Modeling Voltage-gated KcsA Ion Channels as Solid-State Nanodevices

Abstract

In this paper, we present a novel method of simulating KcsA ion channels using a TCAD solid-state device simulator [1]. The ion transport in these channels has interesting similarities with the flow of carriers in electronic nanodevices. With this perspective, we have modeled ion channels as solid-state nanodevices and obtained self-consistent solutions of the axial potential and ion fluxes. Models of cylindrical and KcsA channels are built with the TCAD simulation tools and their steady-state characteristics are studied. The simulation results are compared with the reported experimental results in the literature to verify the efficacy of our method. The ability to simulate realistic ion channel models with such computational ease and reasonable accuracy provides a powerful tool for studying the biological functions of these channels with deeper insight. Ion channels are the ultimate in natural nanotubes that regulate the ion flux across a cell membrane, and thereby play an integral role in cellular signaling mechanisms (Figure 1) [2]. Among the approaches for ion channel modeling, molecular dynamics are the most accurate but presently limited by long simulation times [3]. Continuum electrostatics provides an alternative approach, which involves solving the PoissonNernst-Planck (PNP) equations for the charge distribution in the channel. This technique has proven suitable in predicting the behavior of KcsA and Ca 2+ voltage-gated channels [4]. We have simulated a cylindrical channel and a KcsA channel made of two materials: silicon to mimic the conducting water continuum and SiO2 to mimic the nonconducting protein walls. The carrier concentration, degree of ionization, and diffusion coefficients are adjusted in each conducting region to emulate the realistic motion of K + ions in the channel’s vestibule. Two electrodes are placed at either ends of the reservoirs. The simulations are performed by solving the discretized PNP equations with a computation time ~5sec. Figure 2 shows our TCAD cylindrical channel model (35A long and and 6A wide). Figure 3 shows a decreased carrier concentration in the channel’s vestibule, as also predicted by Brownian dynamics simulations [3]. We also considered the effect of protein surface charges in manipulating the energy profile of an ion traversing a channel. These surface charges are known to influence the gating, conductance, and toxinbinding effects of ion channels. As shown in Figure 4, an energy barrier inside the channel is transformed into a potential well by the inclusion of surface charges, thus increasing the chances of ion permeation through the channel [3]. Based on the KcsA ion channel structure revealed by x-ray crystallography [2], we have built a TCAD KcsA channel model as shown in Figure 5. The KcsA channel is 40A long, with a narrow selectivity filter (12A long and 3A wide). Figure 6 shows the axial potential variation under different electrode voltages. Most of the transmembrane potential drop incurs in the selectivity filter, crucial for the selectivity and permeation of various ions. In Figure 7, we compare the current-voltage (IV) results from our simulations with those from experiments [5]. With accuracy within 3% over 200mV, this degree of agreement with the experimental data is better than those reported elsewhere [6].