The Oligostilbene Gnetin H Is a Novel Glycolysis Inhibitor That Regulates Thioredoxin Interacting Protein Expression and Synergizes with OXPHOS Inhibitor in Cancer Cells

Abstract

Since aerobic glycolysis was first observed in tumors almost a century ago by Otto Warburg, the field of cancer cell metabolism has sparked the interest of scientists around the world as it might offer new avenues of treatment for malignant cells. Our current study claims the discovery of gnetin H (GH) as a novel glycolysis inhibitor that can decrease metabolic activity and lactic acid synthesis and displays a strong cytostatic effect in melanoma and glioblastoma cells. Compared to most of the other glycolysis inhibitors used in combination with the complex-1 mitochondrial inhibitor phenformin (Phen), GH more potently inhibited cell growth. RNA-Seq with the T98G glioblastoma cell line treated with GH showed more than an 80-fold reduction in thioredoxin interacting protein (TXNIP) expression, indicating that GH has a direct effect on regulating a key gene involved in the homeostasis of cellular glucose. GH in combination with phenformin also substantially enhances the levels of p-AMPK, a marker of metabolic catastrophe. These findings suggest that the concurrent use of the glycolytic inhibitor GH with a complex-1 mitochondrial inhibitor could be used as a powerful tool for inducing metabolic catastrophe in cancer cells and reducing their growth.

Keywords: Warburg effect; glycolysis; gnetin H; natural products; oligostilbene.

Figure 1

P. suffruticosa seed extract characterization. LC-MS analysis of the P. suffruticosa seed extract fractions (A), their relative percentage (B), and their chemical structures (C).

Figure 2

T98G cell proliferation, cell metabolism, and acidification of the cell culture medium following GH treatment. T98G cells treated with incremental concentrations (4–20 µM) of paeoniflorin (A), Suffructicosols A (B), and B (C), ε-viniferin (D), and GH (E) up to 3 days, and proliferation was measured by using a WST-8 cell proliferation kit). Data are representative of three biological repeats (±SEM) analyzed by 2-WAY ANOVA/Dunnett’s multiple comparisons test; +++ p ≤ 0.001, ++++ p ≤ 0.0001, ** p ≤ 0.01, **** p ≤ 0.0001, # p ≤ 0.05, ## p ≤ 0.01, ### p ≤ 0.001, #### p ≤ 0.0001, ^^ p ≤ 0.01, ^^^^ p ≤ 0.0001, $ p ≤ 0.05, $$$$ p ≤ 0.0001. (if the symbol is: +) CTRL VH vs. GH 20 µM, (if the symbol is: *) CTRL VH vs. GH 10 µM, (if the symbol is: #) CTRL VH vs. GH 8 µM, (if the symbol is: ^) CTRL VH vs. GH 6 µM, (if the symbol is: $) CTRL VH vs. GH 4 µM. GH inhibited cell culture media acidification, and the effect was more distinct after 3 days at higher concentrations (8 µM), suggesting a modulation of cell metabolism and inhibition of glycolysis (F). T98G cells were also tested using a cell viability assay based on protease activity, which is independent of cell metabolism (G). Data are representative of three biological repeats (±SEM) analyzed by 2-WAY ANOVA/Dunnett’s multiple comparisons tests; ++++ p ≤ 0.0001, ** p ≤ 0.01, **** p ≤ 0.0001, ## p ≤ 0.01, #### p ≤ 0.0001, ^ p ≤ 0.05, $$$$ p ≤ 0.0001. (if the symbol is: +) CTRL VH vs. GH 20 µM, (if the symbol is: *) CTRL VH vs. GH 10 µM, (if the symbol is: #) CTRL VH vs. GH 8 µM, (if the symbol is: ^) CTRL VH vs. GH 6 µM, (if the symbol is: $) CTRL VH vs. GH 4 µM.

Figure 3

T98G and B16 cell viability following GH treatment. Human glioblastoma T98G and murine melanoma B16 cells were treated with an incremental concentration of GH (4–10 µM) for 24 h and a WST8 assay was used to determine cell viability. Data are representative of three biological repeats (±SEM) 2 WAY ANOVA/Dunnett’s multiple comparisons test; **** p ≤ 0.0001.

Figure 4

T98G and B16 intracellular and extracellular lactic acid production and glucose transport measurement. B16 (A) and T98G (B) cells were treated with 4 µM GH and the level of extracellular lactate was measured by sampling the culture medium at 0, 1, 3, and 6 h. Intracellular lactic acid levels were measured in B16 cells (C) following 3 h of treatment with 4 µM GH, AZD3965 (100 nM), an MCT-1 inhibitor, or cotreated with GH and AZD3965. Data are representative of three biological repeats (±SEM); one-way ANOVA Dunnet’s multiple comparisons test; CTRL VH vs. single treatment; *** p ≤ 0.001. Flow cytometric analysis of fluorescent glucose analog 2-NBDG (100 μg/mL) following treatment with 4 µM GH or phloretin (PHL), a glucose transporter inhibitor in B16 and T98G cells (D). Data are representative of three biological repeats (±SEM); (one-way ANOVA, Dunnet’s post-test, *** p < 0.001, **** p < 0.0001).

Figure 5

Effects on T98G cells of combined treatment with glycolysis inhibitors and the mitochondrial complex-I inhibitor phenformin. (A) T98G cells treated for 48 h with either 4 µM GH or other known glycolysis inhibitors, such as 25 mM dichloroacetate (DCA), 25 mM oxamate, 25 mM 2-deoxy-d-glucose (2DG), 25 µM 3- bromopyruvate (3BrPA), alone or in combination with 100 µM of the mitochondrial complex-I inhibitor phenformin. Data is representative of three biological repeats (±SEM) 2-way ANOVA/Tukey’s multiple comparisons test; * p ≤ 0.05, *** p ≤ 0.001, **** p ≤ 0.0001 (B) Assessment of apoptosis (early/late apoptosis) and necrosis following treatment with 4 µM GH alone or in combination with 100 µM phenformin for 48 h by annexin-V/propidium iodide (PI) staining. (C) Analysis of T98G viability independent from the effect on cell metabolism using a protease activity-based cell viability assay. T98G cells were treated for 3 days with 8 µM GH, 100 µM phenformin, or the combination of the two was assessed at 24 h. Data is representative of three biological repeats, analyzed by 2-WAY ANOVA/Dunnett’s multiple comparisons test; CTRL VH vs. GH/time point, * p ≤ 0.05,** p ≤ 0.01, *** p ≤ 0.001; CTRL VH vs. Phen/time point, ### p ≤ 0.001; CTRL VH vs. GH + Phen/time point, ◊◊ p ≤ 0.01, ◊◊◊ p ≤ 0.001.

Figure 6

The treatment with GH impairs the glycolytic activity of T98G cells. The glycolytic activity was measured by Agilent SeaHorse XFe96 Analyzer after 20 min of treatment with either phenformin (100 µM), gnetin H (10 µM), or the combination injected into the cell media following basal ECAR measurements. (*) indicates the time of treatment with phen, GH, or the combination. ECAR data are shown as mpH/min normalized to proteins and represent the mean ± SD of n = 5 independent measurements under normal cell culture conditions. Glycolytic profiles were obtained using the Agilent SeaHorse Glycolysis Stress Test. The ECAR was measured under glucose starvation and after the sequential addition of glucose (basal glycolysis), oligomycin (maximal glycolysis or glycolytic reserve), and 2-DG (glycolysis specificity).

Figure 7

RNA-Seq analysis of T98G cells treated with 2DG, Phen, 2DG + Phen, GH, or GH + Phen. (A) The experimental variation between RNA-Seq experiments performed on 3 biological replicate samples was analyzed with a PCA plot. (B) Venn diagrams comparing the number of up- or downregulated genes that significantly responded (≥2-fold induction, p-value ≤ 0.05) to the different treatments. (C) Hierarchical cluster analysis was performed on 566 genes that displayed a ≥2-fold induction (p-value ≤ 0.05) response to at least one of the three treatments compared to the solvent control. Blue and yellow colors represent up- and downregulated genes, respectively. The genes listed responded to GH + Phen treatment, but not to 2DG or 2DG + Phen treatment.

Figure 8

Western blot expression analysis of key mediators of pathways involved in the regulation of cell metabolism following GH treatment in T98G cells. (A) Treatment of samples (alone or in combination with GH and phenformin) is indicated above the protein bands. (B) Treatment of samples (alone or in combination with phenformin and 2DG) is indicated above the protein bands. Antibodies (TXNIP, c-myc, p-AMPK, and p-AKT) are indicated on the side, β-Actin was used as the internal standard. Data is representative of three biological repeats. (C) Treatment of samples (alone or in combination with phenformin, GH, and 2DG) is indicated above the protein bands. Antibodies (p-JNK, p-ERK, p-p38, p38) are indicated on the side, β-Actin was used as the internal standard. Data are representative of three biological repeats.

Figure 8

Western blot expression analysis of key mediators of pathways involved in the regulation of cell metabolism following GH treatment in T98G cells. (A) Treatment of samples (alone or in combination with GH and phenformin) is indicated above the protein bands. (B) Treatment of samples (alone or in combination with phenformin and 2DG) is indicated above the protein bands. Antibodies (TXNIP, c-myc, p-AMPK, and p-AKT) are indicated on the side, β-Actin was used as the internal standard. Data is representative of three biological repeats. (C) Treatment of samples (alone or in combination with phenformin, GH, and 2DG) is indicated above the protein bands. Antibodies (p-JNK, p-ERK, p-p38, p38) are indicated on the side, β-Actin was used as the internal standard. Data are representative of three biological repeats.