BASIC MECHANISMS OF VOLTAGE-DEPENDENT ION CHANNELS 
Christopher Miller, Ph.D., Investigator 
Ion channels are the most basic elements of mo- 
lecular hardware in the nervous system. They are the 
membrane-spanning proteins that directly mediate 
the transmembrane ionic fluxes giving rise to elec- 
trical signals in neurons and other electrically active 
cells. All proteins of this type have the same struc- 
tural plan: that of a water-filled pore spanning the 
cell membrane. Thus channels act as leakage path- 
ways for ions down their thermodynamic transmem- 
brane gradients. The ion channels involved in neuro- 
nal function select strongly among the diff"erent 
species of inorganic ions in the aqueous solutions 
bathing the cell membrane. In addition, channels 
must have the ability to open and close their con- 
duction pores in response to external signals, such 
as binding of neurotransmitters (ligand-gated chan- 
nels) or changes in transmembrane electric field 
(voltage-dependent channels) . 
Dr. Miller's research is aimed at questions of fun- 
damental molecular mechanisms of ion channel 
operation and of the underlying protein structures. 
The absence of direct structural information makes 
it necessary to draw inferences from close examina- 
tion of function. This can be done because ion chan- 
nels, unique among proteins, can be studied at the 
single-molecule level, both in the cellular environ- 
ment with patch-recording techniques and after re- 
constitution into biochemically defined "artificial" 
membranes. Dr. Miller's laboratory is currently fo- 
cusing on several voltage-dependent ion channels 
that provide opportunities to address mechanisti- 
cally important questions about channel structure 
and function. 
Peptide Neurotoxins as Probes 
of Channel Structure 
Charybdotoxin (CTX), a peptide derived from 
scorpion venom, blocks a small family of K^-spe- 
cific channels. Having recently shown that CTX acts 
as a physical plug in the channel's outer "mouth," 
Dr. Miller is currently using the peptide as a probe 
of this important region of channels. Work on the 
Drosophila Shaker channel, using site-directed 
mutagenesis, has identified residues that specifi- 
cally and locally alter the binding of CTX. These 
residues are therefore located near the ion entry- 
way, a region of the protein contributing to the 
transmembrane pore. 
The toxin binds to the high-conductance Ca^"^- 
activated channel with high affinity, and the in- 
teraction of CTX with this channel may be studied at 
high resolution by reconstituting single channels 
into planar bilayer membranes. Recently, in a re- 
search project supported by the National Institutes 
of Health, Dr. Miller and Dr. Per Stampe have ex- 
pressed a gene coding for CTX at high levels in Esch- 
erichia coli. The recombinant peptide is rendered 
fully active by several in vitro post-translational 
modifications. Specific residues required for chan- 
nel blocking have been identified by mutagenesis 
of the synthetic gene followed by single-channel 
analysis. 
These studies are enhanced by knowledge of the 
solution structure of CTX, recently determined by 
two-dimensional nuclear magnetic resonance 
(NMR) . With this approach, the entire surface of the 
toxin has been functionally mapped. The function- 
ally important residues all lie on the same side of the 
molecule and thus define an interaction surface that 
is recognized by the receptor in the channel mouth. 
Dr. Miller and Dr. Steven Goldstein are currently 
seeking to use CTX as a molecular caliper to deter- 
mine physical distances between residues in the 
mouth of a voltage-gated K"^ channel from Drosoph- 
ila, the genetically manipulable Shaker. While the 
naturally occurring Shaker channel is only weakly 
sensitive to CTX, a mutated channel has been engi- 
neered to build a highly sensitive CTX receptor into 
the channel's outer "vestibule." Now, with the 
functionally critical residues on both toxin and 
channel known, the aim is to pair these up, to iden- 
tify specific residues forming interaction partners in 
the toxin-channel complex. This is now being ap- 
proached by complementary mutagenesis of both 
channel and toxin. With the toxin structure known, 
identification of only a few such residues will place 
strong constraints on the channel's structure. 
An additional use of CTX as a structural probe of a 
channel is being carried out by Dr. Miller and 
Amir Naini, a graduate student. Previous results 
identify Lys27 as a mechanistically rich residue, in 
that the e-amino group protrudes slightly, from the 
channel-bound toxin into the narrower conduc- 
tion pathway. This €-amino group can sense the pres- 
ence of ions within the selectivity regions of the 
pore. Naini has now shown that Lys27 may be re- 
placed by a chemically reactive Cys, and that unnatu- 
ral lysine analogues may be replaced at this position 
while maintaining toxin activity. Thus "tethered" 
with diff^erent chain lengths, amino groups on toxin 
NEUROSCIENCE 419 
