ErbB2 activation upregulates glutaminase 1 expression which promotes breast cancer cell proliferation

Abstract

Active glutamine utilization is critical for tumor cell proliferation. Glutaminolysis represents the first and rate-limiting step of glutamine utilization and is catalyzed by glutaminase (GLS). Activation of ErbB2 is one of the major causes of breast cancers, the second most common cause of death for women in many countries. However, it remains unclear whether ErbB2 signaling affects glutaminase expression in breast cancer cells. In this study, we show that MCF10A-NeuT cell line has higher GLS1 expression at both mRNA and protein levels than its parental line MCF10A, and knockdown of ErbB2 decreases GLS1 expression in MCF10A-NeuT cells. We further show that in these cells, ErbB2-mediated upregulation of GLS1 is not correlated to c-Myc expression. Moreover, activation of neither PI3K-Akt nor MAPK pathway is sufficient to upregulate GLS1 expression. Interestingly, inhibition of NF-κB blocks ErbB2-stimulated GLS1 expression, whereas stimulation of NF-κB is sufficient to enhance GLS1 levels in MCF10A cells, suggesting a PI3K-Akt-independent activation of NF-κB upregulates GLS1 in ErbB2-positive breast cancer cells. Finally, knockdown or inhibition of GLS1 significantly decreased the proliferation of breast cancer cells with high GLS1 levels. Taken together, our data indicate that ErbB2 activation promotes GLS1 expression via a PI3K-Akt-independent NF-κB pathway in breast cancer cells, identifying another oncogenic signaling pathway which stimulates GLS1 expression, and thus promoting glutamine utilization in cancer cells. These findings, if validated by in vivo model, may facilitate the identification of novel biochemical targets for cancer prevention and therapy.

Keywords: BREAST CANCER; ErbB2; GLS1; GLUTAMINASE; GLUTAMINE; NUCLEAR FACTOR κ B.

Fig. 1

ErbB2 activation up-regulates GLS1 expression at both mRNA and protein levels. A: Western blotting analysis of total cell lysates prepared from MCF10A-NeuT and its parental cell line, MCF10A. Both ErbB2 and GLS1 were detected, and α-tubulin was used as a loading control. B: Quantitative analysis of GLS mRNA levels by real time PCR. The total RNA samples were isolated from MCF10A and MCF10A-NeuT cells and reverse-transcribed to cDNA. GLS1 and GLS2 expression at mRNA level was determined by qRT-PCR (ΔRQ). Error bars shown are standard deviations; *P < 0.05. C: ErbB2 knockdown reduces GLS1 levels in MCF10A-NeuT cells. Control and ErbB2 specific siRNA were transfected into MCF10A-NeuT cells. Twenty-four hours post transfection cell lysates were prepared and analyzed by Western blots. D: GLS1 expression after EGF treatment. MCF10A were treated with indicated doses of EGF. Cells were harvested 6 h post treatment. GLS1 was determined by Western blot and pErk was used as an indicator of the activation of ErbB2 signaling pathway.

Fig. 2

ErbB2 enhanced GLS1 expression does not involve c-Myc. A: ErbB2 and its downstream signaling pathways. MAPK/Erk and PI3K/Akt pathways are two key signaling pathways activated by ErbB2. ErbB2 may induce c-Myc expression through either MAPK/Erk, or PI3K/Akt or both. In addition, ErbB2 activation stimulates NF-κB pathway through induction of IKK, which is independent of PI3K. Inhibitors and activators of the signaling pathways studied are shown in the diagram. PD98059 is an MAPK/Erk inhibitor; LY294002 is an Akt inhibitor; BAY 11-7082 is an NF-κB inhibitor; and PMA is an NF-κB activator. (B) ErbB2 does not upregulate c-Myc expression in MCF10A-NeuT cells. The protein levels of ErbB2, GLS1 and c-Myc in MCF10A and MCF10A-NeuT cells were determined by Western blotting analysis. C: Stable overexpression c-Myc in MCF10A cells fails to upregulate GLS1. D: c-Myc knockdown does not affect GLS1 expression in MCF10A-NeuT cells.

Fig. 3

Activation of PI3K-Akt or MAPK/Erk is not sufficient to enhance GLS1 expression. A: GLS1 expression in MCF10A cells transiently transfected with empty vector, and plasmids expressing MEK or Myr-Akt. Western blots show that the activation of either MAPK/Erk or Akt pathway is not sufficient to upregulate GLS1. B: Inhibition of PI3K pathway does not suppress GLS1 expression. MCF10A-NeuT cells were treated with 0, 5, 10, and 20mM of LY294002 for 24 h prior to cell harvest and Western blotting analysis. Phosporylated Akt was detected as an indicator of PI3K pathway inhibition. C: MCF10A-NeuT cells were treated with 0, 25, 50, and 100 mM of PD98059 for 24 h. Both Erk activation and GLS1 expression is suppressed.

Fig. 4

NF-κB pathway is involved in GLS1 expression in MCF10A-NeuT cells. A: MCF10A and MCF10A-NeuT cells were treated with 0, 5, 10, and 20mM of BAY 11-7082 (BAY) for 24 h, GLS1 expression was determined by Western blotting analysis. B: Knockdown of p65 down-regulates GLS1 in MCF10A-NeuT cells. C: Nuclear localization of p65 in MCF10A-NeuT cells. MCF10A and MCF10A-NeuT cells were cultured in eight well chamber slides, fixed and stained for p65, ErbB2, and DNA (DAPI).

Fig. 5

Activation of p65 is sufficient to upregulate GLS1 in MCF10A cells. A: MCF10A cells were treated with 100 nM of PMA in DMSO or DMSO alone for 6 h, followed by immunofluorescent staining for p65. B: MCF10A cells were treated with 20 ng/ml of TNF-α in PBS or PBS alone for 6 h, followed by immunofluorescent staining for p65. C: MCF10A cells were treated with 0, 50, or 100nM of PMA in DMSO for 6 h. GLS1 expression was determined by Western blots. D: MCF10A cells were treated with 0, 10, and 20 ng/ml of TNF-α for 6 h, GLS1 expression was determined by Western blots. E: MCF10A cells were treated with 10mMof BAY 11-7082 for 18 h, followed by treatment with 10 ng/ml of TNF-α or 100 nM of PMA for another 6 h. The whole cell lysates were collected for Western blotting analysis. The numbers under the blots are quantification of each detected bands; GLS1 or α-tubulin of control groups were set at 1.00, and treated groups were compared to the control group.

Fig. 6

Suppressing ErbB2 down regulates GLS1 expression in human breast cancer cells. A: Knockdown of ErbB2 down-regulates GLS1 levels in SK-BR-3 and MDA-MB-453 cells. Twenty-four hours after transfection, protein levels of ErbB2 and GLS1 in SK-BR-3 and MDA-MB-453 cells were examined by Western blots. B: Trastuzumab represses the expression of GLS1. SK-BR-3 and MDA-MB-453 cells were treated with trastuzumab (50 or 100 mg/ml) for 24 h. Protein levels of ErbB2 and GLS1 were determined by Western blotting. The numbers under the blots are quantification of each detected bands; ErbB2, GLS1, or α-tubulin of control group were set at 1.00, treated groups were compared to the control group. C,D: Inhibition or knockdown of p65 reduces GLS1 expression in ErbB2-positive breast cancer cells. NF-κB activity in SK-BR-3 and MDA-MB-453 cells were inhibited by treat cells with 5 and 10mM BAY 11-7082, or by siRNA knockdown. GLS1 expression levels and the p65 knockdown effects were monitored by Western blotting analysis.

Fig. 7

GLS1 knockdown attenuates the proliferation rates of MCF10A-NeuT, SK-BR-3, and MDA-MB-453 cells. A–C: Western blots show the knockdown effects of two different GLS1 siRNAs in MCF10A-NeuT (A), SK-BR-3 (B), and MDA-MB-453 (C) cells (left panels). Cell proliferation curves of MCF10A-NeuT, SK-BR-3, and MDA-MB-453 cells are shown in the right panels. *P < 0.05 for GLS1 siRNA1 versus control siRNA group; **P < 0.05 for GLS1 siRNA2 versus control siRNA groups.

Fig. 8

Small molecule compound BPTES suppresses the proliferation of ErbB2-positive human breast cancer cells. A: Schematic illustration of BPTES treatment. BPTES concentrations used are: 0, 5, 10, and 20mM. Treatment: change of media containing BPTES with indicated concentrations. Test: cell number determination. B–D: Effects of BPTES on the proliferation rates of MCF10A-NeuT cells (B), SK-BR-3 cells (C), and MDA-MB-453 cells (D).