1. Introduction
Altered metabolic activity is required to satisfy the increased needs of highly proliferating cancer cells. Moreover, metabolic alterations are also mechanistically related to other hallmarks of cancer. As with other tumor types, the Warburg effect, which is a typical change in energy metabolism associated with enhanced lactate production (aerobic glycolysis), was also observed in head and neck squamous cell carcinomas (HNSCC) [
1]. Indeed, glucose was identified as the most important source of energy and carbon skeletons in HNSCC cells [
2]. The Warburg effect can be forced via the increased activity of transcription factors, such as c-Myc and Hif-1α, which act by promoting the transcription of specific isoforms of glycolytic enzymes, e.g., hexokinase II (HKII) and pyruvate kinase M2 [
1,
3,
4,
5]. Our recent study showed the beneficial effects of c-Myc inhibition in hypopharyngeal cancer cells [
6]. However, while c-Myc inhibition may block cell proliferation, it may also promote cell survival [
7,
8,
9]; thus, it might be useful as an anticancer strategy, but rather in combination with other chemicals. Apart from c-Myc transcription factor, Akt kinase is also responsible for enhancing glucose metabolism, and the inhibition of c-Myc and Akt downregulated glucose transporters and crucial glycolytic enzymes [
10]. Indeed, targeting Akt kinase seems to be a promising strategy for reducing the survival of tongue cancer cells [
11].
The effective inhibition of proteins orchestrating cellular metabolism is still difficult and requires the development of appropriate small molecule inhibitors. The attenuation of excessive glucose metabolism may be reached by direct targeting of glucose flux in the glycolytic pathway. One of the crucial checkpoints of glycolysis is the activity of hexokinase catalyzing its initial step—the phosphorylation of glucose to glucose-6-phosphate. Tongue cancer cells, in comparison to noncancerous cells, are characterized by increased expression of HKII [
12]. Therefore, HKII is a potentially beneficial molecular metabolic target in tongue carcinoma.
2-Deoxyglucose (2-DG) is a well-known hexokinase inhibitor [
13], but its clinical use in sufficiently high doses is limited by side effects concerning the central nervous system. To bypass the problem, the 2-DG dose can be lowered, but to reach satisfactory effects, some combinatorial treatment with 2-DG should be introduced. Our previous study showed the therapeutic potential of targeting Wnt signaling in tongue cancer CAL 27 and SCC-25 cell lines [
14]. Moreover, our recent study revealed a significant influence of the attenuation of Wnt signaling on glycolytic activity, which was partly potentiated by cotreatment with Akt inhibitor [
11]. Indeed, the canonical Wnt pathway is often upregulated in tumor cells [
15] and cancer stem cells [
16]. Wnt/β-catenin signaling promotes the PI3K/Akt pathway (responsible, for instance, for the activation of energy metabolism) by tyrosine kinase receptor activation, e.g., EGFR [
17]. Moreover, PI3K/Akt pathway inactivates GSK-3β kinase, leading to the increased stability of β-catenin [
18]. Therefore, Wnt/β-catenin signaling and the PI3K/Akt pathway control each other in a positive feedback loop leading to the increased stability of β-catenin, which promotes the expression of Wnt pathway target genes (e.g., c-Myc), which are able to activate Hif-1α in a hypoxia-independent manner [
15]. Both Hif-1α and c-Myc transcription factors upregulate the expression of glycolytic genes, which in turn induce metabolic reprogramming, known as the Warburg effect. The direct influence of Wnt signaling on the expression of glycolytic genes was observed for
MCT-1 and
LDHA genes in colorectal cancer [
19]. In addition, the inhibition of Wnt signaling by niclosamide was associated with the reduced expression of genes related to glycolysis in FaDu hypopharyngeal carcinoma cells [
20].
Thus, the combination of 2-DG—a direct metabolic inhibitor—with small molecules targeting the Wnt pathway seems to be a promising strategy. Apart from 2-DG, other modulators of energy metabolism can also be useful in the context of anticancer therapy. Lonidamine is a hexokinase inhibitor that shows activity against additional molecular targets associated with energy metabolism, including the reduction of mitochondrial respiration and ATP production in cancer cells [
21].
The aim of this study was to assess the effects of the combinatorial use of inhibitors of glycolysis (2-deoxyglucose and lonidamine) together with inhibitors of the Wnt pathway (PRI-724 and IWP-O1) on cell viability, proliferation and apoptosis in tongue carcinoma cells. Moreover, the effects on glycolytic activity were also assessed in order to verify whether the modulation of energy metabolism is an important contributor of the mode of action of compounds inhibiting the Wnt pathway. Based on the experimental results, we have shown that the combined inhibition of glycolysis and Wnt signaling is beneficial in the attenuation of tongue cancer cell growth.
3. Discussion
New effective molecular targeted therapy is needed for the successful treatment of HNSCC patients, especially those with advanced, recurrent tumors. An important role in HNSCC cells biology is played by changes in energy metabolism, including the widely known Warburg effect. In addition, the deregulation of Wnt signaling was also shown to be important for the growth of oral carcinoma [
14,
22], and a potential association between Wnt signaling and energy metabolism was observed [
11]. In the current study, the effects of the co-inhibition of glycolysis and Wnt signaling were evaluated in tongue squamous cell carcinoma cell lines derived from primary tumors (CAL 27 and SCC-25 cells) or a lymph node metastasis of tongue cancer (BICR 22 cells). The metabolic activity of tongue cancer cells was modified by 2-deoxyglucose and lonidamine, while PRI-724 and IWP-O1 inhibitors were used for Wnt pathway modulation.
Glycolytic flux is potently decreased by 2-DG via the inhibition of hexokinase, the first and key regulatory enzyme of glycolysis. Furthermore, 2-DG reduces cellular ATP levels and antioxidant power by decreasing the flow of metabolites into the pentose phosphate pathway. Moreover, it induces autophagy and disrupts protein N-glycosylation [
13]. Despite the known efficacy of 2-DG action against glycolytic flux, the main problem with the use of clinically relevant 2-DG doses is related to the depression of the central nervous system as a result of the brain’s dependence on glucose. However, it may be solved by the introduction of a ketogenic diet [
23]. On the other hand, the beneficial properties of 2-DG could be achieved by using lower doses in combination with other chemicals to improve the therapeutic effects. The addition of 2-DG to standard chemo- and radiotherapy can enhance sensitivity of cells in vitro, but more potent antimetabolites are required for in vivo use [
24]. Cotreatment of FaDu cells with 2-DG and cisplatin enhanced cytotoxic effects with concomitant increased oxidative stress [
25], while in Jurkat cells, the combination with borasertib or everolimus induced apoptosis without an elevation in reactive oxygen species production [
26]. In contrast, 2-DG seems to decrease the efficacy of anti-EGFR therapy by erlotinib application in a xenograft model of HNSCC tumors, possibly via induction of cytoprotective autophagy [
27]. In our study, the comparison of the results of the viability and cytotoxicity tests revealed that the effects of 2-DG in BICR 22 cells are predominantly related to cytotoxic response induction, while in the two other tumor cell lines the effects were related mostly to reduced proliferation. This is in agreement with a slight accumulation of cells in the G1/G0 phase. In all three cell lines, 2-DG increased the number of apoptotic cells by 6–8% compared to DMSO treated cells after 48 h of incubation, so even the IC25 concentrations seemed to be able to induce the programmed death of tongue cancer cells.
Lonidamine is an example of another hexokinase inhibitor, for which several mechanisms of action were proposed. Similarly to 2-DG, it attenuates the flux of metabolites through glycolysis and the pentose phosphate pathway, but additionally it inhibits mitochondrial respiration by blocking succinate–ubiquinone reductase activity of respiratory complex II and mitochondrial pyruvate carrier [
21]. Moreover, lonidamine causes intracellular tumor acidification as a result of inhibition of MCT-1, 2 and 4 lactate transporters [
21,
28]. The limitation of the clinical use of lonidamine is predominantly related to its toxicity against the liver, which may be reduced by using liposome formulations of this compound [
29]. Antiglycolytic therapy based on lonidamine in combination with other chemicals may be more advantageous [
29,
30]. The reduction in cell viability by lonidamine in CAL 27 and BICR 22 cells resulted from the induction of cytotoxic and antiproliferative effects. In comparison to other chemicals used in monotherapy, lonidamine caused the most effective accumulation of cells in the G1/G0 phase in BICR 22 cell lines, but apoptosis induction was weaker than for 2-DG.
The modulation of canonical Wnt signaling was shown to alter energetic metabolism, e.g., in colorectal cancer cells [
19,
31]. In a previous study we observed that the inhibition of the Wnt canonical pathway can be efficient in lowering glycolytic activity in oral cancer cells, including SCC-25 cells, in which it enhanced the effect of the Akt inhibitor [
11]. In this study, we evaluated the potential benefits of combinatorial treatment with metabolic modulators and Wnt signaling inhibitors. PRI-724 targets the nuclear part of the Wnt/β-catenin-dependent pathway via disruption of the interaction between β-catenin and CBP histone acetyltransferase [
32], while IWP-O1 targets porcupine—the
O-acyltransferase crucial for maturation of Wnt ligands—thus attenuating both β-catenin-dependent and independent pathways [
33]. We previously described the benefits of Wnt pathway inhibition in HNSCC [
6,
20,
34,
35]; however, the simultaneous targeting of Wnt signaling and other molecular pathways is necessary for the potentiation of the anticancer effects.
In this study, we performed experiments using combinations of glycolytic inhibitors—2-DG or lonidamine—with small molecule inhibitors of the Wnt signaling pathway—PRI-724 or IWP-O1—at low (IC25) concentrations. In general, the results showed significant improvement of the observed effects after the application of mixtures in comparison to single 2-DG/lonidamine use. Beneficial results were observed for viability, cytotoxicity, apoptosis and cell cycle analysis, and only some differences were reported for mixes of 2-DG and lonidamine with PRI-724/IWP-O1. Comparing particular cell lines, SCC-25 cell viability was more effectively inhibited by combinations of metabolic and Wnt signaling inhibitors than CAL 27 cells. Moreover, IC25 concentrations of 2-DG, lonidamine and PRI-724 were lower in SCC-25 cells. In previous research assessing the migration potential of CAL 27 and SCC-25 cells [
14], the migration of unaffected SCC-25 cells was higher, similar to its higher basal glycolytic activity in the current research. Thus, SCC-25 cells appear to show higher proliferation, metabolic rate and motility. This can partly be the reason for SCC-25 cells’ higher susceptibility to viability reduction by the compounds. Moreover, the pleiotropic mechanism of action of 2-DG/lonidamine mixtures with PRI-724 may be responsible for the appearance of an antiproliferative, rather than proapoptotic, effect in this cell line. On the other hand, CAL 27 cells were also affected, e.g., in regard with apoptosis induction and cell cycle inhibition, so beneficial effects were shown for both primary tongue cancer cell lines. Moreover, in metastatic BICR 22 cells, we observed the highest cytotoxic effect, which shows the potential of this strategy in the search for a beneficial treatment of more advanced tumors.
Metabolic analysis revealed important differences in the actions of Wnt inhibitors. PRI-724 in combination with 2-DG/lonidamine significantly reduced glucose consumption by tongue cancer cells, and decreased lactate release to a culture medium. In turn, IWP-O1—an inhibitor of porcupine—had no significant effect on glucose consumption, and tended to increase lactate production, with the exception of its mixture with 2-DG in CAL 27 cells. That suggests a differential influence on energy metabolism, depending on whether only the canonical or both the canonical and noncanonical Wnt signaling are affected. The molecular mechanisms of these phenomena need to be further elucidated.
4. Materials and Methods
4.1. Cell lines and Culture
Three commercially available tongue squamous cell carcinoma cell lines were used in this study: CAL 27 and SCC-25 cell lines (derived from primary tumors) were purchased from American Type Culture Collection (ATCC), while BICR 22 cells (derived from a lymph node metastasis) were purchased from European Collection of Authenticated Cell Cultures (ECACC).
The CAL 27 and BICR 22 cells were grown in a high-glucose DMEM medium (Biowest, Nuaillé, France), supplemented with 10% FBS (EURx, Gdańsk, Poland) and 1% antibiotic solution (penicillin and streptomycin; Biowest, Nuaillé, France). SCC-25 cells were grown in a 1:1 mixture of DMEM medium with F12 medium containing 1.2 g/L sodium bicarbonate, 2.5 mM L-glutamine, 15 mM HEPES and 0.5 mM sodium pyruvate (Biowest, Nuaillé, France) supplemented with 10% FBS (EURx, Gdańsk, Poland), 1% antibiotics solution (penicillin and streptomycin; Biowest, Nuaillé, France) and 400 ng/mL hydrocortisone (Sigma-Aldrich, St. Louis, MI, USA). Cells were cultured under standard conditions (37 °C and 5% CO2) in a humidified incubator (Memmert, Schwabach, Germany).
4.2. Chemicals and MTS Viability Assay
Four small-molecule inhibitors were used in the experiments. 2-Deoxyglucose (2-DG) and lonidamine (Sigma-Aldrich, St. Louis, MI, USA) were used to directly modulate energy metabolism. Metabolic modulators were combined with Wnt signaling inhibitors—PRI-724 (Selleck Chemicals, Pittsburgh, PA, USA) and IWP-O1 (Sigma-Aldrich, St. Louis, MI, USA). Each time directly before use, 2-DG was dissolved in PBS buffer, and stock solutions of the other compounds were prepared in DMSO and stored in aliquots at −20 °C.
The MTS assay was performed in order to assess the effect of single compounds on the viability of CAL 27, SCC-25 and BICR 22 cells, and to determine IC25 values. Then, viability reduction after treatment with combinations of chemicals at equipotent concentrations (IC25 + IC25) was evaluated. The assay was performed using CellTiter 96® AQueous One Solution Reagent (Promega, Madison, WI, USA) according to the manufacturer’s protocol. Briefly, the cells were seeded in 96-well plates (104 cells per well) and left for 24 h preincubation. Then, the appropriate growth medium was replaced with fresh medium containing various concentrations of the tested compounds and the cells were incubated for an additional 48 h. The cells incubated with solvent (DMSO not exceeding 0.2% or PBS) served as a control (100% viability). Afterwards, the wells were rinsed with PBS buffer and fresh medium containing MTS solution (100 µL + 20 µL) was added. After 60 min of incubation, the absorbance at 490 nm was read using an Infinite M200 multiplate reader (Tecan, Grödig, Austria). The assay was performed in triplicate, each time with four replicates per assay. The IC25 values, which were determined based on the MTS assay, were used in further experiments for single glycolysis inhibitors and their combinations with Wnt signaling inhibitors (IC25 + IC25).
4.3. Cytotoxicity Assay
The CellTox™ Green Cytotoxicity Assay (Promega, Madison, WI, USA) was used to assess the cytotoxic effects of single compounds and their combinations. The assay, which measures the changes in membrane integrity by detecting the binding of a fluorescent dye to DNA, was performed according to the manufacturer’s recommendations. Briefly, the cells were seeded in black 96-well plates (104 cells per well) and left for 24 h preincubation. Then, the appropriate growth medium was replaced with fresh medium (100 µL per well) containing IC25 concentrations of glycolysis inhibitors or their combinations with Wnt signaling inhibitors and the cells were incubated for an additional 24 or 48 h. The cells incubated with solvent served as a negative control. Additionally, the cells that were incubated 30 min before end of the experiment with lysis solution (4 µL per well) served as a positive control of cytotoxic effects. Then, freshly prepared dye solution in assay buffer was directly added (100 µL per well), plates were orbitally shaken for 1 min, incubated for 15 min at room temperature in the dark, once again orbitally shaken for 1 min and the signal (ex/em −500/530 nm) was measured in an Infinite M200 multiplate reader (Tecan, Grödig, Austria). The assay was performed in triplicate, each time with three replicates per assay.
Relative fluorescence values representing IC25 values of cytotoxic effects (as the fold of control, which is equal to 1) were calculated based on the fluorescence of the positive control using the equation:
where RFU denotes Relative Fluorescence Units.
4.4. Mitochondrial Transmembrane Potential
Proapoptotic effects of single compounds and their combinations were analyzed by the measurement of mitochondrial membrane potential (ΔΨm) using the Muse® MitoPotential Kit (Merck, Darmstadt, Germany) according to the manufacturer’s recommendations. Briefly, 2 × 105 cells per well were seeded in 6-well plates and preincubated for 24 h. Then, the appropriate growth medium was replaced with fresh medium (2 mL per well) containing IC25 concentrations of the tested glycolysis inhibitors or their combinations with Wnt signaling inhibitors, and the cells were incubated for an additional 48 h. The cells incubated with solvent served as a negative control. Additionally, cells incubated with 10 µM CCCP (carbonyl cyanide 3-chlorophenylhydrazone; Sigma-Aldrich, St. Louis, MI, USA) served as a positive control of mitochondrial membrane depolarization. Afterwards, the cells were collected by trypsinization, resuspended in 100 µL of assay buffer, 100 µL of MitoPotential working solution and incubated for 20 min at 37 °C in a humidified CO2 incubator. Fluorescence was analyzed by flow cytometry on a Muse® Cell Analyzer (Merck, Darmstadt, Germany). Data were evaluated using Muse® 1.5 analysis software (Merck, Darmstadt, Germany). All the experiments were done in triplicate.
4.5. Cell Cycle Analysis
The effect of single compounds and their combinations on cell cycle distribution was analyzed using the Muse® Cell Cycle Kit (Merck, Darmstadt, Germany) based on propidium iodide interaction with DNA. Briefly, 2 × 105 cells per well were seeded in 6-well plates and preincubated for 24 h. Then, the appropriate growth medium was replaced with fresh medium (2 mL per well) containing IC25 concentrations of the glycolysis inhibitors or their combinations with Wnt signaling inhibitors, and the cells were incubated for an additional 48 h. The cells incubated with solvent served as a negative control. Additionally, cells incubated with 100 nM topotecan (Sigma-Aldrich, St. Louis, MI, USA) served as a positive control of cell cycle arrest. Then, cells were collected by trypsinization, washed with PBS buffer and fixed in ice-cold 70% ethanol. After at least overnight storage at −20° C, fixed cells were collected by centrifugation and washed with PBS buffer. The distribution of cells, depending on the cell cycle phase (G1/G0, S, G2/M), was analyzed with the Muse® Cell Analyzer (Merck, Darmstadt, Germany) after 30 min incubation with propidium iodide solution in the presence of RNase A at room temperature in the dark. Data were analyzed using Muse® 1.5 analysis software (Merck, Darmstadt, Germany). All the experiments were done in triplicate.
4.6. Glucose Concentration Analysis
The analysis of the glucose concentration in the culture medium was performed using the Glucose-Glo™ Assay (Promega, Madison, WI, USA). Briefly, the cells were seeded in 96-well plates (10
4 cells per well) and left for 24-h preincubation. Then, the appropriate growth medium was replaced with fresh medium containing IC25 concentrations of the glycolysis inhibitors or their combinations with Wnt signaling inhibitors, and the cells were incubated for an additional 48 h. Cells incubated with solvent served as a negative control. Afterwards, 2 µL of medium from each well was diluted in 1998 µL of PBS buffer. Luminometric analysis of the glucose concentration was performed on a GloMax
® Discover microplate reader (Promega, Madison, WI, USA) after 60 min incubation of diluted culture medium (50 µL per well) with a glucose detection reagent (50 µL per well). The glucose concentration in the media was determined based on a linear standard curve, which was prepared in parallel (
Supplementary Figure S1). Glucose consumption was calculated as the difference between glucose concentration in cell-free media and media samples obtained after 48 h of cell incubation in the presence of the compounds. Data were normalized based on the viability assay results. The assay was performed in triplicate, each time with three replicates per assay.
4.7. Lactate Concentration Analysis
The analysis of the lactate concentration in the culture medium was performed using the Lactate-Glo™ Assay (Promega, Madison, WI, USA). Briefly, the cells were seeded in 96-well plates (10
4 cells per well) and left for 24-h pre-incubation. Then, the appropriate growth medium was replaced with fresh medium containing IC25 concentrations of the glycolysis inhibitors or their combinations with Wnt signaling inhibitors and the cells were incubated for an additional 48 h. Cells incubated with solvent served as a negative control. Afterwards, 10 µL of medium from each well was diluted in 990 µL of PBS buffer. Luminometric analysis of the lactate concentration was performed on a GloMax
® Discover microplate reader (Promega, Madison, WI, USA) after 60 min incubation of the diluted culture medium (50 µL per well) with lactate detection reagent (50 µL per well). The lactate concentration in the media was determined based on a linear standard curve, which was prepared in parallel (
Supplementary Figure S2). Lactate release was calculated as the level of lactate present in media samples obtained after 48 h of incubation. Data were normalized based on the viability assay results. The assay was performed in triplicate, each time with three replicates per assay.
4.8. Statistical Analysis
For the analysis of the significance of differences between controls, glycolysis inhibitors and their combinations with Wnt signaling inhibitors, the one-way ANOVA test with the Tukey post hoc test were performed, with p < 0.05 considered as significant. The analyses were performed using STATISTICA 11 software.