| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biophys J, July 2002, p. 2-4, Vol. 83, No. 1
Department of Chemistry, Brandeis University, Waltham, Massachusetts 02454-9110 USA
Ion channels are characterized by three main
properties: their gating characteristics, their selectivities, and
their conductivity patterns. The crystal structure of the KcsA
potassium channel pore (Doyle et al., 1998 The crystallographic pore structure represents the channel in a closed
state, the constriction on the intracellular side (the inner pore)
being too narrow to permit ion (or water) passage. As has been pointed
out repeatedly, structural modification of this inner pore is required
for current to flow. In this issue, Chung et al. (2002) In a series of papers Chung and his coworkers have exploited the
advantages of Brownian dynamics (BD) to monitor ionic movement through
channels and mimic experimentally observed
current-voltage-concentration (I-V-c) profiles. The great virtue of BD
is that, unlike molecular dynamics, large (picosecond) time steps can
be used so that, with ultra-high-speed computers, multiple simulations
can be run for hundreds of nanoseconds, long enough to ensure
statistical reliability and observe repeated ion passage through the
pore, thus generating sets of concentration dependent I-V curves. Of
course, much molecular detail is suppressed with simplification. In
applications to the potassium channel, the protein is replaced by a few
selected electrical features probably common to all such channels (the
oxygen containing dipolar moieties forming the selectivity filter, the
macrodipoles oriented toward the aqueous cavity, and acidic guard
groups at each mouth) embedded in a low dielectric milieu. The aqueous
pore becomes a high dielectric viscous continuum, in which the filter radius is fixed, and the radius of the inner pore and both pore mouths
are adjustable. In contrast to continuum treatments like Poisson-Nernst-Planck modeling (Eisenberg, 1999 Previous work with this model showed that BD provides a
semiquantitative description of potassium conductance. The mechanism for controlling ionic currents turns out to be marvelously simple. It
hinges first and foremost on the filter and mid-channel aqueous cavity
being a region of high negative charge density, always multiply cation
occupied. In effect the channel is blocked by its own permeant ions.
Conduction entails relief of this block. Outward conductance requires
an ion to enter the inner pore, penetrate the filter and drive out one
of the resident ions. For inward conductance an ion must exit the
filter region and traverse the inner pore. In all cases, glutamates at
the inner pore entrance form an ion-binding site. The inner pore is
ionophobic. As an ion moves outward from the glutamates it must
surmount a substantial energy barrier until it is attracted by the
field of the macrodipoles surrounding the mid-channel cavity; it
then accelerates rapidly, enters the filter and effects conduction by
knock-off. Inward movement begins by overpopulating the filter; the
extra ion then surmounts the pore's internal barrier, resides
near the inner mouth glutamates, and ultimately escapes. Increasing the
inner pore radius reduces its ionophobic barrier height, thus
increasing current flow in either direction. A change from 2 to 2.5 Å would drop the barrier by almost 4 kT, leading to a nearly 25-fold
increase in current. Small alterations in inner pore radius can account for the great diversity in potassium channel conductances.
The paper makes predictions about rectification (unidirectional ion
passage). According to Chung et al. (2002) However, all is not sweetness and light. The model predicts superlinear
I-V relationships, which would seem contradicted by most experiments,
although not all. But this may reflect the presence of exogenous
blockers rather than the electrodiffusive properties of the channel
itself. Inward rectification involves interaction of
Mg2+ or polyvalent amines with acidic sites (Lu
and MacKinnon, 1994 Where do we stand now? The results presented here provide a further
powerful incentive for intensive experimental study of potassium
channels (as if one were needed). The basic observation, that inner
pore size determines ionic current, should encourage development of
improved molecular calipers for sizing inner pores of potassium
channels, similar to work already under way (Guo and Lu, 2001 This mechanism may not be limited to potassium channels. The sequence
homologies among potassium, calcium, and sodium channels suggest they
have similar architectures. Calcium channels exhibit rectification and
require multiple ion occupancy for calcium flow. Calcium channel
conductances span a 10-fold range and sodium channels a 5-fold one.
Does access to the selectivity region in these two families also
involve ionic diffusion through a long ionophobic pore? BD has already
been used to describe coupled Na and Ca movement in model pores. Might
changes in the diameter of the (putative) long inner pore influence
channel resistance much as it seems to in potassium channels? Could
altering the charge of guard groups also influence rectification? Of
course modeling calcium and sodium channels is much more speculative.
At least we have a potassium channel pore structure. Still, there are
homology models based on analogies with known KcsA features. While
these do not agree on the details of the selectivity filter, the BD
treatment of potassium channels indicates that understanding
resistivity hinges on getting the general picture right. Conductance,
unlike selection, is not sensitive to slight structural change in the
filter. Even if a picture suggested by theory were not precisely
correct, intelligent speculation is a goad for more intense
experimental scrutiny. And sometimes theory gets it right.
This work was supported by National Institutes of Health grant
GM-28643.
ARTICLE
) provided immediate
qualitative answers to most questions about potassium channel
selectivity. It has led to an explosion of computational papers that
illustrate in great detail how the various structural features function
to exclude anions, stabilize cations, select for potassium over sodium,
and promote divalent block. What remains wide open is understanding their enormous diversity: they are gated in many ways and differ greatly in their conductance behavior. Selective potassium channels have five strictly conserved residues (the signature sequence) and
similar inner helix sequences, motifs that form the filter and the
mid-channel aqueous cavity. These features promote multi-ion stabilization, a property initially deduced by Hodgkin and Keynes (1955). Regardless of their gating mechanisms, they exhibit
wide-ranging electrical properties. Maximal conductivities span a
nearly 100-fold range.
provide a
reasonable and intuitive, albeit speculative, proposal for specific
changes that could account for the broad spread of limiting
conductances. The finding is striking: a small adjustment of the inner
pore radius drastically alters channel resistivity.
), ions are charged spheres of finite size, assigned their crystal radii. This aspect of
the model permits quite rigorous treatment of two crucial features: ionic repulsion and dielectric variability. From the Brownian perspective ion movement is basically electrodiffusive, driven by the
potential due to the fixed model charges and the applied voltage,
subject to a diffusive viscous drag, to random forces mimicking thermal
coupling to the surroundings, to inter-ionic repulsion and to reaction
fields induced by dielectric variation.
, were the inner mouth
glutamates fully protonated there would be essentially no outward
current; presumably without charged sites to attract ions they
wouldn't enter the pore from the intracellular side. Inward current
would still flow because escape from the central cavity and overcoming
the pore's internal barrier is rate-limiting; the charge state of
inner mouth guard groups matters little here. What about modifying the
aspartates at the outer mouth? Complete protonation would make the
channel slightly outwardly rectifying, but the effect would be much
less dramatic. Unlike at the inner mouth, where guard groups are the
sole force promoting ion entry, entrance at the outer mouth also
reflects the influence of the filter field.
; Lopatin et al., 1994). Could discharging guard
groups contribute as well?
). What
about rectification? The polyamine data suggest that the picture
is complex and that there are other acidic sites deeper within the
pore. Possibly these affect both ion entry into and the ionic energy
profile within the inner pore. Could the prediction of outward
rectification by discharging the outer mouth aspartates be observed? If
so, it would provide powerful confirmation for the basic model. A more
detailed study of high field currents is surely worthwhile. At the very
least it would critically test the model presented here. It might
provide direct evidence for high field block.
| |
FOOTNOTES |
|---|
Address reprint requests to Peter C. Jordan, Department of Chemistry, MS-015, Brandeis University, P.O. Box 549110, Waltham, MA 02454-9110. Tel.: 781-736-2540; Fax: 781-736-2516; E-mail: jordan{at}brandeis.edu.
Submitted March 12, 2002, and accepted for publication March 14, 2002.
| |
REFERENCES |
|---|
Biophys J, July 2002, p. 2-4, Vol. 83, No. 1
© 2002 by the Biophysical Society 0006-3495/02/07/02/03 $2.00
This article has been cited by other articles:
 |
|
 |
|
  G. V. Miloshevsky and P. C. Jordan Anion Pathway and Potential Energy Profiles along Curvilinear Bacterial ClC Cl- Pores: Electrostatic Effects of Charged Residues Biophys. J., February 1, 2004; 86(2): 825 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Garofoli and P. C. Jordan Modeling Permeation Energetics in the KcsA Potassium Channel Biophys. J., May 1, 2003; 84(5): 2814 - 2830. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
