f the phrase "squid giant axon" releases a flood of memories for you–as it does for me–then this month's hottest new paper is sure to prompt a trip back in time. Rod MacKinnon and his colleagues at the Howard Hughes Medical Institute and Rockefeller University in New York have worked out exactly how the potassium channel operates (paper #5).

   If, by contrast, "squid giant axon" means nothing at all, here's a 30-second introduction. More than 50 years ago, Alan Hodgkin and Andrew Huxley showed that nerve cells transmit electrical signals by selectively allowing some ions to cross the cell membrane. The cell maintains a high concentration of potassium ions inside the membrane, while at the same time excluding sodium ions, which are at a high concentration outside the cell. When a nerve fires, sodium ions flood into the cell, to be pumped out by other systems later.

   Hodgkin and Huxley established how nerve cells work with the conveniently large nerve fibres of the squid, big enough, at 0.7 mm across, to accept one of the relatively crude micro-electrodes of the day. It then turned out that potassium channels are present in almost all living cells. They are, in some sense, fundamental to life, in that they allow the cell to concentrate chemicals inside its membrane while remaining osmotically stable. The question remains: how does the channel select potassium but filter out sodium, which is smaller and ought to get through any pore more easily?

   MacKinnon's group has answered the question completely, with a profoundly satisfying analysis of the detailed physical and chemical structure of the channel. It consists of four identical subunits; each with two long, straight alpha helix chains; MacKinnon had discovered as much almost a decade ago. Central to the subunits is a signature sequence that is characteristic of all the different kinds of potassium channel. "We were able to deduce that the signature sequence made the filter by meeting its partners in the middle," MacKinnon tells Science Watch. "But without seeing it, we could not understand how it worked." Even the latest modelling techniques couldn't get from the protein sequence to a properly folded, fully functional protein. That needed crystals, and X-ray crystallography.

   Crystallization, MacKinnon admits, was "tricky." The team ran through 30 or 40 crystal forms before they got one that would diffract X-rays. "It took a lot of tweaking and fiddling." Some of the tweaks involved modifying the DNA of the channel, cloned from Streptomyces lividans and expressed in Escherischia coli. Others were chemical changes that became part of the purification, for example removing part of the C terminus because that was somewhat disordered. "At that point," he says, "the crystals improved."

   The structure reveals that the four subunits lean together to form what MacKinnon describes as an inverted teepee, its cone towards the inside of the cell and the alpha helix chains the teepee's poles. The selectivity filter is effectively the inside of the teepee, where the signature sequence lines the cavity. The physical meshing of the teepee poles and the presence of oxygen atoms from the signature sequence are crucial to the filter's sensitivity.

   The inside of the teepee is filled with water, and ions in solution. A potassium ion can fit neatly inside the filter, the oxygen atoms of the signature sequence replacing the oxygen of the water that normally surrounds the ion. But the sodium ion is too small. The rigidity of the filter prevents the sodium ion moving into the pore. "It would rather be in the water," MacKinnon explains, "while a potassium ion is just as happy in the filter as in water."

   This is not mere speculation. MacKinnon replaced potassium in the channel with rubidium, which is a little larger than potassium but still accepted by potassium channels, and is dense enough to be easily visualized in an x-ray experiment. And there it was, sitting in the pore. "It's quite an achievement of nature, a beautiful device."

   People have waited a long time to see how this most fundamental component of life works, but that alone is not enough to account for the paper's popularity. Potassium channels are also implicated in medical conditions such as heart arhythmias and some kinds of seizure. "The pharmaceutical companies are very interested," MacKinnon tells Science Watch. His own group, meanwhile, is now looking at how different potassium channels control the pore by responding to different stimuli.