Introduction
Lipid metabolism disorders are associated with several diseases, including hyperlipidemia, atherosclerosis, coronary heart disease, and cardiovascular diseases (CVDs). Lipid metabolism disorders are characterized by increased levels of total cholesterol (TC) and low-density lipoprotein (LDL) cholesterol and decreased levels of high-density lipoprotein (HDL) cholesterol [
1,
2]. In addition to dysregulation of lipid metabolism, aberrant inflammatory responses are common in the diseases mentioned above, and are leading causes of mortality resulting from CVD and atherosclerosis [
2‐
4]. Low density lipoproteins (LDLs) are lipoprotein particles that carry cholesterol into peripheral tissue cells, and can become oxidized to form oxidized LDLs. When LDL levels (especially ox-LDLs) become excessive, the cholesterol they carry accumulates on the arterial wall and ultimately causes arteriosclerosis. Oxidized low-density lipoproteins (ox-LDLs) participate in the formation and progression of lesions by eliciting lipid accumulation, and causing inflammation and toxic events [
5,
6]. It was reported that ox-LDLs combines with their major receptor (Lectin-like oxidized low-density lipoprotein receptor-1 [LOX-1]) to activate the NF-κB signaling pathway [
7,
8]. Ox-LDLs also affect the generation and utilization of endothelium-derived nitric oxide (NO), impair endothelium-dependent vascular diastolic function, and initiate the formation and development of AS plaques [
7,
9‐
11].
Atherosclerosis is characterized by the formation of lipid plaques which mainly consist of lipid-laden foam cells transferred from macrophages, and therefore decrease the uptake of LDLs from blood vessel walls [
2,
12]. Several factors are associated with cholesterol efflux; these factors include the ATP-binding cassette (ABC) transporter A1 (ABCA1) protein family [
13], miR-33-5p [
14‐
16], inflammatory factors such as interleukin (IL)-6 and tumor necrosis factor (TNF)-α [
17], and the 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) enzyme. ABCA1 plays a crucial role in removing excess lipids by initiating reverse cholesterol transport (RCT) and mediating the transport of intracellular free cholesterol and phospholipids from peripheral tissues to lipid-poor apolipoprotein A I (ApoA1) [
18]. In addition, ABCA1 also promotes the production of athero-protective HDL cholesterol [
18,
19]. MiR-33-5p targets ABCA1, and knockdown of miR-33-5p decreases the levels of total cholesterol (TC) and TNF-α, and promotes cholesterol efflux from foam cells [
13,
20]. It has been reported that the inflammatory cytokines IL-6 and TNF-α can suppress ABCA1 expression and thereby inhibit cholesterol efflux [
17].
Citrate is a precursor metabolite for lipid and cholesterol synthesis, and accumulation of exogenous citrate promotes inflammation in visceral adipose tissue [
21]. Lower rates of citrate synthase (CS) activity have been reported to be associated with increased levels of cleaved caspase-3 and lipotoxicity in C1C12 muscle cells [
22]. Our previous experiments showed that CS is a direct target of miR-33-5p. However, there have been no reports of CS activity being associated with cholesterol efflux or lipid metabolism disorders such as atherosclerosis or CVD.
This current study investigated how the miR-33-5p/ABCA1/CS axis helps to control cholesterol efflux in vascular endothelial cells (VECs). VECs were treated with ox-LDLs and pretreated with a miR-33-5p inhibitor, ABCA1 and CS expression plasmids, or with siRNAs. The effects on cell apoptosis, cellular cholesterol efflux, and inflammatory cytokines (IL-6 and TNF-α) that occurred in response to gene modulation were detected. Our results provide novel insights into potential mechanisms associated with lipid metabolism.
Materials and methods
Cells, culture conditions, and treatments
Human VECs (CRL-1730, ATCC, Manassas, VA, USA) were maintained in DMEM (Hyclone, Logan, UT, USA) containing 1% antibiotics (penicillin–streptomycin), 10% FBS (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), with or without oxidized low-density lipoproteins (ox-LDLs) at a low (50 μg/mL), moderate (100 μg/mL) or high concentration (200 μg/mL) [
23]. The ox-LDLs (Shanghai Jingke Chemical Technology Co., LTD, Shanghai, China) were dissolved in PBS, and then diluted to different concentrations. The VECs were incubated in DMEM at 37 °C in a 5% CO
2 atmosphere, and then induced with ox-LDLs for 48 h.
Plasmids and cell transfections
Plasmids overexpressing ABCA1 (OE-ABCA1) and CS (OE-CS) were constructed using the pcDNA3.1 vector (Genechem Co. Ltd., Shanghai, China) and full CDS sequence cloning. The polymerase chain reaction (PCR) was used to clone the ABCA1 or CS coding sequences into the plasmid pcDNA3.1 vector. The pcDNA3.1 vector without the ABCA1 and CS genes served as a negative control. pcDNA3.1-ABCA1 or pcDNA3.1-CS was transfected into VECs (6-well plates, 1.0 × 106/cm2). Short interfering RNAs (siRNAs) against human ABCA1 and CS genes, miR-33-5p mimics (B01001), inhibitors (B03001) (chemically modified inhibitors that specifically targeted miR-33-5p), and scramble sequences were obtained from GenePharma Corporation (Shanghai, China). Cell transfections were performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. Cells transfected with the empty pcDNA3.1 vector or scrambled sequences served as negative controls (NCs) for overexpression and for siRNA, mimics or inhibitors, respectively. Cells were transfected with the different agents for 48 h before harvesting. Each experiment was repeated 3 times.
Dual-luciferase reporter assay
Possible interactions between miR-33-5p and ABCA1 or CS were predicated using TargetScanHuman (
http://www.targetscan.org/vert_72/), and subsequently verified by using the dual-luciferase reporter system. Luciferase vectors containing miR-33-5p binding sites (wild type, WT; and mutant, MUT, 7 base substitutions) in the 3′ UTR regions of
ABCA1 or
CS genes were constructed using XhoI and NotI restriction enzymes; Thermo Fisher Scientific Inc., Waltham, MA, USA) and a psiCHECK-2 expression vector (Promega, Madison, WI, USA). Cell transfections were performed using Lipofectamine 2000 (Invitrogen).
Enzyme-linked immuno sorbent assay (ELISA)
Cell culture samples were collected and prepared for ELISA. The levels of CS, interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α) in the samples were detected using ELISA kits (Elabscience Biotechnology Co. Ltd., Wuhan, China). A microplate reader (Thermo Fisher Scientific Inc.) was used to measure optical density (OD) at 450 nm.
Cholesterol efflux determination
The total cholesterol efflux in culture medium and cell lysate was measured using a Cholesterol Efflux Fluorometric Assay Kit (BioVision, San Francisco, CA USA) according to the manufacturer’s protocol [
24]. The fluorescence-labeled cholesterol was washed with 200 μL of phenol red-free, serum-free RPMI medium and then incubated in serum-free RPMI medium containing cholesterol acceptor ApoA 1 (50 μg/well) for 16 h at 37 °C. The total fluorescence intensity at 48 h post-transfection or treatment was detected at excitation and emission wavelengths of 482 nm and 515 nm, respectively. Cholesterol efflux was calculated as: 100% x media fluorescence intensity/total fluorescence intensity.
Flow cytometric analysis and aging assay
Cell apoptosis was detected by flow cytometry. Cells treated or transfected with different agents for 48 h were harvested and fixed. Cell apoptosis was detected by using an Annexin V-FITC/PI Double Staining Kit (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions. The percentages of apoptotic VECs were analyzed using a BD FACS Calibur flow cytometer (BD Biosciences). Aged cells were detected by using a β-Galactosidase staining kit (Jiancheng Biological Company, Nanjing, China) according to the manufacturer's recommended protocol. Senescence-associated β-galactosidase (SA-β-Gal) in aged cells was detected as previously described [
25]. Cellular images were captured using a microscope (Motic, Xiamen, China). Each experiment was repeated 3 times.
Quantitative real-time PCR (qRT-PCR) analysis
Total cellular RNA was extracted using TRIzol reagent (Invitrogen), and first strand cDNA was synthesized using a Bestar qPCR RT Kit (DBI Bioscience, Shanghai, China). MiR-33-5p expression was achieved by reverse transcription that was performed using a stem-loop primer. A DBI Bestar ® SYBR Green qPCR master mix kit (DBI Bioscience) was used for qRT-PCR amplification conducted on an Mx3000P Agilent Stratagene PCR machine (Agilent, Santa Clara, CA, USA). The primers were synthesized by Sangon Biotechnology Co., Ltd. (Shanghai, China; Table
1). The following conditions were used for amplification: 95 °C for 2 min; followed by 40 cycles of 94 °C for 20 s, 58 °C for 20 s, and 72 °C for 20 s. The relative levels of mRNA or miR-33-5p expression were calculated using standard 2
−△△Ct methods. U6 and GAPDH served as internal reference genes.
Table 1
Primers used for quantitative real time PCR analysis
Hsa-miR-33 | F: ACACTCCAGCTGGGGTGCATTGTAGTTGCAT RT: CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGTGCAATG |
All | R: CTCAACTGGTGTCGTGGA |
ABCA1 | F: ACCCACCCTATGAACAACATGA R: GAGTCGGGTAACGGAAACAGG |
CS | F: AACTGCTACCCAAGGCTAAGG R: CTTTTGAGAGCCAAGATACCTGT |
HMGCR | F: AACTGCTACCCAAGGCTAAGG R: CTAAAATTGCCATTCCACGAGC |
U6 | F: CTCGCTTCGGCAGCACA R: AACGCTTCACGAATTTGCGT |
GAPDH | F: TGTTCGTCATGGGTGTGAAC R: ATGGCATGGACTGTGGTCAT |
Western blot analysis
The total cellular proteins were extracted using RIPA lysis buffer (Beyotime, Shanghai, China), and the protein concentration in each extract was determined using a protein quantification kit (Thermo Fisher Scientific). An aliquot of total protein from each extract was separated by 10% SDS-PAGE (Sangon Biotechnology Co., Ltd), and the protein bands were transferred onto PVDF membranes (Millipore Corporation, Burlington, MA, USA), which were subsequently blocked with 5% skim milk powder (Beyotime). Next, the membranes were incubated with specific primary antibodies against HMGCR (1: 1000; ab174830, Abcam, Cambridge, UK), ABCA1 (1: 1000; #96292, CST, Danvers, MA, USA), CS (1: 1000; #14309, CST), Caspase 3 (1: 1000; #14220, CST), Bax (1: 1000; #5023, CST), Bcl-2 (1: 1000; #4223, CST), and GAPDH (1: 2500; ab9485, Abcam), and subsequently incubated with a secondary antibody (HRP Goat anti-rabbit/mouse IgG, 1: 20,000; Boster, Wuhan, China). The immunostained protein bands were detected using enhanced chemiluminescence luminous fluid (Millipore), and band staining intensity was analyzed using a scanner (Microtek, Carson, CA, USA).
Statistical analysis
The GraphPad Prism 8.0 (GraphPad, La Jolla, USA) was used for data analysis. Experiments were repeated in three times, all data are expressed as a mean value ± standard deviation (SD). The distribution normality of all data was verified prior before analysis by one-way ANOVA. Continuous data obtained from multiple groups was analyzed by one-way ANOVA followed by Tukey's post hoc test. A P-value < 0.05 was considered to be statistically significant, P-value < 0.01 was considered as obvious statistical significance.
Discussion
Low density lipoproteins (LDLs) are lipoprotein particles capable of carrying cholesterol into peripheral tissue cells, where it can be oxidized to form oxidized LDL. When the levels of LDLs, and especially ox-LDLs, become excessive, the cholesterol they carry will accumulate on the arterial wall and eventually cause arteriosclerosis. High density lipoproteins (HDLs) have received more attention as a preventive factor of atherosclerosis because they can export cholesterol and promote cholesterol metabolism. Furthermore, HDLs can help to atherosclerosis and protect against coronary heart disease. Many factors are known to be associated with cholesterol efflux, such as ABCA1 and miR-33-5p [
14,
15]. Our present study demonstrated the role played by the miR-33-5p/ABCA1/CS axis in controlling cholesterol efflux in VECs. To the best of our knowledge, this is the first study to demonstrate the crucial role played by CS in cholesterol efflux in VECs.
ABCA1 is a direct target of miR-33-5p [
14,
20]. Knockdown of miR-33-5p has been reported to decrease the levels of TC and TNF-α, and promote cholesterol efflux from macrophage-derived foam cells [
20]. Our present study demonstrated the targeted interaction that occurs between ABCA1 and miR-33-5p. ABCA1 is crucial for removing excess lipids by initiating the RCT process and promoting cholesterol efflux to ApoA1 [
18]. Our present study demonstrated that overexpression of ABCA1 forced by treatment with an miR-33-5p inhibitor or transfection with an overexpression plasmid promoted cholesterol efflux in VECs. In addition, ABCA1 overexpression and the miR-33-5p inhibitor prevented ox-LDL-induced inflammation, apoptosis, and aging in VECs. These data suggested the potential therapeutic effect of ABCA1 expression and miR-33-5p downregulation in modulating lipid metabolism by promoting cholesterol efflux.
CS is essential for cell growth [
26]. Our study also demonstrated the similar crucial roles played by CS and ABCA1 in regulating cholesterol efflux, inflammation, and apoptosis in VECs. We first confirmed that CS was a direct target of miR-33-5p, and then showed that overexpression of CS impeded the ox-LDL-induced aging and apoptosis of VECs. In addition, out data showed that the ox-LDL-induced reductions in cholesterol efflux in VECs could be rescued by CS overexpression. In contrast, inhibition of CS and ABCA1 both promoted VEC apoptosis. These results were inconsistent those reported by Alhindi et al. [
22], who showed that lower levels of CS activity were associated with increased levels of cleaved caspase-3 and lipotoxicity in C1C12 muscle cells. A low level of CS expression is associated with cell apoptosis and oxidative stress [
27]. CS catalyzes the citric acid cycle in mitochondria and can serve as a biomarker of mitochondrial function in mammals [
26]. In aging cells, mitochondrial function declines [
26,
28]. Our present study demonstrated that siRNAs for both CS and ABCA1 could promote VEC aging and apoptosis. CS functions downstream of ABCA1 and was necessary for maintaining ABCA1-mediated cellular functions and cholesterol efflux in VECs. These findings suggested CS as a key and novel mediator of ABCA1-mediated cholesterol efflux.
Citrate plays important roles in innate immunity and inflammation [
29]. It was reported that the natural auto-antibodies against CS (anti-CS IgM and IgG antibodies) were increased in rheumatoid arthritis (RA) patients treated with TNF inhibitors, certolizumab or etanercept therapy [
30]. These antibodies might be associated with the clearance of apoptotic cells and restoration of immunological systems in the body [
30]. Our study showed that inhibition of CS and ABCA1 increased the levels of IL-6 and TNF-α in VECs, while the overexpression of CS and ABCA1 reduced the levels IL-6 and TNF-α. In addition, suppression of CS increased the levels of ABCA1-reduced inflammatory cytokines (IL-6 and TNF-α) in VECs. These data suggested CS as a potent target for managing the inflammation-associated cholesterol efflux in VECs. However, the in vitro model used in our study has limitations, and an in vivo model such as apolipoprotein E-deficient (apoE(-/-)) hyperlipidemic mice or Abca1 binding site mutant (Abca1(BSM)) mice [
31,
32] should to be used to further validate the role of the miR-33-5p/ABCA1/CS axis in cholesterol efflux. Moreover, citrate synthase activity has been shown to decrease after myocardial infarction and be partially regulated by MMP-9 [
33]. Furthermore, as MMP-9 has been shown to cleave citrate synthase ex vivo, the change and involvement of MMP-9 in lipid metabolism disorders should also be further investigated, along with the role played by the miR-33-5p/ABCA1/CS axis.
Conclusions
In conclusion, our present study demonstrated the crucial roles played by the miR-33-5p/ABCA1/CS axis in regulating cholesterol efflux, inflammation, apoptosis, and aging in VECs. We also found that CS had similar functions to ABCA1 in controlling VEC behavior and cholesterol efflux. CS and ABCA1 expression in VECs promoted cholesterol efflux and inhibited ox-LDL-induced apoptosis, inflammation, and aging. Our rescue experiments confirmed that CS acts downstream of ABCA1. The miR-33-5p/ABCA1/CS axis plays pivotal roles in lipid metabolism and the underlying mechanisms of lipid metabolism disorders such as atherosclerosis and CVDs.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.