Apoptosis Induced by Inhibition of Cyclic AMP Response Element–Binding Protein in Vascular Smooth Muscle Cells
Background— The balance between apoptosis and proliferation of vascular smooth muscle cells (VSMCs) is believed to contribute to the vascular remodeling process. Cyclic AMP response element–binding protein (CREB) is a critical transcription factor for the survival of neuronal cells and T lymphocytes. However, the role of CREB in blood vessels is incompletely characterized.
Methods and Results— Nuclear staining with Hoechst 33258 or propidium iodine showed an increase in apoptotic cells with activation of caspase-3 in VSMCs infected with adenovirus expressing the dominant-negative form of CREB (AdCREBM1). Basal expression of Bcl-2 and Bcl-2 promoter activity were decreased by infection with AdCREBM1. Immunohistochemistry revealed that CREB was mainly induced and activated in the neointimal α-smooth muscle actin–positive cells of rat carotid artery after balloon injury. Infection with AdCREBM1 suppressed neointimal formation (intima-media ratio) by 33.8% after 14 days of injury, which was accompanied by an increase in apoptosis as indicated by terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling–positive cells and a decrease in bromodeoxyuridine incorporation.
Conclusions— These results suggest that CRE-dependent gene transcription might play an important role in the survival and proliferation of VSMCs. CREB might be a novel transcription factor mediating the vascular remodeling process and a potential therapeutic target for atherosclerotic disease.
Received December 3, 2002; revision received May 5, 2003; accepted May 9, 2003.
Apoptosis, programmed cell death, is believed to contribute to the maintenance of cell numbers in normal tissues through exclusion of damaged cells. Cells undergoing apoptosis show characteristic morphological changes, such as shrinkage, chromatin condensation followed by internuclosomal DNA fragmentation, and membrane budding. Because these changes occur with maintenance of membrane integrity, apoptotic cells do not cause an inflammatory reaction, which contrasts with necrosis.1 Apoptosis plays an important role in normal development as a means of fine-tuning of growth and is involved in a variety of diseases, including cancer, neurological disorders, and cardiovascular diseases.
See p 1164
Apoptosis of vascular smooth muscle cells (VSMCs) is observed in physiological remodeling of the vasculature and in diseases such as atherosclerosis and restenosis after angioplasty.2,3 After vascular injury, apoptosis of medial VSMCs occurs within several hours.4 The second wave of apoptosis occurs in intimal VSMCs after 1 to 3 weeks of injury and limits the progression of the neointimal lesion.5 These studies suggest that the process of cell growth and apoptosis is tightly linked and that the balance between apoptosis and proliferation of VSMCs regulates the vascular remodeling process.
Apoptosis is regulated by specific intracellular proteins, such as Bcl-2 and caspase.6 The Bcl-2 family members are composed of antiapoptotic proteins, such as Bcl-2 and Bcl-xL, and proapoptotic proteins, such as BAX, Bad, and Bcl-xS. It was previously reported that BAX was expressed in human atherosclerotic plaques,7 suggesting that apoptosis occurs in atherosclerotic plaques. The balance of expression between antiapoptotic and proapoptotic Bcl-2 family member is a critical determinant as to whether a cell undergoes apoptosis or not. Bcl-2 family members regulate the release of cytochrome c from mitochondria. Proapoptotic signals induce the release of cytochrome c, and the released cytochrome c binds to Apaf-1, resulting in the activation of caspase-3.8
In neuronal cells, expression of Bcl-2 is regulated by the cyclic AMP response element–binding protein (CREB),9 which is a nuclear transcription factor. Inhibition of CREB function induces apoptosis in neuronal cells.9 Overexpression of the dominant-negative form of CREB in the heart of mice showed dilatation of the ventricles, which mimicked dilated cardiomyopathy.10 However, the role of CREB in blood vessels is incompletely characterized.
We show in the present study that overexpression of dominant-negative CREB in VSMCs induces apoptosis, with a decrease in Bcl-2 expression. Dominant-negative CREB transduced to injured artery suppresses neointimal formation, concomitant with a decrease in bromodeoxyuridine (BrdU) uptake and an increase in apoptosis.
Dulbecco’s modified Eagle medium (DMEM) and fetal bovine serum were purchased from GIBCO BRL. The annexin V–fluorescein isothiocyanate (FITC) detection kit was obtained from Trevigen, Inc. Thrombin was purchased from Ito Ham Co. One unit per milliliter of thrombin is approximately equivalent to 20 nmol/L. Bcl-2/luciferase fusion DNA was a generous gift from Dr Linda M. Boxer (Stanford University, Stanford, Calif).11 Antibodies against Bcl-2 and α-tubulin were obtained from Upstate Biotechnology and Sigma, respectively. Antibodies against proliferating cell nuclear antigen (PCNA) and α-smooth muscle (SM) actin were purchased from Santa Cruz Biotechnology. The horseradish peroxidase–conjugated secondary antibodies (anti-rabbit and anti-mouse IgG) were obtained from Vector Laboratories Inc. Other antibodies were obtained from New England Biolabs. A recombinant adenovirus vector expressing a mutant of CREB (AdCREBM1), in which the phosphorylation site at Ser133 was changed to alanine, was a gift from Dr Anthony J. Zeleznik (University of Pittsburgh, Pittsburgh, Pa)12. Unless mentioned otherwise, other chemical reagents were purchased from Wako Pure Chemicals.
Preparation and serum starvation of VSMCs were performed as described previously.13
Infection of Cultured VSMCs With AdCREBM1 and AdLacZ
Infection with adenovirus vector (AdCREBM1 and AdLacZ; multiplicity of infection=30) of VSMCs was performed as described previously.14 The infection efficiency of adenovirus is almost 100%, as determined by β-galactosidase activity expressed by AdLacZ (data not shown).
Detection of Apoptosis
Cells were harvested through trypsinization and stained with Hoechst 33258. The number of apoptotic cells (cell shrinkage, chromatin condensation, and nuclear fragmentation) was counted in 1000 cells by fluorescence microscopy. Cells were doubly stained with Annexin V–FITC and propidium iodine (PI) for 15 minutes in the dark. Fluorescence of FITC or PI of 10 000 cells was measured by flow cytometry (FACSCalibur flow cytometer, Becton Dickinson). Cells were incubated with cold 70% ethanol and stained with PI in the presence of RNase. The fluorescence of PI from 10 000 cells was measured by flow cytometry.
Transfection of Bcl-2/Luciferase Fusion DNA Construct to VSMCs
Five micrograms of Bcl-2/luciferase fusion DNA11,9 and 2 μg of LacZ gene driven by the simian virus 40 promoter-enhancer sequence were introduced to VSMCs with the diethylaminoethyl dextran method, as described previously.15 After 48 hours of transfection, the cells were washed with PBS and stimulated with 1 U/mL thrombin for 6 hours in DMEM with 0.1% bovine serum albumin. Then, luciferase activity was measured and normalized against β-galactosidase activity, as described previously.16
Northern Blot Analysis
Western Blot Analysis
Western blot analyses were performed with conventional methods and an enhanced chemiluminescence detection kit (ECL, Amersham Pharmacia Biotech), as described previously.14
Balloon Injury Model and Infection With Adenovirus
All procedures were approved by the institutional animal use and care committee and were conducted in conformity with institutional guidelines. A male Wistar rat (300 to 350 g) was anesthetized by intraperitoneal administration of pentobarbital sodium (50 mg/kg). The left common carotid artery was denuded of the endothelium with a 2F Fogarty balloon catheter (Baxter) that was introduced through the external carotid artery. Inflation and retraction of the balloon catheter were repeated 3 times.20 AdCREBM1 or AdLacZ was introduced into the lumen, and the carotid artery was incubated for 20 minutes without blood flow. Then the viral solution was removed, and blood flow was restored. β-Galactosidase activity was observed in both the neointima and media in the AdLacZ-infected artery (data not shown).
Morphometry and Immunohistochemistry
After the rats (n=5) were killed with an overdose of pentobarbital, the carotid artery was perfusion-fixed with formaldehyde at 100 mm Hg, excised, and embedded in paraffin. The samples were autoclaved at 121°C for 1 minute to retrieve antigen. Serial cross sections of the carotid rings were stained with hematoxylin and eosin and subjected to morphometry for assessing the intima-media area ratio (I/M ratio) and cross-sectional area and to immunohistochemistry with use of a denoted primary antibody and a commercially available detection system (Dako).
Detection of Apoptosis and DNA Synthesis In Vivo
Apoptotic cells were detected by the terminal deoxynucleotidyl transferase (TdT)–mediated dUTP nick end-labeling (TUNEL) method with an apoptosis in situ detection kit (Wako Pure Chemicals). The counterstain was hematoxylin. In vivo labeling with BrdU (0.5 mg/kg), a thymidine analogue that was injected intraperitoneally 24 hours before preparation of the artery, was performed to identify replicating cells by detection of DNA synthesis. Incorporated BrdU was detected immunohistochemically with an anti-BrdU antibody (cell proliferation kit, Amersham Pharmacia Biotech). Quantitative analysis was performed in 5 independent sections from each rat (n=5). The ratio of TUNEL- or BrdU-positive cells to total nucleated VSMCs was expressed as the TUNEL index or BrdU labeling index, respectively.
Statistical analyses were performed by 1-way ANOVA and multiple comparison tests (Fisher) when appropriate. A value of P<0.05 was considered significant. Data were expressed as mean±SEM.
Dominant-Negative CREB Induced Apoptosis in VSMCs
Two days after infection with AdCREBM1 or AdLacZ, VSMCs were stimulated with thrombin (2 U/mL) for an additional 2 days. Staining with Hoechst 33258 showed an increase in apoptotic cells, characterized by chromatin condensation and nuclear fragmentation, in VSMCs infected with AdCREBM1 (4.9%, P<0.01) compared with those infected with AdLacZ (Figure 1a). We previously reported that thrombin induces proliferation of VSMCs through activation of CREB.14 Thrombin, however, increased the proportion of apoptotic cells (7.3% versus 4.9%; P<0.01) in VSMCs infected with AdCREBM1 but not in those infected with AdLacZ (Figure 1a).
Annexin V specifically binds to phosphatidylserine, which is exposed at the cell surface in the preapoptotic stage.21 Staining with PI indicates dead cells. VSMCs that were stained positively with Annexin V–FITC and negatively with PI were increased by infection with AdCREBM1 (20.8% versus 1.9% in the AdLacZ group, P<0.01; Figure 1b).
Flow cytometric analysis of DNA content by PI staining showed an increase in hypodiploid cells in VSMCs infected with AdCREBM1 (32.9% versus 6.2% in the AdLacZ group, P<0.01; Figure 1c), indicating an increase in DNA fragmentation. These hypodiploid cells increased in a time-dependent manner (Figure 1c). These results suggest that inhibition of CREB function induces apoptosis in VSMCs.
Bcl-2 Expression Was Decreased by AdCREBM1
The promoter region of the Bcl-2 gene contains the CRE site (TGACGTTA).11 Infection with AdCREBM1 reduced both the basal and thrombin-induced promoter activity of the Bcl-2 gene (Figure 2a). Deletion of the CRE site or introduction of a point mutation into the CRE site of the Bcl-2 gene promoter abolished thrombin-induced upregulation (Figure 2b). Basal expression of Bcl-2 mRNA and Bcl-2 protein expression was also decreased by infection with AdCREBM1 (Figure 2c and 2d). These results suggest that CREB plays a critical role in Bcl-2 expression. Infection with AdCREBM1 had no effect on expression of Akt (Figure 2d), which is an antiapoptotic gene. Infection with AdCREBM1 as well as application of H2O2 (200 μmol/L, 4 hours) induced cleavage of caspase-3 (Figure 2e), confirming that the caspase pathway is activated by inhibition of CREB function. Expression of PCNA, an auxiliary factor for DNA polymerase-δ required for DNA synthesis, is also regulated by CRE.22 Infection with AdCREBM1 also inhibited thrombin- or serum-induced upregulation of PCNA (Figure 2f).
Expression and Activation of CREB in Balloon-Injured Artery
Immunoreactive CREB was not detected in the intact artery by immunohistochemistry (Figure 3a). However, very weak expression of CREB was detectable in intact arteries by Western blot analysis,23 suggesting that the expression level of CREB in the intact artery might be below the detectable level of our immunohistochemistry methods. At 7 days after balloon injury, expression of CREB was observed in nuclei of the neointima, and to a lesser extent, in medial VSMCs. At 14 days, the number of CREB-positive cells was decreased but was still higher than that of the intact artery. CREB-positive cells were colabeled with a phosphospecific CREB antibody, indicating that CREB is activated in vivo. The number of phosphoCREB-labeled cells declined to lower levels at 14 days (Figure 3a). These changes were consistent with the results of Western blot analysis (Figure 3b). CREB-positive cells were also colabeled with the α-SM actin antibody (Figure 3c), confirming that CREB is expressed in VSMCs.
Infection with AdCREBM1 of balloon-injured arteries suppressed neointimal formation (I/M ratio) by 33.8% compared with infection with AdLacZ after 14 days of balloon injury (Figure 4a). As shown Figure 4b, CREB expression was increased and staining with phosphoCREB was decreased in AdCREBM1-infected arteries. Balloon injury caused negative remodeling, and AdCREBM1 attenuated this negative remodeling (Figure 4c).
Effect of AdCREBM1 on TUNEL Index and BrdU Incorporation in Injured Arteries
It was previously reported that apoptotic cells were detected in neointimal VSMCs and that the TUNEL index peaked after 7 days of vascular injury24. After 7 days of balloon injury, TUNEL index in the neointima of AdCREBM1-infected arteries was significantly increased compared with that of AdLacZ-infected arteries (55.4% versus 25.4%; P<0.01; Figure 4a), whereas AdCREBM1 did not affect the TUNEL index in medial cells (Figure 5a). In the intact artery, BrdU-positive VSMCs were not detected in the media or intima (data not shown). The BrdU labeling index in the neointima reached 50% in the AdLacZ-infected artery at 7 days. Infection with AdCREBM1 reduced the BrdU labeling index in the neointima by 40% compared with infection with AdLacZ at 7 days (Figure 5b). It is not clear at this point whether the decreased BrdU uptake was due to an antiproliferative effect of dominant-negative CREB or merely reflected the decreased cell number after apoptosis. The number of BrdU-positive VSMCs in the media at 7 days was lower than that in the intima, and AdCREBM1 infection had no apparent effect on BrdU labeling index in the media.
We show in the present study that inhibition of CREB function induces apoptosis of VSMCs, possibly through downregulation of Bcl-2 expression. We show that CREB was induced and activated in the neointima after balloon injury of the rat carotid artery and that transduction of dominant-negative CREB by an adenoviral vector to the injured artery attenuated neointimal formation.
It was previously reported that induction of Bcl-2 through activation of CREB was necessary for nerve growth factor–dependent survival of neuronal cells.9 In contrast to neuronal cells, VSMCs can survive in serum-free medium for a substantial period. We found weak phosphorylation of CREB after 2 days of serum deprivation,14 suggesting that basal CREB activity is important for the survival of serum-deprived VSMCs. Dominant-negative CREB decreased basal Bcl-2 expression, which might induce apoptosis. However, further study is necessary to determine why all of the cells infected with AdCREBM1 did not show apoptotic changes.
Tissue-specific expression of dominant-negative CREB in T lymphocytes by use of a T lymphocyte–specific CD2 promoter/enhancer25 showed cell cycle arrest at the G0/G1 stage and apoptotic death in response to mitogenic signals. CRE regulates expression of many genes26 related to metabolism, gene transcription, growth factors, and the cell cycle, such as cyclin D127 and cyclin A,28 which regulate the G1/S transition. Therefore, inhibition of the induction of cyclin D1 and cyclin A by dominant-negative CREB might account for G0/G1 arrest. Inhibition of upregulation of PCNA by AdCREBM1 (Figure 2f) might also contribute to cell cycle arrest.
We previously reported that thrombin induced proliferation of VSMCs through CREB activation.14 However, thrombin increased apoptotic cell death in VSMCs infected with dominant-negative CREB. These data suggest that CREB supports both basal and stimulus-induced survival of VSMCs. It was previously reported that platelet-derived growth factor induced apoptosis of fibroblasts in the absence of serum,29 suggesting that growth factor might regulate cell survival positively or negatively, depending on the potential of cells to transit the cell cycle. Therefore, it is possible that thrombin might induce the survival factor of VSMCs, which permits transition in the cell cycle in a CREB-dependent manner.
Reusch et al30 showed that dominant-negative CREB increased expression of proapoptotic genes, such as caspase-1, and decreased expression of antiapoptotic genes, such as Akt/protein kinase B in adipocytes. The authors failed to see an effect of dominant-negative CREB on Bcl-2 expression. We examined the expression of Akt in VSMCs infected with dominant-negative CREB. We did not find any changes in the expression level of Akt. Although the reason for this discrepancy is not clear, CREB might differentially regulate gene expression, depending on cell type.
The mechanism for the differential response to overexpression of dominant-negative CREB between the neointima and media is not clear. We previously reported that neointimal VSMCs express SMemb, a nonmuscle isoform of SM myosin heavy chain, whereas medial VSMCs express SM2, a mature isoform of SM myosin heavy chain.31 The phenotypic difference of VSMCs between the neointima and media might be one of the mechanisms responsible for the differential response to inhibition of CREB function in arteries. Because expression level of the transduced gene determined by AdLacZ infection was better in the media than in the neointima (data not shown), a more apparent effect of AdCREBM1 on the intima might not be due to a differential expression level of dominant-negative CREB between the intima and media.
A recent report showed that Hex, a homeobox gene, is expressed in neointimal VSMCs after balloon injury in the rat aorta but not in the intact aorta.32 Hex expression was colocalized with the expression of SMemb. Promoter analysis of the SMemb gene revealed that CRE was critical for responsiveness of the promoter to Hex gene expression, suggesting that Hex might function as a transcriptional modulator of CRE-dependent gene transcription. In the present study, we found that CREB was induced and activated in the neointima but not in the intact artery. The expression pattern of CREB is very similar to that of Hex and SMemb, suggesting that CREB and Hex might cooperatively regulate gene expression in the neointima.
Klemm et al33 reported a negative correlation between the expression level of CREB and proliferation of VSMCs. Platelet-derived growth factor decreased the CREB content in VSMCs and increased proliferation. The reason for this apparent discrepancy of results between those of Klemm et al and ours is not clear. Those authors mainly examined VSMCs derived from the pulmonary artery, and the role of CREB might be different in different vascular beds.
In summary, we have shown that inhibition of CREB function induces apoptosis of VSMCs and suppresses neointimal formation after vascular injury. Although further studies are required to determine the target of CREB-regulated genes involved in VSMC proliferation, CREB might be a critical transcription factor mediating the vascular remodeling process.
This study was supported in part by a grant from the Yamanouchi Foundation for Research on Metabolic Disorders, Tsukuba, Japan, and a grant-in-aid for scientific research on priority areas “Medical Genome Science” from the Ministry of Education, Culture, Sports, Science and Technology of Japan (to T. Ichiki) and a grant for Science Frontier Research Promotion Centers from the Ministry of Education, Science, Sports and Culture, Japan (to H. Kai).
↵*The first 2 authors contributed equally to this work.
Fadok VA, Savill JS, Haslett C, et al. Different populations of macrophages use either the vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic cells. J Immunol. 1992; 149: 4029–4035.
Isner JM, Kearney M, Bortman S, et al. Apoptosis in human atherosclerosis and restenosis. Circulation. 1995; 91: 2703–2711.
Walsh K, Smith RC, Kim HS. Vascular cell apoptosis in remodeling, restenosis, and plaque rupture. Circ Res. 2000; 87: 184–188.
Perlman H, Maillard L, Krasinski K, et al. Evidence for the rapid onset of apoptosis in medial smooth muscle cells after balloon injury. Circulation. 1997; 95: 981–987.
Gross A, McDonnell JM, Korsmeyer SJ. BCL-2 family members and the mitochondria in apoptosis. Genes Dev. 1999; 13: 1899–1911.
Kockx MM, De Meyer GR, Muhring J, et al. Apoptosis and related proteins in different stages of human atherosclerotic plaques. Circulation. 1998; 97: 2307–2315.
Riccio A, Ahn S, Davenport CM, et al. Mediation by a CREB family transcription factor of NGF-dependent survival of sympathetic neurons. Science. 1999; 286: 2358–2361.
Wilson BE, Mochon E, Boxer LM. Induction of bcl-2 expression by phosphorylated CREB proteins during B-cell activation and rescue from apoptosis. Mol Cell Biol. 1996; 16: 5546–5556.
Funakoshi Y, Ichiki T, Ito K, et al. Induction of interleukin-6 expression by angiotensin II in rat vascular smooth muscle cells. Hypertension. 1999; 34: 118–125.
Tokunou T, Ichiki T, Takeda K, et al. cAMP response element-binding protein mediates thrombin-induced proliferation of vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 2001; 21: 1764–1769.
Tokunou T, Ichiki T, Takeda K, et al. Thrombin induces interleukin-6 expression through the cAMP response element in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 2001; 21: 1759–1763.
Takeda K, Ichiki T, Tokunou T, et al. Peroxisome proliferator-activated receptor-γ activators downregulate angiotensin II type 1 receptor in vascular smooth muscle cells. Circulation. 2000; 102: 1834–1839.
Ichiki T, Takeda K, Tokunou T, et al. Downregulation of angiotensin II type 1 receptor by hydrophobic 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 2001; 21: 1896–1901.
Yasukawa H, Imaizumi T, Matsuoka H, et al. Inhibition of intimal hyperplasia after balloon injury by antibodies to intercellular adhesion molecule-1 and lymphocyte function-associated antigen-1. Circulation. 1997; 95: 1515–1522.
Koopman G, Reutelingsperger CP, Kuijten GA, et al. Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis. Blood. 1994; 84: 1415–1420.
Huang D, Shipman-Appasamy PM, Orten DJ, et al. Promoter activity of the proliferating-cell nuclear antigen gene is associated with inducible CRE-binding proteins in interleukin 2-stimulated T lymphocytes. Mol Cell Biol. 1994; 14: 4233–4243.
Funakoshi Y, Ichiki T, Takeda K, et al. Critical role of cAMP-response element-binding protein for angiotensin II-induced hypertrophy of vascular smooth muscle cells. J Biol Chem. 2002; 277: 18710–18717.
Shibata R, Kai H, Seki Y, et al. Role of Rho-associated kinase in neointima formation after vascular injury. Circulation. 2001; 103: 284–289.
D’Amico M, Hulit J, Amanatullah DF, et al. The integrin-linked kinase regulates the cyclin D1 gene through glycogen synthase kinase 3β and cAMP-responsive element-binding protein-dependent pathways. J Biol Chem. 2000; 275: 32649–32657.
Kim HR, Upadhyay S, Li G, et al. Platelet-derived growth factor induces apoptosis in growth-arrested murine fibroblasts. Proc Natl Acad Sci U S A. 1995; 92: 9500–9504.
Reusch JE, Klemm DJ. Inhibition of cAMP-response element-binding protein activity decreases protein kinase B/Akt expression in 3T3-L1 adipocytes and induces apoptosis. J Biol Chem. 2002; 277: 1426–1432.
Seki Y, Kai H, Shibata R, et al. Role of the JAK/STAT pathway in rat carotid artery remodeling after vascular injury. Circ Res. 2000; 87: 12–18.
Sekiguchi K, Kurabayashi M, Oyama Y, et al. Homeobox protein Hex induces SMemb/nonmuscle myosin heavy chain-B gene expression through the cAMP-responsive element. Circ Res. 2001; 88: 52–58.
Klemm DJ, Watson PA, Frid MG, et al. cAMP response element-binding protein content is a molecular determinant of smooth muscle cell proliferation and migration. J Biol Chem. 2001; 276: 46132–46141.