Abstracts

Rationale: In vitro brain slice preparations are typically bathed in artificial cerebral spinal fluid (ACSF) with glucose concentrations between 10-20 mM.  These concentrations are relatively high compared to in vivo conditions as lack of cerebral vascular autoregulation in slices requires higher glucose concentration to support diffusion-dependent delivery to achieve physiological concentrations of 2.5-4.4 mM.  In rat hippocampal slices, we evaluated the effects of a low glucose concentration (2.5 mM) and the glucose analogue and glycolytic inhibitor 2D-deoxyglucose (2DG) on normal and epileptiform synaptic network activity in the CA3 region.   Methods: Whole-cell voltage-clamp recordings were made from the CA3 pyramidal neurons.  Spontaneously occurring excitatory and inhibitory post synaptic currents (PSCs) were measured in 3.5 and 7.5 mM extracellular potassium concentrations ([K+]o).   EPSCs were measured at -70 mV and IPSCs at 0 mV.  The effects of lowering glucose from 10 to 2.5 mM and also 2.5 mM glucose with 2.5 mM 2DG were assessed. Results: 2.5 mM glucose significantly decreased the frequency (1.2 ± 0.3 vs 0.48 ± 0.1 Hz, n = 7) but not amplitude (20.1 ± 1.6 vs. 18.7 ± 0.8 pA) of sEPSCs in 3.5 mM [K+]o.   Addition of 2.5 mM 2DG did not further change the frequency (0.63 ± 0.09) or amplitude (17.9 ± 1.4 pA) of sEPSCs.  Similar results were observed for sIPSCs in 3.5 mM [K+]o; 2.5 mM glucose reduced significantly the frequency (6.6 ± 0.8 vs 5.4 ± 0.7 Hz, n = 11) without an accompanying change in amplitude (29.8 ± 1.6 vs. 28.4 ± 1.7 pA) and further addition of 2.5 mM 2DG had no significant effects.  The magnitude of the frequency reduction produced by 2.5 mM glucose was greater for EPSCs than IPSCs (60 vs 18%).In 7.5 mM [K+]o, compound bursts of PSCs occurred and we measured the area of current (the charge) per second over 3 minute epochs.  The excitatory charge was reduced significantly by 2.5 mM glucose and significantly further decreased by the addition of 2.5 mM 2DG (13.4 ± 2.8 vs. 7.2 ± 2.2 vs. 3.2 ± 0.9 nC/s,  n = 7).  A similar effect was observed on the inhibitory charge (33.1 ± 4.5 vs. 25.0 ± 4.0 vs. 16.9 ± 2.8 nC/s, n = 11).  The relative reduction of low glucose was greater on excitatory (47%) compared to inhibitory (25%) charge.  2DG further reduced excitatory (77%) more than inhibitory charge (49%).  Conclusions: :  Low glucose, but not 2DG, decreased the frequency of spontaneously occurring PSCs in 3.5 mM [K+]o demonstrating a presynaptic need for glucose as an energy source.  The effect was greater on excitatory synaptic transmission and appears to be a result of a presynaptic dependence on glucose for transmitter release.  In the presence of epileptiform burst activity, both low glucose and 2DG reduced excitatory more than inhibitory charge.   Reduced glucose and 2DG have synergistic effects in suppressing epileptiform bursts in the hippocampus and demonstrate anticonvulsant actions of reduced glycolytic flux and glycolytic inhibition on epileptic network synchronization that likely contributes to anticonvulsant efficacy of 2DG in multiple acute and chronic rodent models of seizures and epilepsy.  Funding: VA Research and Development