basic research
http://www.kidney-international.org & 2014 International Society of Nephrology
BMP type I receptor inhibition attenuates endothelial dysfunction in mice with chronic kidney disease Hidemi Kajimoto1, Hisashi Kai1, Hiroki Aoki2, Hiroki Uchiwa1, Yuji Aoki1, Suguru Yasuoka1, Takahiro Anegawa1, Yuji Mishina3, Akira Suzuki4, Yoshihiro Fukumoto1 and Tsutomu Imaizumi1,5 1
Division of Cardio–Vascular Medicine, Department of Internal Medicine, Kurume University, Kurume, Japan; 2Cardiovascular Research Institute, Kurume University, Kurume, Japan; 3Department of Biologic and Materials Sciences, School of Dentistry, University of Michigan, Ann Arbor, Michigan, USA; 4Division of Cancer Genetics, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan and 5International University of Health and Welfare/Fukuoka Sanno Hospital, Fukuoka, Japan
The molecular mechanisms of endothelial dysfunction and vascular calcification have been considered independently and potential links are currently unknown in chronic kidney disease (CKD). Bone morphogenetic protein (BMP) receptor signaling mediates calcification of atherosclerotic plaques. Here we tested whether BMP receptor signaling contributes to endothelial dysfunction, as well as to osteogenic differentiation of vascular smooth muscle cells (VSMCs), in a model of short-term CKD. In C57BL/6 mice, subtotal nephrectomy activated BMP receptor and increased phosphatase-and-tensin homolog (PTEN) protein in the endothelial cells and medial VSMCs without vascular remodeling in the aorta. In the endothelial cells, PTEN induction led to inhibition of the Akt-endothelial nitric oxide synthase (eNOS) pathway and endothelial dysfunction. In VSMCs, the PTEN increase induced early osteogenic differentiation. CKD-induced inhibition of eNOS phosphorylation and the resultant endothelial dysfunction were inhibited in mice with endothelial cell-specific PTEN ablation. Knockout of the BMP type I receptor abolished endothelial dysfunction, the inhibition of eNOS phosphorylation, and VSMC osteogenic differentiation in mice with CKD. A small molecule inhibitor of BMP type I receptor, LDN-193189, prevented endothelial dysfunction and osteogenic differentiation in CKD mice. Thus, BMP receptor activation is a mechanism for endothelial dysfunction in addition to vascular osteogenic differentiation in a short-term CKD model. PTEN may be key in linking BMP receptor activation and endothelial dysfunction in CKD. Kidney International (2015) 87, 128–136; doi:10.1038/ki.2014.223; published online 25 June 2014 KEYWORDS: bone morphogenetic protein receptor activation; chronic kidney disease; endothelial dysfunction; nitric oxide synthase; phosphataseand-tensin homolog; signal transduction
Correspondence: Hisashi Kai, Division of Cardio–Vascular Medicine, Department of Internal Medicine, Kurume University, 67 Asahi-machi, Kurume 830-0011, Japan. E-mail: [email protected] Received 10 September 2013; revised 12 April 2014; accepted 8 May 2014; published online 25 June 2014 128
Chronic kidney disease (CKD) is a worldwide health problem because of its high morbidity and mortality caused by cardiovascular complications.1 Endothelial dysfunction and medial osteogenic differentiation are considered the initial steps of major cardiovascular diseases in CKD patients.2–6 The molecular mechanisms for each type of vascular damage have been poorly understood. Furthermore, the interrelation of these vascular damages is largely unknown. In patients with suspected coronary artery diseases, strong correlation has been shown between impairment of endothelium-dependent flow-mediated vasodilation and the degrees of coronary artery calcification.7 Thus, we speculated that these two critical complications of CKD—namely, endothelial dysfunction and vascular calcification—share a common mechanism. Bone morphogenetic protein (BMP) receptor signaling is a key pathway in bone metabolism.8 BMPs interact with type I receptor kinases including BMPR1A, the activin-like kinase (ALK) 3, to induce Smad1/5/8 phosphorylation.9 Phosphorylated Smad1/5/8 translocates into the nucleus to regulate gene transcription in osteoblasts and VSMCs.8,10 BMP2 is expressed in calcified human atherosclerotic plaques.11 Serum obtained from patients undergoing dialysis increases BMP2 secretion from cultured VSMCs and prompts osteogenic differentiation of VSMCs.12 These findings suggest that activation of BMP receptor signaling contributes to osteogenic differentiation of the VSMCs in CKD. We hypothesized that activation of BMP receptor signaling is also a potential mechanism leading to endothelial dysfunction in CKD. RESULTS CKD activates BMP receptor signaling in endothelial cells (ECs) and VSMCs
In the screening assay, serum obtained from CKD patients showed robust induction of phosphorylated Smad1/5/8 in human umbilical vein endothelial cells (HUVECs), as compared with serum from controls (Figure 1a). This finding suggests that BMP receptor signaling is activated in ECs by CKD. To clarify the roles of BMP receptor signaling in vivo, we created 5/6 nephrectomy (Nx) mice as a very short-term Kidney International (2015) 87, 128–136
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
Healthy Pt 1
2
2
CKD 3
4
5
p-Smad1/5/8 / Smad5
a
6 p-Smad1/5/8
49
Smad5
49 38
GAPDH
1.5 1 0.5 0
b
Healthy
CKD
WT+CKD
WT+Sham Merge
Smad5
*
Smad5
L
Merge
L Merge
p-Smad1/5/8
p-Smad1/5/8
Merge
L
L 20 µm
c
Mean intensity of p-Smad1/5/8 in nuclei (AU) 60
**
**
50 40 30 20 10 0
WT+Sham
WT+CKD
EC
WT+Sham
WT+CKD
VSMC
Figure 1 | Chronic kidney disease (CKD) activates bone morphogenetic protein receptor signaling in endothelial cells (ECs) and vascular smooth muscle cells (VSMCs). (a) Representative immunoblots and the pooled data demonstrating the effects of serum obtained from healthy subjects or CKD patients on the expression levels of total Smad5 and phosphorylated Smad1/5/8 (p-Smad1/5/8) in human umbilical vein endothelial cells. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. *Po0.05 versus healthy. (b, c) Immunofluorescence (b) and the pooled data (c) demonstrating the effects of Nx on the expression levels of Smad5 and p-Smad1/5/8 in the aortas of wild-type (WT) þ Sham and WT þ CKD mice. Green, Smad5 or p-Smad1/5/8; red, nuclei; and yellow, merged. Arrows indicate translocation of p-Smad1/5/8 into the nuclei. **Po0.01 versus WT þ Sham. AU, arbitrary unit.
CKD model. In wild-type C57BL/6J (WT) mice, Nx increased blood urea nitrogen and creatinine by 2.5- and 2-fold, respectively, but did not affect systolic blood pressure (SBP; Supplementary Figure S1 online). WT mice that received Nx (WT þ CKD) did not show vascular remodeling, such as intimal thickening, medial thickening, and adventitial fibrosis, and vascular inflammation (infiltration of inflammatory cells) in the aorta, nor did WT mice that had undergone sham operation (WT þ Sham) during the observation period (Supplementary Figure S2a and b online). In addition, CKD did not induce significant changes in mRNA Kidney International (2015) 87, 128–136
levels of cell-cell adhesion molecules and cell-matrix adhesion molecules, such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 in the aorta (Supplementary Table S1 online). In WT þ CKD mice, increased phosphorylation and nuclear translocation of Smad1/5/8, the indicators of BMP receptor activation, were demonstrated both in ECs and in the medial VSMCs without affecting the total amount of Smad5 (Figure 1b). Nuclear phosphorylated Smad1/5/8 levels were increased by fourfold in ECs and by twofold in VSMCs in WT þ CKD mice, compared with those in WT þ Sham mice (Figure 1c). Accordingly, it is suggested 129
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
that very short-term CKD created by Nx activates BMP signaling in ECs and VSMCs, preceding the development of vascular remodeling.
(Figure 3c). Such endothelium-dependent relaxation was impaired in WT þ CKD mice. In EC-Pten þ / mice with sham operation (EC-Pten þ / þ Sham), the endotheliumdependent relaxation was similar to that in WT þ Sham mice. The Nx-induced impairment of endothelium-dependent relaxation was abolished in EC-Pten þ / þ CKD mice. Endothelium-independent relaxation by nitroglycerin was similar among WT þ Sham, WT þ CKD, EC-Pten þ / þ þ/ Sham, and EC-Pten þ CKD (Supplementary Figure S3a online). In addition, the Nx-induced reduction in eNOS phosphorylation was prevented in EC-Pten þ / þ CKD mice (Figure 4). These findings suggest that very short-term CKD induces PTEN upregulation in ECs, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction.
Effects of BMP receptor activation on eNOS signaling in HUVECs
Next, we explored the role and consequences of BMP receptor activation in ECs. We treated HUVECs for 6 h with 30 or 300 ng/ml of BMP2, a known ligand of the BMP receptor. BMP2 treatment increased Smad1/5/8 phosphorylation and decreased eNOS phosphorylation at Ser1177, a key phosphorylation site that positively regulates eNOS enzyme activity, in HUVECs (Figure 2a). Interestingly, BMP2 treatment reduced phosphorylation levels of Ser473 in Akt, one of the major putative kinases for eNOS phosphorylation, in association with increases in protein levels of PTEN, an upstream protein phosphatase that reduces Akt phosphorylation and its activity (Figure 2a). As shown in Figure 2b, the number of PTEN-overexpressing ECs was increased by BMP2. Noteworthy, immunofluorescence staining revealed that BMP2 upregulated PTEN expression in the HUVECs with increased Smad1/5/8 phosphorylation levels (Figure 2c). To clarify the causal relationship between BMP receptor activation and PTEN-mediated Akt inactivation, we depleted PTEN using small interfering RNA. In HUVECs treated with nontargeting small interfering RNA (siControl), BMP2 reduced vascular endothelial growth factor-induced Akt phosphorylations (Figure 2d). PTEN depletion (siPten) abolished BMP2-induced inhibition of vascular endothelial growth factor-induced Akt phosphorylations. These results suggest that BMP receptor activation inhibits the Akt-eNOS pathway through PTEN induction in ECs. In vivo role of PTEN induction in eNOS signaling and endothelial function in CKD mice
For the next step, we investigated the in vivo roles of PTEN in impairment of the eNOS pathway and endothelial dysfunction in the aorta of CKD mice. WT þ Sham mice showed constitutive phosphorylations of eNOS and Akt (Figure 3a). On the other hand, eNOS and Akt phosphorylations were markedly reduced in WT þ CKD mice. Immunofluorescence staining revealed increased PTEN expression in the ECs of WT þ CKD mice, as compared with the ECs of WT þ Sham mice (Figure 3b). To determine the causal relationship between PTEN induction and endothelial dysfunction in CKD mice, ECdependent relaxation was investigated in WT þ Sham, WT þ CKD, EC-specific heterozygous null mice for Pten (ECPten þ / ) þ Sham, and EC-Pten þ / mice that received Nx (EC-Pten þ / þ CKD). EC-Pten þ / mice showed basal and Nx-elevated blood urea nitrogen and creatinine levels similar to those in WT mice (Supplementary Figure S1a and b online). SBP did not differ among the four groups (Supplementary Figure S1c online). In WT þ Sham mice, acetylcholine induced dose-dependent relaxation of aortic rings that were pre-contracted with phenylephrine 130
In vivo role of BMP receptor activation in endothelial function in CKD mice
To further define the causal relation between BMP receptor activation and endothelial dysfunction, the CKD model was created in heterozygous null mice for Alk3 (Alk3 þ / ). ALK3 is a BMP type I receptor9 that binds to a broad range of ligands in the BMP family, including BMP2, 4, 6, and 7.13–15 Alk3 þ / mice showed basal and Nx-elevated blood urea nitrogen and creatinine levels similar to those in WT mice (Supplementary Figure S1d and e online). There were no differences in SBP among WT þ Sham, WT þ CKD, Alk3 þ / mice with sham operation (Alk3 þ / þ Sham), and Alk3 þ / mice that received Nx (Alk3 þ / þ CKD; Supplementary Figure S1f online). The endothelium-dependent relaxation of the aorta did not significantly differ between WT þ Sham and Alk3 þ / þ Sham mice (Figure 5a). On the other hand, Alk3 þ / mice did not show Nx-induced endothelial dysfunction. Endothelium-independent relaxation by nitroglycerin did not differ among WT þ Sham, WT þ CKD, Alk3 þ / þ Sham, and Alk3 þ / þ CKD (Supplementary Figure S3 online). Aortic eNOS phosphorylation levels were similar between WT þ Sham and Alk3 þ / þ Sham mice. The Nx-induced reduction in eNOS phosphorylation was prevented in Alk3 þ / mice (Figure 4). Because the serum phosphate levels might affect the CKD-induced endothelial dysfunction, we investigated the serum phosphate levels. There were no differences among WT þ Sham, WT þ CKD, Alk3 þ / þ Sham, and Alk3 þ / þ CKD (Supplementary Figure S3b online). Roles of BMP receptor activation in VSMC osteogenic differentiation in CKD mice
We investigated the role of the BMP receptor/Smad pathway in VSMCs of CKD mice because the pathway is known to mediate VSMC osteogenic differentiation in the atheromatous plaques.11 As shown in Figure 1b and c, CKD increased phosphorylation and nuclear translocation of Smad1/5/8 in the medial VSMCs without affecting the total amount of Smad5, indicating the activation of the BMP receptor/Smad pathway. Moreover, expression levels of Msx2, an osteogenic Kidney International (2015) 87, 128–136
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
BMP2 30
300
60
12
p-Smad1/5/8 160 60
PTEN mean intensity
p-eNOS (Ser1177) p-Akt (Ser473) PTEN
49
Vehicle
0 0.00 %
12
93.75 %
0
10
0
17
PTEN
Sec ab only
0 42.11 %
0.00 %
4
0.00 %
57.89 %
0
10
17
50
0.00 %
0.00 %
8
4
4
0
GAPDH
BMP2
0
8
8
0.00 %
38
12
6.25 %
% Over threshold
0
0.00 %
100.00 %
0 0
10
17
**
40 30 20 10 0 BMP2
Vehicle
DAPI mean intensity
p-Smad1/5/8
PTEN
DAPI
p-Smad1/5/8+PTEN
Vehicle
BMP2
100 µm siControl –
BMP2 VEGF
–
siPten +
+
–
– +
–
+ +
–
+ PTEN
50 60 60
p-Akt (Ser473)
short exposure
p-Akt (Ser473)
long exposure
GAPDH 38
Figure 2 | Bone morphogenetic protein (BMP) receptor activation inhibits the Akt-endothelial nitric oxide synthase (eNOS) pathway by upregulating phosphatase-and-tensin homolog (PTEN) expression in endothelial cells (ECs). (a) Representative immunoblots showing the effects of BMP2 on the basal phosphorylation levels of eNOS, Smad1/5/8, Akt, and PTEN. (b) Scatter analysis of PTEN-labeled ECs. The pooled data showing the percentages of the mean fluorescence intensity of PTEN over the threshold. **Po0.01 versus vehicle. (c) Immunofluorescence staining demonstrating the effects of BMP2 on the expression levels of p-Smad1/5/8 (green) and PTEN (red). Blue, nuclei. The merged image shows that BMP2 increased p-Smad1/5/8 and PTEN in the same cells with nucleus stained as yellow. (d) Representative immunoblots showing basal and vascular endothelial growth factor (VEGF)-stimulated Akt phosphorylation with or without BMP2 stimulation after transfection with Pten small interfering RNA (siRNA; siPten) or non-targeting siRNA (siControl). Photographs of p-Akt immunoblots were obtained from the same blot with short (upper panel) and long (lower panel) exposures. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; DAPI, 40 -6-diamidino-2-phenylindole. Kidney International (2015) 87, 128–136
131
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
WT+CKD
WT+Sham 160
p-eNOS (Ser1177)
160
p-eNOS/eNOS
1.5
1
eNOS
62
0
p-Akt/Akt
Akt
38
GAPDH
PTEN
WT+CKD
1
** 0.5
0
WT+Sham
WT+Sham
1.5
p-Akt (Ser473)
62
*
0.5
WT+Sham
WT+CKD
WT+CKD PTEN
Merge 100 µm
L
Merge
L
20
Mean intensity of PTEN in EC (AU)
**
15 20 µm
L
10
L
5 0
WT+CKD
§§
0
*
§§
* 20
% Relaxation
WT+Sham
§
* # ¶
40 WT+Sham WT+CKD EC-Pten +/–+Sham EC-Pten +/–+CKD
60
80
100 –8
–7
–6
–5
–4
Ach log (mol/l)
Figure 3 | Nx increases phosphatase-and-tensin homolog (PTEN) expression and inactivates endothelial nitric oxide synthase (eNOS) in mice. (a) Representative immunoblots and the pooled data showing the effects of chronic kidney disease (CKD) on eNOS signaling in the mouse aorta. *Po0.05 versus wild-type (WT) þ Sham. **Po0.01 versus WT þ Sham. (b) Immunofluorescence and the pooled data demonstrating the effects of CKD on the expression levels of PTEN (green) in the aorta. Blue, nuclei. Arrows indicate endothelial cells (ECs) overexpressing PTEN. **Po0.01 versus WT þ Sham. (c) Pooled data of the acetylcholine (Ach)-induced relaxation in phenylephrine-pretreated rings of the descending aorta obtained from WT þ Sham (n ¼ 9), WT þ CKD (n ¼ 4), EC-Pten þ / þ Sham (n ¼ 12), and EC-Pten þ / þ CKD (n ¼ 11) mice. *Po0.05 WT þ Sham versus WT þ CKD, yPo0.05, and yyPo0.01 WT þ CKD versus EC-Pten þ / þ CKD; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham versus WT þ CKD, and zPo0.01 WT þ CKD versus EC-Pten þ / þ CKD; two-way analysis of variance. AU, arbitrary unit.
transcription factor and an established downstream gene of BMP receptor signaling, were upregulated in the aorta of WT þ CKD mice (Supplementary Figure S4a online). In 132
addition, alkaline phosphatase activity was increased in medial VSMCs, particularly in the inner layers of the media, of WT þ CKD mice, but not in those of WT þ Sham mice Kidney International (2015) 87, 128–136
basic research
p-eNOS (Ser1177)
160
eNOS
38
GAPDH
§
*
20
0
§
*
§
§ #¶
40 60 80
** p-eNOS/eNOS
1.5
1
0.5
0 Sham WT
CKD
Sham
CKD
Alk3 +/–
Sham
CKD
EC-Pten +/–
Figure 4 | Representative immunoblots and the pooled data demonstrating the effects of Nx on the expression levels of endothelial nitric oxide synthase (eNOS) and phosphorylated eNOS (p-eNOS) in the aortas of wild-type (WT) þ Sham, WT þ chronic kidney disease (CKD), Alk3 þ / þ Sham, Alk3 þ / þ CKD, EC-Pten þ / þ Sham, and EC-Pten þ / þ CKD mice. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. *Po0.05 WT þ Sham versus WT þ CKD, **Po0.01 WT þ CKD versus Alk3 þ / þ CKD, **Po0.01 WT þ CKD versus EC-Pten þ / þ CKD.
(Supplementary Figure S4b online). On the other hand, heterozygous deletion of Alk3 abolished Nx-induced alkaline phosphatase activation in the aorta. These results suggest that the BMP receptor also has a role in VSMC osteogenic differentiation in very short-term CKD mice without atherosclerotic changes. Effects of PTEN induction on osteogenic differentiation in VSMCs
Interestingly, immunofluorescence staining revealed increased PTEN expression in the VSMCs of WT þ CKD mice, as compared with the VSMCs of WT þ Sham mice (Figure 3b). BMP2 treatment increased mRNA levels of osteogenic transcriptional factor Msx2 and Pten in cultured VSMCs (Supplementary Figure S5a and b online). We investigated the relationship between PTEN induction and osteogenic differentiation in VSMCs by using small interfering RNA targeting Pten (siPten; Supplementary Figure S5c and d online). In VSMCs treated with nontargeting small interfering RNA (siControl), BMP2 increased mRNA levels of Msx2 and Pten. Although the Pten reduction by siPten was partial, siPten significantly reduced the BMP2-induced induction of Msx2 in VSMCs. These results suggest that Kidney International (2015) 87, 128–136
–7 –6 –5 –4 Ach log (mol/l) WT+Sham WT+CKD Alk3 +/– +Sham Alk3 +/– +CKD
*
*
§ 40
#¶
§ §§
60
100 –8
**
*
80
100 2
*
20
*
% Relaxation
160
§
0
% Relaxation
W T+ C KD Al k3 +/ – +S ha Al m k3 +/ – +C KD EC -P te n+ /– +S EC ha -P m te n+ /– +C KD
W
T+ Sh a
m
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
–8
–7 –6 –5 –4 Ach log (mol/l) WT+Sham+Vehicle WT+CKD+Vehicle WT+Sham+LDN WT+CKD+LDN
Figure 5 | Effects of bone morphogenetic protein (BMP) receptor signaling inhibition on the chronic kidney disease (CKD)-induced endothelial dysfunction. (a) Pooled data of acetylcholine (Ach)induced relaxation in phenylephrine (PE)-pretreated rings of the descending aorta obtained from wild-type (WT) þ Sham (n ¼ 9), WT þ CKD (n ¼ 4), Alk3 þ / þ Sham (n ¼ 4), and Alk3 þ / þ CKD (n ¼ 7) mice. *Po0.05 WT þ Sham versus WT þ CKD, yPo0.05 WT þ CKD versus Alk3 þ / þ CKD; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham versus WT þ CKD, zPo0.01 WT þ CKD versus Alk3 þ / þ CKD; two-way analysis of variance (ANOVA). (b) Pooled data of Ach-induced relaxation in PE-pretreated rings of the descending aorta obtained from Sham þ vehicle (n ¼ 8), Sham þ LDN193189 (n ¼ 6), CKD þ vehicle (n ¼ 5), and CKD þ LDN-193189 (n ¼ 7) mice. *Po0.05 WT þ Sham þ vehicle versus WT þ CKD þ vehicle, yPo0.05, and yyPo0.01 WT þ CKD þ vehicle versus WT þ CKD þ LDN-193189; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham þ vehicle versus WT þ CKD þ vehicle, and z Po0.01 WT þ CKD þ vehicle versus WT þ CKD þ LDN-193189; two-way ANOVA.
PTEN activation would have a role in osteogenic differentiation in VSMC. Effects of selective BMP type I receptor inhibitor
To explore potential clinical interventions, we investigated the effects of LDN-193189 (4-[6-(4-piperazin-1-ylphenyl) pyrazolo [1,5-a]pyrimidin-3-yl] quinolone, 3 mg/kg), a small molecule inhibitor of BMP type I receptor kinases,16,17 in Nxinduced endothelial dysfunction and VSMC osteogenic differentiation in WT mice. LDN-193189 administration did not affect renal function and SBP (Supplementary Figure S6a–c online). This LDN-193189 treatment regimen prevented endothelial dysfunction in WT þ CKD mice without affecting endothelial function in WT þ Sham mice (Figure 5b). Also, LDN-193189 inhibited osteogenic differentiation of medial VSMCs in the CKD mice (Supplementary Figure S4b online). DISCUSSION
The novel findings of the present study are as follows: (i) BMP receptor signaling is involved in endothelial dysfunction, in addition to osteogenic differentiation of medial VSMCs, in a mouse model of very short-term CKD without 133
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
apparent vascular inflammation and remodeling. (ii) CKD induces PTEN upregulation through BMP receptor activation in ECs, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction. (iii) A selective inhibitor of BMP type I receptor kinases, LDN193189, prevents endothelial dysfunction and abolishes osteogenic differentiation of VSMCs in CKD mice. These findings suggest that BMP receptor activation may be a common initial mechanism for endothelial dysfunction and vascular calcification, which is activated even in the early stage of CKD. In the present study, we used Nx mice with 2.5- and 2-fold increases in blood urea nitrogen and creatinine, respectively, without SBP elevation and serum phosphate change, which suggests that it is a very short-term CKD model representing moderate renal failure without hypertension. This study was performed in young mice, and the observation period was rather short, because we focused on investigating the relationship between BMP receptor signaling and endothelial dysfunction in the CKD model before cardiovascular damages and remodeling had developed. The most important novel finding of this study is that BMP receptor activation is crucial for the mechanism of endothelial dysfunction in the early stage of CKD. The causal relation between BMP receptor activation and endothelial dysfunction is clearly presented by the facts that the genetic Alk3 deletion and LDN-193189 treatment abolished the CKD-induced endothelial dysfunction (Figure 5). In this study, we focused on the eNOS pathway as the mechanism of endothelial dysfunction, because we have previously demonstrated that impairment of eNOS activation is important for endothelial dysfunction in this CKD model.18 It is noteworthy that PTEN is the key molecule linking BMP receptor activation and endothelial dysfunction, because PTEN was upregulated in the ECs of CKD mice and the Pten deletion prevented the CKD-induced eNOS dephosphorylation and endothelial dysfunction in mice (Figures 3b, c, and 4). As shown in Figure 3a and b, the CKD-induced PTEN upregulation might decrease eNOS phosphorylation levels via the reduction of Akt activity. This hypothesis is supported by the fact that PTEN depletion using siPten abolished the inhibition of the agonist-induced Akt activation by BMP receptor activation in HUVECs (Figure 2d). Accordingly, it is suggested that CKD induces upregulation of PTEN in ECs as a consequence of activation of BMP receptor signaling, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction. There is increasing evidence that the BMP family participates in various types of ectopic ossification.19 With regard to atherosclerosis, upregulated BMP2 expression was found in calcified human atherosclerotic plaques.11 In addition, the role of the BMP receptor signal was suggested in the plaque calcification in apolipoprotein E-deficient mice receiving high-fat diet and bilateral ovariectomy and in highfat diet-fed LDL receptor-deficient mice.17,20 In this study, not only the activation of the downstream of the BMP
receptor (Smad1/5/8 phosphorylation and Msx2 upregulaiton) but also the osteogenic differentiation (alkaline phosphatase activation) was induced in the medial VSMCs in the early stage of CKD mice (Figure 1b and c and Supplementary Figure S4 online). Moreover, the genetic deletion of Alk3 and pharmacological inhibition of the BMP receptor with LDN-193189 abolished the VSMC osteogenic differentiation (Supplementary Figure S4b online). Taken together, the present study has presented for the first time that the BMP receptor signal has a role in the VSMC osteogenic differentiation in the early stage of CKD. It has been established that, upon binding to BMPs, BMP type I receptors are activated by BMP type II receptors and subsequently activate the Smad1/5/8-dependent and -independent signaling pathways.21–23 Derwall et al. have demonstrated that BMP type I receptor inhibition by LDN-193189 prevents vascular inflammation, atherosclerotic plaque formation, and plaque calcification in high-fat diet-fed LDL receptor–deficient mice.16,17 The mechanism for these effects of LDN-193189 was explained by the reductions in vascular oxidative stress, plaque inflammation, and hepatic LDL biosynthesis. Recently, Kim et al. have shown that genetic ablation of BMP type II receptors exaggerated endothelial inflammatory and atherosclerotic plaque formation in highfat diet-fed ApoE-deficient mice.24 They suggested that the BMP type II depletion induced endothelial inflammation in a ligand-independent manner. Currently, there is no explanation for the discrepancy between the anti-inflammatory and antiatherogenic effects of the BMP type I receptor inhibition and the pro-inflammatory and pro-atherogenic effects of the BMP type II receptor reduction in high-fat dietfed mice prone to atherosclerosis. In the current study, CKD mice had no evidence of vascular inflammation and remodeling, probably because this is a very short-term CKD model. Therefore, it is impossible to determine the roles of BMP receptors in vascular inflammation and remodeling in the later stage of CKD from this study. Because endothelial dysfunction is the initial step of atherosclerotic lesion formation, it is interesting to determine whether the BMP receptor–mediated endothelial dysfunction would be also involved in the earlier stage of atherogenesis from analogy to CKD, as shown in the present study. This issue should be addressed in future studies. There are several limitations in the present study. As shown by this study and others (Figure 1a and Chen et al.12), serum obtained from CKD patients highly stimulates BMP receptor signaling activity in ECs and VSMCs. It awaits future endeavor to identify ligands in uremic serum to activate this cascade. Second, in CKD animal studies, the kidney disease is usually allowed to evolve over at least 8–12 weeks. Thus, this model is regarded as a model of very short-term CKD. This study has clearly shown that the BMP receptors have a role in the prevention of endothelial dysfunction and vascular osteogenic differentiation before the development of histological lesions in the early stage of CKD. However, it remains unclear whether blocking the BMP receptor could also have a
134
Kidney International (2015) 87, 128–136
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
basic research
significant effect once vascular lesions already exist and whether the long-term blocking of the BMP receptor would be useful as well. Third, we clearly demonstrated that PTEN is a downstream effector molecule of activated BMP receptor signaling because endothelial-specific removal of Pten prevents CKD-induced endothelial dysfunction. The mechanism whereby BMP receptor activation results in increases in PTEN remains to be elucidated. It is noteworthy that BMP2 treatment increases PTEN levels by decreasing protein degradation in MCF7 human breast cancer cells.25 Thus, it is possible that the post-translational mechanism would be involved in the CKD-induced PTEN protein upregulation. Fourth, it was suggested that PTEN induction would participate in the BMP2-induced VSMC osteogenic differentiation (Supplementary Figure S5 online). Future experiments using VSMC-specific null mice for PTEN are necessary to determine the in vivo role of PTEN in VSMC osteogenic differentiation and calcification in CKD. Finally, we used mice heterozygous null for Pten or Alk3, because both mice homozygous null allele are lethal.26,27 Having phenotypic rescues by heterozygous deletion indicates that small decreases in PTEN expression or BMP receptor signaling may preserve endothelial function in the early stage of CKD. The fact that a suboptimal dose of LDN-193189 can suppress endothelial dysfunction also strongly supports this notion. The present study suggests that BMP receptor signaling has a pivotal role in endothelial dysfunction, in addition to osteogenic differentiation of VSMCs, in the early stage of CKD without vascular remodeling. An increase in PTEN protein is involved in the mechanism by which BMP receptor activation inhibits Akt-eNOS signaling in CKD. Our observations suggest that BMP receptor signaling would be a promising target for the prevention and treatment of both endothelial dysfunction and vascular calcification in the early stage of CKD patients.
week later. Animal experiments included the following groups: WT þ Sham, WT þ CKD, EC-Pten þ / þ Sham, EC-Pten þ / þ þ/ CKD, Alk3 þ Sham, Alk3 þ / þ CKD, WT mice that received Nx and LDN-193189 (WT þ CKD þ LDN), and WT mice that received Nx and the vehicle (WT þ CKD þ vehicle). To abrogate PTEN expression in ECs, EC-Pten þ / mice, namely Tie2CrePtenflox/ þ mice, were used because Tie2CrePtenflox/flox mice die before embryonic day 11.5 because of bleeding and cardiac failure.27 To negate BMP receptor signaling, we used Alk3 þ / mice because the homozygous mutation is embryonic lethal because of failure to produce mesoderm.26
MATERIALS AND METHODS A detailed method section is available in the Supplementary Materials online. Mice and human samples WT mice were purchased from Charles River Laboratories (Yokohama, Japan). EC-Pten þ / mice and Alk3 þ / mice were constructed as described previously.26,27 Mice were fed with normal chow (normal phosphate diet). LDN-193189 was obtained from Axon Medchem BV (Groningen, The Netherlands). HUVECs and VSMCs were purchased from Lonza (Basel, Switzerland). The study protocol of animal studies was approved by the Animal Care and Treatment Committee of Kurume University. The patient study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of Kurume University. CKD models CKD was established by performing 5/6 Nx in mice as described previously.18 Briefly, mice were anesthetized with 1.5% isoflurane by inhalation. Nx was established by surgical resection of the upper and lower thirds of the left kidney at 10–12 weeks, followed by right Nx 1 Kidney International (2015) 87, 128–136
Evaluation of animal model Four weeks after the establishment of Nx, SBP and heart rate were measured using a tail-cuff sphygmomanometer (MK-2000ST; Muromachi, Tokyo, Japan), as described previously.18 One day after measurements of SBP and heart rate, mice were killed by an overdose of pentobarbital (100 mg/kg, intraperitoneal). Blood was collected from the right atrium, and then mice were perfused with ice-cold saline for 5 min. The aorta was immediately removed and subjected to isometric tension study or immunoblot analysis. Expression analysis Expression levels of mRNA and protein were evaluated as described in the Supplementary Materials online. Statistical analysis Results are shown as mean±s.e.m, unless indicated otherwise. Intergroup differences were assessed by the Mann–Whitney U-test or two-way ANOVA followed by Tukey–Kramer’s post-hoc analysis. A value of Po0.05 was considered statistically significant. DISCLOSURE
All the authors declared no competing interest. ACKNOWLEDGMENTS
This study was supported, in part, by a grant for the Science Frontier Research Promotion Centers (Cardiovascular Research Institute), Grants-in-Aid for Scientific Research (HK and TI) from the Ministry of Education, Science, Sports, and Culture, Japan, and research grants from the Kidney Foundation (HK). We thank Katsue Shiramizu, Miyuki Nishigata, Kimiko Kimura, Miho Nakao, and Makiko Kiyohiro for technical assistance. SUPPLEMENTARY MATERIAL Figure S1. Effects of Nx on BUN, serum creatinine, and SBP in WT, EC-Pten+/ , and Alk3+/ mice at 4 weeks after Nx. Figure S2. Effects of Nx on BUN, serum creatinine, and SBP in WT, EC-Pten+/ , and Alk3+/ mice at 4 weeks after Nx. Figure S3. (a) Endothelium independent relaxation of aortic rings obtained from WT, EC-Pten+/ , and Alk3+/ mice. (b) Effects of Nx on serum phosphate in WT and Alk3+/ mice at 4 weeks after Nx. Figure S4. BMP receptor inhibition prevents Nx-induced osteogenic differentiation of medial VSMCs. Figure S5. Effects of PTEN induction on BMP receptor activationinduced osteogenic differentiation in VSMCs. Figure S6. Effects of LDN-193189 on BUN, serum creatinine, and SBP in WT mice at 2 weeks of treatment after Nx. Table S1. Effects of Nx on cell adhesion molecules in WT mice. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki 135
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
REFERENCES
14.
1.
2. 3.
4.
5.
6.
7.
8. 9. 10.
11.
12.
13.
136
Keith DS, Nichols GA, Gullion CM et al. Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004; 164: 659–663. Morris ST, McMurray JJ, Spiers A et al. Impaired endothelial function in isolated human uremic resistance arteries. Kidney Int 2001; 60: 1077–1082. Stam F, van Guldener C, Becker A et al. Endothelial dysfunction contributes to renal function-associated cardiovascular mortality in a population with mild renal insufficiency: the Hoorn study. J Am Soc Nephrol 2006; 17: 537–545. Blacher J, Guerin AP, Pannier B et al. Arterial calcifications, arterial stiffness, and cardiovascular risk in end-stage renal disease. Hypertension 2001; 38: 938–942. London GM, Guerin AP, Marchais SJ et al. Arterial media calcification in end-stage renal disease: impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant 2003; 18: 1731–1740. Okuno S, Ishimura E, Kitatani K et al. Presence of abdominal aortic calcification is significantly associated with all-cause and cardiovascular mortality in maintenance hemodialysis patients. Am J Kidney Dis 2007; 49: 417–425. Huang PH, Chen LC, Leu HB et al. Enhanced coronary calcification determined by electron beam CT is strongly related to endothelial dysfunction in patients with suspected coronary artery disease. Chest 2005; 128: 810–815. Wozney JM, Rosen V, Celeste AJ et al. Novel regulators of bone formation: molecular clones and activities. Science 1988; 242: 1528–1534. Shi Y, Massague J. Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 2003; 113: 685–700. Speer MY, Yang HY, Brabb T et al. Smooth muscle cells give rise to osteochondrogenic precursors and chondrocytes in calcifying arteries. Circ Res 2009; 104: 733–741. Bostrom K, Watson KE, Horn S et al. Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest 1993; 91: 1800–1809. Chen NX, Duan D, O’Neill KD et al. The mechanisms of uremic seruminduced expression of bone matrix proteins in bovine vascular smooth muscle cells. Kidney Int 2006; 70: 1046–1053. Yamashita H, ten Dijke P, Huylebroeck D et al. Osteogenic protein-1 binds to activin type II receptors and induces certain activin-like effects. J Cell Biol 1995; 130: 217–226.
15.
16.
17.
18.
19. 20.
21. 22. 23.
24.
25.
26.
27.
Ebisawa T, Tada K, Kitajima I et al. Characterization of bone morphogenetic protein-6 signaling pathways in osteoblast differentiation. J Cell Sci 1999; 112(Pt 20): 3519–3527. ten Dijke P, Yamashita H, Sampath TK et al. Identification of type I receptors for osteogenic protein-1 and bone morphogenetic protein-4. J Biol Chem 1994; 269: 16985–16988. Yu PB, Deng DY, Lai CS et al. BMP type I receptor inhibition reduces heterotopic [corrected] ossification. Nat Med 2008; 14: 1363–1369. Derwall M, Malhotra R, Lai CS et al. Inhibition of bone morphogenetic protein signaling reduces vascular calcification and atherosclerosis. Arterioscler Thromb Vasc Biol 2012; 32: 613–622. Kajimoto H, Kai H, Aoki H et al. Inhibition of eNOS phosphorylation mediates endothelial dysfunction in renal failure: new effect of asymmetric dimethylarginine. Kidney Int 2012; 81: 762–768. Reddi AH. Bone morphogenetic proteins: from basic science to clinical applications. J Bone Joint Surg Am 2001; 83-A (Suppl 1): S1–S6. Osako MK, Nakagami H, Koibuchi N et al. Estrogen inhibits vascular calcification via vascular RANKL system: common mechanism of osteoporosis and vascular calcification. Circ Res 2010; 107: 466–475. Hogan BL. Bone morphogenetic proteins in development. Curr Opin Genet Dev 1996; 6: 432–438. Massague J. How cells read TGF-beta signals. Nat Rev Mol Cell Biol 2000; 1: 169–178. Fujii M, Takeda K, Imamura T et al. Roles of bone morphogenetic protein type I receptors and Smad proteins in osteoblast and chondroblast differentiation. Mol Biol Cell 1999; 10: 3801–3813. Kim CW, Song H, Kumar S et al. Anti-inflammatory and antiatherogenic role of BMP receptor II in endothelial cells. Arterioscler Thromb Vasc Biol 2013; 33: 1350–1359. Waite KA, Eng C. BMP2 exposure results in decreased PTEN protein degradation and increased PTEN levels. Hum Mol Genet 2003; 12: 679–684. Mishina Y, Suzuki A, Ueno N et al. Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev 1995; 9: 3027–3037. Hamada K, Sasaki T, Koni PA et al. The PTEN/PI3K pathway governs normal vascular development and tumor angiogenesis. Genes Dev 2005; 19: 2054–2065.
Kidney International (2015) 87, 128–136
http://www.kidney-international.org & 2014 International Society of Nephrology
BMP type I receptor inhibition attenuates endothelial dysfunction in mice with chronic kidney disease Hidemi Kajimoto1, Hisashi Kai1, Hiroki Aoki2, Hiroki Uchiwa1, Yuji Aoki1, Suguru Yasuoka1, Takahiro Anegawa1, Yuji Mishina3, Akira Suzuki4, Yoshihiro Fukumoto1 and Tsutomu Imaizumi1,5 1
Division of Cardio–Vascular Medicine, Department of Internal Medicine, Kurume University, Kurume, Japan; 2Cardiovascular Research Institute, Kurume University, Kurume, Japan; 3Department of Biologic and Materials Sciences, School of Dentistry, University of Michigan, Ann Arbor, Michigan, USA; 4Division of Cancer Genetics, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan and 5International University of Health and Welfare/Fukuoka Sanno Hospital, Fukuoka, Japan
The molecular mechanisms of endothelial dysfunction and vascular calcification have been considered independently and potential links are currently unknown in chronic kidney disease (CKD). Bone morphogenetic protein (BMP) receptor signaling mediates calcification of atherosclerotic plaques. Here we tested whether BMP receptor signaling contributes to endothelial dysfunction, as well as to osteogenic differentiation of vascular smooth muscle cells (VSMCs), in a model of short-term CKD. In C57BL/6 mice, subtotal nephrectomy activated BMP receptor and increased phosphatase-and-tensin homolog (PTEN) protein in the endothelial cells and medial VSMCs without vascular remodeling in the aorta. In the endothelial cells, PTEN induction led to inhibition of the Akt-endothelial nitric oxide synthase (eNOS) pathway and endothelial dysfunction. In VSMCs, the PTEN increase induced early osteogenic differentiation. CKD-induced inhibition of eNOS phosphorylation and the resultant endothelial dysfunction were inhibited in mice with endothelial cell-specific PTEN ablation. Knockout of the BMP type I receptor abolished endothelial dysfunction, the inhibition of eNOS phosphorylation, and VSMC osteogenic differentiation in mice with CKD. A small molecule inhibitor of BMP type I receptor, LDN-193189, prevented endothelial dysfunction and osteogenic differentiation in CKD mice. Thus, BMP receptor activation is a mechanism for endothelial dysfunction in addition to vascular osteogenic differentiation in a short-term CKD model. PTEN may be key in linking BMP receptor activation and endothelial dysfunction in CKD. Kidney International (2015) 87, 128–136; doi:10.1038/ki.2014.223; published online 25 June 2014 KEYWORDS: bone morphogenetic protein receptor activation; chronic kidney disease; endothelial dysfunction; nitric oxide synthase; phosphataseand-tensin homolog; signal transduction
Correspondence: Hisashi Kai, Division of Cardio–Vascular Medicine, Department of Internal Medicine, Kurume University, 67 Asahi-machi, Kurume 830-0011, Japan. E-mail: [email protected] Received 10 September 2013; revised 12 April 2014; accepted 8 May 2014; published online 25 June 2014 128
Chronic kidney disease (CKD) is a worldwide health problem because of its high morbidity and mortality caused by cardiovascular complications.1 Endothelial dysfunction and medial osteogenic differentiation are considered the initial steps of major cardiovascular diseases in CKD patients.2–6 The molecular mechanisms for each type of vascular damage have been poorly understood. Furthermore, the interrelation of these vascular damages is largely unknown. In patients with suspected coronary artery diseases, strong correlation has been shown between impairment of endothelium-dependent flow-mediated vasodilation and the degrees of coronary artery calcification.7 Thus, we speculated that these two critical complications of CKD—namely, endothelial dysfunction and vascular calcification—share a common mechanism. Bone morphogenetic protein (BMP) receptor signaling is a key pathway in bone metabolism.8 BMPs interact with type I receptor kinases including BMPR1A, the activin-like kinase (ALK) 3, to induce Smad1/5/8 phosphorylation.9 Phosphorylated Smad1/5/8 translocates into the nucleus to regulate gene transcription in osteoblasts and VSMCs.8,10 BMP2 is expressed in calcified human atherosclerotic plaques.11 Serum obtained from patients undergoing dialysis increases BMP2 secretion from cultured VSMCs and prompts osteogenic differentiation of VSMCs.12 These findings suggest that activation of BMP receptor signaling contributes to osteogenic differentiation of the VSMCs in CKD. We hypothesized that activation of BMP receptor signaling is also a potential mechanism leading to endothelial dysfunction in CKD. RESULTS CKD activates BMP receptor signaling in endothelial cells (ECs) and VSMCs
In the screening assay, serum obtained from CKD patients showed robust induction of phosphorylated Smad1/5/8 in human umbilical vein endothelial cells (HUVECs), as compared with serum from controls (Figure 1a). This finding suggests that BMP receptor signaling is activated in ECs by CKD. To clarify the roles of BMP receptor signaling in vivo, we created 5/6 nephrectomy (Nx) mice as a very short-term Kidney International (2015) 87, 128–136
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
Healthy Pt 1
2
2
CKD 3
4
5
p-Smad1/5/8 / Smad5
a
6 p-Smad1/5/8
49
Smad5
49 38
GAPDH
1.5 1 0.5 0
b
Healthy
CKD
WT+CKD
WT+Sham Merge
Smad5
*
Smad5
L
Merge
L Merge
p-Smad1/5/8
p-Smad1/5/8
Merge
L
L 20 µm
c
Mean intensity of p-Smad1/5/8 in nuclei (AU) 60
**
**
50 40 30 20 10 0
WT+Sham
WT+CKD
EC
WT+Sham
WT+CKD
VSMC
Figure 1 | Chronic kidney disease (CKD) activates bone morphogenetic protein receptor signaling in endothelial cells (ECs) and vascular smooth muscle cells (VSMCs). (a) Representative immunoblots and the pooled data demonstrating the effects of serum obtained from healthy subjects or CKD patients on the expression levels of total Smad5 and phosphorylated Smad1/5/8 (p-Smad1/5/8) in human umbilical vein endothelial cells. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. *Po0.05 versus healthy. (b, c) Immunofluorescence (b) and the pooled data (c) demonstrating the effects of Nx on the expression levels of Smad5 and p-Smad1/5/8 in the aortas of wild-type (WT) þ Sham and WT þ CKD mice. Green, Smad5 or p-Smad1/5/8; red, nuclei; and yellow, merged. Arrows indicate translocation of p-Smad1/5/8 into the nuclei. **Po0.01 versus WT þ Sham. AU, arbitrary unit.
CKD model. In wild-type C57BL/6J (WT) mice, Nx increased blood urea nitrogen and creatinine by 2.5- and 2-fold, respectively, but did not affect systolic blood pressure (SBP; Supplementary Figure S1 online). WT mice that received Nx (WT þ CKD) did not show vascular remodeling, such as intimal thickening, medial thickening, and adventitial fibrosis, and vascular inflammation (infiltration of inflammatory cells) in the aorta, nor did WT mice that had undergone sham operation (WT þ Sham) during the observation period (Supplementary Figure S2a and b online). In addition, CKD did not induce significant changes in mRNA Kidney International (2015) 87, 128–136
levels of cell-cell adhesion molecules and cell-matrix adhesion molecules, such as vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 in the aorta (Supplementary Table S1 online). In WT þ CKD mice, increased phosphorylation and nuclear translocation of Smad1/5/8, the indicators of BMP receptor activation, were demonstrated both in ECs and in the medial VSMCs without affecting the total amount of Smad5 (Figure 1b). Nuclear phosphorylated Smad1/5/8 levels were increased by fourfold in ECs and by twofold in VSMCs in WT þ CKD mice, compared with those in WT þ Sham mice (Figure 1c). Accordingly, it is suggested 129
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
that very short-term CKD created by Nx activates BMP signaling in ECs and VSMCs, preceding the development of vascular remodeling.
(Figure 3c). Such endothelium-dependent relaxation was impaired in WT þ CKD mice. In EC-Pten þ / mice with sham operation (EC-Pten þ / þ Sham), the endotheliumdependent relaxation was similar to that in WT þ Sham mice. The Nx-induced impairment of endothelium-dependent relaxation was abolished in EC-Pten þ / þ CKD mice. Endothelium-independent relaxation by nitroglycerin was similar among WT þ Sham, WT þ CKD, EC-Pten þ / þ þ/ Sham, and EC-Pten þ CKD (Supplementary Figure S3a online). In addition, the Nx-induced reduction in eNOS phosphorylation was prevented in EC-Pten þ / þ CKD mice (Figure 4). These findings suggest that very short-term CKD induces PTEN upregulation in ECs, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction.
Effects of BMP receptor activation on eNOS signaling in HUVECs
Next, we explored the role and consequences of BMP receptor activation in ECs. We treated HUVECs for 6 h with 30 or 300 ng/ml of BMP2, a known ligand of the BMP receptor. BMP2 treatment increased Smad1/5/8 phosphorylation and decreased eNOS phosphorylation at Ser1177, a key phosphorylation site that positively regulates eNOS enzyme activity, in HUVECs (Figure 2a). Interestingly, BMP2 treatment reduced phosphorylation levels of Ser473 in Akt, one of the major putative kinases for eNOS phosphorylation, in association with increases in protein levels of PTEN, an upstream protein phosphatase that reduces Akt phosphorylation and its activity (Figure 2a). As shown in Figure 2b, the number of PTEN-overexpressing ECs was increased by BMP2. Noteworthy, immunofluorescence staining revealed that BMP2 upregulated PTEN expression in the HUVECs with increased Smad1/5/8 phosphorylation levels (Figure 2c). To clarify the causal relationship between BMP receptor activation and PTEN-mediated Akt inactivation, we depleted PTEN using small interfering RNA. In HUVECs treated with nontargeting small interfering RNA (siControl), BMP2 reduced vascular endothelial growth factor-induced Akt phosphorylations (Figure 2d). PTEN depletion (siPten) abolished BMP2-induced inhibition of vascular endothelial growth factor-induced Akt phosphorylations. These results suggest that BMP receptor activation inhibits the Akt-eNOS pathway through PTEN induction in ECs. In vivo role of PTEN induction in eNOS signaling and endothelial function in CKD mice
For the next step, we investigated the in vivo roles of PTEN in impairment of the eNOS pathway and endothelial dysfunction in the aorta of CKD mice. WT þ Sham mice showed constitutive phosphorylations of eNOS and Akt (Figure 3a). On the other hand, eNOS and Akt phosphorylations were markedly reduced in WT þ CKD mice. Immunofluorescence staining revealed increased PTEN expression in the ECs of WT þ CKD mice, as compared with the ECs of WT þ Sham mice (Figure 3b). To determine the causal relationship between PTEN induction and endothelial dysfunction in CKD mice, ECdependent relaxation was investigated in WT þ Sham, WT þ CKD, EC-specific heterozygous null mice for Pten (ECPten þ / ) þ Sham, and EC-Pten þ / mice that received Nx (EC-Pten þ / þ CKD). EC-Pten þ / mice showed basal and Nx-elevated blood urea nitrogen and creatinine levels similar to those in WT mice (Supplementary Figure S1a and b online). SBP did not differ among the four groups (Supplementary Figure S1c online). In WT þ Sham mice, acetylcholine induced dose-dependent relaxation of aortic rings that were pre-contracted with phenylephrine 130
In vivo role of BMP receptor activation in endothelial function in CKD mice
To further define the causal relation between BMP receptor activation and endothelial dysfunction, the CKD model was created in heterozygous null mice for Alk3 (Alk3 þ / ). ALK3 is a BMP type I receptor9 that binds to a broad range of ligands in the BMP family, including BMP2, 4, 6, and 7.13–15 Alk3 þ / mice showed basal and Nx-elevated blood urea nitrogen and creatinine levels similar to those in WT mice (Supplementary Figure S1d and e online). There were no differences in SBP among WT þ Sham, WT þ CKD, Alk3 þ / mice with sham operation (Alk3 þ / þ Sham), and Alk3 þ / mice that received Nx (Alk3 þ / þ CKD; Supplementary Figure S1f online). The endothelium-dependent relaxation of the aorta did not significantly differ between WT þ Sham and Alk3 þ / þ Sham mice (Figure 5a). On the other hand, Alk3 þ / mice did not show Nx-induced endothelial dysfunction. Endothelium-independent relaxation by nitroglycerin did not differ among WT þ Sham, WT þ CKD, Alk3 þ / þ Sham, and Alk3 þ / þ CKD (Supplementary Figure S3 online). Aortic eNOS phosphorylation levels were similar between WT þ Sham and Alk3 þ / þ Sham mice. The Nx-induced reduction in eNOS phosphorylation was prevented in Alk3 þ / mice (Figure 4). Because the serum phosphate levels might affect the CKD-induced endothelial dysfunction, we investigated the serum phosphate levels. There were no differences among WT þ Sham, WT þ CKD, Alk3 þ / þ Sham, and Alk3 þ / þ CKD (Supplementary Figure S3b online). Roles of BMP receptor activation in VSMC osteogenic differentiation in CKD mice
We investigated the role of the BMP receptor/Smad pathway in VSMCs of CKD mice because the pathway is known to mediate VSMC osteogenic differentiation in the atheromatous plaques.11 As shown in Figure 1b and c, CKD increased phosphorylation and nuclear translocation of Smad1/5/8 in the medial VSMCs without affecting the total amount of Smad5, indicating the activation of the BMP receptor/Smad pathway. Moreover, expression levels of Msx2, an osteogenic Kidney International (2015) 87, 128–136
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
BMP2 30
300
60
12
p-Smad1/5/8 160 60
PTEN mean intensity
p-eNOS (Ser1177) p-Akt (Ser473) PTEN
49
Vehicle
0 0.00 %
12
93.75 %
0
10
0
17
PTEN
Sec ab only
0 42.11 %
0.00 %
4
0.00 %
57.89 %
0
10
17
50
0.00 %
0.00 %
8
4
4
0
GAPDH
BMP2
0
8
8
0.00 %
38
12
6.25 %
% Over threshold
0
0.00 %
100.00 %
0 0
10
17
**
40 30 20 10 0 BMP2
Vehicle
DAPI mean intensity
p-Smad1/5/8
PTEN
DAPI
p-Smad1/5/8+PTEN
Vehicle
BMP2
100 µm siControl –
BMP2 VEGF
–
siPten +
+
–
– +
–
+ +
–
+ PTEN
50 60 60
p-Akt (Ser473)
short exposure
p-Akt (Ser473)
long exposure
GAPDH 38
Figure 2 | Bone morphogenetic protein (BMP) receptor activation inhibits the Akt-endothelial nitric oxide synthase (eNOS) pathway by upregulating phosphatase-and-tensin homolog (PTEN) expression in endothelial cells (ECs). (a) Representative immunoblots showing the effects of BMP2 on the basal phosphorylation levels of eNOS, Smad1/5/8, Akt, and PTEN. (b) Scatter analysis of PTEN-labeled ECs. The pooled data showing the percentages of the mean fluorescence intensity of PTEN over the threshold. **Po0.01 versus vehicle. (c) Immunofluorescence staining demonstrating the effects of BMP2 on the expression levels of p-Smad1/5/8 (green) and PTEN (red). Blue, nuclei. The merged image shows that BMP2 increased p-Smad1/5/8 and PTEN in the same cells with nucleus stained as yellow. (d) Representative immunoblots showing basal and vascular endothelial growth factor (VEGF)-stimulated Akt phosphorylation with or without BMP2 stimulation after transfection with Pten small interfering RNA (siRNA; siPten) or non-targeting siRNA (siControl). Photographs of p-Akt immunoblots were obtained from the same blot with short (upper panel) and long (lower panel) exposures. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; DAPI, 40 -6-diamidino-2-phenylindole. Kidney International (2015) 87, 128–136
131
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
WT+CKD
WT+Sham 160
p-eNOS (Ser1177)
160
p-eNOS/eNOS
1.5
1
eNOS
62
0
p-Akt/Akt
Akt
38
GAPDH
PTEN
WT+CKD
1
** 0.5
0
WT+Sham
WT+Sham
1.5
p-Akt (Ser473)
62
*
0.5
WT+Sham
WT+CKD
WT+CKD PTEN
Merge 100 µm
L
Merge
L
20
Mean intensity of PTEN in EC (AU)
**
15 20 µm
L
10
L
5 0
WT+CKD
§§
0
*
§§
* 20
% Relaxation
WT+Sham
§
* # ¶
40 WT+Sham WT+CKD EC-Pten +/–+Sham EC-Pten +/–+CKD
60
80
100 –8
–7
–6
–5
–4
Ach log (mol/l)
Figure 3 | Nx increases phosphatase-and-tensin homolog (PTEN) expression and inactivates endothelial nitric oxide synthase (eNOS) in mice. (a) Representative immunoblots and the pooled data showing the effects of chronic kidney disease (CKD) on eNOS signaling in the mouse aorta. *Po0.05 versus wild-type (WT) þ Sham. **Po0.01 versus WT þ Sham. (b) Immunofluorescence and the pooled data demonstrating the effects of CKD on the expression levels of PTEN (green) in the aorta. Blue, nuclei. Arrows indicate endothelial cells (ECs) overexpressing PTEN. **Po0.01 versus WT þ Sham. (c) Pooled data of the acetylcholine (Ach)-induced relaxation in phenylephrine-pretreated rings of the descending aorta obtained from WT þ Sham (n ¼ 9), WT þ CKD (n ¼ 4), EC-Pten þ / þ Sham (n ¼ 12), and EC-Pten þ / þ CKD (n ¼ 11) mice. *Po0.05 WT þ Sham versus WT þ CKD, yPo0.05, and yyPo0.01 WT þ CKD versus EC-Pten þ / þ CKD; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham versus WT þ CKD, and zPo0.01 WT þ CKD versus EC-Pten þ / þ CKD; two-way analysis of variance. AU, arbitrary unit.
transcription factor and an established downstream gene of BMP receptor signaling, were upregulated in the aorta of WT þ CKD mice (Supplementary Figure S4a online). In 132
addition, alkaline phosphatase activity was increased in medial VSMCs, particularly in the inner layers of the media, of WT þ CKD mice, but not in those of WT þ Sham mice Kidney International (2015) 87, 128–136
basic research
p-eNOS (Ser1177)
160
eNOS
38
GAPDH
§
*
20
0
§
*
§
§ #¶
40 60 80
** p-eNOS/eNOS
1.5
1
0.5
0 Sham WT
CKD
Sham
CKD
Alk3 +/–
Sham
CKD
EC-Pten +/–
Figure 4 | Representative immunoblots and the pooled data demonstrating the effects of Nx on the expression levels of endothelial nitric oxide synthase (eNOS) and phosphorylated eNOS (p-eNOS) in the aortas of wild-type (WT) þ Sham, WT þ chronic kidney disease (CKD), Alk3 þ / þ Sham, Alk3 þ / þ CKD, EC-Pten þ / þ Sham, and EC-Pten þ / þ CKD mice. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. *Po0.05 WT þ Sham versus WT þ CKD, **Po0.01 WT þ CKD versus Alk3 þ / þ CKD, **Po0.01 WT þ CKD versus EC-Pten þ / þ CKD.
(Supplementary Figure S4b online). On the other hand, heterozygous deletion of Alk3 abolished Nx-induced alkaline phosphatase activation in the aorta. These results suggest that the BMP receptor also has a role in VSMC osteogenic differentiation in very short-term CKD mice without atherosclerotic changes. Effects of PTEN induction on osteogenic differentiation in VSMCs
Interestingly, immunofluorescence staining revealed increased PTEN expression in the VSMCs of WT þ CKD mice, as compared with the VSMCs of WT þ Sham mice (Figure 3b). BMP2 treatment increased mRNA levels of osteogenic transcriptional factor Msx2 and Pten in cultured VSMCs (Supplementary Figure S5a and b online). We investigated the relationship between PTEN induction and osteogenic differentiation in VSMCs by using small interfering RNA targeting Pten (siPten; Supplementary Figure S5c and d online). In VSMCs treated with nontargeting small interfering RNA (siControl), BMP2 increased mRNA levels of Msx2 and Pten. Although the Pten reduction by siPten was partial, siPten significantly reduced the BMP2-induced induction of Msx2 in VSMCs. These results suggest that Kidney International (2015) 87, 128–136
–7 –6 –5 –4 Ach log (mol/l) WT+Sham WT+CKD Alk3 +/– +Sham Alk3 +/– +CKD
*
*
§ 40
#¶
§ §§
60
100 –8
**
*
80
100 2
*
20
*
% Relaxation
160
§
0
% Relaxation
W T+ C KD Al k3 +/ – +S ha Al m k3 +/ – +C KD EC -P te n+ /– +S EC ha -P m te n+ /– +C KD
W
T+ Sh a
m
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
–8
–7 –6 –5 –4 Ach log (mol/l) WT+Sham+Vehicle WT+CKD+Vehicle WT+Sham+LDN WT+CKD+LDN
Figure 5 | Effects of bone morphogenetic protein (BMP) receptor signaling inhibition on the chronic kidney disease (CKD)-induced endothelial dysfunction. (a) Pooled data of acetylcholine (Ach)induced relaxation in phenylephrine (PE)-pretreated rings of the descending aorta obtained from wild-type (WT) þ Sham (n ¼ 9), WT þ CKD (n ¼ 4), Alk3 þ / þ Sham (n ¼ 4), and Alk3 þ / þ CKD (n ¼ 7) mice. *Po0.05 WT þ Sham versus WT þ CKD, yPo0.05 WT þ CKD versus Alk3 þ / þ CKD; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham versus WT þ CKD, zPo0.01 WT þ CKD versus Alk3 þ / þ CKD; two-way analysis of variance (ANOVA). (b) Pooled data of Ach-induced relaxation in PE-pretreated rings of the descending aorta obtained from Sham þ vehicle (n ¼ 8), Sham þ LDN193189 (n ¼ 6), CKD þ vehicle (n ¼ 5), and CKD þ LDN-193189 (n ¼ 7) mice. *Po0.05 WT þ Sham þ vehicle versus WT þ CKD þ vehicle, yPo0.05, and yyPo0.01 WT þ CKD þ vehicle versus WT þ CKD þ LDN-193189; Tukey–Kramer’s post-hoc analysis. # Po0.01 WT þ Sham þ vehicle versus WT þ CKD þ vehicle, and z Po0.01 WT þ CKD þ vehicle versus WT þ CKD þ LDN-193189; two-way ANOVA.
PTEN activation would have a role in osteogenic differentiation in VSMC. Effects of selective BMP type I receptor inhibitor
To explore potential clinical interventions, we investigated the effects of LDN-193189 (4-[6-(4-piperazin-1-ylphenyl) pyrazolo [1,5-a]pyrimidin-3-yl] quinolone, 3 mg/kg), a small molecule inhibitor of BMP type I receptor kinases,16,17 in Nxinduced endothelial dysfunction and VSMC osteogenic differentiation in WT mice. LDN-193189 administration did not affect renal function and SBP (Supplementary Figure S6a–c online). This LDN-193189 treatment regimen prevented endothelial dysfunction in WT þ CKD mice without affecting endothelial function in WT þ Sham mice (Figure 5b). Also, LDN-193189 inhibited osteogenic differentiation of medial VSMCs in the CKD mice (Supplementary Figure S4b online). DISCUSSION
The novel findings of the present study are as follows: (i) BMP receptor signaling is involved in endothelial dysfunction, in addition to osteogenic differentiation of medial VSMCs, in a mouse model of very short-term CKD without 133
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
apparent vascular inflammation and remodeling. (ii) CKD induces PTEN upregulation through BMP receptor activation in ECs, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction. (iii) A selective inhibitor of BMP type I receptor kinases, LDN193189, prevents endothelial dysfunction and abolishes osteogenic differentiation of VSMCs in CKD mice. These findings suggest that BMP receptor activation may be a common initial mechanism for endothelial dysfunction and vascular calcification, which is activated even in the early stage of CKD. In the present study, we used Nx mice with 2.5- and 2-fold increases in blood urea nitrogen and creatinine, respectively, without SBP elevation and serum phosphate change, which suggests that it is a very short-term CKD model representing moderate renal failure without hypertension. This study was performed in young mice, and the observation period was rather short, because we focused on investigating the relationship between BMP receptor signaling and endothelial dysfunction in the CKD model before cardiovascular damages and remodeling had developed. The most important novel finding of this study is that BMP receptor activation is crucial for the mechanism of endothelial dysfunction in the early stage of CKD. The causal relation between BMP receptor activation and endothelial dysfunction is clearly presented by the facts that the genetic Alk3 deletion and LDN-193189 treatment abolished the CKD-induced endothelial dysfunction (Figure 5). In this study, we focused on the eNOS pathway as the mechanism of endothelial dysfunction, because we have previously demonstrated that impairment of eNOS activation is important for endothelial dysfunction in this CKD model.18 It is noteworthy that PTEN is the key molecule linking BMP receptor activation and endothelial dysfunction, because PTEN was upregulated in the ECs of CKD mice and the Pten deletion prevented the CKD-induced eNOS dephosphorylation and endothelial dysfunction in mice (Figures 3b, c, and 4). As shown in Figure 3a and b, the CKD-induced PTEN upregulation might decrease eNOS phosphorylation levels via the reduction of Akt activity. This hypothesis is supported by the fact that PTEN depletion using siPten abolished the inhibition of the agonist-induced Akt activation by BMP receptor activation in HUVECs (Figure 2d). Accordingly, it is suggested that CKD induces upregulation of PTEN in ECs as a consequence of activation of BMP receptor signaling, which leads to inhibition of the Akt-eNOS pathway and subsequent endothelial dysfunction. There is increasing evidence that the BMP family participates in various types of ectopic ossification.19 With regard to atherosclerosis, upregulated BMP2 expression was found in calcified human atherosclerotic plaques.11 In addition, the role of the BMP receptor signal was suggested in the plaque calcification in apolipoprotein E-deficient mice receiving high-fat diet and bilateral ovariectomy and in highfat diet-fed LDL receptor-deficient mice.17,20 In this study, not only the activation of the downstream of the BMP
receptor (Smad1/5/8 phosphorylation and Msx2 upregulaiton) but also the osteogenic differentiation (alkaline phosphatase activation) was induced in the medial VSMCs in the early stage of CKD mice (Figure 1b and c and Supplementary Figure S4 online). Moreover, the genetic deletion of Alk3 and pharmacological inhibition of the BMP receptor with LDN-193189 abolished the VSMC osteogenic differentiation (Supplementary Figure S4b online). Taken together, the present study has presented for the first time that the BMP receptor signal has a role in the VSMC osteogenic differentiation in the early stage of CKD. It has been established that, upon binding to BMPs, BMP type I receptors are activated by BMP type II receptors and subsequently activate the Smad1/5/8-dependent and -independent signaling pathways.21–23 Derwall et al. have demonstrated that BMP type I receptor inhibition by LDN-193189 prevents vascular inflammation, atherosclerotic plaque formation, and plaque calcification in high-fat diet-fed LDL receptor–deficient mice.16,17 The mechanism for these effects of LDN-193189 was explained by the reductions in vascular oxidative stress, plaque inflammation, and hepatic LDL biosynthesis. Recently, Kim et al. have shown that genetic ablation of BMP type II receptors exaggerated endothelial inflammatory and atherosclerotic plaque formation in highfat diet-fed ApoE-deficient mice.24 They suggested that the BMP type II depletion induced endothelial inflammation in a ligand-independent manner. Currently, there is no explanation for the discrepancy between the anti-inflammatory and antiatherogenic effects of the BMP type I receptor inhibition and the pro-inflammatory and pro-atherogenic effects of the BMP type II receptor reduction in high-fat dietfed mice prone to atherosclerosis. In the current study, CKD mice had no evidence of vascular inflammation and remodeling, probably because this is a very short-term CKD model. Therefore, it is impossible to determine the roles of BMP receptors in vascular inflammation and remodeling in the later stage of CKD from this study. Because endothelial dysfunction is the initial step of atherosclerotic lesion formation, it is interesting to determine whether the BMP receptor–mediated endothelial dysfunction would be also involved in the earlier stage of atherogenesis from analogy to CKD, as shown in the present study. This issue should be addressed in future studies. There are several limitations in the present study. As shown by this study and others (Figure 1a and Chen et al.12), serum obtained from CKD patients highly stimulates BMP receptor signaling activity in ECs and VSMCs. It awaits future endeavor to identify ligands in uremic serum to activate this cascade. Second, in CKD animal studies, the kidney disease is usually allowed to evolve over at least 8–12 weeks. Thus, this model is regarded as a model of very short-term CKD. This study has clearly shown that the BMP receptors have a role in the prevention of endothelial dysfunction and vascular osteogenic differentiation before the development of histological lesions in the early stage of CKD. However, it remains unclear whether blocking the BMP receptor could also have a
134
Kidney International (2015) 87, 128–136
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
basic research
significant effect once vascular lesions already exist and whether the long-term blocking of the BMP receptor would be useful as well. Third, we clearly demonstrated that PTEN is a downstream effector molecule of activated BMP receptor signaling because endothelial-specific removal of Pten prevents CKD-induced endothelial dysfunction. The mechanism whereby BMP receptor activation results in increases in PTEN remains to be elucidated. It is noteworthy that BMP2 treatment increases PTEN levels by decreasing protein degradation in MCF7 human breast cancer cells.25 Thus, it is possible that the post-translational mechanism would be involved in the CKD-induced PTEN protein upregulation. Fourth, it was suggested that PTEN induction would participate in the BMP2-induced VSMC osteogenic differentiation (Supplementary Figure S5 online). Future experiments using VSMC-specific null mice for PTEN are necessary to determine the in vivo role of PTEN in VSMC osteogenic differentiation and calcification in CKD. Finally, we used mice heterozygous null for Pten or Alk3, because both mice homozygous null allele are lethal.26,27 Having phenotypic rescues by heterozygous deletion indicates that small decreases in PTEN expression or BMP receptor signaling may preserve endothelial function in the early stage of CKD. The fact that a suboptimal dose of LDN-193189 can suppress endothelial dysfunction also strongly supports this notion. The present study suggests that BMP receptor signaling has a pivotal role in endothelial dysfunction, in addition to osteogenic differentiation of VSMCs, in the early stage of CKD without vascular remodeling. An increase in PTEN protein is involved in the mechanism by which BMP receptor activation inhibits Akt-eNOS signaling in CKD. Our observations suggest that BMP receptor signaling would be a promising target for the prevention and treatment of both endothelial dysfunction and vascular calcification in the early stage of CKD patients.
week later. Animal experiments included the following groups: WT þ Sham, WT þ CKD, EC-Pten þ / þ Sham, EC-Pten þ / þ þ/ CKD, Alk3 þ Sham, Alk3 þ / þ CKD, WT mice that received Nx and LDN-193189 (WT þ CKD þ LDN), and WT mice that received Nx and the vehicle (WT þ CKD þ vehicle). To abrogate PTEN expression in ECs, EC-Pten þ / mice, namely Tie2CrePtenflox/ þ mice, were used because Tie2CrePtenflox/flox mice die before embryonic day 11.5 because of bleeding and cardiac failure.27 To negate BMP receptor signaling, we used Alk3 þ / mice because the homozygous mutation is embryonic lethal because of failure to produce mesoderm.26
MATERIALS AND METHODS A detailed method section is available in the Supplementary Materials online. Mice and human samples WT mice were purchased from Charles River Laboratories (Yokohama, Japan). EC-Pten þ / mice and Alk3 þ / mice were constructed as described previously.26,27 Mice were fed with normal chow (normal phosphate diet). LDN-193189 was obtained from Axon Medchem BV (Groningen, The Netherlands). HUVECs and VSMCs were purchased from Lonza (Basel, Switzerland). The study protocol of animal studies was approved by the Animal Care and Treatment Committee of Kurume University. The patient study was conducted in accordance with the Declaration of Helsinki and approved by the ethics committee of Kurume University. CKD models CKD was established by performing 5/6 Nx in mice as described previously.18 Briefly, mice were anesthetized with 1.5% isoflurane by inhalation. Nx was established by surgical resection of the upper and lower thirds of the left kidney at 10–12 weeks, followed by right Nx 1 Kidney International (2015) 87, 128–136
Evaluation of animal model Four weeks after the establishment of Nx, SBP and heart rate were measured using a tail-cuff sphygmomanometer (MK-2000ST; Muromachi, Tokyo, Japan), as described previously.18 One day after measurements of SBP and heart rate, mice were killed by an overdose of pentobarbital (100 mg/kg, intraperitoneal). Blood was collected from the right atrium, and then mice were perfused with ice-cold saline for 5 min. The aorta was immediately removed and subjected to isometric tension study or immunoblot analysis. Expression analysis Expression levels of mRNA and protein were evaluated as described in the Supplementary Materials online. Statistical analysis Results are shown as mean±s.e.m, unless indicated otherwise. Intergroup differences were assessed by the Mann–Whitney U-test or two-way ANOVA followed by Tukey–Kramer’s post-hoc analysis. A value of Po0.05 was considered statistically significant. DISCLOSURE
All the authors declared no competing interest. ACKNOWLEDGMENTS
This study was supported, in part, by a grant for the Science Frontier Research Promotion Centers (Cardiovascular Research Institute), Grants-in-Aid for Scientific Research (HK and TI) from the Ministry of Education, Science, Sports, and Culture, Japan, and research grants from the Kidney Foundation (HK). We thank Katsue Shiramizu, Miyuki Nishigata, Kimiko Kimura, Miho Nakao, and Makiko Kiyohiro for technical assistance. SUPPLEMENTARY MATERIAL Figure S1. Effects of Nx on BUN, serum creatinine, and SBP in WT, EC-Pten+/ , and Alk3+/ mice at 4 weeks after Nx. Figure S2. Effects of Nx on BUN, serum creatinine, and SBP in WT, EC-Pten+/ , and Alk3+/ mice at 4 weeks after Nx. Figure S3. (a) Endothelium independent relaxation of aortic rings obtained from WT, EC-Pten+/ , and Alk3+/ mice. (b) Effects of Nx on serum phosphate in WT and Alk3+/ mice at 4 weeks after Nx. Figure S4. BMP receptor inhibition prevents Nx-induced osteogenic differentiation of medial VSMCs. Figure S5. Effects of PTEN induction on BMP receptor activationinduced osteogenic differentiation in VSMCs. Figure S6. Effects of LDN-193189 on BUN, serum creatinine, and SBP in WT mice at 2 weeks of treatment after Nx. Table S1. Effects of Nx on cell adhesion molecules in WT mice. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki 135
basic research
H Kajimoto et al.: BMPR activation and endothelial dysfunction in CKD
REFERENCES
14.
1.
2. 3.
4.
5.
6.
7.
8. 9. 10.
11.
12.
13.
136
Keith DS, Nichols GA, Gullion CM et al. Longitudinal follow-up and outcomes among a population with chronic kidney disease in a large managed care organization. Arch Intern Med 2004; 164: 659–663. Morris ST, McMurray JJ, Spiers A et al. Impaired endothelial function in isolated human uremic resistance arteries. Kidney Int 2001; 60: 1077–1082. Stam F, van Guldener C, Becker A et al. Endothelial dysfunction contributes to renal function-associated cardiovascular mortality in a population with mild renal insufficiency: the Hoorn study. J Am Soc Nephrol 2006; 17: 537–545. Blacher J, Guerin AP, Pannier B et al. Arterial calcifications, arterial stiffness, and cardiovascular risk in end-stage renal disease. Hypertension 2001; 38: 938–942. London GM, Guerin AP, Marchais SJ et al. Arterial media calcification in end-stage renal disease: impact on all-cause and cardiovascular mortality. Nephrol Dial Transplant 2003; 18: 1731–1740. Okuno S, Ishimura E, Kitatani K et al. Presence of abdominal aortic calcification is significantly associated with all-cause and cardiovascular mortality in maintenance hemodialysis patients. Am J Kidney Dis 2007; 49: 417–425. Huang PH, Chen LC, Leu HB et al. Enhanced coronary calcification determined by electron beam CT is strongly related to endothelial dysfunction in patients with suspected coronary artery disease. Chest 2005; 128: 810–815. Wozney JM, Rosen V, Celeste AJ et al. Novel regulators of bone formation: molecular clones and activities. Science 1988; 242: 1528–1534. Shi Y, Massague J. Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 2003; 113: 685–700. Speer MY, Yang HY, Brabb T et al. Smooth muscle cells give rise to osteochondrogenic precursors and chondrocytes in calcifying arteries. Circ Res 2009; 104: 733–741. Bostrom K, Watson KE, Horn S et al. Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest 1993; 91: 1800–1809. Chen NX, Duan D, O’Neill KD et al. The mechanisms of uremic seruminduced expression of bone matrix proteins in bovine vascular smooth muscle cells. Kidney Int 2006; 70: 1046–1053. Yamashita H, ten Dijke P, Huylebroeck D et al. Osteogenic protein-1 binds to activin type II receptors and induces certain activin-like effects. J Cell Biol 1995; 130: 217–226.
15.
16.
17.
18.
19. 20.
21. 22. 23.
24.
25.
26.
27.
Ebisawa T, Tada K, Kitajima I et al. Characterization of bone morphogenetic protein-6 signaling pathways in osteoblast differentiation. J Cell Sci 1999; 112(Pt 20): 3519–3527. ten Dijke P, Yamashita H, Sampath TK et al. Identification of type I receptors for osteogenic protein-1 and bone morphogenetic protein-4. J Biol Chem 1994; 269: 16985–16988. Yu PB, Deng DY, Lai CS et al. BMP type I receptor inhibition reduces heterotopic [corrected] ossification. Nat Med 2008; 14: 1363–1369. Derwall M, Malhotra R, Lai CS et al. Inhibition of bone morphogenetic protein signaling reduces vascular calcification and atherosclerosis. Arterioscler Thromb Vasc Biol 2012; 32: 613–622. Kajimoto H, Kai H, Aoki H et al. Inhibition of eNOS phosphorylation mediates endothelial dysfunction in renal failure: new effect of asymmetric dimethylarginine. Kidney Int 2012; 81: 762–768. Reddi AH. Bone morphogenetic proteins: from basic science to clinical applications. J Bone Joint Surg Am 2001; 83-A (Suppl 1): S1–S6. Osako MK, Nakagami H, Koibuchi N et al. Estrogen inhibits vascular calcification via vascular RANKL system: common mechanism of osteoporosis and vascular calcification. Circ Res 2010; 107: 466–475. Hogan BL. Bone morphogenetic proteins in development. Curr Opin Genet Dev 1996; 6: 432–438. Massague J. How cells read TGF-beta signals. Nat Rev Mol Cell Biol 2000; 1: 169–178. Fujii M, Takeda K, Imamura T et al. Roles of bone morphogenetic protein type I receptors and Smad proteins in osteoblast and chondroblast differentiation. Mol Biol Cell 1999; 10: 3801–3813. Kim CW, Song H, Kumar S et al. Anti-inflammatory and antiatherogenic role of BMP receptor II in endothelial cells. Arterioscler Thromb Vasc Biol 2013; 33: 1350–1359. Waite KA, Eng C. BMP2 exposure results in decreased PTEN protein degradation and increased PTEN levels. Hum Mol Genet 2003; 12: 679–684. Mishina Y, Suzuki A, Ueno N et al. Bmpr encodes a type I bone morphogenetic protein receptor that is essential for gastrulation during mouse embryogenesis. Genes Dev 1995; 9: 3027–3037. Hamada K, Sasaki T, Koni PA et al. The PTEN/PI3K pathway governs normal vascular development and tumor angiogenesis. Genes Dev 2005; 19: 2054–2065.
Kidney International (2015) 87, 128–136
