Curcumin inhibits interleukin 8 production and enhances interleukin 8 receptor expression on the cell surface

Impact on human pancreatic carcinoma cell growth by autocrine regulation

Authors


Abstract

BACKGROUND

Curcumin, the yellow pigment in turmeric, has been shown to prevent tumor progression in a variety of tissues in rodents. The authors investigated the effect of curcumin on human carcinoma cell lines to determine whether constitutive interleukin-8 (IL-8) production of tumor cells was correlated with nuclear factor κB (NF-κB) activation and cell growth activity.

METHODS

A human pancreatic carcinoma cell line, SUIT-2, was incubated with various concentrations of curcumin for 2 hours. Biologic features, including IL-8 production, DNA binding activity, transactivation of NF-κB, cell growth activity, cell viability, and the expression of IL-8 receptors (CXCR1 and CXCR2) were analyzed.

RESULTS

The constitutive production of IL-8 was inhibited by curcumin at concentrations of 10–100 μM in a dose dependent manner. NF-κB activity was reduced significantly by curcumin treatment. Pretreatment with curcumin inhibited the growth rate of carcinoma cells significantly. Such cell growth inhibition by curcumin was not recovered by exogenous recombinant IL-8. The investigation of expression in IL-8 receptors, CXCR1 and CXCR2, revealed that the expression of both receptors was enhanced remarkably by curcumin. Exogenous IL-8 could not recover this enhancement of IL-8 receptors. These results suggest that curcumin inhibits IL-8-induced receptor internalization.

CONCLUSIONS

The authors concluded that curcumin contributed not only to the inhibition of IL-8 production but also to signal transduction through IL-8 receptors. These data suggest that curcumin reduces numerous IL-8 bioactivities that contribute to tumor growth and carcinoma cell viability. From this point of view, curcumin is a potent anticancer agent that inhibits the production of proinflammatory cytokines, including IL-8, by tumor cells. Cancer 2002;95:1206–14. © 2002 American Cancer Society.

DOI 10.1002/cncr.10812

Interleukin-8 (IL-8), is a multifunctional CXC chemokine that affects human neutrophil functions, including chemotaxis, enzyme release, and expression of surface adhesion molecules. Recently, it was clarified that IL-8 and other ELR motif-containing CXC chemokines are mediators of angiogenesis,1, 2 and the expression of IL-8 was correlated with tumorigenic and metastatic potentials of human carcinoma cells.3, 4 We previously reported that IL-8 and its receptors were produced constitutively and commonly by carcinoma cells.5 Moreover, IL-8 functioned as an autocrine growth factor for carcinoma cells.5–7 Thus, there has been increased awareness of the control of IL-8 expression by carcinoma cells.

Nuclear factor κB (NF-κB) is a key molecule in the transcriptional regulation of the IL-8 gene.8, 9 In unstimulated cells, NF-κB is sequestered in the cytoplasm by the inhibitory protein IκB. Ultraviolet radiation, bacterial lipopolysaccharide, viral gene products, and inflammatory cytokines promote IκB degradation, thereby allowing NF-κB to enter the nucleus and induce gene transcription. Because NF-κB plays a central role in mediating proinflammatory or apoptotic gene expression, there is growing interest in modulating its activity.

Curcumin, which is extracted from the rhizomes of the plant Curcuma longa L., generally is used as a spice and food coloring.10, 11 Curcumin is known as a traditional medicine to treat inflammatory diseases.12 The anti-inflammatory, antioxidant, and anticarcinogenic properties of curcumin have been well documented along with its low toxicity.10, 11, 13–17 Many of the beneficial effects of curcumin correlate with its ability to block the activity of transcription factor NF-κB. It has been shown that curcumin prevents proinflammatory gene expression by inhibiting NF-κB activity and that its molecular mechanism involves blocking of IκB phosphorylation.18, 19 Although the effects of curcumin on inflammatory responses have been examined well in nontransformed cells,20–22 the effects of curcumin on human carcinoma cells remain unclear.

In this study, we investigated the effects of curcumin on human carcinoma cell lines to determine whether IL-8 production of carcinoma cells is correlated with NF-κB activation and cell growth activity. Our data demonstrate that curcumin inhibited IL-8 production due to blocking of NF-κB activation, which was accompanied by cell growth inhibition. Furthermore, curcumin dramatically changed the levels of IL-8 receptor expression on cell surfaces. This change also may be associated with the inhibition of tumor cell growth.

MATERIALS AND METHODS

Cell Lines and Culture Conditions

A human carcinoma cell line, SUIT-2,23 was used for this study. This cell line was cultured routinely in RPMI 1640 medium (Gibco Laboratories, Grand Island, NY) supplemented with 10% fetal calf serum (Gibco BRL, Gaithersburg, MD), 100 units/mL penicillin, and 100 mg/mL streptomycin at 37 °C in a humidified atmosphere of 5% CO2/95% air. Carcinoma cells and culture supernatants were collected when cells reached nearly confluent density. Cells and culture supernatants were stored at −80 °C until they were assayed.

Curcumin was obtained from Sigma Chemical Company (St. Louis, MO). In our experiments, curcumin was dissolved in dimethyl sulfoxide (DMSO) on the day of use and was added to the cells at final concentrations of 0 μM, 10 μM, 50 μM, and 100 μM for 2 hours. After a 2-hour preincubation, the medium was changed, and supernatants and cells were collected after an additional 6-hour incubation. The concentration of DMSO was always less than 0.1% (volume/volume).

Measurement of IL-8

IL-8 concentration was measured by using an enzyme-linked immunosorbent assay (ELISA) kit (BioSource International, Inc, CA). This assay showed no measurable cross reactivity with either human IL-1, IL-6, or tumor necrosis factor. The lower limit of detection was 15.6 pg/mL. The absorbance of the samples was compared with the standard curve.

Preparation of Nuclear and Cytoplasmic Extracts

The isolation of nuclear extracts was performed according to the previously reported method with minor modifications.9 In brief, all procedures were performed on ice. Nearly confluent monolayers of cells, which had been treated with 0–100 μM of curcumin or with the same amount of DMSO for the appropriate time, were washed with ice-cold, phosphate-buffered saline (PBS); harvested by scraping into 1 mL of PBS; and pelleted in a 1.5 mL microfuge tube at 1500 revolutions per minute for 5 minutes. The pellet was washed twice with ice-cold PBS and then suspended in one packed-cell volume of lysis buffer (10 mM N-2-hydroxyethyl-piper-az-ine-N′-2-ethane-sulphonate [HEPES], pH 7.9; 10 mM KCl; 0.1 mM ethylenediamine tetraacetic acid [EDTA]; 1.5 mM MgCl2; 0.25 volume % Nonidet-P40; 1 mM dithiothreitol [DTT], and 0.1 mM phenylmethylsulfonyl fluoride [PMSF]). After a 5-minute incubation on ice, the nuclear pellet was isolated by centrifugation. The supernatant represented the cytoplasmic extract. The nuclear pellet was resuspended in one packed-cell volume of extract buffer (20 mM HEPES, pH 7.9; 420 mM NaCl; 0.1 mM EDTA; 1.5 mM MgCl2; 25% [volume/volume] glycerol; 1 mM DTT; and 0.5 mM PMSF), and the nuclei were incubated on ice for 20 minutes, the nuclear debris was removed by centrifugation, and the protein concentration of the nuclear extract was determined. The nuclear extracts were stored at −70 °C until further use.

Western Blot Analysis

Trypsinized adherent cells were combined and washed with PBS. After collection, the cells were lysed in sodium dodecylsulfate (SDS) solubilization buffer (62.5 mM Tris-Hcl, pH 6.8; 10% glycerol; 5% β-mercaptoethanol; and 1% SDS). Cytosolic protein was isolated from control cells and carcinoma cells transfected with the luciferase reporter gene. Equal amounts of proteins were boiled for 5 minutes and electrophoresed under reducing conditions on 8% (weight/volume) polyacrylamide gel. Proteins were transferred electrophoretically to Hybond-ECL membrane (Amersham, Tokyo, Japan) and incubated with primary rabbit antibodies against p65 (Santa Cruz Biotechnology, Santa Cruz, CA) and IκB-α followed by peroxidase-linked secondary antibody. An Amersham ECL chemiluminescent Western system was used to detect the secondary antibody.

Electrophoretic Mobility Shift Assay

Nuclear extracts were prepared as described above. Then, extracts (5 mg) were incubated with radio-labeled, double-stranded κB sites; separated by nondenaturing electrophoresis; and analyzed by autoradiography. An electrophoretic mobility shift assay (EMSA) was performed according to the manufacturer's protocol (Gel Shift Assay Systems; Promega, Madison, WI).

A double-stranded oligonucleotide containing the NF-κB binding site (5′-AGT TGA GGG GAC TTT CCC AGG C-3′) was end labeled with [γ-32P]adenosyl triphosphate and used as a probe. Equivalent amounts of nuclear protein were incubated on ice for 10 minutes in a buffer containing 12 mM HEPES, pH 7.9; 4 mM Tris-HCl, pH 7.9; 25 mM KCl; 5 mM MgCl2; 1 mM EDTA; 1 mM DTT; 50 ng/mL poly [d (I-C)]; and 0.2 mM PMSF. EMSA probes for NF-κB were obtained from Santa Cruz Biotechnology. The probes were end labeled with 32P using T4 kinase (Promega); then, unincorporated nucleotides were removed with a G-25 Sephadex separation column (Boehringer Mannheim, Indianapolis, IN). The respective labeled probe (100,000 dpm) was then added to the extracts and incubated with the reaction mixture on ice for an additional 20 minutes. The probes were resolved through nondenaturing 4% polyacrylamide gel electrophoresis.

Luciferase Reporter Assay

SUIT-2 cells were transfected using SuperFect™ transfection reagent (QIAGEN, Valencia, CA), as described in the literature.24 For the reporter gene assay, SUIT-2 cells were transfected with luciferase reporter gene, and, after 48 hours of incubation, the cells were harvested in passive lysis buffer (Promega), and the activities of firefly and renilla luciferase were quantified using a dual-luciferase assay system (Promega).

Flow Cytometry

SUIT-2 cells were suspended at a concentration of 1 × 106 cells/mL in ice-cold RPMI 1640. To detect CXCR1, SUIT-2 cells were incubated with 1 mg/mL of carboxyfluorescein succinimidylester (CFS)-conjugated mouse antihuman immunoglobulin 2a (IgG2a) monoclonal antibody (mAb) that is specific for CXCR1. Anti-p65 mouse IgG2b antibody served as a control for nonspecific binding. To detect CXCR2, SUIT-2 cells were incubated with phycoerythrin-conjugated mouse antihuman IgG2a mAb that is specific for CXCR2. Flow cytometry was performed using a FACScan instrument (Becton Dickinson, Mountain View, CA), as described previously.25

Cell Proliferation Assay

SUIT-2 cells were seeded at a density of 1 × 105/mL in 6-well plates, and, 24 hours after incubation, 0 μM, 10 μM, 50 μM, or 100 μM of curcumin were added. Medium was changed after 2 hours. After the beginning of exposure, at 24 hours, 48 hours, and 72 hours, the cells were harvested, and total cell numbers were counted. A methyl-thiazoldiphenyl tetrazolium (MTT) assay was performed as described previously.26

Statistical Analysis

The data are expressed as mean values ± standard error. Statistical significance was determined with Student t tests and correlation coefficients. P values < 0.05 were considered significant.

RESULTS

Effects of Curcumin on IL-8 Production in SUIT-2 Carcinoma Cells

SUIT-2 cells were cultured in the presence of various doses of curcumin for 2 hours, and the concentration of IL-8 was examined in supernatants harvested after 6 hours using an ELISA. The results are shown in Figure 1. Curcumin significantly inhibited the constitutive production of IL-8 at concentrations of 50 μM and 100 μM in a dose dependent manner.

Figure 1.

Effects of curcumin on the production of interleukin 8 (IL-8) by SUIT-2 human pancreatic carcinoma cells. Curcumin significantly inhibited the constitutive production of IL-8 at a concentration of 50–100 μM in a dose dependent manner. Each bar represents the mean ± standard error of three individual experiments. Statistical significance compared with controls: single asterisk, P < 0.05; double asterisks, P < 0.005.

Regulation of p65 and IκB-α Protein Levels by Curcumin

To determine whether IL-8 production in carcinoma cell lines is influenced by curcumin, SUIT-2 cells that were treated with curcumin for 2 hours were harvested, and total protein was extracted. Then, 20 μg of protein were subjected to Western blot analysis to examine the protein level of the transcriptional factor, which is essential for IL-8 gene expression. Curcumin did not alter the protein level of the p65 subunit of NF-κB protein expression (Fig. 2A). Although the protein level of the transcriptional factor remained unchanged in SUIT-2 cells after treatment, curcumin increased the protein level of IκB-α (Fig. 2B). The increase in the protein level of IκB-α suggests the inhibition of IκB-α degradation by proteasomes.

Figure 2.

Effects of curcumin on protein levels of p65 in SUIT-2 carcinoma cells. Curcumin treatment did not alter protein levels of the p65 subunit of nuclear factor κB (A). However, curcumin treatment increased protein levels of IκB-α (B).

Down-Regulation of NF-κB Activity by Curcumin

Because NF-κB activation is essential for IL-8 gene expression, we next determined NF-κB activation by using EMSA. SUIT-2 cells were pretreated with curcumin (100 μM) or with medium alone for 2 hours. The nuclear extracts were isolated for EMSA, as described above (see Materials and Methods). The EMSA revealed that the decrease in NF-κB-binding activation was correlated with the dose dependent manner of curcumin (Fig. 3).

Figure 3.

Effects of curcumin on DNA binding activity of nuclear factor κB (NF-κB) in human carcinoma cells were examined by electrophoretic mobility shift assay (EMSA). Curcumin blocked constitutive NF-κB binding activity in SUIT-2 pancreatic carcinoma cells. SUIT-2 cells were pretreated with curcumin (100 μM) or with medium alone for 2 hours. Nuclear extracts (5 μg) were tested for κB binding activity by EMSA, as described in the text (see Materials and Methods). NF-κB specific complexes are indicated the arrow. This EMSA is representative of three separate experiments.

Furthermore, we investigated the change in transcriptional activity with curcumin using a luciferase expression assay. SUIT-2 cells were transfected by SuperFect with 5 μg of NF-κB-Luc and pRL-SV40 vector. Curcumin was added after the transfection, and an NF-κB reporter assay was performed after 48 hours of incubation. Then, SUIT-2 cells were harvested after 6 hours of pretreatment with or without curcumin for a dual-luciferase expression assay. The activities of firefly luciferase were measured using the luciferase assay system, as described previously.24 Those experiments revealed a significant decrease in NF-κB transcriptional activation in a dose dependent manner of curcumin (Fig. 4).

Figure 4.

Effects of curcumin at the various concentrations on the transcriptional activity of nuclear factor κB (NF-κB) in human carcinoma cells were examined by Luciferase reporter gene assay. The activities of firefly luciferase for NF-κB transcriptional activation were decreased by curcumin significantly in dose dependent manner. Each bar represents the mean ± standard error of three individual experiments. Statistical significance compared with controls: asterisk, P < 0.05.

Effect of Curcumin on Cell Growth

Because, as we reported previously, IL-8 promotes SUIT-2 cell growth in an autocrine manner, the inhibition of IL-8 production by curcumin should have an impact on cell growth. SUIT-2 cells were incubated in media containing 0 μM, 10 μM, 50 μM, or 100 μM of curcumin for 2 hours and were examined for the next 3 days. The incubation of the cells with curcumin slowed their growth rate compared with controls (Fig. 5A). The MTT assay also revealed that curcumin affected cell survival (Fig. 5B).

Figure 5.

(A) Effects of curcumin at various concentrations on the growth of human carcinoma cells were examined. Incubation of cells with curcumin slowed the growth rate compared with controls. (B) The ratio of cell viability by methyl-thiazoldiphenyl tetrazolium assay also revealed that curcumin affected cell survival in dose dependent manner. However, cell growth inhibition by curcumin did not recover exogenous recombinant interleukin 8 (IL-8). Each bar represents the mean ± standard error of three individual experiments. Statistical significance compared with controls: single asterisk, P < 0.05; double asterisks, P < 0.005.

In addition, we determined whether exogenous IL-8 was capable of promoting the recovery of cell growth after treatment. Figure 5A,B shows that cell growth inhibition by curcumin did not recover recombinant human IL-8. These results showed that exogenous IL-8 could not promote cell growth after curcumin treatment.

Enhancement of IL-8 Receptor Expression on SUIT-2 Cell Surfaces by Curcumin

Next, we examined the effect of curcumin on IL-8 receptor expression. SUIT-2 cells were incubated with 0–100 μM of curcumin for 2 hours. Although reverse transcriptase-polymerase chain reaction and Western blot analyses demonstrated IL-8 receptor gene expression in SUIT-2 cells, flow cytometry could not detect the CXC receptor on cell surfaces (Fig. 6A–C). However, as shown in Figure 6, exposure of the cell to curcumin remarkably enhanced the expression of both CXCR1 and CXCR2 in a dose dependent manner, as determined by FACS analysis using anti-CXCR1 or anti-CXCR2 specific antibodies. Conversely, Western blot analysis revealed no significant change in the level of IL-8 receptor protein after curcumin treatment (Fig. 6C).

Figure 6.

Effects of curcumin on interleukin 8 (rIL-8) receptor expression in human carcinoma cells were examined by flow cytometry. Total surface expression of CXCR1 and CXCR2 was determined by flow cytometry using the human anti-CXCR1 or anti-CXCR2 combined with a fluorescein isothiocyanate-conjugated antihuman antibody. In controls, flow cytometry detected small amounts of CXCR on the cell surface (A, top). However, exposure of the cell to curcumin remarkably enhanced the expression of both CXCR1 and CXCR2 in a dose dependent manner (A, middle). Exogenous recombinant IL-8 (rIL-8) could not recover the enhancement of IL-8 on the cell surface (A, bottom). Those results of flow cytometry are expressed as intensity of fluorescence level for CXCR1 and CXCR2 (B). Western blot analysis also showed no significant change in the protein level of IL-8 receptor A (IL-8RA) after curcumin treatment (C).

Because IL-8 receptors on the cell surface internalized and decreased by a constitutive IL-8 autocrine mechanism, we hypothesized that this enhancement is due to blockage of the autocrine mechanism by the inhibition of IL-8 production after treatment with curcumin. Therefore, we checked the change in IL-8 receptor localization in the presence of exogenous IL-8; however, the addition of recombinant IL-8 failed to promote the recovery of IL-8 internalization (Fig. 6A,B).

DISCUSSION

Many tumor cells constitutively secrete higher levels of proinflammatory cytokines, like IL-8; however, the mechanisms and biologic effects of those cytokines remain unclear. Recent studies have revealed that the expression of cytokines in tumor cells contributes to tumor cell proliferation,7, 27 angiogenesis,28, 29 inflammation,30, 31 catabolism,32 and immunosuppression.33, 34 Therefore, the control of cytokine production and its signal transduction in tumor cells may have an important role in therapy for patients with malignant disease.

The current results demonstrate that the food derivative curcumin reduced the constitutive production of IL-8 due to the prevention and inhibition of NF-κB activation. Moreover, curcumin enhanced the expression of IL-8 receptors by blocking their internalization after curcumin treatment. These data suggest that curcumin inhibits both IL-8 production and its functions.

Curcumin is a food derivative that is used generally as a spice and food coloring, and nontoxic consumption of up to 100–180 mg per day has been reported in humans.35, 36 In addition, curcumin has been reported as a nonmutagenic agent; thus, it is considered safe and has relevant efficacy for use in vivo.37 Moreover, numerous experimental studies have demonstrated the possibility of therapy with curcumin for patients with malignant disease. Chen et al. reported that curcumin inhibits tumor cell proliferation by interfering with the cell cycle and inducing apoptosis.26 Arbiser et al. suggested that the activity of curcumin in the inhibition of carcinogenesis may be mediated by the inhibition of angiogenesis.38 Although several pathways, including apoptosis, may contribute to the inhibition of tumor growth by curcumin, the inhibition of NF-κB plays an important role in the inhibition of tumor growth.18

It has been shown that the food derivative curcumin inhibits NF-κB activity, and experimental studies have reported that curcumin blocks a signal upstream of NF-κB-inducing kinase and inhibitory factor-κβ kinase (IKK).19, 39 The effects of curcumin in some endothelial or epithelial cells is confident,40, 41 but its effects on tumor cells remain unclear. For example, we also examined the effect of MG132, which is known generally as an NF-κB inhibitor,9 but MG132 did not inhibit the high activity of NF-κB in SUIT-2 carcinoma cells and failed to suppress constitutive IL-8 production (data not shown). NF-κB is the most crucial transcriptional factor for IL-8 expression for inflammatory responses in lymphatic cells and epithelial cells. Increased activation of NF-κB in tumors has been reported.42–44 Our study also demonstrated the constitutive activation of NF-κB in SUIT-2 cells, in accord with the results of prior studies. In SUIT-2 carcinoma cells, curcumin reduced NF-κB constitutive activation as well as nontransforming cells. It is believed that such suppression is responsible for the inhibition of IL-8 production in SUIT-2 carcinoma cells. Therefore, these results suggest that curcumin is effective for suppression of NF-κB activation in tumor cells.

The current results also showed that the growth of SUIT-2 carcinoma cells was inhibited significantly by curcumin treatment in a dose dependent manner. This finding is consistent with our prior studies, suggesting that IL-8 plays a role in promoting the growth of SUIT-2 cells in an autocrine fashion. Nevertheless, additional recombinant IL-8 could not recover that inhibition of cell growth. These results suggest that curcumin treatment may have an impact on the sensitivity of SUIT-2 carcinoma cells to IL-8. We measured the expression levels of IL-8 receptors on cell surfaces by FACScan; surprisingly, their expression was enhanced significantly after curcumin treatment. The results from Figure 6C indicate that this enhancement was due not to a change in protein amounts but, rather, to the localization of IL-8 receptors. The mechanisms of IL-8 receptor regulation are not defined well, but it is accepted widely that ligand-promoted internalization is one of the most important regulatory pathways. In neutrophils, IL-8 stimulated β-arrestin dependent internalization of the CXCR1 receptor. Upon agonist binding, CXCR1 and CXCR2 receptor activation is followed by receptor phosphorylation on multiple serine residues and subsequent desensitization of the receptor to further stimulation.45 These events usually are accompanied by receptor endocytosis and/or recycling of the receptor. Autocrine mechanism by IL-8 should induce constitutive IL-8 receptor endocytosis and internalization in SUIT-2 cells. After curcumin treatment, the inhibition of IL-8 production may cause the decrease of IL-8-induced receptor internalization. Therefore, IL-8 receptors on cell surfaces seem to increase their expression on cell surfaces after treatment. However, exogenous recombinant IL-8 failed to recover IL-8 receptor internalization. This disagreement suggests that several mechanisms may contribute to the enhancement of receptor expression. CXCR1 and CXCR2 endocytosis and internalization usually are accompanied by receptor activation followed by phosphorylation of their serine residue. Curcumin also is known as an inhibitory factor for several serine/threonine kinase, such as protein kinase C46 and IKK.19, 39 Previous studies did not explain fully the enhancement of CXCR1 and CXCR2 expression on cell surfaces, but it is possible that curcumin regulates the kinase activity for IL-8 receptor phosphorylation, resulting in the increased stability of IL-8 receptors on cell surfaces. Although phosphorylation is a crucial determinant of CXCR internalization, Western blot analysis showed no significant difference in mobility of CXCR by phosphorylation (Fig. 6C). The mechanisms that regulate enhancement of cell surface CXCR should be elucidated further.

In conclusion, our data provide some significant evidence that curcumin inhibits the constitutive activation of NF-κB and the disorderly production of IL-8 in human pancreatic carcinoma cells. These inhibitory effects have an important role in the suppression of tumor cell growth. Moreover, our findings indicate that the enhancement of IL-8 receptor expression may be due to the blockage of receptor internalization and also plays a role in the inhibition of cell growth by curcumin. Further analyses are needed to determine the mechanism of blockage IL-8 receptor internalization, but this is the first report concerning the impact of curcumin on the IL-8 receptor internalization.

The current study suggests that curcumin reduces numerous IL-8 bioactivities that promote tumor cell viability and tumor progression. In this regard, curcumin is capable of working as a potent agent that reduces tumor promotion. In vivo use of curcumin may be beneficial for patients with carcinoma who are affected by the enhanced production of various proinflammatory cytokines.

Ancillary