Plaque rupture is the precipitating event in the majority of

(1)

2449

P

laque rupture is the precipitating event in the majority of clinical sequelae associated with unstable angina, acute myocardial infarction, cerebral embolism, and atherothrom- botic stroke. The fibrous cap is predominantly composed of vascular smooth muscle cells (VSMCs) and their associated extracellular matrix proteins, such as collagen, elastin, and proteoglycans, which together maintain a strong and throm- boprotective fibrous cap.¹ The fibrous cap is crucial for the separation of cholesterol-rich, procoagulant debris from the circulation.² Stable lesions are those with a thick fibrous cap and a high ratio of VSMCs to macrophages and exhibit reduced susceptibility to rupture.^3,4 Because VSMCs are the primary source of the extracellular matrix within the plaque and the fibrous cap, apoptosis of VSMCs within the fibrous cap is considered a major contributor to increased plaque rupture via loss of cells and loss of extracellular matrix synthesis.

The high level of intimal macrophages, inflammatory medi- ators, and residual oxidized lipid causes an imbalance of reactive oxygen species and leads to oxidative stress within the atherosclerotic plaque. Oxidative stress is a well-characterized

inducer of VSMC apoptosis both in vivo and in vitro.^5,6 Apoptotic VSMCs have been identified throughout the atherosclerotic plaque,⁷ particularly within the fibrous cap and underlying media.^8–10 Importantly, increased VSMC apoptosis within the plaque is associated with plaque instability.¹¹ Mouse models of atherosclerosis have directly demonstrated that the induction of VSMC apoptosis results in the development of vulnerable plaques.^12–15 In addition to participating in fibrous cap thinning, VSMC apoptosis in vitro results in various out- comes, which are likely to contribute to plaque progression and instability, including production of thrombin,¹⁶ release of cytokines,¹⁷ and promotion of vascular calcification.¹⁸

The vertebrate family of Wingless/Wnt proteins and their associated complex signaling pathways have been implicated in the cellular regulation of embryogenesis but also regulate various disease pathologies through cell proliferation, migra- tion, and survival. The Wnt pathways have been reviewed comprehensively in the literature by several groups.^19–22 Wnt- induced cell survival signaling almost exclusively involves the canonical Wnt pathway, which is entirely dependent on

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.114.303922

Objective—Apoptosis of vascular smooth muscle cells (VSMCs) contributes to thinning and rupture of the atherosclerotic plaque fibrous cap and is thereby associated with myocardial infarction. Wnt protein activation of β-catenin regulates numerous genes that are associated with cell survival. We therefore investigated Wnt/β-catenin survival signaling in VSMCs and assessed the presence of this pathway in human atherosclerotic plaques at various stages of the disease process.

Approach and Results—Wnt5a induced β-catenin/T-cell factor signaling and retarded oxidative stress (H₂O₂)–induced apoptosis in mouse aortic VSMCs. Quantification of mRNA levels revealed a >4-fold (P<0.05; n=9) increase in the expression of the Wnt/β-catenin responsive gene, Wnt1-inducible secreted protein-1 (WISP-1), which was dependent on cAMP response element–binding protein and sustained in the presence of H₂O₂. Exogenous WISP-1 significantly reduced H₂O₂–induced apoptosis by 43% (P<0.05; n=3) and was shown using silencing small interfering RNA, to be important for Wnt5a-dependent survival responses to H₂O₂ (P<0.05; n=3). WISP-1 protein levels were significantly lower (≈50%) in unstable atherosclerosis compared with stable plaques (n=11 and n=14).

Conclusions—These results indicate for the first time that Wnt5a induces β-catenin survival signaling in VSMCs via WISP-1. The deficiency of the novel survival factor, WISP-1 in intimal VSMCs of unstable coronary plaques, suggests that there is altered Wnt/β-catenin/ T-cell factor signaling with progressive atherosclerosis, and restoration of WISP-1 protein might be an effective stabilization factor for vulnerable atherosclerotic plaques. (Arterioscler Thromb Vasc Biol. 2014;34:2449-2456.)

Key Words: apoptosis ◼ atherosclerosis ◼ WISP-1 protein ◼ Wnt proteins

Received on: May 2, 2014; final version accepted on: August 31, 2014.

From the School of Clinical Sciences, University of Bristol, Bristol Royal Infirmary, Bristol, United Kingdom.

The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.114.303922/-/DC1.

Correspondence to Sarah Jane George, BSc, PhD, School of Clinical Sciences, University of Bristol, Research Floor Level 7, Bristol Royal Infirmary, Upper Maudlin St, Bristol BS2 8HW, United Kingdom. E-mail [email protected]

Wnt5a-Induced Wnt1-Inducible Secreted Protein-1 Suppresses Vascular Smooth Muscle Cell Apoptosis

Induced by Oxidative Stress

Carina Mill, Bethan Alice Monk, Helen Williams, Steven John Simmonds, Jamie Yancey Jeremy, Jason Lee Johnson, Sarah Jane George

Downloaded from http://ahajournals.org by on September 1, 2021

(2)

the complex between Frizzled receptors with the low-density lipoprotein receptor–related protein 5/6 (LRP5/6) coreceptors, and the activation and nuclear translocation the adaptor protein β-catenin. Wnt activation of β-catenin facilitates T-cell factor (TCF) or lymphoid enhancer–binding factor transcriptional activation leading to TCF-responsive gene transcription and modulation of cell behavior including survival.^23,24 Deregulated Wnt signaling is associated with many diseases including cancer²²; however, its involvement in atherosclerosis remains to be fully elucidated.

It has been shown previously that Wnt5a is expressed in atherosclerotic plaques, predominantly in macrophages.^25,26 The effect of Wnt5a in atherosclerotic plaques, and its role as a survival factor for VSMCs, is however still to be determined.

In this study, we investigated the effect of exogenous Wnt5a on canonical Wnt pathway activation in VSMCs and directly assessed the effect that Wnt5a had on VSMC survival under conditions of oxidative stress. We highlighted an association of Wnt1-inducible secreted protein-1 (WISP-1) expression with atherosclerotic plaque stability and propose that it is a distinct survival factor, which has potential as a plaque stabilizing protein.

Materials and Methods

Materials and Methods are available in the online-only Supplement.

Results

Activation of Wnt/β-Catenin Signaling in VSMCs by Wnt5a

Wnt5a protein was not detectable in mouse or human VSMC lysates or conditioned media but was readily detectable in macrophage lysates (Figure IA and IB in the online-only Data Supplement). This was confirmed by quantitative polymerase chain reaction (Figure IC in the online-only Data Supplement). Moreover, treatment with Wnt5a or H₂O₂ did not induce expression of Wnt5a protein (Figure ID in the online-only Data Supplement). Together this indicates that addition of exogenous Wnt5a protein would mimic the exposure of VSMCs to macrophage-derived Wnt5a in atherosclerotic plaques. β-catenin activation after recombinant Wnt5a treatment was assessed using Western blotting for active β-catenin protein and for β-catenin protein in nuclear extracts from C57BL/6J mice aortic VSMCs and assessment of β-catenin/TCF reporter activity in TCF optimal promoter- beta- galactosidase (TOPGAL) VSMCs. Treatment of VSMCs for 10 and 30 minutes with 200 and 400 ng/mL recombinant Wnt5a protein increased the amount of active β-catenin protein (Figure 1A). The greatest increase was observed with

400 ng/mL for 30 minutes (Figure 1A). Moreover, treatment of VSMCs with 400 ng/mL recombinant Wnt5a protein significantly increased the nuclear levels of β-catenin protein at 30 minutes compared with untreated control (Figure 1A and 1B). This was corroborated by a parallel increase in β-catenin/

TCF reporter activity at 24 hours (Figure 1C). Wnt5a- induced nuclear localization of β-catenin was not affected by H₂O₂ cotreatment; however, surprisingly Wnt5a-dependent β-catenin/TCF signaling was markedly reduced by 91.8±19%

(n=3; P<0.05). Treatment with H₂O₂ alone did not affect the levels of nuclear β-catenin or β-catenin/TCF reporter activity (Figure 1A–1C). Wnt5a-depedent nuclear localization of Nonstandard Abbreviations and Acronyms

CREB cAMP response element–binding protein LRP lipoprotein receptor–related protein siRNA small interfering RNA

TCF T-cell factor

VSMC vascular smooth muscle cell WISP-1 Wnt1 inducible secreted protein-1

Figure 1. Activation of β-catenin by Wnt5a. A, Left, Representa- tive Western blot for active β-catenin and β-actin (loading control) proteins in total cell lysates from vascular smooth muscle cells (VSMCs) treated for 10 or 30 minutes with 200 or 400 ng/mL Wnt5a (n=3). Right, Representative Western blot for total β-catenin protein and p84 protein (nuclear protein loading control) in nuclear lysates from VSMCs treated for 30 minutes with 100 μmol/L H₂O₂, 400ng/

mL Wnt5a or both (n=9). B, Densitometric quantification of Western blots (n=9) for nuclear β-catenin protein (normalized by p84) shown as fold change from untreated control. *P<0.05 vs untreated con- trol, ANOVA and Student–Newman–Keuls post hoc test. C, Quanti- fication of β-catenin/T-cell factor signaling activity by measurement of β-galactosidase protein (expressed as arbitrary units [AU] and normalized to total cell lysate protein) in TOPGAL (TCF optimal promoter-beta-galactosidase) VSMCs treated with 100 μmol/L H₂O₂ and 400 ng/mL Wnt5a. *P<0.05 vs control, $P<0.05 vs Wnt5a, ANOVA, and Student–Newman–Keuls post hoc test (n=3).

(3)

β-catenin was confirmed using immunocytochemistry (Figure IIA and IIB in the online-only Data Supplement). Expression of total cellular β-catenin protein was unaffected by any of the treatments (Figure IIC and IID in the online-only Data Supplement). Together these data suggest that despite activation of β-catenin, Wnt5a does not activate TCF-dependent signaling when in the presence of oxidative stress.

Wnt5a provides resistance against H₂O₂-dependent apoptosis, which is dependent on LRP5/6 canonical receptors.

Immunocytochemistry for cleaved caspase-3 demonstrated that H₂O₂ treatment significantly increased VSMC apoptosis as expected (Figure 2A; Figure IIIA–IIID in the online-only Data Supplement). Wnt5a did not affect basal survival when given as a single agent; however, simultaneous treatment of VSMCs with Wnt5a and H₂O₂ significantly reduced apoptosis compared with H₂O₂ treatment alone (Figure 2A). This finding was unexpected after the observation that Wnt5a- induced TCF signaling was retarded by simultaneous addition of H₂O₂ (Figure 1C). These observations were confirmed by Western blotting for inactive and active (cleaved) caspase-3 and in situ end labeling (Figure IIIE and IIIF in the online- only Data Supplement, respectively). In addition, Wnt5a suppressed apoptosis of VSMCs treated with oxidized low- density lipoprotein but was unable to rescue VSMCs treated with Fas ligand (Figure IIIG and IIIH in the online-only Data Supplement).

To demonstrate that active β-catenin is a limiting factor for the TCF signaling, and therefore VSMC survival, an adenovirus construct containing a mutated β-catenin gene, which is resistant to proteasomal degradation and thus constitutively active, was overexpressed in TOPGAL VSMCs, as previously described.²⁷ Treatment of VSMCs with Wnt5a and H₂O₂ significantly reduced β-catenin/TCF signaling compared with VSMCs treated with Wnt5a alone (Figure 2B, white bars).

Overexpression of mutant β-catenin protein in the presence of Wnt5a and H₂O₂ restored TCF signaling levels to those obtained with Wnt5a treatment alone (Figure 2B, black bars).

These results indicate that active β-catenin is a limiting factor for Wnt5a-induced TCF-dependent signaling in the presence of oxidative stress. To determine the contribution of the canonical Wnt signaling pathway in Wnt5a-dependent resistance to H₂O₂-induced apoptosis small interfering RNA (siRNA), we simultaneously silenced LRP5 and LRP6. As LRP5 and LRP6 are essential for canonical Wnt-β-catenin signaling, but are not involved in noncanonical Wnt signaling pathways, silencing of these coreceptors selectively inhibits canonical signaling. Simultaneous silencing of LRP5 and LRP6 mRNA and protein using siRNA was achieved (Figure IV in the online-only Data Supplement). Dual silencing of LRP5/6 resulted in a significant loss of the apoptotic resistance provided by Wnt5a against H₂O₂ (Figure 2C), which indicates that Wnt5a signaling via the Frizzled-LRP coreceptor interaction was required for Wnt5a-induced survival. The culmination of these findings suggest that Wnt5a provides apoptotic resistance in VSMCs; however, it is independent of β-catenin/TCF signaling (as shown in Figure 1C) but is dependent on Wnt/Frizzled/LRP5/6 signaling and β-catenin activation. This suggests that binding of β-catenin

to an alternative transcription factor may be responsible for the Wnt5a-dependent survival response under conditions of oxidative stress.

Figure 2. Wnt5a retards H₂O₂-induced apoptosis in vascular smooth muscle cells (VSMCs) through canonical Wnt receptor activation. A, Percentage of cleaved caspase-3 (CC-3)–positive VSMCs after 24-hour treatment. *P<0.05 vs control, $P<0.05 vs H₂O₂, ANOVA and Student–Newman–Keuls post hoc test (n=3).

B, Quantification of β-catenin/T-cell factor (TCF) signaling activity by measurement of β-galactosidase protein (expressed as arbitrary units [AU] and normalized to total cell lysate protein) in TOPGAL (TCF optimal promoter-beta-galactosidase) VSMCs infected with adenovirus encoding mutant (active) β-catenin, or green fluorescent protein (GFP; control). Data are expressed as the mean fold change from Wnt5a and quantify β-catenin/TCF activity. *P<0.05 vs Wnt5a (GFP), **P<0.05 vs Wnt5a and H₂O₂ (GFP), ANOVA, and Student–

Newman–Keuls post hoc test (n=3). C, Percentage of CC-3–positive VSMCs after 24-hour treatment with 100 μmol/L H₂O₂ and 400 ng/mL Wnt5a in VSMCs subjected to small interfering RNA (siRNA) nucleofection AllStars negative scrambled control (white bars) or combined low-density lipoprotein receptor–related protein 5/6 (LRP5/6) siRNA (black bars), 24 hours before treatment. *P<0.05 vs control (AllStars controls), $P<0.05 vs H₂O₂ (AllStars controls),

**P<0.05 vs control (LRP5/6 siRNA), $$P<0.05 vs H₂O₂ (LRP5/6 siRNA), ANOVA, and Student–Newman–Keuls post hoc test (n=5).

(4)

WISP-1 Is Upregulated by Wnt5a and Is Important for Wnt5a-Dependent VSMC Survival Under Oxidative Stress

WISP-1 is a β-catenin–dependent survival gene. WISP-1 expression is however independent of TCF, it is in fact dependent on cAMP responsive elements within its promoter region.²⁸ Correspondingly WISP-1 mRNA (Figure 3A) was significantly enhanced with Wnt5a treatment and remained elevated with Wnt5a and H₂O₂ cotreatment (Figure 3A), this was confirmed at protein level (Figure 3B and 3C). Similarly, Wnt3a and Wnt9a significantly enhanced WISP-1 mRNA expression and suppressed H₂O₂-induced apoptosis of VSMCs (Figure V in the online-only Data Supplement). Treatment with Wnt5a significantly increased the mRNA expression of 3 β-catenin/TCF-regulated genes, which have been associated with cell survival (WISP-2, insulin-like growth factor-1, and survivin), but this was lost when VSMCs were cotreated with Wnt5a in the presence of H₂O₂ (Figure VI in the online-only Data Supplement). This highlights that the alternative Wnt5a- induced signaling pathway, which occurs under oxidative stress and results in WISP-1 upregulation, is not a widespread effect. Wnt5a induced phosphorylation of cAMP response element–binding protein (CREB), which was also sustained in the presence of H₂O₂ (Figure 3D and 3E), suggesting that concomitant activation of β-catenin and CREB may be

responsible for Wnt5a-induced WISP-1 expression. This was confirmed by the ability of overexpression of A-CREB (dominant negative CREB) to significantly suppress the induction of WISP-1 mRNA by Wnt5a in the presence and absence of oxidative stress (Figure 3F). The dominant negative function of A-CREB was illustrated by its ability to suppress Forskolin- induced upregulation of 2 CREB-responsive genes (cAMP responsive element modulator and nuclear receptor subfamily 4, group A, member 1; Figure VIIA in the online-only Data Supplement).

WISP-1 Is a Prosurvival Factor for VSMCs Under Oxidative Stress

The effect of WISP-1 on H₂O₂-induced VSMC apoptosis was directly examined using recombinant WISP-1 protein and adenoviral-mediated overexpression of WISP-1. WISP-1 recombinant protein and overexpression significantly reduced H₂O₂-induced VSMC apoptosis (Figure 4A; Figure VIIB in the online-only Data Supplement). To further examine the effect of WISP-1 on H₂O₂-induced apoptosis and demonstrate the involvement of CREB, knockdown of WISP-1 and CREB expression was performed using siRNA. Validation of mRNA and protein knockdown using quantitative polymerase chain reaction and Western blotting was achieved for each protein (Figures VIII and IX in the online-only Data Supplement),

Figure 3. Wnt1-inducible secreted pro- tein-1 (WISP-1) expression and cAMP response element–binding protein (CREB) phosphorylation are induced with Wnt5a treatment under oxidative stress.

A, WISP-1 mRNA (fold change from con- trol) normalized to 18s ribosomal RNA in vascular smooth muscle cells (VSMCs) treated for 4 hours with 100 μmol/L H₂O₂ and 400ng/mL Wnt5a. *P<0.05 vs control, ANOVA, and Dunn post hoc test (n=10). B, Representative Western blots for WISP-1 and β-actin (loading control) proteins in VSMCs treated for 24 hours (n=4). C, Densitometric quantification of Western blots for WISP-1 protein normalized to β-actin and shown as the fold change from control. *P<0.05 vs control, ANOVA, and Student–Newman–Keuls post hoc test (n=4). D, Representative Western blots for phosphorylated CREB (pCREB) and β-actin (loading control) proteins in VSMCs treated for 10 minutes (n=4). E, Densitometric quantification of Western blots for pCREB protein normalized to β-actin shown as the fold change from control. *P<0.05 vs control, ANOVA, and Student–Newman–Keuls post hoc test (n=4). F, WISP-1 mRNA (fold change from control) normalized to 18s ribosomal RNA in VSMCs treated for 4 hours with Wnt5a 42 hours after infection with replication incompetent adenoviruses to overexpress dominant negative CREB (A-CREB) or β-galactosidase (control).

*P<0.05 vs control, ANOVA, and Dunn post hoc test (n=10).

(5)

respectively. There was no significant effect of WISP-1 siRNA treatment on WISP-2 mRNA, demonstrating the specificity of the WISP-1 siRNA sequence (Figure VIIIA in the online-only Data Supplement). Wnt5a retarded H₂O₂-induced apoptosis of VSMCs treated with the control (AllStars) siRNA. However, when either WISP-1 or CREB was depleted by siRNA, no significant difference was observed in the rate of apoptosis between VSMCs treated with H₂O₂ alone or simultaneously treated with Wnt5a and H₂O₂ in the VSMCs exposed to the siRNA (Figure 4B). This demonstrates the involvement of both WISP-1 and CREB in the antiapoptotic effect of Wnt5a in VSMCs under oxidative stress.

WISP-1 Expression in Human Coronary Arteries Immunofluorescence for WISP-1 in human coronary artery samples revealed significantly higher levels of WISP-1 protein in stable coronary artery atherosclerotic plaques than unstable plaques (Figure 5). The majority of WISP-1 protein was colocated with α-smooth muscle actin–positive cells in the intima near the lumen (ie, fibrous cap; Figure 5). The ratio of WISP-1 to α-smooth muscle actin in these regions was significantly greater in stable plaques compared with unstable plaques (0.4±0.1 versus 0.1±0.04; P<0.05; n=7).

Dual immunofluorescence for cleaved caspase-3 (CC-3) and WISP-1 was performed in stable and unstable plaques (Figure 6). WISP-1 protein levels were low in areas of apoptosis in the unstable plaques, and vice versa in stable plaques, where WISP-1 protein levels were high and the number of CC-3–positive cells was low. Quantification revealed that WISP-1 protein was inversely correlated with CC-3 protein in the VSMC-rich fibrous cap (linear regression, r=−0.58, r²=0.34; P=0.0294; n=13).

Figure 4. Wnt1-inducible secreted protein-1 (WISP-1) and cAMP response element–binding protein (CREB) reduced vascular smooth muscle cell (VSMC) apoptosis under oxidative stress.

The percentage of cleaved caspase-3–positive cells 24 hours after: (A) treatment with 100 μmol/L H₂O₂ alone and 100 μmol/L H₂O₂ and 2.5 μg/mL WISP-1 recombinant protein. *P<0.05 vs control, $P<0.05 vs H₂O₂ (n=3). B, Small interfering RNA (siRNA) for WISP-1 (black bars), CREB (gray bars), or control siRNA (AllStars, white bars) were introduced into VSMCs before VSMC treatment for 24 hours. *P<0.05 vs control (AllStars), $P<0.05 vs H₂O₂ (AllStars), **P<0.05 vs control (WISP-1 siRNA), $$P<0.05 vs H₂O₂ (WISP-1 siRNA), ***P<0.05 vs control (CREB siRNA),

$$$P<0.05 vs H₂O₂ (CREB siRNA), no significant difference from H₂O₂, ANOVA, and Student–Newman–Keuls post hoc test (n=3).

CC-3 indicates cleaved caspase-3.

Figure 5. Increased Wnt1-inducible secreted protein-1 (WISP- 1) expression in stable atherosclerotic plaques. Dual immunofluorescence for smooth muscle cell α-actin protein (SMA, green) and WISP-1 protein (red) in sections of a stable human atherosclerotic plaque (A and B) and an unstable atherosclerotic plaque (E and F). Merged images of SMA and WISP-1 are shown in C and G, where yellow/orange color indicates colo- cation of both proteins. Macrophages were localized in serial sections with CD68 (green; D and H). Nuclei were stained blue with 4′,6-diamidino-2-phenylindole (Dapi). As a negative control WISP-1 antibody was substituted with nonimmune immunoglob- ulin on a stable plaque (I). The scale bar applies to all panels and represents 50 μm. J, Quantification of WISP-1 protein expression in plaques expressed as a percentage of area. *P<0.05 vs stable, ANOVA, and Dunns post hoc test (n=14 stable; n=11 unstable).

(6)

Macrophage-Derived Wnt5a Affects VSMC Apoptosis

To determine whether macrophage-derived Wnt5a can suppress VSMC apoptosis, we performed in vitro coculture experiments. Wnt5a expression was significantly suppressed in macrophages (Figure XA in the online-only Data Supplement).

VSMCs were cultured in the presence of the conditioned media collected from macrophages subjected to Wnt5a silencing (or control siRNA, AllStars) for 24 hours and apoptosis quanti- fied by cleaved caspase-3. We observed significantly higher apoptotic rates in VSMCs cultured in conditioned media collected from macrophages subjected to Wnt5a silencing compared with control siRNAs (Figure XB in the online-only Data Supplement). Interestingly, the rate of apoptosis was similar to that seen in VSMCs treated with H₂O₂. Similar results were observed when WISP-1 was silenced in macrophages (Figure XC in the online-only Data Supplement).

Discussion

VSMC apoptosis is a major contributor to atherosclerotic plaque instability and as a consequence is considered to

contribute to the occurrence of myocardial infarction. The identification of antiapoptotic molecules may lead to promis- ing therapeutic leads for reducing plaque instability. Extensive research demonstrates that canonical β-catenin–dependent Wnt signaling promotes survival of numerous cell types, including endothelial cells,²⁹ osteoblasts,³⁰ and cancer cells.^31,32 In this study, we examined the hypothesis that Wnt5a, which has been shown to be expressed in atherosclerotic tissue,^25,26 can induce VSMC survival in vitro via β-catenin signaling, and whether this is affected by oxidative stress.

The effect of recombinant Wnt5a protein was assessed in primary VSMCs and was compared against the effect of cotreatment with H₂O₂, a relevant and effective mimetic for vascular oxidative stress.³³ Wnt5a is expressed by macrophages but not VSMCs in vitro and this is sufficient to provide a prosurvival signal for VSMCs. Wnt5a induced β-catenin activation in VSMCs both alone and in the presence of oxidative stress.

Support for activation of canonical signaling was provided by our observation that Wnt5a induced the TCF reporter activity in VSMCs from TOPGAL mice, which corroborates previous findings in mesangial cells.³⁴ This assay, however, revealed the true complexity of Wnt signaling under oxidative stress in VSMCs, because despite sustained β-catenin activation and nuclear translocation with H₂O₂ cotreatment, Wnt5a-induced TCF reporter signaling was reduced to control levels in the presence of oxidative stress. Furthermore, despite a reduction in overall TCF reporter signaling, which is almost exclusively required for Wnt-induced survival signaling,29–32,35–38 Wnt5a significantly attenuated oxidative stress–induced apoptosis stimulated by H₂O₂ and oxidized low-density lipoprotein (but not apoptosis induced by Fas ligand). This survival effect was also shown to be dependent on the canonical coreceptors LRP5/6, which provides further support for the importance of β-catenin in this mechanism. Overall these findings suggest that an alternative Wnt5a signaling pathway involving β-catenin is stimulated in VSMCs under oxidative stress. Oxidative stress has been shown to manipulate Wnt/β-catenin survival signaling in other cell types including HEK293 (human embryonic kidney 293) cells³⁹ and osteoblasts.^40,41

A previous study by Xu et al²⁸ revealed that despite the presence of TCF/lymphoid enhancer–binding factor binding sites, in the Wnt/β-catenin target gene WISP-1, a single cAMP response element site has dominant functionality. Direct evidence for the role of WISP-1–dependent survival has been provided from studies using rat fibroblasts, lung carcinoma cells neuronal cells, and cardiomyocytes; all of which demonstrated the survival effect of WISP-1.^28,42–45 We therefore investigated the hypothesis that a β-catenin–dependent but TCF-independent WISP-1 is involved in Wnt5a survival under oxidative stress. Indeed, the treatment of VSMCs with Wnt5a and cotreatment with Wnt5a and H₂O₂ resulted in significant increases in CREB phosphorylation and WISP-1 mRNA and protein expression. We compared the expression of WISP-1 with other Wnt/β-catenin responsive survival genes and revealed that in contrast the Wnt5a-dependent mRNA expression of WISP-2, insulin-like growth factor-1 (IGF-1), and survivin was all significantly attenuated under simultaneous H₂O₂ treatment. This endorsed the hypothesis for an alternative Figure 6. Decreased Wnt1-inducible secreted protein-1 (WISP-1)

expression in unstable atherosclerotic plaque in areas of apoptosis. Dual immunofluorescence for WISP-1 (green) and cleaved caspase-3 (CC-3; red) proteins in a stable atherosclerotic plaque (A and B) and an unstable atherosclerotic plaque (E and F). Nuclei are stained blue with 4′,6-diamidino-2-phenylindole (Dapi; C and G) and the merged images are shown in D and H. The scale bar applies to all panels and represents 100 μm (n=7 per group).

(7)

Wnt5a survival signaling pathway involving CREB induction of WISP-1 under oxidative stress and demonstrated that WISP- 2, IGF-1, and survivin mRNA induction is not responsible for enhanced survival as a result of Wnt5a treatment. In this study, we revealed that recombinant WISP-1 can induce survival of VSMCs under oxidative stress, supporting a previous finding that WISP-1 treatment of human saphenous vein VSMCs results in increased mitochondrial Bcl-2 (B cell lymphoma-2 protein) production.⁴⁶ In this study using siRNA silencing of WISP-1, we have provided direct evidence that WISP-1 is required for Wnt5a-dependent apoptotic resistance to oxidative stress–induced apoptosis. Collectively these findings support the hypothesis that an alternative β-catenin–dependent mechanism is induced by Wnt5a under oxidative stress, which involves the upregulation of WISP-1 via CREB-dependent pathways. This pathway is not exclusively mediated by Wnt5a as similar findings were observed with Wnt3a and Wnt9a.

The final aim of this study was to assess the expression of WISP-1 protein in vivo. WISP-1 expression has been scruti- nized in several cell types and different pathologies whereby it alters cellular homeostasis through several autocrine^28,47 and paracrine effects,^28,42,46 and thus represents an attractive therapeutic target for medical application. At present, however, little is known of its role in VSMC homeostasis and atherosclerosis. We investigated the expression of WISP-1 in human coronary artery atherosclerotic plaques, which is previously unreported and revealed that crucially WISP-1 protein expression is enhanced in stable plaques compared with those with unstable phenotypes. Moreover, WISP-1 protein/VSMC ratio was higher in the fibrous cap region of stable plaques compared with unstable examples, suggesting that it might be an effective plaque stabilizing factor. Interestingly, WISP-1 and CC-3 in plaques were inversely correlated with the observation of apoptotic cells (cleaved caspase-3 positive) in areas of low WISP-1 protein. Although in vitro evidence suggests that both autocrine and paracrine (macrophage derived) WISP-1 retard VSMC apoptosis, little was detected in macrophages within atherosclerotic plaques.

The mechanism of action of WISP-1 is not well understood at the receptor level, although we do have an insight into the cellular activities of other members of the CCN-family, which operate through the activation of integrin receptors^48–50; indeed, coimmunoprecipitation experiments have demonstrated the functional interaction of WISP-1 with α₅β₁ in human osteo- genic bone marrow cells,⁵¹ providing supporting evidence for integrin-dependent functions of WISP-1 in other cell types.

Follow-up experiments examining the expression of integrin receptors by VSMCs after WISP-1 treatment, and their interaction with WISP-1, is essential to understand the mechanism of action of WISP-1 and its consequences on VSMC survival.

In summary, these findings indicate that activation of Wnt5a signaling can suppress apoptosis of VSMCs subjected to oxidative stress via activation of β-catenin and CREB, which results in enhanced expression of the prosurvival protein WISP-1. Moreover, our data imply that WISP-1 may be an important contributing factor for fibrous cap stability.

Acknowledgments

We would like to thank Dr Steve White for assistance with the coronary artery samples.

Sources of Funding

This work was funded by the British Heart Foundation (FS/08/042/25378 and PG/11/77/29110) and supported by the National Institute for Health Reasearch Bristol Biomedical Reseach Unit in Cardiovascular Disease.

Disclosures

None.

References

1. Wight TN. The extracellular matrix and atherosclerosis. Curr Opin Lipidol. 1995;6:326–334.

2. Davies MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of throm- bosis in human atherosclerotic plaques: role of extracellular lipid, macro- phage, and smooth muscle cell content. Br Heart J. 1993;69:377–381.

3. Virmani R, Kolodgie FD, Burke AP, Farb A, Schwartz SM. Lessons from sudden coronary death: a comprehensive morphological classifica- tion scheme for atherosclerotic lesions. Arterioscler Thromb Vasc Biol.

2000;20:1262–1275.

4. Dhume AS, Soundararajan K, Hunter WJ 3rd, Agrawal DK. Comparison of vascular smooth muscle cell apoptosis and fibrous cap morphology in symptomatic and asymptomatic carotid artery disease. Ann Vasc Surg.

2003;17:1–8.

5. Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular disease. Arterioscler Thromb Vasc Biol. 2005;25:29–38.

6. Irani K. Oxidant signaling in vascular cell growth, death, and survival: a review of the roles of reactive oxygen species in smooth muscle and endo- thelial cell mitogenic and apoptotic signaling. Circ Res. 2000;87:179–183.

7. Kockx MM, De Meyer GR, Muhring J, Jacob W, Bult H, Herman AG.

Apoptosis and related proteins in different stages of human atherosclerotic plaques. Circulation. 1998;97:2307–2315.

8. Björkerud S, Björkerud B. Apoptosis is abundant in human atherosclerotic lesions, especially in inflammatory cells (macrophages and T cells), and may contribute to the accumulation of gruel and plaque instability. Am J Pathol. 1996;149:367–380.

9. Geng YJ, Libby P. Evidence for apoptosis in advanced human atheroma.

Colocalization with interleukin-1 beta-converting enzyme. Am J Pathol.

1995;147:251–266.

10. Han DK, Haudenschild CC, Hong MK, Tinkle BT, Leon MB, Liau G.

Evidence for apoptosis in human atherogenesis and in a rat vascular injury model. Am J Pathol. 1995;147:267–277.

11. Bauriedel G, Hutter R, Welsch U, Bach R, Sievert H, Lüderitz B. Role of smooth muscle cell death in advanced coronary primary lesions: implica- tions for plaque instability. Cardiovasc Res. 1999;41:480–488.

12. von der Thüsen JH, van Vlijmen BJ, Hoeben RC, Kockx MM, Havekes LM, van Berkel TJ, Biessen EA. Induction of atherosclerotic plaque rupture in apolipoprotein E-/- mice after adenovirus-mediated transfer of p53.

Circulation. 2002;105:2064–2070.

13. Zadelaar AS, von der Thüsen JH, Boesten LS, Hoeben RC, Kockx MM, Versnel MA, van Berkel TJ, Havekes LM, Biessen EA, van Vlijmen BJ. Increased vulnerability of pre-existing atherosclerosis in ApoE- deficient mice following adenovirus-mediated Fas ligand gene transfer.

Atherosclerosis. 2005;183:244–250.

14. Clarke MC, Figg N, Maguire JJ, Davenport AP, Goddard M, Littlewood TD, Bennett MR. Apoptosis of vascular smooth muscle cells induces features of plaque vulnerability in atherosclerosis. Nat Med. 2006;12:1075–1080.

15. Clarke MC, Littlewood TD, Figg N, Maguire JJ, Davenport AP, Goddard M, Bennett MR. Chronic apoptosis of vascular smooth muscle cells accel- erates atherosclerosis and promotes calcification and medial degeneration.

Circ Res. 2008;102:1529–1538.

16. Flynn PD, Byrne CD, Baglin TP, Weissberg PL, Bennett MR.

Thrombin generation by apoptotic vascular smooth muscle cells. Blood.

1997;89:4378–4384.

17. Schaub FJ, Han DK, Liles WC, Adams LD, Coats SA, Ramachandran RK, Seifert RA, Schwartz SM, Bowen-Pope DF. Fas/FADD-mediated activation of a specific program of inflammatory gene expression in vascular smooth muscle cells. Nat Med. 2000;6:790–796.

(8)

18. Proudfoot D, Skepper JN, Hegyi L, Bennett MR, Shanahan CM, Weissberg PL. Apoptosis regulates human vascular calcification in vitro:

evidence for initiation of vascular calcification by apoptotic bodies. Circ Res. 2000;87:1055–1062.

19. Kühl M, Sheldahl LC, Park M, Miller JR, Moon RT. The Wnt/Ca2+ path- way: a new vertebrate Wnt signaling pathway takes shape. Trends Genet.

2000;16:279–283.

20. Katoh M. WNT/PCP signaling pathway and human cancer. Oncol Rep 2005;14:1583–1588.

21. Katoh M. WNT signaling in stem cell biology and regenerative medicine.

Curr Drug Targets. 2008;9:565–570.

22. Chien AJ, Conrad WH, Moon RT. A Wnt survival guide: from flies to human disease. J Invest Dermatol. 2009;129:1614–1627.

23. Papkoff J. Regulation of complexed and free catenin pools by distinct mechanisms. Differential effects of Wnt-1 and v-Src. J Biol Chem.

1997;272:4536–4543.

24. Daniels DL, Weis WI. Beta-catenin directly displaces Groucho/TLE repressors from Tcf/Lef in Wnt-mediated transcription activation. Nat Struct Mol Biol. 2005;12:364–371.

25. Christman MA 2nd, Goetz DJ, Dickerson E, McCall KD, Lewis CJ, Benencia F, Silver MJ, Kohn LD, Malgor R. Wnt5a is expressed in murine and human atherosclerotic lesions. Am J Physiol Heart Circ Physiol.

2008;294:H2864–H2870.

26. Kim J, Kim J, Kim DW, Ha Y, Ihm MH, Kim H, Song K, Lee I. Wnt5a induces endothelial inflammation via beta-catenin-independent signaling.

J Immunol. 2010;185:1274–1282.

27. Tsaousi A, Williams H, Lyon CA, Taylor V, Swain A, Johnson JL, George SJ. Wnt4/β-catenin signaling induces VSMC proliferation and is associ- ated with intimal thickening. Circ Res. 2011;108:427–436.

28. Xu L, Corcoran RB, Welsh JW, Pennica D, Levine AJ. WISP-1 is a Wnt- 1- and beta-catenin-responsive oncogene. Genes Dev. 2000;14:585–595.

29. Masckauchán TN, Shawber CJ, Funahashi Y, Li CM, Kitajewski J. Wnt/

beta-catenin signaling induces proliferation, survival and interleukin-8 in human endothelial cells. Angiogenesis. 2005;8:43–51.

30. Almeida M, Han L, Bellido T, Manolagas SC, Kousteni S. Wnt proteins prevent apoptosis of both uncommitted osteoblast progenitors and differentiated osteoblasts by beta-catenin-dependent and -independent signaling cascades involving Src/ERK and phosphatidylinositol 3-kinase/AKT.

J Biol Chem. 2005;280:41342–41351.

31. Tapia JC, Torres VA, Rodriguez DA, Leyton L, Quest AF. Casein kinase 2 (CK2) increases survivin expression via enhanced beta-catenin-T cell fac- tor/lymphoid enhancer binding factor-dependent transcription. Proc Natl Acad Sci USA. 2006;103:15079–15084.

32. Gandhirajan RK, Staib PA, Minke K, Gehrke I, Plickert G, Schlösser A, Schmitt EK, Hallek M, Kreuzer KA. Small molecule inhibitors of Wnt/

beta-catenin/lef-1 signaling induces apoptosis in chronic lymphocytic leu- kemia cells in vitro and in vivo. Neoplasia. 2010;12:326–335.

33. Gille JJ, Joenje H. Cell culture models for oxidative stress: superox- ide and hydrogen peroxide versus normobaric hyperoxia. Mutat Res.

1992;275:405–414.

34. Lin CL, Wang JY, Huang YT, Kuo YH, Surendran K, Wang FS. Wnt/beta- catenin signaling modulates survival of high glucose-stressed mesangial cells. J Am Soc Nephrol. 2006;17:2812–2820.

35. Chen S, Guttridge DC, You Z, Zhang Z, Fribley A, Mayo MW, Kitajewski J, Wang CY. Wnt-1 signaling inhibits apoptosis by acti- vating beta-catenin/T cell factor-mediated transcription. J Cell Biol.

2001;152:87–96.

36. Ueda Y, Hijikata M, Takagi S, Takada R, Takada S, Chiba T, Shimotohno K. Wnt/beta-catenin signaling suppresses apoptosis in low serum medium and induces morphologic change in rodent fibroblasts. Int J Cancer.

2002;99:681–688.

37. Hwang SG, Ryu JH, Kim IC, Jho EH, Jung HC, Kim K, Kim SJ, Chun JS. Wnt-7a causes loss of differentiated phenotype and inhibits apopto- sis of articular chondrocytes via different mechanisms. J Biol Chem.

2004;279:26597–26604.

38. Dehner M, Hadjihannas M, Weiske J, Huber O, Behrens J. Wnt signaling inhibits Forkhead box O3a-induced transcription and apoptosis through up-regulation of serum- and glucocorticoid-inducible kinase 1. J Biol Chem. 2008;283:19201–19210.

39. Shin SY, Chin BR, Lee YH, Kim JH. Involvement of glycogen synthase kinase-3beta in hydrogen peroxide-induced suppression of Tcf/Lef- dependent transcriptional activity. Cell Signal. 2006;18:601–607.

40. Almeida M, Han L, Martin-Millan M, O’Brien CA, Manolagas SC.

Oxidative stress antagonizes Wnt signaling in osteoblast precursors by diverting beta-catenin from T cell factor- to forkhead box O-mediated transcription. J Biol Chem. 2007;282:27298–27305.

41. Hoogeboom D, Essers MA, Polderman PE, Voets E, Smits LM, Burgering BM. Interaction of FOXO with beta-catenin inhibits beta-catenin/T cell factor activity. J Biol Chem. 2008;283:9224–9230.

42. Su F, Overholtzer M, Besser D, Levine AJ. WISP-1 attenuates p53-mediated apoptosis in response to DNA damage through activation of the Akt kinase. Genes Dev. 2002;16:46–57.

43. You Z, Saims D, Chen S, Zhang Z, Guttridge DC, Guan KL, MacDougald OA, Brown AM, Evan G, Kitajewski J, Wang CY. Wnt signaling promotes oncogenic transformation by inhibiting c-Myc-induced apoptosis. J Cell Biol. 2002;157:429–440.

44. Venkatesan B, Prabhu SD, Venkatachalam K, Mummidi S, Valente AJ, Clark RA, Delafontaine P, Chandrasekar B. WNT1-inducible signaling pathway protein-1 activates diverse cell survival pathways and blocks doxorubicin-induced cardiomyocyte death. Cell Signal. 2010;22:809–820.

45. Wang S, Chong ZZ, Shang YC, Maiese K. WISP1 (CCN4) autoregulates its expression and nuclear trafficking of β-catenin during oxidant stress with limited effects upon neuronal autophagy. Curr Neurovasc Res.

2012;9:91–101.

46. Reddy VS, Valente AJ, Delafontaine P, Chandrasekar B. Interleukin-18/

WNT1-inducible signaling pathway protein-1 signaling mediates human saphenous vein smooth muscle cell proliferation. J Cell Physiol.

2011;226:3303–3315.

47. Pennica D, Swanson TA, Welsh JW, Roy MA, et al. WISP genes are members of the connective tissue growth factor family that are up-regulated in wnt-1-transformed cells and aberrantly expressed in human colon tumors.

Proc Natl Acad Sci USA. 1998;95:14717–14722.

48. Leu SJ, Lam SC, Lau LF. Pro-angiogenic activities of CYR61 (CCN1) mediated through integrins alphavbeta3 and alpha6beta1 in human umbili- cal vein endothelial cells. J Biol Chem. 2002;277:46248–46255.

49. Lin CG, Chen CC, Leu SJ, Grzeszkiewicz TM, Lau LF. Integrin-dependent functions of the angiogenic inducer NOV (CCN3): implication in wound healing. J Biol Chem. 2005;280:8229–8237.

50. Hoshijima M, Hattori T, Inoue M, Araki D, Hanagata H, Miyauchi A, Takigawa M. CT domain of CCN2/CTGF directly interacts with fibro- nectin and enhances cell adhesion of chondrocytes through integrin alpha- 5beta1. FEBS Lett. 2006;580:1376–1382.

51. Ono M, Inkson CA, Kilts TM, Young MF. WISP-1/CCN4 regulates osteo- genesis by enhancing BMP-2 activity. J Bone Miner Res. 2011;26:193–208.

Apoptosis of smooth muscle cells within the lining of the atherosclerotic plaque is thought to lead to plaque rupture, which in turn results in myocardial infarction and stroke. Consequently a greater understanding of the underlying mechanisms of smooth muscle cell apoptosis and identification of novel approaches to reduce smooth muscle cell apoptosis are desirable, as they may lead to the development of new therapeutic approaches. In this study, we have identified using cultured cells that Wnt proteins can induce the expression of a survival protein (Wnt1-inducible secreted protein-1) via the activation of 2 signaling molecules, cAMP response element–binding protein and β-catenin.

However, in unstable atherosclerotic plaques that have a propensity to rupture the levels of Wnt1-inducible secreted protein-1 are reduced compared with stable plaques. Together this indicates that Wnt1-inducible secreted protein-1 is an important prosurvival factor and that enhancing Wnt1-inducible secreted protein-1 levels in atherosclerotic plaques may suppress smooth muscle apoptosis and thereby reduce plaque rupture.