Dulbecco's modified Eagle's medium (DMEM) was purchased from Thermo Scientific (hemel hempstead, UK). Fetal bovine serum (FBS) was purchased from HyClone Laboratories (Logan, UT, USA). Phenol red-free DMEM and penicillin (100 U/ml) and 100 mg/ml streptomycin were purchased from Life Technologies (Rockville, MD, USA). UO126 was purchased from Selleck Chemicals (Houston, TX, USA). SB253580 and LY294002 were purchased from Tocris Bioscience (Ellisville, MO, USA). 4-Hydroxytamoxifen (4-OHT) was purchased from Sigma (St. Louis, MO, USA). The secondary horseradish peroxidase (HRP)-conjugated and mouse monoclonal anti-β-actin antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Rabbit polyclonal anti-MEK and anti-Akt antibodies (total and phospho-form) were purchased from Cell Signaling Technology (Beverly, MA, USA). Recombinant human IL-8 was purchased from R&D Systems (Minneapolis, MN, USA). The ECL prime reagents were purchased from Amersham (Buckinghamshire, UK).

We downloaded expression data from a public database [Kaplan-Meier plotter database (http://kmplot.com/breast)] and analyzed the clinical value of IL-8 expression in luminal A and B type breast cancer patients.

TamR was established using a previously reported methodology (6,18). Briefly, to establish TamR, MCF-7 cells were washed with PBS, and the culture medium was changed to phenol red-free DMEM containing 10% charcoal-stripped steroid-depleted FBS and 0.1 mM 4-hydroxytamoxifen (4-OHT). The cells were continuously exposed to this treatment regimen for two weeks, and the 4-OHT concentration was increased gradually to up to 3 mM over a 9-month period. Initially, cell growth rates were depressed. However, after exposure to the medium for 9 months, the rate of cell growth increased gradually, indicating the establishment of tamoxifen-resistant cells.

TamS and TamR cells were seeded at a density of 5×10⁴ cells/well in 6-well plates in growth medium containing 0.7% agar (1.5 ml/well) on top of a layer of growth medium containing 1.4% agar (2 ml/well). Growth medium (500 µl) with 10% FBS was added on top of the agar. The cell suspension was plated and cultured in a 37°C incubator for two weeks. After two weeks, viable colonies were stained with 0.01% crystal violet and were then observed using a CK40 inverted microscope (Olympus, Tokyo, Japan) (6).

Apoptosis assays were performed with the Annexin V-fluorescein isothiocyanate (FITC) apoptosis kit-I (BD Biosciences, San Diego, CA, USA), according to the manufacturer's protocol. Briefly, cells (1×10⁶ cells/ml) were collected and washed twice with PBS and then resuspended in 500 µl of staining solution containing 5 µl FITC-conjugated Annexin V and 5 µl propidium iodide (PI). After incubation for 15 min at room temperature (RT) in the dark, cells were immediately analyzed on a flow cytometer. Apoptotic cells were double stained with Annexin V and PI and were then analyzed using the FACS Vantage system (Becton-Dickinson, San Diego, CA, USA). The percentage of cells undergoing apoptosis was determined (6).

The total RNA was extracted from the cells using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. Isolated RNA samples were then used for RT-PCR. Samples (1 µg of total RNA) were reverse-transcribed into cDNA in 20 µl reaction volumes using a first-strand cDNA synthesis kit for RT-PCR, according to the manufacturer's instructions (MBI Fermentas, Hanover, MD, USA).

The gene expression was quantified by real-time PCR using a SensiMix SYBR kit (Bioline Reagents Ltd., London, UK) and 100 ng of cDNA per reaction. The sequences of the primer sets used for this analysis were: human IL-8 (forward, 5′-AGGGTTGCCAGATGCAATAC-3′; reverse, 5′-AAACCAAGGCACAGTGGAAC-3′) and GAPDH as an internal control (forward, 5′-ATTGTTGCCATCAATGACCC-3′ and reverse, 5′-AGTAGAGGCAGGGATGATGT-3′). An annealing temperature of 60°C was used for all of the primers. The PCR was performed in a standard 384-well plate format with an ABI 7900HT real-time PCR detection system. For data analysis, the raw threshold cycle (C_T) value was first normalized to the housekeeping gene for each sample to get the ΔC_T. The normalized ΔC_T was then calibrated to the control cell samples to get the ΔΔC_T.

Protein levels of IL-8 were measured using an ELISA kit for human IL-8 (KomaBiotech, Seoul, Korea), according to the manufacturer's instructions, and then a microtiter plate reader was used to read the plate at a 450 nm wavelength.

The cell culture media (supernatants) and cell lysates were used in the immunoblot analysis for MEK and β-actin. The proteins were boiled for 5 min in Laemmli sample buffer and then electrophoresed in 8 or 10% SDS-PAGE gels, respectively. The separated proteins were transferred to PVDF membranes and the membranes were then blocked with 10% skim milk in TBS with 0.01% Tween-20 for 15 min. The blots were incubated with anti-MEK, anti-Akt (total and phospho-form), anti-p38 (phospho-form), and β-actin antibodies (1/1,000 dilution) in 1% TBS/T buffer (0.01% Tween-20 in TBS) at 4°C overnight. The blots were washed three times in TBS with 0.01% Tween-20 and they were subsequently incubated with anti-rabbit peroxidase-conjugated antibody (1/2,000 dilution) in TBS/T buffer. After 1 h of incubation at room temperature (RT), the blots were washed three times and ECL prime reagents were used for development.

The empty (Lac Z) and adenoviral human CA-MEK cDNA were gifts from Dr Hyunil Ha (Korea Institute of Oriental Medicine, Daejeon, Korea). Recombinant adenovirus-expressing human CA-MEK was reproduced into 293A cells and the level of MEK phosphorylation was confirmed by western blotting. Adeniviral vectors were transfected into BT474 cells for 24 h and then further incubated with fresh serum-free media for 48 h to detect the level of IL-8 and phospho-MEK and ERK expression.

Statistical significance was determined using the Student's t-test. Data are presented as the mean ± SEM. All quoted P-values are two-tailed and differences were considered significant for P-values <0.05. Microsoft Excel was used for the statistical analyses.

Clinically, we analyzed whether elevated IL-8 levels confer a poor prognosis for human breast cancer patients using a Kaplan-Meier plotter database (http://kmplot.com/breast). In luminal A type breast cancer patients, patients with high expression of IL-8 showed poorer relapse-free survival in comparison to patients with low expression of IL-8 (P=0.0325, Fig. 1A). However, relapse-free survival by IL-8 expression was not significantly different between luminal B type breast cancer patients (P=0.65, Fig. 1B). Based on these results, we demonstrated that IL-8 expression may be associated with tamoxifen resistance.

In the present study, we investigated the functional role and regulatory mechanism of IL-8 expression in tamoxifen-resistant breast cancer. We established a tamoxifen-resistant breast cancer in vitro model and analyzed differential characteristics of tamoxifen-sensitive (TamS) and -resistant (TamR) breast cancer cell lines. In a recent study, we reported the morphological findings that TamS cells stacked up to form colonies and TamR cells were scattered, loosely packed colonies, with many branches (6). The tumorigenicity of TamS and TamR cells by tamoxifen treatment was analyzed using soft agar colony formation assays. As shown in Fig. 2B, the anchorage-independent growth of TamS cells was completely inhibited by 3 µM tamoxifen treatment while growth of TamR cells was maintained. In addition, we also analyzed the apoptotic cell death of TamS and TamR cells after tamoxifen treatment. TamS and TamR cells were treated with or without 3 µM tamoxifen for 24 h. The apoptotic cell population of TamS cells after tamoxifen treatment significantly increased by 47.23% over the control level. However, the apoptotic cell population of TamR cells was not significantly different from that of TamS cells (Fig. 2C).

In a previous study, Shi et al (19) reported that elevated IL-6 and IL-8 expression contributes to multidrug resistance in human breast cancer cells. Therefore, we also compared the level of IL-8 expression between TamS and TamR cells. Our results showed that the levels of IL-8 mRNA and protein expression were significantly increased in TamR cells when compared with TamS cells (Fig. 3). The level of IL-8 mRNA expression in TamR cells was 62.8±5.4-fold higher than that of TamS cells (Fig. 3A). In addition, IL-8 protein expression was also increased by 42.3±7.3 ng/ml of the control level (14.0±2.5 ng/ml) in TamR cells (Fig. 3B). Therefore, we demonstrated that induction of IL-8 may be associated with tamoxifen resistance in luminal type breast cancer.

To verify the regulatory mechanism of IL-8 expression, we analyzed the phosphorylation degree in a variety of signaling molecules in TamS and TamR cells. As shown in Fig. 4A, the activities of MEK and Akt were significantly increased in TamR cells. However, the activity of p38 was not critically different between TamR and TamS cells. Furthermore, cells were treated using specific inhibitors such as the MEK 1 inhibitor UO126, and the PI3-K inhibitor LY294002, for 48 h. After 48 h, the cell lysates were harvested to detect the levels of IL-8 mRNA expression. The levels of IL-8 mRNA expression were significantly increased by 68.6±19.7-fold over that of TamS cells (Fig. 4B). On the contrary, induction of IL-8 was decreased by UO (UO126, a specific MEK inhibitor), but not by LY294002, in TamR cells (Fig. 4B).

Next, we examined whether MEK directly regulates IL-8 expression in TamS cells. We overexpressed adenovirus-delivered constitutively active-MEK (CA-MEK) into TamS cells. As shown in Fig. 4C, we observed that the phosphorylation of MEK was enhanced in CA-MEK-overexpressing cells. Under the same condition, the level of IL-8 mRNA expression was significantly increased by 52.3±2.3-fold over the vector-alone levels with CA-MEK overexpression. Therefore, we demonstrated that the level of IL-8 expression is upregulated through a MEK/ERK-dependent pathway in tamoxifen-resistant breast cancer cells.

To inhibit IL-8-induced tumor cell growth, we treated TamR cells with the specific MEK1/2 inhibitor Uo126 and the specific CXCR1/2 inhibitor SB225002. As shown in Fig. 5A, anchorage-independent growth of TamR cells was significantly decreased by SB225002 treatment, but not by UO126 treatment. In this study, we suggest that UO126 plays an important role in regulating IL-8 transcriptional activity, while it did not heavily affect anchorage-independent growth of TamR cells.

Next, we investigated the co-relation between SB225002 and apoptotic cell death. We treated TamR cells with SB225002 at the indicated concentration for 24 h and then harvested whole cell lysates to detect the levels of PARP-1 expression. As expected, the level of full-length PARP-1 protein expression was dose-dependently decreased by SB225002 (Fig. 5B). Furthermore, there were serious changes in cell morphology and the number of apoptotic cells was significantly increased by SB225002 (Fig. 5C). The number of apoptotic cells was 669.5% greater than that of the control levels after the SB225002 treatment (Fig. 5C). Therefore, we have demonstrated that the complex of IL-8 and its receptors play a pivotal role in the survival of TamR cells.