Twist promotes reprogramming of glucose metabolism in breast cancer cells through PI3K/AKT and p53 signaling pathways

Abstract

Twist, a key regulator of epithelial-mesenchymal transition (EMT), plays an important role in the development of a tumorigenic phenotype. Energy metabolism reprogramming (EMR), a newly discovered hallmark of cancer cells, potentiates cancer cell proliferation, survival, and invasion. Currently little is known about the effects of Twist on tumor EMR. In this study, we found that glucose consumption and lactate production were increased and mitochondrial mass was decreased in Twist-overexpressing MCF10A mammary epithelial cells compared with vector-expressing MCF10A cells. Moreover, these Twist-induced phenotypic changes were augmented by hypoxia. The expression of some glucose metabolism-related genes such as PKM2, LDHA, and G6PD was also found to be upregulated. Mechanistically, activated β1-integrin/FAK/PI3K/AKT/mTOR and suppressed P53 signaling were responsible for the observed EMR. Knockdown of Twist reversed the effects of Twist on EMR in Twist-overexpressing MCF10A cells and Twist-positive breast cancer cells. Furthermore, blockage of the β1-integrin/FAK/PI3K/AKT/mTOR pathway by siRNA or specific chemical inhibitors, or rescue of p53 activation can partially reverse the switch of glucose metabolism and inhibit the migration of Twist-overexpressing MCF10A cells and Twist-positive breast cancer cells. Thus, our data suggest that Twist promotes reprogramming of glucose metabolism in MCF10A-Twist cells and Twist-positive breast cancer cells via activation of the β1-integrin/FAK/PI3K/AKT/mTOR pathway and inhibition of the p53 pathway. Our study provides new insight into EMR.

Keywords: PI3K/AKT; Twist; glucose metabolism; p53.

Figure 1. The altered energy metabolic phenotype in Twist-overexpressing MCF10A cells

A, B. Glucose consumption and lactate production were measured in MCF10A-Vector and MCF10A-Twist cells (MCF10A-Twist cells versus MCF10A-Vector cells. #P < 0.05, under normal oxygen condition; *P < 0.05, under hypoxia condition). C. Fluorescence microscope analysis of mitochondrial mass in MCF10A-Vector and MCF10A-Twist cells after Mito-Tracker Green staining (Magnification, x200. Scale bars, 100 μm). D. Mitochondrial morphological analysis in MCF10A-Vector and MCF10A-Twist cells by transmission electron microscope (Magnification, x25000. Scale bars, 0.5 μm).

Figure 2. Loss of Twist expression reverses the altered energy metabolic phenotype in MCF10A-Twist cells

A, B. Glucose consumption and lactate production were measured in MCF10A-Twist-sh-Ctrl and MCF10A-Twist-sh-Twist cells (MCF10A-Twist-sh-Twist cells versus MCF10A-Twist-sh-Ctrl cells. #P < 0.05, under normal oxygen condition; *P < 0.05, under hypoxia condition). C. Fluorescence microscope analysis of mitochondrial mass in MCF10A-Twist-sh-Ctrl and MCF10A-Twist-sh-Twist cells after Mito-Tracker Green staining (Magnification, x200. Scale bars, 100 μm).

Figure 3. Expression of genes associated with cell energy metabolism is altered in MCF10A-Twist cells

A. The differential expression genes related to cell energy metabolism were identified after bioinformatics analysis of our mRNA microarray data and proteomic data of MCF10A-Twist and MCF10A-Vector cells. B. The mRNA levels of G6PD, PKM2, LDHA, PGK1, ENO1, and TPI1 were analyzed by qRT-PCR in MCF10A-Vector and MCF10A-Twist cells. C, D, E. Western blotting analysis was used to determine the expression of G6PD, PKM2, LDHA, p-AKT and p53. β-actin was used as an internal control. (C) Protein levels in MCF10A-Vector and MCF10A-Twist cells. (*P < 0.05). (D) Protein levels in MCF10A-Vector and MCF10A-Twist cells after hypoxia treatment for indicated time. (E) Protein levels in MCF10A-Twist-sh-Ctrl and MCF10A-Twist-sh-Twist cells. (*P < 0.05).

Figure 4. Molecular mechanisms underlying cell energy metabolism reprogramming in MCF10A-Twist cells

A, B. After treatment with LY294002, glucose consumption and lactate production of MCF10A-Vector and MCF10A-Twist cells were detected. (*P < 0.05). C. Fluorescence microscope analysis of mitochondrial mass in MCF10A-Vector and MCF10A-Twist cells treated with LY294002 by Mito-Tracker Green staining (Magnification, x200. Scale bars, 100 μm). D. After treatment with LY294002, expression of G6PD, PKM2, LDHA, p-AKT, and p53 in MCF10A-Vector and MCF10A-Twist cells was determined by Western blotting. β-actin was used as an internal control (*P < 0.05). E. Western blotting was used to analyze the mTOR, PKM2 and LDHA expression in MCF10A-Vector, MCF10A-Twist transfected with control siRNA, MCF10A-Twist transfected with mTOR siRNA (*P < 0.05). β-actin was used as an internal control. F. Western blotting was applied to analyze the p53, mTOR and G6PD levels in the indicated cells (*P < 0.05). β-actin was used as an internal control.

Figure 5. Twist activates the FAK pathway and its downstream PI3K/AKT through upregulating β1-integrin

The protein levels were determined by Western blotting analysis in the indicated cells, and β-actin was used as an internal control. (*P < 0.05). A. The β1-integrin (ITGB1) and p-FAK expression in MCF10A-Vector cells and MCF10A-Twist cells. B. The levels of β1-integrin (ITGB1) and p-FAK in MCF10A-Vector cells, MCF10A-Twist-sh-Ctrl cells and MCF10A-Twist-sh-Twist cells. C. The β1-integrin (ITGB1), p-AKT, PKM2, LDHA, p53, and G6PD expression in MCF10A-Vector cells, MCF10A-Twist transfected with control shRNA (MCF10A-Twist-sh-Ctrl), MCF10A-Twist transfected with β1-integrin shRNA (MCF10A-Twist-sh-ITGB1). D. After treatment with PF-562271 (FAK inhibitor), expression of p-FAK, p-AKT, PKM2, LDHA, p53, and G6PD in MCF10A-Vector and MCF10A-Twist cells was determined.

Figure 6. Twist inhibits p53 expression via binding to the E-box of p53 promoter

A. Luciferase assays were conducted in HEK293T cells co-transfected pGL3-Luc or pGL3-p53-Luc or pGL3-mut p53-Luc report plasmid with Twist or its control vector respectively. After 30 h, the luciferase activity was determined. The experiments were repeated 3 times. (**P < 0.01). B. MCF10A-Twist cells and BT549 cells was co-transfected with pGL3-Luc or pGL3-p53-luc or pGL3-mut p53-luc report constructer in the presence of Twist shRNA or control shRNA respectively, and about 60 h, the luciferase activity was determined. The experiments were repeated 3 times. (**P < 0.01). C. MCF10A-Twist cells and BT549 cells were processed for ChIP analysis using c-Myc (MCF10A-Twist cells) or Twist (BT549 cells) antibody for immunoprecipitation followed by semi-quantitative PCR with p53 promoter-specific primers. IgG was used as a control antibody.

Figure 7. Energy metabolism reprogramming enhances the ability of cell migration

A, C. Transwell assays were employed to test the cell migratory capacity of MCF10A-Twist, BT549 cells under blockage of FAK activity by PF-562271, or blockage of the PI3K/AKT/mTOR pathway by LY294002 (*P < 0.05). B, D. The migration potential of MCF10A-Twist and BT549 cells was examined by Transwell Chamber. MCF10A-Twist transfected with pCMV-HA-p53 (MCF10A-Twist/p53), BT549 transfected with p53 siRNA (BT549/p53 siRNA) or with pCMV-HA-p53 (BT549/p53) (*P < 0.05).