Human breast cancer cell line MCF7 was obtained from the Cell Bank of Chinese Academy of Sciences, Shanghai, China. SK-BR-3 was a gift from Professor Shengyong Yang of Sichuan University. MCF7 cells were cultured in DMEM/F12 (Hyclone) with 10% fetal bovine serum (FBS, Biological Industries) and penicillin/streptomycin (PS, Hyclone). SK-BR-3 cells were maintained in DMEM (Hyclone) supplemented with 10% FBS and PS. Cells were cultured at 37°C in a 5% CO₂ incubator.

SK-BR-3 cells (1×10⁶ cells/well) were plated in 6-well plates for 24 h and DMSO or cerulenin (10 µg/ml, Cayman) was added into the medium. Twenty-four hours after treatment, media was collected for glucose uptake and lactate production assays and cells were trypsinized for Transwell and western blot assays. MCF7 cells (8×10⁵ cells/well) were plated in 6-well plates for 24 h. The medium was changed to 0.5% FBS medium and cells were starved for 24 h. Cells were treated with vehicle, heregulin-β1 (HRG-β1, 100 ng/ml, PeproTech), HRG-β1 plus cerulenin (10 µg/ml), 2-deoxyglucose (2-DG, 0.5 mM, Sigma) or oxamate (OX, 50 mM, Sigma). 24 h after treatment, media was collected for glucose uptake and lactate production assays and cells were trypsinized for Transwell and western blot assays.

Glucose uptake was measured using Glucose Oxidase Assay kit (Applygen, China) and results were normalized to the total cellular protein amounts. Lactate production in the medium was performed by using Lactate Assay kit (Nanjing Jiancheng Bioengineering Institute, China) according to the manufacturer's protocol. Results were normalized to the total cellular protein amounts.

Cells were harvested and lysed in a buffer containing 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 2 mM EDTA, 1% Triton, 1 mM PMSF and Protease Inhibitor Cocktail (Sigma) for 20 min on ice. Lysates were cleared by centrifugation at 15,000 g at 4°C for 20 min. Supernatants were collected and protein concentrations were determined by Bradford assay (Bio-Rad). Proteins were subjected to electrophoresis on SDS-polyacrylamide gels and were then transferred to PVDF membranes (Bio-Rad). Membranes were blocked in Tris buffered saline (TBS) (pH 7.4) with 5% non-fat dry milk and probed with primary antibodies as indicated in the figure legends. The following antibodies were used: FASN (sc-55580, Santa Cruz), ErbB2 (2156, Cell Signaling Technology), LDHA (3582, Cell Signaling Technology), β-actin (A2228, Sigma), glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (AG019, Beyotime, China). Immunoreactive bands were visualized by horseradish peroxidase (HRP)-conjugated secondary antibodies (Bio-Rad) using ECL Western Blotting Substrate (Pierce). The fold expression of protein was quantified using Image Lab software. Protein expression in control group was set as 1.0.

pcDNA3.0 was purchased from Hangzhou Baosai Biotechnology, China and pcDNA3.0/ErbB2 was purchased from Addgene. Transient transfection was performed using the Lipofectamine 2000 transfection reagent (Invitrogen) according to the manufacturer's protocol. Forty-eight hours after transfection, cells were trypsinized for Transwell and western blot assays, and culture media was collected for glucose uptake and lactate production assays.

MCF7 and SK-BR-3 cells were plated in 6-well plates and allowed to grow to confluence. Wounds were made using a 10 µl pipette tip and indicated reagents were added to the cells. Wound closure was monitored over a 24 or 48 h period and photographs were taken at 0, 24 or 48 h after wounding. Wound width at 0, 24 or 48 h was measured and the results were expressed as wound closure (%) = (width at 0 h − width at × h)/width at 0 h × 100% (x=24 or 48).

Transwell migration assay was performed using Falcon cell culture inserts (Corning). Cells were trypsinized and suspended in medium with 0.5% FBS. MCF7 cells (3×10⁵) or SK-BR-3 cells (1×10⁵) were seeded in the top of the inserts and placed in the wells containing medium with 10% FBS. After 48 h incubation at 37°C, the inserts were washed with PBS twice and cells on the upper surface of the inserts were gently removed with a cotton swab. The inserts were fixed with 10% methanol, rinsed with ultra-pure water, and stained with 0.1% crystal violet for 20 min. Cells that migrated onto the lower surface were counted under a microscope in five random fields.

All experiments were repeated three times. Statistical analysis was performed using GraphPad Prism version 5.0. The data were expressed as means ± SE. The difference between groups was detected using Student's t-test. A value of P<0.05 was considered to indicate a statistically significant difference.

First, we demonstrated that MCF7 cells displayed low levels of FASN and ErbB2 and SK-BR-3 cells had high levels of FASN and ErbB2 (Fig. 1A). Then, we compared LDHA protein levels between these two cell lines. We found that SK-BR-3 cells showed much higher LDHA protein levels than MCF7 cells (Fig. 1A). Next, glucose uptake and lactate production, which are two important markers of glycolysis, were measured and compared between MCF7 and SK-BR-3 cells. SK-BR-3 cells showed a significantly higher glucose uptake and lactate production than MCF7 cells (Fig. 1B). Finally, we compared the migratory potential between MCF7 and SK-BR-3 cells using wound healing assay and Transwell assay. Compared to MCF7 cells, SK-BR-3 cells showed a higher percentage of wound closure (Fig. 1C) and migrated cell numbers (Fig. 1D), indicating that SK-BR-3 cells had a greater basal migration capacity than MCF7 cells. These results suggest a consistent relationship between FASN, ErbB2, glycolysis, and cell migration, indicating that increased glycolysis predicts higher migratory potential in breast cancer cells.

To investigate the role of FASN in glycolysis and cell migration, we treated SK-BR-3 cells with cerulenin, a specific inhibitor of FASN, and measured its effects on ErbB2 and LDHA protein levels, glycolysis, and cell migration. When FASN protein levels were inhibited by cerulenin, both ErbB2 and LDHA proteins levels were downregulated (Fig. 2A). Cerulenin treatment in SK-BR-3 cells also significantly decreased glucose uptake and lactate production (Fig. 2B), indicating that inhibition of FASN by cerulenin inhibits glycol-ysis. We further found that cerulenin significantly decreased migratory capacity of SK-BR-3 cells (Fig. 2C and D). These results suggest FASN may play an important role in glycolysis and migration of breast cancer cells.

To investigate the role of ErbB2-mediated glycolysis in cell migration, we transiently transfected ErbB2 plasmid into MCF7 cells and then detected FASN and LDHA protein levels. We found that overexpression of ErbB2 upregulates FASN and LDHA protein levels (Fig. 3A). Moreover, we found that acute overexpression of ErbB2 increased both glucose uptake and lactate production (Fig. 3B), indicating increased glycolysis. Then, the role of ErbB2-mediated glycolysis in cell migration was investigated, and we found that overexpression of ErbB2 greatly enhanced cell migration (Fig. 3C). These results suggest ErbB2 plays an important role in glycolysis and migration of breast cancer cells.

We treated MCF7 cells with HRG-β1 and detected FASN, ErbB2, and LDHA protein levels. We found that HRG-β1 upregulated FASN, ErbB2 and LDHA protein levels whereas cerulenin reversed such upregulation (Fig. 4A). We further found that HRG-β1 induced both glucose uptake and lactate production with cerulenin reversing such an induction in MCF7 cells (Fig. 4B). Moreover, HRG-β1 treatment in MCF7 cells greatly increased wound closure and number of migrated cells, while cerulenin overcame such increases (Fig. 4C and D). This result suggests that FASN plays an important role in HRG-β1-indcued glycolysis and migration.

In order to further demonstrate that glycolysis is required for cell migration, we treated MCF7 cells with HRG-β1 and detected LDHA protein levels. We found that HRG-β1 greatly upregulated LDHA protein levels while both 2-DG and OX reversed the upregulation (Fig. 5A). Moreover, both 2-DG and OX reversed HRG-β1-induced glycolysis (Fig. 5B) and cell migration (Fig. 5C and D). These results suggest that HRG-β1-induced glycolysis promotes cell migration whereas glycolysis inhibition by 2-DG or OX reverses such effects.