Author + information
- Received June 5, 2013
- Revision received August 30, 2013
- Accepted September 25, 2013
- Published online April 29, 2014.
- Yi-Chu Liao, MD, PhD∗,†,
- Yung-Song Wang, PhD‡,
- Yuh-Cherng Guo, MD, MS§,‖,
- Wen-Lien Lin, MS‡,
- Ming-Hung Chang, MD∗,† and
- Suh-Hang Hank Juo, MD, PhD‡,¶∗ ()
- ∗Section of Neurology, Taichung Veterans General Hospital, Taichung, Taiwan
- †Department of Neurology, National Yang-Ming University School of Medicine, Taipei, Taiwan
- ‡Department of Genome Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
- §Neuroscience Laboratory, Department of Neurology, China Medical University Hospital, Taichung, Taiwan
- ‖School of Medicine, Medical College, China Medical University, Taichung, Taiwan
- ¶Department of Medical Research and Department of Neurology, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan
- ↵∗Reprint requests and correspondence:
Dr. Suh-Hang H. Juo, Department of Genome Medicine, Kaohsiung Medical University, 100 TzYou First Road, Kaohsiung 80708, Taiwan.
Objectives The present study aimed to explore the role of microribonucleic acid (miRNA) Let-7g in regulating endothelial functions.
Background Derangement of miRNAs is implicated in the pathogenesis of cardiovascular diseases. Because the transforming growth factor (TGF)-β pathway plays a regulatory role in endothelial functions, miRNAs targeted at TGF-β signal cascade might affect vascular health.
Methods Bioinformatics software predicted that Let-7g can influence the TGF-β pathway by targeting 3 genes. The Let-7g's effects on multiple endothelial functions were first tested in endothelial cells (ECs) and then in apolipoprotein E knockout mice. Blood samples from lacunar stroke patients were also examined to further support Let-7g's effects on human subjects.
Results Let-7g was experimentally confirmed to knock down the THBS1, TGFBR1, and SMAD2 genes in the TGF-β pathway. PAI-I, one of the downstream effectors of the TGF-β pathway, was also down-regulated by Let-7g. Let-7g decreased EC inflammation and monocyte adhesion and increased angiogenesis via the TGF-β pathway. Furthermore, Let-7g reduced EC senescence through increasing SIRT-1 protein. Venous injection of Let-7g inhibitor into apolipoprotein E knockout mice caused overgrowth of vascular intima-media, overexpression of PAI-1, increased macrophage infiltration, and up-regulation of TGF-β downstream genes in the carotid arteries. Let-7g's beneficial effects on EC were reduced, whereas the TGF-β pathway was suppressed by ribonucleic acid interference. Restoration of the TGF-β pathway also attenuated the effects of Let-7g overexpression. Low serum levels of Let-7g were associated with increased circulating PAI-1 levels.
Conclusions Decreased Let-7g levels impair endothelial function and increase the risks of cardiovascular diseases through targeting TGF-β and SIRT-1 signaling.
The implications of microribonucleic acid (miRNA) in the pathophysiology of cardiovascular diseases have recently been recognized (1). Our group previously reported that the human miRNA Let-7g prevents atherosclerosis by inhibiting the uptake of oxidized low density lipoprotein (ox-LDL) into endothelial cells (ECs) and vascular smooth muscle cells (2). A recent study showed that patients with coronary artery disease had decreased serum levels of Let-7i (3), suggesting that the Let-7 family may play a critical role in maintaining vascular health.
The vascular endothelium is an active, dynamic tissue that is indispensable for the regulation of vascular homeostasis (4,5). The biological functions of healthy endothelium include: 1) promotion of vasodilation; 2) anticoagulation and profibrinolysis effects; 3) inhibition of leukocyte adhesion and anti-inflammatory effects; and 4) migration and angiogenesis (5). Endothelial dysfunction precipitates the atherosclerosis process and is associated with an increased risk of coronary artery disease, lacunar stroke, and cerebral small vessel disease (6–8).
The transforming growth factor (TGF)-β pathway has been recognized as playing a regulatory role in endothelial functions (9,10). The specific aim of the present study was to test whether Let-7g can regulate the TGF-β pathway and influence other mechanisms to affect endothelial functions. First, the TargetScan algorithm, which uses primarily sequence complementarity between messenger ribonucleic acids (RNAs) and the seed region of miRNA (11), was used to search for Let-7g target genes involved in the TGF-β signaling pathway. Experiments were conducted to confirm the 3 predicted target genes and to test for the consequences of EC functions after inhibiting or overexpressing Let-7g. Plasminogen activator inhibitor (PAI)-1, asymmetric dimethyl arginine (ADMA), vascular cell adhesion molecule (VCAM)-1, monocyte chemotactic protein (MCP)-1, and interleukin-6 (IL-6) were used to serve as EC biomarkers (12–14). RNA interference experiments were used to block TGF-β/SMAD signaling to further confirm that Let-7g's beneficial effects on ECs were primarily mediated by this signaling. In addition, the relationship between EC senescence and Let-7g was explored. The present study used experimental animals and human samples to verify the impact of Let-7g in vivo.
Please see the Online Appendix for detailed methods.
Construction of the 3′-untranslated region reporter plasmids
Briefly, the plasmid-containing predicted Let-7g binding site of each gene was reacted with Let-7g mimic or Let-7g inhibitor to confirm that Let-7g can directly bind to the target 3′-untranslated region (UTR).
Human umbilical endothelial cells (HUVECs) were transfected with the following miRNAs (5 nM): mimic control, inhibitor control, Let-7g mimic (MC11758), and Let-7g inhibitor (MH11758) using HiPerFect transfection reagent (Qiagen, Valencia, California).
Two additional approaches were used to further support the fact that Let-7g's effects on ECs are mediated through the TGF-β pathway: 1) restoration of the pathway by treating HUVECs with TGF-β1 (10 ng/ml, Sigma-Aldrich, St. Louis, Missouri) after transfection of Let-7g mimic for 24 h; and 2) suppression of TGF-β/SMAD2 signaling by transfecting SMAD2 small interfering RNA (siRNA) and/or TGFBR1 siRNA while transfecting Let-7g inhibitor.
Assays for endothelial functions
Detailed methods used for monocyte adhesion assay; tubule formation to assess angiogenesis; measurements of VCAM-1, MCP-1, IL-6, and PAI-1 concentrations; assay to estimate endothelial nitric oxide synthase (eNOS) activity; and the β-galactosidase (β-gal) assay for senescent status of HUVECs can be found in the Online Appendix.
See the Online Appendix.
Lentiviral expression vector pCDH-CMV-MCS-EF1-GreenPuro (SBI Mountain View, California) was used to carry Let-7g–expressing plasmids or Let-7g sponge plasmids. The inserts of Let-7g overexpressing plasmids and sponge plasmids were synthetic oligonucleotides containing EcoRI and XhoI overhangs. The empty vector was used as a negative control. Apolipoprotein E-knockout (ApoE-KO) mice were fed 4 g/day of high-fat diet (product code 57BD, Testdiet, St. Louis, Missouri) to elicit fatty streak lesions in the arteries. Lentivirus carrying different types of plasmids was used: empty or Let-7g–expressing plasmids for 12-week treatment and empty or Let-7g sponge plasmids for 9-week treatment. Undiluted viral stocks (200 to 250 μl) were injected into the tail veins of ApoE-KO mice (n = 2 for each group) weekly in the indicated time periods.
Stains for the carotid artery
Briefly, the slices of carotid arteries were stained with anti–PAI-1 antibody (1:100 dilution) or anti-pSMAD2 antibody (1:2,000 dilution), and then the stain intensity was quantified (see details in the Online Appendix). To stain macrophages, the slices were incubated with anti–Mac-3 antibody (1:100 dilution; Millipore Billerica, Massachusetts) and Dylight 649 conjugated goat antirat antibody (1:200 dilution; Rockland Gilbertsville, Pennsylvania).
Sixty patients with lacunar infarction were enrolled from the Taichung Veterans General Hospital in Taiwan. All participants provided written informed consent, and the study protocols were approved by the local Institutional Review Board (VGHTC-C10253). The patients' serum levels of Let-7g and RNA-U6B were determined by quantitative real-time polymerase chain reaction (qPCR). Plasma PAI-1 and ADMA levels were determined by enzyme-linked immunosorbent assay for patients with the lowest or highest tertile of Let-7g levels (n = 20 per group). Five persons were excluded because samples were contaminated by hemolysis and severe hyperlipidemia.
For cellular studies, variables are presented as mean ± SE from 3 independent experiments. Student t test was used to compare the variables between the treatment and control groups and the variables between stroke patients with the lowest and highest tertile of Let-7g. The data from each experiment in the cellular study can be found in the Online Appendix. Three replicates were performed in each cellular experiment, and the number of replicates was used as the unit of analysis. For human blood samples, variables are presented as mean ± SD. A 2-sided p value <0.05 was considered significant. Owing to the small sample size in cellular experiments and human studies, we did not conduct the Bonferroni correction for multiple comparisons or test for normality and homogeneity.
Confirmation of the Let-7g target genes THBS1, TGFBR1, and SMAD2
Let-7g binding sites were predicted at the 3′-UTRs of the thrombospondin 1 (THBS1), transforming growth factor beta receptor 1 (TGFBR1), and SMAD family member 2 (SMAD2) genes (Fig. 1A). Plasmids carrying either the wild-type or mutant sequence of 3′-UTR of each gene were constructed (Fig. 1A, Online Table 1) and transfected to HEK293 cells. For the wild-type sequence of THBS1, Let-7g mimic induced a dose-dependent decrease of luciferase activity, whereas Let-7g inhibitor exerted an opposite effect (Figs. 1B and 1C, left panel). For the mutant sequence, Let-7g mimic had only a mild knock-down effect, and Let-7g inhibitor paradoxically decreased luciferase activity (Figs. 1B and 1C, left panel).
Similarly, Let-7g mimic induced a stepwise decrease of luciferase activity in plasmids carrying the wild-type 3′-UTR sequence of TGFBR1 or SMAD2 but had no effect on plasmids carrying the mutant sequences (Fig. 1B, middle and right panels). On the contrary, Let-7g inhibitor caused a dose-dependent increase in luciferase activity for plasmids carrying the wild-type 3′-UTR sequences but did not have such an effect on plasmids carrying the mutant sequences (Fig. 1C, middle and right panels). The results from the pMIR-reporter assay provided direct evidence of Let-7g′s binding to the 3 target genes.
Let-7g affected expression levels of the 3 target genes
qPCR confirmed successful transfection of Let-7g mimic or inhibitor to alter the intracellular Let-7g levels (Online Fig. 1). As shown in Figure 2A, Let-7g mimic reduced THBS1 expression, whereas Let-7g inhibitor increased THBS1 expression levels (p < 0.01). Similarly, Let-7g mimic knocked down TGFBR1 and SMAD2 expression levels, whereas Let-7g inhibitor increased their transcription (p < 0.05) (Figs. 2B and 2C). In addition, Western blots illustrated the fact that Let-7g mimic significantly reduced the protein amount of total SMAD2 (p < 0.05) (Fig. 2D).
Given that TGFBR1 and SMAD2 behave as check points in the TGF-β signal cascade (15), Let-7g might interfere with the signal pathway by also influencing SMAD2 phosphorylation. To test this hypothesis, HUVECs were treated with TGF-β1, which caused a rapid but transient increase in phosphorylated SMAD2 (pSMAD2). Let-7g mimic diminished TGF-β1–induced pSMAD2, whereas Let-7g inhibitor significantly increased pSMAD2 under TGF-β1 stimulation (Fig. 2E, left panel). The pSMAD2-to-total SMAD2 ratio was used to indicate the changes in SMAD2 activity. Let-7g mimic influenced both total and pSMAD2 but did not significantly affect the pSMAD2/SMAD2 ratio (Fig. 2E, middle panel). However, Let-7g inhibitor significantly increased the ratio (Fig. 2E, right panel), which implied that loss-of-function in Let-7g can be more influential than gain-of-function in regulating the TGF-β pathway.
The immunofluorescence stain also supported the fact that Let-7g had dual effects on SMAD2. HUVECs transfected with Let-7g mimic had less SMAD2 in the nucleus (i.e., less pSMAD2) as indicated by a weaker stain in the nucleus (Fig. 2F, left panel). In contrast, Let-7g inhibitor increased SMAD2 levels in the nucleus and cytoplasm (Fig. 2F, right panel).
Let-7g inhibited cell adhesion and inflammation but increased angiogenesis
Because the TGF-β pathway has pleiotropic effects on vascular pathophysiology (9,10), we further tested the effect of Let-7g on EC functions, including monocyte adhesion, inflammation, and angiogenesis. Our results showed that Let-7g mimic significantly reduced VCAM-1 secretion (Fig. 3A), which subsequently inhibited the monocyte adhesion to HUVECs (Fig. 3B). Monocyte adhesion was reduced when intracellular Let-7g levels were increased by Let-7g mimic (p < 0.001) (Fig. 3B, middle panel). In contrast, monocyte adhesion was aggravated when HUVECs were transfected with Let-7g inhibitor (Fig. 3B, left panel). Activating the TGF-β pathway by TGF-β1 reversed the inhibitory effect of Let-7g mimic on monocyte adhesion, whereas blocking the signal cascade by SMAD2/TGFBR1 siRNAs diminished the detrimental effects of Let-7g inhibitor (Fig. 3B).
The secretion of inflammatory cytokines (MCP-1 and IL-6) was significantly reduced in HUVECs transfected with Let-7g mimic (p < 0.05) (Fig. 3C). Similarly, Let-7g's anti-inflammatory effect was reversed after restoration of the TGF-β signaling (Online Fig. 2). The proinflammatory effects of Let-7g inhibitor were reduced when the TGF-β signal was blocked (Online Fig. 3). Let-7g mimic mildly increased angiogenesis by the matrigel assay (BD biosciences, Franklin Lakes, New Jersey) (Fig. 3D, middle panel), and treatment with TGF-β1 (10 ng/ml) did not significantly reverse Let-7g's effect (Fig. 3D, right panel). In contrast, the tubule formation was significantly inhibited in HUVECs transfected with Let-7g inhibitor (p = 0.005). Suppression of the TGF-β signaling by siRNAs recovered the angiogenic ability (Fig. 3D, right panel)
Let-7g inhibited in vitro and in vivo PAI-1 expression
PAI-1 levels were used to imply the impact of Let-7g on TGF-β–regulated cardiovascular risks. In HUVECs, Let-7g mimic mildly inhibited PAI-1 expression, whereas Let-7g inhibitor significantly increased PAI-1 messenger RNA levels (Fig. 4A). The secretion of PAI-1 was significantly reduced in HUVECs transfected with Let-7g mimic compared with those transfected with the mimic control (p < 0.05) (Fig. 4B).
We then detected the amounts of PAI-1 and pSMAD2 protein in the carotid arteries of ApoE-KO mice fed a high-fat diet for 9 or 12 weeks (Fig. 4C). Lentivirus carrying Let-7g expressing plasmids was injected into the tail veins of mice each week to increase endogenous Let-7g levels, whereas lentivirus carrying Let-7g sponge plasmids reduced Let-7g levels. The treatment of Let-7g–expressing plasmids increased arterial Let-7g levels by 2.6-fold, whereas Let-7g sponge plasmids reduced arterial Let-7g levels to 48%. The immunohistochemistry stain showed reduced amounts of PAI-1 and pSMAD2 protein in mice treated with Let-7g–expressing plasmids (Fig. 4C, left panel). On the contrary, Let-7g sponge plasmids caused overgrowth of the intima-media layer and overexpression of PAI-1 and pSMAD2 in the vessel walls. Quantitative analysis of the PAI-1 and pSMAD2 staining further demonstrated that loss-of-function in Let-7g had a stronger impact than gain-of-function (Fig. 4C, right panel). The qPCR data consistently supported the findings from immunohistochemistry stain in that PAI-1 level was decreased to 28% (p < 0.001) and SMAD2 level was decreased to 44% (p < 0.001) in the Let-7g–treated mice at 12 weeks (Online Fig. 4). For the mice treated with Let-7g sponge, PAI-1 and SMAD2 levels were increased by 2.6-fold (p = 0.02) and 1.9-fold (p = 0.003) at 9 weeks, respectively (Online Fig. 4).
Let-7g inhibited monocyte infiltration and TGF-β downstream genes in vivo
In vivo, monocyte infiltration can be measured by the number of macrophages in the carotid arteries of ApoE-KO mice. Figure 4D shows that the overexpressed Let-7g could reduce macrophage infiltration and vice versa. The messenger RNA levels of THBS1, TGFBR1, VCAM-1, MCP-1, and IL-6 in carotid arteries were in concert with the findings in cellular experiments (Fig. 4E). These TGF-β downstream genes were suppressed in mice with Let-7g overexpression (p < 0.01) (Fig. 4E) and were up-regulated in mice treated with Let-7g sponge (p < 0.05) (Fig. 4E). Our findings provided in vivo support for the Let-7g effects on vascular morphology and downstream effectors of TGF-β signaling pathway.
Let-7g and lacunar stroke
Endothelial dysfunction is one of the pathogeneses underlying lacunar stroke (7,8). To determine whether Let-7g could modulate EC functions in human subjects, we first measured the serum Let-7g levels in 60 patients with lacunar infarction who were likely to have EC dysfunction. These patients, whose Let-7g levels were in the lowest or highest tertile, were selected for measurement of their plasma levels of PAI-1 and ADMA, which are biomarkers of endothelial dysfunction (13,14). The data showed that plasma PAI-1 levels were significantly higher in the patients with low Let-7g levels than in patients with high Let-7g levels (p = 0.04) (Online Table 2). However, there were no differences in plasma levels of ADMA between the patients with high and low Let-7g levels.
Let-7g prevented endothelial senescence through influencing SIRT-1
We further tested whether Let-7g could alleviate endothelial senescence. Sirtuin 1 (SIRT-1) was selected as a senescence marker, although it was not predicted as a Let-7g target gene. Transfection of Let-7g mimic did not change the messenger RNA levels of SIRT-1 in HUVECs (data not shown). However, the SIRT-1 protein levels were significantly increased in HUVECs transfected with Let-7g mimic (Fig. 5A). To determine whether Let-7g prevented EC senescence through increasing SIRT-1 expression, nicotinamide was added to the culture medium to inhibit endogenous SIRT-1 activity (Fig. 5B). Compared to the control group, Let-7g mimic significantly reduced the percentage of senescent cells. As expected, Let-7g inhibitor caused the opposite effect by increasing the percentage of senescent cells. Although SIRT-1 could increase eNOS activity (16), we did not find any differences between eNOS activity in HUVECs transfected with mimic control and that in Let-7g mimic (Fig. 5C).
The present study demonstrated that Let-7g can improve several endothelial functions including decrease of senescence, inflammation, and monocyte adhesion, and increase in angiogenesis. The beneficial effects of Let-7g may be mediated by several regulatory pathways, but the present study focused primarily on TGF-β and SIRT-1 signaling (Fig. 6). First, we proved that Let-7g directly regulated 3 target genes (THBS1, TGFBR1, and SMAD2) in TGF-β signaling. Second, Let-7g was shown to indirectly increase SIRT-1 expression to prevent EC senescence. Third, the interventional study using ApoE-KO mice and the observational study using clinical human samples further verified Let-7g's beneficial effects on endothelial functions. Although miRNAs have been extensively investigated in cancers, their roles and mechanisms in the cardiovascular system are still underexplored. Let-7 family has been heavily studied in different cancers (17), but its role in vascular biology remains unclear. The strengths/merits of the present study are to combine an array of genome and miRNA techniques along with immunohistochemical staining and microscopy images. We used both gain- and loss-of-function of Let-7g to validate its important roles in EC functions. Moreover, RNA interference experiments were conducted to block TGF-β/SMAD signaling to further confirm that Let-7g's beneficial effects on EC were primarily mediated by the TGF-β signaling. The consistent results from cellular experiments, animal models, and human data supported Let-7g as an important regulator to maintain the healthy vascular endothelium.
The present study demonstrated that Let-7g's functions in keeping endothelial homeostasis were mediated primarily through the TGF-β pathway. THBS1 functions as a major activator of TGF-β pathway by converting the latent TGF-β procytokine to its biologically active form (18). THBS1-activated TGF-β pathway can increase the expression of adhesion molecules and promote monocyte adhesion to ECs (19). Our data showed that knock down of THBS1 as well as 2 downstream genes (TGFBR1 and SMAD2) by Let-7g led to a decrease of monocyte adhesion in vivo and in vitro. THBS1 is a potent inhibitor of angiogenesis (20). Indeed, we showed that enhancing the signal transduction of the THBS1/TGFBR1/SMAD2 pathway by Let-7g inhibitor disturbed the EC activation during angiogenesis.
Our animal studies found that a decrease of arterial Let-7g levels resulted in up-regulation of pSMAD2 and PAI-1 and accelerated intima-media overgrowth in the carotid arteries. These findings were in line with reports of increased neointimal formation in the carotid arteries by overexpressed TGF-β1 (21). A recent study reported that pSMAD2 was expressed mainly in ECs covering atherosclerotic plaques (22), which is consistent with our findings from animal models (Online Fig. 5).
It should be noted that our results indicated that loss-of-function in Let-7g can be more influential than gain-of-function in regulating TGF-β pathway and cell senescence. Given that the Let-7 family plays critical roles in cell differentiation (23), normal cells, including ECs, have high levels of endogenous Let-7. Therefore, the supply of exogenous Let-7g may not affect cellular functions as substantially as the supply of Let-7g inhibitor. These observations suggested that more attention should be paid to a human subject with decreased Let-7g levels.
Lacunar stroke accounts for one-fourth of all ischemic stroke events. Lacunar stroke is caused by occlusion of the small and deep perforating (lenticulostriate) arteries. The pathogenic changes in these types of arteries have been referred to as “lipohyalinosis” or “fibrinoid necrosis” (24), but the exact initiating cause of these vascular abnormalities remains unknown. Endothelial dysfunction can be an underlying cause of lacunar stroke (7,8). Among our lacunar stroke patients, higher PAI-1 levels were associated with lower serum levels of Let-7g. Elevated plasma levels of PAI-1 have been documented in patients with acute ischemic stroke (14). The present study elucidated how Let-7g can regulate multiple EC functions; therefore, decreased levels of Let-7g may be a risk factor for lacunar infarction due to its involvement in endothelial functions.
Endothelial dysfunction can be attributed partially to EC senescence. SIRT-1 has been implicated in antiaging and anti-inflammation (25) and also may regulate EC functions through other downstream targets (26). Although studies have implied Let-7's effect on antiaging process (27), our finding of up-regulation of SIRT-1 by Let-7g has not been reported before. It is likely that Let-7g is an indirect regulator for SIRT-1 because there is no Let-7g binding site at the 3′-UTR of SIRT-1. The MetaCore network prediction identified several possible pathways to account for Let-7g's effect on SIRT-1 (Online Fig. 6, Online Appendix). For example, Let-7g could inhibit Fas, which activates the death domain association protein (DAXX). DAXX is a negative regulator of tumor protein p53, which increases SIRT-1 expression via 2 validated binding sites at the SIRT-1 promoter (28). Further studies are warranted to depict Let-7g relationship with the complex SIRT-1 signaling.
First, we did not test for the potential pathways linking let-7g and SIRT-1. Second, we cannot deduce let-7g's effect toward all stroke subtypes since only patients with lacunar infarction were included in the present study. Third, let-7a, let-7b, and let-7c also have been related to stroke; therefore, other members of let-7g family could also affect EC functions.
Our cellular, animal, and human studies confirmed that the regulation of TGF-β signaling and SIRT-1 expression by Let-7g leads to beneficial effects on EC. The present study provides new evidence of the involvement of Let-7g in multiple endothelial functions. Given that healthy ECs are crucial to maintain normal functions of small arterioles and to prevent atherosclerosis in large and medium-sized arteries, maintenance of a sufficient Let-7g level may reduce the risk of cardiovascular diseases.
For an expanded Materials and Methods section as well as supplemental tables and figures, please see the online version of this article.
This work was supported by funding from the National Science Council (Taiwan, R.O.C. NSC 100-2923-B-037-001-MY3, 101-2314-B-075A-013, 101-2628-B-075A-001-MY2, and 101-2628-B-037-003-MY2), National Health Research Institutes (Taiwan, R.O.C. NHRI-Ex101-10107PI), Academia Sinica (BM102021169) Taichung Veterans General Hospital (TCVGH-1013402C and TCVGH-1013404D) and Kaohsiung Medical University Hospital grant (KMUH102-2T02). All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- asymmetric dimethyl arginine
- endothelial cell
- human umbilical endothelial cell
- monocyte chemotactic protein
- microribonucleic acid
- plasminogen activator inhibitor
- quantitative real-time polymerase chain reaction
- SMAD family member 2
- transforming growth factor
- vascular cell adhesion molecule
- Received June 5, 2013.
- Revision received August 30, 2013.
- Accepted September 25, 2013.
- American College of Cardiology Foundation
- Chen K.C.,
- Hsieh I.C.,
- Hsi E.,
- et al.
- Bonetti P.O.,
- Lerman L.O.,
- Lerman A.
- Hassan A.,
- Hunt B.J.,
- O'Sullivan M.,
- et al.
- Bobik A.
- Badimon L.,
- Romero J.C.,
- Cubedo J.,
- Borrell-Pages M.
- Tjarnlund-Wolf A.,
- Brogren H.,
- Lo E.H.,
- Wang X.
- Mattagajasingh I.,
- Kim C.S.,
- Naqvi A.,
- et al.
- Boyerinas B.,
- Park S.M.,
- Hau A.,
- Murmann A.E.,
- Peter M.E.
- Narizhneva N.V.,
- Razorenova O.V.,
- Podrez E.A.,
- et al.
- Otsuka G.,
- Agah R.,
- Frutkin A.D.,
- Wight T.N.,
- Dichek D.A.
- Nemoto S.,
- Fergusson M.M.,
- Finkel T.