Structure and Function of the KcsA Potassium Channel from

Download Report

Transcript Structure and Function of the KcsA Potassium Channel from

Structure and Function of the KcsA
Potassium Channel
from Streptomyces lividans
Typical voltage-gated potassium channels contain four alpha-subunits in a tetrameric
assembly. Each subunit is composed of six transmembrane segments, labeled S1-S6.
Both the N and C-termini are located on the cytoplasmic side of the membrane. The
selectivity filter on each subunit is in the membrane-embedded loop between S5 and S6.
The S4 region is probably the voltage sensor. Potassium channels that activate and later
inactivate in response to membrane depolarization (like the classic sodium channel)
often have a "ball-and-chain„ segment at the N-terminus which occludes the inner pore
during inactivation.
Until 1998, models of ion channel pores were based on hypotheses which came from
biophysical and biochemical analyses. In 1998, Rod MacKinnon's group reported the xray crystal structure of a prokaryotic K+ channel from Streptomyces lividans (Doyle et
al. (1998) Science 285, 69-77). The homologies between the KcsA protein and the
channels of eukaryotic cells, especially excitable cells, has allowed confident use of this
structure to elaborate many important aspects of channel activity at the molecular level.
The KcsA protein of Streptomyces lividans is a bacterial potassium channel protein that
is gated by pH instead of voltage. Neverthless, the KcsA protein sequence is similar to
many other types of potassium channels. KcsA channels are a passive transport
pathway through which potassium leaves or enters the cell, down its electrochemical
gradient. They exhibit the two characteristic features of an ion channel:
1) high selectivity and
2) high throughput.
These properties can be understood in terms of certain parts of the protein structure.
Sequence homologies and functional characterizations show that the KcsA channel is
very similar to many voltage-gated potassium channels, such as the Shaker family. In
particular, the pore loop, which exists within the membrane and is responsible for the
high flux rate of K+ ions despite strong selectivity for K+ over other monovalent cations,
is strongly conserved among other K+ selective ion channels. The sequence domain of
the selectivity filter is shown in red (below). The blue residues line the pore cavity and
the inner pore, and play a strong role in ionic permeation as well as activation gating.
The functional channel is a homotetramer like most other potassium channels. Unlike
voltage-gated potassium channels, this channel lacks the first 4 membrane-spanning
segments termed S1-S4. However, the KcsA sequence shares very strong homology
with voltage-gated ion channels in the S5, S6 helices - the outer and inner
(transmembrane, TM1 and TM2) helices, respectively, in KcsA.
Overall Architecture
Structurally, the KcsA channel is a homotetramer composed of four αsubunits arranged with four-fold symmetry around the pore. The channel
complex resembles a tea cup, with the mouth facing the extracellular
environment and the bottom facing the cytoplasm. Each of the four
subunits consists of two transmembrane α-helices (TM1 and TM2,
residues 1-52 and 85-119, respectively). The two transmembrane αhelices are connected by a segment of about 30 amino acids comprising
the pore region (P), which includes three main structural domains: the
'turret', the pore helix and the selectivity filter.
The inner transmembrane helix of each subunit faces the central pore
while the outer helix faces the phospholipid membrane. The inner helices
are tilted with respect to the membrane such that there is a relatively
wide outer vestibule. The inner helices also mediate the subunit
interactions which maintain the critical homotetramer structure of KcsA.
Ribbon representation
of two subunits of
the KcsA tetramer.
The two transmembrane helices,
TM1 and TM2, and the
pore loop, P, are
shown. The two loops
in the upper corners
represent the ‘turret’,
associated with the
binding of many
channel blockers. The
spheres show positions
seen for Rb+ and Cs+
ions in the X-ray
structure. The lower
two spheres represents
two overlapping
positions. A third, nonoverlapping site,
observed lower down
in the central cavity, is
not shown.
Top and side views of the KcsA tetramer in ribbon representation.
Each subunit is in a different color.
Between the outer helix and the pore helix is the turret region (residues
53-60), forming the outer vestibule, which interacts with most channel
blockers. Both the extracellular and intracellular openings (outer and
inner vestibules) are lined with acidic amino acids. Their negative charge
at physiological pH attracts the potassium ion to the pore opening and
facilitates ion conduction through the channel.
The selectivity filter (at the extracellular side of the channel) is in a short
extended chain region between the pore helix and the inner trans-membrane helix. At this point the channel is so narrow that a K+ ion cannot
enter without being fully dehydrated. The filter is comprised of the backbone carbonyls from residues: threonine (T75), valine (V76), glycine (G77) and
tyrosine (Y78), and glycine (G79). The signature motif GYG is completely
conserved in all K+ selective channels. Two of the residues of this short
extended chain are characterized by unusual dihedral angles corresponding to the left handed α (helix) conformation of the Ramachandran
diagram. These alternate with residues in the more usual right handed a
conformation, and the effect is to direct the carbonyl oxygens of 3 consecutive peptide bonds into the channel, while the side chains point away.
Mutagenesis studies on the
Shaker potassium channel
mapped onto the KcsA
structure (mutation sites
labeled only on one of the two
subunits shown). White residues
represent the binding site for
charybdotoxin; the yellow
residue (Y82) is the external
TEA binding site; the
mustard/orange residue at the
base of the selectivity filter
(T74) is the internal TEA
binding site. The green residues
are accessible in both open and
closed states; the purple
residues are accessible only
when the channel is open. Red
residues are absolutely required
for K+ selectivity.
Stick model of KcsA.
The internal channel
surface (minimal
radius) is red.
The actual channel, which is formed by the inner helices, is lined with
hydrophobic amino acids, creating a situation that does not seem to favor
ion conduction. However, the inner chamber (cavity) is certainly filled
with water - it is large enough to accommodate 50-60 water molecules,
and this provides a polar and polarizable environment that stabilizes a
cation here.
The Pore Structure
The overall length of the pore is 45 Å and the diameter varies greatly
along the length. The inner helices form a narrow tunnel at the intracellular side of the membrane, with a length of approximately 18 Å. This
tunnel leads into a wide cavity (about 10 Å wide) near the middle of the
membrane. The tunnel and the cavity are wide enough that a K+ ion can
move throughout this region while remaining hydrated. However, at the
extracellular end, the cavity narrows into the selectivity filter which is so
narrow that for a K+ ion to move through the filter, it must shed its
hydration shell. The cavity is lined primarily with hydrophobic residues
while the selectivity filter seems to be primarily polar due to the main
chain peptide carbonyl groups of the GYG signature sequence which line
this narrowest part of the pore.
Ionic Binding Within the Pore
Selectivity for the potassium ion (Pauling radius 1.33 Å) over a smaller ion such as
sodium (radius, 0.95 Å) or lithium (radius, 0.60 Å), arises from the fact that the filter
requires a large sphere for interaction. The energy needed to compensate for the
stripping away of water molecules can only be equaled by interactions with all four
subunit carbonyls, and the interactions between the ion and the carbonyls from all four
subunits are maximized for the potassium ion. Since the sodium ion is too small to
interact with all four subunits, it is energetically unfavorable for the ion to dehydrate
and move through the channel.
Smaller cations such as Li+ and Na+ are excluded because their small dehydrated radius
do not allow binding to the selectivity filter sufficient to compensate for the energetic
penalty of desolvation. The larger alkali cations, Rb+ (radius 1.48Å) and Cs+ (radius
1.69 Å) do permeate the channel and their much higher electron density allows them to
be imaged in the channel to reveal single ion binding sites. The selectivity filter
contains two cation binding sites 7.5 Å apart. A third ion is seen in the central cavity,
below the selectivity filter (MacKinnon suggests that the high electron density within
this cavity may not be an ionic binding site per se, but rather a "hydrated cation cloud").
Molecular surface of
KcsA and contour of
the pore
as a cutaway view
displaying the solvent
accessible surface of the K+
channel. The electrostatic
potential is color coded
red for negative and
blue for positive. The
yellow areas indicate
hydrophobic side chains of
several semi-conserved
residues in the inner cavity.
The green spheres represent
K+ ion positions in the
conduction pathway,
identified from Rb+ and Cs+
diffraction density.
For an ion crossing the membrane through a simple pore, the electrostatic energy is
maximal at the mid-point. This is quite contrary to what is seen here, where a
relatively favorable state is seen as net binding, inspite of the fact that the centre of
the pore is composed mainly of hydrophobic residues. Thus, the channel design
provides mechanisms to counteract the destabilizing influence of the low dielectric.
One design factor is the sheer size of the central cavity, which allows a substantial
number (50-60) of water molecules to be present at all times. By surrounding the
cation with water, the cation can polarize the water, rather than the hydrophobic
environment which has a very low dielectric constant, and the destabilizing influence
of the low dielectric environment is minimized.
Additionally, like all α-helices, the pore helix has a macroscopic dipole moment, the
negative end of which points towards the center of the cavity and further reduces the
destabilizing effect of the hydrophobic cavity residues. In fact, the electric field from
the pore helix dipole is greater (≥ 0.01 V/Å) than that due to the membrane potential
(100 mV over the membrane thickness).
This architecture ensures a relatively low resistance pathway from the cytoplasm to
the selectivity filter, allowing for a high throughput rate of permeation.
The Selectivity Filter
The side chains of the amino acids in the selectivity filter do not
participate directly in cation-coordination. Rather this is accomplished by
their main chain carbonyl groups. The main chain atoms of the signature
sequence create a stack of sequential oxygen rings and allow many sites
of proper dimensions for coordinating a K+ ion. A water molecule is
apparently located between the two K+ ions within the selectivity filter,
but the bound cations remain dehydrated when bound to this extremely
narrow region of the permeation pathway. The inner site of the two may
actually consist of two overlapping sites, with rapid movement of an ion
between them indicative of the flat energy landscape necessary for rapid
motion through the filter.
Detailed (stereo) view of the K+ selectivity filter, with the nearest subunit
removed. The Val (V) and Tyr (Y) side chains are directed away from the
filter. Two K+ ions are shown (green), roughly 7.5 Å apart, with a single
water molecule between them (red). The lower (inner) K+ is depicted as
in rapid equilibrium between adjacent coordination sites.
Interaction between the side chains of the TVGY(G) (Thr-Val-Gly-TyrGly) and the inner helices forms a large "sheet" of aromatic amino acids
which, via hydrogen bonding between Tyr OH groups and van der Waals
interactions within this sheet, ensure the pore diameter remains
constant. A glutamate residue, behind the selectivity filter, probably
hydrogen bonds to one of the peptide NH groups of the GYG sequence
and also holds the filter in place.
Such features may play an important role in preventing Na+ ions from
being accommodated by the selectivity filter. Unless there is a
mechanism to stabilize the diameter of the selectivity filter, a Na+ ion
may induce the filter to narrow, thereby solvating the Na+ more
effectively and allowing its permeation.
Section of the KcsA
channel perpendicular
to the pore, through the
selectivity filter, and
viewed from the
cytoplasm (from
“below”).
The view highlights the
network of aromatic
residues surrounding the
filter and proposed to
hold it at a fixed
diameter. Tyrosine-78
(Y78) interacts through
hydrogen bonding and
van der Waals contacts
with two Trp residues
(W67 and W68) from the
pore helix.
When an ion enters the selectivity filter, it must become dehydrated. Dehydration of a
cation is a very energetically expensive process and thus, the carbonyl oxygens must
take the place of the water, stabilizing the dehydrated cation. K+ ions apparently fit into
this filter so precisely that the energetic cost of dehydration is almost balanced by the
stabilization effect of the main chain carbonyl groups. However, this strong
coordination is incommensurate with rapid flux of K+ ions through the pore. Several
models have been proposed which involve repulsion between the two coordinated
cations within the filter, thus allowing rapid permeation. The 7.5 Å distance between
cation binding sites within the filter is consistent with such a repulsive effect. Up to
three potassium ions can fit into the channel at one time - two in the selectivity filter
and one below it at the top of the vestibule. If only one ion is present in the selectivity
filter, the interactions are too strong to allow rapid conductance through the channel.
This is consistent with the fact that conductance rates are dependent on ion
concentrations. Once a second potassium ion enters the filter, the ions repel one another.
The repulsive force between ions equals the attractive forces between the first ion and
the filter. This allows the first ion to pass into the inner chamber, where it can be
hydrated by water and stabilized by the pore helix dipole moment, expelling any ion
already in the chamber in to the intracellular space.
This represents strong evidence for a multi-ion (single file) conduction mechanism, as
originally proposed many years ago on the basis of biophysical studies of channel
conductances. The rate of conductance is therefore limited by the movement of the
potassium ion through the selectivity filter (12 Å), but, in this way, it is possible for the
channel to be highly specific for potassium ions and yet maintain high rates of
conduction.
Summary
(i) The pore is constructed of the inner helices arranged as an inverted teepee, with the
selectivity filter held at its wide end. This architecture also describes the pore of cyclic
nucleotide-gated channels and probably Na+ and Ca2+ channels as well.
(ii) The narrow selectivity filter is only 12 Å long, whereas the remainder of the pore
is wider and has a relatively inert hydrophobic lining. These structural and chemical
properties favor a high K+ throughput by minimizing the distance over which K+
interacts strongly with the channel.
(iii) A large water-filled cavity and helix dipoles help to overcome the high
electrostatic energy barrier facing a cation in the low dielectric membrane center.
(iv) The K+ selectivity filter is lined by carbonyl oxygen atoms, which provide
multiple closely spaced sites. The filter is constrained in an optimal geometry so that a
dehydrated K+ ion fits with proper coordination but the Na+ ion is too small.
(v) Two K+ ions at close proximity in the selectivity filter repel each other. The
repulsion overcomes the otherwise strong interaction between ion and protein and
allows rapid conduction in the setting of high selectivity.