Title

 

Authors

M. Chisari^1,2, K. Wu², C. F. Zorumski^2,3, S. Mennerick^2,3
 
1Department of Clinical and Molecular Biomedicine, Section of Pharmacology and Biochemistry, University of Catania, Italy,  Departments of ²Psychiatry and ³Anatomy & Neurobiology, Washington University School of Medicine, St. Louis, MO, USA.

 

Abstract

Many compounds that modulate GABA_A receptors are experimentally and clinically useful as anesthetics, anticonvulsants and anxiolytics. Improving our understanding of these receptors might lead to better drug design and enhance both research and therapy. Among non-competitive antagonists, sulfated neurosteroids show a high potency at GABA_A receptors, but a structure-activity relationship is unclear since enantioselectivity is weak. Thus, general properties (negative charge, lipophilicity) rather than specific properties may be important. Here we tested dipicrylamine (DPA), a probe of membrane excitability and of voltage-gated ion channel function. Like sulfated neurosteroids, DPA is a hydrophobic anion that increases membrane capacitance. Based on these broad similarities, we tested DPA on GABA_A receptor function in Xenopus laevis oocytes and hippocampal neurons using voltage-clamp techniques and found that it is a very potent GABA_A receptor antagonist, with an IC₅₀ near 50 nM at half-maximal GABA concentration. Antagonism exhibited dependence on channel activation, like sulfated neurosteroids and picrotoxin. Like pregnenolone sulfate, DPA antagonism was dramatically reduced by a point mutation of the alpha1 subunit. This mutation does not affect picrotoxin antagonism. Kinetics of DPA removal were slow, and the application of bovine serum albumin speeded DPA washout in both capacitance changes and GABA antagonism suggesting that an interaction with a putative binding site is very weak or absent. DPA translocates across the membrane leaflets in a voltage-dependent manner, and we exploited this feature to explore the relationship between membrane interactions and antagonism. Although DPA-induced membrane capacitance changes were strongly voltage dependent, steady-state antagonism was not. However washout of GABA_A receptor antagonism in neurons was severely retarded at positive potentials compared to negative potentials, paralleling measures of membrane DPA loss. We conclude that membrane interaction is likely important for DPA antagonism, but DPA in either membrane leaflet is an effective inhibitor. At concentrations that effectively antagonize exogenous GABA responses, DPA failed to inhibit evoked inhibitory postsynaptic currents (IPSCs) or speed their decay time. Antagonism of IPSCs was, however, evident on pharmacologically prolonged events. These observations support slow, activation-dependent antagonism. Taken together, these results are consistent with the idea that plasma membrane interactions participate in the modulation of GABA_A receptors, and antagonism of these ionotropic receptors should be accounted for in studies using DPA to monitor neuronal excitability.