Peroxisome Proliferator-Activated Receptor γ Level Contributes to Structural Integrity and Component Production of Elastic Fibers in the Aorta

Abstract

Loss of integrity and massive disruption of elastic fibers are key features of abdominal aortic aneurysm (AAA). Peroxisome proliferator-activated receptor γ (PPARγ) has been shown to attenuate AAA through inhibition of inflammation and proteolytic degradation. However, its involvement in elastogenesis during AAA remains unclear. PPARγ was highly expressed in human AAA within all vascular cells, including inflammatory cells and fibroblasts. In the aortas of transgenic mice expressing PPARγ at 25% normal levels (Pparg(C) (/-) mice), we observed the fragmentation of elastic fibers and reduced expression of vital elastic fiber components of elastin and fibulin-5. These were not observed in mice with 50% normal PPARγ expression (Pparg(+/-) mice). Infusion of a moderate dose of angiotensin II (500 ng/kg per minute) did not induce AAA but Pparg(+/-) aorta developed flattened elastic lamellae, whereas Pparg(C/-) aorta showed severe destruction of elastic fibers. After infusion of angiotensin II at 1000 ng/kg per minute, 73% of Pparg(C/-) mice developed atypical suprarenal aortic aneurysms: superior mesenteric arteries were dilated with extensive collagen deposition in adventitia and infiltrations of inflammatory cells. Although matrix metalloproteinase inhibition by doxycycline somewhat attenuated the dilation of aneurysm, it did not reduce the incidence nor elastic lamella deterioration in angiotensin II-infused Pparg(C/-) mice. Furthermore, PPARγ antagonism downregulated elastin and fibulin-5 in fibroblasts, but not in vascular smooth muscle cells. Chromatin immunoprecipitation assay demonstrated PPARγ binding in the genomic sequence of fibulin-5 in fibroblasts. Our results underscore the importance of PPARγ in AAA development though orchestrating proper elastogenesis and preserving elastic fiber integrity.

Keywords: aneurysm; angiotensin II; collagen; doxycycline; fibroblasts.

Figure 1

Expression of PPARγ in human and mouse AAA. A–F, Immunohistochemical staining of PPARγ in normal human aorta (A–C) and human AAA (D–F). Filled arrows in (C) and (F) indicate cells resembling inflammatory cells. Open arrows in (E) resemble VSMCs. Open arrowheads in (F) indicate elongated endothelial cells or fibroblasts in or near microvessels. Int: intima; Med: media; Adv: adventitia. G–I, Normal AA of WT mice. Magnification of the black square in (G) is shown in (H) and (I). PPARγ expression is detected in some VSMCs of WT mice. J–L, AngII-induced dissecting AAA in WT mice. Section (J) shows prominent PPARγ expression in the inflammatory cells (filled arrows) of the luminal surface (K) and adventitia (L) as well as in VSMCs (open arrow in K), and fibroblasts (open arrowheads in L). Magnification of the black square is shown in the indicated picture. Scale bars in (B, C, E, F, H, I, K, and L) are 100 μm and in (A, D, G and J) are 1000 μm.

Figure 2

Elastic fiber fragmentation and expression of elastic fiber components in Pparg^C/− aorta. A, Immunoblot of PPARγ, cathepsin S and MMP-9 in the aorta (N=3 in each group). The relative intensities of the bands are indicated by densitometric quantification with WT. B, Tail-cuff and (C) telemetry systolic BP measurement in Pparg^+/+ and Pparg^C/− mice. N=6 in (B) and 5 in (C). D, Representative images of the elastic network and quantification of elastic fiber breaks in the longitudinal section of aorta. Arrows indicate breaks in the elastic fiber. Numbers of breaks of elastic layers per 10,000 μm² are shown in the right (N=5 in each group). E, mRNA levels of elastolytic enzymes in Pparg^C/− aorta are shown relative to the mean levels in Pparg^+/+ aorta as 1.0 (+/+ = 7, C/− = 6). F, Immunohistochemical staining of MMP-9 in the longitudinal section of aorta. Lu: lumen. G, mRNA levels of inflammatory cytokines and macrophage markers relative to the mean levels in Pparg^+/+ aorta as 1.0 (+/+ = 20, C/− = 16). H, Immunoblot and quantification of ECM components in the soluble fraction of aortic lysates (N=3 in each group) I, mRNA levels of ECM components (N=12–14 in each group).*P<0.05, **P<0.01, and ***P<0.001. Scale bars in (D) are 100 μm and in (F) are 50 μm.

Figure 3

Effects of moderate-dose AngII (500 ng/kg/min) infusion. A, Tail-cuff systolic BP measurement in AngII-infused Pparg^+/+ and Pparg^C/− mice (N=9 in each group). B, Representative images of the elastic network in the TA and AA. Scale bars are 100 μm. C, Parameters of elastic fiber integrity, including thickness, waviness, consistency, and breaks of elastic lamellae, in the AA (+/+ = 5, +/− = 3, C/− = 7). *P<0.05, **P<0.01, and ***P<0.001 by one-way ANOVA with Tukey HSD test. D and E, mRNA levels for inflammatory mediators and for elastolytic proteases, respectively, in the aorta. Data are expressed relative to the mean in Pparg^+/+ aorta as 1.0 (N=5 in each group). *P<0.05 and **P<0.01 by Student’s t-test. F, Gelatin zymography of protein lysate from TA and AA.

Figure 4

Suprarenal aneurysms in high-dose AngII (1000 ng/kg/min) infused Pparg^C/− mice. A, Representative photographs of the aorta. White arrow indicates aneurysm. B, Magnified view of AA region. CA: celiac artery. SMA: superior mesenteric artery. RR: right renal artery. LR: left renal artery. C, Maximal external aortic diameter and percentage of severity grade in AngII-infused mice (+/+ = 6, +/− = 5, C/− = 6). D and E, Elastin, collagen (PS red) staining and immunohistochemistry staining for SMactin of AA. ***P<0.001 by one-way ANOVA with Tukey HSD test. Scale bars in (D) are 200 μm and in (E) are 1000 μm.

Figure 5

Effects of MMP inhibition on AngII-infused Pparg^C/− aneurysm. A, Gelatin zymography of protein lysates from AA. B, Representative photographs of the AAA. C, Maximal aortic diameter, (D) percentage of severity grade and (E) incidence of AAA in AngII-infused Pparg^C/− mice. F, Representative images of the elastic network and (G) parameters of elastic fiber integrity in AA (N=5–6 in each group). Locations of sections in (F) were indicated by dash lines in (B). Scale bars are 200 μm in upper panels and 100 μm in lower panels of (F). *P<0.05.

Figure 6

Effects of PPARγ antagonism on expression of elastic fiber components. Expression of Eln and Fbln5 after treatment with GW9662 or TNF-α in rat ASMCs (A, gray bars) and MEFs (B, white bars). N=3 in each group. C, Expression of Pparg, Eln and Fbln5 in MEFs and (D) HAoAFs treated with scrambled siRNA or siRNA against PPARγ (siPPARγ). N=7–8 in each group. E, ChIP assay in MEFs. Sequences containing the potential PPARγ binding sites in Eln and Fbln5 were amplified by real-time PCR. Rosi: rosiglitazone. Expression of Mmp9 after treatment with GW9662, AngII, or TNF-α in rat ASMCs (F) and MEFs (G). *P<0.05, **P<0.01, and ***P<0.001 in (A–B and E–G) by one-way ANOVA with Tukey HSD test. *P<0.05 in (C and D) by Student’s t-test.