The mechanisms of potentiation and inhibition of GABAA receptors by non-steroidal anti-inflammatory drugs, mefenamic and niflumic acids

Fenamates mefanamic and niflumic acids (MFA and NFA) induced dual potentiating and inhibitory effects on GABA currents recorded in isolated cerebellar Purkinje cells using the whole-cell patch-clamp and fast-application techniques. Regardless of the concentration, both drugs induced a pronounced prolongation of the current response. We demonstrated that the same concentration of drugs can produce both potentiating and inhibitory effects, depending on the GABA concentration, which indicates that both processes take place simultaneously and the net effect depends on the concentrations of both the agonist and fenamate. We found that the NFA-induced block is strongly voltage-dependent. The Woodhull analysis of the block suggests that NFA has two binding sites in the pore - shallow and deep. We built a homology model of the open GABA_AR based on the cryo-EM structure of the open α1 GlyR and applied Monte-Carlo energy minimization to optimize the ligand-receptor complexes. A systematic search for MFA/NFA binding sites in the GABA_AR pore revealed the existence of two sites, the location of which coincides well with predictions of the Woodhull model. In silico docking suggests that two fenamate molecules are necessary to occlude the pore. We showed that MFA, acting as a PAM, competes with an intravenous anesthetic etomidate for a common binding site. We built structural models of MFA and NFA binding at the transmembrane β(+)/α(-) intersubunit interface. We suggested a hypothesis on the molecular mechanism underlying the prolongation of the receptor lifetime in open state after MFA/NFA binding and β subunit specificity of the fenamate potentiation.

Block of GABA(A) receptor ion channel by penicillin: electrophysiological and modeling insights toward the mechanism

GABA(A) receptors (GABA(A)R) mainly mediate fast inhibitory neurotransmission in the central nervous system. Different classes of modulators target GABA(A)R properties. Penicillin G (PNG) belongs to the class of noncompetitive antagonists blocking the open GABA(A)R and is a prototype of β-lactam antibiotics. In this study, we combined electrophysiological and modeling approaches to investigate the peculiarities of PNG blockade of GABA-activated currents recorded from isolated rat Purkinje cells and to predict the PNG binding site. Whole-cell patch-сlamp recording and fast application system was used in the electrophysiological experiments. PNG block developed after channel activation and increased with membrane depolarization suggesting that the ligand binds within the open channel pore. PNG blocked stationary component of GABA-activated currents in a concentration-dependent manner with IC50 value of 1.12mM at -70mV. The termination of GABA and PNG co-application was followed by a transient tail current. Protection of the tail current from bicuculline block and dependence of its kinetic parameters on agonist affinity suggest that PNG acts as a sequential open channel blocker that prevents agonist dissociation while the channel remains blocked. We built the GABA(A)R models based on nAChR and GLIC structures and performed an unbiased systematic search of the PNG binding site. Monte-Carlo energy minimization was used to find the lowest energy binding modes. We have shown that PNG binds close to the intracellular vestibule. In both models the maximum contribution to the energy of ligand-receptor interactions revealed residues located on the level of 2', 6' and 9' rings formed by a bundle of M2 transmembrane segments, indicating that these residues most likely participate in PNG binding. The predicted structural models support the described mechanism of PNG block.