| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular Biology |
in Injured Arteries Controls SDF-1
–Mediated Neointima Formation in Apolipoprotein E–Deficient MiceFrom the Division of Cardiology (E.K., G.S., A.M., F.K., H.-Y.S., V.K., A.S.), Medizinische Poliklinik, University of Munich; Division of Nephrology (C.C., H.S.), Medizinische Poliklinik, University of Munich; and Institute of Molecular Cardiovascular Research (IMCAR) (A.Z., M.H., C.W.), University Hospital Aachen, Germany.
Correspondence Dr Andreas Schober, Division of Cardiology, Medizinische Poliklinik, University of Munich, Pettenkoferstrasse 8a, 80336 Munich, Germany. E-mail andreas.schober{at}med.uni-muenchen.de
| Abstract |
|---|
|
|
|---|
is the regulatory subunit of a transcriptional complex, which controls the recruitment of multipotent progenitor cells and tissue repair in ischemic tissue by inducing stromal cell-derived factor (SDF)-1
expression. Because HIF-1
can be activated under normoxic conditions in smooth muscle cells (SMCs) by platelet products, we investigated the role of HIF-1
in SDF-1
–mediated neointima formation after vascular injury.
Methods and Results— Wire-induced injury of the left carotid artery was performed in apolipoprotein E–deficient mice. HIF-1
expression was increased in the media as early as 1 day after injury, predominantly in SMCs. Nuclear translocation of HIF-1
and colocalization with SDF-1
was detected in neointimal cells after 2 weeks. HIF-1
mRNA expression was induced at 6 hours after injury as determined by real-time RT-PCR. Inhibition of HIF-1
expression by local application of HIF-1
-siRNA reduced the neointimal area by 49% and significantly decreased the neointimal SMCs content compared with control-siRNA. HIF-1
and SDF-1
expression were clearly diminished in neointimal cells of HIF-1
-siRNA treated arteries.
Conclusions— HIF-1
expression is directly involved in neointimal formation after vascular injury and mediates the upregulation of SDF-1
, which may affect the stem cell–based repair of injured arteries.
We studied the role of hypoxia-inducible factor (HIF)-1
in neointima formation after vascular injury in apolipoprotein E–deficient mice. Inhibition of injury-induced HIF-1
upregulation reduced the neointimal area and stromal cell–derived factor (SDF)-1
expression. This suggests a direct contribution of HIF-1
to SDF-1
–mediated neointima formation after vascular injury.
Key Words: atherosclerosis restenosis vascular biology chemokines progenitor cells smooth muscle cells
| Introduction |
|---|
|
|
|---|
The traditional idea of neointima formation focused on the concept that neointimal SMCs originate from medial cells within the vessel wall, which migrate into the intima and start to proliferate.5,6 In recent years, evidence has accumulated that bone marrow–derived vascular progenitor cells participate significantly to neointima formation after selective recruitment to the injury site, thereby directing the vascular wound healing.7 Smooth muscle progenitor cells (SPCs) have been detected in the peripheral blood of humans8 and appear to be the critical vascular stem cell subtype for vascular repair after mechanical injury. Neointimal lesions after endothelial denudation consist of approximately 50% to 60% bone marrow–derived SMCs.7 Among the multiple chemokines directing vascular cell trafficking, the CXC chemokine stromal cell-derived factor (SDF)-1
and its receptor CXCR4 represent an important molecular signaling axis for the mobilization and recruitment of SPCs.9,10 In a "remote control" function, upregulation of SDF-1
in the injured vessel induces mobilization by increased serum SDF-1
levels and local recruitment of SPCs.11 This SDF-1
–mediated vascular repair by SPCs contributes significantly to neointima formation, because inhibition of SDF-1
or CXCR4-deficiency resulted in reduced neointimal plaques.11,12
Similar findings were recently reported in a model of vascular injury with wire-injury and M-colony stimulating factor (CSF) treatment, where the application of a CXCR4-antagonist inhibited M-CSF–induced neointima formation.13 In addition, platelet-derived SDF-1
has been shown to be critically involved in the recruitment of progenitor cells to arterial thrombi.14
The molecular mechanisms of enhanced SDF-1
gene expression in injured arteries are unknown, but hypoxia-induced SDF-1
upregulation in endothelial cells has clearly been ascribed to the activation of the transcription factor hypoxia-inducible factor (HIF)-1.15 HIF-1 binds directly to specific binding sites in the SDF-1 promoter in hypoxic conditions thereby inducing SDF-1 expression.15
HIF-1 is composed of 2 subunits, HIF-1
and HIF-1β, where HIF-1β is constitutively expressed in normoxic cells and HIF-1
is rapidly degraded by ubiquitination and proteosomal degradation under normoxic conditions.16 Interestingly, nonhypoxic stimuli of HIF-1
activation have been detected in vascular SMCs, such as thrombin and platelet-derived growth factor, which are known to play a relevant role after vascular injury.17–19 These secretory platelet products increase HIF-1
gene expression and protein translation17–19 without being affected by hypoxia. This leads to a substantial accumulation of transcriptionally active HIF-1
in SMCs, which can reach higher levels than after hypoxic stimulation.17 Furthermore, hypertensive medial thickening of the aorta can be accompanied by increased HIF-1 expression in SMCs, suggesting a role of HIF-1
in vascular remodelling.20 Therefore, we were prompted to study the expression and activation of HIF-1
after mechanical vascular injury and its contribution to neointima formation.
| Materials and Methods |
|---|
|
|
|---|
siRNA, 7 nmol (140 µmol/L) Lamin A/C siRNA or non-targeting siRNA (siSTABLEv2, Dharmacon) were dissolved in a lipid transfection reagent and mixed with 35% F127 pluronic gel (Sigma-Aldrich), which transforms to a gel at physiological temperatures.22,23 The siRNA containing F127 was applied to the carotid artery immediately after vascular injury. At indicated time points, carotid arteries were excised after in situ perfusion with RNase-free PBS or with 4% paraformaldehyde. For immunoblotting carotid arteries were immediately frozen in liquid nitrogen (LN2). Animal experiments were approved by the local authorities and complied with the German animal protection law.
Quantitative Real-Time Polymerase Chain Reaction
Details are described in the data supplement (available online at http://atvb.ahajournals.org).
Western Blot Analysis
Details are described in the data supplement.
Histomorphometry, Immunohistochemistry, and Immunofluorescence
Details are described in the data supplement.
Quantification of Circulating Sca-1+/Lin– Cells
Sca-1+/lin– cells were quantified in anticoagulated blood collected 24 hours after wire injury from HIF-1
siRNA- or control siRNA-treated apoE–/– mice. After erythrocyte lysis (BD fluorescence-activated-cell sorter lysing solution), cells were incubated with phycoerythrin-conjugated mouse Sca-1 (clone E13–161.7, Pharmingen) and fluorescein isothiocyanate (FITC)-conjugated mouse lineage panel (CD11b, CD45, Gr-1, TER-119, CD3
; BD Pharmingen) and quantified by flow cytometry (FACSCalibur, BD Bioscience).11
Statistical Analysis
Data represent mean±SEM and were compared by either 2-tailed Student t test and Welch correction if appropriate or 1-way ANOVA followed by Newman-Keuls post-test (Prism 4.0, GraphPad). Differences with P<0.05 were considered statistically significant.
| Results |
|---|
|
|
|---|
After Vascular Injury in ApoE–/– Mice
mRNA levels in carotid arteries at 2 hours, 6 hours, and 7 days after vascular injury and in uninjured control arteries by quantitative real-time RT-PCR. In uninjured carotid arteries and 2 hours after wire injury HIF-1
mRNA expression was barely detectable. At 6 hours after arterial injury a significant upregulation of HIF-1
mRNA compared with control and 2 hours after injury was found (n=3, P<0.05; Figure 1A). After this peak at 6 hours, the HIF-1
mRNA expression in carotid tissue declined at 7 days after injury, achieving levels statistically not different from those of uninjured controls (Figure 1A). Accordingly, HIF-1
protein expression was significantly upregulated in carotid arteries after vascular injury as demonstrated by Western blot analysis (Figure 1B and 1C). Compared with uninjured arteries, HIF-1
protein was increased 24 hours after wire-induced endothelial denudation. Peak HIF-1
protein levels were observed at 2 weeks after injury (Figure 1B and 1C).
|
Localization of HIF-1
Expression in Injured Arteries
Next we studied the localization HIF-1
in sham-operated and injured carotid arteries of apoE–/– mice by immunostaining. HIF-1
was weakly expressed as compared with negative-control sections (Figure 2B) in the media of uninjured arteries but not in endothelial cells (Figure 2A); a distinct expression of HIF-1
was detectable in medial SMCs directly adjacent to the injury site 1 day after injury (Figure 2C). At 2 weeks after injury, a more robust staining of HIF-1
was observed in neointimal SMCs and luminal cells (Figure 2D).
|
In addition, the cellular localization of HIF-1
after vascular injury was investigated by double immunofluorescence microscopy for HIF-1
and the SMC-marker
-SMA (Figure 2E and 2F). Most of the HIF-1
expressing cells at 1 day after wire injury were
-SMA positive, indicating that HIF-1
was primarily induced in SMCs (Figure 2G). To verify the specificity of
-SMA staining for neointimal SMCs, colocalization of
-SMA with myocardin, which regulates SMC-specific protein expression,24 was demonstrated by double immunofluorescence staining (supplemental Figure I). In addition, in some sections significant HIF-1
expression was also detectable in luminal
-SMA–negative cells, most likely representing endothelial cells.
Nuclear Translocation of HIF-1
Immunofluorescence staining for HIF-1
in carotid arteries at 1 day after injury demonstrated a distinct cytoplasmatic expression of HIF-1
in medial SMCs (supplemental Figure IIB and IIC), whereas an irrelevant control antibody showed no specific staining (supplemental Figure IIA and IID). Nuclear HIF-1
, however, was only detectable in a small number of cells after 1 day. When Western blot analysis of nuclear extracts of injured carotid arteries was performed, an increase of nuclear HIF-1
protein was observed after 1 hour and 3 hours, suggesting early HIF-1
activation after vascular injury (supplemental Figure III). Interestingly, 2 weeks after injury HIF-1
was clearly detected in the nucleus and the perinuclear region in nearly all neointimal cells including luminal cells, most likely representing endothelial cells (supplemental Figure IIE and IIF).
Colocalization of HIF-1
and SDF-1
Because HIF-1
activation has been shown to induce SDF-1
expression in ischemic tissue, we determined whether cellular colocalization of HIF-1
and SDF-1
is detectable in carotid arteries after wire injury. At 1 day after injury, HIF-1
and SDF-1
expression was clearly evident in the majority of medial cells by double immunofluorescence staining; however, some endothelial cells were HIF-1
–positive without detectable amounts of SDF-1
(supplemental Figure IVA through IVC). Furthermore, 2 weeks after vascular injury most neointimal cells exhibited a concomitant expression of HIF-1
and SDF-1
(supplemental Figure IVD through IVF).
Effect of HIF-1
siRNA on HIF-1
Expression
To establish successful gene silencing by the periadventitial application of siRNA dissolved in pluronic gel, the housekeeping gene Lamin A/C was used as positive control. After 72 hours Lamin A/C protein expression was significantly reduced to 35% by Lamin A/C siRNA compared with nontargeting siRNA, as demonstrated by Western blot analysis (supplemental Figure V). Accordingly, the application of HIF-1
specific siRNA in pluronic gel after wire-induced carotid injury resulted in a marked downregulation of HIF-1
expression as compared with control siRNA-treated mice. Immunofluorescence staining for HIF-1
carotid sections obtained 4 weeks after wire injury unveiled an intense cytoplasmic and nuclear HIF-1
expression in medial and neointimal cells of control siRNA-treated mice (Figure 3A), which was identical to the staining pattern of HIF-1
at 2 weeks after injury in mice without siRNA treatment (supplemental Figures IIE and IVD). In HIF-1
siRNA–treated carotid arteries, HIF-1
expression was almost absent 4 weeks after injury as determined by immunofluorescence (Figure 3B). Quantification of HIF-1
specific immunofluorecence by planimetry revealed a reduction of HIF-1
expression in neointimal cells by 82% in HIF-1
siRNA–treated mice compared with control (0.11±0.05 versus 0.64±0.04 HIF-1
/neointima ratio, n=3, P<0.05; Figure 3C). This was confirmed by Western analysis, demonstrating that HIF-1
protein was significantly decreased in HIF-1
-siRNA–treated mice (Figure 3D and 3E).
|
SDF-1
Expression and SPC Mobilization in HIF-1
siRNA–Treated Mice
SDF-1
immunofluorescence staining was also reduced in carotid sections of HIF-1
siRNA–treated compared with control siRNA-treated mice. A strong SDF-1
expression was identified by immunofluorescence in neointimal and medial cells 4 weeks after injury (Figure 4A), whereas in HIF-1
siRNA–treated mice, SDF-1
was barely detectable (Figure 4B). Planimetric analysis demonstrated a reduction of SDF-1
specific staining by 78% in HIF-1
siRNA–treated compared with control siRNA-treated mice (0.12±0.002 versus 0.56±0.03 SDF-1
/neointima ratio; n=3 to 5, P<0.05) (Figure 4C).
|
The functional significance of reduced SDF-1
expression after HIF-1
siRNA treatment was determined by studying the mobilization of Sca-1+/lin– cells after wire-induced injury, a process which has been previously shown to be SDF-1
–dependent.11 The application of HIF-1
siRNA clearly reduced the percentage of circulating Sca-1+/lin– cells compared with control siRNA (1.8±0.3% versus 0.62±0.12% of mononuclear cells, n=3, P<0.05) (Figure 4D through 4F). This suggests that the inhibition of SDF-1
expression by HIF-1
siRNA results in diminished mobilization of SPCs after carotid injury.
HIF-1
siRNA Treatment Reduces Neointima Formation
To investigate the functional role of increased HIF-1
expression in injured arteries, neointimal and medial areas were determined in control siRNA- and HIF-1
siRNA–treated mice 4 weeks after carotid injury. Treatment with HIF-1
siRNA resulted in a significant reduction of neointimal area by 49% compared with control siRNA (0.015±0.003 mm2 versus 0.007±0.002 mm2, n=5, 10 sections per mouse, P<0.05; Figure 5A through 5C), whereas the medial area after HIF-1
siRNA application remained unchanged (0.05±0.007 mm2 versus 0.06±0.014 mm2, n=5; Figure 5A, 5B and 5D).
|
Effect of HIF-1
siRNA on Neointimal Cell Composition
Cellular neointima composition was determined by quantitative immunohistochemistry using
-SMA (Figure 6A through 6D) and Mac-2 (Figure 6E through 6G) to detect SMCs and macrophages/monocytes, respectively. The neointimal SMC content was significantly decreased by 58% (0.4±0.03 versus 0.23±0.04, n=5, P<0.05; Figure 6A through 6D) in HIF-1
siRNA–treated mice 4 weeks after injury, whereas the relative macrophage content in the neointima remained unchanged (0.15±0.04 versus 0.16±0.02, n=5; Figure 6E through 6H). The negligible background staining in negative-control sections (Figure 6A and 6E) and the intense staining in positive-control sections (supplemental Figure VI) demonstrate the specificity of the
-SMA and Mac-2 immunostaining.
|
| Discussion |
|---|
|
|
|---|
in vascular repair and neointima formation. Because HIF-1
induces SDF-1
expression in hypoxic conditions15 and SDF-1
regulates stem cell–based neointimal growth,11,12 we hypothesized that HIF-1
could mediate neointima formation by the regulation of SDF-1
expression. Indeed, we found that the HIF-1
was upregulated after carotid injury and nuclear translocation of HIF-1
was evident in neointimal cells, suggesting the activation of HIF-1
–dependent signaling pathways. Furthermore, HIF-1
was localized in SDF-1
–expressing neointimal cells, and the application of HIF-1
–specific siRNA significantly reduced neointimal SDF-1
expression. This was accompanied by reduced neointimal lesions in mice treated with HIF-1
–specific siRNA mainly by diminished neointimal SMC content. In conclusion, our data suggest the direct involvement of HIF-1
in neointimal formation after vascular injury by mediating the upregulation of SDF-1
and thereby the stem cell–based repair of injured arteries.
The cellular response to hypoxia is predominantly mediated by HIF-1, a ubiquitously expressed heterodimeric transcription factor, which regulates the expression of more than 60 genes involved in the cellular adaptation to low oxygen availability.25 Whereas the HIF-1β subunit is readily found in many cell types regardless of the oxygen level, the HIF-1
protein levels are low because of a very short half-life under normoxic conditions. In the presence of oxygen HIF-1
is hydroxylated on specific residues, ubiquinitinated and degraded by the proteasome. Under hypoxic conditions HIF-1
escapes from degradation, enters the nucleus, and forms a heterodimer with HIF-1β, which is transcriptionally active.25 In addition to hypoxic HIF-1 activation, evidence has accumulated that increased HIF-1
protein levels can be induced by nonhypoxic stimuli as well. The treatment of vascular SMCs for instance by angiotensin II, platelet-dependent factors like thrombin or platelet-derived growth factor–AB (PDGF-AB) increases HIF-1
mRNA expression levels and the nuclear accumulation of transcriptionally active HIF-1
.17–19 In contrast to hypoxic stimulation, the increased cellular HIF-1
levels after nonhypoxic activation appear to solely rely on enhanced transcription and translation. Two key mechanisms are involved in the nonhypoxic upregulation of HIF-1
: increased HIF-1
translation is mediated by reactive oxygen species (ROS)-dependent activation of phosphatidylinositol-3 kinase (PI3K) pathway, and enhanced HIF-1
gene transcription by activated diacylglycerol-sensitive protein kinase C (PKC).19 These results suggest a potential role of HIF-1
in the vascular response to injury mediated by hypoxia-independent control mechanisms. After wire-mediated vascular injury, we found an early upregulation of HIF-1
mRNA in the vessel wall and a sustained increase of HIF-1
protein expression predominantly in vascular SMCs. Furthermore, nuclear translocation of HIF-1
was evident in neointimal cells, which supports the concept of nonhypoxic activation of HIF-1
after vascular injury. Although the HIF-1
-inducing molecular pathway after vascular injury remains to be determined, the early adhesion of activated platelets with the subsequent release of growth factors and cytokines, and the generation of thrombin at the injured vessel may affect the HIF-1
expression in SMCs, as previously shown in vitro.
Whereas the intracellular pathway of HIF-1
upregulation in SMCs has been studied in detail, the functional role of HIF-1
in vascular remodeling is largely unknown. In a mouse model of pulmonary hypertension hif1a+/– mice display decreased wall thickness of muscularized pulmonary arterioles compared with wild-type mice.26 Furthermore, hypertension has been shown to induce transient expression of HIF-1
in medial smooth muscle cells of rats with aortic constriction, which was accompanied by medial thickening.20 To study the functional role of HIF-1
in neointima formation after vascular injury, we treated the injured vessels with HIF-1
–specific siRNA and found a significant inhibition of neointimal growth. This reduction in neointimal area was attributable to a diminished accumulation of neointimal SMCs, whereas the intimal macrophage content was not affected. Because neointimal SMC accumulation after wire-induced vascular injury is predominantly attributable to the SDF-1
–dependent recruitment of circulating SMC progenitor cells (SPCs),9 increased HIF-1
expression in injured vessels may be involved in SPC-mediated vascular repair.
In contrast to other CXC chemokines, such as GRO-
or ENA-78, the SDF-1 promoter contains no active binding sites for inflammatory mediators such as NF-
B or NF-IL6.27,28 Initially a promoter structure has been described, which is common in constitutively expressed housekeeping genes and contains binding sites for the ubiquitous Sp1 transcription factor. However, induction of SDF-1
transcription has been described in response to various stimuli, such as VEGF,29 interleukin (IL)-1β, or irradiation. Interestingly, the SDF-1
inducing stimuli appear to be specific for the cell type investigated, but the responsible promoter sequences remain to be identified.28 Recently, the molecular mechanism of hypoxia-induced SDF-1
expression in endothelial cells has been elucidated. The interaction of activated HIF with 2 HIF-1 specific binding sites in the SDF-1 promoter mediates the upregulation of SDF-1
under hypoxic conditions.15 This HIF-1–induced SDF-1
expression increased the recruitment of progenitor cells to ischemic tissue and subsequent neovascularization.15 After mechanical vascular injury an early upregulation of SDF-1
expression in the vessel wall and increased SDF-1
plasma levels has been clearly demonstrated.11 Whereas circulating SDF-1
mediates the mobilization of SPCs, neointimal SDF-1
is essential for local SPC recruitment in vascular repair.11 Moreover, blocking of SDF-1
after vascular injury significantly reduces neointima formation through inhibition of intimal SPC accumulation.12 However, the transcriptional regulation of SDF-1
expression after vascular injury is unknown. Because HIF-1
was upregulated in injured vessels presumably by nonhypoxic stimulation, we studied the role of HIF-1
in neointimal SDF-1
expression. Notably, HIF-1
and SDF-1
were colocalized in neointimal cells and nuclear HIF-1
-translocation was associated with increased neointimal SDF-1
expression. A direct regulation of SDF-1
expression by HIF-1
could be demonstrated by treatment of injured arteries with siRNA specific for HIF-1
. HIF-1
siRNA treatment almost completely abolished the neointimal expression of HIF-1
and SDF-1
, suggesting that SDF-1
expression after vascular injury is predominantly regulated by the transcription factor HIF-1
.
As to the mechanisms responsible for the induction of HIF-1
, the upregulation of SDF-1
in lesional SMCs after arterial injury has been linked to the abundant occurrence of SMC apoptosis, because SDF-1
expression was preceded by apoptosis and inhibited by blocking caspase-dependent apoptosis.12 In addition, some of the pathways involved in nonhypoxic HIF-1
upregulation have been implicated in SMC apoptosis, ie, PKC was found to convert ROS-induced signals from necrotic cell death to the activation of caspase-mediated apoptotic cell death in vascular SMCs.30 Although HIF-1
apparently shares signaling pathways with apoptosis-related events in SMCs to regulate SDF-1
expression, the upstream elements feeding this pathway, possibly conferred by apoptotic or necrotic cell bodies,12 remain to be identified.
In summary, we demonstrated that HIF-1
expression is upregulated after vascular injury and directly contributes to neointima formation. Because HIF-1
regulates SDF-1
expression in injured vessels, HIF-1
could be involved in SDF-1
–mediated neointimal SPC recruitment. These results provide evidence for HIF-1
as a new target in the prevention and treatment of restenosis after angioplasty or stent implantation of stenotic arteries.
| Acknowledgments |
|---|
Sources of Funding
This study was supported by Deutsche Forschungsgemeinschaft (SCHO1056/2-1, WE1913/7-1), the Medical Faculty of the University of Munich (FöFoLe 409), the IZKF "Biomat" (TV-B113d), and by the Friedrich-Baur Foundation (grant to A.S.).
Disclosures
None.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Yau, P. Molnar, M. C. Moon, S. Buhay, J. P. Werner, K. Molnar, L. Saward, D. Del Rizzo, and P. Zahradka meta-Iodobenzylguanidine, an Inhibitor of Arginine-Dependent Mono(ADP-ribosyl)ation, Prevents Neointimal Hyperplasia J. Pharmacol. Exp. Ther., September 1, 2008; 326(3): 717 - 724. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |