K Channels

Molecular Biochemistry I

K⁺ Channels

Contents of this page:
Introduction to voltage-gated K⁺ channels
Selectivity
Gating & inactivation

Note: Rather than try to cover the many different types of membrane channels, this class will focus on K⁺ channels whose structures have been solved at atomic resolution. Channels discussed elsewhere in this web include porins, gramicidin, and endoplasmic reticulum calcium-release channels.

Voltage-gated K⁺ channels mediate outward K⁺ currents during nerve action potentials. Important advances in understanding have come from:

physiological studies, including the use of patch clamping,
mutational studies of the Drosophila voltage-gated K⁺ channel protein, product of the Shaker gene,
crystallographic analysis of the structure of bacterial K⁺ channels,
molecular dynamics modeling of permeation dynamics.

Four identical copies of the K⁺ channel protein, arranged as a ring, form the walls of a K⁺ channel.

Tetrameric channel

Hydropathy analysis and topology studies predicted the presence of six transmembrane a-helices in the voltage-gated K⁺ channel protein.

The core of the channel consists of helices 5 & 6 & the intervening H5 segment of each of the 4 copies of the protein.
Helices 1-4 function as a voltage-sensing domain, with helix #4 having a special role in voltage sensing (see below). This domain is absent in K⁺ channels that are not voltage-sensitive.

Selectivity:

K⁺ channels are highly selective for K⁺, e.g., relative to Na⁺. The selectivity filter that determines which cation (e.g., Na⁺ or K⁺) can pass through a channel is located at the narrowest part. Mutation studies showed that the H5 segment is essential for K⁺ selectivity. H5 includes a consensus sequence (Thr-Val-Gly-Tyr-Gly) found in all K⁺channels, with only minor changes through evolution.

Voltage-gated K⁺ channel protein

The first K⁺ channel for which an X-ray crystallographic structure was solved is the KcsA K⁺ channel from Streptomyces lividans, depicted in cartoon display at right. In this channel protein there are only two transmembrane a-helices, corresponding to helices # 5 & 6 in the voltage-gated channel. An intervening segment includes the K⁺ channel consensus sequence.

The left diagram is a side view of the KcsA channel (normal to the membrane), while the other view looks down the channel from the extracellular side of the membrane.

The K⁺ channel consensus sequence (shown in black) forms the selectivity filter at the narrowest part of the channel.
K⁺ ions are bound within the selectivity filter (colored pink).
The gate is where the innermost a-helices of the 4 subunits meet at the narrow end of the "teepee" structure. This is the closed channel.

As K⁺ enters the channel, its waters of hydration are replaced by oxygen atoms lining the selectivity filter.

Four binding sites for K⁺ within the selectivity filter are defined by 5 rings of oxygen atoms (see Chime exercise below). Each ring consists of 4 oxygen atoms, one from each subunit. Most are backbone oxygen atoms of the consensus sequence. Each bound K⁺ interacts with eight oxygen atoms, as it sits between two of the rings of oxygen atoms.

At right, a K⁺ ion (colored pink) is seen closely interacting with the outermost ring of oxygen atoms (colored red) of the selectivity filter.

For additional images see a webpage describing work of R. MacKinnon maintained by the Howard Hughes Medical Institute, plus figures p. 736-737 of Biochemistry, 3rd Ed., Voet & Voet.

Solved X-ray structures such as that depicted above, which average over many copies of the channel in the crystal, show K⁺ in all 4 binding sites. Based on predicted electrostatic repulsion and other evidence, K⁺ is assumed to occupy actually every other binding site, with a water molecule in each intervening site. K⁺ ions and water molecules alternately pass single file through the channel.

As a K⁺ ion binds at one end of the selectivity filter, a K⁺ ion exits at the other end. The arrangement of oxygen atoms surrounding each K⁺ ion within the selectivity filter mimics that of a hydrated K⁺ ion. Thus the energy barrier for entry and exit of K⁺ ions is low.

Na⁺, being smaller than K⁺, would be able to interact with oxygen atoms on only one side of the channel. Thus removal of the waters of hydration from Na⁺, for entry into the channel, would be less favored energetically. Furthermore, the presence of K⁺ within the selectivity filter is required to stabilize its ion-conducting conformation. If Na⁺ is substituted for K⁺ in the medium, the selectivity filter undergoes a conformational change to a collapsed state.

Explore at right the structure of the bacterial KcsA channel.

Bacterial K⁺ Channel

Channel gating: The structure of a Ca⁺⁺-dependent K⁺ channel from Methanobacterium thermoautrophicum, designated MthK, was the first to be solved in the open state, providing information about the mechanism of channel gating.

The selectivity filter (K⁺ channel consensus sequence) is colored black in the diagram at right.

Channel opening is associated with a bend at a glycine residue in the innermost helix of each copy of the MthK channel protein. In the mammalian voltage-gated K⁺ channel, a Pro-Val-Pro (PXP) motif in the same helix functions as the gating hinge.

In contrast to the teepee shape of the closed KcsA channel (above), the outward splaying of bent helices provides a large opening at the cytosolic end of the open MthK channel (side view in diagram at near right).

MthK includes a large extracellular protein assembly involved in regulation of gating by Ca⁺⁺. The Ca⁺⁺-binding gating ring does not itself block ion flow. Instead it induces the channel conformational changes responsible for gating.

Explore at right the open conformation of the bacterial MthK channel and its regulatory gating ring.

Ca⁺⁺-Gated K⁺ Channel

Voltage sensing: Mutational analysis first established that positively charged residues in helix #4 are essential for voltage gating. In helix #4, every 3^rd residue is arginine or lysine, while intervening residues are hydrophobic.

Decreased transmembrane potential causes helix #4 to change position, resulting in more of its (+) charges being accessible to the aqueous phase outside the cell. A small "gating current" is measurable, as positive charges effectively move outward.

Crystal structures have been determined for a bacterial voltage-gated K⁺ channel KvAP, and a mammalian equivalent of the Shaker channel designated Kv1.2.

The core of both voltage-gated channels (selectivity filter & two transmembrane a-helices of each of four copies of the protein) is similar to that of other K⁺ channels.

The bacterial KvAP channel, in the open conformation, is depicted at right.

The four copies of the KvAP channel protein are in cartoon display with different colors, but voltage-sensor 'paddle' domains are colored magenta. K⁺ is shown in spacefill display within the selectivity filter, which is colored black. All K⁺ binding sites appear to be occupied because of summing of locations for multiple copies of the channel in the crystal.

Some a-helices of the voltage-sensing domain of the bacterial KvAP channel differ from the expected transmembrane orientation. The positively charged voltage-sensing helix #4, and part of what was earlier designated helix #3, associate in a helix-turn-helix to form a "paddle" shape at the periphery of the channel (colored magenta in the diagram above). For the channel structure depicted above, Fab fragments of monoclonal antibodies raised to the voltage-sensor domain of the channel protein were used to promote crystallization.

The structure of the mammalian equivalent of the Shaker channel has been determined in the absence of FAB fragments but in complex with a regulatory protein. In this channel, a-helices of the voltage-sensing domain are tilted relative to the plane of the membrane but have a more transmembrane orientation.
For diagrams see a website of the Brookhaven National Laboratory, and an article by Long et al. (requiring a subscription to Science).

Some differences between solved KvAP and Kv1.2 Shaker channel structures are postulated to be due to altered conformation of the KvAP channel on being removed from the lipid membrane. A genetically engineered chimaeric channel, in which the voltage sensor paddle of another K⁺ channel is incorporated into the Shaker-type Kv1.2 channel, has been crystallized in the presence of lipids. The structure of this channel in a lipid membrane-like environment is found to be similar to that of the previously solved Kv1.2 structure.

According to current models, a voltage change drives movement of each positively charged voltage sensor paddle complex across the membrane. This exerts tension, via a linker segment, on the end of each inner helix of the channel core to promote bending, and thus channel opening. Recent high-resolution structural studies permit predictions of how acidic residues may stabilize positive charges on the paddle as it moves within the membrane.

See a diagram postulating conformational changes based on the high-resolution crystal structure determined in the presence of lipids (Fig. 6 of Long et al., 2007, requiring a subscription to Nature).
For a simplified diagram see article in a webpage of F. Sigworth.
An animation (movie S1) shows a model of conformational changes during voltage gating, based on the crystal structure of the mammalian Shaker channel, fluorescence assays of conformational changes and molecular dynamics simulation by Pathak et al.

Inactivation: Many channels have multiple open and/or closed states. There may be an inactivated state, as in the hypothetical example at right.

Voltage-gated K⁺ channels undergo transient inactivation after opening. In the inactivated state, the channel cannot open, even if the voltage is favorable. This results in a time delay before the channel can reopen.

The N-terminus of the Shaker channel (or part of a separate subunit in some voltage-activated channels) is essential for inactivation. Mutants that lack this domain do not inactivate. Adding back a peptide equivalent to this domain restores the ability to inactivate.

A "ball & chain" mechanism of inactivation has been postulated, in which the N-terminus of one of the four copies of the channel proteins enters the channel from the cytosolic side of the membrane to inhibit ion flow.
A more detailed diagram is in a review (Fig. 3, article by Kurata & Fedida; requires a subscription to Progr. Biophys. Molec. Biol.).

In some voltage-gated K⁺ channels, entrance of the N-terminus into the channel is followed by a conformational change in the selectivity filter that contributes to the process of inactivation.

Additional material on K⁺ Channels:
Readings, Test Questions & Tutorials

Molecular Biochemistry I

K+ Channels

K⁺ Channels