Abstracts

RATIONALE: Non-synaptic mechanisms, which comprise glial functions and modulation of the composition of the extracellular space, may play important roles in the initiation and propagation of abnormal electrical discharges. In particular, astrocytes have a well-established role in maintenance of the extracellular space, both through control of extracellular potassium levels and pH, and as a supportive framework for neural tissue. Because mice lacking the gene for GFAP, an intermediate filament protein, have abnormal glial morphology and a lower electrical threshold for the development of LTP, we postulated that these animals might have (i) altered hippocampal excitability and (ii) abnormal responsiveness to compounds that normally reduce in vitro seizure activity.
METHODS: We compared acute hippocampal slices (400 [mu]m; horizontal) obtained from GFAP KO and age-matched, wild-type (WT) control mice. Input-output curves and paired-pulse stimulation responses were obtained by stimulating the Schaffer collaterals and recording from CA1. Hippocampi were perfused with artificial CSF containing 4-aminopyridine (4-AP,100 [mu]M) until spontaneous epileptiform bursts occurred at regular intervals (~30 min). Next, the following compounds were added to the bathing medium: furosemide (2.5 mM), carbenoxolone (100 [mu]M), acetazolamide (2 mM) and mannitol (5.0 g/l to increase osmolarity from 297 [plusminus] 4 to 325 [plusminus] 5). Recording was continued until there was cessation of burst activity or for 90 min.
RESULTS: Although hippocampi from GFAP KO and WT mice had identical population spike waveforms and input/output function, slices from GFAP KO mice exhibited a heightened response to paired-pulse excitation (KO: 14 slices from 5 mice; WT: 17 slices from 7 mice). Bath application of furosemide, carbenoxolone or acetazolamide were equally effective in reducing or eliminating 4-AP induced burst discharge in slices from KO and WT mice. However, hippocampal slices from GFAP KO animals showed much less inhibition of 4-AP induced bursts when exposed to mannitol (KO: 7 slices in 4 mice; WT: 11 slices in 6 mice).
CONCLUSIONS: We interpret our findings as evidence that astrocytes play a role in modulating brain excitability, at least in part, through regulation of the extracellular space. This conclusion is based on the following observations: (i) the reported larger extracellular space in GFAP KO mice [Gimenez et al. Glia 31:69-83, 2000] does not alter the response to furosemide (a chloride cotransporter blocker), acetazolamide (which reduces pH), or carbenoxolone (a gap junction blocker) and (ii) the response to mannitol, which presumably reduces glial-mediated cell swelling associated with seizures, is significantly impaired. In conclusion, our data suggest a disruption of non-synaptic mechanisms in mice lacking glial fibrillary acidic protein.
Support: Sandler Family Supporting Foundation