Age-Dependent and -Independent Effects of Perivascular Adipose Tissue and Its Paracrine Activities during Neointima Formation

Figures

• 

  Figure 1. Age-associated changes of the arterial vessel wall. Cross-sections through the uninjured common carotid artery of adult (n = 11) and middle-aged (n = 11) mice were stained with VES-MTC to visualize elastic fibers (black), muscle cells (red) and extracellular matrix (blue), or with H&E to visualize cell nuclei (blue-black). Antibodies against smooth muscle alpha-actin (SMA) or proliferating cell nuclear antigen (PCNA) were used to immunostain smooth muscle cells (red) or proliferating cells (brown), respectively. The media area (A) and adventitia area (B) were quantified using ImagePro Plus analysis software. The total number of cells was counted manually and expressed per mm² media area (C). The SMA-immunopositive area per total media area was determined using the count-size function (D). Individual data points and the mean ± SD are shown. * p < 0.05, ** p < 0.01 and *** p < 0.001 vs. adult mice (as determined by Student’s t-test for unpaired means (in B) or the Mann–Whitney test (in A,C,D)). Representative findings are shown in (E). Size bars represent 100 µm. Total mRNA was isolated from the aorta of adult (n = 4) and middle-aged (n = 4) mice and the mRNA expression of cyclin…

• 

  Figure 2. Neointima formation in adult and middle-aged mice. Vascular injury was induced at the carotid artery of C57BL6/JRj WT mice aged 16 weeks (adult; n = 9) or 52 weeks (middle-aged; n = 9). The neointima area (A), lumen stenosis (B), media area (C), total vessel area (D) and adventitia area (E) were quantified on at least three Verhoeff’s elastica (VES)-stained paraffin cross-sections through the injured carotid artery as well as the contralateral, uninjured artery (control), and results averaged per mouse. Individual mean values per mouse and the mean ± SD are shown. * p < 0.05, ** p < 0.01 and *** p < 0.001 vs. controls and # p < 0.05 and ## p < 0.01 vs. adult mice (as determined by the Kruskall–Wallis, followed by Dunn’s multiple comparisons test (A,B,E) or one-way ANOVA, followed by Sidak’s multiple comparisons test (C,D). Representative images in adult and middle-aged mice are shown in (F). Size bars represent 100 µm.

• 

  Figure 3. Effect of perivascular adipose tissue age on neointima formation. Vascular injury was induced at the carotid artery of NMRI nu/nu mice aged ten weeks immediately followed by transplantation (Tx) of PVAT freshly harvested from the thoracic aorta of adult (n = 9) or middle-aged (n = 8) C57BL6JRj mice. Nu/nu mice with their own PVAT left intact and not transplanted were used as control. Three weeks later, the neointima area (A), lumen stenosis (B), media area (C) and total vessel area (D) was morphometrically quantified on at least three VES-stained cross-sections and results averaged per mouse. Individual mean values per mouse and the mean ± SD are shown. * p < 0.05 vs. NMRI nu/nu not transplanted with PVAT (control; n = 11), as determined by the Kruskall-Wallis (A,C) or one-way ANOVA (B,D) test. Representative images are shown in (E). Size bars represent 100 µm.

• 

  Figure 4. Differential effects of age on adipose tissue gene expression patterns. Visceral (VAT) and perivascular (PVAT) adipose tissues were harvested from adult (n = 5–16) and aged (n = 5–16) mice and total RNA isolated. Real-time PCR analysis was employed to quantify the mRNA expression of (A) interleukin-1alpha (Il1α), (B) macrophage inflammatory protein 1-alpha (Mip1α), (C) matrix metalloproteinase-3 (Mmp3), (D) Mmp10, (E) tumor necrosis factor-alpha (Tnfα), (F) Il18, (G) angiotensin-converting enzyme-2 (Ace2), (H) adiponectin (Adipoq), (I) sirtuin-1 (Sirt1), (J) Il10, (K) transforming growth factor-beta (Tgfβ), and (L) Mmp12. Individual data points and the mean ± SD are shown. * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001 vs. adult mice and # p < 0.05, ## p < 0.01 and ### p < 0.001 vs. VAT (as determined by two-way ANOVA).

• 

  Figure 5. Perivascular adipose tissue expression of negative regulators of STAT1 and NFκB. (A,B) Western blot analysis of protein inhibitor of activated STAT-1 (PIAS1), (A,C) suppressor of cytokine signaling-1 (SOCS1), (D,E) myeloid differentiation primary response-88 (MyD88), and (D,F) tumor suppressor p53 (p53) in visceral (VAT) and perivascular (PVAT) adipose tissue of adult (n = 3–8) and middle-aged (n = 3–8) mice. Results were normalized to GAPDH protein levels and are expressed as -fold change vs. findings obtained in VAT of adult mice. Individual data points and the mean ± SD are shown. # p < 0.05 and ## p < 0.01 vs. VAT (as determined using two-way ANOVA). Representative Western blot membranes are shown in panels (A,D).

• 

  Figure 6. Expression of PIAS1 in human perivascular adipose tissue. Quantitative analysis of PIAS1 protein levels in visceral (VAT) and perivascular (PVAT) adipose tissue surrounding the aorta of patients (n = 5) with abdominal aortic aneurysm (AAA) using an enzyme-linked immunoassay (A). Results are expressed per ng/mg total protein. Quantitative analysis of PIAS1 in PVAT surrounding the aortic root and coronary arteries of patients (n = 8) with coronary artery disease (CAD) compared to PVAT surrounding the internal mammary artery (IMA-PVAT) (B). Quantitative analysis of PIAS1 in PVAT surrounding the aortic root and coronary arteries of patients (n = 9) with aortic valve stenosis (AS) compared to epicardial adipose tissue (EAT) from the same patient (C). Exact p-values, as determined using the Wilcoxon matched-pairs signed rank test (A) or paired Student’s t-test (B,C) are shown within the graphs. Immunohistochemical detection of PIAS1 (red signal) in human VAT and PVAT (×400 magnification) (D). Results after the omission of the first antibody (neg control) are also shown (×200 magnification).