Neurobiology of Disease (Oct 2019)
Attenuation of epileptogenesis by 2-deoxy-d-glucose is accompanied by increased cerebral glucose supply, microglial activation and reduced astrocytosis
Abstract
Rationale: Neuronal excitability and brain energy homeostasis are strongly interconnected and evidence suggests that both become altered during epileptogenesis. Pharmacologic modulation of cerebral glucose metabolism might therefore exert anti-epileptogenic effects. Here we provide mechanistic insights into effects of the glycolytic inhibitor 2-deoxy-d-glucose (2-DG) on experimental epileptogenesis by longitudinal 2-deoxy-2[18F]fluoro-d-glucose positron emission tomography ([18F]FDG PET) and histology. Methods: To imitate epileptogenesis, 6 Hz-corneal kindling was performed in male NMRI mice by twice daily electrical stimulation for 21 days. Kindling groups were treated i.p. 1 min after each stimulation with either 250 mg/kg 2-DG (CoKi_2-DG) or saline (CoKi_vehicle). A separate group of unstimulated mice was treated with 2-DG (2-DG_only). Dynamic 60-min [18F]FDG PET/CT scans were acquired at baseline and interictally on days 10 and 17 of kindling. [18F]FDG uptake (%injected dose/cc) was quantified in predefined regions of interest (ROI) using a MRI-based brain atlas, and kinetic modelling was performed to evaluate glucose net influx rate Ki and glucose metabolic rate MRGlu. Furthermore, statistical parametric mapping (SPM) analysis was applied on kinetic brain maps. For histological evaluation, brain sections were stained for glucose transporter 1 (GLUT1), astrocytes, microglia, as well as dying neurons. Results: Post-stimulation 2-DG treatment attenuated early kindling progression, indicated by a reduction of fully-kindled mice, and a lower overall percentage of type five seizures. While 2-DG treatment alone led to globally increased Ki and MRGlu values at day 17, kindling progression per se did not influence glucose turnover. Kindling accompanied by 2-DG treatment, however, resulted in regionally elevated [18F]FDG uptake as well as increased Ki at days 10 and 17 compared both to baseline and to the 2-DG_only group. In hippocampus and thalamus, higher MRGlu values were found in the CoKi_2-DG vs. the CoKi_vehicle group at day 17. t maps resulting from SPM analysis generally confirmed the results of the ROI analysis, and additionally revealed increased MRGlu restricted to the ventral hippocampus when comparing the CoKi_2-DG and the 2-DG_only group both at days 10 and, more distinct, day 17. Immunohistochemical staining showed an attenuated kindling-induced regional activation of astrocytes in the CoKi_2-DG group. Interestingly, 2-DG treatment alone (and also in combination with kindling, but not kindling alone) led to increased microglial activation scores, whereas neither staining of GLUT1 nor of dying neurons revealed any differences to untreated controls. Conclusions: Post-stimulation treatment with 2-DG exerts disease-modifying effects in the mouse 6 Hz corneal kindling model. The observed local increase in glucose supply and turnover, the alleviation of astroglial activation and the activation of microglia by 2-DG might contribute separately or in combination to its positive interference with epileptogenesis.
