|Home | About | Journals | Submit | Contact Us | Français|
Peroxisome proliferator-activated receptor γ (PPARγ) agonists attenuate atherosclerosis and abdominal aortic aneurysms (AAAs). PPARγ, a nuclear receptor, is expressed on many cell types including smooth muscle cells (SMCs).
To determine whether a PPARγ agonist reduces angiotensin II (AngII)-induced atherosclerosis and AAAs via interaction with SMC-specific PPARγ.
LDL receptor −/− mice with SMC-specific PPARγ deficiency were developed using PPARγ floxed (PPARγf/f) and SM22 Cre+ mice. PPARγf/f littermates were generated that did not express Cre (Cre0/0) or were hemizygous for Cre (Cre+/0). To assess the contribution of SMC-specific PPARγ in ligand-mediated attenuation of AngII-induced atherosclerosis and AAAs, both male and female Cre0/0 and Cre+/0 mice, were fed a fat-enriched diet with or without PPARγ agonist, pioglitazone (Pio; 20 mg/kg/day) for 5 weeks. After 1 week of feeding modified diets, mice were infused with AngII (1,000 ng/kg/min) for 4 weeks. SMC-specific PPARγ deficiency or Pio administration had no effect on plasma cholesterol concentrations. Pio administration attenuated AngII-increased systolic blood pressure equivalently in both Cre0/0 and Cre+/0 groups. SMC-specific PPARγ deficiency increased atherosclerosis in male mice. Pio administration reduced atherosclerosis only in the Cre0/0 mice, but not in mice with SMC-specific PPARγ deficiency. SMC-specific PPARγ deficiency or Pio administration had no effect on AngII-induced AAA development. Pio also did not attenuate AngII-induced MCP-1 production in PPARγ deficient SMCs.
Pio attenuates AngII-induced atherosclerosis via the interaction with SMC-specific PPARγ, but has no effect on the development of AAAs.
Thiazolidinediones (TZDs), including rosiglitazone and pioglitazone (Pio), are used widely to improve insulin sensitivity in patients with type 2 diabetes. Experimentally, TZDs reduce atherosclerosis in both LDL receptor −/− and apoE −/− mice.1,2 Recent studies have demonstrated that TZDs also reduce AngII-induced abdominal aortic aneurysm (AAA) development in apoE −/− mice.3,4 The molecular target for TZD is PPARγ; a nuclear receptor that is highly expressed in all cell types involved in vascular pathologies, including macrophages, endothelial cells, and smooth muscle cells (SMCs).5 Currently, it has not been defined whether the beneficial effects of TZDs are attributable to PPARγ agonism in a specific cell type. Furthermore, it has been suggested that TZDs may exert some of their biological effects through PPARγ-independent mechanisms, although this has not been defined in vascular pathologies.6
TZDs have been demonstrated to regulate important SMC functions, including proliferation and migration.7 SMC-specific genetic manipulations have resulted in changes in both atherosclerosis 8,9 and AAAs in mice.9 Furthermore, SMC-specific gene deletion of PPARγ results in changes in blood pressure and injury-induced vascular hyperplasia.10,11 However, no studies have currently determined whether the benefits of TZDs on vascular pathologies are mediated via a SMC-specific PPARγ-dependent mechanism.
To elucidate a role of SMC-specific PPARγ expression on TZD-induced reductions in atherosclerosis and AAAs, we bred female LDL receptor −/− mice harboring PPARγ floxed genes to similarly genetically manipulated males that were hemizygous for Cre regulated by the SM22 promoter. This breeding strategy generated littermate controls that were either wild type or SMC-specific deficient in PPARγ. Using these mice, we determined the contribution of PPARγ expression in SMCs to the effects of Pio on AngII-induced atherosclerosis and AAAs.12 The results demonstrate that SMC-PPARγ deficiency resulted in increased AngII-induced atherosclerosis. Furthermore, these data demonstrate that PPARγ expression in SMCs is a major contributor to Pio-induced reduction in atherosclerosis. Contrary to previous studies, we did not discern an effect of Pio on AngII-induced AAAs.
Materials and methods are described in the online supplement.
To verify the genotype of mice, aortas were dissected free, adventitia and endothelium removed, and DNA was isolated from SMC-containing media. PCR analyses were performed on DNA isolated from the arch, thorax, suprarenal, and infrarenal aortic regions to determine the uniformity of Cre-based exon excision. These analyses demonstrated the presence of non-functional alleles (240 bp amplicon) throughout aortas of Cre-expressing mice. In contrast, aortas from non transgenic littermates generated 215 bp amplicons derived from intact floxed genes (Figure 1A).
RT-PCR analyses showed complete deletion of PPARγ mRNA in SMC aortic medias of Cre+/0 mice (Figure 1B), indicating that functional PPARγ transcripts were ablated. Western blot analyses demonstrated that PPARγ protein was ablated in aortic SMCs from Cre+/0 mice, while not influencing abundance in liver, kidney, and adipose tissue (Figure 1C and Online Figure II).
SMC-specific PPARγ deficiency in LDL receptor−/− mice resulted in significant (P<0.05) increases in AngII-induced atherosclerotic lesion areas in male mice, but had no effect in females (Figure 2). SMC-specific deletion of PPARγ had no effect on body weight, plasma total cholesterol concentrations (Online Table I), or lipoprotein-cholesterol distributions (data not shown). AngII infusion significantly increased systolic blood pressure (SBP) in male mice of both groups (Online Table I). SMC-specific PPARγ deficiency had no effect on AngII-induced AAA formation (Figure 2C) or aortic rupture (Cre0/0 − 25% vs. +/0 − 28%) in either gender.
In AngII-infused mice fed a fat-enriched diet, PPARγ mRNA abundance was not significantly increased in peritoneal macrophages (Online Figure III). Pio administration to these mice induced PPARγ mRNA abundance and activity in selected cell types and tissues, including macrophages, liver, kidney, and adipose. These inductions did not differ between Cre0/0 and Cre+/0 mice (Figure 3A,B and Online Figure IV). Increased PPARγ activity was demonstrable by increased mRNA abundance of PPARγ target genes; AP2 and CD36 in macrophages (Online Figure V) and selected tissues (Online Figure VI).
Pio administration profoundly reduced atherosclerosis only in Cre0/0 mice, but not in mice with SMC-specific PPARγ deficiency (Figure 3C and 3D). In contrast, Pio administration significantly attenuated AngII-increased SBP equivalently in both Cre0/0 and +/0 groups (Online Table 2). Pio administration had no effect on body weight, plasma total cholesterol concentrations (Online Table II) or lipoprotein cholesterol distributions (data not shown). AAA formation (Figure 3E), or aortic rupture (Cre0/0 −11% versus +/0 − 20%) was not different between groups
Immunostaining of atherosclerotic lesions with α-actin demonstrated uniform reactivity throughout the medial intralaminar spaces of all groups, but minimal SMC immunostaining was detected in atherosclerotic lesions from any group. Although PPARγ deficiency increased lesion size, immunostaining for macrophages was dominant in atherosclerosis from both Cre0/0 or Cre +/0 mice.
To define potential mechanisms of Pio reducing atherosclerosis, plasma MCP-1 concentrations were measured. No significant difference was observed among groups demonstrating no systemic effect on MCP-1 (Online Figure VII).
Aortic SMCs cultured from either Cre0/0 or Cre +/0 mice were incubated with Pio (20 μM) for 24 hours, and with or without AngII (1 μM) for a further 18 hours. AngII significantly increased MCP-1 concentrations from Cre+/0 SMCs, but had no significant effect on Cre0/0 SMCs (Figure 4). Co-incubation with Pio had no effect on AngII-induced MCP-1 production in Cre+/0 SMCs.
Consistent with SMCs harvested from Cre0/0 and Cre+/0 mice, AngII increased MCP-1 concentrations in media of SMCs cultured from mice expressing a dominant negative mutation of PPARγ P465L (PPARγL+)13 but not in cells isolated from non transgenic littermates. To determine whether PPARγ has a dominant effect on MCP-1 secretion, SMCs were incubated with IFNγ. In contrast to AngII, IFNγ (300 U/ml) significantly increased MCP-1 concentrations in media of SMCs from both strains (Figure 4B). Pio had no effect on IFNγ induced MCP-1 (Figure 4C).
To confirm that the effects of Pio on MCP-1 were due to interactions with PPARγ, Cre0/0 and +/0 SMCs were incubated with Pio and AngII as described above. The absence of PPARγ in SMCs significantly lowered AP2 mRNA abundance, but failed to affect CD36 (Online Figure VIII). AngII incubation significantly reduced mRNA abundance of both these target genes in Cre+/0 SMCs. Co-incubation of AngII and Pio significantly attenuated the reduced mRNA abundance of target genes in SMCs from Cre0/0 but not Cre+/0 mice.
In the present study, we demonstrate that SMC-specific PPARγ deficiency augments AngII-induced atherosclerosis in male LDL receptor −/− mice. Interestingly, Pio administration attenuates AngII-induced atherosclerosis only in wild type mice but not in SMC-specific PPARγ-deficient mice, which characterizes SMC-specific PPARγ as the key molecular target for the ligand-mediated attenuation of atherosclerosis.
SMC-specific PPARγ deficiency augmented AngII-induced atherosclerosis only in male mice. This is in agreement with the study of Li and colleagues, in which the attenuation of atherosclerosis by a PPARγ ligand was only observed in male LDL receptor −/− mice.1 The basis for these gender differences have not been defined.
Pio administration activates PPARγ in both Cre0/0 and Cre+/0 genotypes, which was evidenced by increased PPARγ expression observed in peritoneal macrophages and other tissues. Previous in vitro studies demonstrated that TZDs inhibited SMC proliferation and induced apoptosis through PPARγ dependent mechanisms.14 In the present study, Pio administration attenuates AngII-induced atherosclerosis only in Cre0/0 mice, but not in mice with SMC-specific PPARγ deficiency. Considering that SMC proliferation constitutes an important cellular mechanism for atherosclerosis initiation,15 our findings demonstrated not only SMC-specific PPARγ as an endogenous inhibitor of atherosclerosis, but also established that TZDs exert anti-atherosclerotic effects through this pathway.
Pio administration significantly suppresses AngII-induced SBP in both genotypes. This result indicates that Pio-mediated SBP lowering effect is independent of SMC-specific PPARγ. In support of this observation, a recently published paper using both SM22-Cre+ and Tie2-Cre+ PPARγ flox mice, showed that TZD-mediated the SBP lowering effects via PPARγ expressed in endothelium.16 Since endothelial PPARγ is intact, Pio administration attenuates AngII-induced SBP in both Cre0/0 and Cre+/0 groups in our study.
SMC-specific PPARγ deficiency or Pio administration did not influence aneurysm formation in LDL receptor −/− mice, which is contrary to a recent publication in which Pio reduced suprarenal aortic expansion in AngII-infused ApoE−/− mice.4 The differences may be due to the lower dose used in the present study.4 Our dietary delivery was estimated to be ~20 mg/kg/day, while the drinking water delivery in the study of Golledge et al.4 was estimated to be 50 mg/kg/day. In another study, rosiglitazone attenuated AngII-induced AAA formation in ApoE−/− mice, which was mainly associated with decreased expression of inflammatory mediators.3 The basis for the inconsistent effects of TZDs on AngII-induced AAAs is unclear.
To further understand the mechanism by which Pio mediates its effect via SMC-PPARγ on atherosclerosis, we examined the effect of AngII on MCP-1 production in cultured Cre+ and PPARγL+ SMCs. Interestingly, AngII activates MCP-1 production only in Cre+/0 and PPARγL+ SMCs, but not in control SMCs, suggesting that endogenous SMC-PPARγ regulates AngII-induced MCP-1 production. In addition, Pio had no effect on AngII-induced MCP-1 production in Cre+/0 SMCs which is consistent with this TZD requiring interaction with PPARγ to reduce AngII-induced atherosclerosis. The specificity of this pathway was demonstrated by the continued induction of MCP-1 secretion in PPARγL+ cells during IFNγ incubation that signals via CD74 pathway in SMCs.17 This SMC-PPARγ dependent effect of AngII is localized to SMCs that is not reflected by plasma concentrations of MCP-1.
In summary, this study provides evidence that lack of PPARγ in vascular SMCs results in significant increases in atherosclerosis associated with increased MCP-1 production. Furthermore, the study reveals that SMC-specific PPARγ expression is a novel mediator of ligand-mediated attenuation of atherosclerosis.
PPARγ is a nuclear receptor that is highly expressed in many of cell types involved in vascular pathologies, including macrophages, endothelial cells and smooth muscle cells (SMCs). The (TZDs agoinsts of PPARγ have been shown to inhibit the development of atherosclerosis in male animals. Currently, it is unclear whether the beneficial effects of TZDs could be attributed to PPARγ agonism in a specific cell type. In vitro, TZDs inhibit SMC proliferation and migration, the key events that promote intimal hyperplasia during atherogenesis; however, the contribution of SMC PPARγ to the anti-atherogenic effects of TZD has not been assessed. Because TZDs regulate SMC proliferation, which is a key step in the development of atherosclerosis, we hypothesized that SMC-specific PPARγ is responsible for the beneficial effects of TZD on atherosclerosis. By generating SMC-specific PPARγ deficient mice, we show that SMC-specific PPARγ plays a critical role in the development of angiotensin IIinduced atherosclerosis. We demonstrate that PPARγ expression in SMCs is required for the reduction in AngII-induced atherosclerosis by pioglitazone. This is the first study to report that pioglitazone exerts its beneficial effect on atherosclerosis via a SMC-specific PPARγ-dependent mechanism.
We acknowledge Deborah Howatt, Jessica Moorleghen, Debra Rateri, and Anju Balakrishnan for technical assistance, Richard Charnigo Jr for assistance with statistics, and Takeda Pharmaceuticals for providing pioglitazone. We thank Manikandan Panchatcharam, Susan Smyth, and Nobuyo Maeda, (UNC) for providing PPARγL+ cells.
SOURCES OF RESEARCH SUPPORT
These studies were supported by NHLBI grants (HL80010 to JG and HL80100 to AD) and an AHA Great Rivers Affiliate Postdoctoral Fellowship (0825592D) to VS.