7a. Gramicidin - The Simplest Channel?

How might an integral membrane protein behave like a channel or a pore, to facilitate the selective diffusion of small ions so well documented in many cells? Certainly, in completely spanning the membrane an IMP such as glycophorin could provide a path for solute transport (and as we shall see in later chapters, IMPs certain can function as solute carriers).  Very rapid ion movement along the outside of the spanning portion of IMP is unlikely, however, due to the hydrophobic outer surface of the helical secondary structure and the closely associated bilayer lipid.  Alternatively, ions might diffuse along the "conduit" formed by the interior atoms of the alpha helix, but its diameter is too small to allow even small ions to pass readily.   Are other secondary, tertiary or possibly quaternary IMP structures involved in channel functions?  To begin addressing this question, let's examine gramicidin A, a simple channel.

Many bacteria secrete defensive molecules called ionophores that cause neighboring cells to become leaky to ions.  The most thoroughly characterized of these is gramicidin A, a linear pentadecameric peptide (15 amino acids long) containing glycine and alternating L- and D-forms of other hydrophobic or apolar amino acids (specifically, D- and L-valine, L-tryptophan, and D-leucine) and blocked (non-polar) amino and carboxy terminal residues.  This peptide readily dissolves in lipid bilayers, forming a helix large enough to allow the rapid, yet selective passage of water and small ions (Hille, 1992).  Note the end-on view of the slowly rotating PDP model of gramicidin on the opposite page.

The currently accepted model for the conducting structure of gramicidin is actually an antiparallel"double helix" consisting of 2 identical, interdigitating peptides oriented in the opposite direction, as illustrated on the right.  The double helix is stabilized by H-bonding<HotLink to H-bond option/Chime> between carbonyl oxygens and amino nitrogens of the adjacent antiparallel strands projecting parallel to their helical axes.  The resulting helix has an inner "channel" with the appropriate dimensions required for both membrane traversal and ion conductance (Burkhart, et al., 1998).  As with other IMP alpha helices, the amino acids forming the outer surface of the gramicidin double helix are hydrophobic, providing appropriate interaction with surrounding bilayer lipid.  The conduit formed along the inner surface of the gramicidin helix, however, is much wider than the one found within an alpha helix, and sufficient amino nitrogens and carbonyl oxygens are available along the inner helix surface to coordinate water molecules or conduct 3 cations in single file along the channel. 

Carefully examine the molecular details of this double helix using the various RasMol commands available by pointing the cursor at the PDB model  on the opposite page and right-clicking the mouse (or clicking and holding the mouse button down if you are using a Macintosh). 

This curious double helical arrangement, with its wide bore and membrane-spaning length, derives in part from the presence of D-amino acids and their alternation with L-forms in the primary structure of each peptide subunit.  Eukaryotic cells don't make or use D-amino acids, however, and as we've seen, the portions of IMP traversing the bilayer are organized into much more constricted alpha-helical secondary structures.  Consequently, while providing very useful information concerning atomic interactions between channel and permeating ions and about the molecular basis of selectivity, gramicidin is not a generally useful model for understanding the structure of other, "native" membrane channels. 

Using what you've learned so far about IMP structure, how might other IMP be organized to explain the rapid diffusion of small ions across eukaryotic membranes?  (Recall that glycophorin apparently forms dimers in erythrocyte plasma membranes.  Could similar associations between other IMP subunits create aqueous channels?)  What exactly are the structures of other membrane channels?