chemical information (e.g., phosphorylation sites) 
allows the construction of such models, but com- 
plex phenomena such as ligand-dependent channel 
gating are probably more delocalLzed within the 
protein and are harder to guess at. Ideally, one 
could randomly mutagenize a large protein exten- 
sively, select mutants that exhibit a particular func- 
' tional difference, and then determine the location 
and nature of the mutants. This procedure would 
give relatively unbiased and model-independent in- 
formation about the parts of the protein important 
for this function. 
To this end. Dr. Yellen's laboratory has been 
working on expression systems for ion channels 
that could be used for such a purpose. Previously, 
the yeast Saccharomyces cerevisiae was tried as a 
selectable expression system. It was possible to 
produce all four subunit proteins of the Torpedo 
nicotinic AChR in yeast cells, but no ion channel 
function was detected. For that reason, Dr. Yellen 
and Dr. Mark West have turned to mammalian cell 
expression systems, which have successfully ex- 
pressed other similar ion channel proteins. 
IV Biophysical Studies on Reconstituted Channels. 
Functional studies on single ion channels still 
PUBLICATIONS 
provide the most detailed clues into the operation 
of channel proteins. Dr. Yellen and Susan Demo 
have been examining the interaction of permeant 
ions with the gating mechanism of the calcium-acti- 
vated potassium channel from rat muscle, reconsti- 
tuted in planar lipid bilayers. When the occupancy 
of the pore is increased by raising the concentra- 
tion of potassium or rubidium, the channels tend 
to stay open more of the time. This observation is 
consistent with a "foot in the door" model that sup- 
poses that a channel cannot close when occupied 
by an ion. Other results with ammonium ions, how- 
ever, suggest that occupancy of the channel cannot 
by itself explain the effect on gating and that the 
identity of the occupying ion and the particular lo- 
cation occupied within the channel are also impor- 
tant. Cesium ions, which block this channel, pro- 
duce the same effect in a highly voltage-dependent 
manner. By correlating the voltage dependence of 
the gating effect with the voltage dependence of 
the block, Demo has shown that the gating pertur- 
bation is not a general allosteric effect but a specific 
effect of ions in the permeation pathway. 
Dr. Yellen is also Assistant Professor of Neurosci- 
ence and of Biophysics at The Johns Hopkins Uni- 
versity School of Medicine. 
Article 
Tomaselli, G.F., Marban, E., and Yellen, G. 1989. Sodium channels from human brain RNA expressed in 
Xenopus oocytes: basic electrophysiologic characteristics and their modification by diphenylhydantoin. 
J Clin Invest 83:1724-1732. 
558 
