Narrow Results

The interaction of ladder polyethers of marine origin, like ciguatoxin 3C and brevenal, as well as hypothetic ent-brevenal, with the human voltage-gated Kv1.5 potassium ion channel is investigated in this work using homology modeling, automated docking, and energy scoring from molecular dynamics (MD) simulations. A 3D homology model of the pore region of the Kv1.5 channel, previously developed from the 2.9 Å resolution crystal structure of the mammalian Kv1.2 channel — which has a very similar pore sequence — is used here. While ciguatoxin 3C did not enter the pore, both brevenal and ent-brevenal were found into the pore, the latter one with the best score. Binding is attended by notable strain in the ligands, and the corresponding energy increase was evaluated for ent-brevenal by self consistent field (SCF) and density functional theory (DFT) procedures. Egress of ent-brevenal from the pore, as a microscopical reversal of the ingress, was investigated by a smart form of biased MD simulations. While this study indicates ample room and attractive interactions for both brevenal and its enantiomer into the pore, whether these molecules will be found to inhibit voltage-gated potassium ion currents depends upon the barriers in the real system to access the pore, with their thermodynamic and kinetic requirements.

The interaction of the K⁺-sparing agent amiloride — a synthetic chlorinated pyrimidine derivative — with the hASIC1a ion channel is investigated here along homology modeling of the pore region (using the crystal structure of the cASIC1 channel as a template and the known sequence of hASIC1a), automated docking (using the NMR solution structure of amiloride and its conjugated acid, refined by computations), and molecular dynamics simulations. This represents the first modeling and computational chemistry of the pore region of ASIC/DEG/ENaCs/FaNaCh channels. The results agree with the putative amiloride binding site for alphaENaC channel chimeras once the amiloride free base is considered, while its conjugated acid — in contrast with literature beliefs — is poorly scored on a nearby protein pocket. Different protonation conditions of the pore region are irrelevant because histidine residues are far from the binding sites. Mapping the amino acids of the homology model closest to amiloride can have heuristic value in stimulating in silico search of new pore-blocking agents, experimental studies of ASIC channels themselves, and development of code for constant-pH MD simulations.