The knockout of miR-143 and -145 alters smooth muscle cell maintenance and vascular homeostasis in mice: correlates with human disease

Abstract

Mechanisms controlling vascular smooth muscle cell (VSMC) plasticity and renewal still remain to be elucidated completely. A class of small RNAs called microRNAs (miRs) regulate gene expression at the post-transcriptional level. Here, we show a critical role of the miR-143/145 cluster in SMC differentiation and vascular pathogenesis, also through the generation of a mouse model of miR-143 and -145 knockout (KO). We determined that the expression of miR-143 and -145 is decreased in acute and chronic vascular stress (transverse aortic constriction and in aortas of the ApoE KO mouse). In human aortic aneurysms, the expression of miR-143 and -145 was significantly decreased compared with control aortas. In addition, overexpression of miR-143 and -145 decreased neointimal formation in a rat model of acute vascular injury. An in-depth analysis of the miR-143/145 KO mouse model showed that this miR cluster is expressed mostly in the SMC compartment, both during development and postnatally, in vessels and SMC-containing organs. Loss of miR-143 and miR-145 expression induces structural modifications of the aorta, because of an incomplete differentiation of VSMCs. In conclusion, our results show that the miR-143/145 gene cluster has a critical role during SMC differentiation and strongly suggest its involvement in the reversion of the VSMC differentiation phenotype that occurs during vascular disease.

Figure 1. Expression of miR-143 and miR-145

(a) Localization of miR-143 and miR-145 in the genome of the mouse. (b) miR-143 and -145 sequence alignment in different species. (c) Representative Northern blot of miR-143 and miR-145 in different tissues (top); band intensities were quantified using ImageJ software version 1.34 (http://rsb.info.nih.gov/ij/) and normalized to U6 (bottom). (d) Radioactive in situ hybridization for miR-143, -145 and -208 on adult mouse heart. Hearts were pseudo-colored in red to enhance the contrast using Adobe Photoshop.

Figure 2. Expression of miR-143 and miR-145 in vascular diseases

(a) qRT-PCR of samples of aorta from sham and pressure-overloaded mice generated by aortic constriction (TAC); Pre, portion of aorta proximal to the constriction; Post, portion of an aorta distal to the constriction; Sno22 RNA was used as internal control; All measurements were calculated as percent of control (Sham) and error bars calculated as propagated standard errors of the mean of triplicate measurements from each experiment; *, p<0.05. (b) Expression of miR-143 and miR-145 in the aorta of apolipoprotein E knockout (ApoE KO) mouse fed on normal (ND) and high-cholesterol diets (HFD), analyzed by qRT-PCR, Sno22 RNA was used as internal control; *, p<0.03. (c) qRT-PCR of expression of miR-143, -145 and -199 in aortic aneurysm in humans, U6 snRNA was used as internal control; *, p<0.05.

Figure 3. Targeting of miR-143(145)

(a) Strategy used to generate miR-143 mutant mice by homologous recombination. (b) Southern blot of genomic DNA from ES cells. (c) Northern blot of miR-143 and miR-145 in aortas of mice wild-type (+/+), heterozygous (+/−) and homozygous (−/−) for miR-143/145 deletion. U6 RNA serves as a loading control. (d) Visualization of lacZ-expressing structures in an embryo at E13.5. Arrows indicate β-galactosidase staining of aorta and bladder. (e) lacZ expression on paraffin sections of embryo at E13.5. (f) lacZ expression on whole heart and brain of P1 mice. (g) Immunofluorescence for smooth muscle α-actin and β-gal on a frozen section of P1 heart. DAPI was used to visualize nuclei. Scale Bar = 20μm.

Figure 4. Morphological changes in the vasculature of miR-143(145) KO mice

(a) Van Gieson staining for elastin in aorta from miR-143(145) knockout (KO) and wild-type (WT) mice (top), and analysis of the arterial thickening (bottom) through calculation of the media-area/lumen-area ratio. Scale Bar = 40μm. (b) Masson trichrome (whole section and top inset) and smooth muscle α-actin immunofluorescence staining (bottom inset) of a cross-section of aorta from a miR-143(145) KO mouse. Original magnification ×10 (whole section), ×40 (insets). (c) Proliferation assay of FBS-treated A7r5 cells. (No); no adenovirus; (−), Ad-Empty, (143) AdmiR-143, (145) Ad-miR-145, (208) Ad-miR-208; *, p<0.05. (d) Chemotaxis assay of primary VSMCs from WT and KO mice; All measurements were calculated as percent of control (WT) and error bars calculated as propagated standard errors of the mean of triplicate measurements from each experiment; *, p<0.05.

Figure 5. Electron microscopy on mouse aorta

(a) Electron microscopy micrographs of miR-143(145) KO (KO) and wild-type (WT) aorta. M, mitochondria; RER, rough endoplasmic reticulum; SER, smooth endoplasmic reticulum; N, nuclei. Arrows indicate aligned ribosomes present on active RER, whereas they are organized in clusters all around the SER in WT aorta. Scale Bar = 1000nm. (b) Micrograph of WT and miR-143/145 KO aorta (top panel), higher magnification image of the KO aorta on the bottom panel. N, nucleus; BL, basal lamina; IL, intima layer. Arrows indicate areas of ECM around a KO cell which shows cell protrusions. Scale Bar = 5μm.

Figure 6. Functional changes in the vasculature of miR-143(145) KO mice

(a) Blood pressure of saline- (Sham) and angiotensin II (AngII)-treated WT and miR-143/145 KO mice after 2 weeks; *, P < 0.03. (b) qRT-PCR on adult aorta from WT and KO mice for smooth muscle myosin heavy chain (Myh11) and smooth muscle α-actin (Acta2). GAPDH RNA was used as internal control; data were plotted and analyzed as described in Fig. 2; *, P<0.05.

Figure 7. Effects of miR overexpression in rat carotid arteries after balloon injury

Representative cross sections (14 days after balloon injury in vessels transduced with different miR expressing adenoviral vectors) (top panel). Neointimal area and neointimal/media ratio quantifications (bottom panel); *, P<0.05 vs contralateral untreated arteries.