Ion channel “gating” refers to the conformational changes involved in the opening and closing of the ion conducting pathway. In KcsA potassium channel, gating involves a conformational change between the closed and open conformation in response to intracellular pH. Recently, it was found that channel opening is followed by an inactivation process which resembles the C-type inactivation observed in voltage-gated eukaryotic channels (Gao et al 2005, Cordero-Morales et al 2006). This inactivation is suppressed by mutation E71A. The crystal structure of E71A at 2.5 Å suggested that interactions between residues E71, D80, W67 and R64 might be involved in the distortion of the selectivity filter responsible for the inactivation process. Using patch clamp, EPR spectroscopy, X-ray crystallography and MD simulations, we have characterized the effect of a wide range of side chain substitutions at position 71, in an attempt to establish the role of the selectivity filter in steady-state gating. Single channel analysis demonstrated that at position 71, side-chain polarity and charge affect activation and inactivation gating in KcsA. We classified these mutants in three kinetic groups; the first with Po ≈ 0.9 and mean open times > 70 ms, the second group, represented by the WT KcsA with an apparent Po ≈ 0.1 and mean open time < 30 ms and the last group with Po ≈ 0.5 and a mean open times of less than 6 ms. Interestingly, all the substitutions had a mean closed times of 500 ms). Moreover, macroscopic currents from these mutants are similar to the non-inactivating E71A mutant than WT KcsA. We solved the crystal structure of mutants in each representative group and no significant difference was observed compared to the WT structure, suggesting that the conformational changes at the selectivity filter might occur only when the inner helix bundle is in the open conformation. We have carried out molecular dynamics simulations of E71A and wild-type KcsA channel, to address the basic principles underlying the changes that take place near the selectivity filter and their relationship with the inactivation mechanism. In the wild-type channel, the presence of hydrogen bond network behind the selectivity filter formed by E71, D80 and W67 residues acts like a molecular “spring” that modulates the conformation of the filter, leading to inactivation. In E71A mutant, the absence of this molecular spring due to E71 mutation, the filter is unchanged and hence the channel is constitutively active. Even though the analysis is carried out on the closed state of the channel (in regards to the conformation of the inner helix bundle), we believe that interactions behind the filter also play a crucial role in the open state.