MDA-MB-231 (human mammary carcinoma), HCT116 (human colorectal carcinoma), H1299 (human non-small lung carcinoma), A375 (malignant melanoma), and HepG2 (human hepatocellular carcinoma) cell lines were obtained from the ATCC (Manassas, VA, USA). Cancer cells were cultured in high glucose DMEM containing 10% fetal bovine serum (FBS) and antibiotics in the presence of 20% O₂/5% CO₂ (normoxia) or 1% O₂/5% CO₂ (hypoxia). Small interfering RNAs against human IL-11 or STAT3 was transiently transfected into cancer cells using Lipofectamine 2000 (Life Technologies, Carlsbad, CA, USA) according to the manufacturer's instructions, and transfected cells were then incubated for 48 h before being used in experiments.

Small interfering RNAs against IL-11 (sc-39636) or STAT3 (sc-29493) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Recombinant human IL-11 and IL-11 neutralizing antibody were purchased from R&D Systems (Minneapolis, MN, USA). Recombinant human IL-11 was reconstituted in PBS containing 0.5% bovine serum albumin (BSA) before being used in experiments.

Total proteins were extracted in a buffer containing 1% IGEPAL, 150 mM NaCl, 20 mM HEPES (pH 7.9), 10 mM NaF, 0.1 mM EDTA and protease inhibitor cocktail. Total proteins were electrophoresed on SDS-polyacrylamide gels and transferred to Immobilon-P membrane (Millipore, Billerica, MA, USA). The membranes were then incubated with primary antibodies overnight at 4℃, incubated with HRP-conjugated secondary antibodies for 1 h at room temperature, and visualized using an Enhanced Chemiluminescence Detection kit (Pierce, Rockford, IL, USA). Antibodies recognizing STAT3 and pTyr-STAT3 (Tyr705) were purchased from Cell Signaling Technology (Danvers, MA, USA).

Total RNA was isolated with Trizol (Life Technologies), and used for cDNA synthesis using high capacity cDNA reverse transcription kit (Life Technologies). qRT-PCR was performed using SYBR Green PCR Master Mix (Life Technologies). Experimental Ct values were normalized to 36B4, and mRNA expression was calculated relative to 36B4 expression. The primer sequences were as follows: IL-11; 5'-CTCAGCACGACCAGGACC-3' (right) and 5'-GGACAGGGAAGGGTTAAAGG-3' (left), CAIX; 5'-CAGCAACTGCTCATAGGCAC-3' (right) and 5'-CTTTGCCAGAGTTGACGAGG-3' (left), GLUT1; 5'-GGCATTGATGACTCCAGTGTT-3' (right) and 5'-ATGGAGCCCAGCAGCAA-3' (left), VEGF; 5'-AGCTGCGCTGATAGACATCC-3' (right) and 5'-CTACCTCCACCATGCCAAGT-3' (left), LOX; 5'-TGGCAGTCTATGTCTGCACC-3' (right) and 5'-CTATGGCTACCACAGGCGAT-3' (left), MMP-2; 5'-GGAAAGCCAGGATCCATTTT-3' (right) and 5'-ATGCCGCCTTTAACTGGAG-3' (left), MMP-3; 5'-CAATTTCATGAGCAGCAACG-3' (right) and 5'-AGGGATTAATGGAGATGCCC-3' (left), MMP-9; 5'-TTGGTCCACCTGGTTCAACT-3' (right) and 5'-ACGACGTCTTCCAGTACCGA-3' (left), E-cadherin; 5'-GACCGGTGCAATCTTCAAA-3' (right) and 5'-TTGACGCCGAGAGCTACAC-3' (left), N-cadherin; 5'-CCACCTTAAAATCTGCAGGC-3' (right) and 5'-GTGCATGAAGGACAGCCTCT-3' (left), vimentin; 5'-ATTCCACTTTGCGTTCAAGG-3' (right) and 5'-CTTCAGAGAGAGGAAGCCGA-3' (left), fibronectin; 5'-ACCTCGGTGTTGTAAGGTGG-3' (right) and 5'-CCATAAAGGGCAACCAAGAG-3' (left).

To prepare conditioned media, MDA-MB-231 cells were seeded into 100 mm tissue culture dishes and allowed to reach 60% confluence. The medium was then replaced by serum-free Dulbecco's Modified Eagle Media (DMEM) and the cells were cultured under normoxic or hypoxic conditions for 72 h. Next, conditioned medium fractions were collected and centrifuged at 13,000 g at 4℃ for 10 min, and concentrated 10 times using centrifugal filter unit (Millipore). The concentrated conditioned medium was used freshly for in vitro cell migration, invasion assay, or for secreted IL-11 protein measurement. IL-11 protein levels were measured using the human IL-11 Quantikine ELISA kit purchased from R&D Systems according to the manufacturer's instructions. IL-11 level was normalized to the total protein level in each sample.

Cancer cell migration and invasion assays were performed using Transwell chambers purchased from Sigma-Aldrich (St. Louis, MO, USA). For the migration assay, cells in 0.1 ml of FBS-free medium were seeded in the upper chamber and the lower chamber was filled with complete culture medium as a chemotactic agent, and cells were then incubated for 6 h. For invasion assay, cells (3×10³) in 0.1 ml of FBS-free medium were seeded in the upper chamber of an 8-µm Matrigel coated chamber (BD Biosciences, San Jose, CA, USA) and incubated for 24 h. Cells that migrated and invaded were then stained with hematoxylin and eosin. The filter membranes, containing the migrating cells, were then placed on a glass slide and analyzed using an Olympus IX51 microscope (Olympus, Tokyo, Japan).

The data are presented as means and standard deviations. All the experiments were analyzed using Prism 5.0 software (GraphPad Software, Inc., San Diego, CA, USA), and Student t-test was used to compare the differences between the groups, with p<0.05 considered to be statistically significant.

To understand the functional role of IL-11 in hypoxic cancer cells, we confirmed whether IL-11 was elevated in various cancer cell lines. We found that IL-11 as well as hypoxia inducible factor (HIF)-1α target genes were strongly increased in hypoxic MDA-MB-231 cells (Fig. 1A). To examine the time-dependent effects of hypoxia on IL-11 expression, MDA-MB-231 cells were incubated under hypoxic conditions for specific times. Fig. 1B shows that IL-11 mRNA levels increased under hypoxic conditions at the 24-h time point. Because cancer is heterogeneous in nature, different types of cancer cell lines were tested. As expected, IL-11 mRNA levels strongly increased in several cancer cell lines under hypoxic conditions (Fig. 1C). To examine the effect of hypoxia on the production of IL-11, we measured IL-11 protein levels in the culture medium. Consistent with the mRNA levels, levels of IL-11 protein, which is produced and secreted from cancer cells, increased under hypoxic conditions (Fig. 1D). These results suggest that IL-11 production, both at the transcriptional and protein level, increases under hypoxic conditions in several cancer cell lines.

Because IL-11 is thought to be involved in tumor metastasis [8,9,10], we hypothesized that hypoxia-induced IL-11 may contribute to cancer cell migration and invasion. To prove this hypothesis, we tested IL-11 effect on cell migration and invasion in hypoxic cancer cells. Cell migration and invasion strongly increased when the cells were incubated in the presence of conditioned medium collected and 10 times concentrated from cancer cells cultured under hypoxic conditions (hypoxia-conditioned medium; Fig. 2A). In addition, we confirmed that human recombinant IL-11 treatment increased cancer cell migration and invasion under normoxia condition (Fig. 2A). To examine whether hypoxia-induced IL-11 expression contributed to the hypoxic cancer cell invasiveness, we tested the effect of an IL-11 neutralizing antibody on cancer cell invasiveness in the presence of hypoxia-conditioned medium. Fig. 2B shows that IL-11 sequestration by the neutralizing antibody significantly inhibited cancer cell invasion, originally increased by incubation in presence of hypoxia-conditioned medium. IL-11 silencing using a siRNA was performed to determine the role of IL-11 during cell invasion in hypoxic cancer cells. Fig. 2C shows that IL-11 siRNA completely suppressed IL-11 expression levels in hypoxic cancer cells. Consistent with the effect of the IL-11 neutralizing antibody (Fig. 2B), siRNA-mediated IL-11 inhibition also dramatically suppressed the cancer cell invasion induced by culture of the cells in the presence of hypoxia-conditioned medium (Fig. 2D). These results suggest that hypoxia-induced IL-11 expression is required for cancer cell migration and invasion under hypoxic conditions.

To determine the effective concentration of IL-11 required to induce STAT3 phosphorylation, MDA-MB-231 breast cancer cells were stimulated with increasing concentration of IL-11. Fig. 3A indicates that 2 ng/ml of recombinant IL-11 was sufficient to increase STAT3 phosphorylation. Because IL-11 has been shown to be strongly correlated with increased STAT3 activation in human cancers [7,12,13], we tested the effect of IL-11 produced by hypoxic cancer cells on STAT3 phosphorylation and activation. STAT3 phosphorylation was strongly increased by human recombinant IL-11 treatment (Fig. 3B) and by incubation of the cells with conditioned medium derived from hypoxic cancer cells (Fig. 3C). Moreover, hypoxia-conditioned medium, collected from IL-11 suppressed cancer cells, did not increase STAT3 phosphorylation (Fig. 3C). Consistent with this finding, STAT3 phosphorylation increased by hypoxia-conditioned medium treatment was completely abolished by the IL-11 neutralizing antibody (Fig. 3D), suggesting that IL-11 plays an essential role in hypoxia-induced STAT3 phosphorylation. STAT3 activation has been shown to promote tumor progression by transcriptionally regulating the expression of genes related to the extracellular matrix remodeling, such as MMP-2 and MMP-9 [14]. Here, we examined the effect of IL-11 on MMP mRNA expression. Fig. 3E shows that human recombinant IL-11 strongly increased MMP-2, MMP-3, and MMP-9 mRNA levels in highly metastatic breast cancer cells. To examine whether IL-11 is required for MMP expression in hypoxic cancer cells, we measured MMP-2 and MMP-9 mRNA levels in IL-11 suppressed cells. We here found that IL-11 suppression decreases hypoxia-induced MMP-2 and MMP-9 expression (Fig. 3F). Because hypoxia leads to mesenchymal properties through EMT, causing metastasis in several cancer types, we examined whether IL-11 is involved in hypoxia-induced EMT. Fig. 3G shows that recombinant IL-11 strongly reduced epithelial marker, E-cadherin, and increased mesenchymal markers, N-cadherin, and vimentin in MDA-MB-231 breast cancer cells. We additionally found that IL-11 suppression altered EMT-related gene expression under hypoxia (Fig. 3H), indicating that IL-11 is required for EMT in hypoxic cancer cells.

STAT3 silencing using a siRNA significantly decreased MMP-2 and MMP-9, but not MMP-3 mRNA expression in cancer cells cultured under hypoxic conditions (Fig. 4A and B) and in IL-11 treated cancer cells (Fig. 4C and D). Additionally, we tested whether the IL-11-STAT3 pathway caused EMT under hypoxia. Fig. 4E shows that STAT3 silencing significantly increased E-cadherin, but decreased N-cadherin and vimentin, indicating that IL-11 stimulates EMT through the STAT3 pathway in MDA-MB-231 breast cancer cells. Consistently, cell invasion was completely abolished by STAT3 inhibition in cancer cells cultured under hypoxic conditions (Fig. 4F) and in IL-11-treated MDA-MB-231 breast cancer cells (Fig. 4G).