Voltage-gated potassium (Kv) channels are tetramers of 6-transmembrane domain (S1-S6) α-subunits. The activation gate that seals off the ion conducting pore is under control of the voltage-sensing domain and locates at the bundle crossing of the S6 segments. After channel opening most Kv channels display slow inactivation, a process that involves rearrangements of the selectivity filter (SF) resulting in a non-conducting channel although the S6-gate is open. Recent evidence argued for a strong coupling between inactivation and activation: after gate opening the S6 segment undergoes structural rearrangements that would be transmitted up to the level of the SF destabilizing its conformation. Substituting in Kv1.5 residue T480 that locates at the bottom of the SF by an alanine generated mutant channels which instead of inactivating displayed a second open state, characterised by slowly increasing current. Several molecular dynamics (MD) simulations showed a MD trajectory that displays a hydrophobic collapse of the central cavity with S6 dynamics decreasing the pore radius of the S6-gate. The simulations further showed that the SF did constrict in WT channels but not in the mutant. Ionic current measurements of the mutant T480A channels showed that after prolonged depolarizations - pushing the channels into the second conducting state - the activation gate could close and reopen with the channel remaining in the second conducting state. To convert the channels back to their original conducting state longer repolarizing times were needed. Thus, using the T480A pore mutant we could directly determine from ionic currents that gate closure in Kv1.5 channels does not depend on the status of the SF which, as suggested by Deutsch et al. implies the existence of a channel state with both a closed gate and inactivated selectivity filter. (Support: FWO-G025608)