Page 1 of 41
Diabetes
1 2 3
Wnt Signaling Regulates Blood Pressure by Downregulating a GSK-3β β -Mediated Pathway to Enhance Insulin Signaling in the Central Nervous System
4 5 6
Pei-Wen Cheng, PhD1; Ying-Ying Chen, MD2; Wen-Han Cheng, PhD1; Pei-Jung Lu, PhD3; Hsin-Hung Chen, MS4; Bo-Rong Chen MS1 ; Tung-Chen Yeh, MD, PhD5; Gwo-Ching Sun, MD3; Michael Hsiao, DVM, PhD6; Ching-Jiunn Tseng, MD, PhD1, 4,7*
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
1
Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 2Department of Ophthalmology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 3Institute of Clinical Medicine, National Cheng-Kung University, Tainan, Taiwan; 4Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan; 5 Department of Internal Medicine, Division of Cardiology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 6Genomics Research Center, Academia Sinica, Taipei, Taiwan; 7 Department of Medical Research, China Medical University Hospital, China Medical University, Taichung 40402, Taiwan *
Correspondence: Ching-Jiunn Tseng, MD, PhD Department of Medical Education and Research, Kaohsiung Veterans General hospital 386, Ta-Chung 1st Rd., Kaohsiung, Taiwan 813 E-mail: [email protected] Telephone: 886-7-3422121ext.1505 Fax: 886-7-3468056 Pei-Wen Cheng and Ying-Ying Chen contributed equally to this work.
28 29 30 31 32 33 34 35 36 37
Running title: Cross-talk between Wnt and Insulin Signaling in Blood Pressure Regulation.
38 1
Diabetes Publish Ahead of Print, published online April 16, 2015
Diabetes
Page 2 of 41
1
Abstract
2
Aberrant Wnt signaling appears to play an important role in the onset of diabetes. Moreover,
3
the insulin signaling pathway is defective in the nucleus tractus solitarii (NTS) of
4
spontaneously hypertensive rats (SHRs) and fructose-fed rats. Nevertheless, the relationships
5
between Wnt signaling and the insulin pathway and the related modulation of blood pressure
6
(BP) in the central nervous system have yet to be established. The aim of this study was to
7
investigate the potential signaling pathways involved in Wnt-mediated blood pressure
8
regulation in the NTS. Pretreatment with the low density lipoprotein receptor-related protein
9
(LRP) antagonist Dickkopf-1 (DKK1) significantly attenuated the Wnt3a-induced depressor
10
effect and nitric oxide production. Additionally, inhibition of LRP6 activity using DKK-1
11
significantly abolished Wnt3a-induced glycogen
12
extracellular signal-regulated kinases 1/2T202/Y204, ribosomal protein S6 kinaseT359/S363 and
13
AktS473 phosphorylation and increased IRS1S332 phosphorylation. GSK-3β was also found to
14
directly bind to IRS1 and induce the phosphorylation of IRS1 at Ser332 in the NTS. By
15
contrast, administration of the GSK-3β inhibitor TWS119 into the brain decreased the BP of
16
hypertensive rats by enhancing IRS1 activity. Taken together, these results suggest that the
17
GSK-3β-IRS1 pathway may play a significant role in Wnt-mediated central BP regulation.
18
Key words: Wnt, nucleus tractus solitarii, insulin, insulin receptor substrate 1, low density
19
lipoprotein receptor-related proteins 2
synthase
kinase 3β (GSK-3β)S9,
Page 3 of 41
Diabetes
1
INTRODUCTION
2
The nucleus tractus solitarii (NTS), which is located in the dorsal medulla of the
3
brainstem, is the primary site of blood pressure (BP) and sympathetic nerve activity
4
modulation. The NTS participates in cardiovascular, gastric, and gustatory regulation. Our
5
previous studies revealed that several neuromodulators are involved in BP control of the NTS,
6
including insulin (1; 2) and nitric oxide (NO) (3).
7
Insulin is primarily produced by pancreatic beta-cells. Margolis and Altszuler reported
8
that insulin can cross the blood-brain barrier, as demonstrated by sensitive and specific
9
radioimmunoassays (4). However, insulin mRNA and immunoreactive insulin are present in
10
the rat central nervous system (5; 6). The insulin receptor (IR) is a tetrameric glycoprotein
11
that belongs to the receptor tyrosine kinase superfamily. Insulin can bind to the IR, leading to
12
the autophosphorylation of the beta subunit of the IR at tyrosine residues, which induces the
13
tyrosine phosphorylation of insulin receptor substrates (IRS) (7). IR/IRS can activate two
14
major signaling pathways: the phosphatidylinositol 3-kinase (PI3K)-Akt pathway and the
15
Ras-mitogen-activated protein kinase (MAPK) pathway. Insulin has also been implicated in
16
the regulation of the baroreceptor reflex in the NTS (8; 9). We previously found that insulin
17
is involved in the control of central BP via the PI3K-Akt-NO synthase (NOS) signaling
18
pathway in the NTS (1; 2).
19
Wnt proteins are a family of secreted glycoproteins that bind to the receptor Frizzled
20
(Fzd) and to co-receptors referred to as low density lipoprotein receptor-related protein (LRP) 3
Diabetes
1
5 and 6 (10). The canonical Wnt pathway involves the activation of Fzd receptors and
2
LRP5/6, which results in the stabilization of β-catenin and importation into the nucleus,
3
where it acts as a co-factor of T cell factor (TCF)/lymphoid enhancer factor (LEF) family
4
transcription factors (11). Dickkopf (DKK) molecules have been revealed to function as
5
specific antagonists of canonical Wnt signaling by directly binding to LRP5/6 (12). Wnt
6
signaling regulates various biological activities including cancer (13; 14), stem cell
7
maintenance (15), cardiac hypertrophy(16) and neural differentiation (17; 18). In addition,
8
recent clinical and in vitro studies revealed that aberrant Wnt signaling appears to play an
9
important role in the onset of hypertension and diabetes (19-23). Nevertheless, the
10
relationship between Wnt signaling and the insulin pathway and the modulation of BP in the
11
central nervous system has yet to be established.
12
In the present study, we investigated whether Wnt stimulation promotes central insulin
13
signaling in the NTS to modulate BP. We clarified whether aberrant Wnt signaling causes a
14
neuronal insulin signaling defect that induces hypertension. In addition, we investigated
15
which molecular mechanisms are essential for the fructose-induced hypertension-mediated
16
dysfunction of the canonical Wnt signaling pathway in the NTS. Overall, our results suggest
17
that this neuronal insulin signaling defect is a core mechanism that induces hypertension and
18
that activation of the Wnt signaling pathway may alleviate this defect.
19 4
Page 4 of 41
Page 5 of 41
Diabetes
1
MATERIALS AND METHODS
2
Experimental chemicals
3
All experimental drugs were purchased from Sigma-Aldrich (St. Louis, MO, USA),
4
unless otherwise noted.
5
Animals
6
Twenty-week old male WKY rats and SHRs were obtained from the National Science
7
Council Animal Facility and housed in the animal room of Kaohsiung Veterans General
8
Hospital (Kaohsiung, Taiwan). Humane treatment was administered at all times. The rats
9
were maintained in individual cages in a room in which the lighting was controlled (12 hours
10
on/12 hours off) and the temperature was maintained between 23 °C and 24 °C. The rats were
11
acclimatized to the housing conditions for one week and were then trained for one week to
12
acclimate the animals to the procedure of indirect blood pressure measurement. The rats were
13
randomly assigned to nine groups of six rats per group: 1) SHR group: the SHRs received an
14
intracerebroventricular (ICV) injection of artificial cerebrospinal fluid (aCSF) as a vehicle
15
control; 2) SHR + Wnt group: the SHRs received an ICV injection of Wnt; 3) SHR + Wnt +
16
DKK1 group: the SHRs received an ICV injection of Wnt and DKK1; 4) WKY group: the
17
WKY rats received an ICV injection of aCSF as a vehicle control; 5) WKY + Wnt group: the
18
WKY rats received an ICV injection of Wnt; 6) WKY + Wnt + DKK1 group: the WKY rats
19
received an ICV injection of Wnt and DKK1; 7) the WKY group: WKY rats received an ICV
5
Diabetes
1
injection of aCSF as a vehicle control; 8) Fructose group: the WKY rats were provided with
2
10% fructose water for two weeks; and 9) Fructose + Wnt group: the fructose-fed WKY rats
3
received an ICV injection of Wnt. The rats were provided with normal rat chow (Purina, St.
4
Louis, MO, USA) and tap water ad libitum. All animal research protocols were approved by
5
the Research Animal Facility Committee of Kaohsiung Veterans General Hospital.
6 7
ICV injection procedure
8
ICV infusion experiments were performed following a stabilization period of at least 30
9
minutes after insertion of the microinjector into the ventricular-guided cannula. The BP was
10
monitored for 3 days after drug infusion. As a vehicle control, the effect of ICV injection of
11
aCSF (142 mmol/L NaCl, 5 mmol/L KCl, 10 mmol/L glucose and 10 mmol/L HEPES, pH 7.4)
12
was analyzed. Wnt3a (0.9 pmol/per day) and DKK1 (1 µg/per day) were dissolved in aCSF;
13
the GSK-3β inhibitor (TWS119, 0.173µg/per day) was initially dissolved in DMSO and then
14
diluted in aCSF at a final concentration of 1% DMSO. The basal BP was examined prior to
15
injection. The daily ICV drug infusions were performed over a 2-min period and delivered as
16
a single bolus of a final volume of 5 µL from day 0 to day 14. Wnt3a and the inhibitors were
17
injected simultaneously.
18 19
BP measurement 6
Page 6 of 41
Page 7 of 41
Diabetes
1
Using a previously described tail-cuff method (Model MK-2000 Storage Pressure Meter,
2
Muromachi Kikai, Tokyo, Japan), the systolic BP (SBP) and the heart rate were measured
3
prior to Wnt3a or DKK1 treatment (day 0). The animals were placed in the chamber for 30
4
min. In this method, the reappearance of pulsation on a digital display of the BP cuff was
5
detected using a pressure transducer and was amplified and recorded as the SBP. During the
6
measurement, a series of 10 individual readings were rapidly obtained. The highest and
7
lowest readings were dropped from consideration, and the remaining eight readings were
8
averaged.
9 10
Measurement of the NO and insulin levels in the NTS
11
Total protein was prepared by homogenizing the NTS tissue in lysis buffer and was
12
deproteinized using Microcon YM-30 centrifugal filter units (Millipore, Bedford, MA, USA).
13
The total amount of NO in the samples was determined using a modified procedure based on
14
the purge system of a Sievers Nitric Oxide Analyzer (NOA 280i) (Sievers Instruments,
15
Boulder, CO, USA), which involves the use of chemiluminescence (24). The samples (10 µl)
16
were injected into a reflux column containing 0.1 mol/L VCl3 in 1 mol/L HCl at 90 °C to
17
reduce any nitrates and nitrites (NOx) to NO. Then, the NO reacted with the O3 produced by
18
the analyzer to form NO2. The resulting emission from the excited NO2 was detected using a
19
photomultiplier tube and was digitally recorded (mV). Then, the emission values were 7
Diabetes
1
interpolated to a concurrently determined standard concentration curve of NaNO3. The
2
measurements were performed in triplicate for each sample. The measured NO levels were
3
corrected for the volume of the examined rats. The NTS insulin levels were measured using
4
an Ultrasensitive Rat Insulin ELISA kit (Mercodia, Uppsala, Sweden) and detected with a
5
Biochrom Anthos Zenyth 200RT Microplate Reader (Cambridge, UK).
6
TOP/FOP flash Wnt reporter assay
7
The cells were transfected with TOP flash or FOP flash Wnt reporter plasmids (Millipore
8
Corporation) containing wild-type or mutant T-cell factor (TCF) DNA binding sites by using
9
Lipofectamine 2000. The cells were also cotransfected with β-gal reporter plasmids. The
10
reporter activity was analyzed as described above.
11
Immunoblotting analysis
12
Total protein was prepared by homogenizing the NTS tissue in lysis buffer containing a
13
protease inhibitor cocktail and a phosphatase inhibitor cocktail and subsequently incubating
14
the sample for 1 hour at 4 °C. The protein extracts (20 µg/sample based on the BCA protein
15
assay, Pierce Chemical Co., Rockford, IL, USA) were resolved on a 6% polyacrylamide gel
16
and transferred to a PVDF membrane (GE Healthcare, Buckinghamshire, UK). The
17
membranes were incubated in the appropriate anti-P-LRP6S1490 (2568), anti-P-AktS473 (4060),
18
anti-P-ribosomal protein S6 kinase (RSK)T359/S363 (9344), anti-P-ERK1/2T202/Y204 (4370),
19
anti-LRP6 (2560), anti-Akt (9272), anti-RSK (9341), anti-ERK1/2 (137F5), anti-P-IRS1S332 8
Page 8 of 41
Page 9 of 41
Diabetes
anti-P-GSK-3βS9
1
(2580)
2
anti-GSK-3β (07-389), anti-IRS1 (05-784R), anti-Un-P-β-catenin (05-665), anti-β-catenin
3
(AB19022)(Millipore, Billerica, MA, USA) and anti-P-β-cateninSer33 (SC-16743), anti-Dvl1
4
(Santa Cruz Biotechnology, Dallas, TX, USA) antibodies. Then, the membranes were then
5
incubated in an HRP-labeled goat anti-rabbit secondary antibody at 1:10,000. The
6
membranes were developed using the ECL-Plus detection kit (GE Healthcare).
(Cell
Signaling
Technology,
MA,
USA),
(05-643),
7 8
Immunohistochemistry analysis
9
The sections were deparaffinized, quenched in 3% H2O2/methanol, heated (using a
10
microwave) in citrate buffer (10 mmol/L, pH 6.0), blocked in 5% goat serum and incubated
11
in the anti-P-LRP6S1490 antibody overnight at 4 °C. Next, the sections were incubated in a
12
biotinylated secondary antibody (1:200; Vector Laboratories, Burlingame, CA, USA) for 1
13
hour and in AB complex (1:100) for 30 min at room temperature. The sections were
14
visualized using a DAB substrate kit (Vector Laboratories) and were counterstained with
15
hematoxylin. Then, the sections were photographed using a microscope equipped with a
16
charge-coupled device camera.
17 18 19
Co-immunoprecipitation and in vitro kinase assays The Catch and Release Reversible Immunoprecipitation System (Millipore) was used 9
Diabetes
1
according to the manufacturer's instructions. The proteins were eluted in 70 µl of elution
2
buffer and subjected to immunoblotting analysis using the anti-GSK-3β and anti-P-IRS1S332
3
antibodies.
4
For the in vitro kinase assay, co-immunoprecipitation using the anti-GSK-3β and
5
anti-IRS1 antibodies was performed, and the GSK-3β-IRS1 complex was eluted using the
6
Catch and Release Reversible Immunoprecipitation System. The kinase reaction was initiated
7
by adding kinase buffer and the terminated by adding 2X sample buffer and boiling for 10
8
minutes. The phosphorylation of IRS1 was determined via immunoblotting analysis using the
9
anti-P-IRS1S332 and anti-IRS1 antibodies.
10 11
Immunofluorescence staining analysis
12
The brainstem sections were incubated in the anti-GSK-3β (1:200) and anti-IRS1S332
13
(1:50) antibodies. After washing with PBS, the sections were incubated in green-fluorescent
14
Alexa Fluor-488 anti-rabbit IgG (1:200; Invitrogen) at 37 °C for 2 hours. The sections were
15
analyzed using a confocal microscope and Zeiss LSM Image (Carl Zeiss Micro Imaging)
16
software.
17 18 19
Statistical analysis Unpaired Student's t-test was used to compare protein levels (SHR and WKY group, 10
Page 10 of 41
Page 11 of 41
Diabetes
1
WKY and fructose treated group, SHR and SHR + TWS119 group or fructose and fructose +
2
TWS119 group) and the BP measurements (SHR and SHR + TWS119 group or fructose and
3
fructose + TWS119 group). One-way analysis of variance (ANOVA) followed by Scheffe
4
post-hoc analysis were applied to compare the differences between groups. Differences in
5
which P
Diabetes
1 2 3
Wnt Signaling Regulates Blood Pressure by Downregulating a GSK-3β β -Mediated Pathway to Enhance Insulin Signaling in the Central Nervous System
4 5 6
Pei-Wen Cheng, PhD1; Ying-Ying Chen, MD2; Wen-Han Cheng, PhD1; Pei-Jung Lu, PhD3; Hsin-Hung Chen, MS4; Bo-Rong Chen MS1 ; Tung-Chen Yeh, MD, PhD5; Gwo-Ching Sun, MD3; Michael Hsiao, DVM, PhD6; Ching-Jiunn Tseng, MD, PhD1, 4,7*
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
1
Department of Medical Education and Research, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 2Department of Ophthalmology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 3Institute of Clinical Medicine, National Cheng-Kung University, Tainan, Taiwan; 4Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan; 5 Department of Internal Medicine, Division of Cardiology, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; 6Genomics Research Center, Academia Sinica, Taipei, Taiwan; 7 Department of Medical Research, China Medical University Hospital, China Medical University, Taichung 40402, Taiwan *
Correspondence: Ching-Jiunn Tseng, MD, PhD Department of Medical Education and Research, Kaohsiung Veterans General hospital 386, Ta-Chung 1st Rd., Kaohsiung, Taiwan 813 E-mail: [email protected] Telephone: 886-7-3422121ext.1505 Fax: 886-7-3468056 Pei-Wen Cheng and Ying-Ying Chen contributed equally to this work.
28 29 30 31 32 33 34 35 36 37
Running title: Cross-talk between Wnt and Insulin Signaling in Blood Pressure Regulation.
38 1
Diabetes Publish Ahead of Print, published online April 16, 2015
Diabetes
Page 2 of 41
1
Abstract
2
Aberrant Wnt signaling appears to play an important role in the onset of diabetes. Moreover,
3
the insulin signaling pathway is defective in the nucleus tractus solitarii (NTS) of
4
spontaneously hypertensive rats (SHRs) and fructose-fed rats. Nevertheless, the relationships
5
between Wnt signaling and the insulin pathway and the related modulation of blood pressure
6
(BP) in the central nervous system have yet to be established. The aim of this study was to
7
investigate the potential signaling pathways involved in Wnt-mediated blood pressure
8
regulation in the NTS. Pretreatment with the low density lipoprotein receptor-related protein
9
(LRP) antagonist Dickkopf-1 (DKK1) significantly attenuated the Wnt3a-induced depressor
10
effect and nitric oxide production. Additionally, inhibition of LRP6 activity using DKK-1
11
significantly abolished Wnt3a-induced glycogen
12
extracellular signal-regulated kinases 1/2T202/Y204, ribosomal protein S6 kinaseT359/S363 and
13
AktS473 phosphorylation and increased IRS1S332 phosphorylation. GSK-3β was also found to
14
directly bind to IRS1 and induce the phosphorylation of IRS1 at Ser332 in the NTS. By
15
contrast, administration of the GSK-3β inhibitor TWS119 into the brain decreased the BP of
16
hypertensive rats by enhancing IRS1 activity. Taken together, these results suggest that the
17
GSK-3β-IRS1 pathway may play a significant role in Wnt-mediated central BP regulation.
18
Key words: Wnt, nucleus tractus solitarii, insulin, insulin receptor substrate 1, low density
19
lipoprotein receptor-related proteins 2
synthase
kinase 3β (GSK-3β)S9,
Page 3 of 41
Diabetes
1
INTRODUCTION
2
The nucleus tractus solitarii (NTS), which is located in the dorsal medulla of the
3
brainstem, is the primary site of blood pressure (BP) and sympathetic nerve activity
4
modulation. The NTS participates in cardiovascular, gastric, and gustatory regulation. Our
5
previous studies revealed that several neuromodulators are involved in BP control of the NTS,
6
including insulin (1; 2) and nitric oxide (NO) (3).
7
Insulin is primarily produced by pancreatic beta-cells. Margolis and Altszuler reported
8
that insulin can cross the blood-brain barrier, as demonstrated by sensitive and specific
9
radioimmunoassays (4). However, insulin mRNA and immunoreactive insulin are present in
10
the rat central nervous system (5; 6). The insulin receptor (IR) is a tetrameric glycoprotein
11
that belongs to the receptor tyrosine kinase superfamily. Insulin can bind to the IR, leading to
12
the autophosphorylation of the beta subunit of the IR at tyrosine residues, which induces the
13
tyrosine phosphorylation of insulin receptor substrates (IRS) (7). IR/IRS can activate two
14
major signaling pathways: the phosphatidylinositol 3-kinase (PI3K)-Akt pathway and the
15
Ras-mitogen-activated protein kinase (MAPK) pathway. Insulin has also been implicated in
16
the regulation of the baroreceptor reflex in the NTS (8; 9). We previously found that insulin
17
is involved in the control of central BP via the PI3K-Akt-NO synthase (NOS) signaling
18
pathway in the NTS (1; 2).
19
Wnt proteins are a family of secreted glycoproteins that bind to the receptor Frizzled
20
(Fzd) and to co-receptors referred to as low density lipoprotein receptor-related protein (LRP) 3
Diabetes
1
5 and 6 (10). The canonical Wnt pathway involves the activation of Fzd receptors and
2
LRP5/6, which results in the stabilization of β-catenin and importation into the nucleus,
3
where it acts as a co-factor of T cell factor (TCF)/lymphoid enhancer factor (LEF) family
4
transcription factors (11). Dickkopf (DKK) molecules have been revealed to function as
5
specific antagonists of canonical Wnt signaling by directly binding to LRP5/6 (12). Wnt
6
signaling regulates various biological activities including cancer (13; 14), stem cell
7
maintenance (15), cardiac hypertrophy(16) and neural differentiation (17; 18). In addition,
8
recent clinical and in vitro studies revealed that aberrant Wnt signaling appears to play an
9
important role in the onset of hypertension and diabetes (19-23). Nevertheless, the
10
relationship between Wnt signaling and the insulin pathway and the modulation of BP in the
11
central nervous system has yet to be established.
12
In the present study, we investigated whether Wnt stimulation promotes central insulin
13
signaling in the NTS to modulate BP. We clarified whether aberrant Wnt signaling causes a
14
neuronal insulin signaling defect that induces hypertension. In addition, we investigated
15
which molecular mechanisms are essential for the fructose-induced hypertension-mediated
16
dysfunction of the canonical Wnt signaling pathway in the NTS. Overall, our results suggest
17
that this neuronal insulin signaling defect is a core mechanism that induces hypertension and
18
that activation of the Wnt signaling pathway may alleviate this defect.
19 4
Page 4 of 41
Page 5 of 41
Diabetes
1
MATERIALS AND METHODS
2
Experimental chemicals
3
All experimental drugs were purchased from Sigma-Aldrich (St. Louis, MO, USA),
4
unless otherwise noted.
5
Animals
6
Twenty-week old male WKY rats and SHRs were obtained from the National Science
7
Council Animal Facility and housed in the animal room of Kaohsiung Veterans General
8
Hospital (Kaohsiung, Taiwan). Humane treatment was administered at all times. The rats
9
were maintained in individual cages in a room in which the lighting was controlled (12 hours
10
on/12 hours off) and the temperature was maintained between 23 °C and 24 °C. The rats were
11
acclimatized to the housing conditions for one week and were then trained for one week to
12
acclimate the animals to the procedure of indirect blood pressure measurement. The rats were
13
randomly assigned to nine groups of six rats per group: 1) SHR group: the SHRs received an
14
intracerebroventricular (ICV) injection of artificial cerebrospinal fluid (aCSF) as a vehicle
15
control; 2) SHR + Wnt group: the SHRs received an ICV injection of Wnt; 3) SHR + Wnt +
16
DKK1 group: the SHRs received an ICV injection of Wnt and DKK1; 4) WKY group: the
17
WKY rats received an ICV injection of aCSF as a vehicle control; 5) WKY + Wnt group: the
18
WKY rats received an ICV injection of Wnt; 6) WKY + Wnt + DKK1 group: the WKY rats
19
received an ICV injection of Wnt and DKK1; 7) the WKY group: WKY rats received an ICV
5
Diabetes
1
injection of aCSF as a vehicle control; 8) Fructose group: the WKY rats were provided with
2
10% fructose water for two weeks; and 9) Fructose + Wnt group: the fructose-fed WKY rats
3
received an ICV injection of Wnt. The rats were provided with normal rat chow (Purina, St.
4
Louis, MO, USA) and tap water ad libitum. All animal research protocols were approved by
5
the Research Animal Facility Committee of Kaohsiung Veterans General Hospital.
6 7
ICV injection procedure
8
ICV infusion experiments were performed following a stabilization period of at least 30
9
minutes after insertion of the microinjector into the ventricular-guided cannula. The BP was
10
monitored for 3 days after drug infusion. As a vehicle control, the effect of ICV injection of
11
aCSF (142 mmol/L NaCl, 5 mmol/L KCl, 10 mmol/L glucose and 10 mmol/L HEPES, pH 7.4)
12
was analyzed. Wnt3a (0.9 pmol/per day) and DKK1 (1 µg/per day) were dissolved in aCSF;
13
the GSK-3β inhibitor (TWS119, 0.173µg/per day) was initially dissolved in DMSO and then
14
diluted in aCSF at a final concentration of 1% DMSO. The basal BP was examined prior to
15
injection. The daily ICV drug infusions were performed over a 2-min period and delivered as
16
a single bolus of a final volume of 5 µL from day 0 to day 14. Wnt3a and the inhibitors were
17
injected simultaneously.
18 19
BP measurement 6
Page 6 of 41
Page 7 of 41
Diabetes
1
Using a previously described tail-cuff method (Model MK-2000 Storage Pressure Meter,
2
Muromachi Kikai, Tokyo, Japan), the systolic BP (SBP) and the heart rate were measured
3
prior to Wnt3a or DKK1 treatment (day 0). The animals were placed in the chamber for 30
4
min. In this method, the reappearance of pulsation on a digital display of the BP cuff was
5
detected using a pressure transducer and was amplified and recorded as the SBP. During the
6
measurement, a series of 10 individual readings were rapidly obtained. The highest and
7
lowest readings were dropped from consideration, and the remaining eight readings were
8
averaged.
9 10
Measurement of the NO and insulin levels in the NTS
11
Total protein was prepared by homogenizing the NTS tissue in lysis buffer and was
12
deproteinized using Microcon YM-30 centrifugal filter units (Millipore, Bedford, MA, USA).
13
The total amount of NO in the samples was determined using a modified procedure based on
14
the purge system of a Sievers Nitric Oxide Analyzer (NOA 280i) (Sievers Instruments,
15
Boulder, CO, USA), which involves the use of chemiluminescence (24). The samples (10 µl)
16
were injected into a reflux column containing 0.1 mol/L VCl3 in 1 mol/L HCl at 90 °C to
17
reduce any nitrates and nitrites (NOx) to NO. Then, the NO reacted with the O3 produced by
18
the analyzer to form NO2. The resulting emission from the excited NO2 was detected using a
19
photomultiplier tube and was digitally recorded (mV). Then, the emission values were 7
Diabetes
1
interpolated to a concurrently determined standard concentration curve of NaNO3. The
2
measurements were performed in triplicate for each sample. The measured NO levels were
3
corrected for the volume of the examined rats. The NTS insulin levels were measured using
4
an Ultrasensitive Rat Insulin ELISA kit (Mercodia, Uppsala, Sweden) and detected with a
5
Biochrom Anthos Zenyth 200RT Microplate Reader (Cambridge, UK).
6
TOP/FOP flash Wnt reporter assay
7
The cells were transfected with TOP flash or FOP flash Wnt reporter plasmids (Millipore
8
Corporation) containing wild-type or mutant T-cell factor (TCF) DNA binding sites by using
9
Lipofectamine 2000. The cells were also cotransfected with β-gal reporter plasmids. The
10
reporter activity was analyzed as described above.
11
Immunoblotting analysis
12
Total protein was prepared by homogenizing the NTS tissue in lysis buffer containing a
13
protease inhibitor cocktail and a phosphatase inhibitor cocktail and subsequently incubating
14
the sample for 1 hour at 4 °C. The protein extracts (20 µg/sample based on the BCA protein
15
assay, Pierce Chemical Co., Rockford, IL, USA) were resolved on a 6% polyacrylamide gel
16
and transferred to a PVDF membrane (GE Healthcare, Buckinghamshire, UK). The
17
membranes were incubated in the appropriate anti-P-LRP6S1490 (2568), anti-P-AktS473 (4060),
18
anti-P-ribosomal protein S6 kinase (RSK)T359/S363 (9344), anti-P-ERK1/2T202/Y204 (4370),
19
anti-LRP6 (2560), anti-Akt (9272), anti-RSK (9341), anti-ERK1/2 (137F5), anti-P-IRS1S332 8
Page 8 of 41
Page 9 of 41
Diabetes
anti-P-GSK-3βS9
1
(2580)
2
anti-GSK-3β (07-389), anti-IRS1 (05-784R), anti-Un-P-β-catenin (05-665), anti-β-catenin
3
(AB19022)(Millipore, Billerica, MA, USA) and anti-P-β-cateninSer33 (SC-16743), anti-Dvl1
4
(Santa Cruz Biotechnology, Dallas, TX, USA) antibodies. Then, the membranes were then
5
incubated in an HRP-labeled goat anti-rabbit secondary antibody at 1:10,000. The
6
membranes were developed using the ECL-Plus detection kit (GE Healthcare).
(Cell
Signaling
Technology,
MA,
USA),
(05-643),
7 8
Immunohistochemistry analysis
9
The sections were deparaffinized, quenched in 3% H2O2/methanol, heated (using a
10
microwave) in citrate buffer (10 mmol/L, pH 6.0), blocked in 5% goat serum and incubated
11
in the anti-P-LRP6S1490 antibody overnight at 4 °C. Next, the sections were incubated in a
12
biotinylated secondary antibody (1:200; Vector Laboratories, Burlingame, CA, USA) for 1
13
hour and in AB complex (1:100) for 30 min at room temperature. The sections were
14
visualized using a DAB substrate kit (Vector Laboratories) and were counterstained with
15
hematoxylin. Then, the sections were photographed using a microscope equipped with a
16
charge-coupled device camera.
17 18 19
Co-immunoprecipitation and in vitro kinase assays The Catch and Release Reversible Immunoprecipitation System (Millipore) was used 9
Diabetes
1
according to the manufacturer's instructions. The proteins were eluted in 70 µl of elution
2
buffer and subjected to immunoblotting analysis using the anti-GSK-3β and anti-P-IRS1S332
3
antibodies.
4
For the in vitro kinase assay, co-immunoprecipitation using the anti-GSK-3β and
5
anti-IRS1 antibodies was performed, and the GSK-3β-IRS1 complex was eluted using the
6
Catch and Release Reversible Immunoprecipitation System. The kinase reaction was initiated
7
by adding kinase buffer and the terminated by adding 2X sample buffer and boiling for 10
8
minutes. The phosphorylation of IRS1 was determined via immunoblotting analysis using the
9
anti-P-IRS1S332 and anti-IRS1 antibodies.
10 11
Immunofluorescence staining analysis
12
The brainstem sections were incubated in the anti-GSK-3β (1:200) and anti-IRS1S332
13
(1:50) antibodies. After washing with PBS, the sections were incubated in green-fluorescent
14
Alexa Fluor-488 anti-rabbit IgG (1:200; Invitrogen) at 37 °C for 2 hours. The sections were
15
analyzed using a confocal microscope and Zeiss LSM Image (Carl Zeiss Micro Imaging)
16
software.
17 18 19
Statistical analysis Unpaired Student's t-test was used to compare protein levels (SHR and WKY group, 10
Page 10 of 41
Page 11 of 41
Diabetes
1
WKY and fructose treated group, SHR and SHR + TWS119 group or fructose and fructose +
2
TWS119 group) and the BP measurements (SHR and SHR + TWS119 group or fructose and
3
fructose + TWS119 group). One-way analysis of variance (ANOVA) followed by Scheffe
4
post-hoc analysis were applied to compare the differences between groups. Differences in
5
which P
