Caffeine Inhibits Adenosine-Induced Accumulation of Hypoxia- Inducible Factor-1 , Vascular Endothelial Growth Factor, and Interleukin-8 Expression in Hypoxic Human Colon Cancer Cells

Frequent coffee consumption has been associated with a reduced risk of colorectal cancer in a number of case-control studies. Coffee is a leading source of methylxanthines, such as caffeine. The induction of vascular endothelial growth factor (VEGF) and interleukin-8 (IL-8) is an essential feature of tumor angiogenesis, and the hypoxia-inducible factor-1 (HIF-1) transcription factor is known to be a key regulator of this process. In this study, we investigated the effects of caffeine on HIF-1 protein accumulation and on VEGF and IL-8 expression in the human colon cancer cell line HT29 under hypoxic conditions. Our results show that caffeine significantly inhibits adenosineinduced HIF-1 protein accumulation in cancer cells. We show that HIF-1 and VEGF are increased through A3 adenosine receptor stimulation, whereas the effects on IL-8 are mediated via the A2B subtype. Pretreatment of cells with caffeine significantly reduces adenosine-induced VEGF promoter activity and VEGF and IL-8 expression. The mechanism of caffeine seems to involve the inhibition of the extracellular signal-regulated kinase 1/2 (ERK1/2), p38, and Akt, leading to a marked decrease in adenosine-induced HIF-1 accumulation, VEGF transcriptional activation, and VEGF and IL-8 protein accumulation. From a functional perspective, we observe that caffeine also significantly inhibits the A3 receptor-stimulated cell migration of colon cancer cells. Conditioned media prepared from colon cells treated with an adenosine analog increased human umbilical vein endothelial cell migration. These data provide evidence that adenosine could modulate the migration of colon cancer cells by an HIF-1 /VEGF/IL-8-dependent mechanism and that caffeine has the potential to inhibit colon cancer cell growth. Coffee and tea are the most commonly consumed beverages in the world (Fredholm, 1999). Results of epidemiological studies have not resolved whether coffee consumption is related to colorectal cancer risk. A report by the World Cancer Research Fund concluded that the available evidence was not sufficient to draw any firm conclusions about a decreased risk of colorectal cancer associated with coffee consumption (World Cancer Research Fund/American Institute for Cancer Research, 1997). However, some researchers contend that a link between high consumption of coffee and a low incidence Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. doi:10.1124/mol.106.032920. ABBREVIATIONS: IL-8, interleukin-8; Cl-IB-MECA, N(3-iodobenzyl)2-chloroadenosine-5 N-methyluronamide; DPCPX, 1,3-dipropyl-8-cyclopentylxanthine; MRE 2029F20, N-benzo[1,3]dioxol-5-yl-2-[5-(2,6-dioxo-1,3-dipropyl-2,3,6,7-tetrahydro-1H-purin-8-yl)-1-methyl-1H-pyrazol-3-yloxy]-acetamide; MRE 3008F20, 5N-(4-methoxyphenyl-carbamoyl)amino-8-propyl-2-(2-furyl)-pyrazolo-[4,3e]1,2,4-triazolo[1,5c] pyrimidine; HIF-1, hypoxia-inducible factor-1; HUVEC, human umbilical vein endothelial cell; MAP, mitogen-activated protein; MAPK, mitogen-activated protein kinase; MEK, mitogen-activated protein kinase kinase; siRNA, small interfering RNA; siRNAA2B, small interfering RNA that targets A2B receptor mRNA; siRNAA3, small interfering RNA that targets A3 receptor mRNA; VEGF, vascular endothelial growth factor; ZM 241385, 4-[2-[7amino-2-[furyl][1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl]phenol; ERK1/2, extracellular signal-regulated kinase 1/2; Ab, antibody; U0126, 1,4-diamino-2,3-dicyano-1,4-bis(methylthio)butadiene; PBS, phosphate-buffered saline; MTS, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; NECA, 5 -N-ethylcarboxamidoadenosine; DMSO, dimethyl sulfoxide; SB202190, 4-(4-fluorophenyl)-2-(4-hydroxyphenyl)-5-(4pyridyl)-1H-imidazole; SH5, D-3-deoxy-2-O-methyl-myo inositol 1-[(R)-2-methoxy-3-(octadecyloxy)propyl hydrogen phosphate]; B64, 4-[3-(2-furan-2yl-8-alkyl-8H-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-5-yl)ureido]benzenesulfonic acid derivative. 0026-895X/07/7202-395–406$20.00 MOLECULAR PHARMACOLOGY Vol. 72, No. 2 Copyright © 2007 The American Society for Pharmacology and Experimental Therapeutics 32920/3230203 Mol Pharmacol 72:395–406, 2007 Printed in U.S.A. 395 at A PE T Jornals on Sptem er 8, 2017 m oharm .aspeurnals.org D ow nladed from of colorectal cancer has been firmly established (Ekbom, 1999; Woolcott et al., 2002). Coffee is a leading source of methylxanthines, such as caffeine. A cup of coffee contains approximately 100 mg of caffeine (Fredholm, 1999); thus, caffeine can be found in micromolar concentrations in the human circulation as a result of dietary intake or pharmacological use. Most solid tumors develop regions of low oxygen tension because of an imbalance in oxygen supply and consumption. Clinical and experimental evidence suggests that tumor hypoxia is associated with a more aggressive phenotype (Hockel and Vaupel, 2001). Hypoxic tumor cells are resistant to conventional chemotherapy and radiotherapy. It is therefore rational to target the hypoxic regions of tumors or disrupt events initiated by hypoxia (Melillo, 2004). Interleukin-8 (IL-8), originally discovered as a chemotactic factor for leukocytes, has been shown recently to contribute to human cancer progression through its potential functions as a mitogenic, angiogenic, and motogenic factor (Xie, 2001). Although it is constitutively detected in human cancer tissues and established cell lines, IL-8 expression is regulated by various tumor microenvironment factors, such as hypoxia, acidosis, nitric oxide, and cell density. Furthermore, hypoxia is a potent stimulator of vascular endothelial growth factor (VEGF) expression, a key proangiogenic factor, and this induction is believed to be mediated primarily through hypoxia-inducible factor-1 (HIF-1) (Maxwell et al., 1997). HIF-1 is one of the master regulators that orchestrate the cellular responses to hypoxia. It is a heterodimer composed of an inducibly expressed HIF-1 subunit and a constitutively expressed HIF-1 subunit. A growing body of evidence indicates that HIF-1 contributes to tumor progression and metastasis (Giaccia et al., 2003; Semenza, 2003). Immunohistochemical analyses have shown that HIF-1 is present in higher levels in human tumors than in normal tissues (Zhong et al., 1999). HIF-1 is a potent activator of angiogenesis and invasion through its up-regulation of target genes critical for these functions (Carmeliet et al., 1998; Kung et al., 2000; Ratcliffe et al., 2000). Such genes share the presence of hypoxia response elements, which contain binding sites for HIF-1 (Semenza, 2003). Therefore, because HIF-1 expression and activity seem central to tumor growth and progression, HIF-1 inhibition becomes an appropriate anticancer target (Maxwell et al., 1997; Kung et al., 2000; Giaccia et al., 2003; Semenza, 2003). It is interesting that VEGF is overexpressed not only in advanced colon cancers but also in premalignant colonic adenomas (Wong, 1999). The factors that may contribute to this enhanced VEGF expression are not defined fully. Although the mechanism of the possible protective effect of coffee or its products is unclear, potential protective effects could include antagonistic effects of the adenosine receptors, named A1, A2A, A2B, and A3 (Fredholm et al., 2001). These receptors belong to the P1 subclass of the purinergic family of G protein-coupled receptors, which are activated by adenosine. Adenosine is an ubiquitous autacoid that accumulates to high levels in hypoxic tissues as a result of ATP breakdown (Fredholm et al., 2001). This nucleoside has been involved in the regulation of the cellular response to hypoxia. It is recognized that significant levels of adenosine are present in the extracellular fluid of solid tumors (Fredholm et al., 2001), suggesting a role for this autacoid in tumor growth. In particular, the A3 subtype is highly expressed in tumor cells (Gessi et al., 2001, 2002; Merighi et al., 2001) and is able to significantly up-regulate the expression of HIF-1 in hypoxic tumors (Merighi et al., 2005a, 2006), suggesting that A3 receptor overexpression may be a good candidate as a tumor cell marker (Gessi et al., 2004; Madi et al., 2004). Adenosine also plays a role in the promotion of angiogenesis (Montesinos et al., 2004). Regulation of expression of VEGF through adenosine receptors has been demonstrated in different cell types (Feoktistov et al., 2002, 2003, 2004; Leibovich et al., 2002). The aim of this study is to determine whether caffeine may regulate HIF-1 , VEGF, and IL-8 in colon cancer cells during hypoxia. Materials and Methods Cell Lines, Reagents, and Antibodies. HT29 human tumor colon cells were obtained from American Type Culture Collection (Manassas, VA). Human umbilical vein endothelial cells (HUVEC), tissue culture media and growth supplements were obtained from Lonza Bioscience (Bergamo, Italy). Antiadenosine A2B and antiadenosine A3 receptor antibodies (pAb) were from Alpha Diagnostic (Milano, Italy). Human anti-HIF-1 and human anti-HIF1 antibodies (mAb) were obtained from BD Transduction Laboratories (Milan, Italy). Anti-human vascular endothelial growth factor (VEGF) antibody was developed in goat using recombinant human VEGF165 as immunogen. U0126 (inhibitor of MEK-1 and MEK-2), SB202190 (inhibitor of p38 MAP kinase), human anti-ACTIVEMAPK, and human anti-ERK1/2 antibodies (pAb) were from Promega (Milan, Italy). SH5 (inhibitor of Akt) was from Vinci-Biochem (Florence, Italy). Human phospho-p38 and human p38 MAP kinase antibodies were from Cell Signaling Technology (Milan, Italy). P11w, a firefly luciferase reporter plasmid, comprising the 5 -flanking 985 to 939 base pairs of the human VEGF gene that include an HIF-1-binding site, and p11m, the mutated version of p11w containing a nonfunctional HIF-1-binding site (Forsythe et al., 1996), were obtained from the American Type Culture Collection. BriteLite Ultra-High Sensitivity Luminescence Reporter Gene Assay System kit was obtained from PerkinElmer Life and Analytical Sciences (Milan, Italy). Fugene 6 transfection reagent was purchased from Roche Molecular Biochemicals (Milan, Italy). ZM 241385 and [H]ZM 241385 (specific activity, 17 Ci/mmol) were obtained from Tocris Cookson Ltd. (Bristol, UK). MRE 2029F20, MRE 3008F20, and B64 were synthesized by Dr. Pier Giovanni Baraldi (Department of Pharmaceutical Sciences, University of Ferrara, Ferrara, Italy). [H]MRE 2029F20 (specific activity, 123 Ci/mmol) and [H]MRE 3008F20 (specific activity, 67 Ci/mmol) were obtained from GE Healthcare (Chalfont St. Giles, Buckinghamshire, UK). [H]DPCPX (specific activity, 120 Ci/mmol) was obtained from PerkinElmer Life and Analytical Sciences (Waltham, MA). Adenosine A2B and A3 receptor small interfering RNAs (siRNAs) were from Santa Cruz Biotechnology (Santa Cruz, CA). Unless otherwise noted, all other chemicals were purchased from Sigma (Milan, Italy). Cell Culture. HT29 human tumor colon cells were maintained in RPMI 1640 medium containing 10% fetal calf serum, penicillin (100 U/ml), streptomycin (100 g/ml), and L-glutamine (2 mM) at 37°C in 5% CO2/95% air. HUVEC used in this study were from passages 2 to 7. Establishment of Hypoxic Culture Condition. For hypoxic conditions, cells were placed for the indicated times in a modular incubator chamber and flushed with a gas mixture containing 1% O2, 5% CO2, and balance N2 (MiniGalaxy, RSBiotech, Irvine, Scotland). Maintenance of the desired O2 concentration was constantly monitored during incubation using a microprocessor-based oxygen controller. Caffeine Treatment of Cancer Cells. Exponentially growing cells (70–80% confluence) in complete medium were pretreated for 396 Merighi et al. at A PE T Jornals on Sptem er 8, 2017 m oharm .aspeurnals.org D ow nladed from 1 h with different concentrations of caffeine, followed by continual incubation in normal culturing conditions or exposure to hypoxia (1% O2) for indicated time intervals according to the purpose of the experiment. Membrane Preparation. For membrane preparation, the culture medium was removed. The cells were washed with PBS and scraped off of T75 flasks in ice-cold hypotonic buffer (5 mM Tris HCl and 2 mM EDTA, pH 7.4). The cell suspension was homogenized with a Polytron homogenizer (Kinematica, Basel, Switzerland), and the cell suspension was centrifuged for 10 min at 1000g. The supernatant was then centrifuged again for 30 min at 100,000g, and the membrane pellet was frozen at 80°C until the use in competition binding experiments. Competition Binding Experiments at A1, A2A, A2B, and A3 Adenosine Receptors. Binding of [H]DPCPX to A1 receptors expressed in HT29 cells was performed for 120 min at 25°C in 50 mM Tris-HCl buffer, pH 7.4, containing 1 nM [H]DPCPX, diluted membranes (100 g of protein per assay), and caffeine. Nonspecific binding was determined in the presence of 1 M DPCPX and was always 10% of the total binding. Binding of 1 nM [H]ZM 241385 to human A2A expressed in HT29 membranes (100 g of protein per assay) was performed using 50 mM Tris-HCl buffer, 10 mM MgCl2 pH 7.4, and different concentrations of caffeine for an incubation time of 60 min at 4°C. Nonspecific binding was determined in the presence of 1 M ZM 241385 and was approximately 20% of total binding. Competition experiments to human A2B expressed in HT29 membranes were performed using 3 nM [H]MRE 2029F20 for an incubation time of 60 min at 4°C. Nonspecific binding was defined as binding in the presence of 1 M MRE 2029F20 and was 25% of total binding. Binding of [H]MRE 3008F20 to human A3 expressed in HT29 membranes was carried out in 50 mM Tris-HCl buffer, 10 mM MgCl2, and 1 mM EDTA, pH 7.4, containing 1 nM [H]MRE 3008F20, membranes (100 g of protein per assay), and caffeine for 120 min at 4°C. Nonspecific binding was defined as binding in the presence of 1 M MRE 3008F20 and was approximately 25 to 30% of total binding. Eight different concentrations of caffeine were studied. Measurement of cAMP Levels. HT29 cells in exponential growth were exposed to drugs for 2 h. After the incubation, the HT29 cells were collected, washed three times in ice-cold PBS, lysed, and centrifuged. The supernatants were assayed for cAMP determination using an R&D cAMP assay kit following the manufacturer’s instructions (Parameter kit; R&D Systems, Minneapolis, MN). Conditioned Medium. To obtain conditioned medium from N(3iodobenzyl)2-chloroadenosine-5 N-methyluronamide (Cl-IB-MECA)treated HT29 human tumor colon cells, we plated 10 HT29 cells in a 10-cm diameter plate containing RPMI 1640 medium with 10% fetal bovine serum. After 24 h, the medium of these cells was replaced with fresh growth medium containing Cl-IB-MECA (0 or 100 nM). The plates were then incubated under normoxic or hypoxic conditions. After 1 day of incubation, conditioned medium was removed and centrifuged at 4000g for 20 min at 4°C through an Amicon Ultra-4 centrifugal filter (Millipore, Billerica, MA) to remove any trace of Cl-IB-MECA. The molecular mass cutoff of the filters was 5 kDa, and the molecular mass of Cl-IB-MECA is 0.544 kDa. The flow-through containing excess Cl-IB-MECA was discarded, and the retentate was collected. Furthermore, to exclude that Cl-IB-MECA itself may have an inhibitory effect on the migration assay, we treated HUVECs directly with Cl-IB-MECA 100 nM, which was insufficient to modulate HUVEC migration. The final filter retentate was concentrated 40-fold for use in the migration and proliferation assays. JAM Test. This assay measures cell death by quantifying the amount of fragmented DNA, as described previously (Merighi et al., 2005b). Target cells were labeled with 1 Ci/ml [H]thymidine for 20 h in RPMI 1640 medium containing 10% fetal calf serum, penicillin (100 U/ml), streptomycin (100 g/ml), L-glutamine (2 mM). The cells were then washed and treated with new unlabeled medium containing caffeine for 24 h. At the end of the incubation period, the cells were trypsinized, dispensed in 4 wells of a 96-well plate, and filtered through Whatman GF/C glass-fiber filters using a MicroMate 196 cell harvester (PerkinElmer). The filter-bound radioactivity was counted on a TopCount Microplate Scintillation Counter (efficiency 57%) with MicroScint-20 (both from PerkinElmer Life and Analytical Sciences). The amount of apoptotic and necrotic cells, measured as the loss of radioactivity associated with the loss of fragmented and degraded DNA, was detected by filtration and subsequent washing with a MicroMate 196 cell harvester followed by quantification with a TopCount Microplate Scintillation Counter. The percentage of cell death is expressed as 100 (dpm(U) dpm(T))/ dpm(U), where dpm(U) is the radioactivity of untreated cells, and dpm(T) is the radioactivity of treated cells (Merighi et al., 2005b). MTS Assay. The MTS assay was performed to determine colon cell viability and proliferation according to the manufacturer’s protocol from the CellTiter 96 AQueous One Solution (Promega) cell proliferation assay, as described previously (Merighi et al., 2005b). Cells (10) were plated in 24-multiwell plates; 500 l of complete medium was added to each well with different concentrations of caffeine. The cells were then incubated for 24 h. At the end of the incubation period, MTS solution was added to each well. The optical density of each well was read on a spectrophotometer at 570 nm. For each experiment, four individual wells of each drug concentration were prepared. Each experiment was repeated three times. Migration Assay. Cell migration was performed with the Transwell system (Chemicon International, Temecula, CA), which allows cells to migrate through 8m pore size polycarbonate membrane. In brief, cells were trypsinized, washed, and resuspended in serum-free Dulbecco’s modified Eagle’s medium (5 10 cells/ml). This suspension (300 l) was added to the upper chamber of Transwells. The lower chamber was filled with 500 l of conditioned medium. After the incubation (6–24 h), filters were removed, and cells remaining on the upper surface of the membrane (i.e., that had not migrated through the filter) were removed with a cotton swab. Then, membranes were washed with PBS, and cells present beneath the membrane were fixed with ice-cold methanol for 15 min and stained with the Cell Stain Solution (QCM Colorimetric Cell Migration Assay; Chemicon International). The stained insert was transferred to a well containing the extraction buffer. The dye mixture was transferred to a 96-well microtiter plate suitable for colorimetric measurement. Analysis was performed on three wells for each condition, and each experiment was repeated three times. Western Blot Analysis. Whole-cell lysates, prepared as described previously (Merighi et al., 2005b), were resolved on a 10% SDS gel and transferred onto the nitrocellulose membrane. Western blot analyses were performed as described previously (Merighi et al., 2005a) with anti-HIF-1 (1:250 dilution) and anti-HIF-1 antibodies (1:1000 dilution) in 5% nonfat dry milk in PBS/0.1% Tween 20 overnight at 4°C. Aliquots of total protein sample (50 g) were analyzed using antibodies specific for phosphorylated (Thr183/ Tyr185) or total p44/p42 MAPK (1:5000 dilution), phosphorylated (Thr180/Tyr182) or total p38 MAPK (1:1000 dilution), and for phosphorylated Akt (Ser473) (1:1000 dilution). The protein concentration was determined using BCA protein assay kit (Pierce, Rockford, IL). Membranes were washed and incubated for 1 h at room temperature with peroxidase-conjugated secondary antibodies against mouse and rabbit IgG (1:2000 dilution). Specific reactions were revealed with the Enhanced Chemiluminescence Western blotting detection reagent (GE Healthcare). The membranes were then stripped and reprobed with antitubulin antibodies (1:250) to ensure equal protein