Abstracts

Rationale: Female reproductive cycle-associated hormonal fluctuations underlie catamenial seizure exacerbation. Progesterone is believed to have anti-convulsant action. In the experimental animals progesterone protects against kindling or chemoconvulsant-induced seizures, via its conversion to allopregnanolone, which enhances GABA-A receptor-mediated inhibition. However, a 14-center phase III clinical trial revealed that a two week progesterone therapy was not better than placebo. In menopausal women hormone therapy even worsened the seizures. Thus the effects of chronic progesterone treatment on neurotransmission in an epileptic brain need to be carefully evaluated. We determined the effect of chronic progesterone treatment on seizure frequency, and AMPA receptor (AMPAR)-mediated neurotransmission in the CA1 pyramidal neurons (PN).Methods: Adult female rats which became epileptic following lithium-pilocarpine-induced status epilepticus were used. The progesterone levels were elevated using a pseudo pregnancy regime (PMSG 20 IU, followed 48 hrs later by -HCG 10 IU). The sEPSCs were recorded from PN using standard whole cell patch clamp technique. The surface expression of GluR1 subunit of AMPARs was determined on the 8th day of -HCG administration using a BS3 assay. In some animals progesterone levels were elevated by twice daily injections of progesterone (50 mg/kg, i. p). Seizure activity was recorded in these animals by continuous video-EEG recording via cortical and hippocampal electrodes. Results: Prolonged progesterone treatment was ineffective in controlling seizures, which became more frequent with continued administration of progesterone. The seizures were completely suppressed during 24-36 hrs after the first progesterone injection in some animals; however the seizures returned to basal frequency, determined during 10 days prior to start of the treatment, and became more frequent with subsequent progesterone injections. Excitatory neurotransmission was also enhanced in pseudo pregnant animals. The sEPSCs recorded from PN of pseudo pregnant epileptic animals (n = 14) were larger (55.8 4.4 pA) and more frequent (2.5 0.4 Hz) than those in non-pregnant epileptic animals (14.3 2.9 pA and 0.64 0.4 Hz, n= 5, p<0.05). The surface GluR1 subunit expression also increased in pseudo pregnant epileptic animals, the intracellular fraction of GluR1 subunit normalized to total GluR1 expression in the whole hippocampus was 47 6% (n=3, p<0.05) of that in non-pregnant epileptic animals. Progesterone treatment also enhanced excitatory neurotransmission in the non-epileptic pseudo pregnant animals. The amplitudes of sEPSCs recorded from PN of non-epileptic pseudo pregnant animals were larger (19.8 3 pA n = 7, vs 10.75 0.4 pA, n = 16, p<0.05) than those in na ve animals, and the normalized intracellular fraction of GluR1 subunit was also reduced (67 8% of controls, n=3, p<0.05). Conclusions: These results suggest that an up-regulation of excitatory neurotransmission could underlie a loss of anti-convulsant action of prolonged progesterone treatment in epileptic animals.