Angiogenic and Vasoprotective Effects of Adrenomedullin on Prevention of Cognitive Decline After Chronic Cerebral Hypoperfusion in Mice
Background and Purpose—Although subcortical vascular dementia, the major subtype of vascular dementia, is caused by a disruption in white matter integrity after cerebrovascular insufficiency, no therapy has been discovered that will restore cerebral perfusion or functional cerebral vessels. Because adrenomedullin (AM) has been shown to be angiogenic and vasoprotective, the purpose of the study was to investigate whether AM may be used as a putative treatment for subcortical vascular dementia.
Methods—A model of subcortical vascular dementia was reproduced in mice by placing microcoils bilaterally on the common carotid arteries. Using mice overexpressing circulating AM, we assessed the effect of AM on cerebral perfusion, cerebral angioarchitecture, oxidative stress, white matter change, cognitive function, and brain levels of cAMP, vascular endothelial growth factor, and basic fibroblast growth factor.
Results—After bilateral common carotid artery stenosis, mice overexpressing circulating AM showed significantly faster cerebral perfusion recovery due to substantial growth of the capillaries, the circle of Willis, and the leptomeningeal anastomoses and reduced oxidative damage in vascular endothelial cells compared with wild-type mice. Vascular changes were preceded by upregulation of cAMP, vascular endothelial growth factor, and basic fibroblast growth factor. White matter damage and working memory deficits induced by bilateral common carotid artery stenosis were subsequently restored in mice overexpressing circulating AM.
Conclusions—These data indicate that AM promotes arteriogenesis and angiogenesis, inhibits oxidative stress, preserves white matter integrity, and prevents cognitive decline after chronic cerebral hypoperfusion. Thus, AM may serve as a strategy to tackle subcortical vascular dementia.
Ischemic white matter (WM) lesions, which are most likely caused by cerebrovascular insufficiency after atherosclerosis and/or arteriosclerosis, are an established marker of risk for cognitive deterioration.1 Thus, therapeutic vascular growth and vasoprotection, resulting in the preservation of WM integrity, may serve to maintain cognitive function in subjects at risk of developing dementia.
Adrenomedullin (AM) has a variety of effects on the vasculature that include vasodilation, regulation of permeability, inhibition of endothelial cell apoptosis and oxidative stress, regulation of smooth muscle cell proliferation, and promotion of angiogenesis.2,3
Thus, the purpose of this study was to investigate the mechanisms and therapeutic potential of AM-induced neovascularization and/or vasoprotection after chronic cerebral hypoperfusion in a mouse model of subcortical vascular dementia.4,5
Materials and Methods
An expanded Methods section is available in the Online Data Supplement (http://stroke.ahajournals.org).
Adrenomedullin Facilitates Recovery of Cerebral Blood Flow After Placing Microcoils Bilaterally on the Common Carotid Arteries
Immediately after bilateral common carotid artery stenosis (BCAS), cerebral blood flow (CBF) decreased to the lowest values but thereafter began to recover in all groups. On Days 1 and 3 post-BCAS, there was a slight, but not significant, increase in CBF (average±SEM) in mice overexpressing circulating AM (AM-Tg) compared with wild-type (WT) mice. On Day 7 post-BCAS, AM-Tg mice showed significantly faster CBF recovery: CBF was significantly higher in AM-Tg mice (93%±2%) compared with WT mice (79%±2%) and hydralazine-treated WT mice (71%±2%; Figure 1A–C). This trend continued on Days 14 and 28 post-BCAS (Figure 1C; Supplemental Figure IA). Slower CBF recovery in hydralazine-treated WT mice suggests that AM-induced CBF recovery may not be associated with the hypotensive effect of AM.
We further examined the effects of postoperative exogenous infusion of AM.6,7 Continuous intraperitoneal injection of recombinant human AM at a rate of 50 ng/hr for 2 weeks, beginning on Day 1 post-BCAS, resulted in a significantly faster CBF recovery compared with the vehicle-treated mice (Figure 1D). These effects were comparable to those seen in BCAS-operated AM-Tg mice.
Thus, both genetically overproduced AM and postoperative exogenous administration of AM facilitated recovery of CBF after BCAS.
Adrenomedullin Enhances Arteriogenesis After BCAS
At the dorsal surface of the brain, a significant increase in diameter of the leptomeningeal anastomoses was found in AM-Tg mice (28.7±1.6 μm) compared with WT mice (22.4±1.3 μm) on Day 7 post-BCAS (Figure 2A–C; Supplemental Figure IB, a–d). The number of leptomeningeal anastomoses was not different among the 4 groups. The diameter of the internal carotid artery, anterior cerebral artery, middle cerebral artery, and posterior communicating artery was significantly enlarged at the level of the circle of Willis in AM-Tg mice compared with WT mice (AM-Tg versus WT; anterior cerebral artery, 193±26 versus 161±24 μm; middle cerebral artery, 184±24 versus 153±13 μm; internal carotid artery, 206±30 versus 175±25 μm; posterior communicating artery, 191±16 versus 150±15 μm) on Day 7 post-BCAS (Figure 2D; Supplemental Figure IB, e–h).
To evaluate monocyte recruitments and proliferation of smooth muscle cells, both of which are essential in arteriogenesis, the immunofluorescent analysis of Ki-67 and F4/80, together with α-smooth muscle actin, was performed. BCAS-operated AM-Tg mice showed a significant increase in Ki-67-positive vascular smooth muscle cells compared with BCAS-operated WT mice (Supplemental Figure IC). In addition, a significant increase in vascular smooth muscle cells surrounded by F4/80-positive monocyte/macrophages was found in BCAS-operated AM-Tg mice compared with sham-operated WT mice (Supplemental Figure ID).
Adrenomedullin Enhances Angiogenesis After BCAS
A significant increase in platelet-endothelial cell adhesion molecule-1-positive capillary density of the cortex, corpus callosum, and caudoputamen was found in AM-Tg mice compared with WT mice (AM-Tg versus WT; cortex, 540±55/mm2 versus 473±38/mm2; corpus callosum, 273±7/mm2 versus 213±18/mm2; caudoputamen, 499±36/mm2 versus 455±26/mm2) on Day 7 post-BCAS. There was no significant difference in capillary density between AM-Tg and WT mice after sham operation (Figure 3A–C; Supplemental Figure IE).
Taken together, these results suggest that both chronic ischemic stress and AM overexpression are required to induce arteriogenesis and angiogenesis in the brain.
Adrenomedullin Attenuates Oxidative Damage in Cerebral Microvessels After BCAS
To evaluate oxidative damage in cerebral microvessels, double immunofluorescence staining for platelet-endothelial cell adhesion molecule-1 and 8-hydroxy-deoxyguanosine was performed on Day 3 post-BCAS. A significant decrease in oxidative damage in the cerebral microvessels was found in BCAS-operated AM-Tg mice compared with BCAS-operated WT mice (Figure 3D; Supplemental Figure IF).
Adrenomedullin Preserves WM Integrity After BCAS
BCAS-operated WT mice showed an increased density of glial fibrillary acidic protein-positive astrocytes and ionized calcium binding adapter molecule 1-positive microglia and a decreased density of glutathione-S-transferase-pi-immunoreactive mature oligodendrocytes in the corpus callosum and the anterior commissure compared with sham-operated WT or AM-Tg mice on Day 28 (Figure 4A–C; Supplemental Figure IIA, a–c, e–g, and i–k). In BCAS-operated AM-Tg mice, by contrast, the density of astrocytes and microglia significantly decreased and that of mature oligodendrocytes significantly increased compared with BCAS-operated WT mice (Figure 4A–C; Supplement Figure IIA, d, h, and l). Similarly, BCAS-operated AM-Tg mice on Day 7 showed significantly decreased density of microglia and increased density of mature oligodendrocytes but no difference in the density of astrocytes compared with BCAS-operated WT mice (Supplemental Figure IIB–D).
Klüver–Barrera staining on Day 28 revealed that BCAS-induced WM lesions were predominant in the corpus callosum and the caudoputamen in WT mice (Supplemental Figure IIA, m–o). In AM-Tg mice, such WM lesions became far less severe (Supplemental Figure IIAp). In the hydralazine-treated WT mice, BCAS-induced WM lesions were as severe as those in WT mice, suggesting that such positive effects of AM may be independent of the hypotensive effect of AM (Figure 4D).
Thus, BCAS-induced WM lesions were restored in the AM-Tg mice in parallel with inhibition of glial activation and preservation of mature oligodendrocytes.
Adrenomedullin Prevents Working Memory Deficits After BCAS
To evaluate working memory, we examined a Y maze test and an 8-arm radial maze test. The Y maze test was performed 1 month after the surgery. Alternations of entries in the arms of the Y maze were significantly increased in BCAS-operated AM-Tg mice (64.6%±7.5%) compared with BCAS-operated WT mice (58.2%±8.0%), although alternations of entries were not significantly different between WT and AM-Tg mice after sham operation (Figure 5A). Spontaneous activity was not significantly different among the 4 groups of mice (Figure 5B). In another set of mice, the 8-arm radial maze test was started 1 month after BCAS. BCAS-operated AM-Tg mice showed a significant reduction in the number of revisiting errors compared with BCAS-operated WT mice (Figure 5C). We have found a significant correlation of the averaged number of revisiting errors 1 month after BCAS with CBF on Days 7, 14, and 28, but not with CBF immediately after BCAS, or on Days 1 and 3 (Supplemental Figure III).
Taken together, these results suggest that AM restores working memory deficits induced by BCAS.
Adrenomedullin Increases cAMP Level in the Forebrain After BCAS
The restorative effect of AM described led to the investigation of the underlying mechanisms behind angio-/arteriogenesis and antioxidative activity. The brain level of cAMP, a second messenger known to associate with AM, was therefore measured. A significant increase in the brain level of cAMP was found in AM-Tg mice after BCAS (AM BCAS, 1.6±0.2 pmol/mg wet tissue) compared with WT mice after BCAS (WT BCAS, 1.3±0.2 pmol/mg wet tissue) on Day 5, although the level of cAMP was not different between WT and AM-Tg mice after sham operation (Figure 6A).
These results suggest that chronic ischemic stress induced AM-mediated elevation of cAMP in the brain.
Adrenomedullin Increases mRNA and Protein Levels of Vascular Endothelial Growth Factor and Basic Fibroblast Growth Factor in the Forebrain After BCAS
The reasons behind the apparent AM-initiated signaling pathway-led arteriogenesis and angiogenesis were next examined. Brain levels of vascular growth factors, including vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF), were therefore measured. The mRNA and protein levels of brain VEGF and bFGF were significantly increased in AM-Tg mice on Days 1 and 5 post-BCAS compared with sham-operated WT mice (Figure 6B–C; Supplemental Figure IV).
Chronic Ischemic Insult Upregulates Brain mRNA Level of Adrenomedullin and Abolishes Receptor Activity-Modifying Protein-2 Suppression Induced by Adrenomedullin
The status of mouse AM, high-affinity AM receptors, calcitonin receptor-like receptors, and Subtypes 2 and 3 of a family of receptor activity-modifying proteins (RAMP2 or 3) were then measured on Day 5. The brain mRNA level of mouse AM was significantly increased (3.1-fold) in BCAS-operated WT mice compared with sham-operated WT mice. In addition, brain RAMP2 mRNA level was significantly lower (0.6-fold) in sham-operated AM-Tg mice compared with sham-operated WT mice. Such downregulation of RAMP2 mRNA was abolished after the AM mice were subjected to BCAS operation, suggesting that feedback inhibition is a plausible cause for the downregulation (Figure 6D).
Three major conclusions may be drawn from the present study. First, it was demonstrated that increased levels of circulating AM restored cerebral hemodynamics, promoted arteriogenesis as well as angiogenesis, alleviated oxidative damage in the cerebral microvessels, and preserved WM integrity; this subsequently attenuated working memory deficits in a mouse model of chronic cerebral hypoperfusion. Second, AM selectively upregulated brain levels of cAMP, VEGF, and bFGF in the hypoperfused brain but not in the normoperfused brain. Finally, it was found that such proangiogenic/arteriogenic changes did not occur in sham-operated AM-Tg mice in which the expression of AM receptor component RAMP2 was significantly suppressed, possibly through feedback inhibition.
We have found a significant correlation in the averaged number of revisiting errors at 1 month after BCAS with CBF on Days 7, 14, and 28, but not with CBF immediately after BCAS or on Days 1 and 3. Therefore, the recovery of CBF is one of the substrates for the functional improvements, whereas several phenomena other than CBF recovery may play roles in the pathophysiology of this BCAS model. In fact, we demonstrated that AM induced not only CBF recovery as a result of arteriogenesis, but also angiogenesis (not associated with CBF recovery), antioxidative activity in the microvessels, and attenuation of microglial inflammatory responses. The other effects of AM, including antiapoptotic effects and regulation of endothelial permeability or the blood–brain barrier,3 need further investigation. Positive effects of AM may be mediated by multiple pathways.
Previous reports showed that the AM/cAMP/PKA cascade blocks oxidative damage in ischemic injury8 and promotes angiogenic effects of the endothelial cells in vitro.9 We found that chronic ischemic insult and circulating AM are both required to raise cAMP levels in the brain; this may be associated with alleviating oxidative damage and promoting angiogenesis.
The elevation of VEGF is consistent with the previous report that AM administration upregulates the expression of VEGF in both in vitro and in vivo hindlimb ischemia models.6 Although no previous studies have reported that AM enhances the expression of bFGF after ischemia, we demonstrated the AM-induced upregulation of bFGF after BCAS in vivo. AM was also found to upregulate bFGF as well as VEGF in the cultured endothelial cells (unpublished data). Previous reports have demonstrated that combined gene delivery of VEGF and bFGF produces additive or synergistic effects on angiogenesis or collateral development, probably due to the protective effects of bFGF against VEGF-induced fluid leakage.10 Thus, AM-induced elevation of bFGF may be associated with the development of functional vessels.
AM acts through 2 subtypes of receptor (AM1 and AM2), which derive from the interaction of the calcitonin receptor-like receptors with RAMP2 or 3.11 Interestingly, RAMP2 mRNA level in sham-operated AM-Tg mice was significantly decreased compared with sham-operated WT mice but nearly reached normalization after BCAS. This may explain why arteriogenesis and angiogenesis were significantly promoted in AM-Tg mice only after ischemic insult. Shindo et al have reported that RAMP2 rather than RAMP3 is a key determinant of the effects of AM on the vasculature and is essential for angiogenesis and vascular integrity.11 These results suggest that the AM-initiated signaling pathway is suppressed by downregulation of RAMP2 in the normoperfused brain but that such suppression is abolished by chronic ischemic stress, leading to AM-induced arteriogenesis and angiogenesis. Such tissue selectivity could be an advantage for clinical application of AM in patients with subcortical vascular dementia.
Recently, the concept of an “oligovascular niche” has been proposed, in which crosstalk between endothelial cells and oligodendrocytes, mediated by an exchange of soluble signals such as fibroblast growth factor, are thought to play an important role in sustaining oligodendrocyte homeostasis and WM integrity.12 Because cerebral endothelial cells contribute to numerous signaling cascades that help regulate brain homeostasis and function,13 angio-/arteriogenesis and inhibition of oxidative damage in the cerebral endothelial cells induced by AM might lead to oligovascular protection—namely, successful vascular growth and vasoprotection and preservation of white matter/oligodendrocyte integrity—and prevention of cognitive decline after chronic cerebral hypoperfusion in mice.
In conclusion, this study demonstrates that circulating AM is a highly potent and effective modality for restoring perfusion, promoting arteriogenesis and angiogenesis in the chronically ischemic brain, inhibiting oxidative damage in the cerebral microvessels, preserving ischemic WM integrity, and attenuating working memory deficits in a mouse model of subcortical vascular dementia. Future clinical studies are required to evaluate and confirm the efficacy of AM in chronic cerebral vascular diseases, especially subcortical vascular dementia.
Sources of Funding
This work was supported in part by a grant from the Translational Research Center at Kyoto University Hospital (M.I.), grants from the Mitsubishi Pharma Research Foundation, the Uehara Memorial Foundation, and the Takeda Science Foundation (M.I.), Grants-in-Aid for Young Scientists (start-up; M.I.) and for scientific research (C; H. Ito and R.T.) from the Ministry of Education, Culture, Sports, Science and Technology in Japan, and a Health Labor Sciences Research Grant from the Ministry of Health, Labor and Welfare, Japan (H. Itoh and R.T.).
We are very grateful to Dr Ahmad Khundakar for his thoughtful comments and to Ms Hitomi Nakabayashi for her excellent technical assistance; Mr Kazuo Nakanishi for statistical analysis; Dr Jun Takahashi, Dr Tetsuhiro Kikuchi, and Dr Kenichi Todo for technical advice; and Dr Tsuyoshi Miyakawa and Dr Keizo Takao for their advice on behavioral experiments.
The online-only Data Supplement is available at http://stroke.ahajournals.org/cgi/content/full/STROKEAHA.110.603399/DC1.
- Received September 22, 2010.
- Revision received November 11, 2010.
- Accepted November 15, 2010.
- © 2011 American Heart Association, Inc.
- Pantoni L
- Kato J,
- Tsuruda T,
- Kita T,
- Kitamura K,
- Eto T
- Shibata M,
- Ohtani R,
- Ihara M,
- Tomimoto H
- Shibata M,
- Yamasaki N,
- Miyakawa T,
- Kalaria RN,
- Fujita Y,
- Ohtani R,
- Ihara M,
- Takahashi R,
- Tomimoto H
- Iimuro S,
- Shindo T,
- Moriyama N,
- Amaki T,
- Niu P,
- Takeda N,
- Iwata H,
- Zhang Y,
- Ebihara A,
- Nagai R
- Miyashita K,
- Itoh H,
- Arai H,
- Suganami T,
- Sawada N,
- Fukunaga Y,
- Sone M,
- Yamahara K,
- Yurugi-Kobayashi T,
- Park K,
- Oyamada N,
- Taura D,
- Tsujimoto H,
- Chao TH,
- Tamura N,
- Mukoyama M,
- Nakao K
- Kondoh K,
- Koyama H,
- Miyata T,
- Takato T,
- Hamada H,
- Shigematsu H
- Ichikawa-Shindo Y,
- Sakurai T,
- Kamiyoshi A,
- Kawate H,
- Iinuma N,
- Yoshizawa T,
- Koyama T,
- Fukuchi J,
- Iimuro S,
- Moriyama N,
- Kawakami H,
- Murata T,
- Kangawa K,
- Nagai R,
- Shindo T
- Arai K,
- Lo EH