Ovarian Cancer Relies on Glucose Transporter 1 to Fuel Glycolysis and Growth: Anti-Tumor Activity of BAY-876

Abstract

The recent progresses in understanding of cancer glycolytic phenotype have offered new strategies to manage ovarian cancer and other malignancies. However, therapeutic targeting of glycolysis to treat cancer remains unsuccessful due to complex mechanisms of tumor glycolysis and the lack of selective, potent and safe glycolytic inhibitors. Recently, BAY-876 was identified as a new-generation inhibitor of glucose transporter 1 (GLUT1), a GLUT isoform commonly overexpressed but functionally poorly defined in ovarian cancer. Notably, BAY-876 has not been evaluated in any cell or preclinical animal models since its discovery. We herein took advantage of BAY-876 and molecular approaches to study GLUT1 regulation, targetability, and functional relevance to cancer glycolysis. The anti-tumor activity of BAY-876 was evaluated with ovarian cancer cell line- and patient-derived xenograft (PDX) models. Our results show that inhibition of GLUT1 is sufficient to block basal and stress-regulated glycolysis, and anchorage-dependent and independent growth of ovarian cancer cells. BAY-876 dramatically inhibits tumorigenicity of both cell line-derived xenografts and PDXs. These studies provide direct evidence that GLUT1 is causally linked to the glycolytic phenotype in ovarian cancer. BAY-876 is a potent blocker of GLUT1 activity, glycolytic metabolism and ovarian cancer growth, holding promise as a novel glycolysis-targeted anti-cancer agent.

Keywords: BAY-876; glucose transporter 1; glycolysis; ovarian cancer; patient-derived xenograft.

Figure 1

Glucose transporter 1 (GLUT1) activity is required for maintenance of hyperactive glycolysis in ovarian cancer. Ovarian cancer cell lines were treated for 24 h with indicated concentrations of BAY-876. The glycolytic rate (A) and lactate concentrations in supernatants (B) were measured as described in Materials and Methods. The lactate levels are presented as mM concentrations present in 1 mL culture supernatants normalized to 5 × 10⁵ cells over 24 h incubation. In (C) GLUT1 was silenced by siRNA in SKOV-3 and OVCAR-3 cells and effects of siRNA knockdown on glycolysis and lactate production were determined. For this and all the following figures, the data presented were mean ± SD of triplicates, representative of 3 independent experiments unless otherwise indicated. * p < 0.05; ** p < 0.01.

Figure 2

GLUT1 mediates hypoxia- and energy stress-induced glycolysis. Ovarian cancer cell lines were treated for 24 h with CoCl₂ (150 µM) in the presence or absence of BAY-876 (75 nM) (A). In (B), Ovarian cancer cell lines were treated for 24 h with vehicle, metformin (2 mM), etomoxir (0.25 mM), or MNS-B (50 nM) in the absence or presence of BAY-876 (75 nM). Expression of GLUT1 and HIF-1α proteins was analyzed by immunoblotting. Glycolytic rates were quantified as in Figure 1. The statistical significances of glycolytic induction and BAY-876-dependent inhibition were indicated with red and black asterisks, respectively. n.s.: * p < 0.05; ** p < 0.01.

Figure 2

GLUT1 mediates hypoxia- and energy stress-induced glycolysis. Ovarian cancer cell lines were treated for 24 h with CoCl₂ (150 µM) in the presence or absence of BAY-876 (75 nM) (A). In (B), Ovarian cancer cell lines were treated for 24 h with vehicle, metformin (2 mM), etomoxir (0.25 mM), or MNS-B (50 nM) in the absence or presence of BAY-876 (75 nM). Expression of GLUT1 and HIF-1α proteins was analyzed by immunoblotting. Glycolytic rates were quantified as in Figure 1. The statistical significances of glycolytic induction and BAY-876-dependent inhibition were indicated with red and black asterisks, respectively. n.s.: * p < 0.05; ** p < 0.01.

Figure 3

Inhibition of GLUT1 decreases ATP, activates AMPK and enhances OCR. SKOV-3, OVCAR-3 and A2780 cells were treated with vehicle or BAY-876 (75 nM). Cellular ATP contents (nMol of ATP in 10⁶ cells) (A), and levels of phosphorylated AMPKα (B) were determined with an ATP bioluminescence assay and immunoblotting analysis, respectively. In (C), OCR in these cell lines treated with vehicle or BAY-876 (75 nM) was measured with the Seahorse XF24 Analyzer as detailed in Materials and Methods. The background in cell-free wells was subtracted from readings of all other wells. The presented in upper panel were real time OCRs (pMoles/min, mean ± SD of triplicate assay) over the course of measurement. In lower panel, basal mitochondrial respiration (Basal), ATP-linked respiration (ATP-linked), maximal respiratory capacity (Maximum) and Reserve capacity were calculated and statistically analyzed to show the effects of BAY-876. * p < 0.05; ** p < 0.01.

Figure 4

GLUT1 activity is crucial for cell growth. SKOV-3, OVCAR-3, HEY, and A2780 were cultured in 12-well plates with or without BAY-876 at indicated concentrations. The effects of BAY-876 on cell proliferation were analyzed by determination of cell numbers after 24 h (A) and 72-h growth curves (B). In (C), ovarian cancer cell lines were treated with indicated concentrations of BAY-876 for 72 h before the MTT assay to determine the cytotoxic effect of the compound. IC₅₀ in each cell line was calculated using SigmaPlot 13.0 (Systat Software, San Jose, CA, USA). The effects of BAY-876 (75 nM for SKOV-3 and OVCAR-3, and 200 nM for the less sensitive HEY) on anchorage-independent growth were determined by soft agar assay (D). The data was presented as number of colonies per well. * p < 0.05; ** p < 0.01.

Figure 5

BAY-876 inhibits growth of SKOV-3 xenografts in mice. Four groups of female NSG mice bearing SKOV-3 xenografts were treated with BAY-876 for 28 days at dose of 0, 1.5, 3, or 4.5 mg/kg/day. Data shown were growth curves of tumor volumes (A), body weights over the course of experiment (B), images of end point tumors and final tumor weights (C). The data of all animal experiments were presented as mean ± SD. ** p < 0.01.

Figure 6

BAY-876 suppresses ovarian cancer PDXs in mice. Two serous ovarian carcinoma PDXs grown s.c. in female NSG mice were treated for 30 days with BAY-876 (4 mg/kg/day). Data shown were growth curves of tumor volumes (A,E), body weights over the course of experiment (B,F), images and weights of end point xenografts (C,G) and immunoblotting analysis of GLUT1 protein expression in PDXs from 3 vehicle- and 3 BAY-876-treated mice (D,H). ** p < 0.01.