Abstracts

Rationale: Reduced levels of the progesterone metabolite allopregnanolone (AP) during menses may be involved in catamenial epilepsy. AP enhances GABA_A receptor (GABAR) currents and alters GABAR subunit expression both in vivo and in vitro by an unknown mechanism. AP-induced expression changes were blocked by bicuculline, suggesting mediation by GABARs. We hypothesized that AP alters neuronal transcription by increasing intracellular Ca²⁺ levels, possibly via depolarizing GABA currents. Methods: Fura-2- loaded rat primary cortical neurons (13-15 DIV) in combination with calcium imaging were used to test the ability of AP and GABA to alter intracellular Ca²⁺ levels. Pharmacologic agents were applied by superfusion during Ca²⁺ imaging. Results: Acute 5 s application of AP (1 pM - 30 μM) evoked transient concentration-dependent Ca²⁺ signals with an EC₅₀ of 72 nM. Similarly, 5 s applications GABA (1 nM - 3 mM) evoked Ca²⁺ signals with an EC₅₀ of 96 μM. Treatment with cholesterol or 3β-AP, an inactive stereoisomer of AP, failed to evoke Ca²⁺ transients, suggesting stereospecificity for AP. Likewise, AP application following removal of extracellular Ca²⁺ or treatment with 2-aminoethoxydiphenyl borate (2-APB, 100 μM), an antagonist of inositol 1,4,5-trisphosphate receptors (IP3Rs) failed to evoke Ca²⁺ transients, indicating that the Ca²⁺ was not derived from internal stores. In contrast, AP or GABA-induced Ca²⁺ transients were blocked by nifedipine treatment (10 μM) indicating involvement of L-type voltage-gated Ca²⁺ channels (L-VGCCs). To determine whether L-VGCC currents were activated by depolarizing GABA-evoked currents, we tested whether Ca²⁺ transients were altered by chloride reversal potential (E_Cl) manipulation. GABA- and AP-evoked Ca²⁺ signals were blocked by preincubation with bumetanide (10 μM), an inhibitor of the Na/K/Cl cotransporter NKCC-1, which causes Cl^- accumulation in immature neurons. Moreover, Ca²⁺ signals were enhanced by substitution of NaPropionate for NaCl in the extracellular medium, which would cause a positive shift of E_Cl. GABA- and AP-evoked Ca²⁺ signals were partially blocked by the GABA antagonist, picrotoxin, but not affected by the GABAR-specific antagonist, bicuculline (10 μM), or the GABA_B receptor antagonist, saclofen (100 μM), nor were they blocked by the glycine receptor antagonist, strychnine. Conclusions: These data suggest a novel pathway by which AP and GABA evoke L-VGCC mediated Ca²⁺ signals via depolarizing GABA currents with unusual pharmacology, which may mediate AP transcriptional signaling. This pathway could participate in altered GABAR subunit expression associated with catamenial epilepsy.