The Central Role of Glycolysis in Glioblastoma

Normal, differentiated cells rely primarily on mitochondrial oxidative phosphorylation for energy production. In contrast, many cancer cells, including glioblastoma multiforme (GBM), the commonest primary brain tumor in adults, appear to rely on anaerobic glycolysis instead.¹ Moreover, therapeutic agents already in clinical use such as anti-vascular endothelial growth factor (anti-VEGF) antibodies create a tumor microenvironment that further favors a metabolic shift in tumor cells toward glycolysis.² Glycolysis inhibition is therefore a potential therapeutic strategy, and a recent analysis³ of GBM glycolytic enzymes reveals promising findings and potential pharmacological targets for GBM therapy.

Sanzey et al³ exposed several different GBM cell types to hypoxic and normoxic conditions and used genome-wide transcription analysis to identify any genes that were upregulated during hypoxia but not during normoxia. Among these were several glycolysis-related genes, and they selected 7 promising glycolysis-related genes for further analysis. They modified GBM cells by silencing these 7 glycolysis-related genes and used a cell viability assay to show that, under hypoxic conditions, GBM cells with impaired glycolysis demonstrate decreased proliferation and increased death in vitro compared with nonmodified cells.

In addition, the investigators implanted a mixture of modified (glycolysis-silenced) and control GBM cells into mouse brain, harvested the xenograft tumor after 7 weeks, and used quantitative polymerase chain reaction to show that modified cells, but not control GBM cells, were relatively depleted, indicating that glycolysis-silenced cells had a growth disadvantage in vivo. They further customized GBM cells by creating an independent knockdown cell model for each of the 7 glycolysis-related genes and used a mouse xenograft model to show a significant increase in host survival with impaired glycolysis (eg, survival benefit, +21.8% phosphofructokinase[PFK] knockdown, +20.9% pyruvate dehydrogenase kinase 1 [PDK1] knockdown; Figure).

Figure.

Mouse survival study. Kaplan-Meier curves demonstrate the effect of glycolytic gene knockdown on mouse survival. HK2, hexokinase 2; PFKP, phosphofructokinase platelet (encodes PFK1, phosphofructokinase 1); ALDOA, aldolase A; PGAM1, phosphoglycerate mutase 1; ENO1, enolase 1; ENO2, enolase 2; PDK1, pyruvate dehydrogenase kinase 1. Modified from Sanzey M, Abdul rahim SA, Oudin A, et al. Comprehensive analysis of glycolytic enzymes as therapeutic targets in the treatment of glioblastoma. PLoS One. 2015;10(5):e0123544.

Finally, they attempted to pharmacologically target glycolysis using several different available compounds. For example, clotrimazole is known to interfere with PFK. In a mouse xenograft GBM model, the investigators demonstrated that clotrimazole 150 mg/kg 3 times weekly improved mouse survival (+7 days; P = .02) compared with controls.

These experiments reaffirm the central role of glycolysis in GBM, reveal several potential therapeutic targets (eg, PFK, PDK1), and demonstrate how pharmacologically inhibiting glycolysis may prove to be an effective treatment strategy. In addition to developing antiglycolytic compounds,⁴ other approaches to inhibiting glycolysis such as dietary modification are gaining increased attention.⁵ Any treatment strategy that undermines glycolysis will also need to be balanced against the disproportionately high blood glucose metabolism of the normal brain.⁶ Especially because anti-VEGF antibody therapy encourages GBM cells to become increasingly dependent on glycolysis, it is interesting to consider how anti-angiogenic therapy coupled with antiglycolytic therapy could potentially provide a synergistic one-two punch.