218 GROUP I AND II METALS IN BIOLOGICAL SYSTEMS
conserved in many K + channels as is shown in reference 18 for 15 different
channels. The conclusion reached here is that the voltage - sensor paddle is a
conserved unit in voltage - dependent K + channels and that the S4 – S5 region
can bend easily depending on the relative gating state of the channel. The
S4 – S5 region contains many hydrophilic residues and is comfortable at the
interface with aqueous solution on the intracellular side of the membrane.
Additionally, voltage - sensor paddles appear to be fl exibly attached to S2 and
S5, and S1 and S2 are loosely packed against the channel pore. Indeed the
voltage sensor seems to fl oat within the membrane, leading the authors to
propose a new model for gating charge movements in which the arginine
gating charge carriers can move the voltage - sensor paddle through the mem-
brane from the intracellular side to the extracellular side. (See Figure 8 of
reference 18, a news focus article in Science a month after reference 18 was
published,^24 and Figure 5.8 for another view of the newly proposed movements
of the voltage sensor paddle.)
Conclusions reached in reference 18 include: (1) Voltage - sensor paddles are
helix – turn – helix structures made from helices S3b and the N - terminal half of
S4; (2) voltage - sensor paddles ’ amino acid composition is mainly hydrophobic
with the important exceptions of S4 arginine residues; (3) paddles are movable
with respect to the pore because of their fl exible attachments to the rest of
the voltage sensor; (4) movements of the paddle in response to changes in
membrane voltage could open the pore; and (5) observations and experimen-
tation led to a proposal that the voltage - sensor paddles move across the
membrane carrying their gating charges (arginine residues) through the elec-
tric fi eld. This last proposal led to experiments described in the group ’ s accom-
panyingNature publication.^25
The MacKinnon group next exploited the Fab attachments to the full - length
KvAP channel (Fab6E1) and to the isolated voltage sensor (Fab33H1) to
examine the position of voltage - sensor paddles with the KvAP channel func-
tioning in a lipid membrane. First, it should be said that the Fabs attach them-
selves to the same sector of the voltage - sensor paddle (the same epitope)
between S3b and S4 in both structures. The experiments sought to assess
whether the voltage - sensor paddles change their position when the channel
gates open under conditions of depolarization (change from a holding voltage
of− 100 mV to + 100 mV). The researchers could show that both Fabs inhibited
channel function when applied to the solution external to the lipid membrane,
but not when the Fabs were applied internal to the membrane. It was also
shown that inhibition by the externally applied Fabs fi rst required membrane
depolarization. Experimentally the channel was held closed at − 100 mV for 10
minutes, during which time little or no inhibition was observed. Additionally,
no inhibition was observed in the absence of a Fab when the membrane was
depolarized from− 100 to 100 mV (typically for 200 ms). Then a Fab was added
and depolarizations from− 100 to 100 mV were continued. Inhibition began
after some time lag — perhaps because the channel had to open before Fabs
could bind and begin to cause inhibition. The researchers concluded that the