![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 61, Issue 2, 407-414, February 2002
Department of Pharmacology, Medical Faculty, University of Tübingen, Tübingen, Germany
 |
Abstract |
|---|
 Top
 Abstract
 Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Sulfonylurea receptors (SURs) are
subunits of ATP-sensitive K+ channels (KATP
channels); they mediate the channel-closing effect of sulfonylureas
such as glibenclamide and the channel-activating effect of
KATP channel openers such as the pinacidil analog P1075. We
investigated the inhibition by MgATP and P1075 of glibenclamide binding
to SUR2B, the SUR subtype in smooth muscle. To increase specific
binding, experiments were also performed using SUR2B(Y1206S), a mutant
with higher affinity for glibenclamide than for the wild-type (KD = 4 versus 22 nM,
respectively) but otherwise exhibiting similar pharmacological
properties. In the absence of MgATP, [3H]glibenclamide
binding to both SURs was homogenous. MgATP inhibited [3H]glibenclamide binding to both SURs to 25% by
reducing the apparent number of glibenclamide binding sites, leaving
the affinity unchanged. In the absence of MgATP, P1075 inhibited
[3H]glibenclamide binding in a monophasic manner with
Ki 
1 µM. In the presence of MgATP (1 mM), inhibition was biphasic with one Ki
value resembling the true affinity of P1075 for SUR2B (2-6 nM) and the
other resembling Ki in the absence of MgATP
(1 µM). The data show that (1) MgATP induces heterogeneity in the
glibenclamide sites; (2) the high-affinity glibenclamide sites
remaining with MgATP are linked to two classes of P1075 sites; and (3)
P1075 interacts specifically with SUR2B also in the absence of MgATP. The data are discussed with the assumption that SUR2B, expressed alone,
forms tetramers; that MgATP induces allosteric interactions between the
subunits; and that mixed SUR2B-glibenclamide-P1075 complexes can exist
at equilibrium.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
KATP
channels are a class of K+ channels composed of
pore-forming subunits of type Kir6.x and sulfonylurea receptors (SURs) arranged as a hetero-octameric complex
(Kir6.x/SURx)4 (Clement et al., 1997
; Shyng and
Nichols, 1997; Gonoi and Seino, 2000). These channels are closed by
intracellular ATP and are opened by MgADP, thereby linking the
metabolic state of the cell to membrane potential and excitability
(Ashcroft and Ashcroft, 1990). In addition, they are the target of the
hypoglycemic sulfonylureas (SUs) such as glibenclamide, which induce
channel closure, and of the KATP channel openers,
a structurally heterogeneous class of compounds (Ashcroft and Ashcroft,
1990), which is exemplified here by the pinacidil analog P1075 (Quast
et al., 1993).
SUR, a member of the ATP-binding cassette proteins, is endowed with two
nucleotide-binding domains that exhibit ATPase activity and control
channel opening (Aguilar-Bryan et al., 1995; Bienengraeber et al.,
2000; Zingman et al., 2001). SUR also affords the binding sites for SUs
and openers (Hambrock et al., 1998; Schwanstecher et al., 1998). SUR1
is very sensitive to the SUs but shows little sensitivity to most
openers, whereas the contrary is true for the two SUR2 subtypes, which
differ only in the last 42 amino acids (Inagaki et al., 1996; Isomoto
et al., 1996). Essential parts of the binding sites for P1075 and for
glibenclamide are located in close proximity on the last transmembrane
domain of SUR (Ashfield et al., 1999; Uhde et al., 1999; Babenko et
al., 2000; Moreau et al., 2000; Mikhailov et al., 2001).
These structural features are the foundation of allosteric coupling
effects that exist between nucleotide, SU, and opener binding to SUR.
First, there is the positive allosteric interaction between MgATP and
opener binding. MgATP and hydrolyzable analogs induce
high-affinity opener binding to SUR2 and sensitivity of SUR1 to
pinacidil and diazoxide (Niki and Ashcroft, 1991; Schwanstecher et al.,
1991, 1992a, 1998; Quast et al., 1993; Hambrock et al., 1998, 1999).
Conversely, openers enhance the ATPase activity of SUR2A, thereby
promoting channel activation (Bienengraeber et al., 2000). However,
high concentrations of openers activate KATP channels also in the absence of MgATP, and MgATP sensitizes the channel
for the opener by slowing down the rate of channel closing upon washout
of pinacidil, an effect that is much more pronounced at the
Kir6.2/SUR2B than at the Kir6.2/SUR2A channel (Ashcroft and Gribble,
2000; Reimann et al., 2000). However, binding of openers such as P1075
in the absence of MgATP has not yet been studied.
Second, there is the negative allosteric interaction between nucleotide
and SU binding. Nucleotides reduce glibenclamide binding to
SUR1-containing channels by decreasing the affinity without affecting
the number of binding sites (Niki et al., 1990; Schwanstecher et al.,
1991, 1992b). For the SUR2 subtypes, the nucleotide regulation of
glibenclamide binding has not yet been determined. Third, there is a
negative allosteric coupling between opener and SU binding [SUR1 (Niki
and Ashcroft, 1991; Schwanstecher et al., 1992a, 1998) and SUR2 (Bray
and Quast, 1992; Hambrock et al., 1998, 1999)]. Generally, this leads
to mutually exclusive binding between the two. However, we have
observed that in rat glomeruli and in A10 cells (a cell line derived
from rat aorta), glibenclamide was unable to completely inhibit binding
of [3H]P1075 (Metzger and Quast, 1996; Russ et
al., 1997). Hence, the relationship between opener and SU binding to
SUR requires re-examination.
Here, we study the regulation of glibenclamide binding to SUR2B by
MgATP and the inhibition of this binding by P1075. We show that the
relatively low affinity of glibenclamide for SUR2B
(KD 22 nM), together with a high
level of nonspecific binding, makes the analysis of experiments,
particularly in the presence of MgATP, difficult. Therefore,
experiments were also performed using a SUR2B mutant in which Tyr 1206 of SUR2 (mouse numbering) is replaced by Ser, the amino acid located in
the corresponding position of SUR1 (Ashfield et al., 1999; Toman et
al., 2000). This mutation leads to a 5-fold increase in affinity for
glibenclamide (KD 4 nM) but keeps
the high affinity of SUR2B for P1075 almost unchanged (6.5 nM) (Toman
et al., 2000; Hambrock et al., 2001). Hence, this mutant allows
radioligand binding studies with both [3H]glibenclamide and
[3H]P1075 to be performed with sufficient
precision. Examining the properties of
[3H]glibenclamide binding in the presence of
MgATP to SUR2B (wild-type and mutant) revealed unexpected complexities,
which are discussed while assuming that SUR2B forms homomultimers with
intersubunit interactions.
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Site-Directed Mutagenesis and Cell Transfection.
The mutant
SUR2B(Y1206S) was constructed from the mouse clone (GenBank accession
number D86038; Isomoto et al., 1996) using the QuikChange Site-Directed
Mutagenesis System (Stratagene, Amsterdam, The Netherlands) as
described previously (Hambrock et al., 2001). Human embryonic kidney
(HEK) 293 cells were cultured in minimal essential medium containing
glutamine, supplemented with 10% fetal bovine serum and 20 µg/ml
gentamicin. Cells were transfected with the mammalian expression vector
pcDNA3.1 (Invitrogen, Carlsbad, CA) containing the coding sequence of
wild-type or mutant SUR2B using LipofectAMINE and Opti-MEM
(Invitrogen) according to the manufacturer's instructions
(Hambrock et al., 1998). Stably transfected cells were isolated in the
presence of 700 µg of geneticin/ml of medium for the first 3 weeks
and 300 µg/ml later; 1 week before membrane preparation, the
antibiotic was withdrawn. Membranes were prepared as described
previously (Hambrock et al., 1998).
Equilibrium Binding Experiments.
For saturation binding
experiments of [3H]glibenclamide to
SUR2B(Y1206S), membranes (0.1-0.4 mg of protein/ml) were incubated with [3H]glibenclamide (0.7-20 nM) in a total
volume of 1 ml at 37°C and pH 7.4 for 15 min in an incubation buffer
containing 5 mM HEPES, 139 mM NaCl, and 5 mM KCl; the buffer was
supplemented with MgCl2 (0/2.2 mM), EDTA (1/0
mM), and Na2ATP (0/1 mM). For competition
experiments, membranes (0.4-0.5 and 0.1-0.2 mg of protein/ml for
wild-type and mutant, respectively) were added to the same incubation
buffer (+MgATP or EDTA), supplemented with the radioligand
[3H]glibenclamide 5 nM (wild-type)
or 2.5 nM (mutant), [3H]P1075 1.5 nM], and the inhibitor of interest. After equilibrium had been reached
(15 min for [3H]glibenclamide and 25 min for
[3H]P1075), incubation was stopped by diluting
0.3-ml aliquots (in triplicate) in 8 ml of ice-cold quench solution
[50 mM Tris, 154 mM NaCl, pH 7.4]. Bound and free ligands were
separated by rapid filtration over GF/B filters (Whatman, Clifton, NJ),
washed twice with quench solution, and counted for
3H in the presence of 6 ml of scintillant (Ultima
Gold; Packard Instrument Co., Meriden, CT). Nonspecific binding
(BNS) of
[3H]glibenclamide/[3H]P1075
was determined in the presence of 100/10 µM P1075 (Hambrock et al.,
2001). In the absence of MgATP, specific
[3H]glibenclamide binding to wild-type SUR2B
was approximately 30% of total binding (BTOT)
(Fig. 1), to mutant SUR2B, approximately 75%.
|
Data Analysis, Modeling and Statistics.
In saturation
experiments, nonspecific binding (BNS) was
proportional to the free-label concentration (L) and was fitted to the
equation, BNS = a × L, where a is the
proportionality constant. Total binding (BTOT)
was then analyzed as the sum of specific and nonspecific binding and
was fitted to the equation,
|
(1) |
|
(2) |
logIC50,i; x is
the concentration of the compound under study with px = logx.
IC50 values were converted into inhibition constants (Ki) by correcting for the
presence of the radioligand, L, according to the equation
|
(3) |
), with both tests producing
identical results. Amplitudes and pIC50 values
are normally distributed and were compared by the use of a one-way
analysis of variance. In the case of only two groups, a two-tailed
unpaired Student's t test was used. In this article,
IC50 values are given with the 95% confidence
interval in parentheses. In calculations involving two mean values with
standard errors, propagation of errors was taken into account.
Materials and Solutions. [3H]P1075 [specific activity, 4.5 TBq (117 Ci)/mmol] was purchased from Amersham Buchler (Braunschweig, Germany) and [3H]glibenclamide [specific activity, 1.85 TBq (52 Ci)/mmol] from PerkinElmer Life Sciences (Boston, MA). The reagents and media used for cell culture and transfection were from Invitrogen. Na2ATP was from Roche Molecular Biochemicals (Mannheim, Germany); glibenclamide was from Sigma Chemie (Deisenhofen, Germany); and P1075 was a kind gift from Leo Pharmaceuticals (Ballerup, Denmark). Glibenclamide and P1075 were dissolved in dimethyl sulfoxide/ethanol (50:50, v/v) to produce stock solutions of 50 mM. These were further diluted with the same solvent or with incubation buffer; the final solvent concentration was lower than 0.3%.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
MgATP and [3H]Glibenclamide Binding to SUR2B.
Fig. 1a (
) shows the inhibition of
[3H]glibenclamide binding by unlabeled
glibenclamide in the absence of MgATP (1 mM EDTA, no
Mg2+, no ATP added) in membranes from HEK 293 cells expressing SUR2B. Glibenclamide displaced approximately 55% of
the radioactivity bound. For an analysis of the inhibition curve, one
has to consider that membranes from nontransfected HEK 293 cells and
from cells transfected with the expression vector alone possess
endogenous glibenclamide binding sites with
KD 300 nM (Hambrock et al., 2001).
Hence, the [3H]glibenclamide inhibition curve
is expected to be biphasic with a specific component representing
displacement of the radioligand from SUR2B and a second component
representing the endogenous glibenclamide sites. The opener P1075
completely displaces [3H]glibenclamide from the
KATP channel in rat aorta (presumably Kir6.1/SUR2B) and from SUR2B(Y1206S), but it does not affect
glibenclamide binding to the endogenous sites (Löffler and Quast,
1997; Hambrock et al., 2001; Fig. 1c). The inhibition curve in Fig. 1a
was therefore analyzed for two binding components, with the amplitude
of one component kept fixed at the level of P1075-sensitive binding; Hill coefficients were set to 1. The P1075-sensitive component, which
represents binding to SUR2B, contributed 30 ± 2% to total binding (Fig. 1a) with a KD value of
22 nM (95% confidence interval: 8,63) (Table
1). This
KD value, obtained in membranes in the absence of MgATP, is in excellent agreement with the value of 32 nM
(16,65) obtained from intact HEK 293 cells stably expressing SUR2B
(i.e., in the presence of MgATP) (Russ et al., 1999). The low-affinity
component (i.e., the binding to the endogenous HEK 293 cell membrane
proteins) contributed approximately 25 ± 4% to total binding
(Fig. 1a).
|
) shows that MgATP reduced glibenclamide-sensitive [3H]glibenclamide binding to 50 ± 10% of
that in the absence of MgATP. Although this was quite small in
amplitude, two-component analysis was performed as described above, and
a KD value of 22 nM (13,38) was
obtained for glibenclamide binding to SUR2B at 1 mM MgATP (i.e., the
same value as that obtained in the absence of MgATP). A comparison of
the amplitudes of the specific binding component showed that MgATP
reduced [3H]glibenclamide binding to SUR2B to
25 ± 5%. Because MgATP did not affect the
KD value, it seems that MgATP reduced
[3H] glibenclamide binding to SUR2B by
decreasing the apparent number of binding sites without affecting the
affinity of the remaining sites. Direct confirmation of this point
requires saturation binding experiments; in such experiments, however,
nonspecific binding would totally obscure specific binding at the
radioligand concentrations required.
Fig. 1b presents the concentration dependence of the MgATP effect. In
these experiments, MgATP reduced
[3H] glibenclamide binding to SUR2B to
26 ± 8% of that in the absence in MgATP, with
IC50 = 6.2 µM and Hill coefficient 1 (Table 1).
MgATP and [3H]Glibenclamide Binding to
SUR2B(Y1206S).
To obtain more precise information on
glibenclamide binding to SUR2B in the presence of MgATP, an improved
ratio of specific to nonspecific binding was required. Therefore,
further experiments were conducted using the mutant SUR2B(Y1206S). We
showed previously that this mutant binds glibenclamide in the presence
of MgATP (1 mM) with KD = 3.4 nM
(Hambrock et al., 2001) (i.e., with an approximately 7-fold higher
affinity than wild-type SUR2B). Figure 2
shows that MgATP reduced [3H]glibenclamide
binding to 26 ± 3% (n = 3), with an
IC50 of 8.7 µM (6.3,12.0) and a Hill
coefficient of 1.20 ± 0.02 (>1, p < 0.05). Table 1 shows that these values are in excellent agreement with those
obtained with wild-type SUR2B. In 15 additional experiments, it was
confirmed that MgATP (1 mM) reduced
[3H]glibenclamide binding to 24.0 ± 0.1%.
|
MgATP/+MgATP): KD= 4.1 (3.0,5.6)/3.9 (3.0,4.8) nM and BMAX = 1150 ± 176/350 ± 29 fmol/mg, respectively
(n = 4). Additional experiments (not illustrated)
showed that in the absence of Mg2+ (presence of 1 mM EDTA), neither ATP nor ADP (1 mM) affected [3H]glibenclamide binding, nor did
Mg2+ (1.2 mM) in the absence of nucleotides.
Inhibition of [3H]Glibenclamide Binding to Wild-Type
and Mutant SUR2B by P1075.
The experiments using wild-type SUR2B
are illustrated in Fig. 1c. In the absence of MgATP, the inhibition
curve was regular, with Hill coefficient 1 and
Ki = 720 nM (300,1740), showing that
the opener bound to SUR2B also in the absence of MgATP. In the presence
of MgATP, the amplitude was small and the curve very shallow, giving an
approximate IC50 of 100 nM and Hill
coefficient 0.3. The latter indicated either negative
cooperativity or heterogeneity of binding sites. The two-component fit
shown in Fig. 1c was statistically superior to the Hill fit and gave
two amplitudes of similar size with Ki
values of 1.7 (0.5,6.3) and 1100 (980,1290) nM (Table 1).
and , and Table 1). In the
absence of MgATP, the inhibition curve was monophasic (Hill
coefficient, 1.1 ± 0.1), with Ki = 1.2 µM (Table 1). In the presence of MgATP, the inhibition curve was strongly biphasic with the high-affinity component comprising 46 ± 4% of the total amplitude with a
Ki value of 5.9 nM; the low-affinity
component gave a Ki value of 710 nM
(Fig. 3 and Table 1) (Hambrock et al., 2001).
|
/+P1075): KD = 3.9 (3.2,4.7)/9.3 (7.1,12.3) nM and BMAX = 454 ± 24/395 ± 18 fmol/mg of protein. Figure
4 gives a typical example of these
experiments, which were difficult because in the presence of MgATP and
P1075, [3H]glibenclamide binding is low. The
data show that by the completion of the high-affinity component, P1075
has increased the KD value of
[3H]glibenclamide binding 2.4 times, leaving
the BMAX value essentially unchanged. An
easy calculation based on eq. 1 shows that increasing the
KD value from 3.9 to 9.3 nM at
constant BMAX decreases the binding of 2 nM
[3H]glibenclamide by 50%, thus
quantitatively accounting for the amplitude of the high-affinity
component.
|
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Modeling. This study has produced two major findings. First, MgATP decreased [3H]glibenclamide binding to 25% by reducing the number of binding sites. This means that MgATP induced a heterogeneity in the previously homogeneous and independent glibenclamide sites, leaving one of four sites unchanged and shifting the other three to a low-affinity state so that they were no longer detected by the binding assay. Second, for both mutant and wild-type SUR2B, the class of high-affinity glibenclamide sites remaining in the presence of saturating MgATP was inhibited by P1075 in a biphasic manner.
As an explanation, one might envisage a two-state scheme in which SUR exists in two states, R1 and R2. R1, which predominates in the absence of MgATP, has a high affinity for SUs and a low affinity for openers. MgATP then acts as the allosteric modifier, shifting the equilibrium toward R2, which has a low affinity for SUs and a high affinity for openers. If each reaction step of the scheme is at equilibrium (principle of detailed balance), the model predicts that MgATP shifts the glibenclamide binding curve in a homogeneous manner toward higher concentrations, leaving no fraction with unchanged affinity; in addition, monophasic inhibition curves of glibenclamide binding by openers are predicted (Janin, 1973;
Boeynaems and Dumont, 1980). A more realistic model takes into account
that the transition R1 
R2 is fueled by the hydrolysis of ATP
(Bienengraeber et al., 2000; Zingman et al., 2001) and is irreversible
at high MgATP concentrations.1 Conversely,
glibenclamide binding leads to the release of ATP from SUR (Ueda et
al., 1999) and the SUR-glibenclamide complex R2G
may return to the ground state: R2G R1G. In such an open system, in which an influx
of energy drives the reaction cycle, a monomeric receptor with a single
ligand binding site can display cooperativity in ligand binding at
steady state (Boeynaems and Dumont, 1980).1 This
particular model does not fit the data. Ironically, the scheme in which
MgATP hydrolysis drives the
R1
R2 transitions in the opposite direction (i.e., glibenclamide binding favors MgATP
hydrolysis) and in which the high-affinity state for openers is
R1 (i.e., the state on the absence of MgATP)
could be perfectly fitted to the data in Figs. 1 through 3 (U. Quast,
unpublished observations). We have not found a modification of such an
energy-driven mechanism that is in accordance with both the data and
the biochemical knowledge of SUR.
A mechanism meeting these requirements assumes that SUR forms multimers
with allosteric coupling of binding sites on different subunits. Taking
into account that the completely assembled channel contains four SURs
(Clement et al., 1997; Shyng and Nichols, 1997) and that isolated
nucleotide binding fold-1 of SUR1 spontaneously forms tetramers
(Mikhailov and Ashcroft, 2000), such homomers might be tetramers (Fig.
5). One could speculate that such a
tetramer serves as an outer shell in which the four Kir channels
assemble to form the complete channel. It should be noted that although the tetramer model explains the data (see below), biochemical evidence
for tetramer formation of SUR expressed alone has not yet been
presented and that alternative explanations of the data may be
possible.
|
Negative Allosteric Coupling between Nucleotide and Glibenclamide
Sites.
First, the tetramer model has to explain that MgATP affects
three of the four binding sites of tetrameric SUR2B, leaving one site
unchanged. One may speculate that the nucleotide induces a strong
negative allosteric coupling between the four glibenclamide sites such
that binding of a first glibenclamide molecule greatly impedes binding
of the other three, and an asymmetry is induced. Alternatively, MgATP
could lead to a major symmetrical rearrangement of the tetramer so that
only one site remains sterically accessible. In this case, however, it
is not easy to explain that the remaining site has the same affinity
for glibenclamide as the four sites in the absence of MgATP. Whatever
the mechanism, if in the presence of MgATP only one glibenclamide
molecule binds with high affinity to tetrameric SUR2B, then this most
likely holds true for the complete channel. If so, binding of one
glibenclamide molecule per channel should induce high-affinity channel
block (in the presence of MgATP). This stoichiometry is in agreement
with that found by Dörschner et al. (1999) for glibenclamide
block of the Kir6.2/SUR1 channel in the absence of MgATP, but it is in
contrast to our earlier proposal that binding of four molecules of
glibenclamide is required for block of the Kir6.1/SUR2B channel in
intact cells (Russ et al., 1999). Obviously, this proposal, determined
from the comparison of binding and channel inhibition curves, is not compatible with the tetramer model.
;
Schwanstecher et al., 1991, 1992b); this was recently confirmed using
recombinant SUR1 expressed alone (A. Hambrock, C. Löffler-Walz, and U. Quast, in preparation). Conversely,
glibenclamide induced the release of bound ATP from SUR1 (Ueda et al.,
1999). In contrast to our observations with SUR2B, however, nucleotides
acted at SUR1 by homogeneously reducing the affinity of all
glibenclamide sites, leaving their number unchanged [MgATP
(Schwanstecher et al., 1992b; A. Hambrock, C. Löffler-Walz, U. Quast, in preparation) and MgADP (Niki et al.,
1990)]. This indicates a profound difference between the SUR subtypes.
Inhibition of [3H]Glibenclamide Binding by P1075. In the presence of MgATP, the inhibition curve of P1075 was biphasic. Within the framework of the tetramer model, one has to assume that binding of one glibenclamide molecule per tetramer not only affects the other three glibenclamide sites (see above), but it creates an asymmetry in the previously homogeneous four opener sites: it leaves some in the original high-affinity state and moves others into a low-affinity state. It seems plausible to assume that all opener sites are in the high-affinity state except for the one at the subunit occupied by glibenclamide (Fig. 5). The high-affinity component of the [3H]glibenclamide-opener inhibition curve in the presence of MgATP then reflects opener binding to the three high-affinity sites. This weakens the affinity of glibenclamide binding by a negative allosteric interaction between subunits of the tetramer, and the increase in KD causes the reduction in [3H]glibenclamide binding of this component. The low-affinity component reflects opener binding to the last (low-affinity) site and leads to complete displacement of [3H]glibenclamide from SUR. The experiments presented in Fig. 4 quantitatively support this model.
The model described in Fig. 5 predicts that at equilibrium, mixed states of tetrameric SUR2B exist in which P1075 and glibenclamide sites at different subunits are occupied simultaneously. Hence, tetrameric SUR is not described by the concerted transition model of Monod et al. (1965) (Janin, 1973), which predicts that the four subunits are all
either in the opener- or the glibenclamide-binding conformation, but by
the induced-fit model (sequential model) of Koshland et al. (1966). The
negative allosteric interactions between the glibenclamide sites are
strong (see above); in contrast, interactions between P1075 and
glibenclamide sites on different subunits are subtle: the occupation of
the (high-affinity) opener sites by P1075 increases the
KD of glibenclamide binding to an opener-free subunit 2.4 times (Fig. 4). Such allosteric interactions between SUR subunits in the presence of MgATP have also been observed in [3H]glibenclamide-P1075 and
[3H]P1075-glibenclamide competition experiments
upon coexpression of mutant and wild-type SUR2B with Kir6.x. They
depended on the subtype of Kir6.x and were more prominent in whole
cells than in membranes (Hambrock et al., 2001). Hence, in the whole
channel, allosteric interactions between SUR2B subunits exist and are
modulated by the Kir subunit in a subtype-specific way.
MgATP-Free State of SUR. In the absence of MgATP, the binding sites of wild-type and mutant SUR2B for [3H]glibenclamide were homogenous and independent; in addition, P1075 displaced the radioligand with monophasic inhibition curves. This suggested that in the absence of MgATP, there was no allosteric interaction between the glibenclamide sites neither within the tetramer nor between the glibenclamide and the P1075 sites on different subunits. The Ki value of P1075 against [3H]glibenclamide in the absence of MgATP was similar to that of the low-affinity component of the inhibition curve in the presence of MgATP. In both cases, one monitors opener binding to a SUR subunit occupied by glibenclamide. It seems that the presence of glibenclamide modified the opener site of this subunit profoundly, rendering opener binding difficult. This modification was similar in the presence or absence of MgATP. The lack of subunit interactions in the absence of MgATP also explains that [3H]glibenclamide-P1075 inhibition experiments under these conditions do not provide information on how the opener interacts with the glibenclamide-free subunits of the tetramer.
In conclusion, the negative allosteric coupling between MgATP and glibenclamide binding and the biphasic opener inhibition curve can be explained by a tetrameric model of SUR2B in which MgATP induces allosteric interactions between the subunits. Direct evidence for tetramer formation of SUR2B is desirable.| |
Acknowledgments |
|---|
We thank Drs Y. Kurachi and Y. Horio (Osaka, Japan) for the generous gift of the vector encoding murine SUR2B. The excellent technical assistance of C. Müller with cell culture and transfections and helpful discussions with Dr. Ulrich Russ (Tübingen, Germany) are gratefully acknowledged.
| |
Footnotes |
|---|
Received May 9, 2001; Accepted November 7, 2001
1 We thank an anonymous reviewer for indicating this.
Dr. Ulrich Quast, Department of Pharmacology, University of Tübingen, Wilhelmstr. 56, D-72074 Tübingen, Germany. E-mail: ulrich.quast{at}uni-tuebingen.de
| |
Abbreviations |
|---|
SUR, sulfonylurea receptor; P1075, N-cyano-N'-(1,1-dimethylpropyl)-N''-3-pyridylguanidine); SU, sulfonylurea; KATP channel, ATP-sensitive K+ channel; BNS, nonspecific binding; BTOT, total binding; HEK, human embryonic kidney; BMAX, maximum concentration of binding sites; BS, specific binding; KD, equilibrium dissociation constant of the radioligand; Kir, inwardly rectifying K+ channel.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
cell high-affinity sulfonylurea receptor: a regulator of insulin secretion.
Science (Wash DC)
268:
423-426
separate domains of the regulatory subunits of KATP channel isoforms are required for selective interaction with K+ channel openers.
J Biol Chem
275:
717-720-cell KATP channels.
J Biol Chem
275:
3360-3364-cell KATP channel.
FEBS Lett
499:
154-160[CrossRef][Medline].-cell line, HIT T15.
Biochim Biophys Acta
1133:
95-101[Medline].-cell glibenclamide receptor is an ADP-binding protein.
Biochem J
268:
713-718[Medline].-cell line and rat cerebral cortex.
Br J Pharmacol
106:
295-301[Medline].This article has been cited by other articles:
|
|
 |
|
  M. Winkler, D. Stephan, S. Bieger, P. Kuhner, F. Wolff, and U. Quast Testing the Bipartite Model of the Sulfonylurea Receptor Binding Site: Binding of A-, B-, and A + B-Site Ligands J. Pharmacol. Exp. Ther., August 1, 2007; 322(2): 701 - 708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hambrock, T. Kayar, D. Stumpp, and H. Osswald Effect of Two Amino Acids in TM17 of Sulfonylurea Receptor SUR1 on the Binding of ATP-Sensitive K+ Channel Modulators Diabetes, December 1, 2004; 53(suppl_3): S128 - S134. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Proks, F. Reimann, N. Green, F. Gribble, and F. Ashcroft Sulfonylurea Stimulation of Insulin Secretion Diabetes, December 1, 2002; 51(90003): S368 - 376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Lange, C. Loffler-Walz, H. C. Englert, A. Hambrock, U. Russ, and U. Quast The Stereoenantiomers of a Pinacidil Analog Open or Close Cloned ATP-sensitive K+ Channels J. Biol. Chem., October 18, 2002; 277(43): 40196 - 40205. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. K. Hansen, I. T. Christensen, J. B. Hansen, R. D. Carr, F. M. Ashcroft, and P. Wahl Differential Interactions of Nateglinide and Repaglinide on the Human {beta}-Cell Sulphonylurea Receptor 1 Diabetes, September 1, 2002; 51(9): 2789 - 2795. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
) and with [3H]glibenclamide in the presence
(
