MicroRNA-98 inhibition accelerates the development of atherosclerosis via regulation of dysfunction of endothelial cell

ABSTRACT

Background

Atherosclerosis has been recognized as a chronic inflammation initiated by dysfunction of endothelial cell that contributes to the increased morbidity and mortality of severe cardiovascular events. The reported important role of microRNA-98 (miR-98) in regulation of endothelial cell behaviors prompt us to hypothesize that miR-98 could be involved in the process of atherosclerosis.

Methods and Results

The current research showed the miR-98 expression was gradually down-regulated in atherosclerotic mouse arteries isolated from ApoE ablation mice subjected to high fat diet. Additionally, a dramatically reduced miR-98 expression in endothelial cells administrated to oxidized low-density lipoprotein (Ox-LDL) but a slight down-regulated level was found in macrophages. Functionally, attenuated miR-98 expression promoted secretion of chemokines and adhesion molecules in human umbilical vein endothelial cells (HUVECs) induced by Ox-LDL, which subsequently increased infiltration and pro-inflammatory genes expression of macrophages, as well as the foam cell formation. Mechanistically, in vitro experiments indicated that the endothelial cell dysfunction regulated by miR-98 knockdown was partially contributed by upregulated expression of HMGB1. Furthermore, the animal experiment with ApoE^−/− mice administrated with miR-98 inhibitor demonstrated that miR-98 silencing enhanced the atherosclerotic lesions in aorta and aortic sinus that were accompanied with increased adhesion molecules, chemokines, and pro-inflammatory markers expression.

Conclusion

MicroRNA-98 knockdown promoted endothelial cell dysfunction to affect the inflammatory state of macrophage and the development of atherosclerosis, at least partially, through direct targeting HMGB1. Collected, these data suggested that miR-98 could be a novel drug target for atherogenesis management.

KEYWORDS:

• MircroRNA-98
• Huvecs
• macrophages
• HMGB1
• atherosclerosis

Introduction

Atherosclerosis is considered a chronic and progressive inflammatory disease that accumulates in the large arteries and is responsible for the increased morbidity and mortality associated with severe cardiovascular events, including acute myocardial infarction, malignant arrhythmia, sudden cardiac death and stroke (1). Flow conditions in branching, curvatures, and bifurcations in large arteries are normally disturbed by oscillatory shear stress, which facilitates atherosclerosis formation (2). In response to pathological stimulation, endothelial cell dysfunction occurs and results in increased permeability and upregulated adhesion molecule expression and chemokines, initiating the development of atherosclerosis (3). Subsequently, circulating leukocytes penetrate the subendothelium due to endothelial cell dysfunction. In response to exposure to Ox-LDL, leukocytes/monocytes differentiate into macrophages and then phagocytose Ox-LDL to form foam cells that persistently induce inflammatory responses, enhancing necrosis and vulnerable lesion formation during atherogenesis (4).

MicroRNAs (miRNAs) are 20- to 22-nucleotide-long endogenous RNA molecules that are noncoding and highly conserved and negatively regulate target gene expression by attenuating the translation of proteins or accelerating mRNA degradation by specifically binding to the 3’-UTRs of targets (5). MiRNAs have been shown to play essential roles in multiple pathophysiological processes and various cardiovascular diseases (6). A previous study used a wide array of techniques and showed that miR-98 may be a novel biomarker for the diagnosis and treatment of cardiovascular diseases (7). Functionally, miR-98 overexpression reduces proliferation but induces apoptosis in HUVECs by targeting MAPK6 (8), while inhibiting miR-98 expression abolishes HUVEC injury induced by Ox-LDL; this effect is regulated by OIP5-AS1 downregulation and characterized by the induction proliferation and inhibition of apoptosis, inflammation, and oxidative stress injury (9). In addition, miR-98 can directly target LOX-1 to alleviate HUVEC apoptosis in response to Ox-LDL and regulate foam cell formation and lipid accumulation (10,11). However, the specific role and mechanism of miR-98 in the regulation of endothelial cell dysfunction during atherogenesis remain to be deeply investigated.

Current research showed that the expression of miR-98 was downregulated in the atherosclerotic lesions of ApoE^−/− mice treated with HFD, and this factor was primarily expressed by endothelial cells. A functional study demonstrated that miR-98 inhibition accelerated endothelial dysfunction and proinflammatory macrophage activation, which subsequently promoted the development of atherosclerosis. Mechanistically, we demonstrated that high mobility group box-1 (HMGB1) was a direct target of miR-98 that was responsible for the regulation of endothelial dysfunction.

Method and material

Animal model and treatment

For in vivo study, ApoE^−/− mice at 8-weeks-old were randomized into two groups which were treated with 2’-fluoro-methoxyethyl modified miR-98 inhibitor oligonucleotides obtained from GenePharma (Shanghai, China). Briefly, mice received two subcutaneous injections of miR-98 inhibitor or control (10 mg/kg in PBS) starting at first week (spaced 2 days apart), and followed injection of the above treatment weekly, that concomitantly fed with HFD for 12 weeks. At sacrifice, isoflurane was used to anesthetize mice and the heart was collected and embedded in OCT medium and snap-frozen. The aortas snap-frozen in liquid nitrogen were used for analysis of RNA and protein, while tissues stored in paraformaldehyde were used for morphology analysis. The animal procedures were followed by the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. All the animal protocols were approved by the Animal Care and Use Committee of The 2nd Affiliated Hospital of Harbin Medical University.

Cell culture

HUVECs were transfected with 80 to 120 nM miR-98 inhibitors and controlled overnight via Lipofectamine RNAiMax Transfection Reagent (Invitrogen). The medium 199 including 25 mmol/L HEPES, 1% heparin, 50 mg/L EC growth factor, 1% glutamine, 1% penicillin-streptomycin, and 5% FCS on low-pyrogen fibronectin (1.5 mg/cm²) was used for culturing confluent HUVECs.

Real-time PCR and western blotting

Using TRIzol reagent (Invitrogen), total RNA from mouse tissue or treated cells were collected and then used for synthesizing cDNA via Transcriptor First Strand cDNA Synthesis Kit (Roche) constituted the cDNA synthesis reaction. Real-Time PCR was analyzed using a LightCycler 480 Real-time PCR System (Roche) according to the manufacturer’s protocol. GAPDH was used as a housekeeping gene. Proteins were extracted in RIPA assay buffer. The following primer sequences were used: GAPDH, 5′-TGTGAGGGAGATGCTCAGTG-3′and 5′TGTTCCTACCCCCAATGTGT-3′; IL-6, 5′-TAGTCCTTCCTACCCCAATTTCC-3′and5′-TTGGTCCTTAGCCACTCCTTC-3′; IL-1 β,5′-GCAACTGTTCCTGAACTCAACT-3′and5′-ATCTTTTGGGGTCCGTCAACT-3′;TNF- α,5′-ATGGCCTCCCTCTCATCAGT-3′and5′-ATAGCAAATCGGCTGACGGT-3′;IL-10,5′-CATCGATTTCTCCCCTGTGA-3′and5′-CATTCATGGCCTTGTAGACACC-3′;Arg1,5′-CTCCAAGCCAAAGTCCTTAGAG-3′and5′-AGGAGCTGTCATTAGGGACATC-3′;PPARγ,5′-GCTTGTGAAGGATGCAAGGG-3′and5′-GATATCACTGGAGATCTCCGCC-3′; ICAM-1, 5′-CCTTACTCACGGTGTCCTCG-3′ and 5′-AGCGTCGGATTCGGTTCTAGC-3′;VCAM-1,5′-CCCAAACAGAGGCAGAGTGT-3′and5′-CAGGATTTTGGGAGCTGGTA-3′;MCP-1,5′-GGAAAAGGTAGTGGATGCAATTAGC-3′and5′-AACTGCATCTGCCCTAAGGTCT-3′;E-selectin,5′-ACCAGCCCAGGTTGAATG-3′and5′-GGTTGGACAAGGCTGTGC-3′;CCR2,5′-GGAATCTTCTTCATTATCCTCCTGAC-3′and5′-TGACTACACTTGTTATTACCCCAAAGG-3′;CXCL2,5′-ACCAAACGGAAGTCATAGCCA-3′and5′-TGAGACAAACTTCCTGACCATTCTT-3′; SR-A, 5′-TGGAGGAGAGAATCGAAAGCA-3′and5′- CTGGACTGACGAAATCAAGGAA-3′;CD36,5′-GACTGGGACCATTGGTGATGA-3′and5′-AAGGCCATCTCTACCATGCC-3′;ABCA1,5′-AGGCACTCAAGCCACTGCTTGT-3′and5′-TGCCTCTGCTGTCTAACAGCGT-3′;ABCG1,5′-GGTTGCGACATTTGTGGGTC-3′and5′-TTCTCGGTCCAAGCCGTAGA-3′;HMGB1,5′-GGGATGGCAAAGTTTTTCCCTTTA-3′ and 5′-CACTAACCCTGCTGTTCGCT-3′. The supernatants were electrophoresed on SDS-PAGE (Invitrogen) and transferred to a PVDF membrane (Millipore), which was then blocked with milk and incubated overnight with the specified antibodies at 4°C. The membrane was then incubated with the appropriate secondary antibody for one hour at room temperature and visualized using a FluorChem E imager (ProteinSimple). The results were normalized to GAPDH.

Macrophages migration assay

Modified Boyden chambers with Costar Transwell inserts were utilized to observe the migration ability of macrophages. Bone marrow-derived macrophage (BMDMs) isolated from ApoE^−/− mice were placed in the upper chamber of the insert, the opposite bottom was filled with HUVECs which were previously transfected with miR-98 inhibitor or controls and administrated with Ox-LDL for 12 h. After co-incubation for 12 h, macrophages migrated to the other side of the upper well were fixed and stained with crystal violet/20% methanol and counted using Image-Pro Plus 6.0.

Statistical analysis

All data were analyzed using SPSS software version 20.0 and expressed as mean ± SD. Comparisons between two groups were evaluated using two-tailed Student’s t-tests and one-way ANOVA analysis for multiple groups. A P-value<.05 was considered to be statistically significant.

Result

Down-regulated miR-98 level in atherosclerotic lesion and endothelial cells stimulated by Ox-LDL

To identify the involvement of miR-98 in development of atherosclerosis, expression change of miR-98 in pathological aortas isolated from ApoE^−/− mice administrated with HFD, respectively, for 0, 6 and 12 weeks, were first tested. A significantly gradually decreased miR-98 mRNA expression was found in HFD-treated atherosclerotic aorta tissues (Figure 1a,b). Since the published papers have indicated that miR-98 exerts critical role in regulation of endothelial cell behaviors, the expression of miR-98 in endothelial cells subjected to Ox-LDL was followed to test. We noticed a gradually decreased miR-98 expression in endothelial cells with Ox-LDL treatment followed the increased cultured time (Figure 1c). Additional, miR-98 level in macrophage administrated with Ox-LDL was also examined and exhibited a slight decreased expression (Figure 1d). Thus, these observations suggested miR-98 was further studied to determine its role in endothelial cells regulation and the development of atherosclerosis.

Figure 1. Downregulated miR-98 expression in the atheromatous aortas and endothelial cells.

a, b. RT-PCR analysis of miR-98 mRNA expression in the aortas from ApoE^−/− mice fed with HFD for 0, 6, and 12 weeks. n = 3. *P < 0.05 versus HFD for 0W, #P < 0.05 versus HFD for 6W. c, d. RT-PCR analysis of miR-98 mRNA expression in HUVECs or macrophages treated with Ox-LDL for 0, 4, 8, and 12 h. *P < 0.05 versus Ox-LDL for 0 h; #P < 0.05 versus Ox-LDL for 4 h; §P < 0.05 versus Ox-LDL for 8 h.

MiR-98 inhibition increases adhesion molecule and chemokine expression

An important aspect of endothelial cell dysfunction is the increased secretion of multiple adhesion molecules and chemokines that initiate atherogenesis and promote the infiltration of macrophages, which induce the inflammatory response. The RT‒PCR results showed that miR-98 knockdown significantly increased the mRNA expression of E-selectin, MCP-1, ICAM-1, VCAM-1, CCR2, and CXCL2 in HUVECs stimulated with Ox-LDL in a time-dependent manner compared with that in the control group (Figure 2a-f). The similar results were also obtained in protein level (Figure 2g).

Figure 2. Increased adhesion molecules and chemokines secretion by miR-98 inhibition.

a- f. RT-PCR analysis of adhesion molecules (ICAM-1, VCAM-1) and chemokines (E-selectin, MCP-1, CCR2, and CXCL2) markers in HUVECs transfected with miR-98 inhibitor and control inhibitor treated with OX-LDL for 0, 4, 6, and 12 h. n = 3. g. Western blot analysis of E-selectin, MCP-1, ICAM-1, CXCL2 in HUVECs transfected with miR-98 inhibitor and control inhibitor treated with OX-LDL for 12 h. n = 3. *P < 0.05 versus control group.

MiR-98 knockdown promotes migration, the proinflammatory response, and foam cell formation in macrophages

Next, a coculture system was used to test whether endothelial cell dysfunction regulated by miR-98 inhibition could affect macrophage activation. The medium used to culture HUVECs transfected with the miR-98 inhibitor or control inhibitor and PBS or Ox-LDL treatment was seeded in the bottom chambers of cell culture plates or was collected, and the BMDMs of ApoE^−/− mice were seeded in the upper chambers of the wells or were cultured with the collected medium. By using this coculture system, an increase in macrophage migration was observed in the medium used to culture HUVECs transfected with the miR-98 inhibitor and treated with Ox-LDL compared with control medium (Figure 3a). Moreover, we examined the mRNA levels of inflammatory mediators. In response to the medium used to culture HUVECs transfected with the miR-98 inhibitor and controls, we noticed that miR-98 knockdown increased the expression of proinflammatory genes (IL-6, IL-1β, and TNF-α) but decreased anti-inflammatory gene expression (IL-10, Arg-1 and PPARγ) compared to the controls (Figure 3b). In addition, the induction of foam cell formation in macrophages was also observed after treatment with a medium used to culture HUVECs transfected with the miR-98 inhibitor (Figure 3c). Moreover, the mRNA levels related to cholesterol influx (SR-A and CD36) were upregulated, but markers of cholesterol efflux (ABCA1 and ABCG1) were decreased by miR-98 inhibition (Figure 3d).

Figure 3. Macrophage activation induced by miR-98-mediated endothelial dysfunction.

a. Tranwell analysis of the accumulated migrated macrophages induced by co-culturing with HUVECs transfected with miR-98 inhibitor and controls followed with OX-LDL treatment in the bottom for 12 h. n = 7-8. Scale bar = 100 μm. b. Pro- and anti-inflammatory genes expression of macrophages co-culture with medium used for culturing HUVECs transfected with miR-98 inhibitor and controls followed with OX-LDL treatment for 12 h. n = 3. c. Foam cell formation in macrophages co-culture with medium used for culturing HUVECs transfected with miR-98 inhibitor and controls followed with OX-LDL treatment for 12 h. n = 3. Scale bar = 50 μm. d. Cholesterol efflux and uptake genes expression of macrophages co-culture with medium used for culturing HUVECs transfected with miR-98 inhibitor and controls followed with OX-LDL treatment for 12 h. n = 3. *P < 0.05 versus control inhibitor with Ox-LDL group.

MiR-98 knockdown promotes HMGB1 expression

We performed bioinformatics analysis through TargetScan to identify the potential target of miR-98. Among the potential targets, the results showed that HMGB1 contained a conservative putative binding site for miR-98 in its 3”UTR (Figure 4a), and HMGB1 has been shown to play a key role in the regulation of endothelial cell function and atherogenesis (12). A luciferase reporter assay showed that miR-98 inhibitor transfection increased HMGB1 luciferase activity, which was largely abolished when the predicted binding sites within the HMGB1 3”UTR were mutated (Figure 4b). We also noticed that HMGB1 mRNA and protein expression were increased in endothelial cells transfected with the miR-98 inhibitor and treated with Ox-LDL (Figure 4c,d). Furthermore, HMGB1 knockdown attenuated the positive effect of miR-98 inhibition on endothelial cell dysfunction (Figure 4e).

Figure 4. MiR-98 directly targets to HMGB1.

a. Predicted target sites through TargetScan of miR-98 in the 3’UTR of HMGB1 mRNA. b. HMGB1 activity treated with miR-98 inhibitor or negative control tested by Luciferase. n = 3. *P < 0.05 compared to control inhibitor group. c, d. RT-PCR and western blot analysis of HMGB1 expression in HUVECs transfected with miR-98 inhibitor or control inhibitor administrated with Ox-LDL. *p < 0.05 compared to control inhibitor group. n = 3. E. adhesion molecules and chemokines secretion in HUVECs transfected with miR-98 inhibitor or control inhibitor and infected with AdshHMGB1 administrated with Ox-LDL. n = 3. *P < 0.05 compared to control inhibitor group. #P < 0.05 compared to miR-98 inhibitor group; †P < 0.05 compared to control inhibitor with AdshHMGB1 group.

MiR-98 inhibition promotes atherosclerotic plaque and inflammatory response

Since miR-98 downregulation plays an important role in the regulation of endothelial cell dysfunction in vitro, in vivo analysis of ApoE^−/− mice treated with miR-98 inhibitor or controls was performed, and these mice were fed a HFD for 12 weeks. Morphological analysis of the whole aortas (Figure 5a) and aortic sinus (Figure 5b) showed dramatic induction of plaques in miR-98 inhibitor-treated ApoE^−/− mice that were fed a HFD for 12 weeks in comparison with the control groups. To verify the effect of miR-98 knockdown on adhesion molecule, chemokine, and inflammatory cytokine mRNA expression in vitro, aortic tissues were collected. We showed that the expression of markers including ICAM-1, VCAM-1 and MCP-1, IL-6, IL-1β, TNF-α, IL-10, Arg-1 and PPARγ was induced in the atherosclerotic aortas of miR-98 inhibitor-treated ApoE^−/− mice compared with those in the control group (Figure 5c,d).

Figure 5. The miR-98 silencing accelerates atherogenesis in ApoE^−/− mice.

a. En face analysis of atherosclerotic lesion in the aortas of miR-98 inhibitor treated ApoE^−/− mice with HFD treatment for 12 weeks stained with Oil Red O. n = 8–10. b. Formation of atherosclerotic lesion in aortic sinus in miR-98 inhibitor treated ApoE^−/− mice fed with HFD stained with H&E. n = 3–5. Scale bar = 100 μm. c, d. RT-PCR analysis of mRNA expression of adhesion molecules, chemokines, and inflammatory mediators in artery tissues from miR-98 inhibitor treated ApoE^−/− mice with HFD treatment for 12 weeks. n = 3. *P < 0.05 compared to ApoE^−/− group.

Discussion

The current study demonstrated a novel role of miR-98 in the development of atherosclerosis. The downregulation of miR-98 expression was shown in atherosclerotic lesions, and a significant reduction in miR-98 levels was observed in endothelial cells. The in vitro and in vivo experiments demonstrated that miR-98 inhibition dramatically increased the production of adhesion molecules and chemokines in HUVECs, as well as the inflammatory response in macrophages, which accelerated atherogenesis. Mechanistically, HMGB1 may be the direct target of miR-98. Thus, the miR-98/HMGB1 axis may regulate endothelial cell dysfunction to manage atherosclerosis.

The normal and deterministic roles of endothelial cells are to maintain homeostasis and the physiological function of the vasculature, which is characterized by the inhibition of thrombosis form, barrier function, and anti-inflammatory responses (13). In response to hypercholesterolemia stimulation and oscillatory shear stress, endothelial cell dysfunction occurs and triggers atherosclerosis by promoting leukocyte adhesion, lipoprotein infiltration, and inflammation (14). Previous studies have shown that miR-98 expression is significantly decreased in single-, double- and multivessel-occluded CAD patients, as well as in the endothelial cells of patients with atherosclerosis (7). Moreover, a wide array of techniques has shown the potential association of miRNAs with cardiovascular diseases and diabetes mellitus, and the results show that miR-98 may be a novel biomarker to improve the diagnosis and treatment of CVD (15). Our present study showed the downregulation of miR-98 expression in atherosclerotic plaques and consistent expression in Ox-LDL-induced HUVECs. Collectively, this evidence suggested that miR-98 expression in endothelial cells could affect atherogenesis. Notably, accumulating evidence has demonstrated that miR-98 is widely involved in the regulation of endothelial cell functions. MiR-98 overexpression reduces proliferation but induces apoptosis in HUVECs by targeting MAPK6 (8), while miR-98 alleviates HUVEC apoptosis in response to Ox-LDL by suppressing LOX-1 (10). MiR-98 is responsible for the effects of OIP5-AS1 downregulation on HUVEC injury in the presence of Ox-LDL (9). The secretion of adhesion molecules and chemokines mediated by endothelial cell dysfunction facilitates the rolling and penetration of circulating monocytes (16,17). A previous study showed that VCAM-1 could bind to miR-98 and reverse the protective effects of miR-98 on HUVECs induced by Ox-LDL (18). Consistently, our study showed that miR-98 inhibition dramatically increased the production of adhesion molecules and chemokines in HUVECs. This evidence strongly suggests that miR-98 is a novel target for the treatment of atherogenesis, particularly in an endothelial cell-dependent manner.

The target screen performed by the program TargetScan showed that HMGB1 contained a conservative putative binding site for miR-98 in its 3’UTR and may be required for the regulation of endothelial cell dysfunction mediated by miR-98. HMGB1, which is a nuclear protein, is a nonhistone DNA-binding protein that is expressed in eukaryotic cells (19). HMGB-1 plays an essential role in diseases including sepsis, collagen disease, cancers, arthritis, acute lung injury, epilepsy, myocardial infarction, and systemic inflammation (20). In addition, accumulating evidence has demonstrated that HMGB1 is implicated in atherosclerosis due to its role in multiple cell types involved in atherogenesis, including endothelial cells, vascular smooth cells, and macrophages. HMGB1 expression is closely associated with thrombogenesis in a microminipig model of atherosclerosis induced by hyperlipidemia (21). Endothelial HMGB1 expression regulates LDL transcytosis, which may be a novel mechanism by which HMGB1 is implicated in atherosclerosis (22). HMGB1 interacts with advanced glycation end products (RAGE) or Toll-like receptor 4 (TLR4) and activates the expression of proinflammatory cytokines, chemokines, and adhesion molecules in endothelial cells and macrophages/monocytes to contribute to atherosclerosis progression. An increase in HMGB1 expression has been found in the nuclei and cytoplasm of macrophages, as well as smooth muscle cells in atherosclerotic lesions, and plays an important role in the progression of atherosclerotic plaque (12). Moreover, as a critical mediator of Ox-LDL-induced atherosclerosis, HMGB1 is associated with atherosclerotic plaque composition and burden in patients with stable coronary artery disease (23). Overall, HMGB plays an essential role in the regulation of endothelial cells and macrophages to participate in atherogenesis. In addition, miR-98 can directly target HMGB1 and is responsible for the effects of OIP5-AS1 downregulation on HUVEC injury induced by Ox-LDL-induced (9). Previous studies and our current study provide strong evidence that the miR-98/HMGB1 axis may be important for regulating the development of atherosclerosis.

In summary, we provided evidence that a decrease in miR-98 expression was a critical molecular switch in atherosclerosis development, which involved the upregulation of adhesion molecule secretion and an increase in the inflammatory response in macrophages. Our work highlights the important role of targeting miR-98 as a potential pharmacological therapeutic approach for reversing the development of atherosclerosis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data used to support the findings of this study are available from the corresponding author upon request.

Additional information

Funding

This work was supported by grants from Relationship between oct stent restenosis and serum untargeted metabolomics substrate peaks (YYWS4791)