, 51). Richards in mediating LY294002-induced ATF3 expression (Fig. 6). Up- et al. (41) reported that phosphorylation does not change the DNA regulation of ATF3 may provide a novel explanation for the anti- binding affinity of CREB. As shown in our study, LY294002 alters tumorigenic properties of LY294002 in colorectal cancer cells. EGR-1 phosphorylation and this alteration could play an important role for interaction with coactivators rather than increased DNA binding affinity. Indeed, CREB-binding protein and p300 have been Acknowledgments reported to interact with EGR-1 to modulate gene transcription Received 6/7/2005; revised 11/11/2005; accepted 12/13/2005. (52, 53). Casein kinase II has been suggested to phosphorylate Grant support: NIH grant ES011657. The costs of publication of this article were defrayed in part by the payment of page EGR-1 although its molecular mechanism has not been elucidated charges. This article must therefore be hereby marked advertisement in accordance in detail (49). We examined whether inhibition of casein kinase II with 18 U.S.C. Section 1734 solely to indicate this fact. prevents LY294002-induced ATF3 expression. Casein kinase II We thank Dr. Romaine Ingrid Fernando for supporting adenoviral infection experiment, Dr. Michael M. Fry and Jada Huskey (University of Tennessee) for their inhibitor (4,5,6,7-tetrabromo-2-azabenzimidazole) did not affect critical reading of manuscript, Dr. Tsonwin Hai for providing ATF3 expression vector, LY294002-induced ATF3 expression although staurosporine (a broad and Nancy Neilsen for technical assistance.

References tion factor 3 as a novel contributor to the proapoptotic irradiation-induced signalling. Nat Cell Biol 2001;3: effect of curcumin. Mol Cancer Ther 2005;4:233–41. 1124–8. 1. Vivanco I, Sawyers CL. The phosphatidylinositol 14. Baek SJ, Kim JS, Jackson FR, Eling TE, McEntee MF, 27. Huang RP, Darland T, Okamura D, Mercola D, 3-kinase AKT pathway in human cancer. Nat Rev Lee SH. Epicatechin gallate-induced expression of NAG- Adamson ED. Suppression of v-sis-dependent transfor- Cancer 2002;2:489–501. 1 is associated with growth inhibition and apoptosis in mation by the transcription factor, Egr-1. Oncogene 2. Khaleghpour K, Li Y, Banville D, Yu Z, Shen SH. colon cancer cells. Carcinogenesis 2004;25:2425–32. 1994;9:1367–77. Involvement of the PI 3-kinase signaling pathway in 15. Lee SH, Kim JS, Yamaguchi K, Eling TE, Baek SJ. 28. Cai Y, Zhang C, Nawa T, et al. Homocysteine- progression of colon adenocarcinoma. Carcinogenesis Indole-3-carbinol and 3,3V-diindolylmethane induce ex- responsive ATF3 gene expression in human vascular 2004;25:241–8. pression of NAG-1 in a p53-independent manner. endothelial cells: activation of c-Jun NH(2)-terminal 3. Semba S, Itoh N, Ito M, Harada M, Yamakawa M. Biochem Biophys Res Commun 2005;328:63–9. kinase and promoter response element. Blood 2000;96: The in vitro and in vivo effects of 2-(4-morpholinyl)-8- 16. Mashima T, Udagawa S, Tsuruo T. Involvement of 2140–8. phenyl-chromone (LY294002), a specific inhibitor of transcriptional repressor ATF3 in acceleration of 29. Baek SJ, Horowitz JM, Eling TE. Molecular cloning phosphatidylinositol 3V-kinase, in human colon cancer caspase protease activation during DNA damaging and characterization of human nonsteroidal anti- cells. Clin Cancer Res 2002;8:1957–63. agent-induced apoptosis. J Cell Physiol 2001;188:352–8. inflammatory drug-activated gene promoter. Basal 4. Yamaguchi K, Lee SH, Eling TE, Baek SJ. Identification 17. Thiel G, Cibelli G. Regulation of life and death by the transcription is mediated by Sp1 and Sp3. J Biol Chem of nonsteroidal anti-inflammatory drug-activated gene zinc finger transcription factor Egr-1. J Cell Physiol 2002; 2001;276:33384–92. (NAG-1) as a novel downstream target of phosphatidy- 193:287–92. 30. Baek SJ, Kim JS, Nixon JB, DiAugustine RP, Eling TE. linositol 3-kinase/AKT/GSK-3h pathway. J Biol Chem 18. Fujino H, Xu W, Regan JW. Prostaglandin E2 induced Expression of NAG-1, a transforming growth factor-h 2004;279:49617–23. functional expression of early growth response factor-1 superfamily member, by troglitazone requires the early 5. Wang Q, Wang X, Hernandez A, Hellmich MR, Gatalica by EP4, but not EP2, prostanoid receptors via the growth response gene EGR-1. J Biol Chem 2004;279: Z, Evers BM. Regulation of TRAIL expression by the phosphatidylinositol 3-kinase and extracellular signal- 6883–92. phosphatidylinositol 3-kinase/Akt/GSK-3 pathway in regulated kinases. J Biol Chem 2003;278:12151–6. 31. Baek SJ, Wilson LC, Hsi LC, Eling TE. Troglitazone, human colon cancer cells. J Biol Chem 2002;277: 19. Haas TL, Stitelman D, Davis SJ, Apte SS, Madri JA. a peroxisome proliferator-activated receptor g (PPAR g) 36602–10. Egr-1 mediates extracellular matrix-driven transcription ligand, selectively induces the early growth response-1 6. Wang Q, Li N, Wang X, Kim MM, Evers BM. of membrane type 1 matrix metalloproteinase in gene independently of PPAR g. A novel mechanism Augmentation of sodium butyrate-induced apoptosis endothelium. J Biol Chem 1999;274:22679–85. for its anti-tumorigenic activity. J Biol Chem 2003;278: by phosphatidylinositol 3V-kinase inhibition in the KM20 20. Vidal F, Aragones J, Alfranca A, de Landazuri MO. 5845–53. human colon cancer cell line. Clin Cancer Res 2002;8: Up-regulation of vascular endothelial growth factor 32. Fernando RI, Wimalasena J. Estradiol abrogates 1940–7. receptor Flt-1 after endothelial denudation: role of apoptosis in breast cancer cells through inactivation 7. Latchman DS. Transcription factors as potential transcription factor Egr-1. Blood 2000;95:3387–95. of BAD: Ras-dependent nongenomic pathways requiring targets for therapeutic drugs. Curr Pharm Biotechnol 21. Liu C, Rangnekar VM, Adamson E, Mercola D. signaling through ERK and Akt. Mol Biol Cell 2004;15: 2000;1:57–61. Suppression of growth and transformation and induc- 3266–84. 8. Hai T, Curran T. Cross-family dimerization of tion of apoptosis by EGR-1. Cancer Gene Ther 1998;5: 33. Tolloczko B, Turkewitsch P, Al-Chalabi M, Martin JG. transcription factors Fos/Jun and ATF/CREB alters 3–28. LY-294002 [2-(4-morpholinyl)-8-phenyl-4H -1-benzo- DNA binding specificity. Proc Natl Acad Sci U S A 22. Muthukkumar S, Han SS, Rangnekar VM, Bondada S. pyran-4-one] affects calcium signaling in airway smooth 1991;88:3720–4. Role of Egr-1 gene expression in B cell receptor-induced muscle cells independently of phosphoinositide 3-kinase 9. Hai TW, Liu F, Coukos WJ, Green MR. Transcription apoptosis in an immature B cell lymphoma. J Biol Chem inhibition. J Pharmacol Exp Ther 2004;311:787–93. factor ATF cDNA clones: an extensive family of leucine 1997;272:27987–93. 34. Nobori K, Ito H, Tamamori-Adachi M, et al. ATF3 zipper proteins able to selectively form DNA-binding 23. Muthukkumar S, Nair P, Sells SF, Maddiwar NG, inhibits doxorubicin-induced apoptosis in cardiac heterodimers. Genes Dev 1989;3:2083–90. Jacob RJ, Rangnekar VM. Role of EGR-1 in thapsigargin- myocytes: a novel cardioprotective role of ATF3. J Mol 10. Hai T, Hartman MG. The molecular biology and inducible apoptosis in the melanoma cell line A375–6. Cell Cardiol 2002;34:1387–97. nomenclature of the activating transcription factor/ Mol Cell Biol 1995;15:6262–72. 35. Nakagomi S, Suzuki Y, Namikawa K, Kiryu-Seo S, cAMP responsive element binding family of transcrip- 24. Nair P, Muthukkumar S, Sells SF, Han SS, Sukhatme Kiyama H. Expression of the activating transcription tion factors: activating transcription factor proteins and VP, Rangnekar VM. Early growth response-1-dependent factor 3 prevents c-Jun N-terminal kinase-induced homeostasis. Gene 2001;273:1–11. apoptosis is mediated by p53. J Biol Chem 1997;272: neuronal death by promoting heat shock protein 27 11. Liang G, Wolfgang CD, Chen BP, Chen TH, Hai T. 20131–8. expression and Akt activation. J Neurosci 2003;23:5187–96. ATF3 gene. Genomic organization, promoter, and 25. Pignatelli M, Luna-Medina R, Perez-Rendon A, 36. Choi EK, Park HJ, Ma JS, et al. LY294002 inhibits regulation. J Biol Chem 1996;271:1695–701. Santos A, Perez-Castillo A. The transcription factor monocyte chemoattractant protein-1 expression 12. Hartman MG, Lu D, Kim ML, et al. Role for activating early growth response factor-1 (EGR-1) promotes through a phosphatidylinositol 3-kinase-independent transcription factor 3 in stress-induced h-cell apoptosis. apoptosis of neuroblastoma cells. Biochem J 2003;373: mechanism. FEBS Lett 2004;559:141–4. Mol Cell Biol 2004;24:5721–32. 739–46. 37. Pasapera Limon AM, Herrera-Munoz J, Gutierrez- 13. Yan C, Jamaluddin MS, Aggarwal B, Myers J, Boyd DD. 26. Virolle T, Adamson ED, Baron V, et al. The Egr-1 Sagal R, Ulloa-Aguirre A. The phosphatidylinositol 3- Gene expression profiling identifies activating transcrip- transcription factor directly activates PTEN during kinase inhibitor LY294002 binds the estrogen receptor www.aacrjournals.org 2383 Cancer Res 2006; 66: (4). February 15, 2006

and inhibits 17h-estradiol-induced transcriptional ac- 44. Walker EH, Pacold ME, Perisic O, et al. Structural 49. Jain N, Mahendran R, Philp R, Guy GR, Tan YH, Cao tivity of an estrogen sensitive reporter gene. Mol Cell determinants of phosphoinositide 3-kinase inhibition by X. Casein kinase II associates with Egr-1 and acts as a Endocrinol 2003;200:199–202. wortmannin, LY294002, quercetin, myricetin, and staur- negative modulator of its DNA binding and transcrip- 38. Wu X, Senechal K, Neshat MS, Whang YE, Sawyers osporine. Mol Cell 2000;6:909–19. tion activities in NIH 3T3 cells. J Biol Chem 1996;271: CL. The PTEN/MMAC1 tumor suppressor phosphatase 45. Andrew SD, Capes-Davis A, Delhanty PJ, Marsh DJ, 13530–6. functions as a negative regulator of the phosphoinosi- Mulligan LM, Robinson BG. Transcriptional repression 50. Chrivia JC, Kwok RP, Lamb N, Hagiwara M, tide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A of the RET proto-oncogene by a mitogen activated Montminy MR, Goodman RH. Phosphorylated CREB 1998;95:15587–91. protein kinase-dependent signalling pathway. Gene binds specifically to the nuclear protein CBP. Nature 39. Russo MW, Sevetson BR, Milbrandt J. Identification 2002;298:9–19. 1993;365:855–9. of NAB1, a repressor of NGFI-A- and Krox20-mediated 46. Davis W, Jr., Chen ZJ, Ile KE, Tew KD. Reciprocal 51. Kwok RP, Lundblad JR, Chrivia JC, et al. Nuclear transcription. Proc Natl Acad Sci U S A 1995;92:6873–7. regulation of expression of the human adenosine 5V- protein CBP is a coactivator for the transcription factor 40. Thiel G, Kaufmann K, Magin A, Lietz M, Bach K, triphosphate binding cassette, sub-family A, transporter CREB. Nature 1994;370:223–6. Cramer M. The human transcriptional repressor protein 2 (ABCA2) promoter by the early growth response-1 52. Mouillet JF, Sonnenberg-Hirche C, Yan X, Sadovsky Y. NAB1: expression and biological activity. Biochim (EGR-1) and Sp-family transcription factors. Nucleic p300 regulates the synergy of steroidogenic factor-1 and Biophys Acta 2000;1493:289–301. Acids Res 2003;31:1097–107. early growth response-1 in activating luteinizing hor- 41. Richards JP, Bachinger HP, Goodman RH, Brennan 47. Raychowdhury R, Schafer G, Fleming J, et al. mone-h subunit gene. J Biol Chem 2004;279:7832–9. RG. Analysis of the structural properties of cAMP- Interaction of early growth response protein 1 (Egr-1), 53. Silverman ES, Du J, Williams AJ, Wadgaonkar R, responsive element-binding protein (CREB) and phos- specificity protein 1 (Sp1), and cyclic adenosine 3V5V- Drazen JM, Collins T. cAMP-response-element-binding- phorylated CREB. J Biol Chem 1996;271:13716–23. monophosphate response element binding protein protein-binding protein (CBP) and p300 are transcrip- 42. Yu YL, Chiang YJ, Chen YC, et al. MAPK-mediated (CREB) at a proximal response element is critical for tional co-activators of early growth response factor-1 phosphorylation of GATA-1 promotes Bcl-XL expression gastrin-dependent activation of the chromogranin A (Egr-1). Biochem J 1998;336:183–9. and cell survival. J Biol Chem 2005;280:29533–42. promoter. Mol Endocrinol 2002;16:2802–18. 54. Baek SJ, Kim JS, Moore SM, Lee SH, Martinez J, Eling 43. Vlahos CJ, Matter WF, Hui KY, Brown RF. A specific 48. Huang RP, Adamson ED. The phosphorylated forms TE. Cyclooxygenase inhibitors induce the expression of inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpho- of the transcription factor, Egr-1, bind to DNA more the tumor suppressor gene EGR-1, which results in the linyl)-8-phenyl-4H -1-benzopyran-4-one (LY294002). efficiently than nonphosphorylated. Biochem Biophys up-regulation of NAG-1, an antitumorigenic protein. J Biol Chem 1994;269:5241–8. Res Commun 1994;200:1271–6. Mol Pharmacol 2005;67:356–64.

Cancer Res 2006; 66: (4). February 15, 2006 2384 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2006 American Association for Cancer Research. Activating Transcription Factor 3 and Early Growth Response 1 Are the Novel Targets of LY294002 in a Phosphatidylinositol 3-Kinase−Independent Pathway

Kiyoshi Yamaguchi, Seong-Ho Lee, Jong-Sik Kim, et al.

Cancer Res 2006;66:2376-2384.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/66/4/2376

Cited articles This article cites 53 articles, 30 of which you can access for free at: http://cancerres.aacrjournals.org/content/66/4/2376.full#ref-list-1

Citing articles This article has been cited by 13 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/66/4/2376.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/66/4/2376. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.