![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 282, Issue 3, 1247-1252, 1997
Department of Veterinary Physiology and Pharmacology (C.Y., B.L., T.C., W.H.H.), Iowa State University, Ames, Iowa, Department of Veterinary Medicine National Chung Hsing University, Taichung, Taiwan 40227 (C.Y.), and Department of Comparative Medicine (T.H., W.H.H.), Pig Research Institute of Taiwan, Chunan, Miaoli, Taiwan 35099, Republic of China
 |
Abstract |
|---|
 Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
We investigated the mechanisms underlying bradykinin (BK)-induced rise in intracellular Ca++ concentration [Ca++]i and insulin secretion using clonal beta cell line RINm5F. Incubation with a range of concentrations of BK increased in concentration-dependent manners both insulin secretion (BK of 10 nM to 10 µM) and [Ca++]i (BK of 100 nM to 100 µM). In Ca++-containing medium, BK (1 µM) induced a biphasic [Ca++]i rise, which was characterized by a Ca++ peak and a sustained Ca++ phase. In the Ca++-free medium, BK failed to increase insulin secretion and induced only a Ca++ peak without the sustained Ca++ phase. Thapsigargin (1 µM), an inhibitor of the Ca++ pump in the endoplasmic reticulum, abolished the Ca++ peak and the sustained phase. Nimodipine (1 µM), a voltage-dependent Ca++ channel blocker, abolished the BK-induced sustained Ca++ phase and inhibited BK-induced insulin release. The BK1 receptor agonist des-Arg9-BK (1 µM) did not change either [Ca++]i or insulin secretion. Both the BK-induced insulin secretion and rise in [Ca++]i were inhibited by a selective BK2 receptor antagonist, HOE 140 (3.3-100 nM), in concentration-dependent manners but were not by a BK1 receptor antagonist des-Arg9,Leu8-BK (1 µM). Pretreatment with pertussis toxin (0.1 µg/ml) did not block the BK-induced insulin secretion or increase in [Ca++]i. U-73122 (4, 6 and 8 µM), a phospholipase C inhibitor, antagonized both the BK-induced insulin secretion and the increase in [Ca++]i in a concentration-dependent and parallel manner. BK increased intracellular concentrations of inositol-1,4,5-trisphosphate (IP3). Neither (p-amylcinnamoyl)anthranilic acid (100 µM), a phospholipase A2 inhibitor, nor NG-nitro-L-arginine methylester (100 µM), a nitric oxide synthase inhibitor, inhibited these effects of BK. Taken together, these findings suggested that in beta cells, BK activates BK2 receptors, which, in turn, activate a pertussis toxin-insensitive G protein. The G protein couples to phospholipase C, which promotes the formation of IP3 and diacylglycerol. IP3 releases [Ca++]i from the intracellular Ca++ store, probably the endoplasmic reticulum, which triggers Ca++ influx via voltage-dependent Ca++ channels and thus increases insulin secretion.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
The
autacoid BK is a potent vasoactive nonapeptide that influences a number
of biological processes; it regulates blood pressure and local blood
flow (Pan et al., 1993
), produces pain (Steranka et
al., 1988) and inflammation (Figueroa et al., 1990),
increases vascular permeability and localized edema (Carter et
al., 1974), contracts smooth muscle (Collier et al.,
1962), increases cell proliferation (Marceau and Tremble, 1986),
increases glucose uptake (Sharp and Debnam, 1992) and decreases blood
glucose concentration (Wicklmayr and Dietze, 1977). The effects of BK
are mediated by at least two groups of BK receptors,
BK1 and BK2.
BK1 receptors mediate the acute inflammatory
response, whereas BK2 receptors are responsible
for most of the biological activities of kinins (Regoli et
al., 1993).
In a previous report, we demonstrated that BK stimulated insulin
secretion via BK2 receptors in a
concentration-dependent manner from the perfused rat pancreas (Yang and
Hsu, 1995). However, the mechanisms underlying BK-induced insulin
secretion remain unknown. In general, kinin receptors are coupled to G
proteins that may be PTX sensitive or insensitive (Bhoola et
al., 1992). At least four possible signal transduction pathways
may have been associated with the effects of BK in other cellular
systems: (1) PLC, which hydrolyzes PIP2 to form
IP3 and DAG; IP3 mobilizes [Ca++]i from the
intracellular Ca++ store of the ER (Takeuchi
et al., 1988); (2) PLA2, which
increases the formation of arachidonic acid (Birch and Axelrod, 1987).
Arachidonic acid, in turn, mobilizes Ca++ from
the ER and opens VDCCs (Fernandez and Balsinde, 1991); (3) adenylyl
cyclase, which increases cAMP formation (Suidan et al., 1991); and (4) NO synthase, which increases the formation of NO (Mombouli and Vanhoutte, 1995). Therefore, in this study, we intended to determine which signal transduction pathway would mediate the effects of BK on insulin secretion from clonal beta cell
line RINm5F. Specifically, we determined whether (1) BK induced insulin secretion through an increase in
[Ca++]i, (2) the effects
of BK were mediated by BK1 or
BK2 receptors, (3) a PTX-sensitive G protein and
PLC mediated the effects of BK, (4) VDCCs mediated BK-induced
Ca++ influx, and (5) PLA2
and NO played a role in BK-induced insulin secretion.
| |
Methods |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Cell culture.
A clonal beta cell line RINm5F
(donated by Dr. S. B. Pek of the University of Michigan, Ann
Arbor, MI) was maintained in RPMI-1640 medium supplemented with 10%
fetal bovine serum and aerated with 5% CO2/95%
air, as previously described (Chen and Hsu, 1994). The cells were
cultured for 5 days, and passages from 42 to 55 were used in these
experiments.
Insulin secretion.
RINm5F cells were plated onto 24-well
plates (Corning Glass Works, Corning, NY) at 2 × 105 cells/130-mm well and grown for 5 days.
During the experiments, the culture medium was removed and replaced
with KRB solution containing (in mM) 136 NaCl, 4.8 KCl, 1.2 CaCl2, 1.2 KH2PO4, 1.2 MgSO4, 5 NaHCO3, 10 HEPES,
4 D-glucose and 0.1% bovine serum albumin, pH 7.4. The
cells were preincubated for 15 min at 37°C and then incubated in KRB
with the test agent. When needed, cells were pretreated with PTX for 16 hr, HOE 140, des-Arg9,Leu8-BK, U-73122
(1-[6-[[17
-3-methoxyestra-1,3,5(10)-trien-17-yl]amino]hexyl]-1H-pyrrole-2,5dione), and U-73343
(1-(6-((17-3-methoxystra-1,3,5(10)-trien-17-yl)amino)hexyl)-2,5-pyrrolidine-dione), ACA, L-NAME or nimodipine for 5 min before BK administration. The
supernatant fluids were collected, kept at 4°C and subsequently assayed within 12 hr for insulin by using radioimmunoassay as previously described (Hsu et al, 1991a).
Measurement of
[Ca++]i.
Cultured RINm5F cells of ~30 × 106 cells
were loaded with 2 µM fura-2 AM in KRB for 30 min at 37°C. The
loaded cells were centrifuged (200 × g), resuspended
in KRB at a concentration of 106/ml and kept at
24°C for [Ca++]i
measurement. Aliquots of 1.5 × 106 cells
were used for [Ca++]i
measurement at 24°C. In the absence of extracellular
Ca++, cells were centrifuged (200 × g) and resuspended in Ca++-free/EGTA
(10 µM) medium. The 340 nm/380 nm fluorescence ratios were monitored
in an SLM-8000 fluorescence spectrophotometer (SLM, Urbana, IL).
[Ca++]i was calibrated
after cell lysis as previously described (Hsu et al.,
1991a). When needed, cells were pretreated with HOE 140, des-Arg9,Leu8-BK, U-73122,
ACA, L-NAME or nimodipine for 100 sec before BK administration. Cells
were pretreated with PTX for 2 hr and TG for 30 min before
[Ca++]i measurement. None
of the test agents interacted with fura-2 at the concentration(s) used
in the present study.
Measurement of IP3. Intracellular IP3 concentration of RINm5F cells was measured using a radioreceptor binding assay kit purchased from DuPont (Boston, MA). Then, 1.5 × 106 cells in 1 ml KRB were placed in a polypropylene tube, and the treatment with BK was terminated in 10 sec by the addition of 20% (w/v) ice-cold trichloroacetic acid. The concentration of IP3 was determined according to instructions of the manufacturer.
Data analyses. Data are expressed as mean ± S.E. Analysis of variance was used to determine the treatment or dose effect. The least significant difference test was used to determine the differences between means of end points for which the analysis of variance indicated a significant (P < .05) F ratio.
Materials. All chemicals were purchased from Sigma Chemical Co. (St. Louis, MO), except HOE 140, which was donated by Hoechst-Roussel Pharmaceuticals (Somerville, NJ); U-73122, U-73343 and ACA were purchased from BIOMOL Research Laboratory (Plymouth Meeting, PA), nimodipine was purchased from Research Biochemicals Inc. (Natick, MA) and fura-2 AM was purchased from Molecular Probes (Eugene, OR).
| |
Results |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Effects of BK on insulin secretion and
[Ca++]i
increase.
BK (10 nM to 10 µM) increased insulin secretion in a
concentration-dependent manner (fig. 1A).
BK (1 µM) increased insulin concentration in the wells to 3.7 times
the basal level. BK (1 µM) failed to induce insulin secretion in
Ca++-free medium (control = 0.34 ± 0.02 ng/well/min; BK = 0.35 ± 0.01 ng/well/min,
n = 3, P > .05). BK (100 nM to 100 µM)
increased [Ca++]i in a
concentration-dependent manner (fig. 1B). BK (1 µM) increased [Ca++]i to ~2 times (at
the peak) over the basal level. In
Ca++-containing medium, BK (1 µM) induced a
transient Ca++ increase, which reached the peak
(238 ± 6 nM, n = 8) at ~20 sec. This peak was
followed by a sustained Ca++ plateau phase that
gradually declined to the basal level over 4 min (fig.
2). In Ca++-free
medium, BK (1 µM) induced only a transient increase in
[Ca++]i (168 ± 5 nM, n = 8) without the sustained
Ca++ phase (fig. 2). The basal
[Ca++]i in
Ca++-free/EGTA medium was 88 ± 4 nM
(n = 8) (fig. 2).
|
|
Effects of BK receptor agonists and antagonists on insulin
secretion and [Ca++]i
increase.
BK (1 µM) increased insulin secretion (fig.
3) and
[Ca++]i (at the peak)
(fig. 4), but
des-Arg9-BK (1 µM), a BK1
receptor agonist, did not increase insulin secretion (fig. 3) or
[Ca++]i (fig. 4). HOE 140 (1 µM), a BK2 receptor antagonist, inhibited both the BK-induced insulin secretion (fig.
3) and increase in [Ca++]i (fig. 4). In
contrast, des-Arg9,Leu8-BK (1 µM), a
BK1 receptor antagonist, failed to alter this
effect of BK (figs. 3 and 4). HOE 140 (3 to 100 nM) inhibited both the BK-induced insulin secretion (fig. 5A) and
[Ca++]i increase in a
concentration-dependent manner (fig. 5B). HOE 140 at 100 nM abolished
the BK-induced insulin secretion (fig. 5A) and
[Ca++]i increase (fig.
5B).
|
|
|
Effects of PTX and U-73122 on BK-induced insulin secretion and
[Ca++]i
increase.
Pretreatment with PTX (0.1 µg/ml) for 16 and 2 hr,
respectively, failed to inhibit either BK (1 µM)-induced insulin
secretion (BK = 1.31 ± 0.04 ng/well/min; PTX + BK = 1.26 ± 0.05 ng/well/min, n = 3; P > .05) or
an increase in [Ca++]i
(BK = 212 ± 5 nM; PTX + BK = 210 ± 4 nM,
n = 6; P > .05). PTX (0.1 µg/ml) alone did not
induce insulin secretion (control = 0.35 ± 0.03 ng/well/min;
PTX = 0.36 ± 0.02 ng/well/min, n = 3; P > .05) or increase
[Ca++]i (control = 110 ± 5 nM; PTX = 112 ± 3 nM, n = 6;
P > .05). U-73122 (4, 6 and 8 µM), a PLC inhibitor, inhibited
both the BK-induced insulin secretion (fig.
6A) and
[Ca++]i increase in a
concentration-dependent manner (fig. 6B). U-73122 at 8 µM prevented
the BK-induced insulin secretion and
[Ca++]i increase by
~80%, but the analog U-73343 (8 µM) failed to alter the BK-induced
[Ca++]i increase (data
not shown). U-73122 alone slightly but significantly increased
[Ca++]i values, which
were 8.6 ± 1 and 11.4 ± 2 nM over the basal level after
U-73122 at 6 and 8 µM, respectively (n = 8 for each
concentration level).
[Ca++]i returned to the
basal level within 100 sec of U-73122 administration.
|
Effect of BK on intracellular concentrations of IP3. The basal level of IP3 was 36 ± 2 pmol/106 cells. BK significantly increased intracellular IP3 concentrations to 52 ± 2 pmol/106 cells within 10 sec of administration.
Effects of TG and nimodipine on BK-induced insulin secretion and
[Ca++]i
increase.
Pretreatment of RINm5F cells with TG (1 µM), an
inhibitor of the Ca++ pump in the ER, for 30 min
abolished the BK-induced transient Ca++ release
phase and sustained Ca++ influx phase (fig.
7A). TG (1 µM) alone significantly
increased [Ca++]i with an
onset of 5 sec and reached the peak that was 118 ± 9 nM
(n = 6) over the basal level. TG-induced increase in
[Ca++]i lasted <10 min.
Pretreatment of RINm5F cells with nimodipine (1 µM), a VDCC blocker,
for 5 min and 100 sec also inhibited BK-induced insulin release (fig.
8) and abolished the
Ca++ influx (fig. 7B), respectively. Nimodipine
(1 µM) alone did not change basal insulin secretion (control = 0.35 ± 0.02 ng/well/min; nimodipine = 0.39 ± 0.04 ng/well/min, n = 3; P > .05) or
[Ca++]i (control = 112 ± 4 nM; nimodipine = 114 ± 2 nM, n = 6; P > .05).
|
|
Failure of ACA and L-NAME to alter BK-induced insulin secretion. Neither the PLA2 inhibitor ACA (100 µM) nor the NOS inhibitor L-NAME (100 µM) affected BK (1 µM)-induced insulin secretion (ACA + BK = 1.24 ± 0.04 ng/well/min; L-NAME + BK = 1.20 ± 0.04 ng/well/min; BK = 1.27 ± 0.03 ng/well/min, n = 3; P > .05, compared with each other).
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
The results of the present study suggested that BK increases
[Ca++]i and insulin
secretion in RINm5F cells by activating BK2
receptors because BK, the nonselective BK receptor agonist (Farmer,
1992), increased insulin secretion and
[Ca++]i, but a
BK1 receptor agonist,
des-Arg9-BK (Farmer, 1992), did not. In addition,
a specific BK2 receptor antagonist, HOE 140 (Hock
et al., 1991), inhibited the BK-induced insulin secretion
and [Ca++]increase, but a
specific BK1 receptor antagonist.
des-Arg9,Leu8-BK (Farmer,
1992). did not. These findings are consistent with those of our
previous studies in the perfused rat pancreas (Yang and Hsu, 1995) and
a clonal beta cell line, HIT-T15 (Saito et al.,
1996). Further studies are needed to quantify BK2
receptors of beta cells using the radioligand binding
technique.
The BK2 receptor protein has seven-transmembrane
spanning domains, which are coupled to a G protein (Hess et
al., 1994). There are PTX-sensitive G proteins, such as
Gi and Go (Helper and
Gilman, 1992; Schmidt et al., 1991), and PTX-insensitive G
proteins, such as the Gq family (Helper and
Gilman, 1992). In bovine aortic endothelial cells, BK activates both
Gi and Gq (Liao and Homcy,
1993). We found that PTX at 0.1 µg/ml failed to inhibit the
stimulatory effects of BK in the present study, but it abolished the
inhibitory effect of alpha-2
adrenoceptor agonists on insulin secretion from RINm5F cells (Chen and
Hsu, 1994a) and [Ca++]i
in HIT cells (Hsu et al., 1991b). These results suggested
that in beta cells, BK2 receptors are
coupled to a PTX-insensitive G protein, probably
Gq, thereby increasing
[Ca++]i and insulin
secretion.
The Gq proteins are usually coupled to PLC
(Helper and Gilman, 1992). The specific PLC inhibitor U-73122 inhibits
PLC in a variety of cells, such as human neutrophils and platelets
(Bleasdale et al., 1990), rat hepatocytes (Galan et
al., 1991), pancreatic acinar cells (Yule and Williams, 1992) and
beta cells (Chen and Hsu, 1994b). In the present study,
BK-induced insulin secretion and
[Ca++]i increase were
inhibited by U-73122 in a concentration-dependent manner but were not
altered by the analog U-73343, which does not inhibit PLC activity
(Smith et al., 1990). By using neomycin (0.2 to 10 mM), a
less specific PLC inhibitor that also inhibits VDCC (Redman and
Silinsky, 1994), Saito et al. (1996) found that it inhibited
BK-induced increase in
[Ca++]i of HIT cells.
Thus, these results suggested that BK activates PLC, which catalyzes
the formation of IP3. IP3,
in turn, increases Ca++ release from the ER
(Berridge, 1993).
Intracellular Ca++ is a major signal in insulin
secretion. In many cell types, the stimulus-response usually couples to
the elevation of [Ca++]i
through Ca++ release from the intracellular store
and Ca++ influx. Secretagogues, including
hormones and neurotransmitters, increase
[Ca++]i, which lead to
insulin secretion (Komatsu et al., 1989; Li et
al., 1992). In RINm5F cells, BK induced a transient
Ca++ peak that was immediately followed by a
sustained Ca++ phase attributable to
Ca++ influx. The transient
Ca++ peak is partly attributed to
Ca++ release from the intracellular store,
probably the ER, because in Ca++-free medium, BK
induced only a Ca++ peak without the sustained
Ca++ phase. TG inhibits
Ca++ uptake by the Ca++
pump into the ER, thereby depleting the intracellular
Ca++ store (Thastrup et al., 1990).
Our findings showed that TG abolished both the BK-induced transient
Ca++ peak and sustained
Ca++ phase, suggesting that the BK-induced
Ca++ influx depends on the BK-induced
Ca++ release. The BK-induced
[Ca++]i peak in
Ca++-containing medium was higher than that in
Ca++-free medium. Thus, the BK-induced
Ca++ peak in
Ca++-containing medium came partly from
Ca++ release and partly from
Ca++ influx. The opening of voltage-independent
Ca++ channels and/or VDCCs may greatly contribute
to the Ca++ influx (Fasolato et al.,
1994). Our findings suggested that the BK-induced
Ca++ influx is predominantly mediated through
VDCCs because the sustained Ca++ influx was
abolished by a VDCC blocker nimodipine.
BK stimulates the formation of NO in vascular endothelial cells by
activating a PTX-insensitive G protein (Gq) (Liao
and Homcy, 1993). NO may increase insulin secretion by activating
guanylyl cyclase and thus generating cGMP (Laychock et al.,
1991). Our present results suggested that NO signal transduction
pathway is not involved in BK-induced insulin secretion in RINm5F
cells, because the NO synthase inhibitor L-NAME (100 µM) did not
alter BK-induced insulin secretion or
[Ca++]i increase. In
Swiss 3T3 fibroblasts, BK also activates the PLA2 signal transduction pathway via a PTX-insensitive G protein
(Birch and Axelrod, 1987). The activated PLA2
increases the synthesis of arachidonic acid, which induces insulin
secretion (Band et al., 1992). However, our findings
indicated that PLA2 is not involved in the
BK-induced insulin secretion, because the PLA2
inhibitor ACA (100 µM) failed to alter the effects of BK. ACA
inhibits glucose-induced activation of PLA2 in
beta cells (Konrad et al., 1992). In cultured tracheal smooth muscle cells, BK activates adenylyl cyclase and thus
increases cAMP (Stevens et al., 1994). However, in the
present study, BK (1 µM) did not alter the intracellular
concentration of cAMP in RINm5F cells.2 Our
results further suggest that an increase in cAMP concentration is not
involved in BK-induced insulin secretion.
In summary, our results suggested that in beta cells, bradykinin activates BK2 receptors, which, in turn, activate a receptor-coupled PTX-insensitive G-protein (Gq). The Gq protein activates PLC, which increases the formation of IP3 and DAG. IP3 releases Ca++ from the ER and triggers the Ca++ influx, leading to insulin secretion.
| |
Footnotes |
|---|
Accepted for publication May 12, 1997.
Received for publication September 3, 1996.
1 This work was partially supported by National Science Council, ROC(NSC86-2313-B-005-113-T).
2 C. Yang, B. Lee, T.-H. Chen and W. H. Hsu, unpublished obervations.
Send reprint requests to: Dr. Walter H. Hsu, Department of Biomedical Sciences, Iowa State University, Ames, IA 50011.
| |
Abbreviations |
|---|
ACA, (p-amylcinnamoyl)anthranilic acid; AM, acetoxymethyl ester; BK, bradykinin; DAG, diacylglycerol; ER, endoplasmic reticulum; HOE 140, IP3, inositol-1,4,5-trisphosphate; KRB, Krebs-Ringer bicarbonate solution; L-NAME, NG-nitro-l-arginine methylester; PIP2, phosphatidylinositol-4,5-bisphosphate; PTX, pertussis toxin; PLA2, phospholipase A2; PLC, phospholipase C; TG, thapsigargin; VDCC, voltage-dependent Ca++ channel.
| |
References |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
-TC3 cells.
Life Sci.
56: PL103-108, 1994b.
2-adrenergic receptors decreases Ca2+ influx to inhibit insulin secretion in a hamster cell line: an action mediated by a GTP-binding protein.
Endocrinology
128: 958-964, 1991b[Abstract].
,5-monophosphate formation in rat islets of Langerhans and RINm5F insulinoma cells: evidence for l-arginine:nitric oxide synthase.
Endocrinology
129: 3043-3052, 1991[Abstract].i and Gq family couple the bradykinin receptor to the release of endothelium-derived relaxing factor.
J. Clin. Invest.
92: 2168-2172, 1993. cell line, HIT-T15.
Biochem. Biophys. Res. Commun.
221: 577-580, 1996[Medline].This article has been cited by other articles:
|
|
 |
|
  K. Tsuda, I. Nishio, and M. M. Gironacci Angiotensin-(1-7) and Bradykinin in Norepinephrine Release in the Central Nervous System of Hypertension Hypertension, February 1, 2005; 45(2): e8 - e8. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wang, A. Lukinius, Y. Zhou, P. Stalberg, A. Gobl, K. Oberg, and B. Skogseid Subcellular Distribution of Phospholipase C Isoforms in Rodent Pancreas and Gastric Mucosa Endocrinology, July 1, 2000; 141(7): 2589 - 2593. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
