Antiproliferative Action of Conjugated Linoleic Acid on Human MCF-7 Breast Cancer Cells Mediated by Enhancement of Gap Junctional Intercellular Communication through Inactivation of NF-κB

The major conjugated linoleic acid (CLA) isomers, c9,t11-CLA and t10,c12-CLA, have anticancer effects; however, the exact mechanisms underlying these effects are unknown. Evidence suggests that reversal of reduced gap junctional intercellular communication (GJIC) in cancer cells inhibits cell growth and induces cell death. Hence, we determined that CLA isomers enhance GJIC in human MCF-7 breast cancer cells and investigated the underlying molecular mechanisms. The CLA isomers significantly enhanced GJIC of MCF-7 cells at 40 μM concentration, whereas CLA inhibited cell growth and induced caspase-dependent apoptosis. CLA increased connexin43 (Cx43) expression both at the transcriptional and translational levels. CLA inhibited nuclear factor-κB (NF-κB) activity and enhanced reactive oxygen species (ROS) generation. No significant difference was observed in the efficacy of c9,t11-CLA and t10,c12-CLA. These results suggest that the anticancer effect of CLA is associated with upregulation of GJIC mediated by enhanced Cx43 expression through inactivation of NF-κB and generation of ROS in MCF-7 cells.

Gap junctions are transmembrane channels that connect the cytoplasm of neighboring cells composed of two hemichannels and each consists of six connexin proteins [10].

Preparations of CLA Isomers.
The c9,t11-CLA and t10,c12-CLA isomers were isolated from synthetic CLA methyl ester by low-temperature crystallization at −68 and −71 ∘ C in conjunction with urea treatment [30]. The purity of the CLA isomers was 94.5% for c9,t11-CLA and 98.5% for t10,c12-CLA when analyzed by gas chromatography [30]. Each CLA isomer or LA was complexed with fatty acid-free BSA according to the method described by van Greevenbroek et al. [31].

Cell Culture and Sample
Treatments. The MCF-7 human breast cancer cells (Korean Cell Line Bank, Seoul, Republic of Korea) were cultured in RPMI 1640 medium supplemented with 10% FBS and penicillin-streptomycin in a humidified atmosphere with 5% CO 2 at 37 ∘ C as described previously [8]. The cells were grown in Nunc culture dishes (35 mm, i.d., Rochester, NY, USA) to a confluency of 80% and trypsinized with 0.25% trypsin containing 2 mM EDTA to disperse the cells. The cells were subsequently collected by centrifugation (1000 g for 10 min) and resuspended in fresh culture media.
The cells were seeded either into a 96-well culture plate (Nunc) at 1 × 10 4 cells per well for the cell proliferation test using the MTT assay or Nunc culture dish (10 mm, i.d.) at 1 × 10 5 cells for other assays. After 24 h incubation, the cells were serum starved by an additional 24 h incubation in RPMI 1640 medium supplemented with 5 g/mL transferrin, 5 ng/mL sodium selenite, and 0.1 mg/mL BSA.
Two experiments were conducted for the cytotoxicity assay. In the first experiment, the serum-starved MCF-7 cells were incubated with c9,t11-CLA (0-40 M) in RPMI 1640 medium, supplemented with ascorbic acid (50 ng/mL) andtocopherol phosphate (20 ng/mL) for 72 h to determine the optimal concentration and incubation time. In the second experiment, the cytotoxicity of c9,t11-CLA was compared to that of t10,c12-CLA and LA for 48 h using a 40 M concentration. For the other experiments, the serum-starved cells were incubated with 40 M c9,t11-CLA, t10,c12-CLA, and LA for 48 h. After incubation with the CLA isomers, the cells were trypsinized with 0.25% trypsin containing 2 mM EDTA and collected by centrifugation (1000 g for 10 min). These cells were washed with PBS and collected again by centrifugation before assay. Fresh medium was replaced in every 48 h. The exponentially growing cells were used throughout the experiments.

Cell Viability
Assay. MCF-7 cell proliferation was measured using the MTT assay as described previously [29]. Briefly, the cells treated with the CLA isomers were exposed to MTT solution (5 mg/mL of PBS) for 4 h. The MTT solution Evidence-Based Complementary and Alternative Medicine 3 was removed, and 200 L DMSO was added to each well and mixed to dissolve the MTT formazan crystals formed by the viable cells. The absorbance of each well was measured at 570 nm using an Anthos 2020 microplate reader (Wals, Austria). Data are represented as relative cell viability against the cells at time 0.

Flow Cytometry Analysis.
The cell cycle was analyzed as described previously [32]. Briefly, the cells treated with the CLA isomers were collected in 1 mL ice-cold PBS and fixed with 4 mL 70% ethanol for 30 min at 4 ∘ C. The fixed cells were washed with PBS and then incubated with 500 L PI solution (1 mg/mL of RNAase, 50 g/mL of PI, and 0.1% TritonX-100) for 20 min at 37 ∘ C. Flow cytometric analysis was performed using a BD FACSCalibur Flow Cytometer (BD Science, San Jose, CA, USA) equipped with CellQuest Pro Software (BD Science).

Hoechst 33258
Staining. The cells treated with the CLA isomers were fixed with 4% paraformaldehyde in PBS for 10 min and permeabilized in PBS containing 0.1% Triton X-100 for 10 min, after which the nuclei were stained with Hoechst dye (10 g/mL) for 10 min as described previously [8,9]. Finally, the cells were observed under an Olympus IX70 inverted fluorescent microscope (Okaya, Japan).

GJIC Assay.
GJIC of cells treated with CLA isomers was measured using the scrape loading/dye transfer (SL/DT) technique as previously described [29]. Briefly, cells were treated with 0.05% LY solution and scraped with a surgical steel scalpel blade at low light intensity. After 3 min of incubation (5% CO 2 , 37 ∘ C), the cells were washed four times with 2 mL PBS and fixed with 4% paraformaldehyde. The number of dye-communicating cells perpendicular to the scraped-line was counted under an Olympus IX70 inverted fluorescent microscope.

RNA Extraction and Reverse Transcription Polymerase
Chain Reaction (RT-PCR) for Cx43 Gene Expression. Total RNA of cells treated with the CLA isomers was isolated using Trizol reagent according to the method described by the manufacturer. RT-PCR analysis was performed as described previously [33]. cDNA was synthesized from total RNA using an Omniscript Reverse Transcriptase kit according to the manufacturer's instructions. The Cx43 gene was amplified by RT-PCR using 5 -ACATCAGGTGGACTGTTTCCT-3 (sense) and 5 -ACGACTGCTGGCTCTGCTT-3 (antisense) as primers in a 50 L final volume containing Super-Therm DNA polymerase, 1.5 mM MgCl 2 , and 2 mM dNTP for 30 cycles (denaturation at 94 ∘ C for 30 seconds, annealing at 50 ∘ C for 1 min, and extension at 72 ∘ C for 1 min, followed by final incubation at 72 ∘ C for 7 min). -actin primers 5 -CGTGACATCAAGGAGAAGCT-3 (sense) and 5 -ATC-CACATCTGCTGGAAGGT-3 (antisense) were used as the control under the same PCR conditions. PCR products were analyzed by electrophoresis on 1% agarose gels.

Western Blot Analysis.
Western blot analysis was performed as described previously [29]. Cells treated with CLA isomers were lysed at 4 ∘ C by shaking for 15 min followed by homogenization with RIPA buffer. After centrifugation at 13,000 g for 15 min, the supernatant was collected and stored at −70 ∘ C until use. Protein concentration was determined using the Bradford reagent (Hercules, CA, USA). Proteins were separated by 12.5% SDS-polyacrylamide gel electrophoresis, and Western blot analysis of Cx43, caspase-3, and NF-B was performed according to the manufacturer's instructions. Protein bands were detected using an Enhanced Chemiluminescence Detection kit (Thermo Scientific, Rockford, IL, USA).

Preparation of Nuclear Protein for Western Blotting of NF-B.
Proteins in the nuclear fraction of cells were isolated as described previously [34]. Cells treated with CLA isomers were incubated for 20 min at 48 ∘ C in a buffer containing 20 mM N-(2-hydroxyethyl)-piperazine-N -(2-ethanesulfonic acid) (HEPES) KOH (pH 7.9), 25% glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dichlorodiphenyltrichloroethane (DTT), and 0.2 mM PMSF. Cell nuclei were collected as a pellet by centrifugation at 14,000 g for 5 min at 48 ∘ C. Nuclear proteins were extracted at 4 ∘ C by gently resuspending the nuclear pellets in a buffer solution containing 20 mM Tris (pH 7.5), 20% glycerol, 1.5 mM MgCl 2 , 420 mM NaCl, 0.2 mM EDTA, and 0.1% Triton X-100, followed by a 1 h incubation at 4 ∘ C with occasional vortexing. After centrifugation at 14,000 g for 15 min at 4 ∘ C, the supernatant, which contained the nuclear protein fraction, was collected. Protein concentration was determined using the Bradford reagent.

Immunocytochemistry Assay of Cx43
Protein. The cells treated with CLA isomers were seeded and cultured on chamber slides for immunocytochemistry analysis as described previously [35]. Cells were fixed with 4% paraformaldehyde on chamber slides. The fixed cells were permeabilized with 0.5% Triton X-100 in PBS and incubated with a blocking solution (PBS containing 2% BSA) at room temperature for 30 min. The cells were then incubated with anti-connexin 43 IgM as a primary antibody and FITC-conjugated goat-antimouse IgM as the secondary antibody. Fluorescent staining of Cx43 was imaged using an Olympus IX70 inverted fluorescent microscope.

Intracellular Reactive Oxygen Species (ROS) Assay.
Intracellular ROS positive cells were determined as described previously [36]. Cells treated with CLA isomers were incubated with 50 M DCFH-DA for 30 min at 37 ∘ C in the dark. After the incubation, the cells were washed with Locke's buffer (154 mM NaCl, 25 mM KCl, 2.3 mM CaCl 2 , 3.6 mM NaHCO 3 , 8.6 mM HEPES, and 5.6 mM glucose, pH 7.4). The number of ROS positive cells was measured using an Olympus FV-1000 confocal microscope equipped with an argon laser at 485 nm excitation and 530 nm emission wavelengths. variance followed by Duncan's multiple range test. Differences were considered significant at < 0.05.

Growth Inhibition of MCF-7 Cells.
First, we determined the optimal concentration of c9,t11-CLA that showed an anticancer effect on MCF-7 cells. The MTT test was performed on cells grown at various concentrations (5-40 M) of c9,t11-CLA for 72 h (Figure 1(a)). The growth inhibitory effect of c9,t11-CLA was dose and time dependent. At 40 M, cell viability was 70.6 and 65.9% at 48 and 72 h, respectively ( < 0.05).
The inhibitory efficacy of c9,t11-CLA was compared to that of either t10,c12-CLA or LA at a concentration of 40 M for 48 h (Figure 1(b)). t10,c12-CLA and LA significantly inhibited cell growth by 32.6% and 3.8%, respectively, as compared to that in the control. The potency of c9,t11-CLA and t10,c12-CLA was similar. Hence, 40 M c9,t11-CLA, t10,c12-CLA, and LA for 48 h incubation were selected as the optimal conditions for subsequent experiments.

Induction of Apoptosis in MCF-7 Cells.
We evaluated whether growth inhibition was related to apoptosis of the cells. The effect of c9,t11-CLA and t10,c12-CLA isomers on the cell cycle and apoptotic parameters was evaluated in MCF-7 cells (Figure 2(a)). The cell cycle analysis revealed that c9,t11-CLA and t10,c12-CLA significantly increased the percentage of cells in the sub-G1 phase to 35% and 33.6%, respectively, which was an indicator of DNA fragmentation due to cell death (Figure 2(b)). No significant difference in the sub-G1 phase cell population was found between cells treated with the c9,t11-CLA and t10,c12-CLA isomers.
To determine whether cell death was related to apoptosis, we further performed Hoechst 33258 staining (Figure 2(c)). As a result, cell shrinkage and nuclear condensation were visible in cells treated with c9,t11-CLA and t10,c12-CLA, which indicated apoptosis. To verify whether the apoptosis was caspase dependent, we assessed the level of caspase-3 protein, a key executionary protease, in the process. Western blotting revealed that the c9,t11-CLA and t10,c12-CLA isomers significantly increased caspase-3 levels (Figure 2(d)). Taken together, these results indicate that c9,t11-CLA and t10,c12-CLA induced cell death by inducing apoptosis.

Enhancement of GJIC in MCF-7 Cells.
Next, we determined whether the CLA isomers could reverse the reduced GJIC of MCF-7 cells. We evaluated the status of GJIC in MCF-7 cells treated with c9,t11-CLA, t10,c12-CLA, and LA ( Figures  3(a) and 3(b)). CLA isomers and LA increased GJIC, relative to control, but the efficacy of the c9,t11-CLA and t10,c12-CLA isomers was similar and greater than that of LA. This finding suggests that the inhibition of cell growth might be associated with increased GJIC.

Increased Cx43 Gene Expression in MCF-7 Cells.
Cx43 is a major protein in the gap junction channel that regulates GJIC. Previous results have shown that some chemical compounds upregulate GJIC by upregulating Cx43 expression in human cancer cells [32,37,38]. Thus, we assessed the expression levels of the Cx43 gene at the transcriptional and translational levels to investigate the mechanism of GJIC restoration by the CLA isomers ( Figure 4). The CLA isomers upregulated the expression of the Cx43 gene at the transcriptional (Figure 4(a)) and translational levels (Figures 4(b) and 4(c)). Both c9,t11-CLA and t10,c12-CLA significantly enhanced the level of Cx43 mRNA in MCF-7 cells. No significant difference was observed in the upregulated expression of Cx43 mRNA between c9,t11-CLA and t10,c12-CLA. A similar pattern was observed for Cx43 protein expression. These results indicate that both the c9,t11-CLA and t10,c12-CLA isomers equally enhanced Cx43 expression. Immunofluorescence staining was performed to determine the distribution and expression levels of Cx43 in the cell membranes and to further validate the influence of the CLA isomers on Cx43 expression (Figure 4(c)).

Generation of Intracellular ROS in MCF-7 Cells.
There is evidence that elevated ROS depolarizes the cell membrane and opens connexin hemichannels to increase connexin gene expression, resulting in increased cell death [39]. Hence, we evaluated intracellular ROS generation in MCF-7 cells after CLA treatment. The effect of c9,t11-CLA and t10,c12-CLA isomers on ROS generation was measured by DCFH-DA dye method at 40 M concentration for 48 h (Figures 5(a) and 5(b)). The levels of DCF fluorescence in the cells treated with c9,t11-CLA and t10,c12-CLA increased significantly relative to the control and LA-treated cells. No significant difference in ROS accumulation was found between cells treated with c9,t11-CLA and t10,c12-CLA. These results indicate that the CLA isomers induced the formation of ROS. [40,41]. To analyze if the effects of the CLA isomers on inhibited cell growth are related to NF-B activity, the expression of NF-B in MCF-7 cells treated with the c9,t11-CLA and t10,c12-CLA isomers was determined by Western blotting (Figures 6(a) and 6(b)). The c9,t11-CLA and t10,c12-CLA isomers significantly decreased the level of the NF-B nuclear fraction relative to the control and LA-treated cells. These results indicate that inhibiting NF-B activity might be involved in the inhibition of MCF-7 cell growth by these CLA isomers.

Discussion
We showed previously that both c9,t11-CLA and t10,c12-CLA isomers inhibit the growth of human MCF-7 breast cancer cells by inducing apoptosis [8]; however, the exact mechanism was not clearly elucidated. It is evident that enhancing GJIC through Cx43 upregulation in tumor cells is positively correlated with anticancer activities by inhibiting the growth of cells through the induction of apoptosis [32,37,38,42,43]. Hence, the main questions addressed by these studies are whether the growth inhibitory effect of c9,t11-CLA and t10,c12-CLA isomers on MCF-7 cells, which reduces GJIC, is mediated by enhanced GJIC to induce apoptosis and whether the effects are associated with upregulation of Cx43.
Apoptosis is one of the most common mechanistic actions of anticarcinogenic CLA isomers in animal and cancer cells [7][8][9]. The efficacy of individual CLA isomers on the induction of apoptosis is partly mediated by mitochondrial dysfunctional apoptosis, but the mechanism of action is completely unknown. Rakib et al. [13,29] demonstrated an antipromotional action of c9,t11-CLA, t10,c12-CLA, and t,t-CLA isomers mediated by the inhibition of downregulation of GJIC in human MCF-10A cells, a nontumorigenic mammary epithelial cell line, treated with TPA, and they suggested that CLA isomers protected phosphorylation of Cx43 to maintain GJIC. These results suggest that CLA isomers could restore the down-regulated GJIC in cancer cells, similar to MCF-7 cells. This hypothesis was solved in the present study as the c9,t11-CLA and t10,c12-CLA isomers induced apoptosis by enhancing GJIC through Cx43 gene expression (Figures 3 and  4). Our results agree with reports that anticarcinogens, such as caffeic acid phenethyl ester, coleusin factor, tellimagrandin, and fucoxanthin, induce apoptosis via enhanced GJIC with elevated Cx expression [26,32,37,38].
ROS are elevated in apoptotic cancer cells induced by CLA isomers [44], whereas NF-B activation is inhibited [45]. Hence, it is of significance to elucidate that the elevated GJIC through Cx43 protein expression in MCF-7 cells by c9,t11-CLA and t10,c12-CLA isomers (Figure 4) is associated with enhanced ROS levels ( Figure 5) and/or reduced NF-B activity ( Figure 6). Many chemical compounds that exhibit anticarcinogenic activity can reduce NF-B activity and elevates ROS level in cancer cells. For example, anticarcinogenic niclosamide reduces NF-B activity and elevates ROS concentration in acute myelogenous leukemia stem cells [34]. Evidently, the loss of NF-B results in the accumulation of ROS because NF-B activity suppresses the expression of some antioxidant genes [46][47][48]. The elevated ROS, which decreases the fluidity of the biological membranes and increases membrane permeability, depolarizes the cell membrane, opens connexin hemichannels, and expresses connexin genes, resulting in increased cell death [39]. Moreover, the NF-B inhibitor BAY11-7082 stimulates Cx43 gene expression and restores GJIC in the Toll-like receptor (TLR3) ligand polyinosinic cytidylic acid-induced downregulation of Cx43 gene expression and GJIC [49]. These results suggest that compounds exhibiting anticarcinogenic action produce ROS to induce Cx43 gene expression and enhance GJIC in  cancer cells. Taken together, the apoptosis induced by the c9,t11-CLA and t10,c12-CLA isomers in MCF-7 cells was associated with upregulation of GJIC through Cx43 expression mediated by inactivation of NF-B and production of ROS as shown in Figure 7.
One of the known mechanisms of restoring GJIC is the upregulation of connexin gene expression. Down-regulated connexin gene expression in many cancer cells is related to impaired Cx43 expression, including transcriptional silencing through DNA methylation at the connexin gene promoter sites [50]; lower concentrations of cAMP or estrogen [51]; and instability of mRNA and the Cx43 protein [52]. Although further studies are needed to clarify these mechanistic issues involved in the CLA isomers regarding expression of Cx43 in MCF-7 cells, we found that both the c9,t11-CLA and t10,c12-CLA isomers clearly reversed the reduced GJIC in MCF-7 cells by upregulating Cx43 genes at the transcriptional and translational levels ( Figure 4). This finding was consistent with the result that fucoxanthin [32] and quinoline [53] compounds upregulate GJIC and Cx43 expression in human SK-HP-1 hepatoma cells and human T47D breast cancer cells in relation to growth inhibition. The open hemichannels allow direct entry of intracellular ROS in connexinexpressing cells leading to cancer cell death by altering cell survival mechanisms, including mitochondrial dysfunction [54]. In the present study, increased intracellular ROS were observed in the CLA isomer-treated MCF-7 cells along with increased DNA fragmentation, DNA condensation, and caspase-3 expression (Figure 2). Consequently, the influx of ROS into the cells through the open hemichannels might be involved in the caspase-dependent MCF-7 cell death.
It has been revealed that the antiproliferative activity of individual CLA isomers is dependent upon cancer cells. The anticarcinogenic efficacies of t10,c12-CLA and c9,t11-CLA isomers are nearly identical in MG-63 osteosarcoma cells and human MCF-7 breast cancer cells [8,9], but different efficacies are observed in HT-29 colorectal cancer and PC-3 prostate cancer cells with the stronger potential of t10,c12-CLA [55,56]. In the present study, we found no differences in efficacy of c9,t11-CLA and t10,c12-CLA to inhibit proliferation of MCF-7 cells with enhanced GJIC by inactivating NF-B and reducing ROS. These results might be partly due to the similar structural configuration of the c9,t11-CLA and t10,c12-CLA isomers. Incorporation of these CLA isomers in the plasma membrane of MCF-7 cells is believed to be one of the factors dictating their similar anticancer activity, including Cx43 gene expression and inactivation of NF-B, which affect the distribution and function of several membrane raft associated proteins [57,58]. Furthermore, these CLA isomers may compete as ligands for the NF-B transcription factor in cancer cells [34], and hence, the similar effects of c9,t11-CLA and t10,c12-CLA isomers on inhibiting MCF-7 cell proliferation might be due to similar binding affinities of the CLA isomers to NF-B expressed in the cells. Further studies are needed to clarify these issues.
In conclusion, both c9,t11-CLA and t10,c12-CLA isomers effectively inhibited growth of MCF-7 cells, and this effect was associated with enhanced GJIC through the upregulation of Cx43 expression in conjunction with inactivation of NF-B. Our results support the hypothesis that CLA isomers upregulate reduced GJIC, which plays a role in propagating cell death as a result of upregulated caspase-3 activity.