0163-769X/99/$03.00/0
Endocrine Reviews 20(2): 101–135
Copyright © 1999 by The Endocrine Society
Printed in U.S.A.
Molecular Biology of Adenosine Triphosphate-Sensitive
Potassium Channels*
LYDIA AGUILAR-BRYAN
AND
JOSEPH BRYAN
Departments of Medicine (L.A.-B.) and Cell Biology (J.B.), Baylor College of Medicine, Houston, Texas
77030
I. Introduction
II. How Are KATP Channels Defined?
III. How Do KATP Channels Affect the Membrane Potential of Pancreatic b-Cells?
IV. KATP Channel Subunits
A. The KIR family of inwardly rectifying K1 channels
B. Sulfonylurea receptors
V. Reconstitution of KATP Channel Activity from SUR1
and KIR6.2
A. The question of “promiscuous coupling” of SUR1
with other inward rectifiers
VI. KATP Channel Structure
A. KIR6.2 forms the pore of a KATP channel
B. SUR1 and KIR6.x are physically associated
C. Coexpression with KIR6.2 affects the maturation of
SUR1
D. Complex glycosylated SUR1 and KIR6.2 assemble
a large multimer
E. A 1:1 stoichiometry of SUR to KIR is both necessary
and sufficient to make KATP channels
F. Other KIR channels are tetramers
G. The stoichiometry of active b-cell KATP channels is
(SUR1/KIR6.2)4
VII. Regulation of KATP Channel Activity
A. How do ATP and ADP exert their effects on KATP
channels?
B. Where are the nucleotide binding sites located?
C. C-terminally truncated KIR6.2 channels show abnormal kinetics
D. Coexpression of KIR6.2DC subunits with SUR restores normal KATP channel activity
E. Why are KIR6.2 channels silent?
F. The N terminus of KIR6.2 limits burst duration
G. Where do the openers bind and how do they work?
H. Do SURs have adenosine triphosphatase (ATPase)
activity?
I. Do SURs have transport activity?
J. Is there an endogenous substrate?
VIII. Human SUR1 and KIR6.2 Genes
IX. KATP Channels and Persistent Hyperinsulinemic Hypoglycemia of Infancy (PHHI)
X.
XI.
XII.
XIII.
XIV.
A. HI-GK
B. HI-GlnDH
C. HI-“unknown”
D. HI-KIR6.2
E. HI-SUR1
Linking PHHI to Defects in KATP Channel Activity
A. b-Cells from newborns diagnosed with “sporadic”
PHHI lack KATP channel activity
B. PHHI b-cells with the SUR1 exon 35 mutation lack
KATP channel activity
C. Why is there a lack of dominant negative mutations?
D. Development of mouse models
Other Issues
A. Nesidioblastosis does not cause PHHI
B. “Diffuse” vs. “focal” forms of PHHI
KATP and Non-Insulin-Dependent Diabetes Mellitus
(NIDDM)
A. b-Cell type KATP channels in the brain
The Leptin Connection
Summary and Conclusions
I. Introduction
I
ON CHANNELS are present in the plasma membrane and
intracellular organelles of all cells, where they coordinate
such diverse functions as neurotransmission, contraction,
secretion, and control of cell volume. Over the past decade
it has become increasingly clear that mutations in the genes
that encode the subunits of ion channels can result in pathological states in a wide variety of tissues, including those not
generally thought of as “excitable.” The number of ion channels that give rise to “channelopathies” is increasing. The list
currently spans all of the major channel types, including the
following nonexhaustive survey. Perhaps the most prevalent
ion channelopathy is cystic fibrosis (Online Inheritance in
Man, OMIM 219700), which is due to reduced chloride ion
transport resulting from mutations in the cystic fibrosis
transconductance regulator, CFTR, a member of the large
ATP-binding cassette (ABC) superfamily of proteins. A dominant form of myotonia congenita (OMIM 160800), Thomsen’s disease, is associated with mutations in CLCN1, a voltage-gated chloride channel. Mutations in the gene encoding
another member of this family, CLCN5, give rise to hypercalciuric nephrolithiasis (Dent’s disease; OMIM 300009), an
X-linked recessive disorder. Mutations in the genes encoding
the b- and g-subunits of the amiloride-sensitive, non-voltage-
Address reprint requests to: Joseph Bryan, Ph.D., Department of Cell
Biology, Baylor College of Medicine, Houston, Texas 77030 USA. E-mail:
[email protected]
* Supported by NIH Grants DK-44311, DK-52771, and DK-50750, the
Juvenile Diabetes Foundation International, and the American Diabetes
Association.
101
102
AGUILAR-BRYAN AND BRYAN
gated epithelial Na1 channels, ENac, activate channel activity resulting in Liddle’s syndrome (OMIM 177200), which is
associated with hypertension and hypokalemia. Various episodic disorders have been shown to arise from mutations of
voltage-dependent ion channels in nerve and muscle. Periodic paralysis I, or hypokalemic periodic paralysis (OMIM
170400), results from mutations in the gene encoding the
a1-subunit of the muscle dihydropyridine-sensitive calcium
channel, while periodic paralysis II, hyperkalemic periodic
paralysis (OMIM 170500), is associated with mutations in the
gene encoding a voltage-gated Na1 channel subunit. Mutations and CAG repeat expansions in the gene encoding a
second Ca21 channel a1-subunit isoform, termed isoform 4,
have been identified in familial hemiplegic migraine 1
(OMIM 141500), episodic ataxia type 2 (OMIM 108500), and
spinocerebellar ataxia 6 (OMIM 183086). Mutations in the
genes that encode potassium channels, perhaps the most
diverse class of ion channels, give rise to a number of disorders; for example, variants of the long QT syndrome, LQT1
(OMIM 192500) and LQT2 (OMIM 152427) are caused by
mutations in a voltage-gated K1 channel, KVLQT1, and in
HERG (human ether-a-go-go related gene), a Ca21-modulated K1 channel, respectively. Mutations in the KCNJ1 gene
encoding ROMK1, the first discovered potassium inward
rectifier (1) give rise to the antenatal variant of Bartter’s
syndrome with renal tubular hypokalemic alkalosis (OMIM
601678). Finally, mutations in the gene encoding a water
channel, aquaporin 2, cause the renal form of diabetes insipidus (OMIM 125800). Ion channelopathies have been reviewed recently (2–17). The reader is referred to the extensive
bibliographies cited in the Online Mendelian Inheritance in
Man database (http://www3.ncbi.nlm.nih.gov/Omim/) for
references to each of the channelopathies cited above and for
current information on others.
Our objective is to review the molecular biology of ATPsensitive K1 channels, or KATP channels, which are present
in pancreatic b-cells in the islets of Langerhans where they
play a key role in stimulus-secretion coupling by providing
a link between changes in metabolism and membrane electrical activity. Evidence suggests this channel is regulated by
changes in adenine nucleotide levels, an increase in the ATP/
ADP ratio, resulting from changes in glucose metabolism,
and is the target for a class of drugs called sulfonylureas used
in the treatment of non-insulin-dependent diabetes mellitus.
Closure of KATP channels as a result of increased glucose
metabolism or by sulfonylureas leads to the release of insulin. Nearly 50 mutations in either of the two subunits that
make up the b-cell KATP channel, namely, the high-affinity
sulfonylurea receptor, SUR1, and the K1 inward rectifier,
KIR6.2, are responsible for a channelopathy, the recessive
genetic form of persistent hyperinsulinemic hypoglycemia of
infancy (PHHI), characterized by the uncoupling of glucose
metabolism from b-cell electrical activity (see Refs. 18 –21 for
reviews). Investigation of this recessive form of PHHI has led
to the identification of at least two other causes of dominant
forms of persistent neonatal hypoglycemia and should provide insight into the regulation of the insulin-secretory pathway in both normal and in diabetic individuals.
The objective of this article is to review developments in
the molecular biology of ATP-sensitive potassium channels.
Vol. 20, No. 2
We will focus mainly on recent studies of the molecular
biology of the “classic” b-cell type KATP channels composed
of the high-affinity sulfonylurea receptor, SUR1 (OMIM
600509), and KIR6.2, a potassium inward rectifier subunit
(OMIM 600937), but will draw comparison with the striated
and smooth muscle type channels where appropriate. Over
the last several years, the subunits of the b-cell KATP channel
have been cloned, expressed, and reconstituted into functional nucleotide-sensitive channels and have been used to
investigate their overall structure and regulation by nucleotides, potassium channel openers, and sulfonylureas. The
availability of cDNAs for SUR1 (22) allowed the isolation of
the low-affinity sulfonylurea receptors, SUR2A and SUR2B
(23, 24). In humans, the mRNAs encoding the SUR2A and
SUR2B receptors originate from a single SUR2 gene by differential splicing of the last exon (19). Two KIR6.x isoforms
have been cloned, KIR6.1 (uKATP-1) (25) and KIR6.2 (BIR)
(26). Interestingly, the SUR and KIR genes are paired on
human chromosomes, with SUR1 and KIR6.2 adjacent on
chromosome 11, and SUR2 and KIR6.1 near each other on
chromosome 12. When the low-affinity receptors are reconstituted with KIR6.1 or KIR6.2, they produce KATP channels
with distinctive pharmacologies which, along with their tissue distribution, suggest they make up the cardiac-skeletal
and vascular smooth muscle KATP channels (23, 24, 27–29).
II. How Are KATP Channels Defined?
The application of ATP to the intracellular face of an excised patch was initially used to identify KATP channels.
These observations were made by Cook and Hales (30) in the
United States, by Noma (31) in Japan, and by Trube and
Hescheler (32) in Germany. The first reference to ATP-sensitive K1 channels that we can uncover is an abstract by
Trube and Hescheler (33) outlining the properties of the
cardiac KATP channel from isolated patches of cardiac cell
membranes. Historically, a large number of studies defined
the electrophysiological properties and pharmacology of
these channels, including their single-channel current properties, pattern of regulation by nucleotides, and response to
channel openers and blockers. It is now possible to begin to
define these channels in terms of their molecular composition. KATP channels are heteromultimers of two types of
subunits, inward rectifiers, KIR6.x, and sulfonylurea receptors, SURs, members of the ABC superfamily. The available
evidence, discussed in more detail below, indicates functional channels are assembled as tetramers (SUR/KIR)4. Table
1 relates the known subunits to channel types identified by
tissue type. The list is provisional and is not meant to be
exhaustive, as it is not known whether all the subunit isoTABLE 1. KATP channel isoforms
Sulfonylurea
receptor
SUR1
SUR2A
SUR2B
Inward
rectifier
Tissue/channel subtype
KIR6.1
KIR6.2
KIR6.1
KIR6.2
KIR6.1
KIR6.2
??
Pancreatic b-cell/neuronal
??
Cardiac/skeletal muscle
Vascular smooth muscle
Vascular smooth muscle
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
forms have been identified. This classification is based
largely on pharmacological criteria, and there remains a
strong need for tissue and cellular localization work to confirm this classification and to look for overlapping expression
of subunit types.
KATP channel activity can be characterized by its electrical
activity, by its sensitivity toward nucleotides, and by its
pharmacology. Figure 1 summarizes parts of this behavior.
During a “burst” the channels rapidly “flicker” between
open and closed states. The exact mechanism that gives rise
to these fast transitions is not known, but based on the determination of the structure and dynamics of the Streptomyces
lividans K1 channel, the “KcsA” channel, by crystallographic
(34) and spectroscopic methods (35–37) the short open and
closed states presumably arise because of stochastic movement of the a-helices that form the pore, or of amino acids
within the gate. Silent intervals or gaps delineate the bursts.
The application of ATP, without Mg21, reduces the time
spent in the open state (the mean open time) by prolonging
the lifetime of the gaps and by shortening the duration of the
bursts (38, 39). MgATP also inhibits channel activity. The IC50
for inhibition of the b-cell channel, SUR1/KIR6.2, by MgATP
is reported to be higher than for ATP42 (40), while the reverse
has been reported for the cardiac KATP channel, SUR2A/
KIR6.2 (41). The IC50 values for inhibition by ATP have generally been reported to be higher for the cardiac than for the
b-cell channel, although the variation of the IC50 values, as
well as the Hill coefficients, is quite large for both (42). Our
own values for the human SUR1/KIR6.2 (b-cell) and SUR2A/
KIR6.2 (ventricular myocyte) channels are in the low micromolar range, ;5 and 20 mm, respectively, with Hill coefficients near 1 with no dependence on Mg21 (28). Whether the
reported variations in ATP sensitivity are due to tissuespecific regulatory factors vs. experimental problems has not
been resolved. In the absence of Mg21, ADP is also inhibitory.
Under physiological conditions with Mg21 present, ADP
stimulates ATP-inhibited channels, providing one route for
regulation (43, 44).
Two pharmacological criteria have been used to define
and classify KATP channels. A diverse group of compounds,
referred to as potassium channel openers or KCOs (see Fig.
2), increase the mean open time, while sulfonylureas like
tolbutamide and glibenclamide reduce channel activity. Although dose-response curves have not always been carried
out, it is generally possible to distinguish two types of KATP
channels based on their sensitivity to sulfonylureas. The
b-cell/neuroendocrine/neuronal type channel is 100- to
1000-fold more sensitive than the SUR2 channels, as expected
from the higher affinity of SUR1 for sulfonylureas [dissociation constants (KDs) in the nanomolar range]. Similarly,
there are differences in response to KCOs, with the SUR1 and
SUR2B channels responding better to diazoxide than cardiac
channels. The differential sensitivity of the SUR2B channels
to pinacidil allows them to be discriminated from the SUR2A
channels. The binding sites for both sulfonylureas and KCOs
are believed to reside on the SUR, but the location of these
sites within the receptors remains to be elucidated.
Analysis of the current-voltage relations of KATP channels
indicate they are moderate inward rectifiers that conduct K1
ions better in the inward than the outward direction (as
103
shown in Fig. 1). As has been discovered for other members
of the KIR family (45– 49), the degree of rectification is dependent upon the presence of Mg21 or polyamines, such as
spermine, spermidine, and putrescine, on the intracellular
side of the channel (42, 49, 50). At positive membrane potentials these charged molecules are thought to enter the
pore, where they bind and slow the passage of potassium
ions. As discussed below, the degree of rectification of SUR1/
KIR6.2 channels was changed by mutating a single residue in
KIR6.2, the first indication that KIR6.2 formed the permeation
pathway. SUR does not appear to be necessary to form a pore
since expression of C-terminally truncated KIR6.2 subunits
alone has been shown to generate potassium channels with
a single-channel conductance similar to that of SUR/KIR6.2
channels (51).
III. How Do KATP Channels Affect the Membrane
Potential of Pancreatic b-Cells?
The function of KATP channels is best understood in pancreatic b-cells, the membrane potential of which is responsive
to external glucose concentration. b-Cells show a remarkably
complex electrical bursting behavior (see Fig. 3) in response
to an increase in glucose level. It is generally agreed that the
oscillatory electrical activity leads to oscillatory changes in
[Ca21]i, which drive pulsatile insulin release in isolated islets
(see for example Ref. 52). Various suggestions on how KATP
channels may participate in this response have been put
forward; these range from maintaining the resting membrane potential when glucose is low (53) to playing a major
role in control of bursting when glucose is high (54, 55) or
terminating individual bursts (56). The origin and control of
the electrical bursting activity in pancreatic b-cells remain
controversial, and a detailed discussion of this area is beyond
the scope of this review. The reader is directed to recent
articles for detailed discussions (57– 63). Our intention is to
provide a framework for understanding the molecular biology and structure of KATP channels, how they affect membrane potential and thus how they can affect bursting and
insulin release, and finally, how their absence leads to familial hyperinsulinism.
In b-cells, as in other cells, the Na1/K1 adenosine triphosphatase (ATPase) uses ATP to generate an asymmetric distribution of ions with [Na1]out . [Na1]in and [K1]in .
[K1]out. The opening of channels that conduct K1 will increase the permeability of the membrane for this ion and
affect the membrane potential, VM, in a predictable way (see
for example Ref. 64), hyperpolarizing the cell by shifting VM
toward the potassium ion equilibrium potential, EK, usually
less than 280 mV in mammalian cells. Closing K1 channels
will depolarize the cell membrane and open voltage-gated
Ca21 channels that allow Ca21 to flow into the cell.
As illustrated in Fig. 3, VM in the b-cell is regulated by
glucose metabolism. When the external glucose concentration is less than 3 mm, VM is near 265 to 270 mV. Increasing
the external glucose causes gradual depolarization to a new
steady-state level. Further increases cause the membrane to
depolarize to a threshold potential, near 250 mV, where
electrical activity, due in part to Ca21 currents through voltage-dependent Ca21 channels, is initiated. Several types of
104
AGUILAR-BRYAN AND BRYAN
Vol. 20, No. 2
FIG. 1. Electrical activity of human SUR1/KIR6.2 ATP-sensitive K1 channels. A, Recordings from excised membrane patches of COSm6 cells
transfected with human SUR1 and human KIR6.2 illustrate channel activity during perfusion of the cytoplasmic face of the patch with the
indicated reagents. Panels B and C are traces plotted at shorter time intervals chosen to show open and closed states of a single channel within
a burst of activity. D, A current vs. voltage plot to show the inward rectification (reduced current at more positive potentials) of KATP channels
and the effect of an added polyamine. The single-channel conductance is approximately 75 pS. E, The ATP inhibition curve is the average of
five experiments, and the bars are the 6SE. The activity in the absence of ATP was taken as 100%; the IC50 was 14.7 mM with a Hill coefficient
of ;1.1.
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
105
FIG. 2. Chemical structures of compounds that activate and block KATP channels.
pumps and ion channels have been identified in b-cells, which
could or have been suggested to contribute to this observed
electrical bursting behavior, including the following.
1. Na1/K1 ATPase. The asymmetric distribution of sodium
and potassium ions is maintained by the b-cell Na1/K1
ATPase, which has been identified and subjected to limited
study (65, 66). The activity of this enzyme is generally con-
sidered insensitive to KATP channel blockers and openers,
although high concentrations of glibenclamide have been
reported to be inhibitory (67). Inhibition of the Na1/K1
ATPase with ouabain results in membrane depolarization
and release of insulin. In a recent report, Ding et al. (56) used
manipulations of the activity of the Na1/K1 ATPase to
change the ATP/ADP ratio and affect KATP channel activity
106
AGUILAR-BRYAN AND BRYAN
Vol. 20, No. 2
FIG. 3. Illustration of electrical activity and [Ca21]i in a pancreatic b-cell. Islets, loaded with fura-2, were stimulated with 12 mM glucose, and
simultaneous recordings were made from the same b-cell using a microelectrode and a microfluorometer (modified from Fig. 1 in Ref. 60). The
earliest change, marked phase 0, is a gradual depolarization of the membrane and a decrease in [Ca21]i. An abrupt initiation of a sustained
period of depolarization with superimposed Ca21-dependent action potentials or spikes marks phase 1. [Ca21]i rises abruptly during phase 1,
the result of Ca21 mobilization from intracellular stores and from Ca21 influx through voltage-dependent Ca21 channels. Phase 1 can be
correlated temporally with first-phase insulin release. Phase 2 is marked by the initiation of steady-state oscillations in both membrane potential
and [Ca21]i, commonly called “bursting.” The frequency of bursting is dependent on the glucose level (see for example Ref. 88). F340/F380 is the
fluorescence ratio for fura-2 used as a measure of [Ca21]i. [Adapted with permission from I. D. Dukes et al.: Curr Opin Endocrinol Diab 4:262–271,
1997 (60). © Lippincott Williams and Wilkins.]
and suggest that a fall in [ATP]i (and presumably an increase
in [ADP]i), terminates a burst of electrical activity.
described recently in the bTC-3 b-cell line (80), but its physiological significance is uncertain.
2. Na1 channels. Early studies on the effects of veratridine and
tetrodotoxin suggested that Na1 channels were functionally
important for insulin release from pancreatic islets (68, 69),
although Donatsch et al. (70) found tetrodotoxin did not
affect insulin release from mouse islets. Several reports have
described a transient, voltage-dependent, inward Na1 current that is blocked by tetrodotoxin (71, 72). The canine channel shows steep activation and inactivation between 250 and
240 mV (73, 74). Philipson et al. (75) identified Na1 channel
subunit cDNAs in canine, human, and rodent islets, and in
hamster and mouse insulinoma cell lines, which are most
closely related to the rat brain III a1 isoform of Na1 channel
subunits. This conductance could play a role in the initial
depolarization, but its role is usually discounted in rodent
b-cells because the voltage dependence of inactivation of the
rodent channel suggests it will be largely inactive (71, 76).
Na1 channels may contribute to the action potential in canine
and human b-cells (73, 74), but their role is generally believed
to be minimal.
4. Ca21 channels. The importance of Ca21 in insulin release
has been reviewed by many others (81– 88). Insulin release is
attenuated in Ca21-deficient media and by the action of Ca21
channel blockers. The early reports describing Ca21 currents
in b-cells (89 –91) were followed by others identifying at least
two types of Ca21 channels in b-cells, which could be distinguished by their kinetics and pharmacology (71, 72, 92–
96). The larger conductance channel has the properties of a
fast deactivating, dihydropyridine-sensitive L-type Ca21
channel with an activation threshold near 230 mV. The
smaller conductance is similar to the T-type Ca21 channel, is
slowly deactivated, and has a lower activation threshold,
near 250 mV. The T-type channel is not sensitive to dihydropyridines, but is inhibited by NiCl2, a selective inhibitor
of other T-type calcium channels (96). It has been suggested
that the slow T-type channel could be responsible for the
slow Ca21 waves, while the L-type channels contribute the
spikes. The evidence for and against this idea has been reviewed by Satin and Smolen (58). The possible involvement
of T-type channels in control of insulin secretion has become
more accessible with the cloning of these channels (97, 98)
and the finding of T-type channel mRNAs in the pancreas.
In addition to the voltage-dependent Ca21 channels, a glucose-activated calcium channel has been reported in b-cells
(99).
3. Chloride channels. Several reports have suggested chloride
ions can affect insulin release and b-cell electrical activity (77,
78). The mechanism(s) of these effects are not known, and it
is unclear whether or not chloride ions are passively distributed across the b-cell membrane. Kinard and Satin (79) have
identified an ATP-sensitive chloride channel in b-cells,
termed ICl,islet, which is activated by cAMP, glibenclamide
(1–10 mm), and cell swelling. Decreasing [ATP]i reduces the
amplitude of the current; thus, Kinard and Satin suggest this
channel may be under metabolic control. ICl,islet mediates a
large inward current that would depolarize the b-cell membrane. A second, Ca21-dependent Cl2 conductance has been
5. Ca21-release-activated nonselective cation channels (ICRAN).
Several lines of evidence indicate the importance of a nonselective cation channel, which is activated by depletion of
internal Ca21 stores, in the regulation of b-cell electrical
bursting. This current has been termed ICRAN (60, 100, 101)
or ICRAC for Ca21 release-activated currents (102–105). The
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
use of the term ICRAC in this context appears to be somewhat
unfortunate and this channel, which has a conductance in the
20 – 40 picosiemens (pS) range, should not be confused with
the low-conductance Ca21 release-activated Ca21 channel,
ICRAC, described previously (106 –108).
Dukes et al. (60) have reviewed the evidence for the role of
ICRAN in bursting. This model underscores the importance of
Ca21 mobilization from the endoplasmic reticulum through
activation of the inositol triphosphate (IP3) receptor by IP3
which is generated by membrane depolarization. Emptying
of internal Ca21 stores signals ICRAN to open. Under physiological conditions Na1 is the primary current carrier; thus,
as described above, the increased Na1 permeability depolarizes the b-cell causing L-type Ca21 channels to open,
which provides Ca21 to help refill internal calcium stores.
Replenishment reduces the signal to and activity of ICRAN,
allowing the cell to repolarize and continue cycling. Since the
extent of repolarization will be determined by the K1 permeability and will determine whether bursting continues,
this is one potential control point for KATP channels. Coupling the control of ICRAN to emptying of internal Ca21 stores
is able to explain the effects of compounds such as maitotoxin
(109 –111), which activates ICRAN, and thapsigargin, an inhibitor of the endoplasmic reticulum Ca21 pump [sarcoplasmic or endoplasmic reticulum Ca-ATPase (SERCA)] (112,
113), on b-cell membrane potential and insulin release (105,
114, 115). How the filling level of the internal Ca21 store is
determined and the nature of the chemical signal that passes
to ICRAN remains to be elucidated.
6. Potassium channels. In addition to KATP channels, other K1
channels have been identified in pancreatic b-cells. This area
has been reviewed recently by Dukes and Philipson (87) and
need not be repeated beyond indicating that delayed rectifier
channels (Kv1.x), Ca21-activated K1 channels (maxi-K), an
a-adrenoreceptor-activated K1 channel, and G protein-gated
K1 channels (KIR3.1, 3.2 and 3.4) have all been identified in
b-cells either by electrical recording or molecular biological
methods. The Ca21-activated K1 channel was proposed to
play a role in bursting, but inhibitors of this channel have
little effect on electrical activity.
What role do KATP channels play in bursting? There is
general agreement that KATP channels are the predominant
conductance in a b-cell in low glucose, and that the initial
depolarization, which starts a train of bursts, results from an
increase in glucose metabolism, which reduces their activity.
The action of sulfonylureas is readily understood within this
context. Sulfonylureas close KATP channels, in effect mimicking the signal from glucose metabolism, thus depolarizing
the b-cell membrane. Titration of KATP channel activity with
tolbutamide will lead to bursting at intermediate glucose
concentrations (116, 117). The action of a KCO is also readily
understood. The application of diazoxide inhibits insulin
release by hyperpolarizing b-cells by opening more KATP
channels, thus shifting VM more toward EK. The question of
how KATP channels participate in the actual dynamics of
bursting is controversial. Using whole-cell recording, Smith
et al. (118) were unable to see changes in conductance in KATP
channels during bursting, indicating they are not involved in
the dynamics of bursting. Larsson et al. (119) have reported
107
the reverse: that the conductance of KATP channels oscillates
during slow bursting in single mouse b-cells and in small
clusters of b-cells. Rosario et al. (120) have demonstrated
bursting in b-cells, incubated with high extracellular Ca21,
when KATP channels were blocked by sulfonylureas. These
results indicate KATP channels are not essential for bursting
to proceed, but that residual channel activity may play a role
in normal repolarization. Finally, if KATP channels are stimulated by diazoxide during an episode of bursting, the increase in K1 permeability hyperpolarizes the b-cell membrane and thereby terminates bursting. These results have
led to the idea that there are both KATP channel-dependent
and independent mechanisms for controlling insulin secretion (see for example Refs. 121–123).
The physiological ligand(s) or mechanism(s) controlling
KATP channels remains somewhat controversial. In excised,
inside-out patches, the application of ATP results in channels
closing (Fig. 1), which led to the early suggestion that
changes in the intracellular concentration of ATP might regulate b-cell KATP channels (30, 124). However, the IC50 for
inhibition of channel activity was determined to be in the
10 –50 mm range, well below the millimolar concentrations
estimated for intracellular ATP levels (125–127); when millimolar concentrations of ATP were applied to the intracellular face of excised patches, KATP channel activity was
strongly inhibited. Subsequent studies showed that MgADP
could open KATP channels inhibited by ATP, implying that
the ADP/ATP ratio was the important variable (43, 44, 128 –
130). As described below, studies on channels reconstituted
from a mutant SUR1 identified in a PHHI patient and wildtype KIR6.2 are consistent with this idea and suggest further
that fluctuations in the intracellular ADP concentration that
occur as a result of glucose metabolism (126, 127, 131) are a
critical regulator in b-cells.
IV. KATP Channel Subunits
A. The KIR family of inwardly rectifying K1 channels
The observations (30, 44, 124, 132–134) that ATP-sensitive
K1 channels are both potassium selective and inwardly rectifying suggested they would have some relationship to the
inwardly rectifying potassium channel superfamily, the definition of which began with the cloning of three distinct
inwardly rectifying K1 channels: KIR1.1 (ROMK1), a weak
inward rectifier found in rat kidney (1), KIR2.1 (IRK1), a
strong inward rectifier cloned from a macrophage cell line
(135), and KIR3.1 (KGA, GIRK1), a G protein-gated, strong
inward rectifier isolated from rat heart (136, 137). The descriptive terms “weak” and “strong” in this context are indicators of the amount of current the channels can pass in the
outward direction when the membrane potential is greater
than EK; weak rectifiers conduct more current. These proteins
are all smaller than the voltage-gated K1 channels and, based
on hydropathy plots, were predicted to have two transmembrane domains (TMDs) or helices, termed M1 and M2. These
segments flanked a sequence with a high degree of similarity
to the P (pore) or H5 loop first identified in the voltage-gated
K1 channel (KV) family (138, 139) and shown to be part of the
potassium selectivity “filter.” Insight into how the pore is
108
AGUILAR-BRYAN AND BRYAN
probably formed has been provided by the determination of
the structure of the pore of the KcsA channel (34, 35) [1BL8
in the Protein Data Bank (140, 141)] which is depicted in Fig.
4. Although the KcsA channel is a pH-regulated, bacterial K1
channel, the general similarity between the pore regions of
the Kv channels, the inward rectifiers, and KcsA, all of which
share an architecture with two transmembrane helices flanking a GYG or GFG sequence responsible for the K1 selectivity, indicates their structures will be similar. As expected
from earlier studies, the M2, or inner pore helix, actually
forms the permeation pathway with the signature GYG sequence positioned to act as a selectivity filter. The structure
FIG. 4. Three-dimensional structure of the Streptomyces lividans K1
channel. The top view looks down into the conduction pathway
through the selectivity filter assumed to be composed of the backbone
carbonyl oxygen atoms of the gly-phe-gly motif found in other potassium channels. The side view conveys the overall structure, including
the vestibule below the selectivity filter and the convergence of the
four M2 helices, which form the conduction path at the cytoplasmic
face of the channel, to form a gate which can open and close (35). The
figure was generated using the coordinates, 1BL8, obtained from the
Protein Data Bank (140, 141) using RasMol (373). [Derived from D.
A. Doyle et al.: Science 280:69 –77, 1998 (34).]
Vol. 20, No. 2
and studies on the dynamics of KcsA suggest the actual gate
is formed by amino acids at the cytoplasmic ends of the M2
helices. Lateral or twisting motions of these helices could
control access or gating of internal K1 ions into the waterfilled vestibule immediately below the selectivity filter. The
availability of a general structural model, even one missing
the cytoplasmic domains that presumably regulate motion of
the M2 helices, should allow real progress toward understanding gating mechanisms.
Six subfamilies within the KIR family, KIR1.x through
KIR6.x, have now been identified based on their size, amino
acid similarities, and functional properties. We will employ
the “KIR” nomenclature originally proposed by Chandy and
Gutman (142) to identify these channels. The overall properties of the family have been summarized by Doupnik et al.
(143), while the mechanism of inward rectification has been
reviewed recently by Nichols and Lopatin (49).
1. Cloning of KIR6.x cDNAs. Using a fragment of KIR3.1
(GIRK1) cDNA as a probe, Inagaki et al. (144) cloned KIR6.1
(ruKATP-1) from a rat library. The tissue distribution of KIR6.1
was broad; hence the designation “u” KATP-1 for ubiquitous.
Expression of KIR6.1 in HEK293 cells suggested it formed a
K1 channel whose activity was blocked by 1 mm ATP, and
activated by diazoxide, but was not inhibited by sulfonylureas. Subsequent expression of KIR6.1 in other cell types has
failed to generate novel K1 channel activity with similar
properties (145, 146), and it is unclear whether the original
observation was due to coupling of KIR6.1 to an unidentified
SUR-like protein in HEK293 cells (21), to mischaracterization
of an endogenous channel, or to having a sufficiently high
level of expression to overcome a KIR retention signal. This
report further added to the confusion surrounding the cloning and identification of KATP channels since KIR1.1 (1),
KIR3.4 (147), and KIR6.1 (144) were all reported initially to be
ATP-sensitive potassium channels.
KIR6.1 was not detected in b-cell lines, and Inagaki et al.
(144) used the KIR6.1 cDNA to clone the cDNA for KIR6.2
(originally designated BIR for b-cell inward rectifier). Subsequent work showed that KIR6.2 mRNA was relatively
abundant in pancreatic islets and b-cell lines and was expressed in brain, heart, and skeletal muscle (see Ref. 21 for
a review). The KIR6.2 gene was found to contain no introns
and is therefore relatively easy to obtain using the PCR.
KIR6.1 and KIR6.2 have significant blocks of identical sequence. Seventy percent, 276 of the 390 residues in human
KIR6.2, are identical to those of KIR6.1 after conservative
alignment (Fig. 5). We have marked the proposed transmembrane helices and the P loop by comparison with the
KcsA structure and have marked the approximate positions
of the PHHI mutations described below. KIR6.1 has several
intriguing features, including a triple repeat at its C-terminal
end, S(M/I)RRNN.
B. Sulfonylurea receptors
1. Identification of a 140-kDa polypeptide as the high-affinity
sulfonylurea receptor (SUR1). A number of studies (148 –154)
described a high-affinity sulfonylurea receptor with KDs in
the low nanomolar range, in membranes isolated from
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
109
FIG. 5. Schematic representation of an
inward rectifier potassium channel.
The diagram is based on the Streptomyces lividans K1 channel described by
Doyle et al. (34) and on data described
in the text. M1 and M2 identify the two
TMDs; the region determining the K1
selectivity contains the gly-phe-gly motif. The asterisks mark the positions of
mutations in KIR6.2 that have been
identified in patients with PHHI. The
dotted circles identify segments of the
KATP channel referred to in the text.
The alignment compares the sequences
of human KIR6.1 with KIR6.2. Identical
residues are shaded. The M1 and M2
segments are boxed. Note the repeated
sequence in the C terminus of KIR6.1
marked by the bar.
b-cells, b-cell lines like the SV40 transformed HIT T15 hamster b-cell line (155) or the RINm5F rat insulinoma cell line,
the aTC-6 pancreatic a-cell line (134, 156), heart (157, 158),
and from brain (148, 149, 159, 160). Some effort was made to
solubilize and purify these proteins, and Bernardi et al. (161)
reported the purification of a 150-kDa high-affinity sulfonylurea receptor from brain.
A significant advance in the biochemistry of these receptors came with the realization that [3H]glibenclamide could
function as a photoaffinity probe and would label a polypeptide with an apparent molecular mass of 140 kDa in rat b-cell
tumor membranes (151). A 125I-labeled derivative of glibenclamide, [125I]iodoglibenclamide (152) (Fig. 2), identified a
similar protein in membranes isolated from HIT T15 cells and
had the advantage of higher specific activity, which made
studies of the receptor and its purification feasible. Nelson et
al. (154) investigated the photolabeling of membrane proteins by this reagent and showed that the 140-kDa protein
accounted for most of its high-affinity binding activity, KD ;
5 nm, in HIT T15 cell membranes. Rajan et al. (134) and
Ronner et al. (156) used [125I]iodoglibenclamide and [3H]glibenclamide, respectively, to identify the high-affinity receptor in aTC-6 cells, a glucagon-secreting cell line. Both groups
characterized the sulfonylurea-sensitive KATP channels in
these cells by rubidium efflux and single-channel recording
and concluded they were equivalent to the b-cell channel.
110
AGUILAR-BRYAN AND BRYAN
Schwanstecher et al. (162) synthesized a 4-azidosalicyloyl
analog of glibenclamide, 125I-iodo-azidoglibenclamide (Fig.
2), and demonstrated that this derivative would photolabel
a 38- to 40-kDa protein in addition to SUR1 (163). Competition binding experiments indicated that both the novel
38-to 40-kDa species and SUR1 were labeled with 125I-iodoazidoglibenclamide with the same apparent KD. As described below, the 38- to 40-kDa species is KIR6.2.
2. SUR1 is differentially glycosylated. Labeling of aTC-6 cells
with [125I]iodoglibenclamide showed two labeled species
with estimated molecular masses of 140 kDa and 150 –170
kDa, the latter being a diffuse band (134). The higher molecular mass receptor accounted for approximately half of the
high-affinity labeling in aTC-6 cells. Competition binding
studies demonstrated that the two receptors had the same
affinity for iodoglibenclamide, KD ; 3.5 nm. Rajan et al. (134)
suggested the 150- to 170-kDa receptor resulted from differential glycosylation, but at the time the possibility of an
additional gene product could not be eliminated. Nelson et
al. (164) demonstrated that the 140 and 150- to 170-kDa receptors were present in several other cell types. In RINm5F
and aTC-6 cells the two glycosylated species are present in
about equal amounts, while in HIT cells the predominant
species was the 140-kDa receptor, with the 150- to 170-kDa
species accounting for less than 10% of the high-affinity
binding activity. Ozanne et al. (165) used [3H]glibenclamide
to label receptors in rat pancreatic islets, insulinomas, and
CRI-G1 cells and were able to identify both forms of the
receptor. The 150- to 170-kDa species was the major form in
islets and insulinomas, while the 140-kDa species was predominant in CRI-G1 cells.
Biochemical studies indicate the mass difference between
the two forms of the receptor is the result of differential
glycosylation (164). Photolabeling of membranes isolated
from RINm5F cells grown in tunicamycin, which blocks Nlinked glycosylation, produced a single photolabeled band
with an estimated mass of 137 kDa, distinct from the 140-kDa
species. A 137-kDa species was also generated by digestion
with endoglycosidases. The 140 and 150- to 170-kDa receptors could be separated using concanavalin A and wheat
germ agglutinin, the 140 kDa receptors binding to concanavalin A-agarose, while the 150- to 170-kDa receptors
bound to wheat germ agglutinin-agarose. The results indicate that the 140-kDa receptors are “core” glycosylated with
mannose-containing glycosyl groups that bind to concanavalin A, while the 150- to 170-kDa receptors are “complex”
glycosylated with terminal sialic acid residues that bind to
wheat germ agglutinin (164). Since core glycosylation takes
place in the endoplasmic reticulum after protein synthesis,
while trimming of the mannose chains and addition of sialic
acid groups takes place in the medial Golgi, the data indicate
that the 150- to 170-kDa species is the “mature” form of the
receptor and is either in, or in transit to, the plasma membrane. This is consistent with the results of Ozanne et al. (165)
showing that the “immature” 140-kDa receptors in insulinomas were found in internal membranes, whereas the 150to 170-kDa receptors were in granules and plasma membranes. The results raise questions about how glycosylation
of SUR1 is regulated, whether the core glycosylated receptors
Vol. 20, No. 2
on internal membranes might have a function apart from
KATP channels, and whether the subunits can traffic independently to the plasma membrane. For example, Eliasson et
al. (166) have suggested that sulfonylureas can stimulate
insulin secretion independently of their blockage of KATP
channels, perhaps acting through SUR1 in granule membranes.
The available sequence and site-directed mutagenesis has
confirmed there are two N-glycosylation sites in SUR1 at
residues 10 and 1050. The N10Q or N1050Q mutations partially eliminate glycosylation, while the double mutant is not
glycosylated. All three of the glycosylation mutants generate
channels when expressed with KIR6.2, although the number
of channels, estimated by 86Rb1 efflux, is reduced (J. P. Clement IV and J. Bryan, unpublished data). Interestingly, the
glycosylation patterns of SUR1 appear to differ between endocrine and neuronal tissues as well as a- and b-cell lines
(164). The SUR1 in brain, for example, appears to be only the
highly glycosylated form (162).
3. Linking the sulfonylurea receptor with KATP channel activity.
The link between sulfonylurea receptors and KATP channels
was the subject of some early controversy. The simple view
that receptor and channel might be a single entity (167) was
questioned by Khan et al. (168), who reported the dissociation
of KATP channel activity and sulfonylurea receptors in a rat
insulin-secreting clonal cell line, CRI-D11. Aguilar-Bryan et
al. (169) produced evidence that receptor and channel activity were lost in parallel during serial passage of HIT cells at
approximately the same passage number at which glucose
sensitivity for insulin release was lost (170). These studies
have not been reexamined using the molecular probes now
available, and it is unclear, for example, whether the CRI-D11
cells actually have active KIR6.2 channels, without receptors,
or whether high passage HIT cells have lost KIR6.2 in addition to SUR1.
4. Receptor purification. Hamster SUR1, prelabeled with
[125I]iodoglibenclamide, was purified from HIT cell membranes using chromatographic and electrophoretic methods
(22, 164). The general strategy was to prelabel the receptor
with [125I]glibenclamide, then purify the iodinated protein.
The end point was obtaining sufficient pure receptor for
N-terminal peptide sequencing. The purification has been
described in detail (164, 171) and will not be discussed further. The availability of cloned cDNAs has allowed introduction of specific markers, for example hexa-histidine tags,
to facilitate purification of expressed recombinant receptors
(146, 172) and KIR6.x subunits.
5. Cloning strategy. The N-terminal protein sequence was
used to design oligonucleotide primers for amplification of
a small segment of the SUR1 cDNA by the PCR (22, 164). The
resulting cDNA fragment was used to screen hamster, rat,
mouse, and human pancreatic cDNA libraries to obtain fulllength clones that encode homologous proteins with molecular masses of approximately 176 kDa (1582 amino acids for
the rodent receptors and 1581 or 1582 amino acids for the
human receptor). The cDNA and gene sequences for SUR1
and SUR2 are readily available through the National Library
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
of Medicine (http://www.ncbi.nlm.nih.gov/Web/Search/
index.html).
6. SUR1 is a member of the ATP-binding cassette superfamily. The
rodent and human SUR1 sequences are quite similar with
absolute identities of .90%. Blast searches (173) of the Genbank database show similarities between SUR’s and a large
number of ABC proteins including multidrug resistance proteins (174), multidrug resistance associated proteins (MRPs)
(175), CFTRs (176), canalicular multispecific organic anion
transporter (177, 178), and the cadmium-binding protein
YCF1 (179). Dendrograms and the overall topology of SURs
indicate they can be placed within the MRP subfamily of
ABC proteins (180) with two potential nucleotide-binding
folds (NBFs), multiple TMDs, and a very hydrophobic segment at their N termini. The greatest sequence similarities are
in the NBFs that have the Walker A (-GlyXXGlyXGlyLysSer/
Thr-, where X is any amino acid) and B (-YYYYAsp-, where
Y is a hydrophobic amino acid) consensus motifs (181) and
a conserved -LeuSerGlyGlyGln- sequence in the segment
linking the two Walker motifs. NBF1 matches this motif
exactly, while NBF2 has a somewhat degenerate sequence,
-PheSerGlnGlyGln- (SUR1), -PheSerValGlyGln- (SUR2).
Manavalan et al. (182) have noted the similarity of the NBF
sequence between CFTR and G proteins. X-ray crystal structures for the three major classes of GTPases, the small p21raslike proteins, the heterotrimeric G proteins, and EF-Tu, indicate that the Walker A motif interacts with Mg21 and with
the oxygen atoms of the a- and b- phosphates (see Ref. 183
for a review). The Gln in the -LeuSerGlyGlyGln- linker sequence has been proposed to act as a general base during
nucleotide hydrolysis, but this has been difficult to establish
(184). The lack of a structure for the NBFs of the ABC cassette
proteins has hindered progress. The recent preliminary report on the determination of the structure of an NBF from the
ribose transporter by Armstrong et al. (185) should allow
major advances in this area. A major question with regard to
the sulfonylurea receptors is whether they can hydrolyze
ATP, and, if so, how this hydrolysis is coupled to the regulation of channel activity. The purification and ATPase activity of the closely related MRP1 protein have been described (186). Interestingly, the ATPase activity of this ABC
protein is stimulated several fold by low concentrations of
nucleoside diphosphates (187).
7. Proposed membrane topology for SUR1. Aguilar-Bryan et al.
(22) predicted a topology for SUR1 based on plots of hydrophobicity and hydrophobic moments using the analysis of
Eisenberg and colleagues (188). The constraints on the proposed topology were that the N terminus, glycosylated at
N10 (22), was extracellular and the two NBFs were intracellular. The resulting model consisted of nine TMDs followed
by a nucleotide-binding fold, NBF1, a second transmembrane region with four spanning helices followed by a second
nucleotide-binding fold, NBF2 (i.e., 9 TMD–NBF1– 4 TMD–
NBF2). This topology resembled the model originally put
forward for MRP1 (175). Tusnády et al. (180) have revised the
model of the MRP subfamily using multisequence alignments of recently cloned ABC proteins. Their model, 11
TMD–NBF1– 6 TMD–NBF2, emphasizes the existence within
111
the MRP subfamily proteins of an multidrug resistance protein-like core consisting of 6 TMD–NBF1– 6 TMD–NBF2. This
core is preceded, in the MRP subfamily, by a highly hydrophobic N-terminal extension with 5 or more TMDs. Work is
needed to establish the topology experimentally and to determine what function the additional TMDs might have. We
have discussed the topology of the receptors in more detail
elsewhere (19).
8. Expression of SUR1 cDNAs. Transfection of SUR1 cDNAs
into COSm6 cells lacking endogenous sulfonylurea receptors
led to the expression of proteins that could be photolabeled
with [125I]glibenclamide and had an apparent mobility on
SDS polyacrylamide gels equivalent to the native receptor
(22). The discrepancy in mass between 176 kDa predicted
from the cDNA sequence and 140 kDa estimated by gel
electrophoresis is artifactual. Purification of receptors tagged
with hexa-histidine at either the N- or C termini yield the
same size protein. Membranes isolated from COS cells expressing the recombinant hamster and rat receptors contained high-affinity [125I]glibenclamide-binding activity with
KDs of 10 and 2 nm, respectively (22). These values were in
good agreement with a value of 5 nm determined for the
native hamster receptor in isolated HIT cell membranes (19,
154, 169). Efforts to show that COS cells expressing SUR1
alone had novel K1 channel activity were unsuccessful, suggesting other components were required to form a KATP
channel. A puzzling result, from these early studies, was that
only the 140-kDa core glycosylated protein was expressed,
while the 150- to 170-kDa receptors were not detected.
V. Reconstitution of KATP Channel Activity from
SUR1 and KIR6.2
Coexpression of SUR1 and KIR6.2 in COS cells and in
Xenopus oocytes produced a novel K1 conductance that was
characterized by 86Rb1 efflux and single-channel recording
(26). Metabolic poisoning with 2-deoxyglucose and oligomycin revealed a novel 86Rb1 efflux pathway not detected in
cells transfected with b-galactosidase, or with either SUR1 or
KIR6.2 alone. Efflux was blocked by tolbutamide and glibenclamide and was activated by diazoxide, but was poorly
activated by cromakalim or pinacidil (23). Single-channel
recordings characterized a moderate, inwardly rectifying potassium-selective channel with a conductance of about 65 pS
in 140 mm KCl at 260 mV (26). The extent of inward rectification was dependent upon the presence of Mg21 or polyamines on the intracellular side of the channel (19, 146).
Channel activity was half-maximally inhibited by ATP at
approximately 10 mm (26). Later measurements indicated the
recombinant channels were activated by MgADP in the presence of ATP (189, 190). Table 2 provides a comparison of the
properties of recombinant and native KATP channels. The
recombinant channels appear to mimic all of the basic electrophysiological and pharmacological properties of native
b-cell ATP-sensitive potassium channels. These results have
been confirmed by studies in oocytes (145) and HEK cells
(191). Additional work will be required to determine whether
and by what means the recombinant KATP channels are coupled to the receptors for the various peptides, such as gala-
112
AGUILAR-BRYAN AND BRYAN
Vol. 20, No. 2
TABLE 2. Comparison of native and recombinant b-cell/neuronal type KATP channels (SUR1/KIR6.2)
ATP (mM)
Conductance (pS)a
Native
b
50 – 65
Recombinant
Nativeb
69 –73
10 –20
Recombinant
8 –10
Ref.
(26)
Potassium channel openers
Glibenclamide
IC50
,10 nM
KD ; 7 nMd
a
IC50
,10 nM
KD ; 9 nMd
Diaz/Pin/Cromc
Diaz/Pin/Cromc
20 –100 mM/high/high
65 mM EC50/high/high
(26)
(19)
1
Quasisymmetrical high K (135–145 mM) at negative potentials.
Values for native channels taken from Ref. 375.
Diaz, Diazoxide; Pin, pinacidil, Crom, cromakalim (see Fig. 2 for structures).
d
Determined for 125I azido iodoglibenclamide in HIT cell membranes and membranes isolated from COSm6 cells transfected with hamster
SUR1 (19).
b
c
nin, reported to affect their activity (Ref. 192 but see Ref. 193),
somatostatin (194), GLP-1 (195), CGRP (196), and leptin
(197–201).
The recombinant SUR1/KIR6.2 channels displayed the
characteristic bursting pattern observed in native pancreatic
b-cell channels. A single channel might, for example, be in a
closed state, and then enter a bursting state characterized by
flickering between open and closed states. As we have reviewed in detail elsewhere (19), the expression of the SUR
subtypes with KIR6.2 indicates that the inward rectifier determines the mean open time of the flickering, the fast closures, within a burst, while the SUR subtype determines the
frequency of transitions between bursts.
A. The question of “promiscuous coupling” of SUR1 with
other inward rectifiers
The full range of inward rectifiers that SUR1 will partner
with has not been established. Ämmälä et al. (202) showed
that KIR6.1 and SUR1 would form KATP channels, but these
were not characterized. These authors also argued for greater
promiscuity and pairing of the receptor with KIR1.1
(ROMK1) and KIR3.4 (rcKATP-1/CIR/GIRK4). Clement et al.
(146) have confirmed that KIR6.1 and SUR1 produce metabolically activated KATP channels, while coexpressing SUR1,
and KIR1.1 (ROMK1) or KIR3.4 (CIR, rcKATP) failed to show
association or KATP channel activity. Using Xenopus oocytes,
Gribble et al. (145) were unable to reproduce the original
observations on coupling of SUR1 to ROMK1 and showed
that KIR2.1 also failed to couple to SUR1. We infer that,
although all the potential inward rectifier candidate subunits
have not been screened, the degree of promiscuity involved
in KATP channel formation will not be large. Wellman et al.
(203) have reached a similar conclusion using coronary arterial myocytes. Interestingly, the reports on the SUR1/
KIR6.1 channel have all been with intact cells, since observations on excised patches have been hampered by rapid
channel rundown (145).
VI. KATP Channel Structure
The reconstitution experiments demonstrated that both
SUR and KIR subunits were required to form KATP channels,
but raised a number of questions about the interactions,
stoichiometry, and functional role(s) of each subunit within
a channel and about the trafficking and activity of individual
subunits. It was not at all clear, for example, why KIR6.2 was
silent in the absence of SUR. Some of these questions have
been addressed, and it is becoming clear that essentially all
of the kinetic, pharmacological, and regulatory properties of
KATP channels including their sensitivity to inhibitory ATP
and trafficking to the plasma membrane are highly dependent on interactions between SUR and KIR subunits.
A. KIR6.2 forms the pore of a KATP channel
An early question was which subunit formed the pore or
permeation pathway of the channel and whether a part of
SUR was involved. Two lines of evidence have been developed that show that KIR6.2 subunits are sufficient to form the
permeation pathway. Clement et al. (146) and Shyng et al.
(204) showed that substitution of an aspartate for the asparagine at position 160, N160D, produced strongly rectifying
channels; whereas asparagine, in the wild-type channel produced weaker rectification. The result did not eliminate the
possibility that SUR1 could contribute directly to the permeation pathway, but did indicate that the M2 segments of
KIR6.2, like those of other inward rectifiers, were part of the
permeation pathway. The discovery that truncated KIR6.2
subunits, missing 16 –35 amino acids from their C-terminal
ends, can assemble a K1 channel provided a direct demonstration that SUR1 was not required to form the pore (51).
These findings suggested that the underlying architecture of
the pore of KATP channels would resemble that of the other
members of the KIR channel family.
B. SUR1 and KIR6.x are physically associated
The requirement for coexpression of both SUR and KIR to
make an ATP-sensitive K1 channel implied these subunits
were associated, but other alternatives were possible.
Al-Awqati (205), for example, proposed that SUR might regulate a physically separate potassium channel by releasing
ATP, as CFTR has been proposed to regulate the outwardly
rectifying chloride channel (206, 207). The first direct evidence that SUR1 was associated with a second subunit came
from the work of Schwanstecher, Panten, and colleagues (162,
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
163) as mentioned above. These authors developed an azido
derivative of glibenclamide, [125I]azidoglibenclamide (Fig.
2), and demonstrated that in addition to labeling SUR1 in
brain membranes, an additional 38-kDa protein displayed
high-affinity labeling. The identification of the 38-kDa component as KIR6.2 became clear from [125I]azidoglibenclamide
labeling studies on COS cells expressing the cloned subunits
(146). In these experiments SUR1 photolabeled independently of KIR6.2, but KIR6.2 did not bind glibenclamide directly and was only labeled when coexpressed with SUR1.
These findings implied that 125I-iodo-azidoglibenclamide
was bound to SUR1 and that KIR6.2 must be in close proximity, since the lifetime of the photo-activated species is brief.
C. Coexpression with KIR6.2 affects the maturation of SUR1
The finding that coexpression of SUR1 and KIR6.2 affected
the maturation or glycosylation state of the receptor provided additional support for subunit association, provided a
means to demonstrate complex formation directly, and was
the first indication that SUR-KIR assembly was needed for
SUR to traffick to the plasma membrane. Expression of SUR1
alone generated the immature 140-kDa core glycosylated
receptor, while coexpression with KIR6.2 also generated mature complex glycosylated receptors (146). Chromatography
on wheat germ agglutinin was used to show that KIR6.2 was
preferentially associated with mature SUR1 and was assembled into a stable multimer that could be isolated after solubilization in digitonin (146). These findings suggested a
plausible assembly pathway in which the immature receptor
associates with KIR6.2 in the endoplasmic reticulum and maturation to the complex glycosylated form occurs in the medial Golgi as the assembled channel moves to the plasma
membrane. These results agree with the observations of
Ozanne et al. (165) that the 150-kDa species of SUR1 is in
plasma membranes. It was not clear whether KIR6.2 could
traffic independently to the plasma membrane, but the result
indicated SUR1 did not move efficiently through the medial
Golgi in the absence of KIR. The later observation that Cterminally truncated KIR6.2 subunits could generate K1
channels raised the question of whether KIR subunits could
traffic independently to the plasma membrane (51).
113
proteins with defined stoichiometries were engineered to
determine whether a 1:1 SUR/KIR stoichiometry was both
sufficient and necessary for the formation of functional KATP
channels (146). Expression of SUR1;KIR6.2, a fusion of the N
terminus of KIR6.2 with the C terminus of SUR1 through a
6-glycine linker produced potassium-selective channels that
were activated by metabolic poisoning and inhibited by sulfonylureas. Single-channel recording showed that the
SUR1;KIR6.2 channels were weak inward rectifiers, which
had the same conductance as unfused channels, although the
sensitivity to both ATP and glibenclamide were reduced.
Sucrose gradient analysis showed that SUR1;KIR6.2, like the
wild-type subunits, produced a 950-kDa multimer. The results showed that a 1:1 stoichiometry was sufficient for formation of KATP channels.
Expression of a “triple” fusion construct, SUR1;(KIR6.2)2,
with a defined 1:2 stoichiometry, did not generate ATPsensitive potassium currents. However, the triple fusion
could be rescued by coexpression with monomeric SUR1,
implying that the additional receptor(s) reestablished the
required 1:1 stoichiometry. The SUR1 1 SUR1;(KIR6.2)2
channels had the same conductance properties as unfused
channels, but were less sensitive to both ATP and sulfonylureas. The results showed that a 1:1 stoichiometry was necessary for assembly of KATP channels and, together with the
mass estimates, argued for a tetrameric organization similar
to other members of the KIR channel family.
F. Other KIR channels are tetramers
The finding that maturation of SUR1 was coupled with
expression of KIR6.2 and that these subunits assembled a
stable complex argued that the multimer might be the channel itself. The mass of the stable multimer, estimated by
sucrose gradient sedimentation, was approximately 950 kDa,
consistent with a large channel composed of four complex
glycosylated SUR1 subunits plus four KIR6.2 subunits (an
expected protein mass of ; 885 kDa 5 4 3 176 kDa 1 4 3
45 kDa, plus an unspecified carbohydrate mass) (146).
Several reports (208, 209) indicate that the KIR channels,
like the KV channels (138, 139, 210 –212), are tetrameric.
Glowatzki et al. (208) coexpressed weak (KIR1.1, ROMK1) and
strong (KIR4.1, BIR10) inward rectifiers and observed intermediate rectification properties expected from the assembly
of heterologous channels. The differential sensitivity of these
channels to voltage-dependent block by spermine was used
to study the distribution of channel types assembled in cotransfected cells. Five types were identified that approximated the binomial distribution expected for a pore composed of four subunits. The polyamine binding site, in the
pore, is therefore assumed to be assembled from amino acids
contributed by four M2 segments. Yang et al. (209) studied
the subunit stoichiometry of the strongly rectifying KIR2.1
channel (IRK1) by fusing subunits to form multimers, a strategy that had been used earlier by Liman et al. (211) on KV1.1
(RCK1) channels. The trimeric and tetrameric constructs,
(KIR2.1)3 and (KIR2.1)4, both produced channels. Coexpression of the strongly rectifying wild-type trimers with a
weakly rectifying mutant monomer produced a heterologous channel with intermediate rectification and weaker affinity for spermidine, while coexpression of the mutant
monomer with the wild-type tetramer did not affect rectification or polyamine binding. The results were consistent
with a tetrameric channel.
E. A 1:1 stoichiometry of SUR to KIR is both necessary and
sufficient to make KATP channels
G. The stoichiometry of active b-cell KATP channels is
(SUR1/KIR6.2)4
The simplest model for the assembly of SUR1 with KIR6.2
was a 1:1 association. Therefore, cDNAs encoding fusion
A strategy based on the formation of heterologous channels was used to show that functional KATP channels are
D. Complex glycosylated SUR1 and KIR6.2 assemble a
large multimer
114
AGUILAR-BRYAN AND BRYAN
tetramers (146). cDNAs were engineered to express triplefusion proteins, SUR1;(KIR6.2)2 and SUR1;(KIR6.2N160D)2,
which generated moderate or strongly rectifying channels,
respectively, when rescued by SUR1 monomers. When these
constructs were coexpressed, three classes of channels could
be identified: the parental weak and strong rectifiers plus a
third species with intermediate rectification. The presence of
a species with intermediate rectification implies the pore is
constructed from two triple-fusion proteins, one with wildtype KIR6.2, the other with KIR6.2N160D. Shyng and Nichols
(213) used SUR1;KIR6.2 constructs engineered with and
without the N160D mutation to reach the same conclusion,
and similar results with triple-fusion constructs were obtained
by Inagaki et al. (214).
Clement et al. (146) specifically proposed that active channels are composed of four complex glycosylated receptors
interacting with four KIR6.x subunits, which form the K1
selective pore (SUR1/KIR6.2)4. The observation that the fusion channels were active suggested that the N terminus of
KIR6.2 may be near the C terminus of SUR1 in the b-cell
channel, where it may serve a role in the transduction of
nucleotide-induced conformational changes from the receptor to the pore. This model, illustrated in Fig. 6, further
suggests the possibility of extensive cooperative interactions
between the eight nucleotide-binding domains, four sulfo-
FIG. 6. Topology and assembly of KATP
channels. The topologies of SUR, based
on the model of Tusnády et al. (180), and
KIR6.x are illustrated. The wild-type
channel is assembled from SUR1 and
KIR6.2 subunits as an octameric complex as described by Clement et al.
(146). Truncation of the C-terminal end
of KIR6.2 allows trafficking of the inward rectifier to the cell surface to form
homomeric K1 channels (KIR6.2DC)4
with abnormal properties (51). Truncated KIR subunits can also assemble
with SUR. These channels are compared in Fig. 8.
Vol. 20, No. 2
nylurea-binding sites, and four potassium channel opener
(diazoxide)-binding sites.
VII. Regulation of KATP Channel Activity
The sensitivity of KATP channels to cytosolic nucleotide
concentrations clearly implies nucleotides are involved in the
regulation of channel activity. How this regulation takes
place, however, is far from clear. Electrical recordings from
intact b-cells perfused with increasing concentrations of glucose show a decrease in the opening of KATP channels (44,
215) that coincides with increasing insulin release. These
experiments illustrate the coupling between metabolism and
electrical activity and agree with the idea that the rate of
glucose metabolism, initiated by glucokinase (GK), is regulating openings of KATP channels in b-cells (see Refs. 216 and
217 for reviews). The results do not specify how metabolism
is coupled to channel activity. On the other hand, recordings
from excised patches show KATP channels respond to
changes in ATP and ADP, suggesting that coupling could
occur through nucleotide fluctuations. The present problem
is understanding how the (SUR1/KIR6.2)4 complex interacts
with nucleotides to regulate K1 flux and whether there are
sufficient changes in the ATP/ADP ratio to control KATP
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
channel activity. ATP and ADP have three main effects on
KATP channel activity that offer potential control points: refreshment, inhibition, and stimulation (reviewed in Ref. 42).
1. Refreshment. When a patch containing KATP channels is
excised into an ATP-free solution, channels open, as was first
described by Noma for cardiac myocytes (31) and by Cook
and Hales (30) and Trube and Hescheler (32) for pancreatic
b-cells. These channels lose their activity or “rundown” as a
function of time. Activity can be restored, or “refreshed,” by
brief application of MgATP at millimolar concentrations.
Refreshment can be thought of as switching the channel from
a nonoperational to an operational state. The transition to an
operational state requires hydrolysable nucleotides and is
not supported by nonhydrolysable ATP analogs or by hydrolysable nucleotides in the absence of Mg21 or Mn21. Several possible explanations have been put forward to explain
the biochemical basis for this switching, including phosphorylation/dephosphorylation reactions (see Refs. 218 and 219
but also Refs. 220 –223 for another view), uncoupling of KATP
channels from the actin cytoskeleton (223), and hydrolysis of
anionic phospholipids, which are proposed to stabilize the
channel in the operational state (224, 225). It is unclear
whether one or all of these mechanisms is a critical factor in
rundown and refreshment of channels in excised patches.
Perhaps more importantly, it is not certain whether there is
a physiological analog of rundown in intact cells where the
level of MgATP is in the millimolar range. One could propose
a coupling between metabolism and, for example, a phosphorylation/dephosphorylation cycle that controls KATP
channel openings. This idea has been explored by Ribalet et
al. (226) but has not received general support.
2. Inhibition. The application of ATP, with or without Mg21,
to the intracellular face of an excised patch inhibits KATP
channels. The IC50 or Ki values for this inhibition are quite
low, estimated at 5 to 10 mm for the pancreatic b-cell channel
and for reconstituted SUR1/KIR6.2 channels (26), and in the
range of 8 to .500 mm for the native cardiac channel and from
approximately 20 (28) to 100 mm or more (23, 29) for reconstituted SUR2A/KIR6.2 channels. These values predict approximately 99% inhibition of the b-cell channel, assuming
a cytosolic [ATP]i level near 1 mm. This exquisite sensitivity
to ATP has been a source of confusion when coupled with the
idea that ATP is the regulator of channel activity. It is worth
noting that several studies have suggested, and continue to
suggest, that fluctuations in [ATP]i regulate KATP channels
either through large changes in the concentration of ATP (40)
or through compartmentalization models that propose significant local changes in [ATP] (see for example Ref. 227).
Various mechanisms have been proposed to explain the
discrepancy and render KATP channels less sensitive to ATP.
For example, the inhibitory effect of MgATP has been reported to be weaker than ATP42, for the b-cell channel,
although Findlay (41) has reported the reverse for the cardiac
channel. An early proposal by Ashcroft and Kakei (40), therefore, suggested the idea that a significant fraction of the total
ATP would be complexed with Mg21 at the intracellular free
Mg21 concentration (0.5–1 mm) (see Refs. 228 and 229 for
reviews); therefore, the concentration of ATP42 would be
115
closer to the IC50 determined for KATP channels, thus allowing variations in ATP to regulate channel activity directly.
Our own studies on the Mg21 dependence of the ATP inhibition of KATP channel activity have failed to show significant
effects. In a different vein, Cook et al. (230) pointed out, in
their spare-channel hypothesis, that the high sensitivity of
KATP channels to ATP is required to inhibit the large number
of channels (assumed in the model to be 10,000 per b-cell).
Thus while approximately 99% of the channels were inhibited, the remaining channels were sufficient to maintain the
resting membrane potential in b-cells, which are small and
have a high membrane impedance. According to the sparechannel hypothesis, the exquisite sensitivity of the b-cell
KATP channel to ATP is a necessity rather than a problem.
Studies on cells expressing a mutant of SUR1, identified in
a patient with familial hyperinsulinism, argue against ATP
as the physiological regulator of KATP channels. This mutation, a gly3arg at position 1479 in NBF2 (189), uncouples the
inhibition of KATP channels by ATP from their activation by
MgADP and thus provides a tool to distinguish between
regulation by changes in ATP vs. ADP. Coexpression of
SUR1G1479R with wild-type KIR6.2 produces K1 channels that
are inhibited by ATP42, or MgATP, in excised patches with
nearly the same IC50 as native or wild-type reconstituted
channels, but are not activated by MgADP (189). The
SUR1G1479R/KIR6.2 channels are not activated by metabolic
inhibition in which [ATP]i is reduced and [ADP]i increases,
conditions that strongly activate wild-type channels. Inhibition by 2-deoxyglucose and oligomycin in the absence of
glucose cannot lower the [ATP]i sufficiently to activate these
KATP channels with an altered response to MgADP. Similarly, in PHHI b-cells the G1479R channels are blocked by
ATP and apparently fail to respond to fluctuations in ADP.
The result implies that fluctuations of [ATP]i alone are not
sufficient to trigger insulin secretion in these patients and
indicate ADP is a critical factor.
3. Stimulation. While ATP inhibits channel activity, the application of MgADP was shown to stimulate channel openings in the absence of added ATP and could activate channels
inhibited by ATP (43, 44). The presence of Mg21 is critical:
ADP and other nucleoside diphosphates, in the absence of
Mg21, inhibit KATP channel activity. These observations, on
excised patches, led to the idea that the ADP/ATP ratio was
critical for regulation of KATP channel activity (43, 44). In
b-cells, for example, increased glucose metabolism was proposed to reduce [ADP]i and increase [ATP]i (see, for example,
Ref. 216 for a recent review). The expectation was that the
decrease in [ADP]i would reduce channel activity (and any
increase in [ATP]i would do the same) and thus lead to
membrane depolarization, activation of voltage-gated Ca21
channels, increased [Ca21]i, and initiation of the exocytosis of
insulin. These studies pointed to the importance of ADP as
a potential regulator of KATP channels in b-cells, while similar studies have shown the importance of ADP for regulation of cardiac KATP channels (231, 232).
What is the evidence that the ADP/ATP ratio changes?
Direct measurements of the ADP/ATP ratio under different
metabolic conditions are difficult to make, but several reports
in b-cells (126, 127, 216, 233) and cardiomyocytes (234) are in
116
AGUILAR-BRYAN AND BRYAN
accord with this general model. The specific experiments
support the idea that [ATP]i increases, while [ADP]i is reduced during periods of increased glucose metabolism. The
magnitude of the measured increase in the ATP/ADP ratio
depends on the method of assay. Ghosh et al. (126) reported
an increase, measured by microchemical methods, but questioned its statistical significance. Detimary et al. (235) used
cultured islets to show that degranulation of b-cells, by preincubation in glucose, reduced the level of a nucleotide compartment, presumably insulin granules, with an ATP/ADP
ratio near 1. Cultured islets showed an increase in the ATP/
ADP ratio of 2.4 to .8 when incubated with 2 and 20 mm
glucose, respectively. Detimary et al. (127) also developed a
lysed cell protocol, which distinguished a diffusible cytosolic
pool of nucleotides from a nondiffusible pool located within
the insulin-containing granules. Incubation of islets in 20 mm
glucose did not affect the nondiffusible pool but caused a
change in the ATP/ADP ratio from 3.8 to 12.1, over a range
of 2–20 mm glucose, in the diffusible pool. Matschinsky and
colleagues used 31P-nuclear magnetic resonance techniques
on bTC3 and bHC9 b-cell lines and estimated that the concentration of free ADP dropped from approximately 35 mm
to 20 mm when cells were perfused with a high concentration,
25 mm, of glucose (216).
Cook and colleagues (128) developed a semiquantitative
model to explain the regulation of KATP channel activity in
b-cells. In their model, KATP channels were tonically inhibited by [ATP]i at concentrations estimated to be in the millimolar range, well above the values (10 mm) required to
half-maximally inhibit activity of the b-cell channel. In resting or fasting b-cells in a low glucose milieu, inhibition by
ATP is relieved by elevated MgADP, which opens a small,
but sufficient, number of KATP channels to hold VM near 260
mV, below the threshold for activation of voltage-gated Ca21
channels. Increased glucose metabolism causes [ADP]i to fall;
thus KATP channels close and VM shifts to more positive
values, which triggers insulin release through activation of
Ca21 channels. Cooperativity built into the model increases
the steepness of the response curve, thus reducing the magnitude of the change in [ADP]i required for regulation. This
model relies heavily on the spare channel hypothesis (230)
and the idea that maintaining the resting membrane potential of a b-cell requires openings of only a small number of
K1 channels.
The general wisdom is that b-cells have several thousand
KATP channels, but the evidence in support of this figure is
somewhat difficult to track down. Rorsman and Trube (124)
estimated an average of 500 channels per mouse b-cell from
whole-cell recordings during dialysis with ATP-free solution. Ohno-Shosaku et al. (236) estimated the channel density
at 1.5 per mm2 and calculated an average of 750 channels per
b-cell based on a spherical cell of 13 mm diameter. Misler et
al. (44) reported a greater density, 8 channels per mm2, although it is not clear whether this was an average value. The
number of sulfonylurea receptors in RINm5f cells was estimated from [3H]glibenclamide binding studies at approximately 5120 6 500 per b-cell (237). This would yield a maximum of about 1250 KATP channels per b-cell, assuming all
the receptors are assembled as tetramers and are surface
accessible.
Vol. 20, No. 2
It is interesting to note that, during the nearly 20 yr KATP
channels have been studied, there are numerous published
examples of ATP inhibition curves, but the number of studies
aimed at providing a quantitative basis for understanding
stimulation by MgADP are remarkably few. It is essentially
impossible to draw quantitative conclusions about this
model, although it appears qualitatively correct. The experiments with the SUR1G1479R channels, for example, provide
strong support, and other SUR1 mutations produce a similar
phenotype. Mutations in the conserved lysine residues in the
Walker A motifs in both NBFs of SUR1 have been made
(lys7193ala and lys13843met) and shown to result in loss of
activation by ADP (190, 238) while maintaining sensitivity to
ATP. It seems reasonable to conclude that, in the b-cell, ATP
acts on KATP channels to maintain an operational state, i.e.,
minimize “rundown,” and to inhibit channel activity, while
fluctuations in ADP regulate opening of the channel when
the rate of glucose metabolism changes. What is needed to
validate this model for the b-cell is quantitative data on the
actual physiological variation in [ADP]i with glucose metabolism and data indicating that b-cell KATP channels respond to variations in the ADP/ATP ratio within this physiological range.
A. How do ATP and ADP exert their effects on
KATP channels?
It is not understood, at the molecular level, how ATP and
ADP exert their effects on KATP channels. Several factors
need to be investigated and integrated into a coherent regulatory scheme, including the number and location of the
nucleotide binding sites, a determination of whether sulfonylurea receptors, like other members of the ABC superfamily, have ATPase activity and, if so, how hydrolysis is involved in gating of these channels. It will be necessary to
understand how potassium channel openers exert their effects and what relationship their action has to activation of
KATP channels by MgADP.
B. Where are the nucleotide-binding sites located?
1. SUR binds ATP. Amino acid sequence similarities placed
the sulfonylurea receptors in the ATP-binding cassette superfamily and predicted that SURs have two NBFs, while
KIR6.1 and KIR6.2 were members of the inwardly rectifying
K1 channel family with no obvious nucleotide-binding motif. Thus family lineage, plus the observation that the
SUR2A/KIR6.2 channels displayed a somewhat higher IC50
for ATP inhibition (23) than the SUR1/KIR6.2 channels, suggested the receptor was the nucleotide sensor. As described
below, the situation has become more complicated.
Ueda et al. (239) have shown that [32P] 8-azido ATP can be
used to photolabel SUR1. Based on the pattern of labeling of
receptors with mutations in the conserved Walker A motifs
of NBF1 and NBF2, high- and low- affinity ATP-binding sites
were distinguished, and NBF1 was identified as the highaffinity site. MgADP reduced labeling and was proposed to
antagonize binding of ATP at NBF1 by its interaction with
NBF2. Ueda et al. (239) further suggested that incubation of
SUR1 with 8-azido ATP at 37 C induced a conformational
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
change that made nucleotide binding tighter and more stable. The results demonstrate SUR1 can bind ATP as anticipated and appeared to identify a stable, long-lived ATPliganded state of SUR1 with no apparent relationship to
inhibition of KATP channels.
Figure 7 illustrates several properties of the photolabeling
of SUR1 with [32P] 8-azido ATP and competition with unlabeled 8-azido ATP and ATP. Qualitatively similar results
were obtained with ATP42, ADP32, MgATP, MgADP, and
several nonhydrolysable analogs. It is difficult to interpret
the results of the experiments done in the presence of Mg21
since these membrane preparations contain adenylate kinase, which readily converts ADP to ATP plus AMP. In
addition, it is uncertain whether SURs have an intrinsic
ATPase activity; thus we have not tried to determine an
apparent IC50 for nucleotides in the presence of divalent
cations. The estimated IC50 for 8-azido ATP42 is approximately 1 mm and approximately 10 –15 mm for ATP42, in
general agreement with the data of Ueda et al. (239). Interestingly, this is the range where ATP42 half-maximally inhibits KATP channels (Fig. 1 for example) and would be consistent with binding of ATP42 to SUR1, serving to inhibit
FIG. 7. Labeling of SUR1 with [a32P] 8-azido ATP. A, Loss of 8-azido
ATP from SUR1 at 25 C. Membranes were incubated with 2 mM
[a32P]8-azido ATP in 150 mM NaCl, 50 mM Tris-HCl, 2 mM EDTA, pH
7.5 for 10 min, diluted with 20 volumes of the same buffer (ice cold),
pelleted in a microfuge for 15 min at 4 C, and resuspended in 50 ml
aliquots of ice-cold buffer. Dissociation of the complex was initiated
by transfer to a water bath at 25 C. Time points were taken by addition
of 1 ml of ice-cold buffer and irradiation at 254 nm (total energy 5 1
J/cm2) on ice. Previous experiments indicated the loss of label at 0 – 4
C is negligible of the time course of the experiment (J. Bryan, unpublished data). The estimated half-life of the SUR1;8-azido ATP
complex at 25 C is 7.5 min. B, Competition of 8-azido ATP labeling by
unlabeled 8-azido ATP42. Membranes were incubated with 2 mM
[a32P]8-azido ATP and the indicated concentrations of 8-azido ATP in
150 mM NaCl, 50 mM Tris-HCl, 2 mM EDTA, pH 7.5 for 10 min, diluted
with 20 volumes of the same buffer (ice cold), pelleted in a microfuge
for 15 min at 4 C, resuspended in 20 ml of the same buffer, and then
irradiated at 254 nm (total energy 5 1 J/cm2) on ice. C, Competition
of 8-azido ATP labeling by ATP4-. The conditions were as described in
panel B.
117
channel activity. On the other hand, we estimate the off rate
for 8-azido ATP bound to SUR1 to be approximately 10 min
at 25 C (Fig. 7), several orders of magnitude slower than
would be expected from patch-clamp experiments. Thus,
while the affinity of SUR1 for ATP42 appears to be consistent
with that observed for inhibition of KATP channels, the kinetics of dissociation are markedly slower than expected.
An equally puzzling set of observations (189, 190, 238) is
the failure to see an increase in the IC50 for ATP inhibition in
channels reconstituted from wild-type KIR6.2 and SUR1 subunits with mutations in their NBFs. The general conclusion
is that channels with mutations in NBF2 are not stimulated
by MgADP and diazoxide, while channels with mutations in
NBF1 can be stimulated, but display altered kinetics. In other
ABC proteins, similar mutations have large effects on binding and hydrolysis of ATP, but the mutant SUR1/KIR6.2
channels display near-normal inhibition by ATP. For example, the K719R mutation in the conserved Walker A motif that
is involved in binding of the a- and b-phosphates has no
effect on the ATP sensitivity of reconstituted channels (190),
although Ueda et al. (239) show that SUR1K719R does not
photolabel with [32P] 8-azido ATP. These results imply there
is another site(s) for binding of inhibitory ATP to KATP
channels.
2. ATP can interact directly with KIR6.2. Tucker et al. (51) have
shown that expression of C-terminally truncated KIR6.2 subunits generated K1 channels that were inhibited by ATP.
Since these truncated subunits were not associated with SUR
(see Fig. 6), the result indicated nucleotides could interact
directly with a “primary” inhibitory ATP-binding site on
KIR6.2. Removal of 18, 26, and 36 (DC18, DC26, and DC36,
respectively) amino acids from the C terminus of KIR6.2
produced K1 channels, while DC41 showed no activity. The
DC26 and DC36 channels were the most active and had
conductance properties similar to the SUR/KIR6.2 channels.
The truncated channels showed rundown and refreshment
as described above, but were not activated by MgADP or
potassium channel openers such as diazoxide and were not
inhibited by sulfonylureas. Thus, these regulatory functions
were presumed to require SUR, and coexpression of
KIR6.2DC subunits with SUR1 restored these properties. The
IC50 values for inhibition of the KIR6.2DC channels by ATP42
were 100 –200 mm with no significant Mg21 dependence.
These values were 20- to 40-fold higher than for the SUR1/
KIR6.2 channels. Coexpression of SUR1 with KIR6.2DC26 reduced the IC50 for ATP inhibition, leading to the idea that
SUR1 could “enhance” the inhibitory effect of ATP. The
nature of the ATP-binding site was not clear, although the
authors put forward a simple charge argument and showed
that substitution of glutamine for lysine 185 markedly reduced
the sensitivity to ATP (IC50 . 1 mm). An extensive series of
site-directed mutations did not define a binding site but did
indicate that substitution at positions near the mouth of the
pore and in the N terminus would also reduce the sensitivity
to ATP. The interpretation of the site-directed mutation experiments has been confounded by the realization that a
given substitution can alter both the apparent affinity for
ATP and the open channel probability, making it difficult to
determine whether a particular residue is part of a binding
118
AGUILAR-BRYAN AND BRYAN
site or part of a link between a binding site and the channel
gate. The results with C-terminally truncated channels are
important and provide direct evidence that KIR6.2 forms the
KATP channel pore and that ATP can directly affect its
activity.
Shyng et al. (238) have proposed a model in which SUR1 acts
to “sensitize” the KIR6.2 channel to inhibition by ATP, a process
akin to the enhancement described by Tucker et al. (51). Shyng
et al. (238) propose that SUR1 can be switched into a state,
designated the “off” state, which cannot sensitize KIR6.2. ATP
hydrolysis, at one or both NBFs, switches SUR to the off state,
thus blocking sensitization, while MgADP and potassium channel openers, like diazoxide, are argued to stabilize the off state
and block sensitization. Mutations in the NBFs that prevent or
slow hydrolysis are expected to reduce entry into the off state,
thus maintaining these channels in a sensitized state. Although
other interpretations are possible, in the simplest incarnation of
this model, the function of SUR1 is to control the affinity of the
primary inhibitory ATP-binding site on KIR6.2, with higher
affinity for ATP equating to a reduced open probability, Po, of
a single channel.
C. C-terminally truncated KIR6.2 channels show
abnormal kinetics
While the single-channel conductance of the homomeric
KIR6.2DC channels was equivalent to that of the SUR/KIR6.2
Vol. 20, No. 2
channels, analysis of their single-channel kinetics revealed
substantial differences in basic gating properties (240). As
shown in Fig. 1, SUR1/KIR6.2 channels exhibit bursting behavior with a burst consisting of a number of fast openings
defined by brief closures. Bursting in the KIR6.2DC channels
is markedly attenuated with less than two openings in an
average burst (see Fig. 8). The average lifetime of these fast
openings is also reduced by about 3-fold. The KIR6.2DC
channels spend most of their time in long closed states analogous to the interburst intervals seen in SUR1/KIR6.2 channels. The result is a marked decrease in the maximal open
probability. Under equivalent conditions, we estimate the
maximal Po of the KIR6.2DC channels to be 0.09 vs. 0.69 for
SUR1/KIR6.2 channels.
D. Coexpression of KIR6.2DC subunits with SUR restores
normal KATP channel activity
Just as the study of KIR6.2DC channels has provided data
on the properties of the pore in the absence of SUR, they have
also underscored exactly how dependent the properties of
normal KATP channels are on the interactions between both
subunits. It is clear that coexpression of KIR6.2DC subunits
with SUR1 restores sensitivity to sulfonylureas, potassium
channel openers, and stimulation by nucleoside diphosphates (51). In addition, coexpression nearly restores the
FIG. 8. Comparison of wild-type and truncated KATP channels. Single-channel records from the three types of channels illustrated in Fig. 6
are shown. The (KIR6.2DC35)4 channels show abnormal bursting and have a markedly reduced probability of being in an open state. The maximal
open probability, Po, estimated from the all-points histogram is 0.09. The behavior of the wild-type channel is shown in the middle record.
Coexpression of KIR6.2DC35 with SUR1 restores normal bursting, increasing the maximal Po. The solution conditions were as follows: internal
solution (millimolarM concentration) KCl, 140; MgCl2, 1; EGTA, 5; HEPES, 5; KOH, 10; pH 7.2 (adjusted with KOH); external solution: NaCl,
140; KCl, 5; MgCl2, 1; CaCl2, 1; HEPES, 10; pH 7.4 (adjusted with NaOH). The holding potential was 0 mV. The horizontal dotted lines give
the zero-current level; the upward deflections correspond to outwardly directed currents.
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
normal bursting pattern giving a maximal Po of 0.49 for the
SUR1/KIR6.2DC channels (240). Finally, coexpression reduces the sensitivity to inhibition by ATP, suggesting that
SUR may increase the affinity of KIR6.2 for ATP, either by an
allosteric mechanism or by formation of a shared binding
site. A comparison of single-channel records from KIR6.2DC
and SUR1/KIR6.2DC channels with the wild-type channel
illustrates the highly integrated nature of the two subunits
within a KATP channel (Fig. 8).
E. Why are KIR6.2 channels silent?
The initial reconstitution studies failed to detect K1 channels when KIR6.2 was expressed without SUR1 (26), a result
that has been confirmed by others. The reasons for this silence or latency were unclear, but recent results suggest a
fascinating story, the outlines of which are just becoming
clear. The finding that KIR6.2DC subunits could form K1
channels, albeit with abnormal properties, provided a means
to measure the effects of coexpression with SUR1 on trafficking. Babenko et al. (240) showed that coexpression with
SUR1 increased the transit of KIR6.2DC subunits to the
plasma membrane by about 8-fold. John et al. (241) showed
that very strong overexpression of full-length KIR6.2, or
KIR6.2 tagged with GFP, the green fluorescent protein, in
HEK cells produced K1 channels with properties similar to
the C-terminally truncated KIR6.2 channels. This was the first
demonstration that full-length KIR6.2 could generate channels and effectively ruled out the possibility that the C terminus blocked channel activity. The results suggested that
the C terminus affected trafficking in some unspecified fashion. Schwappach et al. (242) have shown that the C-terminal
sequence of KIR6.2 (and KIR6.1) has an endoplasmic reticulum (ER) retention signal that keeps full-length KIR6.2 from
reaching the plasma membrane under normal conditions.
Using a quantitative assay to measure surface expression of
KIR6.2, Schwappach et al. (242) could demonstrate that coexpression with SUR1 increased transit approximately 500fold. Transfer of the KIR6.2 C-terminal sequence to other
membrane proteins caused their retention in the ER. Truncation of the C terminus of KIR6.2 thus allows its limited
expression at the plasma membrane, yielding the results seen
by Tucker et al. (51). The most likely explanation for the
observations of John et al. (241) is that the retention mechanism can be overcome when levels of expression are sufficiently high. How SUR interacts with KIR to abrogate the
retention signal promises to be an interesting story, but it is
not at all clear why nature has gone to such lengths to control
surface expression of this inward rectifier.
F. The N terminus of KIR6.2 limits burst duration
The C terminus of KIR6.2 is not the only interesting cytoplasmic domain. Truncation of the N terminus of KIR6.2 has
proven to be equally informative. Deletion of 30 – 40 amino
acids does not affect surface expression but has a profound
effect on bursting activity (240). SUR1/DN32KIR6.2 channels,
for example, burst continuously with few interburst intervals. The maximal open probability, Po ;0.95, is the value
expected from a channel displaying only the fast closures
119
observed within a normal burst. We have interpreted these
results to imply that the N terminus of KIR6.2 limits the time
the channel spends in a bursting state and speculate that SUR
affects bursting through its interaction with the N terminus.
G. Where do the openers bind and how do they work?
A diverse group of compounds, referred to as potassium
channel openers or KCOs, are able to open KATP channels and
thus hyperpolarize cells. Clinically, these compounds have
been used for the treatment of hyperinsulinemic states and
as smooth muscle relaxants for hypertension. The literature
in this area is voluminous, and a complete review is beyond
the scope of this article. Reviews may be found in Refs.
243–253. This area has been hampered by the lack of means
to study specific KATP channel isoforms, but this problem
should now be more accessible by reconstitution of the
SUR2A and SUR2B receptors with both KIR6.1 and KIR6.2 (23,
24, 254), as described above.
Electrophysiological studies indicate that activation of
KATP channels in excised patches requires hydrolysable nucleotides and Mg21 (238, 255–263), consistent with a requirement for ATP hydrolysis. Several of these studies point out
that ADP in the presence of ATP or ADP alone will potentiate
the effect of KCOs. The latter observations are consistent with
the presence of adenylate kinase and conversion of ADP to
ATP; thus it is not clear whether ADP and KCOs can act
synergistically. Several reports have argued that the stimulatory effect of KCOs is through direct competition with ATP
(257, 264, 265), although this is not firmly established.
The target site for potassium channel openers is assumed
to be the sulfonylurea receptor, based on observations that
the response of reconstituted KATP channels to either diazoxide, cromakalim, or pinacidil is correlated with the SUR
subtype (23, 24). A comparison of the pharmacology of the
SUR2A/KIR6.2 (23) and SUR2B/KIR6.2 (24) channels indicates the C-terminal end of SUR is a critical determinant for
KCO response and selectivity. The SUR1 and SUR2B C termini, for example, are similar, and channels formed from
these receptors appear to share responsiveness to diazoxide,
while the SUR2A/KIR6.2 channel shows essentially no response to diazoxide.
Attempts to develop membrane-binding assays for channel openers from responsive tissues have been largely unsuccessful, although in some cases tissue and whole-cell
binding has been detected. The most successful efforts have
employed [3H]P1075, a pinacidil analog that activates
smooth muscle KATP channels at concentrations of 20 –30 nm
and channels from isolated rat aortic rings and myocytes
(248, 266). This line of research has recently shown that ATP
is required for [3H]P1075 binding to isolated cardiac and
skeletal muscle membranes (267, 268). Biochemical studies
on membranes isolated from cells transfected with SUR2A
and B (269, 270) suggest two plausible reasons for the earlier
difficulties: low receptor concentration and a requirement for
Mg21 and hydrolysable nucleotides. Schwanstecher et al.
(269) show that SUR2B has a KD of approximately 25 nm for
P1075, while binding of this compound to SUR1 and SUR2A
is low. Binding requires MgATP; nonhydrolysable ATP analogs will not support [3H]P1075 binding. Consistent with
120
AGUILAR-BRYAN AND BRYAN
the electrophysiological data (189, 190, 238), binding of the
opener is impaired by mutations in the NBFs. The suggestion
is that nucleotide binding and hydrolysis by SUR are important for the binding of P1075. The results appear to argue
against direct competition between ATP and KCOs. The
availability of binding assays opens the way for the experiments with different KCOs and for the structure-function
studies with the SUR subtypes needed to establish the number and location of KCO-binding sites on KATP channels and
to correlate binding with channel subtype responses.
H. Do SURs have ATPase activity?
Although there is no direct biochemical evidence that
SURs are ATPases, the indirect evidence is reasonably strong.
First, the similarity of the NBFs of SURs to those of other
members of the ATP-binding cassette superfamily suggests
they will hydrolyze ATP. Second, as discussed, several aspects of channel function require Mg21 and hydrolysable
nucleotides, particularly refreshment and the regulation of
channel activity by MgADP and potassium channel openers.
As indicated above, refreshment is not well understood and
may require additional kinases or cytoskeletal elements that
require ATP. MgADP and KCOs, on the other hand, interact
with SUR, and their effects show a strong requirement for
divalent cations such as Mg21 or Mn21, and for the presence
of a hydrolysable nucleotide. It is not clear, in most cases,
whether ATP is required or other nucleotides will substitute.
For example, MgADP has been observed to substitute for
ATP and support the action of diazoxide, while a,b-methylene ADP does not (263). This could be interpreted as a
requirement for ADP hydrolysis, but probably reflects the
generation of ATP by adenylate kinase. This is an area that
needs to be developed and should provide insight into the
regulation of these channels.
I. Do SURs have transport activity?
Membership in the ATP-binding superfamily suggests
SURs may have an endogenous transport function. To our
knowledge, the possibility that SURs are involved in transport has not been tested in any systematic fashion. One
possibility, based on suggestions from the CFTR field, is that
SUR might transport nucleotides (205). While this has not
been disproved rigorously, there is no direct evidence in
support of nucleotide transport at this time.
J. Is there an endogenous substrate?
A frequently raised question concerns the nature of the
putative “endogenous” ligand, which sulfonylureas may
mimic or replace. Virsolvy-Vergine et al. (271) have addressed this question directly using competition assays in
which peptide fractions isolated from brain were tested for
their ability to displace [3H]glibenclamide from either brain
or b-cell membranes. This work led to the isolation of two
peptides, a- and b-endosulfines (272, 273). Both molecules
were reported to interact with the high-affinity receptor, and
b-endosulfine was reported to stimulate insulin release from
a b-cell line. a-Endosulfine, a 77-amino acid protein (Mr 5
13,196), has been cloned and is similar to two phosphopro-
Vol. 20, No. 2
teins, ARPP-16 (96 amino acids) and ARPP-19 (112 amino
acids), Mr ; 16,000 and 19,000, respectively. The sequence
similarity is striking, with 65 identical amino acids over the
77 residues of a-endosulfine. ARPP-16 and -19 are believed
to be splice variants of a single gene since their amino acid
sequences are identical except for an additional 16 amino
acids at the N terminus of ARPP-19 (274). ARPP-16 is present
mainly in dopamine-innervated regions of the brain, while
ARPP-19 is more widely distributed being present in all
tissues in the adult rat (275). The phosphorylation of both
ARPP-16 and -19 is regulated by cAMP and by vasoactive intestinal peptide in cultured striatal cells (276). a-Endosulfine is believed to be the product of a separate gene and
has a wide tissue distribution (277). Expression of recombinant a-endosulfine in a bacteria produces a protein with an
apparent molecular mass of 23,000 Da, which competes with
[3H]glibenclamide for binding to SUR1 (Ki ;1 mm), inhibits
SUR1/KIR6.2 channels expressed in Xenopus oocytes (EC50
;1 mm) and stimulates insulin release from MIN6 cells in the
same concentration range. Heron et al. (277) suggest that
a-endosulfine may act as an endogenous regulator of KATP
channels in b-cells.
VIII. Human SUR1 and KIR6.2 Genes
The human SUR1 and KIR6.2 genes have been mapped to
the short arm of chromosome 11 (11p15.1). The SUR1 gene
has 39 exons spanning approximately 100 kb of genomic
DNA. The average exon size is 124 bp with a range of 33–243
bp; the known intron sizes range from 100 bp to greater than
6.5 kb. Figure 9 provides an overview and superimposes the
major structural features of the protein on the human gene,
as well as summarizing the known SUR1 mutations associated with familial hyperinsulinism. We have reviewed the
structure of the human SUR1-KIR6.2 gene region elsewhere
along with a comparison of the human SUR1 and SUR2 genes
(19). There is new sequence available that will be of interest
to those screening for disorders associated with the SUR1KIR6.2 complex. Evans et al. have put an 86-kb sequence into
the Genbank database (Accession no. U90583). The sequence
starts in a large intron (.6.5 kb) between SUR1 exon 16 and
17 and spans exons 17–39. KIR6.2 is downstream of the SUR1
gene (4900 bp between the stop codon of SUR1 and the start
codon of KIR6.2). The entire SUR1-KIR6.2 region plus a considerable amount of flanking sequence is available with the
sequence of a pac clone, pDJ239b22, in the Genbank database
(Accession no. U90583). The entire coding region including
both genes is contained within a segment that spans almost
90 kb of DNA. The promoter sequences in the 59-untranslated
regions of the SUR1 and KIR6.2 genes have begun to be
explored, which should lead to understanding of how this
family of channel proteins is regulated (278).
IX. KATP Channels and Persistent Hyperinsulinemic
Hypoglycemia of Infancy (PHHI)
PHHI, also referred to as familial hyperinsulinism or pancreatic nesidioblastosis (Online Mendelian Inheritance in
Man: 256450), is a rare metabolic disorder of neonates and
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
121
FIG. 9. Representation of the human SUR1 gene. An illustration of SUR1 is depicted in the top line. The shaded boxes position the 17 predicted
TMDs, the open boxes, identified with A and B to mark the Walker A and B motifs, position the two NBFs. The 39 exons of the SUR1 gene are
shown below the protein. The exon boundaries are projected onto the protein. The approximate positions of known mutations in SUR1 that have
been identified in patients with PHHI are given. The intron mutations are identified by intron number for clarity, and intron base changes are
lowercase. Exon base changes are given as uppercase and are designated by the use of “nt.”
infants characterized by inappropriate secretion of insulin
despite severe hypoglycemia (see Refs. 18 and 20 for recent
reviews). Plasma insulin levels are frequently elevated, with
the key feature of being inappropriately high for the degree
of hypoglycemia. Infusion of large amounts of glucose, more
than 12 mg/kg/min, is required to maintain euglycemia in
newborns. In severe cases, patients are unresponsive to KATP
channel openers such as diazoxide and to treatment with
somatostatin derivatives such as octreotide. Partial or subtotal pancreatectomy is required to relieve the hypoglycemia
in these cases to avoid mental retardation.
The diagnosis of hyperinsulinism, considered here as an
inappropriately high insulin level, low blood ketone levels,
and low free fatty acid levels at the time of hypoglycemia, can
result from several genetic alterations. At the suggestion of
Dr. Franz Matschinsky, we will name the familial hyperinsulinemic disorders for which a responsible gene has been
identified, first by designating them as hyperinsulinemic and
second by designating the affected molecule. For example,
hyperinsulinism associated with a mutation in the sulfonylurea receptor would be designated “HI-SUR1.” As we are
reviewing KATP channels, we will focus mainly on mutations
in channel subunits that give rise to hyperinsulinism, but
information on the causes of persisting hypoglycemia in
newborns is widening and teaching us much about the heterogeneity of this disease. It is clear that KATP channel defects
are not the sole cause. Additional mutations in other pathways have been identified as discussed briefly below, and
focal and diffuse histopathological forms of the disease have
been identified and partially related to KATP channel defects.
It has also become clear that nesidioblastosis is not the pathognomonic histopathological lesion for PHHI.
A. HI-glucokinase (GK)
GK, or hexokinase IV, found in b-cells is responsible for
conversion of glucose to glucose-6-phosphate and is generally considered to be the b-cell “glucose sensor” (216). Mutations in GK have been identified as one cause of maturity
onset diabetes of the young (MODY 2), where increases in the
Km of GK are correlated with an increase in the concentration
of glucose needed for insulin release. Glaser et al. (279) described an informative mutation in GK which lowers the Km
and results in hypoglycemia. The mutation, val3met, V408
m, causes a rare, mild dominant form of PHHI, which responds clinically to diazoxide. No homozygotes have been
identified, suggesting that two copies of the allele may be
lethal.
B. HI-GlnDH
Zammarchi et al. (280) and Weinzimer et al. (281) described
a mild, dominant form of hyperinsulinism associated with
hyperammonemia. Several patients from unrelated families,
who responded to diazoxide but were poorly responsive to
octreotide, have been identified. This form of hyperinsulin-
122
AGUILAR-BRYAN AND BRYAN
ism appears to be the result of excessive oxidation of glutamate via glutamine dehydrogenase. Enzyme studies indicate
that the glutamine dehydrogenase in two unrelated patients
with this disorder was less sensitive to GTP, a negative allosteric effector, as a result of mutations in the C terminus of
the enzyme (282).
C. HI-“unknown”
Kukuvitis et al. (283) reported on PHHI in a FrenchCanadian kindred. Mutations in KATP channel subunits and
GK were ruled out as potential causes. These patients responded to diazoxide, confirming they have functional KATP
channels. Familial clustering suggested that this may be another mild dominant form of PHHI, but the susceptibility
locus has not been identified. In several populations studied,
approximately 50% of the cases fall within this unknown
category.
D. HI-KIR6.2
Three mutations that produce persisting hypoglycemia
due to hyperinsulinism have been identified in KIR6.2 (Fig.
5). Although as described below, mutations in SUR1 were the
first to be associated with hyperinsulinism and are more
frequent, mutations in KIR6.2 produce the same phenotype
and provide a genetic confirmation of the biochemical and
electrophysiological data that KIR6.2 is the partner for SUR1
in b-cells.
Thomas et al. (284) reported on a KIR6.2 mutation, a
leu3pro change, L147P, near the external side of M2, which
demonstrated that mutations in the inward rectifier led to
PHHI. This mutation was identified in a child of Iranian
origin, the progeny of a marriage of first cousins. Diagnosis
of PHHI was based on an insulin level of more than 30
mU/ml with a glucose level less than 30 mg/dl and a requirement of more than 15 mg glucose kg/min to maintain
euglycemia. The substitution of a proline in M2 has a large
effect on KATP channel activity; when engineered and expressed with wild-type SUR1, KIR6.2L147P does not produce
KATP channels and does not co-photolabel with [125I]iodoazidoglibenclamide (J. Bryan, unpublished data). It is not clear
whether the protein is folded correctly and/or fails to associate with SUR1.
Nestorowicz et al. (285) described a nonsense mutation in
KIR6.2 that truncates the protein after 12 amino acids, Y12X.
The patient was homozygous for this mutation. When Y12X
was engineered into KIR6.2, as expected for a 12-residue
peptide that is missing all of the elements required to form
the K1 channel pore, it did not form a functional channel
when coexpressed with wild-type SUR1.
Sharma and Aguilar-Bryan (unpublished data) identified
a third KIR6.2 mutant, a trp3arg change, W91R, near the
external side of M1. This mutation was identified in a newborn, the product of a marriage of first cousins of Palestinian
descent. Clinical treatment with diazoxide and somatostatin
was not successful; a partial pancreatectomy was performed
when the patient was 2 weeks of age, and a second resection
was necessary 4 weeks later. When coexpressed with wildtype SUR1, KIR6.2W91R failed to produce active channels.
Vol. 20, No. 2
E. HI-SUR1
A number of studies indicated that, in most instances,
PHHI was an autosomal recessive disorder (summarized in
Ref. 20). Two groups (286, 287) independently mapped the
major susceptibility gene to the short arm of chromosome 11
at 11p14 –15.1 using multiplex families. In situ hybridization
mapped the SUR1 gene to the same region (288). It was
apparent that loss of KATP channel activity in pancreatic
b-cells might be expected to give the PHHI phenotype. As
described above, KATP channels set the b-cell resting membrane potential below the threshold for activation of voltagegated Ca21 channels. In the absence of some compensatory
mechanism, the loss of b-cell KATP channel activity would
lead to constitutive membrane depolarization, spontaneous
voltage-gated Ca21 channel activity, and a sustained increase
in intracellular Ca21 levels. The loss of K1 channel activity
would therefore effectively uncouple membrane electrical
activity from metabolism and would result in persistent insulin release, regardless of the blood glucose level (18, 289).
The discovery of two separate splice site mutations that
segregated with the disease phenotype in affected children
from nine different families provided the first evidence that
SUR1 was a PHHI susceptibility gene (288). The first two
mutations discovered were in intron 32 and exon 35, respectively. [Note, in the Permutt review article (20), the authors
have used an early numbering of the SUR1 exons where the
39-most exon is defined as exon 1. The human SUR1 sequence(s) in Genbank begin with exon 1 at the N terminus;
we will use this numbering convention throughout.] The
intron 32 allele is a g3a mutation identified in 1 PHHI
affected children from consanguineous mating. This mutation eliminates an NciI restriction endonuclease recognition
site; homozygous loss of this site cosegregated with the disorder within the two families. The g3a mutation was at a
position 29 from the 39-end of intron 32 (Fig. 9). The resulting
sequence, -gcc cag ccc cagCAC- caused usage of three cryptic
splice sites within exon 33 in place of the normal wild-type
splicing. The “alternative” splicing was expected to produce
transcripts encoding truncated receptors of three different
sizes missing NBF2. The exon 35 allele is also a G3 A mutation in the last position of the exon (Fig. 9). This change
eliminates an MspI site; homozygous loss of this site was
observed to cosegregate with the phenotype in 8 families and
was missing in 12 affected children from 6 families of Saudi
Arabian and 1 of German origin. The G3 A mutation in the
last position of the exon weakens the splice site and results
in the skipping of exon 35. The exon deletion causes a frame
shift that results in a premature stop 24 codons beyond the
end of exon 34, again producing a transcript that encodes a
truncated receptor missing most of NBF2. RT-PCR was used
to clone a pancreatic cDNA product missing the 109-bp segment, showing that the altered transcript was expressed.
A screen for SUR1 mutations in 25 probands from
Ashkenazi Jewish families affected with PHHI uncovered an
additional mutation, a deletion of codon 1388 resulting in a
loss of a phenylalanine residue, DF1388 (290). In some patients this mutation was present in the homozygous state and
in others it was present as a compound heterozygote with
known SUR1 mutations or mutations yet to be identified.
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
Thomas et al. (291) reported on three additional mutations
in the region of NBF1. Two of these mutations are expected
to lead to severe truncations of SUR1 in or near NBF1, while
the third mutation results in a gly3val substitution at position 716, G716V, in the Walker A motif.
Nichols et al. (189) reported on a missense mutation,
G1479R, in the second NBF, which is informative about the
mechanism of nucleotide regulation as discussed below. This
mutation was identified by Dr. Ann Nestorowicz in Dr. Alan
Permutt’s laboratory in a patient of Iraqi and Moroccan Jewish extraction (20) who was diagnosed with PHHI by Drs.
Heddy Landau and Benjamin Glaser (Hebrew University,
Hadassah Medical Center, Jerusalem). The patient is a compound heterozygote, with the G1479R allele on one chromosome and a second, unidentified allele on the other. As
described above, when the G1479R mutation was engineered
into hamster SUR1 and expressed with wild-type KIR6.2, the
resulting channel displayed a novel phenotype with normal
inhibition by ATP, but reduced stimulation by MgADP.
A number of new mutations have been identified in patients with PHHI from different ethnic groups (Ref. 292) and
N. Sharma and L. Aguilar-Bryan, unpublished data). We
have summarized as many of these mutations as are available in Fig. 9. Some of these mutations have been engineered
into the hamster or human SUR1 and examined for KATP
channel activity after coexpression with wild-type KIR6.2
(Ref. 292 and L. Aguilar-Bryan, N. Sharma, A. Crane, G.
Gonzalez, and J. Bryan, unpublished data). A number of
these mutations are associated with severe forms of the disease and show no KATP channel activity in the reconstitution
assays. This includes a failure to show increased sulfonylurea-sensitive 86Rb1 efflux upon metabolic poisoning. Several
of these mutations behave like the G1479R mutation and
retain some sensitivity to diazoxide. With the exception of
the severe truncations and the G716V mutation, the SUR1
mutants retain high-affinity sulfonylurea binding activity,
suggesting their folding is not completely aberrant. Detailed
examination of these mutations should provide further insight into the regulation of KATP channels.
X. Linking PHHI to Defects in KATP Channel Activity
A. b-Cells from newborns diagnosed with “sporadic” PHHI
lack KATP channel activity
Kane et al. (293) studied islet cells derived from partial
pancreatectomy specimens from five patients (four of Caucasian and one of Kuwaiti origin) diagnosed with sporadic
PHHI, differentiated here from familial hyperinsulinism, because there is a single affected sibling in the family and no
knowledge of consanguinity. KATP channels could not be
detected in primary cultured islet cells, although they were
present in normoglycemic human tissue from unaffected
individuals (1 infant, 11 months of age, and 12 adults, median
age 45 yr) treated similarly. The control islet cells were electrically silent, while the patient’s islet cells exhibited spontaneous openings of voltage-gated Ca21 channels that resulted in an estimated increase in [Ca21]i of about 2-fold (;88
nm to 177 nm). The results confirmed the hypothesis put
forward earlier that loss of KATP channel activity should
123
produce the PHHI phenotype (18). The missing link in this
study was a direct connection between a mutation(s) in one
of the KATP channel subunits and the loss of channel activity.
These five patients had no parental consanguinity, no other
affected siblings, and a preliminary, but not exhaustive,
screen for SUR1 and KIR6.2 mutations was negative (L. Aguilar-Bryan, unpublished data). Thus, although the physiology
was compelling, one could argue that the genetic defect lay
elsewhere in these patients.
B. PHHI b-cells with the SUR1 exon 35 mutation lack KATP
channel activity
A direct demonstration that a known SUR1 mutation results in the loss of KATP channel activity in patient b-cells was
made for the exon 35 mutation. Dunne et al. (294) analyzed
islet cells obtained by pancreatectomy from one newborn
diagnosed with PHHI. Analysis of patient DNA indicated
she was homozygous for the exon 35 G3 A mutation. She
was not responsive to diazoxide or to nifedipine and required continuous infusion of glucose (18 mg/kg/min) to
remain euglycemic. Subtotal (95%) pancreatectomy was insufficient to reduce her insulin levels, and removal of 99% of
the pancreas was necessary to control her hypoglycemia.
Electrical recording from b-cells isolated from the surgical
specimens demonstrated KATP channel activity was absent,
but voltage-gated Ca21 channels were spontaneously active.
As in the five infants studied previously (293), b-cell cytosolic
Ca21 levels were elevated beyond control cell values (83 vs.
115 nm) as determined by fluorescence measurements. A
parallel mutation engineered into hamster SUR1 did not
generate KATP channel activity when cotransfected into COS
cells with wild-type KIR6.2. Photolabeling studies on the
parallel mutation indicate that a truncated receptor is produced and retains high-affinity sulfonylurea binding activity. Preliminary results indicate that the truncated receptor
is able to associate with and co-photolabel KIR6.2. These
studies have been repeated with b-cells from a second newborn homozygous for the same allele, with essentially the
same results (M. Dunne, K. Lindley, A. Aynsley-Green, and
L. Aguilar-Bryan, unpublished data). In addition, four different mutations have been identified in patients previously
diagnosed with sporadic PHHI whose b-cells were shown to
lack KATP channel activity. The overwhelming conclusion is
that mutations in SUR1 result in the loss of KATP channel
activity that leads to PHHI. This indicates that KATP channels
are the predominant mechanism that human b-cells use to set
their resting membrane potential; there is no evidence for
redundancy of SUR1, KIR6.2, or KATP channel functions. This
conclusion emphasizes the key link that KATP channels form
in the closed negative feedback loop that regulates insulin
release and validates our initial hypothesis that this recessive
familial form of PHHI is a potassium channel disease (18).
C. Why is there a lack of dominant negative mutations?
Although analysis of PHHI mutations in SUR1 and
KIR6.2 is at an early stage, given the multimeric structure
of KATP channels it is not clear why dominant negative
mutations in either subunit have not been discovered. It is
124
AGUILAR-BRYAN AND BRYAN
Vol. 20, No. 2
not clear, for example, whether the truncated SUR1 subunits produced by the exon 35 mutation can assemble a
tetrameric channel. The exon 35 mutation, and other identified mutations, lead to severe hypoglycemia only in the
homozygous condition. The heterozygous parents are apparently normoglycemic and asymptomatic. Thus, dominant negative effects are either small or the fraction of
normal channels produced in the heterozygotes is sufficient to maintain the b-cell resting membrane potential. If
assembly of a single-mutant receptor into the tetramer is
sufficient to block channel activity, we expect only 6.25%
of the KATP channels to be functional in the heterozygotes
if the assembly of channel subunits is random. One possible explanation for the failure to identify dominant negative mutations could lie in the spare channel hypothesis
(230), if 6 –7% of the KATP channels in a b-cell are sufficient
for adequate regulation. Additional studies on the heterozygous parents could prove informative on this issue.
firm this initial suggestion. Gould et al. (303, 304) reported an
increase in total endocrine cell volume in PHHI patients but
concluded the “typical features of nesidioblastosis” were
also present in normal age-matched controls. They introduced the term “nesidiodysplasia” to define the abnormal
growth, distribution, and regulation of endocrine cells when
present in association with endocrine abnormality. Rahier et
al. (305, 306) reached a similar conclusion after comparing 15
hypoglycemic infants diagnosed with PHHI to 23 normoglycemic controls; “nesidioblastosis was not a specific feature of the pancreas in infantile hypoglycemia, being observed in age-matched controls as well.” Nesidioblastosis
has also been identified in other pathologies such as multiple
endocrine neoplasia, cystic fibrosis, and pancreatitis, without
hypoglycemia; therefore it is clear at this point, that nesidioblastosis is not the pathognomonic lesion of PHHI (307).
D. Development of mouse models
In addition to the heterogeneous genetic causes of PHHI,
at least two histopathological forms of hyperinsulinism have
been distinguished by morphology (308, 309): a focal type,
which shows nodular or adenomatous hyperplasia of a particular area within the normal pancreas, and a diffuse type,
which involves the entire pancreas and is characterized by
irregularly sized islets and ducto-insular complexes, both
with hypertrophied b-cells with larger nuclei. The two forms
of PHHI have different times of onset and degree of severity
(310); the diffuse form is more severe, develops within the
first 48 h after birth, and requires a large resection or total
pancreatomy to control hypoglycemia, while the focal form
is less severe and is identified in patients that develop hypoglycemia several weeks after birth. The focal form can be
controlled by resection of the lesion (310, 311). Sempoux et al.
(310) analyzed the pancreata from 25 PHHI infants with
diffuse or focal forms and evaluated b-cell hyperactivity by
determining the size of the nuclei and cytoplasm. Infants
diagnosed with the diffuse form of hyperinsulinemia had
larger b-cell nuclear radii than infants diagnosed with the
focal form. In addition, the cytoplasmic volume was greater
in the diffuse form. These two parameters allowed discrimination between the two different forms of the disease.
Infants with PHHI who have had a 95% pancreatectomy
frequently develop diabetes mellitus; therefore, distinguishing between the focal and diffuse forms of the disorder is of
importance (see for example Refs. 312 and 313 for discussion). Toward this end, two methods have been recently
employed. Dubois et al. (314) have used pancreatic venous
sampling (315–317), a technique in which catheters are used
to sample blood from different areas of the pancreas and
“map” local insulin release, to differentiate diffuse vs. focal
forms of PHHI at time of surgery. Similarly, the analysis of
small specimens from the head, isthmus, body, and tail by
means of frozen sections has been used to differentiate between the focal and diffuse forms. Rahier et al. (318) employed this approach and used nuclear size to discriminate
the two forms in a study of 20 infants diagnosed with PHHI
in the first few hours after birth who were nonresponsive to
diazoxide. These methods show promise as a means for
discriminating between the focal and diffuse forms of PHHI
Miki et al. (295) have shown that a dominant negative effect
can be observed by overexpression of a mutant KIR6.2
subunit. Transgenic mice expressing KIR6.2 carrying the
“weaver” mutation, G132S, have low blood glucose at birth
as a result of unregulated insulin secretion. Four weeks after
birth these animals develop hyperglycemia, with reduced
insulin secretion as a result of b-cell destruction. Reconstitution experiments indicate KIR6.2G132S can compete with
wild-type KIR6.2 and suppress KATP channel activity. Interestingly the weaver mutation in KIR3.2 (GIRK2) shows altered cation selectivity and leads to neuronal cell death (296,
297), suggesting this mutation in KIR6.2 could have similar
consequences.
Knockout mice missing both KIR6.2 and SUR1 are being
generated and should prove to be interesting and informative regarding KATP channel function in the pancreas and
other tissues.
XI. Other Issues
A. Nesidioblastosis does not cause PHHI
Since the initial description of nesidioblastosis, a term
coined by Laidlaw (298) to refer to a persistent, diffuse,
disseminated endocrine cell budding from pancreatic ducts,
many authors have suggested that this lesion is the specific
histopathological finding in the pancreas of hypoglycemic
neonates. Yackovac et al. (299) described nesidioblastosis in
the pancreas of newborns with persisting hypoglycemia and
made the initial suggestion that the increased b-cell mass
might produce the insulin increase and intractable hypoglycemia (300, 301). Heitz et al. (302) analyzed pancreatic tissue
from patients with PHHI using immunocytochemistry and
electron microscopy. They reported a 5-fold increase in the
mean area occupied by endocrine tissue relative to unaffected control tissue. The ratio of b-cells to endocrine area
was equivalent to that of controls.
More detailed comparisons of pancreatic tissue from patients with PHHI and normal individuals has failed to con-
B. “Diffuse” vs. “focal” forms of PHHI
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
and minimizing the extent of surgery required to control
hypoglycemia (319, 320).
Is there a relationship between the genetic causes of PHHI
and the two morphological forms? A recent report by de
Lonlay et al. (321) suggests this possibility in some instances.
These authors reported on 16 cases of “sporadic” PHHI, 10
identified with the focal form, and 6 with the diffuse form.
Analysis of DNA isolated from the foci of identified focal
PHHI suggested genomic imprinting, with a specific loss of
maternal alleles from the chromosome 11p15.1 region where
the SUR1 and KIR6.2 genes map. A similar loss was not
observed in DNA taken from cells outside the foci or in
samples taken from the pancreata of patients diagnosed with
the diffuse form of PHHI. de Lonlay et al. (321) show that a
somatic event resulting in deletion of the maternal alleles has
occurred and that a clonal expansion of the cells carrying the
deletion gives rise to a focus of b-cells missing KATP channels
as a result of a mutation in SUR1 on the paternal chromosome. The development of this area should prove interesting.
XII. KATP and NIDDM
It was of interest to determine whether mutations in SUR1
were associated with other disorders of glucose homeostasis,
particularly non-insulin-dependent diabetes, NIDDM. The
evidence for a connection is mixed. Hani et al. (322) and Inoue
et al. (323) have suggested that the SUR1 locus may contribute
to the genetic susceptibility of Caucasians to NIDDM, while
Iwasaki et al. (324) and Stirling et al. (325) have concluded that
this locus does not make a major contribution to this susceptibility in Japanese and Mexican-American populations,
respectively. Sakura et al. (326) have described sequence variations in KIR6.2, but failed to find linkage with NIDDM.
A. b-Cell type KATP channels in the brain
Several early binding studies established the presence of
high-affinity glibenclamide receptors, presumably SUR1, in
the brain (148, 158, 160, 327–329), and one of the earliest
reports of purification of the high-affinity receptor was from
porcine brain (161). Autoradiography with 3H- and 125Ilabeled glibenclamide have been used to localize receptors to
regions of the brain (330 –334). These studies localize SUR1
in various areas of the brain. Mourre et al. (330), using [3H]
glibenclamide, indicate that the density of receptors is “particularly important in the substantia nigra reticulata, the
septohippocampal nucleus, the globus pallidus, the neocortex, the molecular layer of the cerebellum, and the CA3 field
and dentate gyrus of the hippocampus. They note that the
”hypothalamic areas, medulla oblongata and spinal cord“
contained lower amounts of glibenclamide receptors. Treherne
and Ashford (331), also using [3H]glibenclamide, note the
”highest levels of glibenclamide binding were found in the
substantia nigra with high levels in the globus pallidus, cerebral cortex, hippocampus and caudate-putamen, intermediate levels in the cerebellum, and low levels in the hypothalamus and pons.“ These authors mention that only low
levels of binding were observed in glucose-responsive
regions of the brain known to respond to sulfonylureas.
Gehlert et al. (334), using [125I]iodoglibenclamide, report the
125
”highest levels of binding were seen in the globus pallidus
and ventral pallidum followed by the septohippocampal
nucleus, anterior pituitary, the CA2 and CA3 region of the
hippocampus, ventral pallidum, the molecular layer of the
cerebellum and substantia nigra zona reticulata. The hilus
and dorsal subiculum of the hippocampus, molecular layer
of the dentate gyrus, cerebral cortex, lateral olfactory tract
nucleus, olfactory tubercle and the zona incerta contained
relatively high levels of binding. A lower level of binding
(approximately 3- to 4-fold) was found throughout the remainder of the brain.“
In situ hybridization has been used to localize SUR1 and
KIR6.2 mRNAs in rodent brain (335). There was extensive
overlap of the two signals with each other and extensive
overlap with the earlier studies using radiolabeled glibenclamide. Figure 10 gives an example of the localization of
SUR1 by in situ hybridization in adult rat brain.
The localization studies have been paralleled by a large
number of reports on the electrophysiology of various neuronal KATP channels (336 –360). The function(s) of these neuronal KATP channels is unclear. They are usually considered
to be of importance during ischemia where they could function to reduce cell death (337, 341, 347, 361). Other reports
indicate that KATP channels are active under physiological
conditions in the hippocampus where they are regulated by
glucose levels (362). g-Aminobutyric acid release in the substantia nigra has also been shown to be affected by local
glucose concentrations, presumably through an action on
KATP channels (363). It will be of great interest to determine
the effects of SUR1 and KIR6.2 knockouts on neural behavior.
XIII. The Leptin Connection
One of the more interesting and speculative areas of research relating the functions of KATP channels in the brain
and in b-cells involves activation of channel activity by leptin, the product of the mouse obesity or ob gene. Leptin, a
16-kDa protein, is produced by adipocytes and is believed to
regulate body weight, food intake, and energy expenditure
through activation of the leptin receptor in hypothalamic
neurons and peripheral tissues (see for example Refs. 364 and
365). Ob/ob mice develop a profound obesity with hyperglycemia and hyperinsulinemia. Leptin treatment of ob/ob mice
improves their glycemia and insulinemia and causes a reduction in food intake along with increased physical activity
and thermogenesis. Leptin has been shown to activate a
potassium conductance sensitive to the sulfonylurea, tolbutamide, in glucose-responsive hypothalamic neurons (198).
Activation of these channels hyperpolarized the neurons inhibiting the firing of action potentials. The electrophysiological properties of these channels suggest they are distinct
from the b-cell/neuronal channels described here as they are
markedly less sensitive to inhibitory ATP, their slope conductance in isolated patches is about twice that observed
with b-cell channel, and tolbutamide fails to block channels
in isolated patches.
The development of hyperinsulinemia as a result of a
leptin deficiency or loss of the leptin receptor in db/db mice
suggested leptin might suppress insulin secretion under nor-
126
AGUILAR-BRYAN AND BRYAN
Vol. 20, No. 2
FIG. 10. Localization of SUR1 mRNA
in rat brain. The localization was done
after Largent et al. (374) using three
45-mer oligonucleotide probes based on
the rat SUR1 sequence (22). The highest levels of expression were seen in the
CA-2 and dentate gyrus (DG) of the hippocampus, and in the granular cell
layer (gr) of the cerebellum (CB). Lower
levels are in the pontine nucleus and
medial habenular nucleus. Light staining is evident throughout the cortex
(CTX) as well as the thalamus and inferior colliculus. [This work was done by
Dr. Noam Cohen in Dr. Solomon Snyder’s laboratory at the Johns Hopkins
University; the authors are indebted to
Dr. Cohen for his help.]
mal conditions (366, 367), and the long form of the leptin
receptor is found in islets and b-cell lines (368). A study on
the isolated, perfused rat pancreas failed to show an effect of
leptin on glucose-stimulated insulin secretion (369, 370),
while leptin is reported to suppress insulin release from
isolated b-cells (197, 371) and from the perfused pancreas of
ob/ob mice (371). This effect appears to involve KATP channels,
as Kieffer et al. (197) and Harvey et al. (372) have reported that
leptin activates a sulfonylurea-suppressible K1 conductance
in the insulinoma bTC3 and CRI-G1 insulin-secreting cell
lines, respectively. The time course of channel activation is
slow, occurring for more than approximately a 10-min period, suggesting an indirect link between activation of the
leptin receptor and the channel. In this regard, Harvey and
Ashford (201) have reported that application of tyrosine kinase inhibitors to CRI-G1 cells produced an activation of a K1
conductance similar to that seen with leptin, while whole-cell
dialysis with the tyrosine phosphatase inhibitor orthovanadate blocked the action of leptin and tyrosine kinase inhibitors. Serine/threonine-specific protein phosphatase inhibitors neither blocked nor reversed the action of leptin on KATP
channels. The authors conclude that leptin activation appears
to require inhibition of tyrosine kinases and subsequent dephosphorylation. Harvey and Ashford (200) further suggest
that opening of KATP channels by leptin may occur through
the activation of PI 3-kinase as wortmannin, a PI 3-kinase
inhibitor, blocked activation of the K1 conductance by tyrosine kinase inhibitors. Interestingly, Harvey and Ashford
(200) report that insulin can reverse the activating effect of
leptin on K1 channels and will block the effect if applied
before leptin. Unraveling how activation of the leptin receptor up-regulates KATP channels promises to be exciting and
to give deeper insight into the metabolic alterations associated with both obesity and NIDDM.
XIV. Summary and Conclusions
KATP channels are a newly defined class of potassium
channels based on the physical association of an ABC protein, the sulfonylurea receptor, and a K1 inward rectifier
subunit. The b-cell KATP channel is composed of SUR1, the
high-affinity sulfonylurea receptor with multiple TMDs and
two NBFs, and KIR6.2, a weak inward rectifier, in a 1:1 stoichiometry. The pore of the channel is formed by KIR6.2 in a
tetrameric arrangement; the overall stoichiometry of active
channels is (SUR1/KIR6.2)4. The two subunits form a tightly
integrated whole. KIR6.2 can be expressed in the plasma
membrane either by deletion of an ER retention signal at its
C-terminal end or by high-level expression to overwhelm the
retention mechanism. The single-channel conductance of the
homomeric KIR6.2 channels is equivalent to SUR/KIR6.2
channels, but they differ in all other respects, including bursting behavior, pharmacological properties, sensitivity to ATP
and ADP, and trafficking to the plasma membrane. Coexpression with SUR restores the normal channel properties.
The key role KATP channels play in the regulation of insulin
secretion in response to changes in glucose metabolism is
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
underscored by the finding that a recessive form of persistent
hyperinsulinemic hypoglycemia of infancy (PHHI) is caused
by mutations in KATP channel subunits that result in the loss
of channel activity. KATP channels set the resting membrane
potential of b-cells, and their loss results in a constitutive
depolarization that allows voltage-gated Ca21 channels to
open spontaneously, increasing the cytosolic Ca21 levels
enough to trigger continuous release of insulin. The loss of
KATP channels, in effect, uncouples the electrical activity of
b-cells from their metabolic activity. PHHI mutations have
been informative on the function of SUR1 and regulation of
KATP channels by adenine nucleotides. The results indicate
that SUR1 is important in sensing nucleotide changes, as
implied by its sequence similarity to other ABC proteins, in
addition to being the drug sensor. An unexpected finding is
that the inhibitory action of ATP appears to be through a site
located on KIR6.2, whose affinity for ATP is modified by
SUR1. A PHHI mutation, G1479R, in the second NBF of SUR1
forms active KATP channels that respond normally to ATP,
but fail to activate with MgADP. The result implies that ATP
tonically inhibits KATP channels, but that the ADP level in a
fasting b-cell antagonizes this inhibition. Decreases in the
ADP level as glucose is metabolized result in KATP channel
closure.
Although KATP channels are the target for sulfonylureas
used in the treatment of NIDDM, the available data suggest
that the identified KATP channel mutations do not play a
major role in diabetes. Understanding how KATP channels fit
into the overall scheme of glucose homeostasis, on the other
hand, promises insight into diabetes and other disorders of
glucose metabolism, while understanding the structure and
regulation of these channels offers potential for development
of novel compounds to regulate cellular electrical activity.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
References
1. Ho K, Nichols CG, Lederer WJ, Lytton J, Vassilev PM, Kanazirska
MV, Hebert SC 1993 Cloning and expression of an inwardly rectifying ATP-regulated potassium channel. Nature 362:31–38
2. Jentsch TJ 1994 Molecular physiology of anion channels. Curr Opin
Cell Biol 6:600 – 606
3. Pusch M, Jentsch TJ 1994 Molecular physiology of voltage-gated
chloride channels. Physiol Rev 74:813– 827
4. Hoffman EP 1995 Voltage-gated ion channelopathies: inherited
disorders caused by abnormal sodium, chloride, and calcium regulation in skeletal muscle. Annu Rev Med 46:431– 441
5. Gutmann L 1996 Axonal channelopathies: an evolving concept in
the pathogenesis of peripheral nerve disorders. Neurology
47:18 –21
6. Keating MT, Sanguinetti MC 1996 Molecular genetic insights into
cardiovascular disease. Science 272:681– 685
7. Keating MT, Sanguinetti MC 1996 Pathophysiology of ion channel
mutations. Curr Opin Genet Dev 6:326 –333
8. Lehmann-Horn F, Rudel R 1996 Channelopathies: the nondystrophic myotonias and periodic paralyses. Semin Pediatr Neurol
3:122–139
9. Greenberg DA 1997 Calcium channels in neurological disease. Ann
Neurol 42:275–282
10. Ptacek LJ 1997 Channelopathies: ion channel disorders of muscle
as a paradigm for paroxysmal disorders of the nervous system.
Neuromuscul Disord 7:250 –255
11. Sanguinetti MC, Spector PS 1997 Potassium channelopathies.
Neuropharmacology 36:755–762
12. Ackerman MJ, Schroeder JJ, Berry R, Schaid DJ, Porter CBJ,
Michels VV, Thibodeau SN 1998 A novel mutation in KVLQT1 is
27.
28.
29.
30.
31.
32.
33.
34.
35.
127
the molecular basis of inherited long QT syndrome in a neardrowning patient’s family. Pediatr Res 44:148 –153
Antes LM, Kujubu DA, Fernandez PC 1998 Hypokalemia and the
pathology of ion transport molecules. Semin Nephrol 18:31– 45
Connolly DL, Shanahan CM, Weissberg PL 1998 The aquaporins.
A family of water channel proteins. Int J Biochem Cell Biol 30:
169 –172
Doyle JL, Stubbs L 1998 Ataxia, arrhythmia and ion-channel gene
defects. Trends Genet 14:92–98
Karolyi L, Koch MC, Grzeschik KH, Seyberth HW 1998 The
molecular genetic approach to “Bartter’s syndrome.” J Mol Med
76:317–325
Oksche A, Rosenthal W 1998 The molecular basis of nephrogenic
diabetes insipidus. J Mol Med 76:326 –337
Aguilar-Bryan L, Bryan J 1996 ATP-sensitive potassium channels,
sulfonylurea receptors and persistent hyperinsulinemic hypoglycemia of infancy. Diabetes Rev 4:336 –346
Aguilar-Bryan L, Clement IV JP, Gonzalez G, Kunjilwar K,
Babenko A, Bryan J 1998 Towards understanding the assembly
and structure of KATP channels. Physiol Rev 78:227–245
Permutt MA, Nestorowicz A, Glaser B 1996 Familial hyperinsulinism: an inherited disorder of spontaneous hypoglycemia in neonates and infants. Diabetes Rev 4:347–355
Seino S, Inagaki N, Namba N, Gonoi T 1996 Molecular biology of
the b-cell ATP-sensitive K1 channel. Diabetes Rev 4:177–190
Aguilar-Bryan L, Nichols CG, Wechsler SW, Clement IV JP, Boyd
III AE, Gonzalez G, Herrera-Sosa H, Nguy K, Bryan J, Nelson DA
1995 Cloning of the b cell high-affinity sulfonylurea receptor: a
regulator of insulin secretion. Science 268:423– 426
Inagaki N, Gonoi T, Clement IV JP, Wang CZ, Aguilar-Bryan L,
Bryan J, Seino S 1996 A family of sulfonylurea receptors determines the pharmacological properties of ATP-sensitive K1 channels. Neuron 16:1011–1017
Isomoto S, Kondo C, Yamada M, Matsumoto S, Higashiguchi O,
Horio Y, Matsuzawa Y, Kurachi. Y 1996 A novel sulfonylurea
receptor forms with BIR (Kir6.2) a smooth muscle type ATP-sensitive K1 channel. J Biol Chem 271:24321–24324
Inagaki N, Inazawa J, Seino S 1995 cDNA sequence, gene structure, and chromosomal localization of the human ATP-sensitive
potassium channel, uKATP-1, gene (KCNJ8). Genomics 30:102–104
Inagaki N, Gonoi T, Clement IV JP, Namba N, Inazawa J, Gonzalez G, Aguilar-Bryan L, Seino S, Bryan J 1995 Reconstitution of
IKATP an inward rectifier subunit plus the sulfonylurea receptor.
Science 270:1166 –1170
Chutkow WA, Simon MC, Le Beau MM, Burant CF 1996 Cloning,
tissue expression, and chromosomal localization of SUR2, the putative drug-binding subunit of cardiac, skeletal muscle, and vascular KATP channels. Diabetes 45:1439 –1445
Babenko AP, Gonzalez G, Aguilar-Bryan L, Bryan J 1998 Reconstituted human cardiac KATP channels: Functional identity with
native channels in the sarcolemma of human ventricular cells. Circ
Res 83:1132–1143
Okuyama Y, Yamada M, Kondo C, Satoh E, Isomoto S, Shindo T,
Horio Y, Kitakaze M, Hori M, Kurachi Y 1998 The effects of
nucleotides and potassium channel openers on the SUR2A/Kir6.2
complex K1 channel expressed in a mammalian cell line, HEK293T
cells. Pflugers Arch 435:595– 603
Cook DL, Hales CN 1984 Intracellular ATP directly blocks K1
channels in pancreatic B-cells. Nature 311:271–273
Noma A 1983 ATP-regulated K1 channels in cardiac muscle. Nature 305:147–148
Trube G, Hescheler J 1984 Inward-rectifying channels in isolated
patches of the heart cell membrane: ATP-dependence and comparison with cell-attached patches. Pflugers Arch 401:178 –184
Trube G, Hescheler J 1983 Potassium channels in isolated patches
of cardiac cell membrane. Naunyn-Schmiedebergs Arch Pharmacol
322:R64
Doyle DA, Cabral JM, Pfuetzner RA, Kuo A, Gulbis JM, Cohen
SL, Chait BT, MacKinnon R 1998 The structure of the potassium
channel: molecular basis of K1 conduction and selectivity. Science
280:69 –77
Perozo E, Cortes DM, Cuello LG 1998 Three-dimensional archi-
128
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
AGUILAR-BRYAN AND BRYAN
tecture and gating mechanism of a K1 channel studied by EPR
spectroscopy. Nat Struct Biol 5:459 – 469
Cuello LG, Romero JG, Cortes DM, Perozo E 1998 pH-dependent
gating in the Streptomyces lividans K1 channel. Biochemistry 37:
3229 –3236
Tatulian SA, Cortes DM, Perozo E 1998 Structural dynamics of the
Streptomyces lividans K1 channel (SKC1): secondary structure characterization from FTIR spectroscopy. FEBS Lett 423:205–212
Kakei M, Noma A 1984 Adenosine-59-triphosphate-sensitive single potassium channel in the atrioventricular node cell of the rabbit
heart. J Physiol (Lond) 352:265–284
Qin DY, Takano M, Noma A 1989 Kinetics of ATP-sensitive K1
channel revealed with oil-gate concentration jump method. Am J
Physiol 257:H1624 –1633
Ashcroft FM, Kakei M 1989 ATP-sensitive K1 channels in rat
pancreatic b-cells: modulation by ATP and Mg21 ions. J Physiol
(Lond) 416:349 –367
Findlay I 1988 ATP42 and ATP.Mg inhibit the ATP-sensitive K1
channel of rat ventricular myocytes. Pflugers Arch 412:37– 41
Babenko AP, Aguilar-Bryan L, Bryan J 1998 A view of SUR/
KIR6.x, KATP channels. Annu Rev Physiol 60:667– 687
Dunne MJ, Petersen OH 1986 Intracellular ADP activates K1 channels that are inhibited by ATP in an insulin-secreting cell line. FEBS
Lett 208:59 – 62
Misler S, Falke LC, Gillis K, McDaniel ML 1986 A metaboliteregulated potassium channel in rat pancreatic B cells. Proc Natl
Acad Sci USA 83:7119 –7123
Lopatin AN, Makhina EN, Nichols CG 1995 Potassium channel
block by cytoplasmic polyamines as the mechanism of intrinsic
rectification. Nature 372:366 –369
Lopatin AN, Makhina EN, Nichols CG 1995 The mechanism of
inward rectification of potassium channels: “long-pore plugging”
by cytoplasmic polyamines. J Gen Physiol 106:923–955
Ficker E, Taglialatela M, Wible BA, Henley CM, Brown AM 1994
Spermine and spermidine as gating molecules for inward rectifier
K1 channels. Science 266:1068 –1072
Fakler B, Brandle U, Glowatzki E, Weidemann S, Zenner HP,
Ruppersberg JP 1995 Strong voltage-dependent inward rectification of inward rectifier K1 channels is caused by intracellular
spermine. Cell 80:149 –154
Nichols CG, Lopatin AN 1997 Inward rectifier potassium channels. Annu Rev Physiol 59:171–191
Nichols CG, Makhina EN, Pearson WL, Sha Q, Lopatin AN 1996
Inward rectification and implications for cardiac excitability. Circ
Res 78:1–7
Tucker SJ, Gribble FM, Zhao C, Trapp S, Ashcroft FM 1997
Truncation of Kir6.2 produces ATP-sensitive K1 channels in the
absence of the sulphonylurea receptor. Nature 387:179 –183
Gilon P, Shepherd RM, Henquin JC 1993 Oscillations of secretion
driven by oscillations of cytoplasmic Ca21 as evidences in single
pancreatic islets. J Biol Chem 268:22265–22268
Ashcroft FM, Rorsman P 1989 Electrophysiology of the pancreatic
b-cell. Prog Biophys Mol Biol 54:87–143
Smolen P, Keizer J 1992 Slow voltage inactivation of Ca21 currents
and bursting mechanisms for the mouse pancreatic b-cell. J Membr
Biol 127:9 –19
Keizer J, Smolen P 1991 Bursting electrical activity in pancreatic b
cells caused by Ca21- and voltage-inactivated Ca21 channels. Proc
Natl Acad Sci USA 88:3897–3901
Ding WG, He LP, Omatsu-Kanbe M, Kitasato H 1996 A possible
role of the ATP-sensitive potassium ion channel in determining the
duration of spike-bursts in mouse pancreatic b-cells. Biochim Biophys Acta 1279:219 –226
Cook DL, Satin LS, Hopkins WF 1991 Pancreatic B cells are bursting, but how? Trends Neurosci 14:411– 414
Satin LS, Smolen PD 1994 Electrical bursting in b-cells of the
pancreatic islets of Langerhans. Endocrine 2:677– 687
Sherman A 1996 Contributions of modeling to understanding stimulus-secretion coupling in pancreatic b-cells. Am J Physiol 271:
E362–372
Dukes ID, Roe MW, Worley III JF, Philipson LH 1997 Glucoseinduced alterations in b-cell cytoplasmic Ca21 involving coupling
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
Vol. 20, No. 2
of intracellular Ca21 stores and plasma membrane ion channels.
Curr Opin Endocrinol Diab 4:262–271
Sherman A, Smolen P 1997 Computer modeling of heterogeneous
b-cell populations. Adv Exp Med Biol 426:275–284
Jonas JC, Gilon P, Henquin JC 1998 Temporal and quantitative
correlations between insulin secretion and stably elevated or oscillatory cytoplasmic Ca21 in mouse pancreatic b-cells. Diabetes
47:1266 –1273
Barbosa RM, Silva AM, Tome AR, Stamford JA, Santos RM,
Rosario LM 1998 Control of pulsatile 5-HT/insulin secretion from
single mouse pancreatic islets by intracellular calcium dynamics.
J Physiol (Lond) 510:135–143
Byrne JH, Schultz SG 1994 An Introduction to Membrane Transport and Bioelectricity, ed 2. Raven Press, New York, p 198
Kemmler W, Loffler G 1977 NaK-ATPase in rat pancreatic islets.
Diabetologia 13:235–238
Hoenig M, Culberson LH, Clement JM, Ferguson DC 1992 Na21/
Ca21 exchange in plasma membrane vesicles from a glucoseresponsive insulinoma. Cell Calcium 13:1– 8
Ribalet B, Mirell CJ, Johnson DG, Levin SR 1996 Sulfonylurea
binding to a low-affinity site inhibits the Na/K-ATPase and the
KATP channel in insulin-secreting cells. J Gen Physiol 107:231–241
Richardson BP, Donatsch P, Lowe DA, Taylor P, Stocklin E 1981
Na21 channels in pancreatic islet cells. Postgrad Med J 57:98 –109
Richardson BP, Donatsch P, Hauser RE, Lowe DA, Marley PD,
Matthews EK 1980 Correlations between chanelled Na21 entry,
Ca21 fluxes and insulin release in pancreatic b-cells. Horm Metab
Res Suppl:81–91
Donatsch P, Lowe DA, Richardson BP, Taylor P 1977 The functional significance of sodium channels in pancreatic b-cell membranes. J Physiol (Lond) 267:357–376
Satin LS, Cook DL 1988 Evidence for two calcium currents in
insulin-secreting cells. Pflugers Arch 411:401– 409
Hiriart M, Matteson DR 1988 Na channels and two types of Ca
channels in rat pancreatic B cells identified with the reverse hemolytic plaque assay. J Gen Physiol 91:617– 639
Pressel DM, Misler S 1990 Sodium channels contribute to action
potential generation in canine and human pancreatic islet B cells.
J Membr Biol 116:273–280
Pressel DM, Misler S 1991 Role of voltage-dependent ionic currents in coupling glucose stimulation to insulin secretion in canine
pancreatic islet B-cells. J Membr Biol 124:239 –253
Philipson LH, Kusnetsov A, Larson T, Zeng Y, Westermark G
1993 Human, rodent, and canine pancreatic b-cells express a sodium channel a 1-subunit related to a fetal brain isoform. Diabetes
42:1372–1377
Plant TD 1988 Na21 currents in cultured mouse pancreatic B-cells.
Pflugers Arch 411:429 – 435
Sehlin J 1978 Interrelationship between chloride fluxes in pancreatic islets and insulin release. Am J Physiol 235:E501–508
Eddlestone GT, Beigelman PM 1983 Pancreatic b-cell electrical
activity: the role of anions and the control of pH. Am J Physiol
244:C188 –197
Kinard TA, Satin LS 1995 An ATP-sensitive Cl2 channel current
that is activated by cell swelling, cAMP, and glyburide in insulinsecreting cells. Diabetes 44:1461–1466
Kozak JA, Logothetis DE 1997 A calcium-dependent chloride current in insulin-secreting b TC-3 cells. Pflugers Arch 433:679 – 690
Findlay I, Ashcroft FM, Kelly RP, Rorsman P, Petersen OH, Trube
G 1989 Calcium currents in insulin-secreting b-cells. Ann NY Acad
Sci 560:403– 409
Rajan AS, Aguilar-Bryan L, Nelson DA, Yaney GC, Hsu WH,
Kunze DL, Boyd III AE 1990 Ion channels and insulin secretion.
Diabetes Care 13:340 –363
Boyd III AE 1992 The role of ion channels in insulin secretion. J Cell
Biochem 48:235–241
Misler S, Barnett DW, Gillis KD, Pressel DM 1992 Electrophysiology of stimulus-secretion coupling in human b-cells. Diabetes
41:1221–1228
Petit P, Loubatieres-Mariani MM 1992 Potassium channels of the
insulin-secreting b cell. Fundam Clin Pharmacol 6:123–134
Ashcroft FM, Proks P, Smith PA, Ammala C, Bokvist K,
April, 1999
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
Rorsman P 1994 Stimulus-secretion coupling in pancreatic b cells.
J Cell Biochem 55:54 – 65
Dukes ID, Philipson LH 1996 K1 channels: generating excitement
in pancreatic b-cells. Diabetes 45:845– 853
Atwater I, Mears D, Rojas E1996 Electrophysiology of the pancreatic b-cell. In: LeRoith D, Taylor SI, Olefsky JM (eds) Diabetes
Mellitus. Lippencott-Raven Publishers, Philadelphia, pp 78 –102
Ribalet B, Beigelman PM 1980 Calcium action potentials and potassium permeability activation in pancreatic b-cells. Am J Physiol
239:C124 –133
Atwater I, Dawson CM, Eddlestone GT, Rojas E 1981 Voltage
noise measurements across the pancreatic b-cell membrane: calcium channel characteristics. J Physiol (Lond) 314:195–212
Rorsman P, Trube G 1986 Calcium and delayed potassium currents in mouse pancreatic b-cells under voltage-clamp conditions.
J Physiol (Lond) 374:531–550
Satin LS, Cook DL 1989 Calcium current inactivation in insulinsecreting cells is mediated by calcium influx and membrane depolarization. Pflugers Arch 414:1–10
Hopkins WF, Satin LS, Cook DL 1991 Inactivation kinetics and
pharmacology distinguish two calcium currents in mouse pancreatic B-cells. J Membr Biol 119:229 –239
Sala S, Matteson DR 1990 Single-channel recordings of two types
of calcium channels in rat pancreatic b-cells. Biophys J 58:567–571
Davalli AM, Biancardi E, Pollo A, Socci C, Pontiroli AE, Pozza G,
Clementi FSE, Carbone E 1996 Dihydropyridine-sensitive and -insensitive voltage-operated calcium channels participate in the control of glucose-induced insulin release from human pancreatic b
cells. J Endocrinol 150:195–203
Bhattacharjee A, Whitehurst Jr RM, Zhang M, Wang L, Li M 1997
T-type calcium channels facilitate insulin secretion by enhancing
general excitability in the insulin-secreting b-cell line, INS-1. Endocrinology 138:3735–3740
Cribbs LL, Lee JH, Yang J, Satin J, Zhang Y, Daud A, Barclay J,
Williamson MP, Fox M, Rees M, Perez-Reyes E 1998 Cloning and
characterization of a1H from human heart, a member of the T-type
Ca21 channel gene family. Circ Res 83:103–109
Perez-Reyes E, Cribbs LL, Daud A, Lacerda AE, Barclay J,
Williamson MP, Fox M, Rees M, Lee JH 1998 Molecular characterization of a neuronal low-voltage-activated T-type calcium channel. Nature 391:896 –900
Rojas E, Hidalgo J, Carroll PB, Li MX, Atwater I 1990 A new class
of calcium channels activated by glucose in human pancreatic
b-cells. FEBS Lett 261:265–270
Mears D, Sheppard Jr NF, Atwater I, Rojas E, Bertram R, Sherman
A 1997 Evidence that calcium release-activated current mediates
the biphasic electrical activity of mouse pancreatic b-cells. J Membr
Biol 155:47–59
Roe MW, Worley III JF, Qian F, Mittal AA, Dralyuk F, Blair NT,
Mertz RJ, Philipson LH, Dukes ID 1998 Characterization of a Ca21
release-activated nonselective cation current (ICRAN) regulating
membrane potential and [Ca21]i oscillations on transgenically derived b-cells. J Biol Chem 273:10402–10410
Bordin S, Boschero AC, Carneiro EM, Atwater I 1995 Ionic mechanisms involved in the regulation of insulin secretion by muscarinic
agonists. J Membr Biol 148:177–184
Silva AM, Rosario LM, Santos RM 1994 Background Ca21 influx
mediated by a dihydropyridine- and voltage-insensitive channel in
pancreatic b-cells. Modulation by Ni21, diphenylamine-2-carboxylate, and glucose metabolism. J Biol Chem 269:17095–17103
Leech CA, Holz GG, Habener JF 1994 Voltage-independent calcium channels mediate slow oscillations of cytosolic calcium that
are glucose dependent in pancreatic b-cells. Endocrinology 135:
365–372
Bertram R, Smolen P, Sherman A, Mears D, Atwater I, Martin F,
Soria B 1995 A role for calcium release-activated current (CRAC)
in cholinergic modulation of electrical activity in pancreatic b-cells.
Biophys J 68:2323–2332
Lewis RS, Cahalan MD 1989 Mitogen-induced oscillations of cytosolic Ca21 and transmembrane Ca21 current in human leukemic
T cells. Cell Regul 1:99 –112
Hoth M, Penner R 1992 Depletion of intracellular calcium stores
activates a calcium current in mast cells. Nature 355:353–356
129
108. Clapham DE 1996 TRP is cracked but is CRAC TRP? Neuron
16:1069 –1072
109. Legrand AM, Bagnis R 1984 Effects of highly purified maitotoxin
extracted from dinoflagellate Gambierdiscus toxicus on action potential of isolated rat heart. J Mol Cell Cardiol 16:663– 666
110. Takahashi M, Ohizumi Y, Yasumoto T 1982 Maitotoxin, a Ca21
channel activator candidate. J Biol Chem 257:7287–7289
111. Gusovsky F, Daly JW 1990 Maitotoxin: a unique pharmacological
tool for research on calcium-dependent mechanisms. Biochem
Pharmacol 39:1633–1639
112. Lytton J, Westlin M, Hanley MR 1991 Thapsigargin inhibits the
sarcoplasmic or endoplasmic reticulum Ca-ATPase family of calcium pumps. J Biol Chem 266:17067–17071
113. Gericke M, Droogmans G, Nilius B 1993 Thapsigargin discharges
intracellular calcium stores and induces transmembrane currents in
human endothelial cells. Pflugers Arch 422:552–557
114. Roe MW, Mertz RJ, Lancaster ME, Worley III JF, Dukes ID 1994
Thapsigargin inhibits the glucose-induced decrease of intracellular
Ca21 in mouse islets of Langerhans. Am J Physiol 266:E852– 862
115. Worley III JF, McIntyre MS, Spencer B, Dukes ID 1994 Depletion
of intracellular Ca21 stores activates a maitotoxin-sensitive nonselective cationic current in b-cells. J Biol Chem 269:32055–32058
116. Cook DL, Ikeuchi M 1989 Tolbutamide as mimic of glucose on
b-cell electrical activity. ATP-sensitive K1 channels as common
pathway for both stimuli. Diabetes 38:416 – 421
117. Misler S, Barnett DW, Pressel DM, Gillis KD, Scharp DW, Falke
LC 1992 Stimulus-secretion coupling in b-cells of transplantable
human islets of Langerhans. Evidence for a critical role for Ca21
entry. Diabetes 41:662– 670
118. Smith PA, Ashcroft FM, Rorsman P 1990 Simultaneous recordings
of glucose dependent electrical activity and ATP-regulated K1currents in isolated mouse pancreatic b-cells. FEBS Lett 261:187–190
119. Larsson O, Kindmark H, Brandstrom R, Fredholm B, Berggren
PO 1996 Oscillations in KATP channel activity promote oscillations
in cytoplasmic free Ca21 concentration in the pancreatic b cell. Proc
Natl Acad Sci USA 93:5161–5165
120. Rosario LM, Barbosa RM, Antunes CM, Silva AM, Abrunhosa AJ,
Santos RM 1993 Bursting electrical activity in pancreatic b-cells:
evidence that the channel underlying the burst is sensitive to Ca21
influx through L-type Ca21 channels. Pflugers Arch 424:439 – 447
121. Straub SG, James RF, Dunne MJ, Sharp GW 1998 Glucose activates both KATP channel-dependent and KATP channel-independent signaling pathways in human islets. Diabetes 47:758 –763
122. Komatsu M, Noda M, Sharp GW 1998 Nutrient augmentation of
Ca21-dependent and Ca21-independent pathways in stimuluscoupling to insulin secretion can be distinguished by their
guanosine triphosphate requirements: studies on rat pancreatic
islets. Endocrinology 139:1172–1183
123. Sato Y, Nenquin M, Henquin JC 1998 Relative contribution of
Ca21-dependent and Ca21-independent mechanisms to the regulation of insulin secretion by glucose. FEBS Lett 421:115–119
124. Rorsman P, Trube G 1985 Glucose dependent K1-channels in
pancreatic b-cells are regulated by intracellular ATP. Pflugers Arch
405:305–309
125. Matschinsky FM, Pagliara AS, Stillings SN, Hover BA 1976 Glucose and ATP levels in pancreatic islet tissue of normal and diabetic
rats. J Clin Invest 58:1193–1200
126. Ghosh A, Ronner P, Cheong E, Khalid P, Matschinsky FM 1991
The role of ATP and free ADP in metabolic coupling during fuelstimulated insulin release from islet b-cells in the isolated perfused
rat pancreas. J Biol Chem 266:22887–22892
127. Detimary P, Jonas JC, Henquin JC 1996 Stable and diffusible pools
of nucleotides in pancreatic islet cells. Endocrinology 137:4671–
4676
128. Hopkins WF, Fatherazi S, Peter-Riesch B, Corkey BE, Cook DL
1992 Two sites for adenine-nucleotide regulation of ATP-sensitive
potassium channels in mouse pancreatic b-cells and HIT cells. J
Membr Biol 129:287–295
129. Keizer J, Magnus G 1989 ATP-sensitive potassium channel and
bursting in the pancreatic b cell. A theoretical study. Biophys J
56:229 –242
130. Weiss JN, Venkatesh N 1993 Metabolic regulation of cardiac ATPsensitive K1 channels. Cardiovasc Drugs Ther 7:499 –505
130
AGUILAR-BRYAN AND BRYAN
131. Detimary P, Gilon P, Henquin JC 1998 Interplay between cytoplasmic Ca21 and the ATP/ADP ratio: a feedback control mechanism in mouse pancreatic islets. Biochem J 333:269 –274
132. Findlay I, Dunne MJ, Petersen OH 1985 ATP-sensitive inward
rectifier and voltage- and calcium-activated K1 channels in cultured pancreatic islet cells. J Membr Biol 88:165–172
133. Ribalet B, Ciani S 1987 Regulation by cell metabolism and adenine
nucleotides of a K channel in insulin secreting b-cells (RINm5F).
Proc Natl Acad Sci USA 84:7119 –7123
134. Rajan AS, Aguilar-Bryan L, Nelson DA, Nichols CG, Wechsler
SW, Lechago J, Bryan J 1993 Sulfonylurea receptors and ATPsensitive K1 channels in clonal pancreatic a cells. Evidence for two
high affinity sulfonylurea receptors. J Biol Chem 268:15221–15228
135. Kubo Y, Baldwin TJ, Jan YN, Jan LY 1993 Primary structure and
functional expression of a mouse inward rectifier potassium channel. Nature 362:127–133
136. Dascal N, Schreibmayer W, Lim NF, Wang W, Chavkin C,
DiMagno L, Labarca C, Kieffer BL, Gaveriaux-Ruff C, Trollinger
D, Lester HA, Davidson N 1993 Atrial G protein-activated K1
channel: Expression cloning and moleular properties. Proc Natl
Acad Sci USA 90:10235–10239
137. Kubo Y, Reuveny E, Slesinger PA, Jan YN, Jan LY 1993 Primary
structure and functional expression of a rat G-protein-coupled
muscarinic potassium channel. Nature 364:802– 806
138. Jan LY, Jan YN 1992 Structural elements involved in specific K1
channel functions. Annu Rev Physiol 54:537–555
139. Pongs O 1993 Structure-function studies on the pore of potassium
channels. J Membr Biol 136:1– 8
140. Abola EE, Sussman JL, Prilusky J, Manning NO 1997 Protein data
bank archives of three-dimensional macromolecular structures.
Methods Enzymol 277:556 –571
141. Abola EE, Bernstein FC, Bryant SH, Koetzle TF, Weng J 1987
Protein data bank. In: Allen FH, Bergerhoff G, Sievers R (eds)
Crystallographic Databases — Information Content Software Systems Scientific Applications. Data Commission of the International
Union of Crystallography, Bonn/Cambridge/Chester, pp 107–132
142. Chandy KG, Gutman GA 1993 Nomenclature for mammalian
potassium channel genes. Trends Pharmacol Sci 14:434
143. Doupnik CA, Davidson N, Lester HA 1995 The inward rectifier
potassium channel family. Curr Opin Neurobiol 5:268 –277
144. Inagaki N, Tsuura Y, Namba N, Masuda K, Gonoi T, Horie M,
Seino Y, Mizuta M, Seino S 1995 Cloning and functional characterization of a novel ATP-sensitive potassium channel ubiquitously
expressed in rat tissues, including pancreatic islets, pituitary, skeletal muscle, and heart. J Biol Chem 270:5691–5694
145. Gribble FM, Ashfield R, Ammala C, Ashcroft FM 1997 Properties
of cloned ATP-sensitive K1 currents expressed in Xenopus oocytes.
J Physiol (Lond) 498:87–98
146. Clement IV JP, Kunjilwar K, Gonzalez G, Schwanstecher M,
Panten U, Aguilar-Bryan L, Bryan J 1997 Association and stoichiometry of KATP channel subunits. Neuron 18:827– 838
147. Ashford ML, Bond CT, Blair TA, Adelman JP 1994 Cloning and
functional expression of a rat heart KATP channel. Nature 370:
456 – 459
148. Kaubisch N, Hammer R, Wollheim C, Renold AE, Offord RE 1982
Specific receptors for sulfonylureas in brain and in a b-cell tumor
of the rat. Biochem Pharmacol 31:1171–1174
149. Geisen K, Hitzel V, Okomonopoulos R, Punter J, Weyer R, Summ
HD 1985 Inhibition of 3H-glibenclamide binding to sulfonylurea
receptors by oral antidiabetics. Arzneimittelforschung 35:707–712
150. Gaines KL, Hamilton S, Boyd III AE 1988 Characterization of the
sulfonylurea receptor on b cell membranes. J Biol Chem 263:2589 –
2592
151. Kramer W, Oekonomopulos R, Punter J, Summ HD 1988 Direct
photoaffinity labeling of the putative sulfonylurea receptor in rat
b-cell tumor membranes by [3H]glibenclamide. FEBS Lett 229:
355–359
152. Aguilar-Bryan L, Nelson DA, Vu QA, Humphrey MB, Boyd III
AE 1990 Photoaffinity labeling and partial purification of the b cell
sulfonylurea receptor using a novel, biologically active glyburide
analog. J Biol Chem 265:8218 – 8224
153. Schwanstecher M, Behrends S, Brandt C, Panten U 1992 The
binding properties of the solubilized sulfonylurea receptor from a
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
173.
Vol. 20, No. 2
pancreatic B-cell line are modulated by the Mg11-complex of ATP.
J Pharmacol Exp Ther 262:495–502
Nelson DA, Aguilar-Bryan L, Bryan J 1992 Specificity of photolabeling of b-cell membrane proteins with an 125I-labeled glyburide
analog. J Biol Chem 267:14928 –14933
Santerre RF, Cook RA, Crisel RM, Sharp JD, Schmidt RJ,
Williams DC, Wilson CP 1981 Insulin synthesis in a clonal cell line
of simian virus 40-transformed hamster pancreatic b cells. Proc
Natl Acad Sci USA 78:4339 – 4343
Ronner P, Matschinsky FM, Hang TL, Epstein AJ, Buettger C 1993
Sulfonylurea-binding sites and ATP-sensitive K1 channels in a-TC
glucagonoma and b-TC insulinoma cells. Diabetes 42:1760 –1772
Fosset M, De Weille JR, Green RD, Schmid-Antomarchi H,
Lazdunski M 1988 Antidiabetic sulfonylureas control action potential properties in heart cells via high affinity receptors that are
linked to ATP-dependent K1 channels. J Biol Chem 263:7933–7936
Gopalakrishnan M, Johnson DE, Janis RA, Triggle DJ 1991 Characterization of binding of the ATP-sensitive potassium channel
ligand, [3H]glyburide, to neuronal and muscle preparations. J Pharmacol Exp Ther 257:1162–1171
Lupo B, Bataille D 1987 A binding site for [3H]glipizide in the rat
cerebral cortex. Eur J Pharmacol 140:157–169
Schwanstecher M, Schaupp U, Loser S, Panten U 1992 The binding
properties of the particulate and solubilized sulfonylurea receptor
from cerebral cortex are modulated by the Mg21 complex of ATP.
J Neurochem 59:1325–1335
Bernardi H, Fosset M, Lazdunski M 1988 Characterization, purification, and affinity labeling of the brain [3H]glibenclamide-binding protein, a putative neuronal ATP-regulated K1 channel. Proc
Natl Acad Sci USA 85:9816 –9820
Schwanstecher M, Loser S, Chudziak F, Bachmann C, Panten U
1994 Photoaffinity labeling of the cerebral sulfonylurea receptor
using a novel radioiodinated azidoglibenclamide analogue. J Neurochem 63:698 –708
Schwanstecher M, Loser S, Chudziak F, Panten U 1994 Identification of a 38-kDa high affinity sulfonylurea-binding peptide in
insulin-secreting cells and cerebral cortex. J Biol Chem 269:17768 –
17771
Nelson DA, Bryan J, Wechsler S, Clement IV JP, Aguilar-Bryan
L 1996 The high-affinity sulfonylurea receptor: distribution, glycosylation, purification, and immunoprecipitation of two forms
from endocrine and neuroendocrine cell lines. Biochemistry 35:
14793–14799
Ozanne SE, Guest PC, Hutton JC, Hales CN 1995 Intracellular
localization and molecular heterogeneity of the sulphonylurea receptor in insulin-secreting cells. Diabetologia 38:277–282
Eliasson L, Renstrom E, Ammala C, Berggren PO, Bertorello AM,
Bokvist K, Chibalin A, Deeney JT, Flatt PR, Gabel J, Gromada J,
Larsson O, Lindstrom P, Rhodes CJ, Rorsman P 1996 PKCdependent stimulation of exocytosis by sulfonylureas in pancreatic
b cells. Science 271:813– 815
Sturgess NC, Ashford ML, Cook DL, Hales CN 1985 The sulphonylurea receptor may be an ATP-sensitive potassium channel. Lancet 2:474 – 475
Khan RN, Hales CN, Ozanne SE, Adogu AA, Ashford ML 1993
Dissociation of KATP channel and sulphonylurea receptor in the
rat clonal insulin-secreting cell line, CRI-D11. Proc R Soc Lond B
Biol Sci 253:225–231
Aguilar-Bryan L, Nichols CG, Rajan AS, Parker C, Bryan J 1992
Co-expression of sulfonylurea receptors and KATP channels in hamster insulinoma tumor (HIT) cells. Evidence for direct association
of the receptor with the channel. J Biol Chem 267:14934 –14940
Zhang HJ, Walseth TF, Robertson RP 1989 Insulin secretion and
cAMP metabolism in HIT cells. Reciprocal and serial passagedependent relationships. Diabetes 38:44 – 48
Aguilar-Bryan L, Clement IV JP, Nelson DA 1998 Sulfonylurea
receptors and ATP-sensitive potassium channels. Methods Enzymol 292:732–744
Mikhailov MV, Proks P, Ashcroft FM, Ashcroft SJ 1998 Expression of functionally active ATP-sensitive K-channels in insect cells
using baculovirus. FEBS Lett 429:390 –394
Altschul SF, Lipman DJ 1990 Protein database searches for multiple alignments. Proc Natl Acad Sci USA 87:5509 –5513
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
174. Gottesman MM, Pastan I 1993 Biochemistry of multidrug resistance mediated by the multidrug transporter. Annu Rev Biochem
62:385– 427
175. Cole SP, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE, Almquist
KC, Stewart AJ, Kurz EU, Duncan AM, Deeley RG 1992 Overexpression of a transporter gene in a multidrug-resistant human
lung cancer cell line. Science 258:1650 –1654
176. Riordan JR, Rommens JM, Kerem B, Alon N, Rozmahel R,
Grzelczak Z, Zielenski J, Lok S, Plavsic N, Chou JL, Drumm ML,
Iannuzzi MC, Collins FS, Tsui LC 1989 Identification of the cystic
fibrosis gene: cloning and characterization of complementary
DNA. Science 245:1066 –1073
177. Buchler M, Konig J, Brom M, Kartenbeck J, Spring H, Horie T,
Keppler D 1996 cDNA cloning of the hepatocyte canalicular isoform of the multidrug resistance protein, cMrp, reveals a novel
conjugate export pump deficient in hyperbilirubinemic mutant
rats. J Biol Chem 271:15091–15098
178. Taniguchi K, Wada M, Kohno K, Nakamura T, Kawabe T,
Kawakami M, Kagotani K, Okumura K, Akiyama S, Kuwano M
1996 A human canalicular multispecific organic anion transporter
(cMOAT) gene is overexpressed in cisplatin-resistant human cancer cell lines with decreased drug accumulation. Cancer Res 56:
4124 – 4129
179. Li ZS, Szczypka M, Lu YP, Thiele DJ, Rea PA 1996 The yeast
cadmium factor protein (YCF1) is a vacuolar glutathione S-conjugate pump. J Biol Chem 271:6509 – 6517
180. Tusnady GE, Bakos E, Varadi A, Sarkadi B 1997 Membrane topology distinguishes a subfamily of the ATP-binding cassette
(ABC) transporters. FEBS Lett 402:1–3
181. Walker JE, Saraste M, Runswick MJ, Gay NJ 1982 Distantly related sequences in the a- and b-subunits of ATP synthase, myosin,
kinases and other ATP-requiring enzymes and a common nucleotide binding fold. EMBO J 1:945–951
182. Manavalan P, Dearborn DG, McPherson JM, Smith AE 1995 Sequence homologies between nucleotide binding regions of CFTR
and G-proteins suggest structural and functional similarities. FEBS
Lett 366:87–91
183. Hilgenfeld R 1995 Regulatory GTPases. Curr Opin Struct Biol
5:810 – 817
184. Hilgenfeld R 1995 How do the GTPases really work? Nature Struct
Biol 2:3– 6
185. Armstrong SR, Tabernero L, Zhang H, Hermodson M, Stauffacher CV 1998 The 2.5A structure of the N-terminal ATP-binding
cassette of the ribose ABC-transporter. Biophys J 74:A338
186. Chang XB, Hou YX, Riordan JR 1997 ATPase activity of purified
multidrug resistance-associated protein. J Biol Chem 272:30962–
30968
187. Chang XB, Hou YX, Riordan JR 1998 Stimulation of ATPase activity of purified multidrug resistance-associated protein by nucleoside diphosphates. J Biol Chem 273:23844 –23848
188. Eisenberg D, Schwarz E, Komaromy M, Wall R 1984 Analysis of
membrane and surface protein sequences with the hydrophobic
moment plot. J Mol Biol 179:125–142
189. Nichols CG, Shyng SL, Nestorowicz A, Glaser B, Clement IV JP,
Gonzalez G, Aguilar-Bryan L, Permutt MA, Bryan J 1996 Adenosine diphosphate as an intracellular regulator of insulin secretion.
Science 272:1785–1787
190. Gribble FM, Tucker SJ, Ashcroft FM 1997 The essential role of the
Walker A motifs of SUR1 in K-ATP channel activation by Mg-ADP
and diazoxide. EMBO J 16:1145–1152
191. Sakura H, Ammala C, Smith PA, Gribble FM, Ashcroft FM 1995
Cloning and functional expression of the cDNA encoding a novel
ATP-sensitive potassium channel subunit expressed in pancreatic
b-cells, brain, heart and skeletal muscle. FEBS Lett 377:338 –344
192. de Weille J, Schmid-Antomarchi H, Fosset M, Lazdunski M 1988
ATP-sensitive K1 channels that are blocked by hypoglycemiainducing sulfonylureas in insulin-secreting cells are activated by
galanin, a hyperglycemia-inducing hormone. Proc Natl Acad Sci
USA 85:1312–1316
193. Ahren B, Berggren PO, Bokvist K, Rorsman P 1989 Does galanin
inhibit insulin secretion by opening of the ATP-regulated K1 channel in the b-cell? Peptides 10:453– 457
194. de Weille JR, Schmid AH, Fosset M, Lazdunski M 1989 Regulation
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
211.
212.
213.
214.
215.
216.
217.
218.
131
of ATP-sensitive K1 channels in insulinoma cells: activation by
somatostatin and protein kinase C and the role of cAMP. Proc Natl
Acad Sci USA 86:2971–2975
Leech CA, Holz GG, Habener JF 1996 Signal transduction of
PACAP and GLP-1 in pancreatic b cells. Ann NY Acad Sci 805:
81–92
Nelson MT, Huang Y, Brayden JE, Hescheler J, Standen NB 1990
Arterial dilations in response to calcitonin gene-related peptide
involve activation of K1 channels. Nature 344:770 –773
Kieffer TJ, Heller RS, Leech CA, Holz GG, Habener JF 1997 Leptin
suppression of insulin secretion by activation of ATP-sensitive K1
channels in pancreatic b-cells. Diabetes 46:1087–1093
Spanswick D, Smith MA, Groppi VE, Logan SD, Ashford MLJ
1997 Leptin inhibits Hypothalamic neurons by activation of ATPsensitive potassium channels. Nature 390:521–525
Harvey J, Ashford ML 1998 Diazoxide- and leptin-activated KATP
currents exhibit differential sensitivity to englitazone and ciclazindol in the rat CRI-G1 insulin-secreting cell line. Br J Pharmacol
124:1557–1565
Harvey J, Ashford ML 1998 Insulin occludes leptin activation of
ATP-sensitive K1 channels in rat CRI-G1 insulin secreting cells.
J Physiol (Lond) 511:695–706
Harvey J, Ashford ML 1998 Role of tyrosine phosphorylation in
leptin activation of ATP-sensitive K1 channels in the rat insulinoma
cell line CRI-G1. J Physiol (Lond) 510:47– 61
Ammala C, Moorhouse A, Gribble F, Ashfield R, Proks P, Smith
PA, Sakura H, Coles B, Ashcroft SJ, Ashcroft FM 1996 Promiscuous coupling between the sulphonylurea receptor and inwardly
rectifying potassium channels. Nature 379:545–548
Wellman GC, Quayle JM, Standen NB 1996 Evidence against the
association of the sulphonylurea receptor with endogenous Kir
family members other than KATP in coronary vascular smooth
muscle. Pflugers Arch 432:355–357
Shyng S, Ferrigni T, Nichols CG 1997 Control of rectification and
gating of cloned KATP channels by the Kir6.2 subunit. J Gen Physiol
110:141–153
Al-Awqati Q 1995 Regulation of ion channels by ABC transporters
that secrete ATP (Perspective). Science 269:805– 806
Schwiebert EM, Egan ME, Hwang TH, Fulmer SB, Allen SS,
Cutting GR, Guggino WB 1995 CFTR regulates outwardly rectifying chloride channels through an autocrine mechanism involving
ATP. Cell 81:1063–1073
Jovov B, Ismailov II, Berdiev BK, Fuller CM, Sorscher EJ,
Dedman JR, Kaetzel MA, Benos DJ 1995 Interaction between
cystic fibrosis transmembrane conductance regulator and outwardly rectified chloride channels. J Biol Chem 270:29194 –29200
Glowatzki E, Fakler G, Brandle U, Rexhausen U, Zenner HP,
Ruppersberg JP, Fakler B 1995 Subunit-dependent assembly of
inward-rectifier K1 channels. Proc R Soc Lond B Biol Sci 261:
251–261
Yang J, Jan YN, Jan LY 1995 Determination of the subunit stoichiometry of an inwardly rectifying potassium channel. Neuron
15:1441–1447
MacKinnon R 1991 Determination of the subunit stoichiometry of
a voltage-activated potassium channel. Nature 350:232–235
Liman ER, Tytgat J, Hess P 1992 Subunit stoichiometry of a mammalian K1 channel determined by construction of multimeric cDNAs. Neuron 9:861– 871
Li M, Unwin N, Stauffer KA, Jan YN, Jan LY 1994 Images of
purified Shaker potassium channels. Curr Biol 4:110 –115
Shyng S, Nichols CG 1997 Octameric stoichiometry of the KATP
channel complex. J Gen Physiol 110:655– 664
Inagaki N, Gonoi T, Seino S 1997 Subunit stoichiometry of the
pancreatic b-cell ATP-sensitive K1 channel. FEBS Lett 409:232–236
Ashcroft FM, Harrison DE, Ashcroft SJ 1984 Glucose induces
closure of single potassium channels in isolated rat pancreatic
b-cells. Nature 312:446 – 448
Matschinsky FM 1996 Banting Lecture 1995. A lesson in metabolic
regulation inspired by the glucokinase glucose sensor paradigm.
Diabetes 45:223–241
Efrat S, Tal M, Lodish HF 1994 The pancreatic b-cell glucose
sensor. Trends Biochem Sci 19:535–538
Olesen SP, Bundgaard M 1993 ATP-dependent closure and reac-
132
219.
220.
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
241.
AGUILAR-BRYAN AND BRYAN
tivation of inward rectifier K1 channels in endothelial cells. Circ
Res 73:492– 495
Light P 1996 Regulation of ATP-sensitive potassium channels by
phosphorylation. Biochim Biophys Acta 1286:65–73
de Weille JR, Muller M, Lazdunski M 1992 Activation and inhibition of ATP-sensitive K1 channels by fluorescein derivatives.
J Biol Chem 267:4557– 4563
Hussain M, Wareham AC 1994 Rundown and reactivation of ATPsensitive potassium channels (KATP) in mouse skeletal muscle. J
Membr Biol 141:257–265
Furukawa T, Virag L, Furukawa N, Sawanobori T, Hiraoka M
1994 Mechanism for reactivation of the ATP-sensitive K1 channel
by MgATP complexes in guinea-pig ventricular myocytes. J Physiol
(Lond) 479:95–107
Furukawa T, Yamane T, Terai T, Katayama Y, Hiraoka M 1996
Functional linkage of the cardiac ATP-sensitive K1 channel to the
actin cytoskeleton. Pflugers Arch 431:504 –512
Hilgemann DW, Ball R 1996 Regulation of cardiac Na1,Ca21 exchange and KATP potassium channels by PIP2. Science 273:956 –959
Fan Z, Makielski JC 1997 Anionic phospholipids activate ATPsensitive potassium channels. J Biol Chem 272:5388 –5395
Ribalet B, Ciani S, Eddlestone GT 1989 ATP mediates both activation and inhibition of KATP channel activity via cAMP-dependent
protein kinase in insulin-secreting cell lines. J Gen Physiol 94:
693–717
Gerbitz KD, Gempel K, Brdiczka D 1996 Mitochondria and diabetes. Genetic, biochemical, and clinical implications of the cellular
energy circuit. Diabetes 45:113–126
Alvarez-Leefmans FJ, Giraldez F, Gamino SM 1987 Intracellular
free magnesium in excitable cells: its measurement and its biologic
significance. Can J Physiol Pharmacol 65:915–925
Murphy E 1993 Measurement of intracellular ionized magnesium.
Miner Electrolyte Metab 19:250 –258
Cook DL, Satin LS, Ashford ML, Hales CN 1988 ATP-sensitive K1
channels in pancreatic b-cells. Spare-channel hypothesis. Diabetes
37:495– 498
Nichols CG, Lederer WJ 1991 Adenosine triphosphate-sensitive
potassium channels in the cardiovascular system. Am J Physiol
261:H1675–H1686
Terzic A, Findlay I, Hosoya Y, Kurachi Y 1994 Dualistic behavior
of ATP-sensitive K1 channels toward intracellular nucleoside
diphosphates. Neuron 12:1049 –1058
Olson LK, Schroeder W, Robertson RP, Goldberg ND, Walseth
TF 1996 Suppression of adenylate kinase catalyzed phosphotransfer precedes and is associated with glucose-induced insulin secretion in intact HIT-T15 cells. J Biol Chem 271:16544 –16552
Weiss JN, Lamp ST 1987 Glycolysis preferentially inhibits ATPsensitive K1 channels in isolated guinea pig cardiac myocytes.
Science 238:67– 69
Detimary P, Jonas JC, Henquin JC 1995 Possible links between
glucose-induced changes in the energy state of pancreatic B cells
and insulin release. Unmasking by decreasing a stable pool of
adenine nucleotides in mouse islets. J Clin Invest 96:1738 –1745
Ohno-Shosaku T, Zunkler BJ, Trube G 1987 Dual effects of ATP
on K1 currents of mouse pancreatic b-cells. Pflugers Arch 408:
133–138
Schmid-Antomarchi H, De Weille J, Fosset M, Lazdunski M 1987
The receptor for antidiabetic sulfonylureas controls the activity of
the ATP-modulated K1 channel in insulin-secreting cells [published erratum appears in J Biol Chem 1989 264:10926]. J Biol Chem
262:15840 –15844
Shyng S, Ferrigni T, Nichols CG 1997 Regulation of KATP channel
activity by diazoxide and MgADP. Distinct functions of the two
nucleotide binding folds of the sulfonylurea receptor. J Gen Physiol
110:643– 654
Ueda K, Inagaki N, Seino S 1997 MgADP antagonism to Mg21independent ATP binding of the sulfonylurea receptor SUR1. J Biol
Chem 272:22983–22986
Babenko A, Gonzalez G, Aguilar-Bryan L, Bryan J 1999 Sulfonylurea receptors set the maximal open probability, ATP-sensitivity and plasma membrane density of KATP channels. FEBS Lett
445:131–136
John SA, Monck JR, Weiss JN, Ribalet B 1998 The sulphonylurea
242.
243.
244.
245.
246.
247.
248.
249.
250.
251.
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
262.
263.
264.
Vol. 20, No. 2
receptor SUR1 regulates ATP-sensitive mouse Kir6.2 K1 channels
linked to the green fluorescent protein in human embryonic kidney
cells (HEK 293). J Physiol (Lond) 510:333–345
Schwappach B, Zerangue N, Jan YN, Jan LY, The last 36 amino
acids of Kir6.2 block surface expression. Program of the 2nd International Conference on ATP-Sensitive Potassium Channels and
Disease, 1998, St. Charles, IL (Abstract 232)
Gehlert DR, Robertson DW 1994 ATP sensitive potassium channels: potential drug targets in neuropsychopharmacology. Prog
Neuropsychopharmacol Biol Psychiatry 18:1093–1102
Davis NW, Standen NB, Stanfield PR 1991 ATP-dependent potassium channels of muscle cells: their properties, regulation, and
possible functions. J Bioenerg Biomembr 23:509 –535
Weston AH, Edwards G 1991 Latest developments in K-channel
modulator pharmacology. Z Kardiol 80:1– 8
Weston AH, Longmore J, Newgreen DT, Edwards G, Bray KM,
Duty S 1990 The potassium channel openers: a new class of
vasorelaxants. Blood Vessels 27:306 –313
Quayle JM, Standen NB 1994 KATP channels in vascular smooth
muscle. Cardiovasc Res 28:797– 804
Quast U, Guillon JM, Cavero I 1994 Cellular pharmacology of
potassium channel openers in vascular smooth muscle. Cardiovasc
Res 28:805– 810
Quast U 1992 Potassium channel openers: pharmacological and
clinical aspects. Fundam Clin Pharmacol 6:279 –293
Grover GJ 1994 Protective effects of ATP-sensitive potassiumchannel openers in experimental myocardial ischemia. J Cardiovasc Pharmacol 24:S18 –27
Edwards G, Weston AH 1995 Pharmacology of the potassium
channel openers. Cardiovasc Drugs Ther 9:185–193
Edwards G, Weston AH 1990 Potassium channel openers and
vascular smooth muscle relaxation. Pharmacol Ther 48:237–258
Challinor-Rogers JL, McPherson GA 1994 Potassium channel
openers and other regulators of KATP channels. Clin Exp Pharmacol
Physiol 21:583–597
Yamada M, Isomoto S, Matsumoto S, Kondo C, Shindo T, Horie
Y, Kurachi Y 1997 Sulfonylurea receptor 2B and KIR 6.1 form a
sulfonylurea sensitive but ATP-insensitive K1 channel. J Physiol
(Lond) 499:715–720
Arena JP, Kass RS 1989 Activation of ATP-sensitive K channels in
heart cells by pinacidil: dependence on ATP. Am J Physiol 257:
H2092–2096
Thuringer D, Escande D 1989 Apparent competition between ATP
and the potassium channel opener RP 49356 on ATP-sensitive K1
channels of cardiac myocytes. Mol Pharmacol 36:897–902
Weik R, Neumcke B 1990 Effects of potassium channel openers on
single potassium channels in mouse skeletal muscle. Naunyn
Schmiedebergs Arch Pharmacol 342:258 –263
Panten U, Heipel C, Rosenberger F, Scheffer K, Zunkler BJ,
Schwanstecher C 1990 Tolbutamide-sensitivity of the adenosine
59-triphosphate-dependent K1 channel in mouse pancreatic B-cells.
Naunyn Schmiedebergs Arch Pharmacol 342:566 –574
Fan Z, Nakayama K, Hiraoka M 1990 Pinacidil activates the ATPsensitive K1 channel in inside-out and cell-attached patch membranes of guinea-pig ventricular myocytes. Pflugers Arch 415:
387–394
Schwanstecher C, Dickel C, Ebers I, Lins S, Zunkler BJ, Panten
U 1992 Diazoxide-sensitivity of the adenosine 59-triphosphatedependent K1 channel in mouse pancreatic b-cells. Br J Pharmacol
107:87–94
Allard B, Lazdunski M 1993 Pharmacological properties of ATPsensitive K1 channels in mammalian skeletal muscle cells. Eur
J Pharmacol 236:419 – 426
Jaggar JH, Harding EA, Ayton BJ, Dunne MJ 1993 Interaction of
diazoxide and cromakalim with ATP-regulated K1 channels in
rodent and clonal insulin-secreting cells. J Mol Endocrinol 10:59 –70
Larsson O, Ammala C, Bokvist K, Fredholm B, Rorsman P 1993
Stimulation of the KATP channel by ADP and diazoxide requires
nucleotide hydrolysis in mouse pancreatic b-cells. J Physiol (Lond)
463:349 –365
Henry P, Escande D 1994 Do potassium channel openers compete
with ATP to activate ATP sensitive potassium channels? Cardiovasc Res 28:754 –759
April, 1999
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
265. Forestier C, Pierrard J, Vivaudou M 1996 Mechanism of action of
K channel openers on skeletal muscle KATP channels. Interactions
with nucleotides and protons. J Gen Physiol 107:489 –502
266. Quast U, Bray KM, Andres H, Manley PW, Baumlin Y, Dosogne
J 1993 Binding of the K1 channel openers [3H]P1075 in rat isolated
aorta: relationship to functional effects of openers and blockers. J
Mol Pharmacol 43:474 – 481
267. Loffler-Walz C, Quast U 1998 Binding of KATP channel modulators
in rat cardiac membranes. Br J Pharmacol 123:1395–1402
268. Dickinson KE, Bryson CC, Cohen RB, Rogers L, Green DW,
Atwal KS 1997 Nucleotide regulation and characteristics of potassium channel opener binding to skeletal muscle membranes. Mol
Pharmacol 52:473– 481
269. Schwanstecher M, Sieverding C, H Dr Gross I, Aguilar-Bryan L,
Schwanstecher C, Bryan J 1998 Potassium channel openers require
ATP to bind to and act through sulfonylurea receptors. EMBO J
17:5529 –5535
270. Hambrock A, Loeffler-Walz C, Kurachi Y, Quast U 1998 Mg21 and
ATP dependence of KATP channel modulator binding to recombinant sulphonylurea receptor, SUR2B. Br J Pharmacol 125:577–583
271. Virsolvy-Vergine A, Leray H, Kuroki S, Lupo B, Dufour M,
Bataille D 1992 Endosulfine, an endogenous peptidic ligand for the
sulfonylurea receptor: purification and partial characterization
from ovine brain. Proc Natl Acad Sci USA 89:6629 – 6633
272. Peyrollier K, Heron L, Virsolvy-Vergine A, Le Cam A, Bataille D
1996 Alpha endosulfine is a novel molecule, structurally related to
a family of phosphoproteins. Biochem Biophys Res Commun 223:
583–586
273. Virsolvy-Vergine A, Salazar G, Sillard R, Denoroy L, Mutt V,
Bataille D 1996 Endosulfine, endogenous ligand for the sulphonylurea receptor: isolation from porcine brain and partial structural
determination of the a form. Diabetologia 39:135–141
274. Horiuchi A, Williams KR, Kurihara T, Nairn AC, Greengard P
1990 Purification and cDNA cloning of ARPP-16, a cAMP-regulated phosphoprotein enriched in basal ganglia, and of a related
phosphoprotein, ARPP-19. J Biol Chem 265:9476 –9484
275. Girault JA, Horiuchi A, Gustafson EL, Rosen NL, Greengard P
1990 Differential expression of ARPP-16 and ARPP-19, two highly
related cAMP-regulated phosphoproteins, one of which is specifically associated with dopamine-innervated brain regions. J Neurosci 10:1124 –1133
276. Girault JA, Shalaby IA, Rosen NL, Greengard P 1988 Regulation
by cAMP and vasoactive intestinal peptide of phosphorylation of
specific proteins in striatal cells in culture. Proc Natl Acad Sci USA
85:7790 –7794
277. Heron L, Virsolvy A, Peyrollier K, Gribble FM, Le Cam A,
Ashcroft FM, Bataille D 1998 Human a-endosulfine, a possible
regulator of sulfonylurea-sensitive KATP channel: molecular cloning, expression and biological properties. Proc Natl Acad Sci USA
95:8387– 8391
278. Ashfield R, Ashcroft SJ 1998 Cloning of the promoters for the
b-cell ATP-sensitive K-channel subunits Kir6.2 and SUR1. Diabetes
47:1274 –1280
279. Glaser B, Kesavan P, Heyman M, Davis E, Cuesta A, Buchs A,
Stanley CA, Thornton PS, Permutt MA, Matschinsky FM, Herold
K 1998 Familial hyperinsulinsim caused by an activating glucokinase mutation. N Engl J Med 338:226 –230
280. Zammarchi E, Filippi L, Novembre E, Donati MA 1996 Biochemical evaluation of a patient with a familial form of leucine-sensitive
hypoglycemia and concomitant hyperammonemia. Metab Clin
Exp 45:957–960
281. Weinzimer SA, Stanley CA, Berry GT, Yudkoff M, Tuchman M,
Thornton PS 1997 A syndrome of congenital hyperinsulinism and
hyperammonemia. J Pediat 130:661– 664
282. Stanley CA, Lieu Y, Hsu B, Poncz M 1997 Hypoglycemia in infants
with hyperinsulinemia and hyperammonemia: gain of function
mutations in the pathway of leucine-mediated insulin secretion.
Diabetes 46:217A
283. Kukuvitis A, Deal C, Arbour L, Polychronakos C 1997 An autosomal dominant form of familial persistent hyperinsulinemic hypoglycemia of infancy, not linked to the sulfonylurea receptor
locus. J Clin Endocrinol Metab 82:1192–1194
284. Thomas P, Ye Y, Lightner E 1996 Mutations of the pancreatic islet
285.
286.
287.
288.
289.
290.
291.
292.
293.
294.
295.
296.
297.
298.
299.
300.
301.
302.
133
inward rectifier also lead to familial persistent hyperinsulinemic
hypoglycemia of infancy. Hum Mol Genet 5:1809 –1812
Nestorowicz A, Inagaki N, Gonoi T, Schoor KP, Wilson BA,
Glaser B, Landau H, Stanley CA, Thornton TS, Seino S, Permutt
MA 1997 A nonsense mutation in the inward rectifier potassium
channel gene, Kir6.2, is associated with familial hyperinsulinism.
Diabetes 46:1743–1748
Glaser B, Chiu KC, Anker R, Nestorowicz A, Landau H, BenBassat H, Shlomai Z, Kaiser N, Thornton PS, Stanley CA, Spielman RS, Gogolin-Ewens K, Cerasi E, Baker L, Rice J, Donis-Keller
H, Permutt MA 1994 Familial hyperinsulinism maps to chromosome 11p14 –15.1, 30 cM centromeric to the insulin gene. Nat Genet
7:185–188
Thomas PM, Cote GJ, Hallman DM, Mathew PM 1995 Homozygosity mapping of the gene for familial persistent hyperinsulinemic
hypoglycemia of infancy to chromosome 11p. Am J Hum Genet
56:416 – 421
Thomas PM, Cote GJ, Wohlik N, Haddad B, Mathew PM, Rabl
W, Aguilar-Bryan L, Gagel RF, Bryan J 1995 Mutations in the
sulfonylurea receptor gene in familial hyperinsulinemic hypoglycemia of infancy. Science 268:426 – 429
Bryan J, Aguilar-Bryan L 1997 ATP-sensitive potassium channels:
more pieces to the puzzle. Curr Opin Cell Biol 9:553–559
Nestorowicz A, Wilson BA, Schoor KP, Inoue H, Glaser B,
Landau H, Stanley CA, Thornton PS, Clement IV JP, Bryan J,
Aguilar-Bryan L, Permutt MA 1996 Mutations in the sulfonylurea
receptor gene are associated with familial hyperinsulinism in Ashkenazi Jews. Hum Mol Genet 5:1813–1822
Thomas PM, Wohllk N, Huang E, Kuhnle U, Rabl W, Gagel RF,
Cote GJ 1996 Inactivation of the first nucleotide-binding fold of the
sulfonylurea receptor, and familial persistent hyperinsulinemic hypoglycemia of infancy. Am J Hum Genet 59:510 –518
Shyng S-L, Nestorowicz A, Permutt MA, Nichols CG 1997 Molecular mechanisms of persistent hyperinsulinemia hypoglycemia
of infancy caused by mutations in the ATP-sensitive potassium
channel. J Neurosci 23:1743
Kane C, Shepherd RM, Squires PE, Johnson PR, James RF, Milla
PJ, Aynsley-Green A, Lindley KJ, Dunne MJ 1996 Loss of functional KATP channels in pancreatic b-cells causes persistent hyperinsulinemic hypoglycemia of infancy. Nat Med 2:1344 –1347
Dunne MJ, Kane C, Shepherd RM, Sanchez JA, James RFL, Johnson PRV, Aynsley-Green A, Lu S, Clement IV JP, Lindley KJ,
Seino S, Aguilar-Bryan L 1997 Familial persistent hyperinsulinemic hypoglycemia of infancy and mutations in the sulfonylurea
receptor. N Engl J Med 336:703–706
Miki T, Tashiro F, Iwanaga T, Nagashima K, Yoshitomi H, Aihara
H, Nitta Y, Gonoi T, Inagaki N, Miyazaki J, Seino S 1997 Abnormalities of pancreatic islets by targeted expression of a dominant-negative KATP channel. Proc Natl Acad Sci USA 94:11969 –
11973
Patil N, Cox DR, Bhat D, Faham M, Myers RM, Peterson AS 1995
A potassium channel mutation in weaver mice implicates membrane excitability in granule cell differentiation. Nat Genet 11:
126 –129
Tucker SJ, Pessia M, Moorhouse AJ, Gribble F, Ashcroft FM,
Maylie J, Adelman JP 1996 Heteromeric channel formation and
Ca21-free media reduce the toxic effect of the weaver Kir 3.2 allele.
FEBS Lett 390:253–257
Laidlaw GF 1938 Nesidioblastoma, the islet tumor of the pancreas.
Am J Pathol 14:125–134
Yakovac WC, Baker L, Hummeler K 1971 Beta cell nesidioblastosis
in idiopathic hypoglycemia of infancy. J Pediat 79:226 –231
Aynsley-Green A, Polak JM, Bloom SR, Gough MH, Keeling J,
Ashcroft SJ, Turner RC, Baum JD 1981 Nesidioblastosis of the
pancreas: definition of the syndrome and the management of the
severe neonatal hyperinsulinaemic hypoglycaemia. Arch Dis Child
56:496 –508
Becker K, Muntefering H, Gruneklee D 1980 Nesidioblastosis of
the pancreas: clinical aspects, diagnosis and therapy. Monatsschrift
Kinderheilkunde 128:335–336
Heitz PU, Kloppel G, Hacki WH, Polak JM, Pearse AG 1977
Nesidioblastosis: the pathologic basis of persistent hyperinsulinemic hypoglycemia in infants. Morphologic and quantitative anal-
134
303.
304.
305.
306.
307.
308.
309.
310.
311.
312.
313.
314.
315.
316.
317.
318.
319.
320.
321.
AGUILAR-BRYAN AND BRYAN
ysis of seven cases based on specific immunostaining and electron
microscopy. Diabetes 26:632– 642
Gould VE, Memoli VA, Dardi LE, Gould NS 1981 Nesidiodysplasia and nesidioblastosis of infancy: ultrastructural and immunohistochemical analysis of islet cell alterations with and without
associated hyperinsulinaemic hypoglycaemia. Scand J Gastroenterol Suppl 70:129 –142
Gould VE, Memoli VA, Dardi LE, Gould NS 1983 Nesidiodysplasia and nesidioblastosis of infancy: structural and functional
correlations with the syndrome of hyperinsulinemic hypoglycemia. Pediatr Pathol 1:7–31
Rahier J, Wallon J, Henquin JC 1981 Cell populations in the endocrine pancreas of human neonates and infants. Diabetologia
20:540 –546
Rahier J, Falt K, Muntefering H, Becker K, Gepts W, Falkmer S
1984 The basic structural lesion of persistent neonatal hypoglycaemia with hyperinsulinism: deficiency of pancreatic d cells or hyperactivity of b cells? Diabetologia 26:282–289
Sempoux C, Guiot Y, Dubois D, Nollevaux MC, Saudubray JM,
Nihoul-Fekete C, Rahier J 1998 Pancreatic B-cell proliferation in
persistent hyperinsulinemic hypoglycemia of infancy: an immunohistochemical study of 18 cases. Mod Pathol 11:444 – 449
Lyonnet S, Bonnefont JP, Saudubray JM, Nihoule-Fekete C,
Brunelle F 1989 Localisation of focal lesion permitting partial pancreatectomy in infants. Lancet 2:671
Goossens A, Gepts W, Saudubray JM, Bonnefont JP, Nihoul F,
Heitz PU, Kloppel G 1989 Diffuse and focal nesidioblastosis. A
clinicopathological study of 24 patients with persistent neonatal
hyperinsulinemic hypoglycemia. Am J Surg Pathol 13:766 –775
Sempoux C, Poggi F, Brunelle F, Saudubray JM, Fekete C, Rahier
J 1995 Nesidioblastosis and persistent neonatal hyperinsulinism.
Diabete Metab 21:402– 407
Craver RD, Hill CB 1997 Cure of hypoglycemic hyperinsulinism
by enucleation of a focal islet cell adenomatous hyperplasia. J Pediat Surg 32:1526 –1527
Leibowitz G, Glaser B, Higazi AA, Salameh M, Cerasi E, Landau
H 1995 Hyperinsulinemic hypoglycemia of infancy (nesidioblastosis) in clinical remission: High incidence of diabetes mellitus and
persistent b-cell dysfunction at long-term follow-up. J Clin Endocrinol Metab 80:386 –392
Schonau E, Deeg KH, Huemmer HP, Akcetin YZ, Bohles HJ 1991
Pancreatic growth and function following surgical treatment of
nesidioblastosis in infancy. Eur J Pediat 150:550 –553
Dubois J, Brunelle F, Touati G, Sebag G, Nuttin C, Thach T,
Nikoul-Fekete C, Rahier J, Saudubray JM 1995 Hyperinsulinism
in children: diagnostic value of pancreatic venous sampling correlated with clinical, pathological and surgical outcome in 25 cases.
Pediat Radiol 25:512–516
Ingemansson S, Lunderquist A, Lundquist I, Lovdahl R, Tibblin
S 1975 Portal and pancreatic vein catheterization with radioimmunologic determination of insulin. Surg Gynecol Obstet 141:
705–711
Ingemansson S, Kuhl C, Larsson LI, Lunderquist A, Lundquist I
1978 Localization of insulinomas and islet cell hyperplasias by
pancreatic vein catheterization and insulin assay. Surg Gynecol
Obstet 146:724 –734
Reichardt W, Ingemansson S 1980 Selective vein catheterization
for hormone assay in endocrine tumours of the pancreas. Technique and results. Acta Radiol Diag 21:177–187
Rahier J, Sempoux C, Fournet JC, Poggi F, Brunelle F, NihoulFekete C, Saudubray JM, Jaubert F 1997 Partial or near-total pancreatomy for persistent neonatal hyperinsulinaemic hypoglycaemia: a pathologist’s role. Histopathology 32:15–19
Sempoux C, Guiot Y, Lefevre A, Nihoul-Fekete C, Jaubert F,
Saudubray JM, Rahier J 1998 Neonatal hyperinsulinemic hypoglycemia: heterogeneity of the syndrome and keys for differential
diagnosis. J Clin Endocrinol Metab 83:1455–1461
Rahier J, Sempoux C, Fournet JC, Poggi F, Brunelle F, NihoulFekete C, Saudubray JM, Jaubert F 1998 Partial or near-total pancreatectomy for persistent neonatal hyperinsulinaemic hypoglycaemia: the pathologist’s role. Histopathology 32:15–19
de Lonlay P, Fournet J, Rahier J, Gross-Morand M, Poggi-Travert
F, Foussier V, Bonnefont J, Brusset M, Brunelle F, Robert J,
322.
323.
324.
325.
326.
327.
328.
329.
330.
331.
332.
333.
334.
335.
336.
337.
338.
339.
340.
341.
Vol. 20, No. 2
Nihoul-Fekete C, Saudubray J, Junien C 1997 Somatic deletion of
the imprinted 11p15 region in sporadic persistent hyperinsulinemic
hypoglycemia of infancy is specific of focal adenomatous hyperplasia and endorses partial pancreatectomy. J Clin Invest 100:
802– 807
Hani EH, Clement K, Velho G, Vionnet N, Hager J, Philippi A,
Dina C, Inoue H, Permutt MA, Basdevant A, North M, Demenais
F, Guy-Grand B, Froguel P 1997 Genetic studies of the sulfonylurea
receptor gene locus in NIDDM and in morbid obesity among
French Caucasians. Diabetes 46:688 – 694
Inoue H, Ferrer J, Welling CM, Elbein SC, Hoffman M, Mayorga
R, Warren-Perry M, Zhang Y, Millns H, Turner R, Province M,
Bryan J, Permutt MA, Aguilar-Bryan L 1996 Sequence variants in
the sulfonylurea receptor (SUR) gene are associated with NIDDM
in Caucasians. Diabetes 45:825– 831
Iwasaki N, Kawamura M, Yamagata K, Cox NJ, Karibe S,
Ohgawara H, Inagaki N, Seino S, Bell GI, Omori Y 1996 Identification of microsatellite markers near the human genes encoding
the b-cell ATP-sensitive K1 channel and linkage studies with
NIDDM in Japanese. Diabetes 45:267–269
Stirling B, Cox NJ, Bell GI, Hanis CL, Spielman RS, Concannon
P 1995 Linkage studies in NIDDM with markers near the sulphonylurea receptor gene. Diabetologia 38:1479 –1481
Sakura H, Wat N, Horton V, Millns H, Turner RC, Ashcroft FM
1996 Sequence variations in the human Kir6.2 gene, a subunit of the
b-cell ATP-sensitive K-channel: no association with NIDDM in
while Caucasian subjects or evidence of abnormal function when
expressed in vitro. Diabetologia 39:1233–1236
Gopalakrishnan M, Triggle DJ 1992 Regulation of ATP-sensitive
K1 channels by chronic glyburide and pinacidil administration.
Biochem Pharmacol 44:1843–1847
Niki I, Ashcroft SJ 1993 Characterization and solubilization of the
sulphonylurea receptor in rat brain. Neuropharmacology 32:
951–957
Zini S, Tremblay E, Roisin MP, Ben-Ari Y 1991 Two binding sites
for [3H]glibenclamide in the rat brain. Brain Res 542:151–154
Mourre C, Widmann C, Lazdunski M 1990 Sulfonylurea binding
sites associated with ATP-regulated K1 channels in the central
nervous system: autoradiographic analysis of their distribution and
ontogenesis, and of their localization in mutant mice cerebellum.
Brain Res 519:29 – 43
Treherne JM, Ashford ML 1991 The regional distribution of sulphonylurea binding sites in rat brain. Neuroscience 40:523–531
Jiang C, Xia Y, Haddad GG 1992 Role of ATP-sensitive K1 channels
during anoxia: major differences between rat (newborn and adult)
and turtle neurons. J Physiol (Lond) 448:599 – 612
Zini S, Tremblay E, Pollard H, Moreau J, Ben-Ari Y 1993 Regional
distribution of sulfonylurea receptors in the brain of rodent and
primate. Neuroscience 55:1085–1091
Gehlert DR, Gackenheimer SL, Mais DE, Robertson DW 1991
Quantitative autoradiography of the binding sites for [125I] iodoglyburide, a novel high-affinity ligand for ATP-sensitive potassium
channels in rat brain. J Pharmacol Exp Ther 257:901–907
Karschin C, Ecke C, Ashcroft FM, Karschin A 1997 Overlapping
distribution of KATP channel-forming Kir6.2 subunit and the sulfonylurea receptor SUR1 in rodent brain. FEBS Lett 401:59 – 64
Ben-Ari Y, Krnjevic K, Crepel V 1990 Activators of ATP-sensitive
K1 channels reduce anoxic depolarization in CA3 hippocampal
neurons. Neuroscience 37:55– 60
Amoroso S, Schmid AH, Fosset M, Lazdunski M 1990 Glucose,
sulfonylureas, and neurotransmitter release: role of ATP-sensitive
K1 channels. Science 247:852– 854
Ashford ML, Boden PR, Treherne JM 1990 Tolbutamide excites rat
glucoreceptive ventromedial hypothalamic neurones by indirect
inhibition of ATP-K1 channels. Br J Pharmacol 101:531–540
Ashford ML, Boden PR, Treherne JM 1990 Glucose-induced excitation of hypothalamic neurones is mediated by ATP-sensitive K1
channels. Pflugers Arch 415:479 – 483
Politi DM, Rogawski MA 1991 Glyburide-sensitive K1 channels in
cultured rat hippocampal neurons: activation by cromakalim and
energy-depleting conditions. Mol Pharmacol 40:308 –315
Murphy KP, Greenfield SA 1991 ATP-sensitive potassium chan-
April, 1999
342.
343.
344.
345.
346.
347.
348.
349.
350.
351.
352.
353.
354.
355.
356.
357.
358.
MOLECULAR BIOLOGY OF ATP-SENSITIVE POTASSIUM CHANNELS
nels counteract anoxia in neurones of the substantia nigra. Exp
Brain Res 84:355–358
Levesque D, Greenfield SA 1991 Psychopharmacological evidence
for a role of the ATP-sensitive potassium channel in the substantia
nigra of the rat. Neuropharmacology 30:359 –365
Hausser MA, de WJ, Lazdunski M 1991 Activation by cromakalim
of pre- and post-synaptic ATP-sensitive K1 channels in substantia
nigra. Biochem Biophys Res Commun 174:909 –914
Tromba C, Salvaggio A, Racagni G, Volterra A 1992 Hypoglycemia-activated K1 channels in hippocampal neurons. Neurosci Lett
143:185–189
Treherne JM, Ashford ML 1992 Extracellular cations modulate the
ATP sensitivity of ATP-K1 channels in rat ventromedial hypothalamic neurons. Proc R Soc Lond B Biol Sci 247:121–124
Ohno-Shosaku T, Yamamoto C 1992 Identification of an ATPsensitive K1 channel in rat cultured cortical neurons. Pflugers Arch
422:260 –266
Murphy KP, Greenfield SA 1992 Neuronal selectivity of ATPsensitive potassium channels in guinea-pig substantia nigra revealed by responses to anoxia. J Physiol (Lond) 453:167–183
Takata Y, Shimada F, Kato H 1993 Possible involvement of ATPsensitive K1 channels in the inhibition of rat central adrenergic
neurotransmission under hypoxia. Jpn J Pharmacol 62:279 –287
Schwanstecher C, Panten U 1993 Tolbutamide- and diazoxidesensitive K1 channel in neurons of substantia nigra pars reticulata.
Naunyn Schmiedebergs Arch Pharmacol 348:113–117
Erdemli G, Krnjevic K 1993 Diazoxide suppresses slowly-inactivating outward and inward currents in CA1 hippocampal neurones. Neuroreport 5:249 –251
Crepel V, Rovira C, Ben-Ari Y 1993 The K1 channel opener diazoxide enhances glutamatergic currents and reduces GABAergic
currents in hippocampal neurons. J Neurophysiol 69:494 –503
Trapp S, Ballanyi K, Richter DW 1994 Spontaneous activation of
KATP current in rat dorsal vagal neurones. Neuroreport 5:1285–1288
Tromba C, Salvaggio A, Racagni G, Volterra A 1994 Hippocampal
hypoglycaemia-activated K1 channels: single-channel analysis of
glucose and voltage dependence. Pflugers Arch 429:58 – 63
Schwanstecher C, Panten U 1994 Identification of an ATP-sensitive
K1 channel in spiny neurons of rat caudate nucleus. Pflugers Arch
427:187–189
Jiang C, Sigworth FJ, Haddad GG 1994 Oxygen deprivation activates an ATP-inhibitable K1 channel in substantia nigra neurons.
J Neurosci 14:5590 –5602
Watts AE, Hicks GA, Henderson G 1995 Putative pre- and
postsynaptic ATP-sensitive potassium channels in the rat substantia nigra in vitro. J Neurosci 15:3065–3074
Trapp S, Ballanyi K 1995 KATP channel mediation of anoxiainduced outward current in rat dorsal vagal neurons in vitro.
J Physiol (Lond) 487:37–50
Erdemli G, Krnjevic K 1995 Actions of diazoxide on CA1 neurons
359.
360.
361.
362.
363.
364.
365.
366.
367.
368.
369.
370.
371.
372.
373.
374.
375.
135
in hippocampal slices from rats. Can J Physiol Pharmacol 73:
608 – 618
Tanaka T, Yoshida M, Yokoo H, Mizoguchi K, Tanaka M 1996
ATP-sensitive K1 channel openers block sulpiride-induced dopamine release in the rat striatum. Eur J Pharmacol 297:35– 41
Jiang C, Haddad GG 1997 Modulation of K1 channels by intracellular ATP in human neocortical neurons. J Neurophysiol 77:
93–102
Lazdunski M 1994 ATP-sensitive potassium channels: an overview. J Cardiovasc Pharmacol 24:S1–5
Fellows LK, Boutelle MG, Fillenz M 1993 ATP-sensitive potassium channels and local energy demands in the rat hippocampus:
an in vivo study. J Neurochem 61:949 –954
During MJ, Leone P, Davis KE, Kerr D, Sherwin RS 1995 Glucose
modulates rat substantia nigra GABA release in vivo via ATPsensitive potassium channels. J Clin Invest 95:2403–2408
Tartaglia LA 1997 The leptin receptor. J Biol Chem 272:6093– 6096
Houseknecht KL, Baile CA, Matteri RL, Spurlock ME 1998 The
biology of leptin: a review. J Anim Sci 76:1405–1420
Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman
JM 1994 Positional cloning of the mouse obese gene and its human
homologue [published erratum appears in Nature 1995 374:479].
Nature 372:425– 432
Lee GH, Proenca R, Montez JM, Carroll KM, Darvishzadeh JG,
Lee JI, Friedman JM 1996 Abnormal splicing of the leptin receptor
in diabetic mice. Nature 379:632– 635
Kieffer TJ, Heller RS, Habener JF 1996 Leptin receptors expressed
on pancreatic b-cells. Biochem Biophys Res Commun 224:522–527
Leclercq-Meyer V, Malaisse WJ 1998 Failure of human and mouse
leptin to affect insulin, glucagon and somatostatin secretion by the
perfused rat pancreas at physiological glucose concentration. Mol
Cell Endocrinol 141:111–118
Leclercq-Meyer V, Considine RV, Sener A, Malaisse WJ 1996 Do
leptin receptors play a functional role in the endocrine pancreas?
Biochem Biophys Res Commun 229:794 –798
Emilsson V, Liu YL, Cawthorne MA, Morton NM, Davenport M
1997 Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes 46:313–316
Harvey J, McKenna F, Herson PS, Spanswick D, Ashford ML 1997
Leptin activates ATP-sensitive potassium channels in the rat insulin-secreting cell line, CRI-G1. J Physiol (Lond) 504:527–535
Sayle RA, Milner-White EJ 1995 RASMOL: biomolecular graphics
for all. Trends Biochem Sci 20:374
Largent BL, Jones DT, Reed RR, Pearson RC, Snyder SH 1988 G
protein mRNA mapped in rat brain by in situ hybridization. Proc
Natl Acad Sci USA 85:2864 –2868
Ashcroft SJ, Ashcroft FM 1990 Properties and functions of ATPsensitive K-channels. Cell Signal 2:197–214
The Cleveland Clinic Foundation Department of Endocrinology Presents
Diabetes Day 1999: The Cutting Edge
May 5, 1999, The Omni International Hotel, Cleveland, Ohio
For further information please contact the Department of Continuing Medical Education at: (216) 444-5696
(Local) 800-862-8163 (Toll Free) or (216) 445-9406 (Fax).
27th European Symposium on Calcified Tissues
The 27th European Symposium on Calcified Tissues, hosted by the Finnish Bone Society and organized in
cooperation with the European Calcified Tissue Society, will be held on May 6 –10, 2000, in Tampere, Finland.
For more information, see www.congcreator.com/ects-2000, or contact the Congress Secretariat:
[email protected] or fax 1358-9-454-21930.