miR-24 limits aortic vascular inflammation and murine abdominal aneurysm development

Abstract

Identification and treatment of abdominal aortic aneurysm (AAA) remain among the most prominent challenges in vascular medicine. MicroRNAs (miRNAs) are crucial regulators of cardiovascular pathology and represent intriguing targets to limit AAA expansion. Here we show, by using two established murine models of AAA disease along with human aortic tissue and plasma analysis, that miR-24 is a key regulator of vascular inflammation and AAA pathology. In vivo and in vitro studies reveal chitinase 3-like 1 (Chi3l1) to be a major target and effector under the control of miR-24, regulating cytokine synthesis in macrophages as well as their survival, promoting aortic smooth muscle cell migration and cytokine production, and stimulating adhesion molecule expression in vascular endothelial cells. We further show that modulation of miR-24 alters AAA progression in animal models, and that miR-24 and CHI3L1 represent novel plasma biomarkers of AAA disease progression in humans.

Figure 1. miRNAs in mouse AAA.

(a) Abdominal aortic diameter (AAD; in % versus baseline) expansion in porcine-pancreatic-elastase (PPE)-induced AAA compared with sham-operated mice. (b) Significantly downregulated aortic miRNAs (miR-23b-24-27b family highlighted), 7 days after PPE induction, compared with sham. (c) DIANA-mirExTra histogram corresponding to (b), ranking downregulated miRNAs by −ln (P value): the probability that their target hexamers are found more often in 3′UTRs of differentially upregulated mRNAs than in unchanged genes by Wilcoxon Rank Sum Test (miR-23b-24-27b family highlighted). (d) miR-24 is the only miR-23b-24-27b family member significantly downregulated in aortic tissue at all three different time points during PPE-induced AAA expansion. (e) Aortic pri-miR-24 expression at 3 and 7 days after PPE induction of AAA compared with sham. (f) ISH for miR-24 (purple chromagen) in untreated (control) aorta, sham and PPE at 14 days after AAA induction (scale bar, 400 μm). (g,h) Expression of the most significant upregulated (day 7) miR-24 target genes at all tested time points during PPE-induced AAA expansion (versus sham). n=5–8 for each group and time point. Data are mean±s.e.m. *P<0.05 versus sham analysed with analysis of variance (ANOVA; one-way repeated measures ANOVA for Fig. 1a) and Bonferroni’s post test.

Figure 2. miR-24 expression and downstream effects in angiotensin II-induced AAAs and in vitro.

(a) Abdominal aortic diameter (AAD; in % versus baseline) expansion in angiotensin II (ANGII)-induced AAA compared with saline-infused control mice. (b) miR-23b-24-27b supra-renal aortic tissue expression in ANGII compared with sham (saline control). (c) Aortic Chi3l1 expression in ANGII compared with sham. (d) miR-24 (purple chromagen) is co-localized with Chi3l1 protein (brown) in PPE-induced AAA, 7 days after AAA induction (‘L’ indicates the luminal side). High-magnification inset shows co-stained macrophage. (e) miR-24 expression in peritoneal macrophages and RAW 264.7 cells, stimulated with interleukin-6 (IL-6) for either 12 or 24 h (versus untreated). (f) pri-miR-24 expression in IL-6-stimulated RAW 264.7 cells (24 h). (g) Chi3l1 expression in IL-6-stimulated and miR-24-modulated RAW 264.7 cells (at 12 and 24 h). (h) Cytokine gene fold-change in IL-6-stimulated and anti-miR-24±siChi3l1-transfected RAW 264.7 s (24 h). (i) miR-24 expression in IL-6-stimulated human aortic SMCs (24 h). n=5–9 for each group and time point. Data are mean±s.e.m. *P<0.05 versus sham (saline controls). ^#P<0.05 versus untreated/control and other treatment groups analysed with analysis of variance (ANOVA; one-way repeated measures ANOVA) and Bonferroni’s post test or Student’s t-test (two-tailed; Fig. 2i).

Figure 3. Regulation and modulation of miR-24 in vitro.

(a) CHI3L1 expression in miR-24-modulated and IL-6-stimulated human aortic SMCs. (b) Cytokine expression in CHI3L1-treated human aortic SMC. Data are mean±s.e.m. (c) Cytokine expression in IL-6-stimulated, miR-24-modulated and siChi3l1-transfected hASMC. (d) Apoptosis in IL-6-stimulated and miR-24-modulated peritoneal mouse macrophages from mice with ANGII-induced AAA. (e) Apoptosis in IL-6-stimulated and miR-24-modulated RAW 264.7 cells. (f) Apoptosis in IL-6-stimulated and miR-24-modulated peritoneal mouse macrophages from mice without ANGII-induced AAA. (g) Percent apoptosis in Chi3l1-treated RAW 264.7 cells. (h) Inhibition of NF-κB via silencing of its components (RelA or Nfkb1) prevents miR-24 downregulation in IL-6-stimulated peritoneal macrophages. n=3–4 for each treatment and cell type; all experiments were repeated three times. Data are mean±s.e.m. *P<0.05 versus control, untreated or untreated/scr-miR. ^#P<0.05 versus untreated/control and other treatment groups. ^P<0.05 versus sham, scr-miR and anti-24, analysed with analysis of variance and Bonferroni’s post test.

Figure 4. miR-24 modulation in vivo.

(a) Chi3l1 aortic expression at different time points in scr-miR and anti-/pre-24-transduced mice in sham or PPE-induced AAA at various time points. (b) Abdominal aortic diameter (AAD; in % versus baseline) time course in scr-miR and anti-/pre-24 PPE-induced AAA. (c) IHC for aortic Chi3l1 (brown chromagen) in anti-/pre-24-treated mice compared with scr-miR, 14 days after PPE induction of AAA. (d) Chi3l1-positive cell counts (n=4 high-power fields (HPFs) on three different sections of three different aortas per group) in miR-24-modulated mice (versus scr-miR control), 14 days after PPE–AAA induction. (e) Representative IHC images (antibody against murine Chi3l1) in different-sized aneurysms of miR-24-modulated mice (pre-/anti-24) compared with sham and scr-miR (scale bar, 50 μm), 28 days after PPE–AAA induction. Data are mean±s.e.m. *P<0.05 versus sham. ^#P<0.05 versus scr-miR and sham. ^P<0.05 versus sham, scr-miR and anti-24; analysed with analysis of variance (ANOVA; one-way repeated measures ANOVA for Fig. 4b and Bonferroni’s post test.

Figure 5. Effects of miR-24 modulation in vivo.

(a) IHC for CD11b/c in anti-/pre-24-treated mice compared with scr-miR and sham, 14 days after PPE induction of AAA (brown chromagen; scale bar, 400 μm). (b) CD11b/c-positive cell counts (n=4 high-power fields on three different sections of three different aortas per group) in miR-24-modulated mice, 14 days after PPE–AAA induction (versus scr-miR and sham). (c) IHC for SMC α-actin (SMA) in anti-/pre-24-treated mice compared with scr-miR and sham, 14 days after PPE induction of AAA (brown chromagen; scale bar, 400 μm). (d) SMA-positive cells (n=4 high-power fields on three different sections of three different aortas per group) in miR-24-modulated mice, 14 days after PPE–AAA induction (versus scr-miR and sham). (e) Confocal microscopy illustrating co-localization and decreased levels of F4/80 and Chi3l1 in pre-24-transfected mice with PPE–AAA (at 14 days) compared with sham (scale bar, 50 μm), scr-miR- and anti-24-transfected mice (‘L’ indicates the luminal side; scale bars, 400 μm). n=4–6 mice for each group. Data are mean±s.e.m. ^#P<0.05 versus scr-miR and sham. *P<0.05 versus sham and anti-24.

Figure 6. miR-24 modulation in ANGII-induced AAA and levels in human AAA.

(a) Abdominal aortic diameter (AAD; in % versus baseline) in scr-miR and anti-/pre-24 ANGII-induced AAA (n=4–6 mice for each time point and group). (b) Chi3l1 aortic expression in scr-miR- and anti-/pre-24-transfected mice compared with sham in the ANGII–AAA model at various time points. (c) miR-23b-24-27b expression in human aortic tissue samples from patients with small AAA (n=12) and large AAA (n=10), compared with tissue from control patients without AAA (n=14). (d) CHI3L1 is significantly upregulated in human tissue samples with large AAA compared with small AAA and non-aneurysmal aortic tissue. (e) miR-24 is significantly downregulated in plasma samples from human patients with AAA (n=43 with small AAA; n= 39 with large AAA) compared with control plasma (n=44), and patients with peripheral vascular occlusive disease (PVOD) but no AAA (n=22). (f) CHI3L1 plasma levels are significantly increased in patients with large AAA compared with patients with small AAA, PVOD and un-diseased controls. Data are mean±s.e.m. *P<0.05 versus sham/controls. ^#P<0.05 versus scr-miR/controls/small AAA. ^P<0.05 versus pre-/anti-24 or versus controls/PVOD; analysed with analysis of variance (ANOVA; one-way repeated measures ANOVA for Fig. 6a) and Bonferroni’s post test.