Fast C-type inactivation confers distinctive functional properties to the hERG potassium channel, and its association to inherited and acquired cardiac arrythmias makes the study of the inactivation mechanism of hERG at the atomic detail of paramount importance. At present, two models have been proposed to describe C-type inactivation in K+-channels. Experimental data and computational work on the bacterial KcsA channel support the hypothesis that C-type inactivation results from a closure of the selectivity filter that sterically impedes ion conduction. Alternatively, recent experimental structures of a mutated Shaker channel revealed a widening of the extracellular portion of the selectivity filter, which might diminish conductance by interfering with the mechanism of ion permeation. Here, we performed molecular dynamics simulations of the wild-type hERG, a non-inactivating mutant (hERG-N629D), and a mutant that inactivates faster than the wild-type channel (hERG-F627Y) to find out which and if any of the two reported C-type inactivation mechanisms applies to hERG. Closure events of the selectivity filter were not observed in any of the simulated trajectories but instead, the extracellular section of the selectivity filter deviated from the canonical conductive structure of potassium channels. The degree of widening of the potassium binding sites at the extracellular entrance of the channel was directly related to the degree of inactivation with hERG-F627Y > wild-type hERG > hERG-N629D. These findings support the hypothesis that C-type inactivation in hERG entails a widening of the extracellular entrance of the channel rather than a closure of the selectivity filter.

The TRPV1 channel is responsible for conducting cations through the cell membrane in response to a variety of stimuli, amongst which noxious heat and chemical ligands, such as the vanilloid compounds capsaicin and its ultrapotent agonist resiniferatoxin (RTX). Predominantly expressed in sensory neurons and nociceptive fibers, TRPV1 is involved in important cellular functions, including heat-sensation and pain. One unresolved question about TRPV1 concerns activation by stimuli acting from the extracellular milieu, such as the binding of the double-knot toxin (Dk/Tx) from spider venom. How is this stimulus coupled to the binding of RTX at the vanilloid site? How do Dk/Tx and RTX synergistically act on the gate? Interestingly, both these biomolecular ligands participate in the formation of the vanilloid-lipid-channel-toxin quadripartite complex, as shown by recent cryoEM experiments. We performed molecular dynamics (MD) simulations and confirmed that annular lipids are permanently bound to and stabilize the open state. While lipid head-groups interact with Dk/Tx, the tail-groups are simultaneously in contact with RTX and with a side chain crucial for the movement of the S4-S5 linker. Importantly, mutations at this position were shown to selectively impair capsaicin ability to activate the channel without appreciably affecting ligand binding. In summary, our results shed light on the mechanism by which vanilloid binding is transduced into gating motion and rationalize the experimentally observed allosteric coupling between Dk/Tx and RTX.

No abstract