| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Biotechnology and Biosciences, University of Milano-Bicocca, I-20126 Milano, Italy;
Department of Endocrinology and Metabolism, Metabolic Unit, Ospedale Cisanello, I-56100 Pisa, Italy; and
Department of Experimental Pathology and Oncology, University of Florence, I-50134 Firenze, Italy
1Correspondence: Dipartimento di Biotecnologie e Bioscienze, Università di Milano-Bicocca, Piazza della Scienza, 2, 20126 Milano, Italy. E-mail: enzo.wanke{at}unimib.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: pancreas • eag gene family • erg gene • nesidioblastosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The action potentials produce pulses of
Ca2+ entry through voltage-dependent
Ca2+ channels that ultimately lead to
[Ca2+]i-dependent
exocytosis (2)
. Ca2+ and
K+ currents have been recognized and
characterized in human ß-cells (3
, 4)
, and some
electrical activity has been recorded using perforated patch techniques
(4
5
6
7)
.
Like rodent islets, human islets respond to a prolonged glucose step of
3–15 mM, with biphasic insulin secretion characterized by a spike-like
first phase (5–10 min), followed by a long-lasting plateau second
phase (7)
. The initial rise is characterized by latency
and a long duration before the beginning of the oscillatory pattern.
Secretion can be negatively modulated by Ca2+
channel blockers and induced in resting ß-cells by tolbutamide, a
reversible oral hypoglycemic sulfonylurea. The correlation between
electrical activity and
[Ca2+]i has been
demonstrated in mouse ß-cells (8)
. The use of
simultaneous microfluorimetric and amperometric techniques in the same
type of preparation has shown (9
10
11)
that a close
correlation exists between
[Ca2+]i and insulin
release.
HERG potassium channels (12
, 13)
are known to regulate the
duration of the heart action potential (14)
.The slowly
decreasing action potential (AP) plateau potential shifts these
channels to a less inactivated state that leads to an increasing
outward K+ current, and this sustains further
repolarization until complete repolarization is attained
(15)
. In addition to their role in the repolarization of
the heart action potential, HERG channels may sustain a process of
spike-frequency adaptation and thus contribute to the control of burst
duration. Chiesa et al. (16)
have shown that ERG channels
in a differentiated neuroblastoma cell sustain spike-frequency
adaptation during long trains of spikes because, at rest, an outward
ERG current (IERG) develops that is
sufficient to inhibit firing; Schönherr et al. (17)
found that the molecular determinants of this process are the specific
properties of the activation gate. This suggests that HERG channels may
also play a crucial role in other tissues in which excitability is of
primary significance for neurotransmitter and hormone release.
We provide evidence that RNAase protection assay experiments performed
in human pancreatic islets show the presence of herg1
transcripts. Moreover, in voltage-clamped human ß-cells, we found a
K+ current that has all of the biophysical and
pharmacological properties of the current sustained by HERG channels
(14
, 18)
. We characterized the properties of the
electrical activity of brief (3–5 min) and repeated applications of
high glucose and arginine in single patch-clamped ß-cells with and
without HERG channel blocker, and found the presence of
hyperexcitability during the blockade of these K+
channels, which suggests an increased insulin release. By means of
radioimmunoassay measurements of insulin in islets under the same
pharmacological conditions as those used during the electrical
recordings, we confirmed the predicted insulin hypersecretion.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 20)
. Briefly, the enzyme
collagenase (type XI, Sigma Chemical Co., St. Louis, Mo.) was used to
digest the gland. The pancreatic duct was cannulated and the digestion
solution [collagenase, 2 mg/ml, dissolved in 300 ml Hanks’ balanced
salt solution (HBSS), a volume equivalent to approximately three times
the weight of the pancreas] was slowly injected in order to distend
the tissue. After distension, the gland was placed into a 500 ml glass
beaker to which was added what remained of the distension solution, and
the beaker was loaded into a shaking 37°C water bath, activated at
120 revolutions/min. After 
10 min, the pancreas was shaken with
forceps for 60 s; the digestate was filtered first through a 300
µm, then through a 90 µm mesh stainless steel filter. The filtered
solution and the tissue remaining in the 300 µm mesh filter were
returned to the water bath for further digestion; the tissue remaining
in the 90 µm filter was washed with HBSS and 10% bovine serum (BS).
The same procedure of filtration, washing, and settling in HBSS
solution was repeated 8–10 times for up to 40–50 min. For the
purification, 3 ml of tissue was loaded into 220 ml conical plastic
containers and resuspended in 60 ml of 80% Histopaque (Sigma) 1.077 in
HBSS, topped up with 40 ml of HBSS. After centrifugation at 800
g for 5 min at 4°C, the islets were recovered at the
interface between the two layers, washed with HBSS 2% BS by
centrifugation at 800 g for 2 min at 4°C, and resuspended
in HBSS. After being put in 15 ml of M199 culture medium (supplemented
with 10% serum and antibiotics), the islets were loaded into 25
cm2 uncoated plastic flasks (2500 islets per
flask, Biobraun, Milan, Italy) and cultured at 37°C in a
CO2 incubator for 7–10 days, at which time the
electrophysiological and secretion studies were performed.
Secretion experiments
The insulin secretory function of the islets was evaluated by
means of static incubation as described previously (19
, 20)
. After a 30' preincubation period at 37°C in Krebs-Ringer
bicarbonate solution, 0.5% albumin, pH 7.4, containing 3.3 mM glucose,
they were incubated for 45 min in Krebs-Ringer solution with either
(mM) 1.1 glucose, 1.1 glucose plus 20 arginine, 5.5 glucose, 5.5
glucose plus 20 arginine, or 11.1 glucose with or without the addition
of 1, 5, or 10 µM WAY-123,398 (Wyeth-Ayerst Research, Princeton,
N.J.). At the end of the incubation time, aliquots of the incubation
medium were taken for insulin immunoassay measurements (Medgenix
Diagnostics, Fleurs, Belgium).
Molecular biology
The RNAase Protection Assays were performed essentially
according to Dixon and McKinnon (21)
. Briefly, 5–10 µg
of total RNA from either human brain or human pancreatic islets was
hybridized overnight at 48°C with
32P-UTP-labeled RNA probes. The herg probe was a
fragment (240 bp long) of the herg1 clone (accession number
NM000238.1KCNH2) in pBluescript SK+, linearized with
HindIII, and transcribed with T7 polymerase. Human
cyclophillin (Ambion, Austin, Tex.) was used as an internal loading
control. Digestion was then performed for 1 h at room temperature
with RNAase A (40 µg/ml) and T1 (2 µg/ml). Five micrograms of yeast
tRNA was used as a negative control for the probe self-protection
bands. The samples were then run on a 6.6% polyacrylamide gel and
exposed for 3 days.
Patch-clamp solutions
The standard extracellular solution contained (mM) NaCl 130, KCl
5, CaCl2 2, MgCl2 2,
HEPES-NaOH 10, D-glucose 3, and pH 7.40. In the high
K+ external solution
([K+]o= 40 mM), NaCl was
replaced by an equimolar amount of KCl. During the glucose-induced
responses, 12 mM glucose was added to the standard solution; during the
arginine-induced responses, 20 mM arginine was added in a 1 mM glucose
extracellular solution. The standard pipette solution at
[Ca2+]i =
10-7 M (pCa 7) contained (mM)
K+-aspartate 130, NaCl 10,
MgCl2 2, CaCl2 1.3,
EGTA-KOH 10, HEPES-KOH 10, ATP (Mg2+ salt) 1,
pH = 7.30. For perforated patch, pipettes were immersed briefly in
the following internal solution (mM):
K+-aspartate 140, NaCl 10,
MgCl2 2, HEPES-KOH 10, pH = 7.30; they were
backfilled with the same solution containing amphotericin B (150
µg/ml) kept at 4°C and made fresh every 2 h from a stock
solution (20 mg/ml in dimethylsulfoxide, prepared before each
experiment). In some experiments (data presented in Figs. 2
and 3
), the
control recordings were repeated (with the same protocol) in the
presence of 1–2 µM WAY-123,398 (WAY; 22
); the
difference between these recordings (called WAY-sensitive currents) was
used for analysis under the hypothesis that it represents the current
flowing in the HERG channels. WAY (received from Dr. W. Spinelli,
Wyeth-Ayerst Research) was dissolved in distilled water in order to
make 10 mM stock solutions.
|
|
Patch-clamp recordings
The currents were recorded at room temperature as described
previously (18
, 23)
. Pipette resistance (3–5 M
), cell
capacitance, and series resistance errors were carefully compensated
(85–95%) before each voltage-clamp protocol run. The APs and firing
were recorded (10 kHz sampling rate) in perforated patch current-clamp
by means of a particular patch-clamp amplifier developed in our
laboratory (24)
or an Axopatch 200A in
Ifast mode (Axon Instruments, Foster City,
Calif.). The extracellular solutions were delivered through a 9 hole
(0.6 mm), remote-controlled linear positioner placed near the cell
under study, which has an average response time of 2–3 s
(25)
.
Data analysis
During data acquisition and analysis, the pClamp suite (Axon
Instruments) and Origin 4.1 (Microcal Inc., Northampton, Mass.)
software were routinely used. The firing frequency was computed by
means of the ‘Peak’ routine in Origin 4.1. The peaks were considered
spikes when they were higher than -30 mV. The time of the successive
peaks was used to calculate the instantaneous frequency, which was then
averaged (‘running’ average over at least 50 points) and plotted (see
Figs. 5
and 6
). The data are expressed as mean values ±
SE.
|
|
To evaluate the total predicted insulin release from the membrane
potential (VM) recordings during each glucose
application, it is necessary to calculate the
Ca2+ currents
(ICa) from the voltage-dependent
activation curves and, given the roughly linear relationship between
ICa and exocytosis (26
, 27)
, evaluate the time integral of the instantaneous release. To
compute this integral, we calculated the hypothetical
ICa, at 2 ms intervals, as the product
of the voltage-dependent activation Boltzmann curve
1/(1+e-(VM-V1/2)/k)
(where V1/2 and k are the half activation
and the slope factor in mV) multiplied by the driving force term
(VM -80), assuming that the Nernst potential for
Ca2+ is +80 mV. The time integral of this
variable is shown in Figs. 5
6
7
. To investigate the importance of
setting the correct activation curve, we made preliminary calculations
using theoretical curves derived from published data (4
, 28)
. The Boltzmann curve parameters V1/2 and
k of these two models had the values (mV) of -12, 11, and
+9, 15. The results of these calculations, expressed as the percentage
increase in release (with the HERG blocking drug) with respect to
control, were 88% and 74% (see Fig. 6C
). On the whole, the
similar results obtained using two theoretical models with reasonably
different V1/2 values led us to conclude that the
overall calculation of total insulin release is relatively weakly
dependent on the absolute properties of the
ICa activation curve, and we finally
decided to use the model of Kelly et al. (28)
for plotting
the data shown in Figs. 5
6
7
.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,
only the human herg1 gene has a completely known sequence.
To demonstrate the presence of this gene in pancreatic ß cells, its
transcript was first checked in RNA extracted from the cell culture
used in the patch-clamp experiments. The herg1 transcript
was analyzed using the RNAase Protection Assay with a probe consisting
of a 240 bp-long fragment belonging to the herg1 sequence
(accession number NM000238.1 KCNH2). The results obtained from two
different pancreatic islet samples can be seen in Fig. 1
, which shows that a protected herg1 transcript is detectable
in human brain (used as a positive control) and, at the same or higher
intensity, in the two islet samples.
|
K(ATP) and HERG channels coexist in human ß-cells
Patch-clamp electrophysiology experiments were performed on 35
human ß-cells identified as being spontaneously silent (in low
glucose) and responding to tolbutamide, including 25 that also
responded to high levels of glucose (only one responded under
nonperforated patch conditions). IHERG
was detected in all of these cells, and we occasionally detected
spontaneously firing cells that did not respond to glucose (probably
-cells) or cells that became silent in high glucose levels and did
not respond to tolbutamide (probably 
-cells). These findings agree
with those of Nadal et al. (11)
in intact mouse Langerhans
islets. In a typical experiment (see Fig. 2A
), a cell perfused under normal glucose concentrations (3
mM) was tested in current-clamp mode using two concentrations (10 and
50 µM) of tolbutamide, a reversible blocker of
K(ATP) channels, in order to verify that some of
these channels are blocked and that the cell consequently depolarizes.
After this test establishing that the cell is a ß-cell, we switched
to the voltage-clamp mode in which we used the relatively high
[K+]o (40 mM), useful to
record consistent HERG K+ currents (HERG
dependence on [K+]o; see
refs 14
15
16
17
18
, 23
). From a holding potential
(VH) of 0 mV, the protocol in Fig. 2G
was used to elicit deactivating tail currents under control conditions
(Fig. 2B
) and during the perfusion of 50 µM tolbutamide
(Fig. 2C
). Tolbutamide produced a decrease in current that
clearly revealed a transient current which completely disappeared in
the presence of 1 µM of the antiarrhythmic drug WAY-123,398 (WAY), a
specific blocker of HERG channels (18
; not shown). The
recordings of the tolbutamide-sensitive current (obtained as recordings
B minus C) are shown in Fig. 2D
.
We used WAY either for isolating IHERG
(see Fig. 3
) or during the experiments in current-clamp (Fig. 6)
, but the blocker
was also used in the insulin release experiments (Table 1
) at relatively high doses. To exclude a blocking action of WAY on the
K(ATP) channels, we performed a specific test
that is shown in Fig. 3E
, F
. For this purpose, we used an
insulin-secreting cell clone RINm5F (30)
, known to express
K(ATP) channels, and did experiments both at
normal [K+]o of 5 mM
(n=5) and high
[K+]o of 40 mM
(n=5) at a maximal concentration of 5 µM of WAY. Figure 2E
, F
shows the tolbutamide-sensitive
(IK(ATP)) and WAY-sensitive currents
([K+]o=40 mM) elicited by
the protocol shown in Fig. 2G
, but from a holding of -60 mV
used for the purpose of not activating ERG channels (recordings are the
result of subtracting traces in the presence of drug from control
traces). It can be seen that in the same cell in which a large
IK(ATP) current can be recorded, the
application of WAY did not produce any measurable effect, thus
excluding that the HERG channels blocker could contribute through the
K(ATP) channels to the effects shown in Fig. 6
and Table 1
. In all other experiments (n=10) the results
were similar.
|
The biophysical properties of HERG channels in ß-cells
The experiments designed to characterize the properties of the
HERG currents were performed in high
[K+]o extracellular
solution (40 mM) and high [ATP]i (5 mM). Figure 3A
shows the results of experiments carried out on 10 cells
to determine the voltage dependence of the activation curve. The
recordings from a typical cell are shown in Fig. 3B
(protocols shown at the bottom of Fig. 3B
) and were obtained
as WAY-sensitive currents (see Materials and Methods). The traces are
tail currents from various steady-state preconditioning levels, and
represent a first and fast (7–8 ms) phase of removal of inactivation
and a second phase of deactivation lasting 100 ms. The data in Fig. 3A
fitted a Boltzmann relationship with the following
parameters (mV): V1/2, -29.5 ± 1.1, slope
9.9 ± 0.9. The voltage dependence of the inactivation curve (Fig. 3C
) was obtained by evaluating the normalized peak cord
conductance (18)
from recordings similar to those shown in
Fig. 3D
, which were obtained after applying the protocol
shown at the bottom of the panel. These traces describe the same
phenomenon (described before) of removal of inactivation and succeeding
deactivation but tested at different membrane potentials. The
experimental points fitted a Boltzmann relationship with the following
parameters (mV): V1/2, -42 ± 1.6, slope
17.4 ± 1.8. On the whole, the voltage-dependent properties of the
currents found in human ß-cells are in line with those described in
the literature (14
, 15
, 17
, 18)
.
B cell electrophysiological identification with tolbutamide,
glucose, and arginine
The ß-cells were identified by means of perforated-patch
current-clamp experiments using tolbutamide-, arginine-, or
glucose-induced electrical depolarization as shown in Fig. 4
. The typical membrane resistance and resting potential respectively
ranged from 1.5 to 4 G (mean 1.8±0.2) and from -49 to -65 mV
(mean -57±4 mV (n=35)). The application of different
concentrations of tolbutamide led to various depolarizing responses
(see also Figs. 5
and 6
), but the application of the highest
concentration (50 µM) generally induced strong firing very similar to
that obtained using the highest glucose (15 mM) or arginine (20 mM)
concentrations. In some cells (2 out of 15), the highest dose of
glucose did not reach the firing level whereas 20 mM arginine almost
always (8 out of 9) led to maximal depolarization. The delay in the
onset of the depolarizing wave after the glucose application varied
from 1.5 to 6 min (Fig. 4D
).
|
To verify that the tolbutamide-induced depolarization was sustained by
an increase in membrane resistance, we applied short (1.4 s)
hyperpolarizing current pulses every 15 s to test the voltage
response (Fig. 4A
): the application of tolbutamide led to a
strong depolarization and a large increase in resistance. The same
experiment was repeated also at a constant voltage (-71 mV) by
manually injecting a hyperpolarizing steady-state current in the
presence of the drug (Fig. 4B
). The increases in membrane
resistance were similar in both experiments and are consistent with the
hypothesis that ATP-dependent K+ channels are
closed by the drug (7)
.
Durability and stability of the glucose-induced response in single
human ß-cells
Given the lack of consistent data concerning the reproducibility
of successive glucose-evoked responses in the same human ß-cell, we
analyzed the variability of the responses to various applications of
glucose using two different criteria of analysis during glucose-evoked
electrical activity (Fig. 5A
): 1) average firing frequency (Fig. 5B
) and 2) predicted cumulative insulin release
(Fig. 5C
), assuming that it directly correlates with
the inflow of Ca2+ (see section on Data analysis
for details). The results of one of the four experiments are shown in
Fig. 5B
, C
. The experiments showed that the variability in
glucose responses after consecutive perfusions and washouts was
relatively small. The mean frequency in Hz (averaged on the number of
action potentials, n) was 2.97 (n=256), 2.76
(n=227), and 2.88 (n=297) in the three
applications; the predicted cumulative release was 0.75, 0.56, and 0.53
arbitrary units. The results obtained in the four cells from which we
could obtain consecutive responses fluctuated around the mean, with an
SE of ±0.04 for the average frequency and ±0.12
for the cumulative predicted release.
On the whole, these data indicate that single human ß-cells are capable of surviving under good conditions for more than 30 min and producing physiological responses to high glucose concentrations. Given that these responses are reproducible, any manipulation interfering with electrical activity during insulin secretion can be detected. In our experiments, we observed a variable latency and a strong initial phase of activity, but never found any evidence of oscillatory phases because we used much shorter glucose applications than those used by other laboratories.
Hyperexcitability during glucose-induced response in the presence
of a HERG channel blocker
The threshold of the activation curve for HERG channels is around
-60 mV, which is the usual level of the resting potential of these
cells. It is therefore reasonable to suppose that HERG channels
blockade would be ineffective in obtaining depolarizations at rest,
i.e., when the K(ATP) channels are still open. To
verify whether the presence of HERG channels is capable of regulating
the firing and consequent insulin release during glucose applications,
we performed experiments in which we added the WAY-123,398 blocker (1
µM) to extracellular solutions with a high glucose concentration (5.5
min application). One of the five experiments of this type is shown in
Fig. 6
. We could repeat the glucose+WAY application twice; the same type of
analysis was made as in the control experiment shown in Fig. 5
.
Voltage-clamp experiments (not shown) revealed that this concentration
of the blocker effectively inhibits 95% of
IHERG in 50–60 s and that 90%
recovery from the drug takes place in 500–700 s (18
, 25)
. In the presence of the drug we did not find any significant
differences in either the resting potential or the properties of the
spikes during firing (16)
, but both the firing frequency
and the predicted cumulative release were greatly enhanced during the
HERG channel blockade. The frequency changed from 1.65 Hz in the
control to 2.58 (+56%) and 2.26 (+37%) during the HERG blockade; the
number of action potentials was 102, 319, and 299, respectively; the
cumulative release increased by 88% and 83% in comparison with the
control value measured during the first glucose application without
WAY. Overall, in the five cells tested under control conditions and
during HERG channel blockade, the firing frequency increased by 32 ± 5% and the predicted release by 77 ± 8%.
Hyperexcitability during arginine-induced response in the presence
of HERG channel blocker
As shown in Fig. 4
, the human ß-cells responded to a high
concentration of arginine. Only a few cells responded to 5 mM (one out
of five) and 10 mM of arginine (three out of five); eight out of nine
promptly, reversibly, and repeatedly responded to 20 mM arginine, thus
suggesting that the response time is as short as in the case of
tolbutamide and much faster than that induced by glucose. Using
arginine instead of glucose, we performed the same set of experiments
as that shown in Fig. 6
, in which we blocked the HERG channel with the
specific blocker. The results of one of six successful experiments are
shown in Fig. 7
. Figure 7A
shows the recording of a 5 min application of 20
mM arginine without and with, after washout, the addition of 1 µM
WAY-123,398. Figure 7B
, C
shows the instantaneous frequency
and predicted insulin release calculated according to the procedures
used in Figs. 5
and 6
, which revealed a fourfold increase in the
average frequency (from 0.22 to 1 Hz, dotted line) and a doubling of
release after the HERG channels blockade. Overall, in the six cells
tested under control conditions and during HERG channel blockade, the
predicted release increased by 89 ± 6%.
Insulin release from intact human islets
To evaluate whether the findings reported above are paralleled by
changes in insulin release from intact human islets, we tested the
insulin secretion in response to varying glucose concentrations (from
1.1 to 11.1 mM) and/or 20 mM arginine with or without the addition of
1, 5, or 10 µM WAY-123,398, during static incubation experiments (see
Table 1
). At the low glucose concentration (1.1 mM), insulin release
clearly increased in a dose-dependent manner after the addition of the
HERG channel blocker. This effect was not found at the higher glucose
concentrations (5.5 and 11.1 mM), suggesting that the HERG channel may
play an important role in reducing insulin secretion only at low
glucose levels.
Blocking the HERG channels definitely increased arginine-potentiated
insulin release at both low (1.1 mM) and medium (5.5 mM) glucose
concentrations, but the present data do not allow any conclusion as to
whether this was due to a direct effect of the blocker on the ß-cells
or to the action of the amino acid on the -cells, with the
subsequent action of arginine-stimulated glucagon on insulin secretion.
Taken together, these data reinforce the evidence (shown in Figs. 6
and 7
) that blocking HERG channels induces a hyperexcitable condition in
ß-cells that leads to more pronounced islets insulin secretion than
under normal conditions.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study was carried after discovering that human ß-cells
constitutively express the herg1 gene (Fig. 1)
and are
endowed with HERG currents (Fig. 3)
whose properties are very similar
to those already described in other tissues. Recordings of
glucose-evoked electrical activity in human ß-cells are relatively
rare (4
5
6)
, and we wanted to examine and analyze some of
their behavioral properties using appropriate tools. This procedure was
necessary in order to be able to compare the responses of single cells
in which the HERG channels were normal or blocked; as far as we know,
no other experiments of this type in human ß-cells have been
documented. Our single cell electrophysiological recordings and the
correlated analyses (Figs. 6
and 7
for the glucose- and
arginine-induced responses, respectively) indicate that the blockade of
HERG channels is sufficient to cause unequivocal hyperexcitability
during the application of insulin releasing drugs.
A number of hypotheses have been proposed to clarify the processes that
underlie the ß-cells bursting pattern (for review see 31). Very
recently Göpel et al. (32)
suggested, in mouse
ß-cells, a novel IK(Ca) current.
They show that an apamin-, charybdotoxin-, and tolbutamide-insensitive
but Cd2+- and nifedipine-sensitive current is
activated during a simulated firing command. Our group
(17)
used a similar type of command to explain the role of
HERG currents in firing accommodation and the deactivation time
constants found by Göpel et al. are perfectly in line with those
of HERG channels (17)
. Moreover,
Cd2+ produces (33)
, a remarkable
activation right-shift that is probably sufficient to virtually block
IHERG at -40 mV; some
Ca2+ channel antagonists (we tested also
nifedipine) are also IHERG blockers
(34)
.
The results of the parallel islet experiments qualitatively agree with the single cell data, but revealed a different pattern at low glucose concentrations because of the basically different environmental conditions of the ß-cells. In the Petri dish used during the electrophysiological experiments, the single patched ß-cell is surrounded by only a few other cells, which prevents any paracrine signaling, whereas the ‘integrated’ system represented by the islets (in which the various endocrine cells can affect their mutual functions) means that the single ß-cells are subject to a complex orchestration of stimuli. No major change in glucose-stimulated insulin secretion was induced in the islet preparation with higher glucose levels by the presence of WAY-123,398, thus suggesting that other signaling pathways are likely to predominate in human ß-cells under our experimental conditions.
Nesidioblastosis (also called persistent hyperinsulinemic hypoglycemia
of infancy) and insulinomas are syndromes that cause severe
hypoglycemia because of the excessive release of insulin, which may
lead to seizures and epileptic events. Standard pancreatectomy and/or
treatment with diazoxide (a K+ ATP-dependent
channel opener) are the usual therapies, but although it has been
discovered that some forms of nesidioblastosis are linked to mutations
in the sulfonylurea receptor or the Kir6.2 ATP-dependent
K+ channel (35
, 36)
, the origin of
others described in the literature (37
38
39)
is not related
to either.
It has been reported (40)
that islets prepared from an
adult subject with nesidioblastosis showed an increase in basal insulin
release that was not further enhanced by higher glucose levels. These
findings resemble the data of the present study, which show that
insulin release increases at low glucose levels after HERG channels
blockade, and suggest the possibility that alterations in HERG channel
function may play a role in the altered insulin release associated with
some hyperinsulinemic conditions.
It has been shown in mouse that the role of various types of amino
acids is to shift the usual glucose dose response curve toward the left
in such a way that a smaller glucose concentration is necessary to
produce insulin secretion and electrical activity (41)
.
The application of 20 mM of arginine in the presence of a normal
glucose concentration led to a marked increase in single ß-cell
electrical activity and in islets insulin release. The HERG channel
blocker considerably increased this effect in both preparations, thus
suggesting that although arginine acts through a different signaling
pathway, it can still depolarize ß-cells in such a way that the HERG
channels retain an inhibitory role that is eliminated when they are
blocked. Furthermore, hypoglycemic conditions associated with
hyperinsulinemia may be induced by amino acid feeding
(42)
, which again suggests that HERG channels may play a
role in regulating insulin release under certain circumstances.
To conclude, the present study, while revealing another task entrusted to the multifaceted activity of HERG channels, discloses a new branching of the complex signaling network that controls the insulin secretion in human ß-cells.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 21, 2000.
Revision received April 24, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-, ß-, and -cells within intact islets of Langherhans in the mouse. J. Physiol. 517,85-93This article has been cited by other articles:
|
|
 |
|
  C. Alonso-Ron, P. de la Pena, P. Miranda, P. Dominguez, and F. Barros Thermodynamic and Kinetic Properties of Amino-Terminal and S4-S5 Loop HERG Channel Mutants under Steady-State Conditions Biophys. J., May 15, 2008; 94(10): 3893 - 3911. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Fernandez, J. Sargent, F. B. Sachse, and M. C. Sanguinetti Structural Basis for Ether-a-go-go-Related Gene K+ Channel Subtype-Dependent Activation by Niflumic Acid Mol. Pharmacol., April 1, 2008; 73(4): 1159 - 1167. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zhang, X.-S. Liu, S. Diochot, M. Lazdunski, and G.-N. Tseng APETx1 from Sea Anemone Anthopleura elegantissima Is a Gating Modifier Peptide Toxin of the Human Ether-a-go-go- Related Potassium Channel Mol. Pharmacol., August 1, 2007; 72(2): 259 - 268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-N. Tseng, K. D. Sonawane, Y. V. Korolkova, M. Zhang, J. Liu, E. V. Grishin, and H. R. Guy Probing the Outer Mouth Structure of the hERG Channel with Peptide Toxin Footprinting and Molecular Modeling Biophys. J., May 15, 2007; 92(10): 3524 - 3540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, G. H. Hockerman, H. W. Green III, C. F. Babbs, S. I. Mohammad, D. Gerrard, M. A. Latour, B. London, K. M. Hannon, and A. L. Pond Merg1a K+ channel induces skeletal muscle atrophy by activating the ubiquitin proteasome pathway FASEB J, July 1, 2006; 20(9): 1531 - 1533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Restano-Cassulini, Y. V. Korolkova, S. Diochot, G. Gurrola, L. Guasti, L. D. Possani, M. Lazdunski, E. V. Grishin, A. Arcangeli, and E. Wanke Species Diversity and Peptide Toxins Blocking Selectivity of Ether-a-go-go-Related Gene Subfamily K+ Channels in the Central Nervous System Mol. Pharmacol., May 1, 2006; 69(5): 1673 - 1683. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chapman, C. Ramstrom, L. Korhonen, M. Laine, K. T. Wann, D. Lindholm, M. Pasternack, and K. Tornquist Downregulation of the HERG (KCNH2) K+ channel by ceramide: evidence for ubiquitin-mediated lysosomal degradation J. Cell Sci., November 15, 2005; 118(22): 5325 - 5334. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Miranda, T. Giraldez, P. de la Pena, D. G Manso, C. Alonso-Ron, D. Gomez-Varela, P. Dominguez, and F. Barros Specificity of TRH receptor coupling to G-proteins for regulation of ERG K+ channels in GH3 rat anterior pituitary cells J. Physiol., August 1, 2005; 566(3): 717 - 736. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cherubini, G. Hofmann, S. Pillozzi, L. Guasti, O. Crociani, E. Cilia, P. Di Stefano, S. Degani, M. Balzi, M. Olivotto, et al. Human ether-a-go-go-related Gene 1 Channels Are Physically Linked to {beta}1 Integrins and Modulate Adhesion-dependent Signaling Mol. Biol. Cell, June 1, 2005; 16(6): 2972 - 2983. [Abstract] [Full Text] [PDF]
|
|
|
|
< |
