Thrombin and NAD(P)H Oxidase–Mediated Regulation of CD44 and BMP4-Id Pathway in VSMC, Restenosis, and Atherosclerosis
http://www.100md.com
Aleksandr E. Vendrov, Nageswara R. Madam
参见附件。
the Carolina Cardiovascular Biology Center, Department of Medicine, University of North Carolina at Chapel Hill.
Abstract
To characterize novel signaling pathways that underlie NAD(P)H oxidase–mediated signaling in atherosclerosis, we first examined differences in thrombin-induced gene expression between wild-type and p47phox–/– (NAD[P]H oxidase–deficient) VSMC. Of the 9000 genes analyzed by cDNA microarray method at the G1/S transition point, 76 genes were similarly and significantly modulated in both the cell types, whereas another 22 genes that encompass various functional groups were regulated in NAD(P)H oxidase–dependent manner. Among these 22 genes, thrombin-induced NAD(P)H oxidase–mediated regulation of Klf15, Igbp1, Ak4, Adamts5, Ech1, Serp1, Sec61a2, Aox1, Aoh1, Fxyd5, Rai14, and Serpinh1 was shown for the first time in VSMC. The role of NAD(P)H oxidase in the regulation of a subset of these genes (CD44, BMP4, Id1, and Id3) was confirmed using modulators of reactive oxygen species (ROS) generation, a ROS scavenger and in gain-of-function experiments. We then characterized regulation of these genes in restenosis and atherosclerosis. In both apoE–/– mice and in a mouse vascular injury model, these genes are regulated in NAD(P)H oxidase–dependent manner during vascular lesion formation. Based on these findings, we propose that NAD(P)H oxidase–dependent gene expression in general, and the CD44 and BMP4-Id signaling pathway in particular, is important in restenosis and atherosclerosis.
Key Words: vascular smooth muscle cells reactive oxygen species microarray redox regulation
Introduction
The development of molecular therapies for atherosclerosis has lagged behind other diseases, perhaps because of the complexity of redundant signaling pathways that govern vascular cell function. Although the precise role of reactive oxygen species (ROS) in vascular smooth muscle cell (VSMC) biology remains controversial, the importance of ROS generation in signaling is not. In VSMC, several signal transduction pathways induced by growth factors and cytokines are mediated by ROS generated by the activation of membrane-bound NAD(P)H oxidase.1 The VSMC NAD(P)H oxidase includes membrane-bound components: Nox1/4 and p22phox; and cytosolic components: Rac1 and p47phox.2 A third cytosolic component, p67phox, found in other vascular cells, is not present in VSMC.3 Recently, p67phox and also p47phox homologs have been reported in VSMC.2 The key role of p47phox in NAD(P)H oxidase activation in response to agonist stimulation and subsequent increase in cell proliferation was demonstrated using wild-type and p47phox–/– VSMC.4
With few exceptions,5 predominant data support strong association of ROS with atherosclerosis. ApoE–/–/ p47phox–/– mice have lower levels of aortic ROS production and less atherosclerosis than apoE–/– mice.4 Other mouse models with altered levels of manganese superoxide dismutase (SOD2)6 and p66Shc7 have produced a consistent theme that increased levels of vascular ROS promote, whereas decreased levels reduce atherogenesis. ROS may enhance atherosclerosis by several mechanisms. Increased ROS production enhances lipid peroxidation and activation of matrix metalloproteinases8 and may induce differentiation of monocytes into macrophages leading to inflammation.9 ROS also regulate the expression of genes involved in the early stages of atherosclerosis by modulating the activity of redox-sensitive transcription factors.10 Furthermore, ROS may induce atherosclerosis by dysregulating VSMC proliferation via activation of tyrosine kinases and mitogen-activated protein kinases.11
However, comprehensive transcriptional responses to oxidative stress that influence VSMC proliferation and migration and atherosclerotic lesion development are not well studied, nor has there been much attention to late-response genes in cell cycle progression that may play critical role in sustaining atherogenesis. Of the many VSMC agonists, thrombin mediates a number of physiologic responses of vascular cells that are involved in atherosclerotic lesion formation. In addition, thrombin like other G protein–coupled receptor agonists and cytokines activates many signaling moieties in NAD(P)H oxidase–derived ROS-dependent manner.1,3
In this study, we examined the role of thrombin-induced NAD(P)H oxidase activation on gene regulation by comparing gene expression in wild-type and p47phox–/– VSMC. Using cDNA microarray and real-time RT-PCR analysis, we identified regulation of 22 ROS-sensitive genes during cell cycle transition from G1 to S phase. We confirm ROS-dependent regulation of a subset of these genes (CD44, BMP4, Id1, and Id3) using DMNQ, a ROS inhibitor, a ROS scavenger and gain-of-function experiments. We also demonstrate ROS-dependent regulation of this subset of genes in arterial injury. Further, our results show decreased expression of CD44 in atherosclerotic lesions of apoE–/–/ p47phox–/– mice compared with lesions in apoE–/– mice. Together, these results suggest that NAD(P)H oxidase–dependent VSMC gene regulation plays an important role in restenosis and atherosclerosis.
Materials and Methods
Cell Culture
VSMC were isolated from aortas of 4-month-old mice and maintained as described previously4 (see the online data supplement available at http://circres.ahajournals.org).
RNA Extraction, cDNA Microarray Analysis, and Real-Time RT-PCR
Detailed information on RNA extraction, cDNA microarray, real-time RT-PCR, and data analysis is available in the online data supplement.
Intracellular ROS Detection
ROS levels in VSMC were detected by staining with dihydroethidium4 (see online data supplement).
Western Blot Analysis
Preparation of cell lysates and immunoblotting were performed as described11 (see online data supplement).
Mice and Femoral Artery Injury
Transluminal femoral artery injury was performed as described.12 All animal procedures were in compliance with the University of North Carolina Institutional Animal Care and Use Committee (see online data supplement).
Statistical Analysis
All numerical data are expressed as mean±SEM. One-way ANOVA with Newman–Keuls post hoc test was used to analyze gene expression levels. Two-tailed unpaired t test was used to compare lesion area. Differences were considered significant at P<0.05.
Results
Thrombin Regulates Gene Expression in VSMC via ROS Generation by NAD(P)H Oxidase
We examined thrombin-induced NAD(P)H oxidase–mediated gene expression at transition from G1-to-S phase of the cell cycle. Maximum increase in phospho-Cdk2 (Thr160) level was observed in VSMC at 8 hours after thrombin treatment (Figure IA in the online data supplement), consistent with this time point representing late G1-to-S phase transition.13 On this basis, we performed the gene expression profiling at 8 hours after thrombin treatment. To identify genes regulated specifically through NAD(P)H oxidase, we compared expression profiles of wild-type and gp91phox–/– VSMC (both posses functional NAD[P]H oxidase)14 to that of p47phox–/– VSMC using cDNA microarray analysis.
We first identified genes with changes in expression similar in all 3 cell types. Expression profiles for 133 genes that passed filtering of raw microarray data were analyzed using 1-class response SAM.15 With 2000 permutations and value at 0.225 corresponding to a false discovery rate (FDR) of 3.04%, 76 genes were identified as potentially significant. Of these, 61 genes were upregulated and 15 were downregulated by thrombin. Thirty-two genes encoded signal transduction proteins; 13 encoded metabolic pathway proteins; another 17 encoded proteins involved in transcriptional regulation; 9 encoded extracellular matrix and cytoskeletal proteins; and 5 genes encoded electron and ion exchange proteins (supplemental Table I).
Then we compared the differences in gene expression profiles of cell types with functional and impaired NAD(P)H oxidase. Microarray data analyzed with multiclass SAM (2000 permutations and value of 0.245 corresponding to a FDR of 5.1%) identified 28 genes as most significantly differentially regulated (supplemental Figure IB) in response to thrombin treatment. Real-time RT-PCR analysis confirmed the transcriptional pattern of 22 of 28 genes identified using SAM (78.5%). Among these 22 genes, 11 were upregulated and 11 were downregulated. These genes fall into various functional groups: 5 genes encoded proteins involved in transcriptional regulation; 5 encoded signal transduction proteins; 5 encoded metabolic pathway proteins; 1 gene encoded electron and ion transport proteins; and another 6 encoded extracellular matrix and cytoskeletal proteins (supplemental Table II). Expression of these genes was significantly changed in wild-type and gp91phox–/– VSMC (P<0.05) but not in p47phox–/– cells. Of the 3 cell types tested, changes in gene expression were greater in gp91phox–/– VSMC. Previously, we observed higher cell growth in gp91phox–/– VSMC compared with wild-type and p47phox–/– VSMC in normal growth conditions.4 It is possible that this accelerated growth phenotype reflects the greater response in gene expression of gp91phox–/– VSMC. Only 5 of the 22 genes, inhibitor of DNA binding 3 (Id3),16 vascular endothelial growth factor receptor-1 (VEGFR-1),17 aldehyde oxidase 1 (Aox1),18 aldehyde oxidase structural homolog 2 (Aoh1),19 and tissue inhibitor of metalloproteinase 3 (Timp-3)19 were previously shown to be redox sensitive. Redox upregulated genes include VEGFR-1, Timp-3, and syndecan-1 and downregulated genes include natriuretic peptide receptor 3 (Npr3). The expression of 12 of the 22 genes, Klf15, Igbp1, Ak4, Adamts5, Ech1, Serp1, Sec61a2, Aox1, Aoh1, Fxyd5, Rai14, and Serpinh1 in VSMC has not been reported before. We chose to further analyze a subset of these 22 genes.
To identify genes that play critical roles in NAD(P)H oxidase–mediated phenotypic and functional features of VSMC, we focused our studies on thrombin- and NAD(P)H oxidase–regulated genes that had similar basal expression levels in p47phox–/– and wild-type VSMC. CD44 expression increased by 3.97-fold (P<0.05) in wild-type VSMC in response to thrombin treatment but did not change significantly in p47phox–/– cells. BMP4, Id1, and Id3 were downregulated by 2.74-, 1.32-, and 1.69-fold (P<0.05), respectively, in wild-type cells but were unchanged in p47phox–/– VSMC following thrombin treatment (Figure 1A).
We then performed Western blot analysis to determine whether the changes in mRNA expression reflected changes in protein expression. CD44 expression was similar in wild-type, gp91phox–/–, and p47phox–/– VSMC (Figure 1B, top). Thrombin treatment induced an increase in CD44 expression in wild-type and gp91phox–/– cells but had no such effect in p47phox–/– VSMC. Basal BMP4 protein levels were similar in all cell types (Figure 1B, top middle). Thrombin induced a significant decrease in this protein levels in wild-type and gp91phox–/– cells, whereas BMP4 levels remain unchanged in p47phox–/– VSMC treated with thrombin. Id1 (Figure 1B, bottom middle) and Id3 (Figure 1B, bottom) protein levels decreased only in wild-type and gp91phox–/– cells in response to thrombin treatment. Thus, the changes in protein expression were similar to those observed by cDNA microarray and real-time RT-PCR.
Confirmation of Redox-Sensitive Regulation of Gene Expression in VSMC
To confirm that the regulation of thrombin-mediated VSMC genes was indeed dependent on redox changes, we performed a number of complementary experiments. First, 2,3-dimethoxy-1,4-naphthoquinone (DMNQ), a redox cycling quinone, was used to simulate NAD(P)H oxidase–mediated transcriptional changes. VSMC were treated with 1 μmol/L DMNQ for 8 hours and gene expression was analyzed by real-time RT-PCR. Changes in expression of CD44, BMP4, Id1 and Id3 in wild-type cells in response to DMNQ treatment were similar to those seen following thrombin treatment. CD44 gene expression was increased by 3.73-fold (P<0.05), whereas expression of BMP4, Id1 and Id3 genes was decreased by 1.93-, 2.34-, 2.12-fold, respectively (P<0.05) (Figure 2A). The regulation of CD44, BMP4, Id1, and Id3 genes in p47phox–/– VSMC treated with DMNQ was similar to that in wild-type cells treated with this quinone, demonstrating that these cells have not lost their capacity to respond to superoxide and confirming that thrombin-mediated superoxide generation does not occur in the cells which lack NAD(P)H oxidase activity. Specifically, CD44 expression was enhanced by 3.1-fold, whereas the expression of BMP4, Id1, and Id3 were decreased by 2.46-, 1.51-, and 1.43-fold, respectively (P<0.05). Our results also indicate that DMNQ generates ROS in VSMC in NAD(P)H oxidase–independent manner.
To corroborate the importance of NAD(P)H oxidase–dependent regulation of gene expression in VSMC treated with thrombin, we pretreated cells with diphenyleneiodonium chloride (DPI), an inhibitor of flavin-containing enzymes including NAD(P)H oxidase, or N-acetyl-L-cysteine (NAC), a ROS scavenger. Wild-type cells pretreated with either NAC or DPI and then treated with thrombin showed little or no changes in expression of CD44, BMP4, Id1, and Id3 genes, similar to that in p47phox–/– cells treated with thrombin. CD44 levels were significantly lower in wild-type VSMC pretreated with NAC or DPI and then treated with thrombin compared with the levels in cells treated with thrombin alone (P<0.001) (Figure 2B, top left). Similarly, BMP4 gene expression levels in wild-type VSMC pretreated with either NAC or DPI and then treated with thrombin were significantly different than the levels in wild-type cells treated with thrombin alone (P<0.001) (Figure 2B, top right). Pretreatment with NAC or DPI before stimulation with thrombin significantly attenuated thrombin-induced decrease in the expression of Id1 (Figure 2B, bottom left) and Id3 (Figure 2B, bottom right) genes in wild-type cells (P<0.05). Taken together, these data confirm NAD(P)H oxidase–derived ROS-dependent regulation of CD44, BMP4, Id1, and Id3 gene expression by thrombin.
Finally, we performed gain-of-function experiments to confirm that replacement of the p47phox restores thrombin-responsive expression of CD44, BMP4, Id1, and Id3 genes in p47phox–/– cells. Western blot analysis showed significantly increased p47phox expression in cells transfected with adenoviral vector containing human p47phox cDNA as compared with control cells (Figure 3A). On thrombin treatment, DHE and Amplex Red fluorescence showed increased ROS generation in transfected p47phox–/– VSMC compared with nontransfected p47phox–/– cells (Figure 3B and supplemental Figure II). Cells transfected with a control vector, Ad-GFP, were not different from nontransfected p47phox–/– cells in ROS generation.
Gene expression changes in p47phox transfected p47phox–/– VSMC were assessed by real-time RT-PCR. There were no significant changes in gene expression after thrombin treatment in nontransfected or p47phox–/– VSMC transfected with Ad-GFP. In contrast, in p47phox–/– cells transfected with Ad-p47phox, changes in CD44, BMP4, Id1, and Id3 gene expression on thrombin treatment (Figure 3C) were similar to those in wild-type cells treated with thrombin. These data reiterate ROS-sensitive regulation of CD44, BMP4, Id1, and Id3 gene expression in VSMC by thrombin and emphasize the role of NAD(P)H oxidase in these events.
Role of BMP4 in Id1 and Id3 Gene Expression in VSMC
Id1 and Id3 are known transcriptional targets of BMP4.20 For this reason, we questioned whether the observed effect of ROS on Id1 and Id3 gene expression was due solely to an effect on BMP4 or whether ROS directly activated Id1 and Id3 gene expression. Treatment with exogenous BMP4 resulted in a significant increase in Id1 and Id3 gene expression at 1 hour followed by a decrease in both wild-type and p47phox–/– VSMC (Figure 4). The most straightforward interpretation of this finding is that thrombin and NAD(P)H oxidase act only on BMP4 gene regulation, and this results in consequent down regulation of Id1 and Id3. This interpretation is consistent with our overall gene expression results and the notion that NAD(P)H oxidase activation induces specific signaling pathways. Surprisingly, Id1 and Id3 gene expression in response to BMP4 was higher in p47phox–/– cells than in wild-type VSMC. The reasons for this discrepancy are not clear. We also observed that BMP4 significantly suppressed thrombin-induced [3H]-thymidine incorporation in wild-type cells (supplemental Figure III). This suggests that higher BMP4 levels in p47phox–/– VSMC treated with thrombin may be 1 of the mechanisms for decreased p47phox–/– VSMC proliferation.4
NAD(P)H Oxidase Regulates Gene Expression in Arterial Wall and Is Necessary for Neointimal Hyperplasia Following Arterial Injury
We next examined whether NAD(P)H oxidase–dependent changes in expression of these genes occurred in arterial VSMC in vivo. The early stages of atherosclerotic lesion formation through plaque rupture and arterial occlusion all depend on VSMC proliferation and migration. Transluminal wire injury of the femoral artery results in endothelial denudation with rapid onset of medial SMC apoptosis, followed by SMC proliferation and migration leading to development of reproducible neointimal hyperplasia.12 In addition, continued and marked thrombin generation was observed following arterial injury,21 suggesting that arterial injury is a reasonable model for examining the role of thrombin in NAD(P)H oxidase–mediated gene expression and neointima formation. Evans blue staining depicted consistent denudation of the endothelium in injured arteries (Figure 5A). Four weeks after injury, statistically significant increase in neointimal hyperplasia (P<0.001) was observed in wild-type mice compared with sham-operated mice (Figure 5B through 5D). There was little or no increase in neointimal hyperplasia in p47phox–/– mice, which is consistent with decreased proliferation of p47phox–/– VSMC compared with wild-type VSMC.4
To determine the molecular basis for VSMC growth and migration in vivo, we compared the CD44, BMP4, and Id protein expression in the femoral artery in wild-type and p47phox–/– mice. CD44 expression was not observed in uninjured arteries of wild-type and p47phox–/– mice, whereas expression of this protein increased substantially in medial and neointimal SMC of wild-type mice 14 days postinjury, consistent with previous findings.22 In contrast, CD44 protein expression was not observed in p47phox–/– mice following injury (Figure 6A). Immunoreactive BMP4 was not seen in sham-operated arteries of either wild-type or p47phox–/– mice. BMP4 expression was also not present in wild-type femoral arteries but markedly increased in the medial VSMC of p47phox–/– mice 14 days after injury (Figure 6B). Id1 and Id3 expression was minimal in uninjured arteries of wild-type and p47phox–/– mice and also in injured arteries from wild-type mice. However, similar to BMP4, expression of Id1 and Id3 was increased following injury in medial VSMC of p47phox–/– mice (Figure 6C and 6D). Changes in protein expression in VSMC were ascertained by smooth muscle -actin staining (Figure 6E). The decrease in CD44 and increase in BMP4, Id1, and Id3 expression in the injured artery of p47phox–/– mice suggest that these proteins might play an important role in restenosis by affecting VSMC proliferation. Further, these results strongly support NAD(P)H oxidase and ROS as important in regulating the expression of these proteins in VSMC in response to arterial injury in a manner similar to that observed in cultured VSMC. Of note, prior studies implicating the NAD(P)H oxidase in the response to arterial injury focused on ROS production in adventitial fibroblasts and endothelial cells.23
ROS Influence Development of Atherosclerotic Lesions Through Regulation of CD44 Expression
ApoE–/–/p47phox–/– mice develop significantly fewer atherosclerotic lesions than apoE–/– littermates.4 Similarly, CD44–/– mice have 50% to 70% less atherosclerotic lesions compared with wild-type mice.24 Following our observation that VSMC CD44 expression was decreased in p47phox–/– mice following injury, we examined expression of CD44 protein in atherosclerotic lesions of apoE–/– and apoE–/–/p47phox–/– animals fed a Western diet.
Immunostaining of representative sections of proximal aorta revealed increased expression of immunoreactive CD44 in lesions of apoE–/– mice compared with apoE–/–/ p47phox–/– mice. CD44 was present in both neointimal and medial VSMC, as well as in macrophages present in lesions. The less intense CD44 staining in apoE–/–/p47phox–/– mice was limited to macrophages; little or no expression of CD44 was observed in arterial VSMC (Figure 7A). Finally, we measured the total area of atherosclerotic lesions in the aortic root. The oil red O–positive area was significantly less in apoE–/–/p47phox–/– mice than in apoE–/– mice (P<0.01) (Figure 7A and 7B). This extends our prior conclusion that atherosclerosis is reduced by deletion of p47phox,4 and, in addition, the diet used in the present study allows to observe differences in lesions in the proximal aortas of apoE–/– and apoE–/–/p47phox–/– mice. The above data further solidify the relationship between NAD(P)H oxidase activity, CD44 expression, and atherosclerosis.
Discussion
The present study was initiated to determine whether ROS-stimulated signaling pathways in VSMC are also activated in the vessel wall during restenosis and atherogenesis. Although targeted inhibition of an intracellular signaling pathway has been successful for the treatment of some malignancies,25 this approach has yet to show the same promise for the treatment or prevention of atherosclerosis. Because many agonists activate VSMC, it is not surprising that inhibition of a single growth factor or cytokine signaling is ineffective in arresting VSMC growth and migration.
The physiological effects of many VSMC agonists require NAD(P)H oxidase activation,1,2 which suggests that analysis of this oxidase-dependent signals might identify potential therapeutic targets for inhibiting VSMC growth and migration. VSMC play important role throughout the atherosclerotic process; therefore, we reasoned that not only would these NAD(P)H oxidase–dependent agonists affect tyrosine/serine phosphorylation but also gene expression of key signal transduction proteins. Based on this notion, the present study was designed to broadly characterize agonist-mediated NAD(P)H oxidase–dependent gene expression in VSMC to understand convergent signaling pathways of importance in atherogenesis. Thrombin was chosen as a model agonist for in vitro studies because the physiologic responses of VSMC to thrombin (ie, contraction, mitogenesis, and migration) all foster atherosclerotic lesion formation and because many signals activated by thrombin require activation of the NAD(P)H oxidase.3
Because little is known about the thrombin-induced late-response genes that likely sustain atherosclerotic lesion formation, we focused on changes in gene expression after 8 hours of thrombin treatment in vitro and on mid-stages of vascular lesion formation in vivo. Microarray analysis of 3 VSMC cell types with functional and nonfunctional NAD(P)H oxidase suggests that 76 genes are significantly and similarly modulated by thrombin. Thrombin-induced NAD(P)H oxidase regulated expression of another 22 genes that belong to various functional groups at the G1/S transition point (supplemental Table II). These data imply that thrombin induces transcriptional networks in VSMC in both NAD(P)H oxidase–dependent and –independent manner. Although we have not investigated the functional relevance of the latter on VSMC phenotype, our results suggest that at least some of the thrombin-induced NAD(P)H oxidase–dependent genes are involved in cell cycle progression. We have shown previously that p47phox–/– VSMC exhibit decreased proliferative response to thrombin treatment compared with wild-type VSMC.4
The 22 NAD(P)H oxidase–dependent genes identified in vitro represented several signaling pathways of known importance for growth factors and cytokines in vascular and nonvascular cells.24,26 Some of these NAD(P)H oxidase–dependent gene products have been show to play an important role in vascular pathophysiology. For example, VEGFR-1 expression was upregulated in injured arteries27 and its enhanced expression in tumor cells was mediated by NAD(P)H oxidase.17 Similarly, Timp-3 was upregulated by ROS20 and also in atherosclerotic tissue and known to influence plaque stability.28 Similar to our observation, Npr3 is downregulated by other inducers of ROS production.29 Only 2 of the 22 NAD(P)H oxidase–dependent genes regulated by thrombin are transcription factors, which indicates temporal regulation of thrombin-mediated gene transcription and interestingly both the genes are downregulated in wild-type cells. To confirm that these genes were indeed regulated in a redox-dependent manner, we characterized expression of a subset of these genes in wild-type and p47phox–/– VSMC treated with DMNQ, an inhibitor of ROS generation, a ROS scavenger and in gain-of-function experiments. These experiments confirmed the importance of ROS generation and the central role of NAD(P)H oxidase activity in the expression of a subset of 22 genes.
CD44, among the thrombin-induced genes in wild-type VSMC, mediates VSMC proliferation induced by hyaluronic acid22 and is expressed in both inflammatory and vascular cells.30 However, our data show for the first time that CD44 expression in mouse VSMC is mediated by NAD(P)H oxidase activation. BMP4, another gene of interest in thrombin-regulated subset of genes, is a member of the transforming growth factor- super family. In concordance with our data, Jeffery et al31 reported inhibition of mitogenesis in human fibroblasts by BMP4 and this was mediated by the upregulation of cell cycle inhibitor p21Cip1. Interestingly, BMP4 stimulates ROS production in endothelial cells in NAD(P)H oxidase–dependent manner.32 The other genes of interest in thrombin-mediated subset of genes are Id1 and Id3. Id class of helix-loop-helix proteins have been linked to cell cycle control,33 and it is suggested that they promote cell proliferation by acting as dominant negative transcription factors for p21Cip1.34 In contrast to our data, upregulation of Id3 expression by angiotensin II leads to VSMC proliferation and this was mediated by ROS production.16 This apparent contradiction could be attributable to temporal regulation of Id proteins in response to agonist treatment. In fact, Forrest et al35 have reported that Id3 expression was upregulated at 1 hour after serum stimulation but declined by 4 hours.
We then further studied the expression of this subset of genes (CD44, BMP4, Id1, and Id3) in vivo. Immunoreactive CD44 levels were higher in the more prominent atherosclerotic lesions found in apoE–/– mice compared with the less prominent lesions found in apoE–/–/p47phox–/– mice (Figure 7). This result is corroborated by the observations that CD44–/– mice have less atherosclerotic lesions than wild-type mice24 and CD44 expression is higher in human atheromatous plaques.36 Neointimal hyperplasia was also significantly greater in wild-type mice than in p47phox–/– mice (Figure 5) and was accompanied by increased levels of immunoreactive CD44 and reduced levels of BMP4 and Id1 and Id3 (Figure 6). In contrast to our data, increased expression of Id3 was reported in injured carotid arteries of rat which correlated with lesion development.37 These contradictions warrant further careful investigation of BMP4-Id1/Id3 pathway in restenosis and atherosclerosis. Because BMP4 has been reported to regulate Id expression,20 we treated VSMC with exogenous BMP4, confirming upregulation of Id genes (Figure 4).
In summary, we have shown that agonist-activated NAD(P)H oxidase mediates expression of a number of VSMC genes which are important in lesion formation. In addition, agonist-induced ROS production and proliferation of VSMC from wild-type mice correlates with injury-induced neointimal hyperplasia and atherosclerotic lesion formation in these mice. A group of genes implicated in VSMC function—CD44, BMP4, Id1, and Id3—are regulated in vascular lesion formation in 2 in vivo models in exactly the manner predicted from studies on cultured cells and directly correlated to the lesion size. These findings support further investigation of redox-regulated gene expression in atherosclerosis and consideration of therapies that modulate specific, convergent intracellular signals, including the CD44 and BMP4/Id signaling pathways for the treatment of restenosis and atherosclerosis.
Acknowledgments
This work was supported by NIH grant HL57352 (to M.S.R.).
Footnotes
Original received October 20, 2005; revision received February 21, 2006; accepted March 29, 2006.
References
Griendling KK, Sorescu D, Ushio-Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494–501.
Brandes RP, Kreuzer J. Vascular NADPH oxidases: molecular mechanisms of activation. Cardiovasc Res. 2005; 65: 16–27. [Order article via Infotrieve]
Patterson C, Ruef J, Madamanchi NR, Barry-Lane P, Hu Z, Horaist C, Ballinger CA, Brasier AR, Bode C, Runge MS. Stimulation of a vascular smooth muscle cell NAD(P)H oxidase by thrombin. Evidence that p47(phox) may participate in forming this oxidase in vitro and in vivo. J Biol Chem. 1999; 274: 19814–19822.
Barry-Lane PA, Patterson C, van der Merwe M, Hu Z, Holland SM, Yeh ET, Runge MS. p47phox is required for atherosclerotic lesion progression in ApoE(–/–) mice. J Clin Invest. 2001; 108: 1513–1522.
Sentman ML, Brannstrom T, Westerlund S, Laukkanen MO, Yla-Herttuala S, Basu S, Marklund SL. Extracellular superoxide dismutase deficiency and atherosclerosis in mice. Arterioscler Thromb Vasc Biol. 2001; 21: 1477–1482.
Ballinger SW, Patterson C, Knight-Lozano CA, Burow DL, Conklin CA, Hu Z, Reuf J, Horaist C, Lebovitz R, Hunter GC, McIntyre K, Runge MS. Mitochondrial integrity and function in atherogenesis. Circulation. 2002; 106: 544–549.
Napoli C, Martin-Padura I, de Nigris F, Giorgio M, Mansueto G, Somma P, Condorelli M, Sica G, De Rosa G, Pelicci P. Deletion of the p66Shc longevity gene reduces systemic and tissue oxidative stress, vascular cell apoptosis, and early atherogenesis in mice fed a high-fat diet. Proc Natl Acad Sci U S A. 2003; 100: 2112–2116.
Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular disease. Arterioscler Thromb Vasc Biol. 2005; 25: 29–38.
Barbieri SS, Eligini S, Brambilla M, Tremoli E, Colli S. Reactive oxygen species mediate cyclooxygenase-2 induction during monocyte to macrophage differentiation: critical role of NADPH oxidase. Cardiovasc Res. 2003; 60: 187–197. [Order article via Infotrieve]
Pueyo ME, Gonzalez W, Nicoletti A, Savoie F, Arnal JF, Michel JB. Angiotensin II stimulates endothelial vascular cell adhesion molecule-1 via nuclear factor-kappaB activation induced by intracellular oxidative stress. Arterioscler Thromb Vasc Biol. 2000; 20: 645–651.
Madamanchi NR, Moon SK, Hakim ZS, Clark S, Mehrizi A, Patterson C, Runge MS. Differential activation of mitogenic signaling pathways in aortic smooth muscle cells deficient in superoxide dismutase isoforms. Arterioscler Thromb Vasc Biol. 2005; 25: 950–956.
Reis ED SS, Coller BS. Mouse model of transluminal femoral artery injury. In: Simon DRC, ed. Vascular Disease and Injury: Preclinical Research. Totowa, NJ: Humana Press; 2000: 103–113.
Ohtsubo M, Theodoras AM, Schumacher J, Roberts JM, Pagano M. Human cyclin E, a nuclear protein essential for the G1-to-S phase transition. Mol Cell Biol. 1995; 15: 2612–2624.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91(phox) homologues in vascular smooth muscle cells: nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888–894. [Order article via Infotrieve]
Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A. 2001; 98: 5116–5121.
Mueller C, Baudler S, Welzel H, Bohm M, Nickenig G. Identification of a novel redox-sensitive gene, Id3, which mediates angiotensin II-induced cell growth. Circulation. 2002; 105: 2423–2428.
Arbiser JL, Petros J, Klafter R, Govindajaran B, McLaughlin ER, Brown LF, Cohen C, Moses M, Kilroy S, Arnold RS, Lambeth JD. Reactive oxygen generated by Nox1 triggers the angiogenic switch. Proc Natl Acad Sci U S A. 2002; 99: 715–720.
Rivera SP, Choi HH, Chapman B, Whitekus MJ, Terao M, Garattini E, Hankinson O. Identification of aldehyde oxidase 1 and aldehyde oxidase homologue 1 as dioxin-inducible genes. Toxicology. 2005; 207: 401–409. [Order article via Infotrieve]
Li WQ, Qureshi HY, Liacini A, Dehnade F, Zafarullah M. Transforming growth factor beta1 induction of tissue inhibitor of metalloproteinases 3 in articular chondrocytes is mediated by reactive oxygen species. Free Radic Biol Med. 2004; 37: 196–207. [Order article via Infotrieve]
Korchynskyi O, ten Dijke P. Identification and functional characterization of distinct critically important bone morphogenetic protein-specific response elements in the Id1 promoter. J Biol Chem. 2002; 277: 4883–4891.
Walters TK, Gorog DA, Wood RF. Thrombin generation following arterial injury is a critical initiating event in the pathogenesis of the proliferative stages of the atherosclerotic process. J Vasc Res. 1994; 31: 173–177. [Order article via Infotrieve]
Jain M, He Q, Lee WS, Kashiki S, Foster LC, Tsai JC, Lee ME, Haber E. Role of CD44 in the reaction of vascular smooth muscle cells to arterial wall injury. J Clin Invest. 1996; 97: 596–603.
Chen Z, Keaney JF Jr, Schulz E, Levison B, Shan L, Sakuma M, Zhang X, Shi C, Hazen SL, Simon DI. Decreased neointimal formation in Nox2-deficient mice reveals a direct role for NADPH oxidase in the response to arterial injury. Proc Natl Acad Sci U S A. 2004; 101: 13014–13019.
Cuff CA, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R, Yeh C, Secreto A, Assoian RK, Rader DJ, Pure E. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest. 2001; 108: 1031–1040.
Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia. Nat Rev Cancer. 2005; 5: 172–183. [Order article via Infotrieve]
Forrest ST, Barringhaus KG, Perlegas D, Hammarskjold ML, McNamara CA. Intron retention generates a novel Id3 isoform that inhibits vascular lesion formation. J Biol Chem. 2004; 279: 32897–32903.
Zhao Q, Egashira K, Hiasa K, Ishibashi M, Inoue S, Ohtani K, Tan C, Shibuya M, Takeshita A, Sunagawa K. Essential role of vascular endothelial growth factor and Flt-1 signals in neointimal formation after periadventitial injury. Arterioscler Thromb Vasc Biol. 2004; 24: 2284–2289.
Fabunmi RP, Sukhova GK, Sugiyama S, Libby P. Expression of tissue inhibitor of metalloproteinases-3 in human atheroma and regulation in lesion-associated cells: a potential protective mechanism in plaque stability. Circ Res. 1998; 83: 270–278.
Sun JZ, Oparil S, Lucchesi P, Thompson JA, Chen YF. Tyrosine kinase receptor activation inhibits NPR-C in lung arterial smooth muscle cells. Am J Physiol Lung Cell Mol Physiol. 2001; 281: L155–L163.
Lazaar AL, Albelda SM, Pilewski JM, Brennan B, Pure E, Panettieri RA Jr. T lymphocytes adhere to airway smooth muscle cells via integrins and CD44 and induce smooth muscle cell DNA synthesis. J Exp Med. 1994; 180: 807–816.
Jeffery TK, Upton PD, Trembath RC, Morrell NW. BMP4 inhibits proliferation and promotes myocyte differentiation of lung fibroblasts via Smad1 and JNK pathways. Am J Physiol Lung Cell Mol Physiol. 2005; 288: L370–L378.
Sorescu GP, Song H, Tressel SL, Hwang J, Dikalov S, Smith DA, Boyd NL, Platt MO, Lassegue B, Griendling KK, Jo H. Bone morphogenic protein 4 produced in endothelial cells by oscillatory shear stress induces monocyte adhesion by stimulating reactive oxygen species production from a nox1-based NADPH oxidase. Circ Res. 2004; 95: 773–779.
Barone MV, Pepperkok R, Peverali FA, Philipson L. Id proteins control growth induction in mammalian cells. Proc Natl Acad Sci U S A. 1994; 91: 4985–4988.
Prabhu S, Ignatova A, Park ST, Sun XH. Regulation of the expression of cyclin-dependent kinase inhibitor p21 by E2A and Id proteins. Mol Cell Biol. 1997; 17: 5888–5896.
Forrest ST, Taylor AM, Sarembock IJ, Perlegas D, McNamara CA. Phosphorylation regulates Id3 function in vascular smooth muscle cells. Circ Res. 2004; 95: 557–559.
Krettek A, Sukhova GK, Schonbeck U, Libby P. Enhanced expression of CD44 variants in human atheroma and abdominal aortic aneurysm: possible role for a feedback loop in endothelial cells. Am J Pathol. 2004; 165: 1571–1581.
Matsumura ME, Li F, Berthoux L, Wei B, Lobe DR, Jeon C, Hammarskjold ML, McNamara CA. Vascular injury induces posttranscriptional regulation of the Id3 gene: cloning of a novel Id3 isoform expressed during vascular lesion formation in rat and human atherosclerosis. Arterioscler Thromb Vasc Biol. 2001; 21: 752–758.
the Carolina Cardiovascular Biology Center, Department of Medicine, University of North Carolina at Chapel Hill.
Abstract
To characterize novel signaling pathways that underlie NAD(P)H oxidase–mediated signaling in atherosclerosis, we first examined differences in thrombin-induced gene expression between wild-type and p47phox–/– (NAD[P]H oxidase–deficient) VSMC. Of the 9000 genes analyzed by cDNA microarray method at the G1/S transition point, 76 genes were similarly and significantly modulated in both the cell types, whereas another 22 genes that encompass various functional groups were regulated in NAD(P)H oxidase–dependent manner. Among these 22 genes, thrombin-induced NAD(P)H oxidase–mediated regulation of Klf15, Igbp1, Ak4, Adamts5, Ech1, Serp1, Sec61a2, Aox1, Aoh1, Fxyd5, Rai14, and Serpinh1 was shown for the first time in VSMC. The role of NAD(P)H oxidase in the regulation of a subset of these genes (CD44, BMP4, Id1, and Id3) was confirmed using modulators of reactive oxygen species (ROS) generation, a ROS scavenger and in gain-of-function experiments. We then characterized regulation of these genes in restenosis and atherosclerosis. In both apoE–/– mice and in a mouse vascular injury model, these genes are regulated in NAD(P)H oxidase–dependent manner during vascular lesion formation. Based on these findings, we propose that NAD(P)H oxidase–dependent gene expression in general, and the CD44 and BMP4-Id signaling pathway in particular, is important in restenosis and atherosclerosis.
Key Words: vascular smooth muscle cells reactive oxygen species microarray redox regulation
Introduction
The development of molecular therapies for atherosclerosis has lagged behind other diseases, perhaps because of the complexity of redundant signaling pathways that govern vascular cell function. Although the precise role of reactive oxygen species (ROS) in vascular smooth muscle cell (VSMC) biology remains controversial, the importance of ROS generation in signaling is not. In VSMC, several signal transduction pathways induced by growth factors and cytokines are mediated by ROS generated by the activation of membrane-bound NAD(P)H oxidase.1 The VSMC NAD(P)H oxidase includes membrane-bound components: Nox1/4 and p22phox; and cytosolic components: Rac1 and p47phox.2 A third cytosolic component, p67phox, found in other vascular cells, is not present in VSMC.3 Recently, p67phox and also p47phox homologs have been reported in VSMC.2 The key role of p47phox in NAD(P)H oxidase activation in response to agonist stimulation and subsequent increase in cell proliferation was demonstrated using wild-type and p47phox–/– VSMC.4
With few exceptions,5 predominant data support strong association of ROS with atherosclerosis. ApoE–/–/ p47phox–/– mice have lower levels of aortic ROS production and less atherosclerosis than apoE–/– mice.4 Other mouse models with altered levels of manganese superoxide dismutase (SOD2)6 and p66Shc7 have produced a consistent theme that increased levels of vascular ROS promote, whereas decreased levels reduce atherogenesis. ROS may enhance atherosclerosis by several mechanisms. Increased ROS production enhances lipid peroxidation and activation of matrix metalloproteinases8 and may induce differentiation of monocytes into macrophages leading to inflammation.9 ROS also regulate the expression of genes involved in the early stages of atherosclerosis by modulating the activity of redox-sensitive transcription factors.10 Furthermore, ROS may induce atherosclerosis by dysregulating VSMC proliferation via activation of tyrosine kinases and mitogen-activated protein kinases.11
However, comprehensive transcriptional responses to oxidative stress that influence VSMC proliferation and migration and atherosclerotic lesion development are not well studied, nor has there been much attention to late-response genes in cell cycle progression that may play critical role in sustaining atherogenesis. Of the many VSMC agonists, thrombin mediates a number of physiologic responses of vascular cells that are involved in atherosclerotic lesion formation. In addition, thrombin like other G protein–coupled receptor agonists and cytokines activates many signaling moieties in NAD(P)H oxidase–derived ROS-dependent manner.1,3
In this study, we examined the role of thrombin-induced NAD(P)H oxidase activation on gene regulation by comparing gene expression in wild-type and p47phox–/– VSMC. Using cDNA microarray and real-time RT-PCR analysis, we identified regulation of 22 ROS-sensitive genes during cell cycle transition from G1 to S phase. We confirm ROS-dependent regulation of a subset of these genes (CD44, BMP4, Id1, and Id3) using DMNQ, a ROS inhibitor, a ROS scavenger and gain-of-function experiments. We also demonstrate ROS-dependent regulation of this subset of genes in arterial injury. Further, our results show decreased expression of CD44 in atherosclerotic lesions of apoE–/–/ p47phox–/– mice compared with lesions in apoE–/– mice. Together, these results suggest that NAD(P)H oxidase–dependent VSMC gene regulation plays an important role in restenosis and atherosclerosis.
Materials and Methods
Cell Culture
VSMC were isolated from aortas of 4-month-old mice and maintained as described previously4 (see the online data supplement available at http://circres.ahajournals.org).
RNA Extraction, cDNA Microarray Analysis, and Real-Time RT-PCR
Detailed information on RNA extraction, cDNA microarray, real-time RT-PCR, and data analysis is available in the online data supplement.
Intracellular ROS Detection
ROS levels in VSMC were detected by staining with dihydroethidium4 (see online data supplement).
Western Blot Analysis
Preparation of cell lysates and immunoblotting were performed as described11 (see online data supplement).
Mice and Femoral Artery Injury
Transluminal femoral artery injury was performed as described.12 All animal procedures were in compliance with the University of North Carolina Institutional Animal Care and Use Committee (see online data supplement).
Statistical Analysis
All numerical data are expressed as mean±SEM. One-way ANOVA with Newman–Keuls post hoc test was used to analyze gene expression levels. Two-tailed unpaired t test was used to compare lesion area. Differences were considered significant at P<0.05.
Results
Thrombin Regulates Gene Expression in VSMC via ROS Generation by NAD(P)H Oxidase
We examined thrombin-induced NAD(P)H oxidase–mediated gene expression at transition from G1-to-S phase of the cell cycle. Maximum increase in phospho-Cdk2 (Thr160) level was observed in VSMC at 8 hours after thrombin treatment (Figure IA in the online data supplement), consistent with this time point representing late G1-to-S phase transition.13 On this basis, we performed the gene expression profiling at 8 hours after thrombin treatment. To identify genes regulated specifically through NAD(P)H oxidase, we compared expression profiles of wild-type and gp91phox–/– VSMC (both posses functional NAD[P]H oxidase)14 to that of p47phox–/– VSMC using cDNA microarray analysis.
We first identified genes with changes in expression similar in all 3 cell types. Expression profiles for 133 genes that passed filtering of raw microarray data were analyzed using 1-class response SAM.15 With 2000 permutations and value at 0.225 corresponding to a false discovery rate (FDR) of 3.04%, 76 genes were identified as potentially significant. Of these, 61 genes were upregulated and 15 were downregulated by thrombin. Thirty-two genes encoded signal transduction proteins; 13 encoded metabolic pathway proteins; another 17 encoded proteins involved in transcriptional regulation; 9 encoded extracellular matrix and cytoskeletal proteins; and 5 genes encoded electron and ion exchange proteins (supplemental Table I).
Then we compared the differences in gene expression profiles of cell types with functional and impaired NAD(P)H oxidase. Microarray data analyzed with multiclass SAM (2000 permutations and value of 0.245 corresponding to a FDR of 5.1%) identified 28 genes as most significantly differentially regulated (supplemental Figure IB) in response to thrombin treatment. Real-time RT-PCR analysis confirmed the transcriptional pattern of 22 of 28 genes identified using SAM (78.5%). Among these 22 genes, 11 were upregulated and 11 were downregulated. These genes fall into various functional groups: 5 genes encoded proteins involved in transcriptional regulation; 5 encoded signal transduction proteins; 5 encoded metabolic pathway proteins; 1 gene encoded electron and ion transport proteins; and another 6 encoded extracellular matrix and cytoskeletal proteins (supplemental Table II). Expression of these genes was significantly changed in wild-type and gp91phox–/– VSMC (P<0.05) but not in p47phox–/– cells. Of the 3 cell types tested, changes in gene expression were greater in gp91phox–/– VSMC. Previously, we observed higher cell growth in gp91phox–/– VSMC compared with wild-type and p47phox–/– VSMC in normal growth conditions.4 It is possible that this accelerated growth phenotype reflects the greater response in gene expression of gp91phox–/– VSMC. Only 5 of the 22 genes, inhibitor of DNA binding 3 (Id3),16 vascular endothelial growth factor receptor-1 (VEGFR-1),17 aldehyde oxidase 1 (Aox1),18 aldehyde oxidase structural homolog 2 (Aoh1),19 and tissue inhibitor of metalloproteinase 3 (Timp-3)19 were previously shown to be redox sensitive. Redox upregulated genes include VEGFR-1, Timp-3, and syndecan-1 and downregulated genes include natriuretic peptide receptor 3 (Npr3). The expression of 12 of the 22 genes, Klf15, Igbp1, Ak4, Adamts5, Ech1, Serp1, Sec61a2, Aox1, Aoh1, Fxyd5, Rai14, and Serpinh1 in VSMC has not been reported before. We chose to further analyze a subset of these 22 genes.
To identify genes that play critical roles in NAD(P)H oxidase–mediated phenotypic and functional features of VSMC, we focused our studies on thrombin- and NAD(P)H oxidase–regulated genes that had similar basal expression levels in p47phox–/– and wild-type VSMC. CD44 expression increased by 3.97-fold (P<0.05) in wild-type VSMC in response to thrombin treatment but did not change significantly in p47phox–/– cells. BMP4, Id1, and Id3 were downregulated by 2.74-, 1.32-, and 1.69-fold (P<0.05), respectively, in wild-type cells but were unchanged in p47phox–/– VSMC following thrombin treatment (Figure 1A).
We then performed Western blot analysis to determine whether the changes in mRNA expression reflected changes in protein expression. CD44 expression was similar in wild-type, gp91phox–/–, and p47phox–/– VSMC (Figure 1B, top). Thrombin treatment induced an increase in CD44 expression in wild-type and gp91phox–/– cells but had no such effect in p47phox–/– VSMC. Basal BMP4 protein levels were similar in all cell types (Figure 1B, top middle). Thrombin induced a significant decrease in this protein levels in wild-type and gp91phox–/– cells, whereas BMP4 levels remain unchanged in p47phox–/– VSMC treated with thrombin. Id1 (Figure 1B, bottom middle) and Id3 (Figure 1B, bottom) protein levels decreased only in wild-type and gp91phox–/– cells in response to thrombin treatment. Thus, the changes in protein expression were similar to those observed by cDNA microarray and real-time RT-PCR.
Confirmation of Redox-Sensitive Regulation of Gene Expression in VSMC
To confirm that the regulation of thrombin-mediated VSMC genes was indeed dependent on redox changes, we performed a number of complementary experiments. First, 2,3-dimethoxy-1,4-naphthoquinone (DMNQ), a redox cycling quinone, was used to simulate NAD(P)H oxidase–mediated transcriptional changes. VSMC were treated with 1 μmol/L DMNQ for 8 hours and gene expression was analyzed by real-time RT-PCR. Changes in expression of CD44, BMP4, Id1 and Id3 in wild-type cells in response to DMNQ treatment were similar to those seen following thrombin treatment. CD44 gene expression was increased by 3.73-fold (P<0.05), whereas expression of BMP4, Id1 and Id3 genes was decreased by 1.93-, 2.34-, 2.12-fold, respectively (P<0.05) (Figure 2A). The regulation of CD44, BMP4, Id1, and Id3 genes in p47phox–/– VSMC treated with DMNQ was similar to that in wild-type cells treated with this quinone, demonstrating that these cells have not lost their capacity to respond to superoxide and confirming that thrombin-mediated superoxide generation does not occur in the cells which lack NAD(P)H oxidase activity. Specifically, CD44 expression was enhanced by 3.1-fold, whereas the expression of BMP4, Id1, and Id3 were decreased by 2.46-, 1.51-, and 1.43-fold, respectively (P<0.05). Our results also indicate that DMNQ generates ROS in VSMC in NAD(P)H oxidase–independent manner.
To corroborate the importance of NAD(P)H oxidase–dependent regulation of gene expression in VSMC treated with thrombin, we pretreated cells with diphenyleneiodonium chloride (DPI), an inhibitor of flavin-containing enzymes including NAD(P)H oxidase, or N-acetyl-L-cysteine (NAC), a ROS scavenger. Wild-type cells pretreated with either NAC or DPI and then treated with thrombin showed little or no changes in expression of CD44, BMP4, Id1, and Id3 genes, similar to that in p47phox–/– cells treated with thrombin. CD44 levels were significantly lower in wild-type VSMC pretreated with NAC or DPI and then treated with thrombin compared with the levels in cells treated with thrombin alone (P<0.001) (Figure 2B, top left). Similarly, BMP4 gene expression levels in wild-type VSMC pretreated with either NAC or DPI and then treated with thrombin were significantly different than the levels in wild-type cells treated with thrombin alone (P<0.001) (Figure 2B, top right). Pretreatment with NAC or DPI before stimulation with thrombin significantly attenuated thrombin-induced decrease in the expression of Id1 (Figure 2B, bottom left) and Id3 (Figure 2B, bottom right) genes in wild-type cells (P<0.05). Taken together, these data confirm NAD(P)H oxidase–derived ROS-dependent regulation of CD44, BMP4, Id1, and Id3 gene expression by thrombin.
Finally, we performed gain-of-function experiments to confirm that replacement of the p47phox restores thrombin-responsive expression of CD44, BMP4, Id1, and Id3 genes in p47phox–/– cells. Western blot analysis showed significantly increased p47phox expression in cells transfected with adenoviral vector containing human p47phox cDNA as compared with control cells (Figure 3A). On thrombin treatment, DHE and Amplex Red fluorescence showed increased ROS generation in transfected p47phox–/– VSMC compared with nontransfected p47phox–/– cells (Figure 3B and supplemental Figure II). Cells transfected with a control vector, Ad-GFP, were not different from nontransfected p47phox–/– cells in ROS generation.
Gene expression changes in p47phox transfected p47phox–/– VSMC were assessed by real-time RT-PCR. There were no significant changes in gene expression after thrombin treatment in nontransfected or p47phox–/– VSMC transfected with Ad-GFP. In contrast, in p47phox–/– cells transfected with Ad-p47phox, changes in CD44, BMP4, Id1, and Id3 gene expression on thrombin treatment (Figure 3C) were similar to those in wild-type cells treated with thrombin. These data reiterate ROS-sensitive regulation of CD44, BMP4, Id1, and Id3 gene expression in VSMC by thrombin and emphasize the role of NAD(P)H oxidase in these events.
Role of BMP4 in Id1 and Id3 Gene Expression in VSMC
Id1 and Id3 are known transcriptional targets of BMP4.20 For this reason, we questioned whether the observed effect of ROS on Id1 and Id3 gene expression was due solely to an effect on BMP4 or whether ROS directly activated Id1 and Id3 gene expression. Treatment with exogenous BMP4 resulted in a significant increase in Id1 and Id3 gene expression at 1 hour followed by a decrease in both wild-type and p47phox–/– VSMC (Figure 4). The most straightforward interpretation of this finding is that thrombin and NAD(P)H oxidase act only on BMP4 gene regulation, and this results in consequent down regulation of Id1 and Id3. This interpretation is consistent with our overall gene expression results and the notion that NAD(P)H oxidase activation induces specific signaling pathways. Surprisingly, Id1 and Id3 gene expression in response to BMP4 was higher in p47phox–/– cells than in wild-type VSMC. The reasons for this discrepancy are not clear. We also observed that BMP4 significantly suppressed thrombin-induced [3H]-thymidine incorporation in wild-type cells (supplemental Figure III). This suggests that higher BMP4 levels in p47phox–/– VSMC treated with thrombin may be 1 of the mechanisms for decreased p47phox–/– VSMC proliferation.4
NAD(P)H Oxidase Regulates Gene Expression in Arterial Wall and Is Necessary for Neointimal Hyperplasia Following Arterial Injury
We next examined whether NAD(P)H oxidase–dependent changes in expression of these genes occurred in arterial VSMC in vivo. The early stages of atherosclerotic lesion formation through plaque rupture and arterial occlusion all depend on VSMC proliferation and migration. Transluminal wire injury of the femoral artery results in endothelial denudation with rapid onset of medial SMC apoptosis, followed by SMC proliferation and migration leading to development of reproducible neointimal hyperplasia.12 In addition, continued and marked thrombin generation was observed following arterial injury,21 suggesting that arterial injury is a reasonable model for examining the role of thrombin in NAD(P)H oxidase–mediated gene expression and neointima formation. Evans blue staining depicted consistent denudation of the endothelium in injured arteries (Figure 5A). Four weeks after injury, statistically significant increase in neointimal hyperplasia (P<0.001) was observed in wild-type mice compared with sham-operated mice (Figure 5B through 5D). There was little or no increase in neointimal hyperplasia in p47phox–/– mice, which is consistent with decreased proliferation of p47phox–/– VSMC compared with wild-type VSMC.4
To determine the molecular basis for VSMC growth and migration in vivo, we compared the CD44, BMP4, and Id protein expression in the femoral artery in wild-type and p47phox–/– mice. CD44 expression was not observed in uninjured arteries of wild-type and p47phox–/– mice, whereas expression of this protein increased substantially in medial and neointimal SMC of wild-type mice 14 days postinjury, consistent with previous findings.22 In contrast, CD44 protein expression was not observed in p47phox–/– mice following injury (Figure 6A). Immunoreactive BMP4 was not seen in sham-operated arteries of either wild-type or p47phox–/– mice. BMP4 expression was also not present in wild-type femoral arteries but markedly increased in the medial VSMC of p47phox–/– mice 14 days after injury (Figure 6B). Id1 and Id3 expression was minimal in uninjured arteries of wild-type and p47phox–/– mice and also in injured arteries from wild-type mice. However, similar to BMP4, expression of Id1 and Id3 was increased following injury in medial VSMC of p47phox–/– mice (Figure 6C and 6D). Changes in protein expression in VSMC were ascertained by smooth muscle -actin staining (Figure 6E). The decrease in CD44 and increase in BMP4, Id1, and Id3 expression in the injured artery of p47phox–/– mice suggest that these proteins might play an important role in restenosis by affecting VSMC proliferation. Further, these results strongly support NAD(P)H oxidase and ROS as important in regulating the expression of these proteins in VSMC in response to arterial injury in a manner similar to that observed in cultured VSMC. Of note, prior studies implicating the NAD(P)H oxidase in the response to arterial injury focused on ROS production in adventitial fibroblasts and endothelial cells.23
ROS Influence Development of Atherosclerotic Lesions Through Regulation of CD44 Expression
ApoE–/–/p47phox–/– mice develop significantly fewer atherosclerotic lesions than apoE–/– littermates.4 Similarly, CD44–/– mice have 50% to 70% less atherosclerotic lesions compared with wild-type mice.24 Following our observation that VSMC CD44 expression was decreased in p47phox–/– mice following injury, we examined expression of CD44 protein in atherosclerotic lesions of apoE–/– and apoE–/–/p47phox–/– animals fed a Western diet.
Immunostaining of representative sections of proximal aorta revealed increased expression of immunoreactive CD44 in lesions of apoE–/– mice compared with apoE–/–/ p47phox–/– mice. CD44 was present in both neointimal and medial VSMC, as well as in macrophages present in lesions. The less intense CD44 staining in apoE–/–/p47phox–/– mice was limited to macrophages; little or no expression of CD44 was observed in arterial VSMC (Figure 7A). Finally, we measured the total area of atherosclerotic lesions in the aortic root. The oil red O–positive area was significantly less in apoE–/–/p47phox–/– mice than in apoE–/– mice (P<0.01) (Figure 7A and 7B). This extends our prior conclusion that atherosclerosis is reduced by deletion of p47phox,4 and, in addition, the diet used in the present study allows to observe differences in lesions in the proximal aortas of apoE–/– and apoE–/–/p47phox–/– mice. The above data further solidify the relationship between NAD(P)H oxidase activity, CD44 expression, and atherosclerosis.
Discussion
The present study was initiated to determine whether ROS-stimulated signaling pathways in VSMC are also activated in the vessel wall during restenosis and atherogenesis. Although targeted inhibition of an intracellular signaling pathway has been successful for the treatment of some malignancies,25 this approach has yet to show the same promise for the treatment or prevention of atherosclerosis. Because many agonists activate VSMC, it is not surprising that inhibition of a single growth factor or cytokine signaling is ineffective in arresting VSMC growth and migration.
The physiological effects of many VSMC agonists require NAD(P)H oxidase activation,1,2 which suggests that analysis of this oxidase-dependent signals might identify potential therapeutic targets for inhibiting VSMC growth and migration. VSMC play important role throughout the atherosclerotic process; therefore, we reasoned that not only would these NAD(P)H oxidase–dependent agonists affect tyrosine/serine phosphorylation but also gene expression of key signal transduction proteins. Based on this notion, the present study was designed to broadly characterize agonist-mediated NAD(P)H oxidase–dependent gene expression in VSMC to understand convergent signaling pathways of importance in atherogenesis. Thrombin was chosen as a model agonist for in vitro studies because the physiologic responses of VSMC to thrombin (ie, contraction, mitogenesis, and migration) all foster atherosclerotic lesion formation and because many signals activated by thrombin require activation of the NAD(P)H oxidase.3
Because little is known about the thrombin-induced late-response genes that likely sustain atherosclerotic lesion formation, we focused on changes in gene expression after 8 hours of thrombin treatment in vitro and on mid-stages of vascular lesion formation in vivo. Microarray analysis of 3 VSMC cell types with functional and nonfunctional NAD(P)H oxidase suggests that 76 genes are significantly and similarly modulated by thrombin. Thrombin-induced NAD(P)H oxidase regulated expression of another 22 genes that belong to various functional groups at the G1/S transition point (supplemental Table II). These data imply that thrombin induces transcriptional networks in VSMC in both NAD(P)H oxidase–dependent and –independent manner. Although we have not investigated the functional relevance of the latter on VSMC phenotype, our results suggest that at least some of the thrombin-induced NAD(P)H oxidase–dependent genes are involved in cell cycle progression. We have shown previously that p47phox–/– VSMC exhibit decreased proliferative response to thrombin treatment compared with wild-type VSMC.4
The 22 NAD(P)H oxidase–dependent genes identified in vitro represented several signaling pathways of known importance for growth factors and cytokines in vascular and nonvascular cells.24,26 Some of these NAD(P)H oxidase–dependent gene products have been show to play an important role in vascular pathophysiology. For example, VEGFR-1 expression was upregulated in injured arteries27 and its enhanced expression in tumor cells was mediated by NAD(P)H oxidase.17 Similarly, Timp-3 was upregulated by ROS20 and also in atherosclerotic tissue and known to influence plaque stability.28 Similar to our observation, Npr3 is downregulated by other inducers of ROS production.29 Only 2 of the 22 NAD(P)H oxidase–dependent genes regulated by thrombin are transcription factors, which indicates temporal regulation of thrombin-mediated gene transcription and interestingly both the genes are downregulated in wild-type cells. To confirm that these genes were indeed regulated in a redox-dependent manner, we characterized expression of a subset of these genes in wild-type and p47phox–/– VSMC treated with DMNQ, an inhibitor of ROS generation, a ROS scavenger and in gain-of-function experiments. These experiments confirmed the importance of ROS generation and the central role of NAD(P)H oxidase activity in the expression of a subset of 22 genes.
CD44, among the thrombin-induced genes in wild-type VSMC, mediates VSMC proliferation induced by hyaluronic acid22 and is expressed in both inflammatory and vascular cells.30 However, our data show for the first time that CD44 expression in mouse VSMC is mediated by NAD(P)H oxidase activation. BMP4, another gene of interest in thrombin-regulated subset of genes, is a member of the transforming growth factor- super family. In concordance with our data, Jeffery et al31 reported inhibition of mitogenesis in human fibroblasts by BMP4 and this was mediated by the upregulation of cell cycle inhibitor p21Cip1. Interestingly, BMP4 stimulates ROS production in endothelial cells in NAD(P)H oxidase–dependent manner.32 The other genes of interest in thrombin-mediated subset of genes are Id1 and Id3. Id class of helix-loop-helix proteins have been linked to cell cycle control,33 and it is suggested that they promote cell proliferation by acting as dominant negative transcription factors for p21Cip1.34 In contrast to our data, upregulation of Id3 expression by angiotensin II leads to VSMC proliferation and this was mediated by ROS production.16 This apparent contradiction could be attributable to temporal regulation of Id proteins in response to agonist treatment. In fact, Forrest et al35 have reported that Id3 expression was upregulated at 1 hour after serum stimulation but declined by 4 hours.
We then further studied the expression of this subset of genes (CD44, BMP4, Id1, and Id3) in vivo. Immunoreactive CD44 levels were higher in the more prominent atherosclerotic lesions found in apoE–/– mice compared with the less prominent lesions found in apoE–/–/p47phox–/– mice (Figure 7). This result is corroborated by the observations that CD44–/– mice have less atherosclerotic lesions than wild-type mice24 and CD44 expression is higher in human atheromatous plaques.36 Neointimal hyperplasia was also significantly greater in wild-type mice than in p47phox–/– mice (Figure 5) and was accompanied by increased levels of immunoreactive CD44 and reduced levels of BMP4 and Id1 and Id3 (Figure 6). In contrast to our data, increased expression of Id3 was reported in injured carotid arteries of rat which correlated with lesion development.37 These contradictions warrant further careful investigation of BMP4-Id1/Id3 pathway in restenosis and atherosclerosis. Because BMP4 has been reported to regulate Id expression,20 we treated VSMC with exogenous BMP4, confirming upregulation of Id genes (Figure 4).
In summary, we have shown that agonist-activated NAD(P)H oxidase mediates expression of a number of VSMC genes which are important in lesion formation. In addition, agonist-induced ROS production and proliferation of VSMC from wild-type mice correlates with injury-induced neointimal hyperplasia and atherosclerotic lesion formation in these mice. A group of genes implicated in VSMC function—CD44, BMP4, Id1, and Id3—are regulated in vascular lesion formation in 2 in vivo models in exactly the manner predicted from studies on cultured cells and directly correlated to the lesion size. These findings support further investigation of redox-regulated gene expression in atherosclerosis and consideration of therapies that modulate specific, convergent intracellular signals, including the CD44 and BMP4/Id signaling pathways for the treatment of restenosis and atherosclerosis.
Acknowledgments
This work was supported by NIH grant HL57352 (to M.S.R.).
Footnotes
Original received October 20, 2005; revision received February 21, 2006; accepted March 29, 2006.
References
Griendling KK, Sorescu D, Ushio-Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494–501.
Brandes RP, Kreuzer J. Vascular NADPH oxidases: molecular mechanisms of activation. Cardiovasc Res. 2005; 65: 16–27. [Order article via Infotrieve]
Patterson C, Ruef J, Madamanchi NR, Barry-Lane P, Hu Z, Horaist C, Ballinger CA, Brasier AR, Bode C, Runge MS. Stimulation of a vascular smooth muscle cell NAD(P)H oxidase by thrombin. Evidence that p47(phox) may participate in forming this oxidase in vitro and in vivo. J Biol Chem. 1999; 274: 19814–19822.
Barry-Lane PA, Patterson C, van der Merwe M, Hu Z, Holland SM, Yeh ET, Runge MS. p47phox is required for atherosclerotic lesion progression in ApoE(–/–) mice. J Clin Invest. 2001; 108: 1513–1522.
Sentman ML, Brannstrom T, Westerlund S, Laukkanen MO, Yla-Herttuala S, Basu S, Marklund SL. Extracellular superoxide dismutase deficiency and atherosclerosis in mice. Arterioscler Thromb Vasc Biol. 2001; 21: 1477–1482.
Ballinger SW, Patterson C, Knight-Lozano CA, Burow DL, Conklin CA, Hu Z, Reuf J, Horaist C, Lebovitz R, Hunter GC, McIntyre K, Runge MS. Mitochondrial integrity and function in atherogenesis. Circulation. 2002; 106: 544–549.
Napoli C, Martin-Padura I, de Nigris F, Giorgio M, Mansueto G, Somma P, Condorelli M, Sica G, De Rosa G, Pelicci P. Deletion of the p66Shc longevity gene reduces systemic and tissue oxidative stress, vascular cell apoptosis, and early atherogenesis in mice fed a high-fat diet. Proc Natl Acad Sci U S A. 2003; 100: 2112–2116.
Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular disease. Arterioscler Thromb Vasc Biol. 2005; 25: 29–38.
Barbieri SS, Eligini S, Brambilla M, Tremoli E, Colli S. Reactive oxygen species mediate cyclooxygenase-2 induction during monocyte to macrophage differentiation: critical role of NADPH oxidase. Cardiovasc Res. 2003; 60: 187–197. [Order article via Infotrieve]
Pueyo ME, Gonzalez W, Nicoletti A, Savoie F, Arnal JF, Michel JB. Angiotensin II stimulates endothelial vascular cell adhesion molecule-1 via nuclear factor-kappaB activation induced by intracellular oxidative stress. Arterioscler Thromb Vasc Biol. 2000; 20: 645–651.
Madamanchi NR, Moon SK, Hakim ZS, Clark S, Mehrizi A, Patterson C, Runge MS. Differential activation of mitogenic signaling pathways in aortic smooth muscle cells deficient in superoxide dismutase isoforms. Arterioscler Thromb Vasc Biol. 2005; 25: 950–956.
Reis ED SS, Coller BS. Mouse model of transluminal femoral artery injury. In: Simon DRC, ed. Vascular Disease and Injury: Preclinical Research. Totowa, NJ: Humana Press; 2000: 103–113.
Ohtsubo M, Theodoras AM, Schumacher J, Roberts JM, Pagano M. Human cyclin E, a nuclear protein essential for the G1-to-S phase transition. Mol Cell Biol. 1995; 15: 2612–2624.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91(phox) homologues in vascular smooth muscle cells: nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888–894. [Order article via Infotrieve]
Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A. 2001; 98: 5116–5121.
Mueller C, Baudler S, Welzel H, Bohm M, Nickenig G. Identification of a novel redox-sensitive gene, Id3, which mediates angiotensin II-induced cell growth. Circulation. 2002; 105: 2423–2428.
Arbiser JL, Petros J, Klafter R, Govindajaran B, McLaughlin ER, Brown LF, Cohen C, Moses M, Kilroy S, Arnold RS, Lambeth JD. Reactive oxygen generated by Nox1 triggers the angiogenic switch. Proc Natl Acad Sci U S A. 2002; 99: 715–720.
Rivera SP, Choi HH, Chapman B, Whitekus MJ, Terao M, Garattini E, Hankinson O. Identification of aldehyde oxidase 1 and aldehyde oxidase homologue 1 as dioxin-inducible genes. Toxicology. 2005; 207: 401–409. [Order article via Infotrieve]
Li WQ, Qureshi HY, Liacini A, Dehnade F, Zafarullah M. Transforming growth factor beta1 induction of tissue inhibitor of metalloproteinases 3 in articular chondrocytes is mediated by reactive oxygen species. Free Radic Biol Med. 2004; 37: 196–207. [Order article via Infotrieve]
Korchynskyi O, ten Dijke P. Identification and functional characterization of distinct critically important bone morphogenetic protein-specific response elements in the Id1 promoter. J Biol Chem. 2002; 277: 4883–4891.
Walters TK, Gorog DA, Wood RF. Thrombin generation following arterial injury is a critical initiating event in the pathogenesis of the proliferative stages of the atherosclerotic process. J Vasc Res. 1994; 31: 173–177. [Order article via Infotrieve]
Jain M, He Q, Lee WS, Kashiki S, Foster LC, Tsai JC, Lee ME, Haber E. Role of CD44 in the reaction of vascular smooth muscle cells to arterial wall injury. J Clin Invest. 1996; 97: 596–603.
Chen Z, Keaney JF Jr, Schulz E, Levison B, Shan L, Sakuma M, Zhang X, Shi C, Hazen SL, Simon DI. Decreased neointimal formation in Nox2-deficient mice reveals a direct role for NADPH oxidase in the response to arterial injury. Proc Natl Acad Sci U S A. 2004; 101: 13014–13019.
Cuff CA, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R, Yeh C, Secreto A, Assoian RK, Rader DJ, Pure E. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest. 2001; 108: 1031–1040.
Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia. Nat Rev Cancer. 2005; 5: 172–183. [Order article via Infotrieve]
Forrest ST, Barringhaus KG, Perlegas D, Hammarskjold ML, McNamara CA. Intron retention generates a novel Id3 isoform that inhibits vascular lesion formation. J Biol Chem. 2004; 279: 32897–32903.
Zhao Q, Egashira K, Hiasa K, Ishibashi M, Inoue S, Ohtani K, Tan C, Shibuya M, Takeshita A, Sunagawa K. Essential role of vascular endothelial growth factor and Flt-1 signals in neointimal formation after periadventitial injury. Arterioscler Thromb Vasc Biol. 2004; 24: 2284–2289.
Fabunmi RP, Sukhova GK, Sugiyama S, Libby P. Expression of tissue inhibitor of metalloproteinases-3 in human atheroma and regulation in lesion-associated cells: a potential protective mechanism in plaque stability. Circ Res. 1998; 83: 270–278.
Sun JZ, Oparil S, Lucchesi P, Thompson JA, Chen YF. Tyrosine kinase receptor activation inhibits NPR-C in lung arterial smooth muscle cells. Am J Physiol Lung Cell Mol Physiol. 2001; 281: L155–L163.
Lazaar AL, Albelda SM, Pilewski JM, Brennan B, Pure E, Panettieri RA Jr. T lymphocytes adhere to airway smooth muscle cells via integrins and CD44 and induce smooth muscle cell DNA synthesis. J Exp Med. 1994; 180: 807–816.
Jeffery TK, Upton PD, Trembath RC, Morrell NW. BMP4 inhibits proliferation and promotes myocyte differentiation of lung fibroblasts via Smad1 and JNK pathways. Am J Physiol Lung Cell Mol Physiol. 2005; 288: L370–L378.
Sorescu GP, Song H, Tressel SL, Hwang J, Dikalov S, Smith DA, Boyd NL, Platt MO, Lassegue B, Griendling KK, Jo H. Bone morphogenic protein 4 produced in endothelial cells by oscillatory shear stress induces monocyte adhesion by stimulating reactive oxygen species production from a nox1-based NADPH oxidase. Circ Res. 2004; 95: 773–779.
Barone MV, Pepperkok R, Peverali FA, Philipson L. Id proteins control growth induction in mammalian cells. Proc Natl Acad Sci U S A. 1994; 91: 4985–4988.
Prabhu S, Ignatova A, Park ST, Sun XH. Regulation of the expression of cyclin-dependent kinase inhibitor p21 by E2A and Id proteins. Mol Cell Biol. 1997; 17: 5888–5896.
Forrest ST, Taylor AM, Sarembock IJ, Perlegas D, McNamara CA. Phosphorylation regulates Id3 function in vascular smooth muscle cells. Circ Res. 2004; 95: 557–559.
Krettek A, Sukhova GK, Schonbeck U, Libby P. Enhanced expression of CD44 variants in human atheroma and abdominal aortic aneurysm: possible role for a feedback loop in endothelial cells. Am J Pathol. 2004; 165: 1571–1581.
Matsumura ME, Li F, Berthoux L, Wei B, Lobe DR, Jeon C, Hammarskjold ML, McNamara CA. Vascular injury induces posttranscriptional regulation of the Id3 gene: cloning of a novel Id3 isoform expressed during vascular lesion formation in rat and human atherosclerosis. Arterioscler Thromb Vasc Biol. 2001; 21: 752–758.
您现在查看是摘要介绍页，详见ORG附件。