ION CHANNELS MEDIATING ELECTROCHEMICAL SIGNALING IN VERTEBRATE NEURONS 
Paul R. Adams, Ph.D., Investigator 
Nerve cells express a panoply of plasma mem- 
brane ion channels that mediate electrical and chem- 
ical signaling on a variety of time scales. These chan- 
nels open and close in response to changes in local 
membrane voltage or chemical environment. Ion 
flow through the open channels then causes further 
changes in local voltage or chemistry. These 
changes can spread to adjacent cell regions, eventu- 
ally leading to coordinated electrical responses, 
such as propagating action potentials or synaptic 
transmission. Dr. Adams's laboratory investigates 
how various types of ion channel contribute to these 
electrical responses and how cell geometry influ- 
ences signal spread . This is achieved with a combina- 
tion of electrophysiological techniques (such as 
patch-clamp recording) and specialized light mi- 
croscopy (such as calcium imaging and three- 
dimensional reconstruction) . 
Nerve Cells in Three Dimensions 
The electrical activity of neurons is played against 
a framework of passive membrane capacity distrib- 
uted in an intricate three-dimensional array. Embed- 
ded within this array is a network of ion channels 
fronting complex cytoplasmic diffusion pathways. 
Dr. Adams's laboratory has combined confocal mi- 
croscopy of sympathetic ganglion cells, hippocam- 
pal neurons, and lateral geniculate relay cells with 
custom voxel-based ray-tracing software ("Volvis," 
developed in collaboration with Dr. Ari Kaufman's 
laboratory in Stony Brook) . 
Use of various fluorescent probes allows diff'erent 
aspects of the cell to be visualized. Negative staining 
with fluorescein dextrans permits the outer cell sur- 
face to be seen, as the cell makes a hole in the fluo- 
rescent medium, which can then be reversed on a 
graphics workstation. Conversely, intracellular 
staining with lucifer yellow reveals the inner cell 
surface and details of intracellular structures. Cell 
membranes are visualized with nile red, synapses 
with mitochondrial stains, and intracellular calcium 
with PLUG 3 . Rendering algorithms are then used to 
provide a vivid three-dimensional view from any de- 
sired aspect. 
Any relevant geometrical parameter is easily ex- 
tracted from these large three-dimensional data sets. 
For example, cell surface area measurement is com- 
pared with electrical capacity measurements to ex- 
plain the enigma that the specific capacitance of the 
neuronal membranes is often considerably greater 
than that allowed by its lipid bilayer structure. 
Because living cells move, three-dimensional visu- 
alization is also necessary to explore quantitatively 
optical signals from small structures like spines 
that are close to the resolution limits of light 
microscopy. 
Calcium Imaging 
In previous work the laboratory characterized 
rapid cytoplasmic calcium redistribution in gan- 
glion cells following brief voltage-clamp depolariza- 
tions by using a line-scanning technique. This 
method is rather insensitive to spatial inhomogenei- 
ties in calcium dynamics. Two-dimensional confo- 
cal imaging of calcium signals across the entire cell 
has now been achieved at five frames per second, 
with improved-resolution optical sectioning and cal- 
cium sensitivity. Work in the laboratory had previ- 
ously shown that a major problem with rapid cal- 
cium imaging is dye redistribution, which tends to 
collapse calcium gradients. This is being solved by 
using lower dye concentrations and slowly diff'using 
dyes. Calcium entry seems to be uniform along the 
cell membrane (after making allowance for surface- 
to-volume ratio effects) . Detection of true calcium 
"hotspots" or "domains," if they exist, will require 
further improvement in imaging speed. 
M-Current Regulation 
Many neurons express M channels, which are non- 
inactivating voltage-dependent potassium channels 
controlled by various neurotransmitters acting 
through an unknown G protein. These channels 
limit action-potential firing frequency unless turned 
off by impinging synaptic activity. The laboratory 
has continued to examine mechanisms underlying 
this synaptic regulation of M current. Fast-flow tech- 
niques were used to establish the time course of 
transmitter action. After a latency of 200 millisec- 
onds, M-channel closure peaks ~2 seconds after 
very brief transmitter applications, and then chan- 
nels reopen over the next minute. At the termination 
of longer pulses, supranormal channel reopening is 
seen, leading to overrecovery. These observations 
suggest that the channels can operate in three dis- 
tinct gating modes. These can be directly observed 
by using on-cell single-channel recording. 
The roles of calcium and arachidonic acid metabo- 
lism in M-channel inhibition and overrecovery have 
been further investigated. It had been shown that 
M-channel activity is optimal at physiological levels 
(80 nM), declining at higher or lower levels. It has 
NEUROSCIENCE 385 
