Adipocytes promote interleukin-18 binding to its receptors during abdominal aortic aneurysm formation in mice

Abstract

Aims: Obesity is a risk factor of abdominal aortic aneurysm (AAA). Inflammatory cytokine interleukin-18 (IL18) has two receptors: IL18 receptor (IL18r) and Na-Cl co-transporter (NCC). In human and mouse AAA lesions, IL18 colocalizes to its receptors at regions rich in adipocytes, suggesting a role of adipocytes in promoting IL18 actions in AAA development.

Methods and results: We localized both IL18r and NCC in human and mouse AAA lesions. Murine AAA development required both receptors. In mouse AAA lesions, IL18 binding to these receptors increased at regions enriched in adipocytes or adjacent to perivascular adipose tissue. 3T3-L1 adipocytes enhanced IL18 binding to macrophages, aortic smooth muscle cells (SMCs), and endothelial cells by inducing the expression of both IL18 receptors on these cells. Adipocytes also enhanced IL18r and IL18 expression from T cells and macrophages, AAA-pertinent protease expression from macrophages, and SMC apoptosis. Perivascular implantation of adipose tissue from either diet-induced obese mice or lean mice but not that from leptin-deficient ob/ob mice exacerbated AAA development in recipient mice. Further experiments established an essential role of adipocyte leptin and fatty acid-binding protein 4 (FABP4) in promoting IL18 binding to macrophages and possibly other inflammatory and vascular cells by inducing their expression of IL18, IL18r, and NCC.

Conclusion: Interleukin-18 uses both IL18r and NCC to promote AAA formation. Lesion adipocyte and perivascular adipose tissue contribute to AAA pathogenesis by releasing leptin and FABP4 that induce IL18, IL18r, and NCC expression and promote IL18 actions.

Keywords: Abdominal aortic aneurysm; Adipocyte; IL18; IL18 receptor; Na-Cl co-transporter.

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: journals.permissions@oup.com.

Figures

Figure 1

Na-Cl co-transporter expression in human and mouse abdominal aortic aneurysm lesions. Immunoblot and quantification (A) and RT-PCR (B) tested interleukin-18 receptor, Na-Cl co-transporter, and interleukin-18 expression in normal human abdominal and ascending aortas (n = 7) and abdominal aortic aneurysm lesions (n = 7). Immunofluorescent staining localized Na-Cl co-transporter to macrophages (C), T cells (D), smooth muscle cells (E), and endothelial cells (F) as summarized (G), and localized interleukin-18 to interleukin-18 receptor in human abdominal aortic aneurysm lesions (H). Data were representative of nine abdominal aortic aneurysm and six normal aortas (Supplementary material online, Figure S2). Immunofluorescent staining also localized Na-Cl co-transporter to macrophages (I), T cells (J), smooth muscle cells (K), and microvessel endothelial cells (L) in mouse abdominal aortic aneurysm lesions as summarized (M), including nine abdominal aortic aneurysm and nine normal aortas (Supplementary material online, Figure S5). Scale: 200 µm, inset: 70 µm. Adv, adventitia; Med, media.

Figure 2

Participation of Na-Cl co-transporter and interleukin-18 receptor in mouse abdominal aortic aneurysm. (A) Blood pressure before and after Ang-II-induced abdominal aortic aneurysm. (B) Abdominal aortic diameter. (C) Mortality rate. (D) Lesion macrophage-positive area. Scale: 200 µm. (E) CD4+ T-cell number. Scale: 100 µm. (F) CD8+ T-cell number. Scale: 100 µm. Representative images for D–F are shown to the right. (G) abdominal aortic aneurysm tissue gelatin gel zymogram. Active MMPs are indicated. Coomassie gel staining ensured equal protein loading. Representative image is shown to the left. Data represent at least three independent experiments. Mouse genotype of each cohort is indicated. The results are expressed as mean ± standard deviation of 9–13 mice per group.

Figure 3

Na-Cl co-transporter and interleukin-18 receptor in abdominal aortic aneurysm lesion matrix remodelling and inflammatory molecule expression. (A) abdominal aortic aneurysm lesion elastica fragmentation in grade. Scale: 100 µm. (B) Sirius red collagen staining in grade. Scale: 250 µm. (C) TUNEL+ cell number. (D) Ki67+ cell number. α-Actin+ (E) and Myh11+ (F) arterial wall smooth muscle cell loss in grade. (G) Myh11+Ki67+ media proliferating smooth muscle cell content. (H) Microvessel number. Scale in panels in C–H: 100 µm. Representative images are shown to the right. (I) Abdominal aortic aneurysm lesion IL6-, interferon-γ-, and TNF-α-positive areas. (J) Plasma IL6 and monocyte chemoattractant protein-1. (K) Plasma interferon-γ from mice with and without abdominal aortic aneurysm. (L) Plasma IL4 and IL5. Mouse genotype of each cohort is indicated. The results are expressed as mean ± standard deviation of 9–13 mice per group.

Figure 4

Immunofluorescent staining of interleukin-18, Na-Cl co-transporter, and interleukin-18 receptor in mouse abdominal aortic aneurysm lesions. Interleukin-18 and Na-Cl co-transporter in lesions with intact aortic wall (A) or colocalization in lesions with elastica fragmentation (B). (C) Interleukin-18–Na-Cl co-transporter colocalization in lesion periaortic adipose tissue. Parallel sections were used for elastin and H&E staining. Arrows indicate elastica fragmentation. (D) Immunofluorescent staining localized FABP4 and adiponectin to adipocytes from parallel sections. (E) Interleukin-18 receptor–Na-Cl co-transporter colocalization. (F) Interleukin-18–interleukin-18 receptor colocalization in lesions with different degrees of elastica fragmentation. (G) interleukin-18—interleukin-18 receptor colocalization in lesion periaortic adipose tissue. The parallel section was stained for FABP4 (H) and adiponectin (I). (J) Pearson’s correlations between interleukin-18–Na-Cl co-transporter colocalization, media elastica fragmentation, and lesion FABP4+ adipocyte content. Data were representative of nine samples per experiment. Scale: 200 µm, inset: 70 µm.

Figure 5

Adipocytes induced interleukin-18, interleukin-18 receptor, and Na-Cl co-transporter expression and interleukin-18 binding to its receptors. Confocal analyses (A) and quantification (B) of FITC-interleukin-18 binding on macrophages from different mice. Scale: 100 µm. (C) Avidin staining detection of biotin-interleukin-18 binding on smooth muscle cells, endothelial cells, and macrophages after different treatments. Scale: 50 µm. Immunoblot (D) and quantification (E) of Na-Cl co-transporter and interleukin-18 receptor-α from smooth muscle cells, endothelial cells, and macrophages. Immunoblot (F) and quantification (G) of interleukin-18 and interleukin-18 receptor-α from CD4+ and CD8+ T cells. (H) Immunofluorescent staining of apoptotic smooth muscle cells. Representative images are shown to the left. Scale: 100 µm. Data in B and H were mean ± standard deviation of 4–8 experiments and data in E and G were mean ± standard deviation of three experiments.

Figure 6

Abdominal aortic aneurysm lesion characterization in mice received white adipose tissue transplants from chow diet-fed Apoe–/– (lean white adipose tissue), high-fat diet-fed Apoe–/– (obese white adipose tissue), or ob/ob mice (ob/ob white adipose tissue). Abdominal aortic diameter (A), cross-section lesion area (B), lesion macrophage area (C), CD4+ (D) and CD8+ T-cell number (E), and medial elastica fragmentation grade (F). Mouse genotype of each cohort is indicated. The results are mean ± standard deviation of 5–10 mice/group. Abdominal aortic aneurysm lesion sections from Apoe–/– recipient mice with attached white adipose tissue transplant were used for immunofluorescent staining to detect caspase-3-positive cells (arrows) and FABP4-positive adipocytes in areas adjacent to the aortic wall (G) and in white adipose tissue transplant (H). Scale: 200 µm, inset: 70 µm. (I) ELISA determined leptin in peri-vascular adipose tissue from normal or abdominal aortic aneurysm mice. The results are mean ± standard deviation of eight mice/group.

Figure 7

Adipose tissue-induced interleukin-18, interleukin-18 receptor, and Na-Cl co-transporter expression and interleukin-18 binding to its receptors. (A) Confocal analysis of macrophage FITC-interleukin-18 binding after treatment with leptin and different white adipose tissue. (B) Representative images. Scale: 100 µm. (C) Confocal analysis of macrophage FITC-interleukin-18 binding after treatment with FABP4 and different white adipose tissue. Data in A–C were from 7–12 samples/group. Immunoblot (D/G) and quantification (E/H) of Na-Cl co-transporter, interleukin-18 receptor, and interleukin-18 in macrophages treated with and without leptin, FABP4, and different white adipose tissue. (F) RT-PCR determined Na-Cl co-transporter, interleukin-18 receptor, and interleukin-18 expression in macrophages from WT and db/db mice. (I) Immunoblot and quantification of leptin and FABP4 in different white adipose tissue. β-Actin and GAPDH blots ensured equal protein loading. (J) RT-PCR determined interleukin-18, interleukin-18 receptor, and Na-Cl co-transporter expression in peri-vascular adipose tissue adipocytes treated with or without leptin or FABP4. Data in E, H, and I were mean ± standard deviation of three experiments. Data in F and J were mean ± standard deviation of six experiments.

Take home figure

Adipocytes in the adventitia and perivascular adipose tissue (PVAT) release leptin that in turn promotes IL18-mediated activation of macrophages, T cells, smooth muscle cells (SMC), and endothelial cells (EC) during AAA development.