Figure 1.
Expression of periostin in human AAA specimens.
A: Protein samples were obtained from the walls of human AAA specimens (n = 42) and non-aneurysmal aortic walls (Control, n = 5). Representative western blot results for periostin are shown (top panel) with quantitative analyses (bottom panel). GAPDH served as an internal control. Data are mean ± SE. *p<0.05 compared to Control. B: Representative histological and immunohistochemical stains are shown for the walls of human AAA specimens. Different regions of the aorta include parts that are non-dilated (ND), transitional (T), and maximally dilated (MD). The luminal surface is oriented toward the top of each panel. Hematoxylin/eosin (HE), elastica van-Gieson (EVG), and Masson trichrome (MT) stains depict cell nuclei (blue-black), elastin network (black), and collagen fibers (blue), respectively. Localization of periostin is indicated by red staining. C: Localization of periostin and α-smooth muscle actin (α-SMA) is indicated by red staining.
Figure 2.
Temporal pattern of periostin expression during the development of abdominal aortic aneurysm (AAA) in a mouse model.
A: Representative photographs show mouse abdominal aortas 42 days after periaortic application of saline (control) or CaCl2 (for AAA induction). B: Changes in the diameters of abdominal aortas are shown for the indicated time points after application of saline or CaCl2 (3 days, n = 3; 7 days, n = 4; 14 days, n = 4; 28 days, n = 3–4; 42 days, n = 7–8). Data are mean ± SE. *p<0.05 compared to saline controls. C: Representative images show aortic walls stained with hematoxylin/eosin (HE), elastica van-Gieson (EVG), or antibody against periostin at the indicated time points after application of saline or CaCl2.
Figure 3.
Role of periostin in linking mechanical strain to inflammatory responses in VSMCs.
Cultured rat VSMCs treated with or without periostin-neutralizing antibody (PN-nAb, 1 µg/ml) were stimulated with cyclic uniaxial strain (20% elongation in length at a frequency of 30 cycles/min for 48 h, n = 6). Protein levels of periostin (A) and MCP-1 (B) in cell lysates were determined by western blotting. Levels of active MMP-2 in the conditioned media were determined by gelatin zymography (C). Levels of phosphorylated FAK (D), ERK (E) and JNK (F) in cell lysates were determined by western blotting. Levels of total FAK (G), ERK (H) and JNK (I) in cell lysates were determined by western blotting. GAPDH served as an internal control. Rat VSMCs were stimulated with recombinant periostin (rPeriostin; 1 µg/ml for 48 h; n = 4). Protein levels of MCP-1 in conditioned media were determined by ELISA (J). Data are mean ± SE. *p<0.05 and **p<0.01 compared to Control; #p<0.05 compared to VSMCs under mechanical strain.
Figure 4.
Role of periostin/FAK axis in mechanotransduction signaling pathway in VSMCs.
Rat VSMCs were treated with or without of FAK inhibitor (PF573228, 10 µM, FAK-I) and stimulated with cyclic uniaxial strain (20% elongation in length at a frequency of 30 cycles/min for 48 h, n = 5). Levels of phosphorylated FAK (A, B), ERK (A, C) and JNK (A, D) in cell lysates were determined by western blotting. Protein levels of MCP-1 (A, E) and periostin (A, G) in cell lysates, and active MMP-2 (A, F) in conditioned media were determined by western blotting and gelatin zymography, respectively. GAPDH served as an internal control. Data are mean ± SE. *p<0.05 and **p<0.01 compared to Control; #p<0.05 and ##p<0.01 compared to VSMCs under mechanical strain.
Figure 5.
Role of periostin/FAK axis in sustaining inflammatory responses in human AAA tissues.
Human AAA tissues were cut into small pieces and cultured with or without of FAK inhibitor (PF573228, 10 µM, FAK-I) for 48 h (n = 5). Levels of phosphorylated FAK (A), ERK (B) and JNK (C) in tissue lysates were determined by western blotting. GAPDH served as an internal control. Protein levels of MCP-1 (D) and MMPs (E-F) in conditioned media were determined by ELISA and gelatin zymography, respectively. Data are mean ± SE. *p<0.05 and **p<0.01 compared to Control.
Figure 6.
Role of periostin in inducing inflammatory responses in vivo.
A-B: Gelfoam patches containing PBS (Control, n = 6) or recombinant periostin (rPeriostin, n = 7) were placed into periaortic spaces of mice for 7 days. Representative images show mouse aortic walls, stained with hematoxylin/eosin (HE) or with antibodies against periostin, phosphorylated FAK (pFAK), or MCP-1 (A). Infiltrating cells were counted in 5-10 high-power fields (B). Data are mean ± SE. *p<0.05 compared to Control. C: Schematic diagram represents the proposed vicious cycle of periostin upregulation driven by mechanical strain through activation of FAK, resulting in the maintenance of inflammation in AAA.