Anti-Warburg Mechanism of Ginsenoside F2 in Human Cervical Cancer Cells via Activation of miR193a-5p and Inhibition of β-Catenin/c-Myc/Hexokinase 2 Signaling Axis

Abstract

Though Ginsenoside F2 (GF2), a protopanaxadiol saponin from Panax ginseng, is known to have an anticancer effect, its underlying mechanism still remains unclear. In our model, the anti-glycolytic mechanism of GF2 was investigated in human cervical cancer cells in association with miR193a-5p and the β-catenin/c-Myc/Hexokinase 2 (HK2) signaling axis. Here, GF2 exerted significant cytotoxicity and antiproliferation activity, increased sub-G1, and attenuated the expression of pro-Poly (ADPribose) polymerase (pro-PARP) and pro-cysteine aspartyl-specific protease (procaspase3) in HeLa and SiHa cells. Consistently, GF2 attenuated the expression of Wnt, β-catenin, and c-Myc and their downstream target genes such as HK2, pyruvate kinase isozymes M2 (PKM2), and lactate dehydrogenase A (LDHA), along with a decreased production of glucose and lactate in HeLa and SiHa cells. Moreover, GF2 suppressed β-catenin and c-Myc stability in the presence and absence of cycloheximide in HeLa cells, respectively. Additionally, the depletion of β-catenin reduced the expression of c-Myc and HK2 in HeLa cells, while pyruvate treatment reversed the ability of GF2 to inhibit β-catenin, c-Myc, and PKM2 in GF2-treated HeLa cells. Notably, GF2 upregulated the expression of microRNA139a-5p (miR139a-5p) in HeLa cells. Consistently, the miR139a-5p mimic enhanced the suppression of β-catenin, c-Myc, and HK2, while the miR193a-5p inhibitor reversed the ability of GF2 to attenuate the expression of β-catenin, c-Myc, and HK2 in HeLa cells. Overall, these findings suggest that GF2 induces apoptosis via the activation of miR193a-5p and the inhibition of β-catenin/c-Myc/HK signaling in cervical cancer cells.

Keywords: apoptosis; cervical cancer; ginsenoside F2; glycolysis; miR193a-5p.

Figure 1

Effect of GF2 on cytotoxicity and colony formation in HeLa and SiHa cells. (A) Chemical structure of GF2 (MW = 785.03). (B) Cytotoxic effect of GF2 in HeLa and SiHa cells. The cells were exposed to various concentrations of GF2 for 24 h, and cell viability was assessed by MTT assay. * p < 0.05, ** p < 0.01, and *** p < 0.001 versus the untreated control. (C) Antiproliferative effect of GF2 in HeLa and SiHa cells. HeLa and SiHa cells were treated with GF2 (0, 50, and 70 μM), and then colony formation took place over 1 week. The colonies were visualized by staining with Diff-Quick solution (Sysmex, Kobe, Japan). The data represent the means ± SD.

Figure 2

Effect of GF2 on apoptosis in HeLa and SiHa cells. (A) Effect of GF2 on PARP and caspase 3 in HeLa and SiHa cells. (B) Effect of GF2 on sub-G1 phase in HeLa and SiHa cells.

Figure 3

Effect of GF2 on band level of Wnt, β-catenin, c-Myc, HK2, PKM2, and LDHA in HeLa and SiHa cells. (A) Effect of GF2 on Wnt, β-catenin, and c-Myc in HeLa and SiHa cells. (B) Effect of GF2 on HK2, PKM2, and LDHA in HeLa and SiHa cells. (C) Effect of GF2 on glucose production in HeLa and SiHa cells using a colorimetric assay. The results represent the ±SD; *** p < 0.001. (D) Effect of GF2 on lactate production in HeLa and SiHa cells using a colorimetric assay. The results represent mean ± SD; ** p < 0.01, *** p < 0.001.

Figure 4

Effect of GF2 on the stability of β-catenin, c-Myc, and HK2 in HeLa cells.

Figure 5

Binding index and correlation efficient between β-catenin and c-Myc and the effect of β-catenin siRNA or pyruvate on glycolysis-related proteins in HeLa cells. (A) RNA-seq data (cBioportal) confirm a strong correlation between β-catenin and c-Myc with a Spearman correlation coefficient of 0.09. (B) Binding between β-catenin and c-Myc in HeLa cells. Immunoprecipitation was performed in HeLa cells treated with GF2 for 24 h and then subjected to Western blotting to detect β-catenin and c-Myc concentrations in whole-cell lysates. (C) Effect of β-catenin depletion on β-catenin, c-Myc, and HK2 in GF2-treated HeLa cells. (D) Effect of pyruvate treatment on β-catenin and PKM2 in GF2-treated HeLa cells.

Figure 6

The important role of miR-139a-5p in GF2-induced apoptosis in HeLa cells. (A) Direct binding of miR-139a-5p to the 3′-Untranslated region of β-catenin. (B) Endogenous expression level of miR139a-5p in HeLa cells. The expression level of miR139a-5p was measured in intact HeLa cells by RT-PCR, *** p < 0.001. (C) Effect of miR139a-5p mimic on β-catenin, HK2, and c-Myc in HeLa cells. The expression of β-catenin, c-Myc, HK2, and pro-PARP was evaluated in HeLa cells with or without the miR139a-5p mimic by Western blotting. (D) Effect of miR139a-5p inhibitor on β-catenin, c-Myc, HK2, and pro-PARP in GF2-treated HeLa cells. The expression of β-catenin, c-Myc, HK2, and pro-PARP was evaluated in HeLa cells with or without GF2 by Western blotting.

Figure 7

Schematic diagram on the anti-Warburg effect of GF2 via the activation of miR-139a-5p and the suppression of the β-catenin/c-Myc/HK2 signaling axis.