Neuroprotection by propofol in acute mechanical injury: role of GABAergic inhibition.

1. Whole cell patch-clamp and extracellular field recordings were obtained from granule cells of the dentate gyrus in 400-microns-thick brain slices of the adult rat to determine the actions of the intravenous general anesthetic 2,6-diisopropylphenol (propofol) on acute neuronal survival and preservation of synaptic integrity after amputation of dendrites (dendrotomy), and to determine the role of gamma-aminobutyric acid-A (GABAA)-receptor-mediated inhibition in the neuroprotective effects of propofol. The actions of propofol were compared with those exerted by another widely used intravenous general anesthetic, 5-ethyl-5-[1-methylbutyl]-2-thiobarbituric acid (thiopental). 2. Propofol (10 microM) increased the frequency (control: 5.9 +/- 0.9 Hz, mean +/- SE; propofol: 10.5 +/- 1.3 Hz) and the single-exponential decay time constant (tau D) (control: 4.5 +/- 0.2 ms; propofol: 15.3 +/- 1.5 ms) of miniature inhibitory postsynaptic currents (mIPSCs) recorded in control neurons. Thiopental (25 microM) also increased the tau D (14.3 +/- 0.9 ms) of mISPCs, but had no effect on mIPSC frequency. Both anesthetics potentiated mIPSCs at low concentrations (propofol: 5 microM; thiopental: 1 microM). Propofol and thiopental did not change the peak amplitude and rise times of mIPSCs. 3. Propofol (10 microM) was able to depress the excitability of control granule cells, as determined by the reduction in the amplitude of the orthodromic population spikes. This depression could be prevented by the GABAA receptor antagonist bicuculline (50 microM), indicating that propofol reduces excitability via GABAA receptor functions. 4. Propofol and thiopental were neuroprotectant (assessed by antidromic population responses 2-5 h after injury) if present before and during the amputation of the granule cell dendrites. The protective actions were dose dependent, and at high doses (propofol: 200 microM; thiopental: 400 microM) the anesthetics were as neuroprotective against dendrotomy-induced cell death as 2-amino 5-phosphovaleric acid (APV) and 6-cyano-7-nitroquinoxaline-2,3 dione (CNQX). The protective effects of the anesthetics were completely blocked with the GABAA receptor antagonists picrotoxin or bicuculline, and were mimicked by the GABAA receptor agonist muscimol (100 microM). 5. Propofol, in contrast to APV and CNQX, could not prevent the dendrotomy-induced Ca(2+)-dependent and long-lasting changes in mIPSC decay kinetics (appearance of a double-exponential, prolonged decay). 6. The protective effects of the anesthetics and those of APV and CNQX on neuronal survival were not significant when the drugs were applied after dendrotomy, indicating that dendrotomy carried out 150-200 microns from the soma without neuroprotective agents rapidly induces irreversible acute degeneration in most injured neurons. The failure to rescue cells from dendrotomy-induced injury did not result from a decreased sensitivity of the GABAA receptors to the anesthetics, because the potentiating effects of the anesthetics on mIPSCs from control and dendrotomized neurons were not different. 7. These data indicate that propofol potentiates synaptic inhibition pre- and postsynaptically, and, when present during dendrotomy, it can protect neurons from acute mechanical-injury induced cell death via potentiation of GABAA receptor functions. However, propofol fails to provide neuroprotection against dendrotomy-induced changes in synaptic physiology.