Abstract
Background/Aim: In glioblastoma multiforme (GBM), a deadly brain tumor, glucose is one of the main fuels for accelerated growth. Patients with GBM are also exposed to excess glucose through hyperglycemia in diabetes mellitus. In addition, dexamethasone (Dex), a corticosteroid commonly administered for controlling cerebral oedema, causes additional excess glucose. Therefore, targeting glucose metabolism is an attractive therapeutic intervention for GBM treatment. We have recently shown that riluzole (Ril), a drug used to treat amyotrophic lateral sclerosis (ALS), has an effect on some detrimental Dex-induced metabolic changes in GBM. Therefore, we examined the effect of the combination of metformin (Met), widely used to treat type 2 diabetes, and Ril on GBM cells.
Materials and Methods: The 3-(4, 5-dimethylthiazol)-2, 5-diphenyltetrazolium bromide (MTT) assay was used to determine cell viability of U87MG after treatment with Ril, Met, Ril plus Met (Ril+Met) and the addition of Dex to this co-treatment. Cell migration was assessed by the xCELLigence system, matrix metalloproteinase 2 (MMP2) activation by zymography assay and gene expression by real-time polymerase chain reaction (RT-PCR).
Results: Co-treatment with Ril and Met was effective in killing GBM cells and reducing the expression of genes involved in glucose and stem cell metabolism. Furthermore, combination of Ril and Met reduced MMP2 activation. But co-administration increased the migration of U87MG cells. The addition of Dex to this combination reversed the unfavorable effects of Ril+Met on cell migration.
Conclusion: Ril+Met co-treatment had a positive effect in terms of GBM cell death, decreased expression of genes involved in glucose metabolism and stemness, and reduced MMP2 activation. Disadvantage of Ril+Met treatment was increased cell migration. Taken together, these drug combinations may also allow the reduction of the concentration of Dex to minimize its side effects.
Introduction
Glioblastoma multiforme (GBM) is biologically complex and thus difficult to treat. Despite the best available treatment – surgical resection and radio/chemotherapy – only approximately 3% of patients survive 3-6 years and the majority of patients survive no longer than 12 months. Glucose and glutamine, the metabolic fuels for almost all brain functions under normal physiological conditions, also fuel rapid tumor growth.
In GBM, high glucose is the substrate for rapid growth, with enhanced chemotaxis and production of VEGF mediated by the chemoattractant and growth factor receptors formyl peptide receptor 1 (FPR1) and epidermal growth factor receptor (EGFR) (1). Glucose transporters are important indicators of poor prognosis in cancer (2). Glucose transporter 3 (GLUT3) levels in gliomas increase with brain tumor grades, with GBM showing the highest expression, and GLUT3 expression is found to be higher in recurrent GBMs than in primary GBMs (3). In addition, many patients with GBM are older and have type 2 diabetes mellitus. High glucose in diabetes mellitus activates intracellular pathways such as pro-cell survival AKT/mTOR, enhances WNT/β-catenin signaling, up-regulates inflammatory cytokine levels in the circulation, and induces epithelial-mesenchymal transition (EMT) (4-7). Currently, GBM oedema is mostly treated with the corticosteroid dexamethasone (Dex), which significantly increases blood glucose concentration (8), invasion, proliferation and angiogenesis in human GBM stem cell (GSC)-derived orthotopic tumours, potentially worsening the prognosis of patients with GBM (9). Overall, controlling glucose levels is one of the strategies used to mitigate its deleterious effect on cancer/GBM progression.
Ril is an approved drug for the treatment of amyotrophic lateral sclerosis ALS (10) and is being tested as a single agent or in combination with other drugs for the treatment of various cancers [reviewed in (11)]. Previous studies have demonstrated the broad-spectrum activity of Ril, including its anti-glutamatergic pharmacological properties (12, 13). In a GBM model using the U87 GBM cell line, Ril blockade of glutamate release inhibited cell proliferation (14). The involvement of Ril in the glucose metabolism of GBM was shown in our previous work, which demonstrated the effect of Ril on the GLUT3 transporter in GBM stem-like cells. Furthermore, Ril was effective in killing brain tumor stem-like cells in vitro and inhibited tumor growth in vivo (15).
Met is a widely used drug in patients with diabetes [reviewed in (16)]. A growing number of studies have shown a beneficial effect of Met in tumors (17-20). The cytotoxic activity of Met is due to a reduction in glucose levels in the tumor milieu (21). In GBM cells, Met inhibits cell proliferation (22), motility/invasion (23), apoptosis and autophagy (23, 24), inhibits glioma cells stemness and epithelial-mesenchymal transition (25, 26). Met also increases the efficacy of standard glioma therapies (27). Many studies have shown that co-administration of Met and temozolomide (TMZ) leads to a synergistic response by GBM cells, with an increase in the mortality of TMZ-sensitive and TMZ-resistant cells (27, 28). To date, there are several clinical trials involving Met in GBM therapy [reviewed in (29)]. Because GBM remains resistant to treatment and leads to relapse, it is difficult to argue that introducing this drug into the standard treatment strategy could be sufficient. Nevertheless, due to the positive effects of this drug, there remains potential interest in using it as an add-on therapy in GBM.
In this study, we tested the effects of Ril in combination with either Met or Met/Dex to a) investigate the effects of Ril co-administered with Met (diabetic) on U87MG GBM cells and b) investigate the effects of Ril + Met + Dex co-treatment, considering that Dex is commonly used to treat oedema in GBM.
Materials and Methods
Materials and cell cultures. Cell culture media for the U87MG cell line, foetal bovine serum (FBS) and other supplements were purchased from ThermoFisher Scientific (Waltham, MA, USA). Met and Dex were purchased from Sigma-Aldrich (St. Louis, MO, USA), Ril from TOCRIS (Bristol, UK) and TMZ from MSD Sharp & Dohme GmbH (Haar, Germany). Ril, Met and Dex were dissolved in double-distilled water. TMZ was dissolved in dimethylsulfoxide (DMSO) (Sigma-Aldrich). The U87MG cells were grown to 60%-80% confluence in Minimum Essential Medium (ThermoFisher Scientific) supplemented with 10% FBS, 2 mM L-glutamine and 1 mM sodium pyruvate in the presence of penicillin/streptomycin (ThermoFisher Scientific).
Experimental setup. A schematic representation of the experimental setup is shown in Figure 1A. For all Ril groups, U87MG cells were pre-treated with 25 μM Ril, whereas the control (Con) and individual Met treatment groups received FBS-free medium for 72 h (Figure 1). After 72 h, the Con group continued to receive FBS-free medium; the Ril groups received an additional 25 μM Ril; the Met group received 1 mM Met; the Ril+Met group received 25 μM Ril + 1 mM Met; and the Ril+Met+Dex group received 25 μM Ril + 1 mM Met + 10 μM Dex. For radiotherapy, the groups were irradiated with 10 Gy on day 5 (96 h). For the chemotherapy simulation, 100 μM TMZ was added to the respective treatment on day 5 (96 h). After a further 72 h culture (144 h in total), the cells were harvested for cell death and gene expression analyses. For the zymographic assay, the medium was also collected after 72 h (144 h in total). Cells were monitored/recorded over a 3-h period (72 h pre-treatment + 3 h co-treatment) for the migration assay.
Schematic representation of the experimental design (A) used to analyze cell viability (B, C, and D), migration, proteolytic activity of MMP2 and gene expression. The MTT assay was used for examining the effect of single drug or drug combinations. Significant cell death induced by Ril remains after co-administration of Met or Met+Dex, (B) cell death induced by the drugs and their combinations after treatment with radiotherapy or (C) TMZ chemotherapy (D). C: Control; R: riluzole (Ril); M: metformin (Met); D: dexamethasone (Dex); T-temozolomide (TMZ). Significantly different at: *p<0.05, **p<0.01, and ***p<0.001.
Cell viability assay. Cell viability assays were performed using 3-(4, 5-dimethylthiazol)-2, 5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich), as described in Keul et al. (30). For the MTT assay, after 24 h of treatment, cells were irradiated using the RS225A X-Ray Research System (Siemens, Munich, Germany) with a radiation dose of 10 Gy.
Migration assay. The rate of cell migration was monitored in real time using the xCELLigence system CIM plates (ACEA Biosciences Inc., San Diego, CA, USA). The method used has been previously described (30). The groups are described above under ‘Experimental setup’. Each well’s electrode impedance was automatically monitored for 3 h using the xCELLigence system and expressed as a Cell Index (CI).
Zymography assay. For the zymography assay, the culture-conditioned medium was filtered and concentrated to 500 μl to increase the protein concentration using a centrifugal filter with a cut-off of 10 kDa. For each sample, 15 μg of protein extract was loaded onto a 10% SDS-polyacrylamide gel containing 2.5 mg/ml gelatin. As an additional control, the samples were also loaded on a gelatin-free gel to further control and normalize the amount of protein in each chamber. ImageLab Software (BioRad, Hercules, CA, USA) was used for the analysis. The intensity of matrix metalloproteinase 2 (MMP2) activity was normalized to the intensity of the corresponding total protein band on the normalization gel.
RNA purification, reverse transcription and real-time polymerase chain reaction for the analysis of gene expression. See Keul et al. (30) for a detailed description of total cellular RNA isolation, reverse transcription and RT-PCR using gene-specific primers (Table I).
Primer sequence used in this study.
Statistical analysis. The results shown are the mean values of at least three independent experiments (±SEM), unless otherwise stated. The significance of the differences was analyzed using two-sided t-tests for two-group comparisons. Prism 5 statistics software (GraphPad Software, La Jolla, CA, USA) was used for the calculations. A probability of p<0.05 was considered statistically significant.
Results
Ril remained effective in killing GBM cells after co-administration with Met or Met+Dex. Our previous results showed that Ril was effective in killing GBM and stem cells. In this study, we monitored the effect of Ril co-administered with Met or Met+Dex on the death of U87MG cells. A schematic representation of the cell viability setup is shown in Figure 1A. A significant reduction in cell viability was observed after 72 h of treatment with a single Ril or Met treatment. The effect of Ril on cell survival remained significant after the addition of Met or Met+Dex (Figure 1B) but did not show any additive effect when the drugs were used in combination. Further experiments using radiotherapy or TMZ chemotherapy, which is usually the standard treatment, showed no effect compared to co-treatments (Figure 1C and D). Significant differences were observed between Ril or Met alone compared to the Ril+Met co-treatment experiments with +/− TMZ (Figure 1D).
U87MG cell migration after Ril co-treatments with Met or Met+Dex. Migration is an important characteristic of cancer cells and therefore the migration ability of U87MG cells treated with a single or combination of drugs was assessed using the xCELLigence system. After 3 h of Ril application, no significant effect was observed, whereas Met alone significantly increased cell migration. Migration was significantly increased upon co-treatment with Ril+Met, whereas it was significantly decreased upon the addition of Dex to Ril+Met, compared to Ril+Met or Met alone. Unfortunately, Ril+Met+Dex co-treatment still significantly increased migration compared to the control or Ril alone (Figure 2A and B).
Effects of combined treatment on the migration of U87MG cells (A) The migration rate of U87MG cells was monitored with the xCELLigence system (0-3 hours). Data are expressed as mean±SEM. (B) Real-time monitoring of cell migration, representative picture. Significantly different at: *p<0.05, **p<0.01, and ***p<0.001.
Decrease in GBM matrix metalloproteinase 2 (MMP2) activation by drug combination treatment. Cell invasion is an important hallmark of cancer, including GBM. We examined the influence of single drugs and drug combinations on the U87MG cell line by measuring the percentage of MMP2 activity. This activity was measured by zymography using gelatin as a substrate for MMP2 gelatinase. The activity of MMP2 was significantly reduced (Figure 3A) following Ril+Met combination treatment compared to Ril treatment alone. The addition of Dex did not change the percentage of the proteolytically active form of MMP2 compared with the control treatment.
Ril/Met effect on the invasion property of U87MG cells. (A) MMP2 activity under different conditions was assessed using zymography. Diagram showing the results of the three independent experiments. (B) TIMP2 mRNA expression analysis in Ril, Ril+Met and Ril+Met+Dex groups using real-time PCR. Significantly different at: *p<0.05, **p<0.01, and ***p<0.001.
It has been shown that the shift in balance between tissue inhibitor of metalloproteinase 2 (TIMP2) and MMP2 activity in GBM is important for the GBM cell invasion (31). Here, we also examined TIMP2 mRNA expression. TIMP2 was found to be down-regulated in cells treated with Ril alone, Ril+Met and Ril+Met+Dex. Met alone significantly increased TIMP2 gene expression (Figure 3B). The direct role of TIMP2 in cell invasion was not examined in this study.
Evaluation of gene expression in the treatment groups. The use of a drug leads to the activation of pathway(s), which can sometimes cause side effects. We tested the expression levels of genes involved in stemness, glucose metabolism and migration/invasion after treating the cells with Ril, Met, Ril+Met and Ril+Met+Dex. The expression of cluster of differentiation 90 (CD90), which has been described as a stem cell marker and a marker of neural-to-mesenchymal transition, was analyzed. Treatment with Ril or Met alone significantly reduced the expression of CD90. The Ril+Met combination had an even greater effect than the single drug treatments and the addition of Dex maintained a low and significantly reduced expression of this gene compared to the control (Figure 4). We have previously shown that Dex significantly increases the expression of CD90 in GBM cells (30).
Gene expression analysis. Monitoring gene expression in Ril, Ril+Met and Ril+Met+Dex groups using real-time PCR. Analysis was performed for the following genes: (A) CD90; (B) N-Cadherin (CDH2); (C) S100A10; (D) GLUT1 (SLC2A1); (E) GLUT3 (SLC2A3); (F) TFPi2. Significantly different at: *p<0.05, **p<0.01, and ***p<0.001.
N-cadherin expression was down-regulated by both Ril and Met alone. A combination of these drugs and their combination with Dex also significantly down-regulated N-cadherin expression. Furthermore, the expression levels of genes involved in glucose metabolism were examined. S100 calcium-binding protein A10 (S100A10) was reduced by Met alone and was also significantly down-regulated by the combination of all three drugs. GLUT1 expression was increased with Met treatment alone. GLUT3 expression was significantly down-regulated upon treatment with Ril but was significantly up-regulated in the Met treatment group. Interestingly, the Ril+Met combination significantly reduced GLUT3 gene expression compared to Met treatment alone and to the control. The addition of Dex to the treatment significantly reduced GLUT3 expression compared to the control (Figure 4).
The down-regulation of tissue factor pathway inhibitor 2 (TFPi2) in GBM has been associated with increased invasion. A significant increase in TFPi2 expression was observed in the Ril group (Figure 4F), whereas Met alone significantly decreased its expression compared to the control group. The addition of Met or Met+Dex reduced the effect of Ril. However, the effect was still positive when compared to the control group.
Discussion
The effect of Ril on cell survival remained stable after adding Met and Met+Dex. The application of standard of care therapy did not influence the Ril effect, which remained the most effective drug in killing U87MG GBM cells. El Hassan et al. reported that Met significantly reduced the viability of two different cell lines in a dose-dependent manner compared to the untreated controls (23). However, Würth et al. observed that differentiated cells were completely insensitive to Met, with only a slight reduction in cell survival at the highest concentration tested (50 mM for 48 h). Nevertheless, the viability of glioblastoma stem-like cells was almost completely suppressed (32).
To improve survival in patients with GBM, it is crucial to screen for GBM-targeted anticancer agents with anti-migration/-invasion potential. Assessment of the migration ability of U87MG after these co-treatments showed a significant increase when Met was included. Combining Ril with Met or Met+Dex increased cell migration compared to the control. The observed significant reduction in migration with Ril+Met+Dex compared to Met or Ril+Met could be relevant when considering the use of Ril+Met for treatment. In our experimental settings, Met alone increased cell migration, in contrast to previous reports in which Met alone showed a decrease in the total distance migrated by U87 cells (23). It is important to note that the treatment regimen in those experiments was different from that of this study. In our experimental settings, 1 mM Met for 3 h was used, whereas Al Hassan et al. used 2.5 mM Met for 24 h. The specific effect of Met on migration was also shown in different GBM stem-like cells, some of which were more or less sensitive to the anti-migratory effect of Met (33).
In contrast, the combination of Ril and Met significantly reduced the levels of the active enzyme MMP2, which is primarily associated with the invasive properties of GBM (34-36). MMPs require proteolytic activation involving cleavage of a pro-peptide domain to exhibit enzymatic activity (37). Both deficiency and excess of TIMP2 can prevent MMP2 activation, and small shifts in the balance between TIMP and MMP levels can strongly modulate the invasive phenotype of tumor cells (31). Lu et al. showed that the GBM more invasive U87-C1 cells have a three-fold increase in MMP2 activation accompanied by increased TIMP2 expression. In this study, Ril+Met treatment significantly reduced MMP2 activation compared to Ril treatment alone. The expression of TIMP2 was also reduced by this treatment, which could also indicate reduced invasion. We did not directly measure the invasion of these cells, but interestingly, migration, which is most often associated with invasive properties, was increased by this treatment. There are also reports that in GBM cells, GLUT3, but not GLUT1, is directly involved in their invasion (38). Changes in the organization of the extracellular matrix induced by GLUT3 over-expression have been reported (38). Previous reports have shown that ionizing radiation increases MMP2 activity and protein secretion in glioma cells, while MMP2 siRNA inhibits radiation-induced invasiveness of U87MG and U-251 cells (39). In our case, it would be interesting to examine whether co-treatment is effective in reducing MMP2 activation after irradiation.
Furthermore, the knockdown of TFPi2 is associated with migration and invasion of GBM cells (40). The effect of co-treatments on TFPi2 expression level was also examined. Ril alone significantly increased TFPi2 expression and the addition of Met and Dex reversed this effect. The direct role of these two particular genes in migration/invasion processes was not investigated. Usually, migration and invasion are tightly connected processes that have been described in GBM. Nonetheless, there are reports of a disconnection between migration and invasion during carcinoma-associated EMT in vitro and in vivo (41).
Important correlations between treatments and alterations in gene expression profiles have been identified and described (9, 42). Met reduced the expression of key genes of the hypoxia gene signature in GBM cells (42). Glucocorticoid signaling regulates the expression of hundreds of genes [reviewed in (43)], whose dysregulation could favor a poor prognosis in patients with GBM (9).
These co-treatments decreased expression of the markers CD90, N-cadherin, S100A10 and GLUT3, which are important for stem cell properties and glucose metabolism. Ril alone significantly decreased the expression of these genes, but co-treatment resulted in a further decrease. The up-regulation of CD90 and GLUT3 is directly associated with the acquisition of a stem cell state and the ability of tumors to propagate in vivo (3, 44). However, GLUT1 expression decreased after Ril treatment, but significantly increased after Ril+Met co-treatment. In the brain, GLUT3 is recognized as the neuronal glucose transporter, whereas GLUT1 is important for glucose uptake in astrocytes mediating the transport of glucose across the blood–brain barrier (45). Therefore, it would be interesting to assess the Ril+Met combination effect in in vivo mouse models. The S100A10 gene was significantly down-regulated by Ril and co-treatments indicating their significant anti-GBM effect, considering that S100A10 is remarkably highly expressed in high-grade glioma (46). All combined treatments were shown to down-regulate CD90. It is interesting to note that the co-administration of Met reversed TMZ-induced CD90 expression (47). However, when given alone, Met increased CD90 in the U251 cell line (47). The addition of Dex to this combination of drugs did not alter the effect on CD90 and GLUT3, but reduced GLUT1 levels after Ril+Met treatment.
Finally, we did not find any changes in N-cadherin gene expression. The role of N-cadherin in gliomas is conflicting. N-cadherin protein expression levels have been shown to increase with the increase of the WHO grade of glioma and to correlate with the Ki-67 labelling index, suggesting a role for cell adhesion signaling in tumor cell proliferation and dedifferentiation [reviewed in (48)].
Conclusion
We showed that the combination of Ril with Met or Met+Dex had positive but also some negative effects on GBM cells. Co-treatment with Ril had a positive effect, as follows: killing GBM cells, reducing the expression of genes involved in glucose metabolism and stemness, and reducing MMP2 activation. The negative effect included the increased migration of these cells. The use of Dex in combination with these two drugs reduced the migration of these cells almost to the control level. Although this study is preliminary and requires further work involving more cell lines and, most importantly, in vivo experiments, it shows the potentially beneficial role of combination treatments for GBM. It would also be interesting to monitor the effect of Ril and Ril+Met on GBM oedema. Ril has been shown to attenuate cytotoxic brain oedema after focal cerebral ischemia and to significantly reduce brain oedema development. The multidimensional effect of Met on the pathomechanism of cerebral oedema has also been documented in several studies, including GBM-induced brain oedema in a mouse model (49). Overall, some of these drug combinations could also be advantageous for reducing the concentration of Dex to minimize its side effects.
Footnotes
Authors’ Contributions
The work reported in this article has been performed by the authors, unless clearly specified in the text. Jonathan Keul: Conceptualization, Methodology, Formal Analysis, Investigation, Writing-Original Draft. Swetlana Sperling: Conceptualization, Methodology, Validation, Writing – Review & Editing. Veit Rohde: Funding, Writing-Review & Editing. Milena Ninkovic: Conceptualization, Methodology, Investigation, Writing – Review & Editing, Supervision.
Conflicts of Interest
The Authors declare no conflicts of interest with respect to the research, authorship and/or publication of this article. They also do not have relevant or non-financial interests to disclose.
Funding
The Authors declare that no funds, grants or other support were received during the preparation of this manuscript.
- Received March 3, 2025.
- Revision received March 24, 2025.
- Accepted March 26, 2025.
- Copyright © 2025 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.