Tuesday, March 06, 2007

Voltage-Gated Ion Channels

Cells have to import and export ions in order to survive. Since ions are charged molecules, they do not pass through the cell membrane. This means they have to be transported by proteins that are embedded in the membrane.

There are many different kinds of transporters. Some of them use up energy to move ions across the membrane by a process called active transport. Some of them simply open a channel and allow ions to pass through. Let's look at some channel proteins.

A typical channel protein consists of several subunits arranged around a central hole that permits passage of ions. In this case we'll be dealing with ion channel proteins that allow movement of Na+ (sodium) or K+ (potassium) ions from the outside of the cell (top) to the cytoplasm (bottom). The part of the protein that spans the membrane is usually an α helical region. In fact, there are usually several α helices in a channel protein.

The flow of ions must be regulated, so these channel proteins must be capable of opening and closing to allow or restrict the flow of Na+ or K+. This is why they're called gated channels. There are several ways of controlling the gate. We are going to be discussing mechanisms that respond to the membrane potential—the difference in charge on either side of the membrane. For example, the wave of membrane depolarization that occurs as a nerve impulse, or action potential, moves along the axon of a nerve cell.

These channel proteins are called voltage-gated ion channels.

The simplest kind of voltage-gated ion channels are the K+ (potassium) channels found in bacteria and in eukaryotes. The role of potassium channels is to allow influx of potassium if the concentration inside the cell drops below some minimal value. Normally the concentration of K+ in the cytoplasm is about 100 times greater than the concentration outside the cell because K+ ions are actively transported into the cell as Na+ ions are pumped out. The concentration of Na+ outside the cell is about 10 times higher than the inside concentration.

From time to time, these concentrations are perturbed (e.g., during the firing of a nerve impulse) and the sodium and potassium channels will open to restore the resting state concentrations. The gates in the ion channels will only open for 1 millisecond but this is enough to allow entry of thousands of ions per channel.

The structures of several bacterial K+ channels are known and recently the structure of the homologous channel from rat was solved. The structures are very similar. Let's look at the rat voltage-gated K+ channel (Longe et al. 2005a; Long et al. 2005b).

The complete channel consists of four identical subunits colored red, green, yellow, and blue in the figure on the left. The crystallized protein has an additional β subunit at the bottom attached to the T1 cytoplasmic domain but I've removed it in order to simplify the description. The β subunit contains an oxidoreductase activity that's coupled to the influx of potassium.

If you look at the structure of the single red subunit you can see that it consists of a bundle of four transmembrane α helices labelled S1-S4 on the left and two others (S5 and S6) on the top right. The S5 and S6 helices are the ones that form the central channel when the four subunits come together. This is most easily seen in the top view at the bottom of the figure. The black dot in the middle represent a K+ ion about to pass through the channel.

The structure shown is the open conformation that allows passage of ions. The closed conformation was modeled from the known structure of the bacterial channels and experimental data that implicated the S4-S5 linker helix in the gating process. You can see this linker helix in the figure above. It joins the S1-S4 bundle to the S5, S6 bundle. It is not one of the transmembrane helices. The results are shown below.

In the closed conformation, the S4-S5 helix responds to changes in the membrane potential by twisting from a horizontal position to a slightly more vertical position. This causes the S5 and S6 transmembrane helices to also adopt a more vertical orientation. Since the same conformational change occurs in all four subunits, the net effect is to make the entire protein more narrow and this closes the channel in the middle of the protein. The switch between the open and closed conformations is extremely rapid and it is very sensitive to changes in the membrane potential.

Long, Stephen B., Campbell, Ernest B., MacKinnon, Roderick (2005a) Crystal Structure of a Mammalian Voltage-Dependent Shaker Family K+ Channel. Science 309:897-903.

Long, Stephen B., Campbell, Ernest B., MacKinnon, Roderick (2005b) Voltage Sensor of Kv1.2: Structural Basis of Electromechanical Coupling. Science 309:903-908.


  1. Ion channels are truly impressive things. For instance in the sodium channel sodium ions pass through the channel at nearly the diffusion rate in water, yet are 10,000 times more permeable to sodium than potassium. Other channels are the same, able to pass their own ions with extreme efficiency but are nearly impermeable to most other ions, even extremely similar ones (sodium and potassium have nearly the same charge and similar size). An interesting exception is barium, which passes through calcium channels better than calcium itself.

    The other interesting thing is the structure. Sodium, potassium, calcium, and chloride channels are nearly identical, based on 4 subunits each with 6 transmembrane alpha helices laid out in roughly the same manner in all organisms. In the case of potassium these are 4 different proteins (one for each subunit), while in sodium it is one protein with 4 subunits, but the basic structure is the same and they are likely descended from a single ancestral channel.

  2. "From time to time, these concentrations are perturbed (e.g., during the firing of a nerve impulse) and the sodium and potassium channels will open to restore the resting state concentrations."

    I haven't gone to the references, but IIRC the Wikipedia piece claims that the mechanism is fairly statistical, as other biochemical machinery. I.e. they can spontaneously open or close in rest and during a pulse. (But at different rates, of course.)

    Seems the whole mechanism is rather robust if it can allow such mishappens, and a fine testament to evolution as the ultimate doodler.

  3. Please, correct the second picture. You've introduced an error. It is K+ outflow in voltage-gated potassium channels, not the inflow, as you have shown with the arrows.

    S5-S6 gate is facing the cytosol side of the membrane. Closed/Open state of the channel is modulated by the hinge in gating S6 helices.

  4. Thank-you very much. I spent a long time trying to figure out which direction potassium ions were flowing. Obviously, I should have spent even more time!

    You've no idea how difficult it is to figure out what's going on. None of the biochemsitry textbooks are very good. Part of this blogging exercise is to prepare for introducing this topic into my own book and I really appreciate the feedback.

    Email me if you're willing to help by reviewing the book.

  5. No mention of
    voltage-gated proton channels?
    The latest addition to the group.
    I think I was in the acknowledgments for
    that Science paper, working with the Japanese group.
    A new twist on VGIC's.

    Tom McCormack