Overexpression of 15-Lipoxygenase-1 Induces Growth Arrest through ...

Overexpression of 15-Lipoxygenase-1 Induces Growth Arrest through Phosphorylation of p53 in Human Colorectal Cancer Cells Jong-Sik Kim,1 Seung Joon Baek,2 Frank G. Bottone, Jr.,1 Tina Sali,1 and Thomas E. Eling1 1

Laboratory of Molecular Carcinogenesis, National Institute of Environmental Health Sciences, NIH, Research Triangle Park, North Carolina and 2Department of Pathobiology, College of Veterinary Medicine, University of Tennessee, Knoxville, Tennessee

Abstract

Introduction

To investigate the function of 15-lipoxygenase-1 (15-LOX-1) in human colorectal cancer, we overexpressed 15-LOX-1 in HCT-116 human colorectal cancer cells. Clones expressing the highest levels of 15-LOX-1 displayed reduced viability compared with the HCT-116-Vector control cells. Further, by cell cycle gene array analyses, the cyclin-dependent kinase inhibitor p21 WAF1/CIP1 and MDM2 genes were up-regulated in 15-LOX-1-overexpressing cells. The induction of p21WAF1/CIP1 and MDM2 were linked to activation of p53 by 15-LOX-1, as there was a dramatic induction of phosphorylated p53 (Ser15) in 15-LOX-1-overesxpressing cells. However, the 15-LOX-1 metabolites 13(S)-hydroxyoctadecadienoic acid and 15(S)-hydroxyeicosatetraenoic acid failed to induce phosphorylation of p53 at Ser15, and the 15-LOX-1 inhibitor PD146176 did not inhibit the phosphorylation of p53 at Ser15 in 15-LOX-1-overexpressing cells. Nonetheless, the growth-inhibitory effects of 15-LOX-1 were p53 dependent, as 15-LOX-1 overexpression had no effect on cell growth in p53 ( / ) HCT-116 cells. Finally, treatment of HCT-116-15-LOX-1 cells with different kinase inhibitors suggested that the effects of 15-LOX-1 on p53 phosphorylation and activation were due to effects on DNA-dependent protein kinase. Collectively, these findings suggest a new mechanism to explain the biological activity of 15-LOX-1, where 15-LOX plays a stoichiometric role in activating a DNA-dependent protein kinase – dependent pathway that leads to p53-dependent growth arrest. (Mol Cancer Res 2005;3(9):511 – 7)

Lipoxygenases are important enzymes in lipid metabolism that convert the polyunsaturated fatty acids, arachidonic acid and linoleic acid, to their corresponding metabolites and are classified with respect to their position of oxygenation of arachidonic acid (1). Studies suggest that lipoxygenases have many significant roles in human disease, including cancer, but the exact biological role of lipoxygenases remains unclear (2). Arachidonate 15-lipoxygenase (15-LOX) can be subclassified according to specificity of tissue distribution and enzymatic characteristics into 15-LOX-1 and 15-LOX-2. 15LOX-1 is expressed in reticulocytes, eosinophils, macrophages, tracheobronchial epithelial cells, skin, and colon (3, 4) and converts linoleic acid and arachidonic acid to their metabolites, 13(S)-hydroxyoctadecadienoic acid [13(S)-HODE] and 15(S)hydroxyeicosatetraenoic acid [15(S)-HETE)], respectively (5). On the other hand, 15-LOX-2 expression is detected in prostate, lung, skin, and cornea but not in colon (6), and it converts arachidonic acid to 15(S)-HETE but metabolizes linoleic acid poorly (5). Previously, it was observed that 15-LOX-1 and its metabolites were involved in several types of cancers, including colorectal and prostate cancer. 15-LOX-1 expression in prostate cancer strongly correlates with the degree of malignancy as assessed by Gleason staging (7). Because 15-LOX-1 is expressed preferentially in prostatic adenocarcinomas and 15LOX-2 is primarily expressed in normal prostatic tissue, these two enzymes seem to have opposing roles in prostatic carcinogenesis (8), with increased 13(S)-HODE and reduced 15(S)-HETE production in neoplastic versus normal prostatic tissues. In addition, Kelavkar et al. reported that overexpression of 15-LOX-1 contributed to the prostate cancer progression by regulating insulin-like growth factor-I receptor (9). In contrast to the prostate, 15-LOX-1 is normally expressed in the colon and has been reported to increase with colorectal tumorigenesis. In the normal colon, expression is highly localized to the mucosal epithelium (10), whereas tumors may express 15-LOX-1 in both the neoplastic epithelium and the infiltrating inflammatory cells (11, 12). Neoplastic colonic epithelium treated in vitro with the short-chain fatty acid sodium butyrate can be induced to differentiate and undergo apoptosis associated with up-regulation of 15-LOX-1 (4). This enzyme is uniquely regulated by histone acetylation in colorectal cells with generally greater levels of acetylation in neoplastic versus nonneoplastic tissues (13). Increased 15-LOX-1 expression in tumors and regulation linked to

Received 2/7/05; revised 7/29/05; accepted 8/3/05. Grant support: Intramural Research Program of the NIH, National Institute of Environmental Health Sciences. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Note: The current address for J.-S. Kim is the Department of Biological Science, Andong National University, Andong, Kyungpook, 760-749, South Korea. Requests for reprints: Thomas E. Eling, Laboratory of Molecular Carcinogenesis, National Institute of Environmental Health Sciences, NIH, MD: E4-09, 111 T.W. Alexander Drive, Research Triangle Park, NC 27709. Phone: 919-541-3911; Fax: 919-541-0146. E-mail: [email protected] Copyright D 2005 American Association for Cancer Research. doi:10.1158/1541-7786.MCR-05-0011

Mol Cancer Res 2005;3(9). September 2005

511

512 Kim et al.

histone acetylation suggest a possible role for 15-LOX-1 in tumor development, but expression in normal mucosal epithelium and induction by butyrate suggest a conflicting antitumorigenic activity in the large intestine. This dichotomy has yet to be resolved but may involve contradictions in reported localization and levels of expression of 15-LOX-1 in neoplastic versus normal colorectal tissues. Shureiqi et al. used immunohistochemistry to show greater expression of 15-LOX1 in normal human colorectal tissues compared with tumors, whereas infiltrating inflammatory cells expressed the enzyme (11). In contrast, Western blot analysis indicates that the tumor mass contains higher levels of 15-LOX-1 than normal tissues perhaps due to the contribution of inflammatory cells (10). Thus, a dichotomy exists between the expression of 15-LOX-1 in colorectal cells and its biological function. The higher expression in tumors, as measured by Western analysis, would seem to suggest a protumorigenic activity; however, overexpression of 15-LOX-1 in xenograft tumors reduces growth, indicating an antitumorigenic activity (10). Further experiments are necessary to resolve these issues, but our working hypothesis is that expression of 15-LOX-1 exerts an antitumorigenic effect in the colon but is protumorigenic in the prostate. In this report, we have investigated the molecular mechanism of 15-LOX-1 in human colorectal carcinogenesis. Our results indicate that 15-LOX-1 overexpression in human colorectal cells induces growth arrest by activation and phosphorylation of p53.

Up-Regulation of p21 and MDM2 in HCT-116-15-LOX-1 Cells To decipher the molecular mechanisms responsible for reduced growth and reduction in tumor volume by the 15LOX-1-overexpressing cells, we decided to determine if the expression of 15-LOX-1 altered the expression of genes related to control of cell proliferation. We selected two different membrane-based microarray systems to measure changes in gene expression, one that examines human cell cycle genes and a second that measures human p53 signaling pathway genes. Each microarray membrane contains 96 genes related with either cell cycle or p53 signaling, respectively. From the microarray experiments with the cell cycle gene array, we found two genes up-regulated and one gene down-regulated in HCT-116-15-LOX-1 cells (Table 1). With the p53 signaling pathway gene array experiment, 11 genes were affected by 15-LOX-1 overexpression (Table 2). Interestingly, p21WAF1/CIP1 and MDM2 were up-regulated in both microarray assays (Fig. 2A). These genes are well established to be involved in cell cycle arrest and are regulated by the tumor suppressor gene p53. This result suggests that

Results Overexpression of 15-LOX-1 in HCT-116 Cells Reduces Cell Growth Previously, we constructed HCT-116 human colorectal cells that ectopically express 15-LOX-1 and reported that these cells exhibited a reduced tumor formation in the nude mouse xenograft model (10). Thus, these cells are a suitable model system to investigate the mechanisms responsible for the antitumorigenic activity of 15-LOX-1. Overexpression of 15LOX-1 in the stable cell lines was confirmed by Western blot analysis and reverse transcription-PCR with four different stable cell lines (HCT-116-15-LOX-1-POOL, HCT-116-15-LOX-1-20, HCT-116-15-LOX-1-22, and HCT-116-Vector). As shown in Fig. 1A and B, high expression of 15-LOX-1 protein as well as mRNA was observed in HCT-116-15-LOX-1-20 and HCT-11615-LOX-1-22 stable cells compared with HCT-116-15-LOX-1POOL or vector-transfected cells. Interestingly, higher expression of 15-LOX-1 was detected in HCT-116-15-LOX-1-20 than in HCT-116-15-LOX-1-22 cells. Therefore, we decided to use HCT-116-15-LOX-1-20 cell line for further experiments. Next, we investigated whether overexpression of 15-LOX-1 would alter cell growth rate. The proliferation of these two stable cell lines was measured with the 3-(4,5-dimethylthiazol-2-yl)-5-(3carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt cell viability assay as described in Materials and Methods. HCT-116-15-LOX-1-20 cells grew at a slower rate than the HCT-116-Vector cells (Fig. 1C), consistent with our previous report that overexpression of 15-LOX-1 in nude mice resulted in the inhibition of tumor growth (10).

FIGURE 1. 15-LOX-1 overexpression in the stable cell lines and growth suppression in HCT-116-15-LOX-1-20 cells. A. The stable cell lines [HCT-116-Vector (VEC ), HCT-116-15-LOX-1-POOL, HCT-116-15LOX-1-20, and HCT-116-15-LOX-1-22] were harvested, and cells were prepared for Western blot analysis. 15-LOX-1 antibody was used. Subsequently, the membrane was stripped and probed with actin antibody. B. Total RNAs were extracted from the stable cell lines and reverse transcription-PCR was done with 15-LOX-1 primers and glyceraldehyde-3-phosphate dehydrogenase (GAPDH ) primers as an internal control. C. Cells were seeded in a 96-well plate and cell viability was measured at 24-hour intervals as described in Materials and Methods. Points, mean of four to five experiments; bars, SD. Mol Cancer Res 2005;3(9). September 2005

Growth Arrest by 15-Lipoxygenase and p53

Table 1. Analysis of Changes in Gene Expression in HCT-116-15-LOX-1 Cells Using the Human Cell Cycle Gene Array Gene ID Genes up-regulated (>1.5) 36 69 Genes down-regulated (>1.5) 43

Gene name

Genbank

Description

Fold change

p21 WAF1/CIP1 MDM2

L47233 Z12020

Cyclin-dependent kinase inhibitor 1A Mouse double minute 2

1.9 1.5

chk1

AF016582

CHK1 (check point, Schizosaccharomyces pombe) homologue

1.9

15-LOX-1 may elicit an antitumorigenic activity via the tumor suppressor p53, thereby leading to the increased expression of p21WAF1/CIP1 and MDM2 observed in the 15-LOX-1-expressing cells. To confirm the microarray data, p21 WAF1/CIP1 and MDM2 protein expression was measured by Western blot analysis (Fig. 2B). p21WAF1/CIP1 was up-regulated in 15-LOX1-containing stable cell lines, whereas we could detect induction of cleaved MDM2 protein (60 kDa) not total MDM2 (90 kDa). In addition, we analyzed expression of nonsteroidal antiinflammatory drug – activated gene-1 (NAG-1), a proapoptotic gene belonging to the transforming growth factor-h superfamily and one of p53 the downstream genes (14). Up-regulation of NAG-1 was also detected in the 15-LOX-1 cells (Fig. 2B). To determine if the inhibition of cell proliferation observed in 15-LOX-1-overexpressing cells required p53, HCT-116 p53 ( / ) cells were transfected with vector or 15-LOX-1 and stable pools of cells were selected for cell proliferation assays. Inhibition of cell proliferation by 15-LOX-1 was abolished in the p53 knockout HCT-116 cells (Fig. 2C). Expression and Phosphorylation of p53 in 15-LOX-1Overexpressing Cells The colorectal cancer cell line HCT-116 contains wild-type p53, which suggests that the up-regulation of p21WAF1/CIP1 and NAG-1 is mediated by activation of p53. It has been known that the phosphorylation of p53 at Ser15 position or increased expression of p53 protein plays a pivotal role in p53 activation (15). The expression and phosphorylation (Ser15) of p53 was then measured in the four cell lines. Interestingly, the phosphorylation of p53 protein was dramatically increased in the HCT-116-15-LOX-1 cells (Fig. 3), whereas total p53 was

not changed. These results indicate that the growth arrest induced by overexpression of 15-LOX-1 in HCT-116 cells is mediated through a p53-dependent pathway via an induced p53 phosphorylation at the Ser15 position. p53 Phosphorylation Is Not Dependent on Metabolites of 15-LOX 15-LOX-1 converts the cis-unsaturated fatty acids linoleic acid and arachidonic acid to the metabolites 13(S)-HODE and 15(S)-HETE, respectively. Previously, we reported 13(S)HODE and 15(S)-HETE production in this 15-LOX-1 cell line (16). To determine whether the metabolites formed by 15-LOX-1 will affect the p53 status, we incubated the wild-type HCT-116 cells with 30 Amol/L of either 13(S)-HODE or 15(S)HETE for 3 and 24 hours, and the cell lysates were analyzed by Western blotting for total p53 and phosphorylated p53. As shown in Fig. 4A, 13(S)-HODE and 15(S)-HETE do not affect p53 expression or its phosphorylation. In addition, there was no change of phosphorylated p53 after treatment of linoleic acid in the HCT-116-15-LOX-1-20 cells (data not shown). These results imply that expression and phosphorylation of p53 in the HCT-116-15-LOX-1 cells is caused by 15-LOX-1 itself rather than the metabolites. To confirm these findings, vector or 15-LOX-1 (HCT-116-15-LOX-1-20) cells were incubated in the presence of vehicle or the 15-LOX-1 specific inhibitor PD146176 at 1 and 10 Amol/L concentrations followed by protein isolation and Western blot analysis. The increase in phosphorylation of p53 in the 15-LOX-1-expressing cells remained dramatically increased even in the presence of the 15-LOX-1 inhibitor, whereas total p53 remained unchanged, further indicating that phosphorylation of p53 by 15-LOX-1 occurred independent of 15-LOX-1 metabolites (Fig. 4B).

Table 2. Analysis of Changes in Gene Expression in HCT-116-15-LOX-1 Cells Using the Human p53 Signaling Pathway Gene Array Gene ID Genes up-regulated (>1.5) 21 73 12 54 14 51 67 Genes down-regulated (>1.5) 47 78 38 63

Gene name

Genbank

Description

Fold change

p21 WAF1 MASPIN PUMA/BBC3 RTP BRAP MDM2 DNA-PKcs

L47233 NM_002639 AF332558 NM_006096 NM_006768 Z12020 NM_006904

Cyclin-dependent kinase inhibitor 1A Serine proteinase inhibitor, clade B Bcl-2-binding component 3 N-myc downstream-regulated gene 1 BRCA1-associated protein Mouse double minute 2 Protein kinase, DNA activated, catalytic

4.4 4.4 4.0 1.9 1.5 1.5 1.5

MKK4 TBP FAF1 PKCa

L36870 M55654 NM_007051 NM_002737

MAPK kinase 4 TATA box – binding protein Fas-associated factor Protein kinase Ca

2.5 1.9 1.7 1.7


513

514 Kim et al.

FIGURE 2. The p53 pathway is involved in the inhibition of cell proliferation observed in 15-LOX-1-overexpressing cells. A. Total RNAs from the stable cells were isolated and then used to probe the Superarray membrane according to the manufacturer’s protocol. Cyclin-dependent kinase inhibitor p21 WAF1/CIP1 and MDM2 genes were induced in 15-LOX-1-overexpressing cell lines in both membrane-based microarrays (Human Cell Cycle Gene Array and Human p53 Signaling Pathway Gene Array, Superarray). B. Proteins from two cell lines (HCT-116-Vector and HCT-116-15-LOX-1-20) were prepared and each cell lysate (30 Ag) was subject to Western blot analysis. p21WAF1/CIP1, MDM2, NAG-1, and actin antibodies were used. C. Cells were seeded in 96-well plates and cell viability was measured at 24-hour intervals as described in Materials and Methods. Points, mean (n = 8) of a representative experiment; bars, SD.

DNA-Dependent Protein Kinase Is Responsible for Phosphorylation of p53 To determine which kinase pathway is involved in p53 phosphorylation, five different kinase inhibitors were incubated with the HCT-116-15-LOX-1 cells and p53 phosphorylation was measured by Western analysis. Mitogen-activated protein kinase (MAPK) inhibitors (SB203580, PD98059, and SP600125) and DNA damage – related kinase inhibitors (wortmannin and caffeine) were used to block MAPK pathways (p38, extracellular signal-regulated kinase, and c-Jun NH2-terminal kinase) and DNA damage – related kinase pathways [DNA-dependent protein kinase (DNA-PK), ataxia telangiectasia mutated, and ataxia telangiectasia mutated and Rad3 related]. Both the p38 kinase inhibitor SB203580 and wortmannin seemed to reduce the expression of phosphorylated p53 (Fig. 5A) compared with the vehicle-treated 15-LOX-1 cells. To further evaluate which MAPK pathway was involved, Western blot analysis for total p38 MAPK, phosphorylated p38 MAPK, and DNA-PKcs was done. Whereas no change in either total or phosphorylated p38 MAPK was observed, significant up-regulation of DNA-PKcs in HCT-116-15-LOX-1 cells was observed (Fig. 5B). Interestingly, DNA-PKcs was also identified as an induced gene in HCT-116-

15-LOX-1 cells by microarray experiment (Table 2), a further indication of the involvement of this pathway. Taken together, these results suggest that the mechanism of p53 phosphorylation in 15-LOX-1-overexpressing cells may, in part, be dependent on DNA-PK.

Discussion Although the expression of 15-LOX-1 has an antitumorigenic effect in human colorectal cancer (17, 18), the mechanisms remain unclear. In this article, we report that the expression of

FIGURE 3. Activation and phosphorylation of tumor suppressor gene p53 . Proteins from four cell lines (HCT-116-Vector, HCT-116-15-LOX-1POOL, HCT-116-15-LOX-1-20, and HCT-116-15-LOX-1-22) were prepared and each cell lysate (30 Ag) was subject to Western blot analysis. p53 and phosphorylated p53 (Ser15) antibodies were used. Mol Cancer Res 2005;3(9). September 2005


FIGURE 4. Phosphorylation of p53 by 15-LOX-1 overexpression is independent of 15-LOX-1 metabolites. A. HCT-116 cells were treated with vehicle (0.1% ethanol), 30 Amol/L 13(S)-HODE, or 30 Amol/L 15(S)HETE for 3 or 24 hours. After incubation, cells were harvested, and cell lysates were prepared for Western blot analysis. p53, phosphorylated p53 (Ser15), and actin antibodies were used. B. Vector or 15-LOX-1 HCT-116-15-LOX-1-20 stable cells were treated in serum-free medium for 24 hours in the presence of vehicle (lanes 1 and 4 ), 1 Amol/L (lanes 2 and 5 ), or 10 Amol/L (lanes 3 and 6) of the 15-LOX-1 inhibitor PD146176 followed by protein isolation and Western blot analysis for total p53, phosphorylated p53 (Ser15), and actin.

15-LOX-1 inhibits cell growth by DNA-PKcs induction followed by the phosphorylation of p53. We also extended our studies to delineate the molecular mechanisms of growth arrest by 15-LOX-1-overexpressing cells. Both p21WAF1/CIP1 and MDM2 were up-regulated in 15-LOX-1-overexpressing HCT-116 cells as determined by both microarray assays. The cyclin-dependent kinase inhibitor p21WAF1/CIP1 has a growth arrest activity and is regulated by both p53-dependent and p53-independent pathways (19). In contrast, MDM2 is another important target gene of p53 as well as a negative regulator of p53. MDM2 binds to p53 and inhibits the accumulation of p53 by ubiquitination and proteasome-dependent degradation (20, 21). Shieh et al. showed that DNA damage induced phosphorylation of p53 at Ser15 and this phosphorylation led to reduced interaction of p53 with MDM2 (22). By our experiments, the p53 protein level and phosphorylation of p53 at Ser15 were highly up-regulated in HCT-116-15-LOX-1 cells compared with HCT-116-Vector cells. This indicates that p53 is activated and functional in HCT-116-15-LOX-1 cells. To further confirm the activation of p53 in 15-LOX-1overexpressing cells, we also measured the expression of NAG-1, which is another target gene of p53 (23). NAG-1 was also up-regulated in 15-LOX-1-overexpressing cells. Taken together, the induction of p21 and NAG-1 expression by 15-LOX-1 may provide the evidence that HCT-116-15-LOX-1 cell showed cell growth retardation compared with HCT-116Vector cells (Fig. 1C). 15-LOX-1 produces 13(S)-HODE and 15(S)-HETE from linoleic acid and arachidonic acid, respectively. In this report, neither metabolite influenced p53 expression or the phosphorylation level of p53. Previous studies with these cells confirmed the metabolism of unsaturated fatty acids, and extensive investigation with a range of concentrations of Mol Cancer Res 2005;3(9). September 2005

metabolites and under several experimental conditions failed to enhance p53 expression or phosphorylation of p53. This result implies that the 15-LOX-1 enzyme can directly activate and phosphorylate p53 independent of its catalytic activity. The phosphorylation of p53 at Ser15 has been reported to occur via several different protein kinases, such as ataxia telangiectasia mutated (24), ataxia telangiectasia mutated and Rad3 related (25), DNA-PK (22), and MAPKs (26). To determine which kinase or kinases are involved in the 15LOX-1-dependent phosphorylation of p53 at Ser15 in this study, we tested five different kinase inhibitors for the inhibition of p53 in 15-LOX-1-overexpressing cells. The results suggest that DNA-PK is a likely candidate responsible for phosphorylation. The DNA-PK is serine/threonine kinase and consists of three components, a large catalytic subunit (DNA-PKcs) and the heterodimeric regulatory subunits (Ku70-Ku80; ref. 27). Thus, DNA-PK is likely responsible for p53 phosphorylation in HCT-116-15-LOX-1 cells. This is supported by Xu et al. who reported that nitric oxide upregulated the expression of DNA-PKcs and increased expression of DNA-PKcs genes correlated with an increase in its activity (28) and Shintani et al. who reported that radiation up-regulates DNK-PKcs expression (29). However, further studies are required to determine the precise molecular mechanisms responsible for the increase of DNA-PK by 15LOX-1 observed in this study.

FIGURE 5. DNA-PK is a potential kinase involved in p53 phosphorylation in HCT-116-15-LOX cells. A. HCT-116-15-LOX-1-20 cells were treated with vehicle (0.1% DMSO), PD98059 (20 Amol/L), SB203580 (20 Amol/L), SP600125 (20 Amol/L), caffeine (4 mmol/L), and wortmannin (10 Amol/L) for 24 hours. After incubation, cells were collected and lysates were prepared as described above and then subjected to Western blot analysis for p53 and phosphorylated p53 (Ser15). B. Proteins from three different cell lines were prepared and each (30 Ag) was used for Western blot analysis. p38, phosphorylated p38, DNA-PKcs, and actin antibodies were used.

515

516 Kim et al.

Cell Viability Assay Cell proliferation was measured using the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2Htetrazolium, inner salt colorimetric assay by Promega (Madison, WI), which estimates the number of viable cells in proliferation. Briefly, 2,000 cells were plated in a final volume of 0.1 mL complete medium in 96-well tissue culture dishes (day 0). Cell viability was measured at 490 nm in an ELISA plate reader following the addition of 0.02 mL 3-(4,5-dimethylthiazol-2-yl)5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt Aqueous One solution per well and a 1-hour incubation at 37jC/5% CO2. Each experiment was carried out in quadruplicate and repeated twice. Percent viability is calculated relative to vehicle treated controls using the mean F SD A 490 nm.

FIGURE 6. Proposed model for growth arrest by 15-LOX-1 overexpression. The 15-LOX-1 protein rather than its metabolites can induce p53 expression and phosphorylation via DNA-PK. The activated p53 induces growth arrest by triggering expression of genes involved in growth arrest, such as p21, NAG-1, or other p53 target genes.

In this report, we provide a novel mechanism, which has not been reported previously, that 15-LOX-1 affects DNA-PK to phosphorylate p53. Thus, 15-LOX-1 induces p53 expression and phosphorylation of p53 via DNA-PK. The activated p53 induces growth arrest by triggering gene expressions involved in growth arrest, such as p21 and NAG-1 (Fig. 6), a gene well characterized as downstream of p53 following a variety of treatments. These results will provide a new mechanism to explain the biological activity of 15-LOX-1 in human colorectal cells.

Materials and Methods Cell Lines and Reagents Human colorectal carcinoma cells, HCT-116, were purchased from American Type Culture Collection (Manassas, VA) and maintained in McCoy’s 5A medium supplemented with 10% fetal bovine serum and penicillin-streptomycin. The construction of the vector and 15-LOX-1 stable cell lines were described previously by this laboratory (16) and were cultured in McCoy’s 5A medium supplemented with 10% fetal bovine serum, penicillin-streptomycin, and zeocin (Invitrogen, Carlsbad, CA). HCT-116 p53 ( / ) cells were kindly provided by Dr. Bert Vogelstein (Johns Hopkins University, Baltimore, MD) and transfected with vector or 15-LOX-1 followed by isolation of stable pools of cells in the presence of zeocin as described above. 13(S)-HODE and 15(S)-HETE were purchased from Cayman Chemical (Ann Arbor, MI). Wortmannin and caffeine were obtained from Sigma Chemical Co. (St. Louis, MO), whereas MAPK inhibitors (SB203580, PD98059, and SP600125) were purchased from AG Scientific (San Diego, CA). Antibodies of p53, MDM2, p21WAF1/CIP1, p38, DNA-PKcs, and actin were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), whereas phosphorylated p53 and phosphorylated p38 antibodies were obtained from Cell Signaling (Beverly, MA) and NAG-1 antibody was described previously (14). The 15-LOX-1 inhibitor PD146176 was from Sigma Chemical and a stock of 10 mmol/L was dissolved in DMSO.

RNA Isolation and Reverse Transcription-PCR Total cellular RNAs were extracted from stable cell lines using TRIzol reagent (Life Technologies, Gaithersburg, MD). Total RNA (10 Ag) was reverse transcribed and a one-tenth volume of synthesized cDNA then added to 50 AL PCR mixture with 15-LOX-1 gene-specific primers (5V-GGCAAGGAGACAGAACTCAA-3V and 5V-TCCTTCCAGCAAGTCAGAAC-3V or glyceraldehyde-3-phosphate dehydrogenase primers (5V-TCAACGGATTTGGTCGTATT-3V and 5V-CTGTGGTCATGAGTCCTTCC-3V). The thermal cycling conditions used consisted of initial denaturation at 94jC for 4 minutes followed by 25 cycles of 94jC for 30 seconds, 58jC for 30 seconds, and 72jC for 45 seconds and final extension at 72jC for 10 minutes. The final PCR products were electrophoresed on a 1% agarose gel and photographed under UV light. Western Blot Analysis Cells lysates were isolated using radioimmunoprecipitation assay buffer [50 mmol/L Tris-HCl (pH 8.0), 150 mmol/L NaCl, 1% NP40, 0.1% SDS, 0.5% sodium deoxycholate] containing protease inhibitors (Sigma Chemical). Whole-cell extracts (30 Ag) were separated in a 4% to 12% gel and transferred onto a nitrocellulose membrane. Membranes were blocked in TTBS and 5% skim milk. Protein bands were probed with primary antibody followed by labeling with horseradish peroxidase – conjugated anti-mouse, anti-rabbit, or anti-goat secondary antibody. Bands were visualized by enhanced chemiluminescence using enhanced chemiluminescence kit (Amersham, Piscataway, NJ) according to the manufacturer’s instruction. Membrane Microarray Experiment The procedure for biotinylated cDNA probe synthesis was done using the Ampho-LPR Labeling kit (Superarray, Inc., Frederick, MD). Briefly, total RNA (5 Ag) was used as a template for cDNA synthesis and the cDNA was labeled with biotin-dUTP (Roche, Indianapolis, IN) during PCR. Then, the reaction was stopped and denatured at 94jC for 2 minutes and the resulting DNA probe was applied to a prehybridized GEArray membrane. The hybridization was done at 60jC for 12 hours in a hybridization oven. After two-step washing at Mol Cancer Res 2005;3(9). September 2005


60jC, the membrane was blocked and treated with alkaline phosphatase – conjugated streptavidin and finally exposed to CDP-Star alkaline phosphatase chemiluminescent substrate. The membrane was exposed to X-ray film. The intensity of the each spot was compared with Scion Image software using glyceraldehyde-3-phosphate dehydrogenase or cyclophilin A as a positive control.

Acknowledgments We thank Drs. John Roberts and Alex Merrick (National Institute of Environmental Health Sciences) for their helpful suggestions and comments.

References

12. Ikawa H, Kamitani H, Calvo BF, Foley JF, Eling TE. Expression of 15-lipoxygenase-1 in human colorectal cancer. Cancer Res 1999;59:360 – 6. 13. Kamitani H, Taniura S, Ikawa H, Watanabe T, Kelavkar UP, Eling TE. Expression of 15-lipoxygenase-1 is regulated by histone acetylation in human colorectal carcinoma. Carcinogenesis 2001;22:187 – 91. 14. Baek SJ, Kim KS, Nixon JB, Wilson LC, Eling TE. Cyclooxygenase inhibitors regulate the expression of a TGF-h superfamily member that has proapoptotic and antitumorigenic activities. Mol Pharmacol 2001;59: 901 – 8. 15. Dumaz N, Meek DW. Serine15 phosphorylation stimulates p53 transactivation but does not directly influence interaction with HDM2. EMBO J 1999; 18:7002 – 10. 16. Hsi LC, Wilson L, Nixon J, Eling TE. 15-Lipoxygenase-1 metabolites downregulate peroxisome proliferator-activated receptor g via the MAPK signaling pathway. J Biol Chem 2001;276:34545 – 52.

1. Yamamoto S. Mammalian lipoxygenases: molecular structures and functions. Biochim Biophys Acta 1992;1128:117 – 31.

17. Shureiqi I, Chen D, Lotan R, et al. 15-Lipoxygenase-1 mediates nonsteroidal anti-inflammatory drug-induced apoptosis independently of cyclooxygenase-2 in colon cancer cells. Cancer Res 2000;60:6846 – 50.

2. Kuhn H, Walther M, Kuban RJ. Mammalian arachidonate 15-lipoxygenases structure, function, and biological implications. Prostaglandins Other Lipid Mediat 2002;68 – 9:263 – 90.

18. Wu J, Xia HH, Tu SP, et al. 15-Lipoxygenase-1 mediates cyclooxygenase-2 inhibitor-induced apoptosis in gastric cancer. Carcinogenesis 2003;24:243 – 7.

3. Funk CD. The molecular biology of mammalian lipoxygenases and the quest for eicosanoid functions using lipoxygenase-deficient mice. Biochim Biophys Acta 1996;1304:65 – 84. 4. Kamitani H, Geller M, Eling T. Expression of 15-lipoxygenase by human colorectal carcinoma Caco-2 cells during apoptosis and cell differentiation. J Biol Chem 1998;273:21569 – 77.

19. Gartel AL, Tyner AL. Transcriptional regulation of the p21(WAF1/CIP1) gene. Exp Cell Res 1999;246:280 – 9. 20. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature 1997;387:296 – 9. 21. Kubbutat MH, Jones SN, Vousden KH. Regulation of p53 stability by Mdm2. Nature 1997;387:299 – 303.

5. Brash AR, Boeglin WE, Chang MS. Discovery of a second 15S-lipoxygenase in humans. Proc Natl Acad Sci U S A 1997;94:6148 – 52.

22. Shieh SY, Ikeda M, Taya Y, Prives C. DNA damage-induced phosphorylation of p53 alleviates inhibition by MDM2. Cell 1997;91:325 – 34.

6. Shappell SB, Boeglin WE, Olson SJ, Kasper S, Brash AR. 15-Lipoxygenase-2 (15-LOX-2) is expressed in benign prostatic epithelium and reduced in prostate adenocarcinoma. Am J Pathol 1999;155:235 – 45.

23. Baek SJ, Wilson LC, Eling TE. Resveratrol enhances the expression of nonsteroidal anti-inflammatory drug-activated gene (NAG-1) by increasing the expression of p53. Carcinogenesis 2002;23:425 – 34. 24. Canman CE, Lim DS, Cimprich KA, et al. Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science 1998; 281:1677 – 9.

7. Kelavkar UP, Cohen C, Kamitani H, Eling TE, Badr KF. Concordant induction of 15-lipoxygenase-1 and mutant p53 expression in human prostate adenocarcinoma: correlation with Gleason staging. Carcinogenesis 2000;21: 1777 – 87. 8. Hsi LC, Wilson LC, Eling TE. Opposing effects of 15-lipoxygenase-1 and -2 metabolites on MAPK signaling in prostate. Alteration in peroxisome proliferatoractivated receptor g. J Biol Chem 2002;277:40549 – 56. 9. Kelavkar UP, Cohen C. 15-Lipoxygenase-1 expression upregulates and activates insulin-like growth factor-1 receptor in prostate cancer cells. Neoplasia 2004;6:41 – 52.

25. Tibbetts RS, Brumbaugh KM, Williams JM, et al. A role for ATR in the DNA damage-induced phosphorylation of p53. Genes Dev 1999;13:152 – 7. 26. She QB, Bode AM, Ma WY, Chen NY, Dong Z. Resveratrol-induced activation of p53 and apoptosis is mediated by extracellular-signal-regulated protein kinases and p38 kinase. Cancer Res 2001;61:1604 – 10. 27. Smith GC, Jackson SP. The DNA-dependent protein kinase. Genes Dev 1999; 13:916 – 34.

10. Nixon JB, Kim KS, Lamb PW, Bottone FG, Eling TE. 15-Lipoxygenase-1 has anti-tumorigenic effects in colorectal cancer. Prostaglandins Leukot Essent Fatty Acids 2004;70:7 – 15.

28. Xu W, Liu L, Smith GC, Charles G. Nitric oxide upregulates expression of DNA-PKcs to protect cells from DNA-damaging anti-tumour agents. Nat Cell Biol 2000;2:339 – 45.

11. Shureiqi I, Wojno KJ, Poore JA, et al. Decreased 13-S-hydroxyoctadecadienoic acid levels and 15-lipoxygenase-1 expression in human colon cancers. Carcinogenesis 1999;20:1985 – 95.

29. Shintani S, Mihara M, Li C, et al. Up-regulation of DNA-dependent protein kinase correlates with radiation resistance in oral squamous cell carcinoma. Cancer Sci 2003;94:894 – 900.


517