RETRACTED ARTICLE: Down-regulation of microRNA-342-5p or Up-regulation of Wnt3a Inhibits Angiogenesis and Maintains Atherosclerotic Plaque Stability in Atherosclerosis Mice

Evidence has demonstrated that microRNA-342-5p (miR-342-5p) is implicated in atherosclerosis (AS), but little is known regarding its intrinsic regulatory mechanisms. Here, we aimed to explore the effect of miR-342-5p targeting Wnt3a on formation of vulnerable plaques and angiogenesis of AS. ApoE−/− mice were fed with high-fat feed for 16 w to replicate the AS vulnerable plaque model. miR-342-5p and Wnt3a expression in aortic tissues of AS were detected. The target relationship between miR-342-5p and Wnt3a was verified. Moreover, ApoE−/− mice were injected with miR-342-5p antagomir and overexpression-Wnt3a vector to test their functions in serum lipid levels, inflammatory and oxidative stress-related cytokines, aortic plaque stability and angiogenesis in plaque of AS mice. miR-342-5p expression was enhanced and Wnt3a expression was degraded in aortic tissues of AS mice and miR-342-5p directly targeted Wnt3a. Up-regulating Wnt3a or down-regulating miR-342-5p reduced blood lipid content, inflammatory and oxidative stress levels, the vulnerability of aortic tissue plaque and inhibited angiogenesis in aortic plaque of AS mice. Functional studies show that depleting miR-342-5p can stabilize aortic tissue plaque and reduce angiogenesis in plaque in AS mice via restoring Wnt3a. Supplementary Information The online version contains supplementary material available at 10.1186/s11671-021-03608-w.


Introduction
Atherosclerosis (AS) is an age-related artery disease featured by the thickening, stenosis, hardening and the formation of atherosclerotic plaques of the arteries [1]. It is the prevailing cause of death and morbidity in developed countries [2]. Histopathologic studies of human atherosclerotic lesions have shown that plaque development and rupture are featured by lipid/necrotic core expansion, reduction in smooth muscle cell number, macrophage infiltration and decreased collagen content [3]. The key cellular elements of AS comprise of hyperlipidemia, foam cell formation, differentiation into macrophages, monocyte recruitment and induced inflammation [4]. Although many medicines for the treatment of AS have been widely used in the clinic, some subgroups of patients are still at high risk of myocardial infarction, myocardial ischemia, heart failure and stroke [5]. Therefore, further exploration of potential molecular mechanisms can offer more evidence for AS treatment.
A single microRNA (miRNA) can simultaneously regulate several gene targets [6]. miR-342-5p is investigated to be contained in the imprinted 14q32 miRNA cluster, acts as an innovative Notch downstream molecule [7] and modulates multiple angiogenic pathways, such as transforming growth factor β signaling and Open Access

A R T I C L E
Page 2 of 13 Sun et al. Nanoscale Research Letters (2021) 16:165 vascular endothelial growth factor [8]. Related immunomodulatory microRNAs, such as miR-342-5p, have multiple important roles in regulating the progression of atherosclerosis [9]. Moreover, some miRNAs have been suggested to involve in the resolution of AS, such as miR-155 and miR-217 [10,11]. A study has reported that miR-342-5p acts as a new modulator for macrophage activation in AS [12]. Another study has revealed that macrophage-derived miR-342-5p facilitates AS and increases the inflammatory stimulation of macrophages [13]. Qu et al. have found that Wnt3a expression is negatively modulated by miR-342-5p in anorectal malformations [14], indicating that there is a target relationship between miR-342-5p and Wnt3a. Wnt signaling serves an essential role during embryogenesis for the modulation of cell polarity, cell proliferation, axis formation and apoptosis [15]. Wnt3a, a key component of the mesoderm gene, plays a crucial role in embryonic development [16]. It has been presented that epigenome-guided analysis of the plaque macrophages transcriptome during AS regression uncovers activation of the Wnt signaling pathway [17]. Moreover, a study has reported that Wnt3a modulates adhesion and migration of vascular smooth muscle cells which contribute to the pathogenesis of AS and restenosis [18]. Therefore, this study for the first time explored the effect of miR-342-5p targeted Wnt3a on formation of vulnerable plaques and angiogenesis of AS.

Ethics Statement
Animals were treated humanely using approved procedures in compliance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Institutional Animal Care and Use Committee of Qinghai Provincial People's Hospital (ethical number: 201870726).

Experimental Animals
Male ApoE −/− mice and C57BL/6J mice (specific pathogen-free grade) aged 8 weeks were available from Beijing Vital Laboratory Animal Technology (Beijing, China). Mice (5-6 mice in a cage) were housed with 12 h/12 h day/night cycle with ad libitum access to food and water.

Establishment of Mice Models of AS
ApoE −/− mice were fed with high-fat feed for 16 w to establish the AS vulnerable plaque model. C57BL/6 J mice were used as the normal group with natural drink and food. ApoE −/− mice had glossy light hair and hair shedding in the back after 12 w. The aortic arch and brachiocephalic artery from 3 model mice were dissected to perform hematoxylin-eosin (HE) staining, and there was no significant plaque deposits on the intima. Another 3 modeled mice were identified again after 4 w, and HE staining showed that there was obvious plaque deposits on the intima of aorta arch, indicating success of the model establishment.

Sample Collection and Treatment
Before sampling, mice were fasted for 12 h and anesthetized by ether inhalation and blood samples were collected from eyeballs. The chest of mice was opened, the thoracic aorta was dissociated to the end of the abdominal aorta and the entire vessel was removed. After cleaned by RNA-free phosphate-buffered saline (PBS), the tissues were embedded for HE staining, oil red O, Sirius red staining and immunohistochemical staining. Some of the vascular tissues were preserved at − 80 °C for RT-qPCR and Western blot.

Blood Lipid Level Detection
The automatic biochemical analyzer (Roche, Basel, Switzerland) was adopted to detect total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) in serum. The detection was implemented following the specification of the kits (NanJing JianCheng Bioengineering Institute, Nanjing, China).

Enzyme-Linked Immunosorbent Assay (ELISA)
Determination of serum cytokine content: commercial interleukin (IL)-5, IL-12p70, tumor necrosis factor alpha (TNF-α) and interferon (IFN)-γ ELISA kits were used. Finally, the optical density (OD) value of each well was tested by a microplate reader at 450 nm. Determination of oxidative stress injury: malondialdeyde (MDA) content and superoxide dismutase (SOD) activity in serum were tested by MDA kit (OD value was tested by spectrophotometer at 532 nm) and SOD kit (OD value was determined by a microplate reader at 450 nm). IL-5, IL-12p70, TNF-α, IFN-γ, MDA and SOD ELISA kits were purchased from MultiSciences (Lianke) Biotechnology Corporate Limited (Hangzhou, Zhejiang, China).

HE Staining, Oil Red O Staining and Sirius Red Staining
After fixation and embedding, the specimens were sliced into consecutive sections of 4-microm thickness. Slices were dewaxed and hydrated, stained by hematoxylin and eosin, differentiated, dehydrated, cleared by xylene, dried and sealed by neutral gum. The nucleus was blue, and other tissues such as cytoplasm and connective tissues were red in different shades. The plaque formation was observed by a fluorescence microscope. The arterial walls of HE-stained sections were selected under the microscope, and the experimental results were collected by a digital camera. The Image Pro Plus6.0 (IPP6.0) image analysis software module was utilized to calculate the plaque area of cross section of each slices and the area of wall as well as their ratio.
Slices with 4-5 μm were chosen for oil red O staining. The slices were dried with excess temperature for 20 min and incubated with 100% isopropanol for 5 min. Then the slices were incubated with 0.5% oil red O staining solution in a 60 °C oven for 8 min, washed in 85% isopropanol for 3 min, dyed with hematoxylin for 1 min, cleaned and sealed. The results of oil red O staining suggested that the lipid was red or orange and the nucleus was light blue. IPP6.0 software was used to reckon the fat area and plaque area in tissue slice's plaque. The lipid content = oil red O positive staining area/plaque area × 100%.
Sirius red staining: slices were dewaxed and hydrated, dyed for 10 min with celestine blue staining solution, with Sirius red staining solution for 20 min and counterstained for 10 min with hematoxylin. Finally, slices were dehydrated by gradient ethanol, cleared by xylene and sealed with neutral gum. The collagen area in tissue slice's plaque was computed by IPP6.0 software. The collagen area = Sirius red positive staining area/plaque area × 100%. The percentage of lipid and collagen in the plaque area was calculated.
Hematoxylin, eosin and Sirius dye were available from China Pharmaceutical Group Shanghai Chemical Reagent Co. Ltd. (Shanghai, China). The oil red O powder was bought from Sigma-Aldrich Chemical Company (St Louis, MO, USA).

Immunohistochemical Staining
Slices of 4-5 μm were placed on the slides coated with 100 mg/L polylysine and fixed by acetone. Endogenous peroxidase was blocked by bovine serum albumin. The tissues were dripped with MOMA-2 antibody

Statistical Analysis
All data were interpreted by SPSS 21.0 software (IBM Corp. Armonk, NY, USA). Measurement data were indicated as mean ± standard deviation. Disparities between two groups were formulated by t-test, while those among multiple groups by one-way analysis of variance (ANOVA) followed by Tukey's multiple comparisons test. Statistical significance was established by P value < 0.05.

miR-342-5p Increases and Wnt3a Decreases in Aortic Tissues of ApoE −/− mice and miR-342-5p Directly Targets Wnt3a
MicroRNA (miRNA) target genes are associated with atherosclerosis-related functions. miR-342-5p, Wnt3a and β-catenin were tested in aortic tissues of AS model mice by RT-qPCR and Western blot assay. It was revealed that in relation to the normal group, miR-342-5p was raised while Wnt3a and β-catenin were decreased in the AS group (both P < 0.05). In comparison with the NC group, miR-342-5p was enhanced as well as Wnt3a and β-catenin were decreased in the miR-342-5p agomir group (both P < 0.05), while miR-342-5p was decreased, Wnt3a and β-catenin were heightened in the miR-342-5p antagomir group (both P < 0.05). Wnt3a and β-catenin expression were heightened in the oe-Wnt3a group relative to the NC group (both P < 0.05). Compared to the miR-342-5p agomir group, Wnt3a and β-catenin expression were raised in the miR-342-5p agomir + oe-Wnt3a group (P < 0.05) (Fig. 1A-D). In addition, in the preliminary experiment, β-catenin expression under the treatment of different concentrations of miR-342-5p agomir, miR-342-5p antagomir and oe-Wnt3a was tested and the results showed (Additional file 1: Fig. S1) the higher the miR-342-5p agomir concentration, the lower the β-catenin expression; the higher the miR-342-5p antagomir concentration, the higher the β-catenin expression; and the higher the oe-Wnt3a concentration, the higher the β-catenin expression. miRNAs could inhibit the translation of specific genes by binding to their messenger RNA 3′UTR. Bioinformatics website predicted that there was a target relationship between miR-342-5p and Wnt3a (Fig. 1E). Dual luciferase reporter gene assay reported that in Wnt3a 3′UTR vector-transfected mouse aortic endothelial cells, the renilla/firefly value of luciferase was dose-dependently decreased by transfection with miR-342-5p agomir, with a significant decrease from 10 to 100 pM miR-342-5p agomir and a 64% decrease occurred at 100 pM miR-342-5p agomir group when compared with the NC group. This indicated the presence of a miR-342-5p target site in the Wnt3a 3′UTR. However, the renilla/firefly value of luciferase activity was not affected in the Wnt3a mutation group (Fig. 1F, G). Thus, it could be confirmed that Wnt3a was a direct target gene of miR-342-5p, and miR-342-5p/Wnt3a could regulate AS progression. lipid levels of AS mice, automatic biochemical analyzer was utilized to observe the change of lipid levels. The results revealed that ( Fig. 2A-D) in contrast with the normal group, TC, TG and LDL-C contents were raised and HDL-C content was decreased in the AS group (all P < 0.05). Versus the NC group, TC, TG and LDL-C contents were heightened and HDL-C content was decreased in the miR-342-5p agomir group (all P < 0.05), while TC,  N = 3). G Relative activity of luciferase in cells with wild-type and mutant Wnt3a 3′-UTR after transfected with miR-342-5p agomir or scramble (N = 3). *P < 0.05 vs. the normal group, # P < 0.05 vs. the NC group. & P < 0.05 vs. the miR-342-5p agomir group. Measurement data were indicated as mean ± standard deviation. Comparisons between two groups were formulated by t-test, while comparisons among multiple groups were assessed by one-way ANOVA followed by Tukey's multiple comparisons test. AS, atherosclerosis; NC, negative control TG and LDL-C contents were decreased and HDL-C content was augmented in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05). In relation to the miR-342-5p agomir group, TC, TG and LDL-C contents were reduced and HDL-C content was raised in the miR-342-5p agomir + oe-Wnt3a group (all P < 0.05). These results suggested that miR-342-5p and Wnt3a regulated blood lipid level of AS mice, and further illustrated the targeted regulation relationship between miR-342-5p and Wnt3a. Overexpression of Wnt3a would reverse overexpressed miR-342-5p-induced effects on AS mice.

Effects of Overexpression of Wnt3a or Low Expression of miR-342-5p on Inflammatory and Oxidative Stress-Related Cytokines in Serum of ApoE −/− Mice
Then, the contents of cytokines in serum of AS mice were tested by ELISA, and the results reported that (Fig. 3A-D) versus the normal group, IL-5, IL-12p70, IFN-γ and TNF-α were enhanced in the AS group (all P < 0.05). Versus the NC group, IL-5, IL-12p70, IFN-γ and TNF-α contents were raised in the miR-342-5p agomir group (all P < 0.05), while IL-5, IL-12p70, IFN-γ and TNF-α contents were degraded in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05). In contrast to the miR-342-5p agomir group, IL-5, IL-12p70, IFN-γ and TNF-α contents were decreased in the miR-342-5p agomir + oe-Wnt3a group (all P < 0.05). It was hinted that miR-342-5p targeted regulation of Wnt3a signaling pathway further regulated the level of related cytokines in serum of AS mice. Furthermore, MDA content and SOD activity in serum of mice were tested, and it was revealed that (Fig. 3E, F) in comparison with the normal group, MDA content was raised and SOD activity was depressed in the AS group (both P < 0.05). Versus the NC group, MDA content was enhanced and SOD activity was decreased in the  miR-342-5p agomir group (both P < 0.05), while MDA content was decreased and SOD activity was increased in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05). Versus the miR-342-5p agomir group, MDA content was depressed and SOD activity was enhanced in the miR-342-5p agomir + oe-Wnt3a group (both P < 0.05). Therefore, a summary was obtained that depleting miR-342-5p and restoring Wnt3a inhibited oxidative stress in AS mice.

Effects of Depletion of miR-342-5p or Restoration of Wnt3a on Lipid and Collagen Content in Aortic Plaque of ApoE −/− Mice
In order to explore the effect of miR-342-5p targeted Wnt3a on plaque area in aortic tissue of mice, HE staining was performed and the results revealed that (Fig. 4A, B) except for the normal group, AS plaques formed in all sections of the other groups. In the AS group, the plaque area was large, the fibrous cap was thinner, the lipid core was enlarged, more foam cells and cholesterol crystal precipitation appeared in the plaque, the inner wall of artery and muscle layer were thickened, and the plaque was unstable. The situation in the NC group was similar to those of the AS group. In the miR-342-5p antagomir and the oe-Wnt3a group, the plaque area was small, the intima of the artery was smooth, and fibrous caps had small number and became thinner. There was no rupture but foam cells in different sizes in the plaque. The cholesterol crystal was asymmetrically distributed and partially calcified, the number of smooth muscle cells and collagen fibers was increased, and the plaque tended to be stable. Compared to the NC group, the plaque area was increased and AS lesions was aggravated in the miR-342-5p agomir group (P < 0.05) while the plaque area was decreased in the miR-342-5p antagomir group and the oe-Wnt3a group with reduced AS lesions (both P < 0.05). Versus the miR-342-5p agomir group, the plaque area was decreased in the miR-342-5p agomir + oe-Wnt3a group (P < 0.05).
Oil red O staining and Sirius red staining were adopted to detect the effect of miR-342-5p targeted Wnt3a on lipid content and collagen content in plaque of aortic tissue of mice, and the results demonstrated that (Fig. 4C-F) Oil red O staining showed red fat and blue nucleus while Sirius red staining showed red collagen fibers and blue nucleus. Versus the NC group, the lipid content was raised and collagen content was reduced in the miR-342-5p agomir group as well as lipid content was reduced and collagen content was accumulated in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05).
With respect to the miR-342-5p agomir group, the lipid content was decreased and collagen content was elevated   16:165 in the miR-342-5p agomir + oe-Wnt3a group (both P < 0.05). The experimental results fully illustrated that miR-342-5p targeted regulation of Wnt3a signaling pathway had a regulatory effect on lipid and collagen content in aortic plaque of AS mice.

Effects of Down-Regulated miR-342-5p or Up-Regulated Wnt3a on Macrophages and Smooth Muscle Cells in Aortic Plaque of ApoE −/− Mice
The degree of AS is directly proportional to the content of mononuclear macrophages [21]. VSMCs are the main cells in the middle layer of arteries and are essential for maintaining the integrity of the arterial wall. VSMCs are involved in arterial wall reconstruction and play importantly in AS at various stages [22]. α-SMA is a specific marker of smooth muscle cells [23]. In this study, a macrophage marker antibody (MOMA-2) was utilized to label macrophages, and immunohistochemistry was applied detect MOMA-2 and α-SMA expression, respectively. Under the microscope, positive immunohistochemical staining of MOMA-2 and α-SMA, respectively, indicates that macrophages and smooth muscle cells are mainly located in the cytoplasm, which is yellow to brown. MOMA-2 immune positive indicated that macrophages was mainly located in the cytoplasm with yellow to brown. Determined by immunohistochemistry, it was manifested that versus the NC group, percentage of plaque macrophages (MAMO-2) positive staining was raised and percentage of positive smooth muscle cells was decreased in the miR-342-5p agomir (both P < 0.05). Percentage of plaque macrophages (MAMO-2) positive staining was reduced and percentage of positive smooth muscle cells was increased in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05). In comparison with the miR-342-5p agomir group, percentage of plaque macrophages (MAMO-2) positive staining was depressed and percentage of positive smooth muscle cells was raised in the miR-342-5p agomir + oe-Wnt3a group (both P < 0.05) (Fig. 5A-D). It was implied that miR-342-5p targeted regulation of Wnt3a signaling pathway could regulate aggregation of macrophages and smooth muscle cells in arterial tissue plaques of AS mice.

Effects of High Expression of Wnt3a or Poor Expression of miR-342-5p on Aortic Plaque Vulnerability of ApoE −/− Mice
Plaque vulnerability index was calculated: (positive percentage of macrophages + positive percentage of lipids)/ (positive percentage of smooth muscle cells + positive percentage of collagen). In relation to the NC group, the plaque vulnerability index was raised in the miR-342-5p agomir group (P < 0.05) and decreased in the miR-342-5p antagomir group and the oe-Wnt3a group (both P < 0.05). Versus the miR-342-5p agomir group, the plaque vulnerability index was decreased in the miR-342-5p agomir + oe-Wnt3a group (P < 0.05) (Fig. 5E). Briefly, miR-342-5p targeted regulation of Wnt3a signaling pathway-mediated vulnerability of plaques in arterial tissues of AS mice.

Effects of Low Expression of miR-342-5p or Overexpression of Wnt3a on Angiogenesis in Aortic Plaque of ApoE −/− Mice
Antibodies against endothelial cell marker CD34 can detect blood vessel density [24]. By immunohistochemistry and Western blot. Versus the NC group, MVD was heightened in the miR-342-5p agomir group and attenuated in the miR-342-5p antagomir group and the oe-Wnt3a group (all P < 0.05). In comparison with the miR-342-5p agomir group, MVD was decreased in the miR-342-5p agomir + oe-Wnt3a group (P < 0.05) ( Fig. 6A-C). Collectively, miR-342-5p targeting and regulating Wnt3a signaling pathway directly participated in regulating the density of neovascularization in plaques of AS mice.

Discussion
AS is an unpredictable disease involving forms of chronic inflammation and vascular remodeling, and is the major cause of mortality and morbidity globally [25]. A previous study has discussed that miR-342-5p is implicated in regulating the progression of AS [26]. Also, it was mirrored that Wnt pathway is involved in facilitating the occurrence and development of diabetic AS [27]. As the related mechanisms of miR-342-5p and Wnt3a in AS remain to be excavated, our study was to inquire the effect of miR-342-5p targeted Wnt3a on formation of vulnerable plaque and angiogenesis of AS.
Our study revealed that highly expressed miR-342-5p and lowly expressed Wnt3a were found in aortic tissues of AS mice. A study has presented that macrophagederived miR-342-5p is dramatically raised in early atherosclerotic lesions in ApoE −/− mice [13]. Another study has presented that miR-342-5p is markedly elevated in atrial fibrillation patients [28]. It has been reported that Wnt3a deficiency irreversibly injures hematopoietic stem cell self-renewal and results in defects in progenitor cell differentiation [29]. A study has shown that depletion of Wnt3a leads to defective cardiac function [30]. It has been revealed that Wnt3a expression in hippocampus of Alzheimer's disease mice is remarkably decreased [31]. Another result from our study is that Wnt3a was directly targeted by miR-342-5p in AS mice. It has been reported that miR-342-5p can target the 3'-UTR of Wnt3a and negatively regulate its expression [14].  16:165 In addition, our study has suggested that TC, TG, LDL-C, IL-5, IL-12p70, IFN-γ, TNF-α and MDA contents were increased in serum, and HDL-C content and SOD activity were decreased. In addition, plaque area, lipid content, collagen content and MVD were enhanced as well as MOMA-2 expression was raised and α-SMA expression was decreased in AS mice. IFN-γ is a soluble cytokine with many functions, including anti-fibrosis, anti-proliferation, immunomodulation, apoptosis and anti-viral activities [32]. It has been revealed that glutamine treatment markedly raises SOD activity and reduces MDA content as well as increases Wnt3a protein levels in Alzheimer's disease [31]. A study has revealed that the plasma levels of TC, TG and LDL-C are notably elevated and HDL-C is markedly reduced in AS [33]. A study has reported that TEMPOL supplementation, which has a value in suppressing metabolic disorders and raising atherosclerotic plaque stability, enhances plaque collagen content and reduces lipid content [34]. Zhou et al. noted that OPCRR treatments dramatically reduces the serum lipid profiles including TC, TG and LDL-C as well as and raises the HDL-C, also decreases MDA content as a product of lipid peroxidation and, moreover, declines serum levels of TNF-α in AS [35]. It has been presented that atherosclerotic samples have obviously reduced expression of α-SMA [36]. A study has presented that raised MVD is found in diseased aortas and especially in ruptured atherosclerotic plaque [37]. Furthermore, our study revealed that poor expression of miR-342-5p and overexpression of Wnt3a decreased he lipid levels, cytokine contents, oxidative stress response, plaque area and lipid content as well as increased collagen content, depleted MOMA-2 expression and restored α-SMA expression in aortic tissues in AS mice. It has been suggested previously that miR-342-5p is found to be positively linked to LDL-C and TNF-α serum levels and has an inverse correlation with HDL-C in coronary artery disease (CAD) patients [12]. Another study has verified that depletion of miR-342-5p inhibits AS [13]. Additionally, an experiment has presented that low serum level of Wnt1 is related to raised TG and LDL-C in premature CAD patient [38]. In addition, a study has showed that up-regulated Wnt3a, enhanced SOD content and decreased MDA content are found in the curcumin groups in Parkinson's disease rats [39].

Conclusion
In brief, our study for the first time discovered the mechanism of miR-342-5p/Wnt3 axis in AS and revealed that depleting miR-342-5p could reduce formation of vulnerable plaque and angiogenesis in AS mice via restoring Wnt3a, which may be a potential candidate for treatment of AS (Additional file 2: Fig. S2). miR-342-5p may have a synergistic effect with other miRNAs in atherosclerotic vascular disease, but due to time and funding constraints, we did not conduct further relevant discussions, which is also a limitation of this study.