Biomedicine & Pharmacotherapy 146 (2022) 112596 Contents lists available at ScienceDirect Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha MiR-125b downregulates macrophage scavenger receptor type B1 and reverse cholesterol transport Miguel Hueso a, b, *, Raquel Griñán c, d, Adrián Mallen b, Estanislao Navarro b, Elvira Purqueras e, Montse Gomá e, Fabrizio Sbraga f, Arnau Blasco-Lucas f, Giovanna Revilla c, David Santos g, Marina Canyelles c, Josep Julve c, g, Joan Carles Escolà-Gil c, g, **, Noemi Rotllan c, g, ** a Department of Nephrology, Hospital Universitari Bellvitge and Bellvitge Research Institute (IDIBELL), L′ Hospitalet de Llobregat, 08907 Barcelona, Spain b Nephrology and Renal Transplantation Group, Infectious Disease and Transplantation Program, Institut d′ Investigació Biomèdica de Bellvitge-IDIBELL, L′ Hospitalet de Llobregat, 08907 Barcelona, Spain c Institut d′ Investigacions Biomèdiques (IIB) Sant Pau, C/ Sant Quintí 77, 08041 Barcelona, Spain d Departament de Bioquímica i Biologia Molecular, Universitat Autònoma De Barcelonna, 08041 Barcelona, Spain e Department of Pathology, Hospital Universitari Bellvitge and Bellvitge Research Institute (IDIBELL), L′ Hospitalet de Llobregat, 08907 Barcelona, Spain f Department of Cardiac Surgery, Hospital Universitari Bellvitge and Bellvitge Research Institute (IDIBELL), L′ Hospitalet de Llobregat, 08907 Barcelona, Spain g CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Objective: To determine whether miR-125b regulates cholesterol efflux in vivo and in vitro through the regulation of Cholesterol efflux scavenger receptor type B1 (SR-B1). Reverse cholesterol transport Approach and results: We demonstrated that miR-125b is up-regulated in the human aortas of patients with CAD Scavenger receptor class B type 1 (SRB1) and is located in macrophages and vascular smooth muscle cells (VSMCs). We identified SCARB1 as a direct MicroRNA MiR-125b target of miR-125b by repressing the activity of the SCARB1 3′ -untranslated region reporter construct. Moreover, Macrophage the overexpression of miR-125b in both human and mouse macrophages as well as VSMCs was found to Vascular smooth muscle cell downregulated the expression of the SCARB1 and the SR-B1 protein levels, thereby impairing α-HDL-mediated Coronary artery disease macrophage cholesterol efflux in vitro. The in vivo reverse cholesterol transport (RCT) rate from non-cholesterol- loaded macrophages transfected with miR-125b to feces was also found to be decreased when compared with that of control mimic-transfected macrophages. Conclusions: Together, these results provide evidence that miR-125b downregulates SCARB1 and SR-B1 in both human and mouse macrophages as well as VSMCs, thereby impairing macrophage cholesterol efflux in vitro and the whole macrophage-specific RCT pathway in vivo. 1. Introduction efflux occurs via several pathways, one of these pathways is called passive diffusion, and it may be facilitated by scavenger receptor class B The accumulation of macrophages in the artery wall, and the sub­ type 1 (SR-B1), and the other pathways are active processes involving sequent foam cells formation are hallmarks of vascular lesions in the ATP-binding cassette (ABC) A1 and G1 transporters [2]. SR-B1 is atherosclerosis (ATH) progression. Since macrophages do not express widely expressed throughout the different cell types and both preclinical pathways for catabolizing cholesterol, the ability of high-density lipo­ experimental and clinical studies have shown that the disruption of proteins (HDL) particles to stimulate the efflux of excess free cholesterol SR-B1 function predisposes to the development of ATH and cardiovas­ is critical in reducing foam cell formation [1]. Macrophage cholesterol cular disease [3,4], although there is a controversy in different human Abbreviations: ABCA-1, ATP-binding cassette transporter A1; ATH, atherosclerosis; BOECS, Blood Outgrowth Endothelial Cells; CAD, coronary artery disease; ECs, Endothelial Cells; HDL, high-density lipoprotein; miRNA, microRNA; RCT, reverse cholesterol transport; SCARB1, scavenger receptor class B type 1 gene; SR-B1, scavenger receptor class B type 1; 3′ UTR, 3′ untranslated region; VSMCs, Vascular smooth muscle cells; WT, Wild-type. * Correspondence to: L′ Hospitalet de Llobregat, 08907 Barcelona, Spain. ** Correspondence to: C/ Sant Quintí 77, 08041 Barcelona, Spain. E-mail addresses: mhueso@idibell.cat (M. Hueso), JEscola@santpau.cat (J.C. Escolà-Gil), nrotllanv@santpau.cat (N. Rotllan). https://doi.org/10.1016/j.biopha.2021.112596 Received 29 September 2021; Received in revised form 16 December 2021; Accepted 23 December 2021 Available online 28 December 2021 0753-3322/© 2021 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 genetic studies with rare SCARB1 mutations and the association with considering the predictive Tm values of the probe used). After the hy­ CAD and levels of HDL cholesterol (HDL-C) [5–7]. Interestingly, the bridization incubation, the coverslips were removed and the slides were contribution of macrophage SR-B1 to the pathogenesis of ATH appears washed three times in a Coplin jar with 0.01 × SSC at 78 ◦ C (4–8 ◦ C to be highly dependent on the foam cell formation stage [3]. In addition, above the hybridization temperature) with 10 min per wash in agitation. other cell types, such endothelial (ECs) and vascular smooth muscle cells Then, the slides were switched to a Coplin jar with 2 × SSC at RT and (VSMCs) can also become foam cells [8,9], although the contribution of incubated for 5 min with agitation. The tissue slides were dried for 5 min SR-BI-mediated cholesterol efflux pathways on foam cell formation in at RT, embedded with a DAPI mounting medium, and covered with a these particular cell types remains unknown. glass coverslip to keep the slides at 4 ◦ C under dark conditions until Small non-coding miR-125b was associated with ATH development microscope imaging. Fluorescence imaging was performed using a Carl in an experimental model, and it was also found to be upregulated in Zeiss LSM880 spectral confocal microscope. We used a 63 × objective human atherosclerotic plaques of abdominal aortas from deceased pa­ with NA of 1.4. For the excitation of the DAPI, we used 405 nm-laser tients [10]. Furthermore, the relative abundance of miR-125b was found line, and for Cy5, we used a 633 nm-laser line. The images were to be up-regulated by the CD40-NF-κB signaling axis in activated mouse analyzed with Zen Black 2.3 Software. macrophages [10]. Also, a microRNA expression quantitative trait loci-mapping study of whole blood from a large cohort identified the 2.3. Immunohistochemistry polymorphisms of cis-acting regulatory elements of miR-125b highly associated with HDL-C [11]. In this context, miR-Base screening detec­ Consecutive sections to the AISH aortas were cut into of 8 µm-thick ted that the 3′ -untranslated region (3′ -UTR) from the SCARB1 gene, sections on a cryostat and placed on a Ventana Ultra View Autostainer. which encodes SR-B1, was a predicted target for miR-125a and The sections were dried at 95 ◦ C for 8 min. The slides were pretreated at miR-125b [12]. This latter study also reported that miR-125a down­ 95 ◦ C for heat-induced antigen retrieval in a Tris_EDTA Cell Condi­ regulated SR-B1 expression by direct targeting and inhibiting tioning Buffer (CC1) (pH = 8.4) for 64 min followed by incubation with SR-B1-mediated selective HDL uptake in mouse and rat steroidogenic the ready-to-use mouse monoclonal antibody, CONFIRM™ anti-CD68 cells [12]. (KP-1) Primary Antibody (Ventana Medical Systems, Arizona, USA), at Since SR-B1 is highly expressed in foam cells within human aortic 37 ◦ C for 32 min, or alpha smooth muscle actin (GA61161-2, Agilent ATH [13], here, we aimed to evaluate the role of miR-125b in regulating Technologies, Inc., Glostrup, Denmark) at 37 ◦ C for 20 min. Next, the SR-B1-mediated macrophage as well as VSMCs cholesterol efflux to HDL slides were counterstained with hematoxylin for 12 min and a bluing particles in vitro and the entire macrophage-specific reverse cholesterol reagent for 4 min. transport (RCT) pathway in vivo. 2.4. Cell culture and miRNA mimic transfections 2. Materials and methods Human monocytic (THP-1) and mouse macrophages (J774A.1 and The data, analytical methods, and study materials that support the RAW264.7) as well as mouse vascular smooth muscle (VSMC, CRL- findings of this study will be available to other researchers upon 2797) cells were obtained from American Type Tissue Collection reasonable request. A detailed research materials is listed in the online (ATTC).CHO (Chinese-hamster ovary) cells were kindly provided by Dr. Supplemental materials. Joerg Heere. The THP-1 cells were maintained in RPMI 1640 media supplemented with 10% fetal bovine serum (FBS) (Pan Biotech, Aiden­ 2.1. Patients and histology bach, DE) and 1% penicillin-streptomycin (Dominique Dutscher, Bru­ math, France) at 37 ºC and 5% CO2. THP-1 differentiation into This study analyzed blood and samples from ascendant aortas ob­ macrophages was induced using 50 ng/mL PMA for 24 h. The J774A.1 tained in the surgery scenario. Samples and data from the patients cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) included in this study were provided by the Biobank HUB-ICO-IDIBELL supplemented with 10% FBS and 1% penicillin-streptomycin. The (PT17/0015/0024), which is integrated within the Spanish National VSMCs were maintained in DMEM supplemented with 10%FBS and Biobanks Network. The study was approved by the ethics committee of 0.2% G418 antibiotic. The CHO cells were maintained in F-12 media Bellvitge University Hospital (PR149/14, approved on May 19th, 2016) (HAM) supplemented with 10% FBS and 1% penicillin-streptomycin. and all participants provided their written informed consent. The de­ The THP-1 and J774A.1 macrophages, along with the VSMCs were mographic characteristics have been previously published [14]. seeded at a density of 1.5 × 106, 1 × 106, and 2 × 105 cells per 6-well plates, respectively, and transfected the following day with 40 nM of 2.2. Histology and in situ hybridization control mimic (CM) or hsa-miR-125b-5p mimic (miR-125b) for 48 h utilizing the RNAimax Transfection Reagent (Invitrogen, Carlsbad, CA) The human aorta samples were fixed in 10% neutral buffered according to previously established protocols [16]. formalin for two days, followed by one day in 15% sucrose. The typical Human aortic cell and blood outgrowth endothelial cells (BOEC) protocol for embedded tissues in paraffin was performed as well as that isolation and culturing are detailed in the on line Supplemental for hematoxylin-eosin (HE) staining. For antagomir fluorescence in situ materials. hybridization (AISH), fresh-frozen human aorta samples were fixed in 4% paraformaldehyde (PFA) and embedded in OCT. Tissue sections of 8 2.5. Dual luciferase assays µm were cut by using a microtome-cryostat and stored on SuperFrost slides at − 80 ◦ C until the AISH was performed. The cryosections were In order to determine whether SCARB-1 3′ UTR activity is regulated thawed 10 min at room temperature (RT), dried for 10 min at 50◦ C, and by miR-125b, RAW264.7 cells (2 × 105 cells/well) were seeded sepa­ then fixed in 4% PFA for 10 min followed by two washes with DEPC- rately in 96-well opaque plates (#6005660; PerkinElmer, Massachu­ treated 1 × PBS at RT for 3 min in a Coplin jar with agitation. An in setts, USA) overnight to reach 80–90% confluence. Twenty-four hours situ hybridization assay was performed as previously described [15]. after seeding, the cells were transfected with 100 ng pMirTarget reporter Briefly, 5′ -fluorescein Cy5 probes (25 nM AntagomiR-125b or 25 nM constructs, both p_SCARB1. WT with the wild type 3′ UTR region of AntagomiR Cel-miR-67 scrambled control provided by Dharmacon) SCARB1 (#SC211509; Origene pSCARB1–3′ UTR SCARB1 (NM_005505) were added to 10 μL of IQFISH Hybridization Buffer. The probe mixture Human 3′ UTR Clone) and p_SCARB1. Mut with the mutated miR-125b was added to the tissue sample and covered with a coverslip, incubated predicted binding site from SCARB1 3′ UTR (#SC211509_CW306882 on the hybridizer for 90 min at 70 ◦ C (temperature was adjusted Origene Technologies, Maryland USA) alone as reference controls or 2 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 together with 40 nM mirVana hsa-miR-125b-5p miRNa mimic fluorescence)] × 100. (#4464066-MC10148; Thermofisher Scientifics, Massachusetts USA) using 0.3 μL TransIT-X2 (#6004 MirusBio, Wisconsin USA) according to the manufacturer’s protocol. Forty-eight hours after transfection, when 2.8. In vitro cholesterol influx assay the cells reached the desired 100% confluence, the luminescence ac­ tivity was measured using the Luciferase Assay System (#E4530; Radiolabeled lipoproteins were prepared in a manner adapted from Promega, Wisconsin USA). Briefly, the cells were lysed with 20 μL 1 × the method developed by Cedo et al. [19]. First, 5 μCi of [1α,2α(n)-3H] Reporter Lysis Buffer. Then, the RFP fluorescence signal of the cells was cholesterol (Perkin Elmer) was mixed with absolute ethanol, and the measured with a FLUOstar OPTIMA Plate Reader (BMG LabTech) at Ex solvent was evaporated under a stream of N2. The total HDL was added 530 nm and Em 620 nm. Next, 100 μL of Luciferase Assay Reagent was to the [1α,2α(n)-3H]cholesterol-containing tubes (0.5 mL, 2.11 mg/mL added to each lysate, and the luminescence signal was immediately of apoAI) and then incubated for 4 h in a 37 ◦ C bath. The labeled HDL measured with the FLUOstar OPTIMA Plate Reader (BMG LabTech). All was reisolated by ultracentrifugation at 1.063–1.210 g/mL and dialyzed samples were measured in triplicate. For further evaluation, the lumi­ against PBS via gel filtration chromatography. Cells were incubated with nescence values were normalized to the RFP values, and the miR-125b 25 μg of HDL for 4 h in a free medium. The percentage of cholesterol mimic values were normalized to the p_SCARB1 controls. infflux was calculated by measuring the radioactivity in the cells after adding 0.2 N NaOH [17]. 2.6. In vitro cholesterol efflux radioactivity assays HDL (1.063–1.210 g/mL) was isolated from the normolipidemic 2.9. In vivo macrophage-specific reverse cholesterol transport plasma via sequential ultracentrifugation (using potassium bromide for density adjustment) at 100,000 × g for 24 h with an analytical fixed- All animal procedures were conducted in accordance with the pub­ angle rotor (50.3, Beckman Coulter, Fullerton, CA, USA). The cellular lished regulations and reviewed and approved by the Institutional Ani­ cholesterol efflux was determined after seeding the 1 × 106 human THP- mal Care Committee of the Institut de Recerca of the Hospital de la Santa 1 and 0.5 × 106 mouse J774A.1 macrophage, as well as the 0.2 × 106 Creu i Sant Pau (ref. 10626). We used 24-week-old male C57BL/6 mice mouse VSMC, respectively, per well in 6-well plates; these were allowed fed with a Western-type diet (TD.88137, Envigo, Indianapolis, IN, USA) to grow for 24 h in a complete RPMI 1640 medium containing 2 mmol/L containing 21% fat and 0.2% cholesterol for four weeks. The mouse L-glutamine (Pan Biotech, Aidenbach, DE) supplemented with 10% FBS J774A.1 macrophages were cultured in 75 cm2-tissue culture plates in a and 100 U/mL penicillin/streptomycine. At that time point, the cells RPMI-supplemented medium. The cells were then transfected with 40 were transfected with 40 nM of CM or miR-125b for 48 h. Then, the cells nM of CM or miR-125b for 48 h. After that, the cells were incubated for were labeled for 24 h in the presence of 1 µCi/well of [1α,2α(n)-3H] 24 h in the presence of 5 µCi/mL of [1α,2α(n)-3H]cholesterol and 10% cholesterol (Perkin Elmer, Boston, MA, USA) and 5% FBS, as previously inactivated human lipoprotein-depleted serum (LPDS). These cells were reported [17]. The cells were then equilibrated with 0.2% bovine serum washed, equilibrated with a medium containing 0.2% bovine serum albumin (Sigma-Aldrich) in the medium for 1 h and incubated for 4 h albumin, detached via scraping, resuspended in PBS, and pooled before with the total HDL (25 µg/mL of APOA1) in the free medium. The being intraperitoneally (i.p.) injected into the mice (2.76 × 106 mac­ radioactivity was then measured in both the medium and the cells, and rophages transfected with CM and containing 4.53 ×106 cpm per mouse; the percentage of cholesterol efflux was calculated [17]. This experi­ and 3.05 × 106 macrophages transfected with miR-125 and containing ment was also performed after incubating purified APOA1 (25 µg/mL) 4.62 ×106 cpm per mouse); the cell viability was 85% in both macro­ with macrophages after stimulating the ABCA1-dependent efflux phages, as measured via trypan blue staining [17]. The mice were then pathway by treating the cells with 0.3 mmol/L 8-(4-Chlorophenylthio) individually housed, and their stools were collected over the next 48 h. adenosine 3′ ,5′ -cyclic monophosphate (pCPT-cAMP, Sigma-Aldrich) Serum radioactivity was determined via liquid scintillation counting. during the equilibration period. The [3H] HDL-cholesterol levels were measured in plasma obtained after the precipitation of APOB-containing lipoprotein particles. Liver and 2.7. In vitro cholesterol efflux fluorescent assay fecal lipids were extracted with isopropyl alcohol-hexane (2:3, v:v) and the [3H]tracer detected in the cholesterol and fecal bile acids was The in vitro cellular cholesterol efflux was also determined using determined as previously described [19]. TopFluor-cholesterol, a fluorescent cholesterol probe in which the cholesterol molecule is linked to a boron dipyrromethene difluoride (BODIPY) moiety (Avanti Polar Lipids, Alabaster, AL, USA). The 2.10. RNA isolation and quantitative real-time PCR J774A.1 cells (1 × 105/well) were seeded in 48-well plates and allowed to grow for 24 h in a DMEM-supplemented medium. The cells were The total RNA was isolated from the cells using TRIzol Reagent transfected with CM or miR-125b for 48 h as previously explained in (Invitrogen, Carlsbad, CA) according to the manufacturers protocol and Section 2.4. Next, the macrophages were labeled for 1 h in a high- purified using the EZ-10 DNAaway RNA Miniprep Kit (Bio Basic, glucose DMEM medium (Lonza, Waltham, MA, USA) containing Markham, Canada). One microgram of total RNA was reverse- 0.125 mmol/L total cholesterol, where the fluorescent cholesterol transcribed using EasyScript First-Strand cDNA Synthesis SuperMix accounted for 20% of the total cholesterol complexed with 10 mmol/L (Transgen Biotech, Beijing, China) following the manufacture’s proto­ methyl-β-cyclodextrin (Sigma-Aldrich, Madrid, Spain), as reported pre­ col. Quantitative real-time PCR amplification was performed in dupli­ viously [18]. The labeled cells were subsequently equilibrated for 18 h cates using the GoTaq(R) Probe qPCR Master Mix (Promega, Madison, with DMEM containing 0.2% fatty-acid free BSA (Sigma-Aldrich, WI). Specific TaqMan probes (Applied Biosystems, Foster City, CA) were Madrid, Spain) and then incubated for 24 h with the total HDL (25 used for human SCARB1 (Hs00969827_m1) and mouse Scarb1 µg/mL of APOA1) in the free medium. The fluorescence intensity was (Mm00450236_m1); for human ABCG1 (Hs00245154_m1) and mouse then measured in the medium using the microplate reader, Synergy HT Abcg1 (Mm00437390_m1); and human (Hs03928985_G1) or mouse (BioTek Instruments, Winooski, VT, USA), at λEx/Em = 485/530 nm. Rn18s rRNA (Mm03928990_g1) were used as internal control genes. The The cells were solubilized with 1% cholic acid and mixed on a plate reactions were run on a the CFX96TM Real-Time System (Bio-Rad, shaker for 4 h at room temperature, and the fluorescence intensity was Hercules, CA, USA). The relative mRNA expression levels were calcu­ quantified. The cholesterol efflux capacity was calculated according to lated using the ΔΔCt method. The mRNA levels were normalized to 18S the formula: [media fluorescence/(media fluorescence + cells rRNA. 3 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 2.11. miRNA quantification 3. Results For miRNA quantification, TaqMan mature miRNA detection kits 3.1. Hsa-miR-125b expressed in the macrophages and VSMCs of the were used. Primers specific for human and mouse miR-125b (hsa-miR- arteries obtained from patients with CAD and targeted the 3′ -UTR of 125b-5p, assay ID 002198, TaqMan microRNA Assays, ThermoFisher SCARB1 mRNA Scientifics, MA USA) were used for cDNA production and qRT-PCR analysis and normalized to U6 as a housekeeping gene. The miR-125b We previously reported the upregulation of hsa-miR-125b in quantification and quantitative real-time PCR were performed in atherosclerotic plaques from abdominal aortas obtained from deceased duplicate using TaqMan RT mix on an iCycler real-time detection system patients [10]. Here, we first validated these results in a second inde­ (Eppendorf). pendent cohort of patients with CAD by testing the hsa-miR-125b expression in ascendant aortas collected in the surgery scenario 2.12. Western blot analysis (representative histological sections from those aortas can be found in Fig. IA–B in the online Supplemental materials). As shown in Fig. 1A, The human THP-1 and mouse J774.1macrophages as well as the hsa-miR-125b was up-regulated, confirming that miR-125b may actu­ VSMC cells were lysed in ice-cold buffer containing 50 mM Tris–HCl, pH ally be activated during ATH progression. We next aimed to identify 7.5, 0.1% SDS, 0.1% deoxycholic acid, 0.1 mM EDTA, 0.1 mM EGTA, 1% which cells were overexpressing hsa-miR-125b in the vascular lesions by NP-40, 5.3 mM NaF, 1.5 mM NaP, a 1 mM orthovanadate and 1 mg/mL performing in situ hybridization for hsa-miR-125b in three independent of protease inhibitor cocktail (Roche Diagnostics, Basel, Switzerland), aorta samples from ascendant aortas obtained during cardiac surgery and 0.25 mg/mL AEBSF (Roche). Cell lysates were centrifuged at (Fig. 1B–C). We found that hsa-miR-125b was predominantly expressed 12,000 g for 15 min at 4 ◦ C, and the protein concentrations of the su­ in macrophages (Fig. 1B–C) and VSMCs (Fig. 1C). More detailed images pernatants were determined using the BCA Protein Assay Reagent Kit of a single Z-section of monocytes labeled with a Cy5-conjugated (A53227, Thermo Scientific, Waltham, MA, USA). The protein extracts anti-hsa-miR-125b antagomir in a vasa vasorum from the aortic were mixed with a corresponding volume of 4 × Laemmli loading buffer adventitia are shown in Fig. 1B. To confirm that the cells were mono­ containing 62.5 mM Tris-HCl pH 6.8, 2% SDS, 25% glycerol, and 0.01% cytes, staining of the correlative section was performed with anti-CD68 bromophenol blue and heated at 94 ºC for 4 min. Then, 20 μg of total antibody (Fig. 1B). We also analyzed the expression of miR-125b in protein from each sample was separated onto a 7.5% TGX Stain Free BOECs and ECs; miR-125b was not expressed in these cells (data not PreCast Gel (4568026, Bio-Rad, Carson, CA, USA) and transferred to a shown). 0.22 µm PVDF membrane (1704157, Bio-Rad, Carson, CA, USA). Since macrophage cholesterol efflux is essential for homeostasis and Nonspecific sites on the membrane were blocked with either 3% dried critical in ATH pathogenesis [2], we focused on these cells and we milk in Tris-buffered saline containing 0.05% Tween-20 (TBST buffer) performed a bioinformatics analysis to identify potential hsa-miR-125b for 15 min. Membranes were incubated overnight at 4 ºC with optimized target genes related to cholesterol metabolism using the TargetScan dilutions of the primary antibody SCARB1 (Abcam 36970, Cambridge, v.7.2 data base. We detected that SCARB1, LIMA1 (LIM Domain and UK), of 1:1000 or, for ABCG1 (Novus Biologicals, Centennial, CO, Acting Binding 1, NCBI = 51474) and VPS4B (Vacuolar Protein Sorting NB400-132SS), of 1:500. The following day, the membranes were 4B, NCBI = 9525) had predicted binding sites for hsa-miR-125b. Spe­ washed with the TBST buffer and reincubated with IgG HRP-conjugated cifically, we focused on studying SCARB1, since its importance of anti-rabbit secondary antibody (W401B, Promega, Madison, USA) for 1 SCARB1 during cholesterol efflux is well known in addition to its h. After that, the membranes were washed with the TBST buffer and dysfunction that predispose to the pathogenesis of ATH. The human revealed using the Immun-Star WesternC Chemiluminescence Kit (170- SCARB1 3′ UTR gene has only one computationally predicted miR-125b 5070, Bio-Rad, Carson, CA). Imaging and data analysis: TGX Stain-Free binding site in the second 3′ UTR (Fig. 2A), and, for the murine Scarb1 gels were activated for 1 min after SDS–electrophoresis, and images gene, there are three predicted miRNA binding sites in the three were captured using ChemiDOC™XRS Gel Documentation Systems (Bio- different 3′ UTR variants (Fig. IIA–C in the online Supplemental mate­ Rad, Hercules, CA) and ImageLab software (version 6.0.1, Bio-Rad, rials). To directly assesses whether the hsa-miR-125b targets the 3′ UTR Richmond, CA). Data normalization analysis was performed using the of SCARB1, we transfected mouse macrophage cells with a reporter stain free images. The approximate molecular weights of the target construct with the luciferase coding sequence fused to the 3′ UTR of proteins were determined via comparison with a prestained protein SCARB1. Interestingly, hsa-miR-125b markedly repressed the activity of ladder (ab116028, abcam, Cambridge, UK). the SCARB1 3′ UTR reporter construct (Fig. 2B). Importantly, when we mutated the binding site of miR-125b in the 3′ UTR of SCARB1, the 2.13. Statistical analysis luciferase activity increased (Fig. 2B), thereby providing direct evidence that miR-125b regulates SCARB1 via directly binding to SCARB1-3′ UTR. The in vitro experiments were routinely repeated at least three times Furthermore, we observed that miR-125b was inversely related to but — (with three replicates per experiment) unless otherwise noted. The data not significantly associated with—SCARB1 expression in the ascendant is presented as the mean ± SEM (n is noted in the figure legends). aortas of the patients (Fig. 2C). D′ Agostino & Pearson and Shapiro-Wilk normality tests were performed to identify Gaussian distributions. When the data followed a Gaussian 3.2. miR-125b downregulates macrophage SCARB1 and impairs HDL- distribution, Student’s t-test was used to compare the differences be­ mediated macrophage cholesterol efflux and macrophage-to-feces RCT in tween two groups. The number of mice per group was estimated vivo considering α = 0.05, power = 80%, and an effect size of 0.5% in the fecal excretion of the macrophage-derived tracer. The subjects were We next investigated the physiological role of miR-125b in vitro and randomized into groups and all animals were included in the analyzes. in vivo. Specifically, in order to establish whether the increased The experiments were conducted blinded regarding the origin of the expression of miR-125b in macrophages would directly regulate the specimens in order to reduce bias. Significance was accepted at the level expression of SCARB1 and α-HDL-mediated macrophage cholesterol of p < 0.05. GraphPad Prism version 8.0 for Windows (GraphPad Soft­ efflux in vitro, we transfected human THP-1 monocyte–derived macro­ ware, San Diego, CA) was used to perform all statistical analyses. phages, J774A.1 mouse macrophages and VSMCs with miR-125b. Gene and protein expression analysis demonstrated that the overexpression of hsa-miR-125b significantly reduced the SCARB1 mRNA and protein levels in the three cell types studied (Fig. 3A–C, left showing mRNA and 4 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 Fig. 1. Aortic miR-125b expression in patients with CAD and the regulation of miR-125b accessibility to the 3′UTRs of the different scarb1 transcript variants. (A) Boxplot of hsa-miR-125b expression in the aortas of patients with CAD (n = 33) vs those without CAD (non-CAD) (n = 5). Hsa-miR-125b expression was normalized to U6 levels. (B) Representative in situ single Z-section of a vase from the vasa-vasorum of the aortic adventitia showing a monocyte adhered to ECs labeled with 25 mM of a Cy5-conjugated anti-miR-125b antagomir and their scrambled negative control (right image). Nuclei were stained in blue with 4,6-diami­ dino-2-phenylindole (DAPI), and miR-125b was stained in red with Cy5. Staining with anti-CD68 antibody reveal their monocyte origin. (C) Representative transverse histological sections from an ascending aortic wall of a patient with CAD and stained with H&E (left panel) where the in situ hybridization was performed (middle panel) and alpha-actin showing a diffuse intimal thickening characterized by an increase in fusiform cells (right panel). The black arrows indicate the tunica intima (Ti), tunica media (Tm), and tunica adventitia (Ta). Scale bars represents 100 µm. A fluorescent in situ hybridization (AISH) (middle panel) shows the localization of miR-125b in the cytoplasm of human monocyte (Mc) and VSMCs. The white arrows indicate positive cells for miR-125b. Scale bars represents 10 µm. upper panel showing protein). However, miR-125b expression did not 4. Discussion significantly affect the abcg1 mRNA and protein levels (Supplementary Fig. III in the online Supplemental materials). Also, we found very low Our study was originally set up to deepen understanding of the levels of macrophage abca1 mRNA, and protein was not detected by mechanism by which miR-125b affects the pathogenesis of ATH. We western blot in these non-cholesterol-loaded cells. This reduction in previously detected that miR-125b was upregulated in a mouse model of SCARB1 mRNA and protein level was associated with an impaired ATH and in the abdominal aorta of deceased patients [10]. Here, in a macrophage radioactive cholesterol efflux to α-HDL (Fig. 3A–C, right). second independent cohort, we validated this previous result, and, in the This was also confirmed via TopFluor-cholesterol efflux assay, where the ascendant aortas of patients with ischemic heart disease, we detected cholesterol efflux in J774 cells transfected with CM was 19.40% vs that the expression of miR-125b was also highly up-regulated. 14.75% in cells transfected with miR-125b. Furthermore, the sole miR-125b has also been reported to be highly expressed in vascular stimulation of ABCA1 by pCPT-cAMP did not affect the cholesterol efflux murine ECs [20], and it was also physiologically expressed in murine promoted by lipid-free ApoA1 (Fig. 3D). Importantly, the miR-125b steroidogenic cells, although at much lower levels than miR-125a, which expression did not affect the α-HDL-mediated cholesterol efflux from targeted the SCARB1 gene and downregulated the SR-B1-mediated se­ CHO cells expressing SCARB1, which does not contain the target SR-B1 lective HDL cholesterol esters uptake [12]. However, human ECs and 3′ -UTR sequence (Fig. 3E). BOECs did not express miR-125b, rather indicating a differential We also investigated the potential of miR-125b in regulating expression of this miRNA across both species and cell types. Since macrophage cholesterol influx. The total HDL was labeled and incubated vascular ECs and circulating BOECs from patients expressed SCARB1 but with J774.1 macrophages after transfection with CM or miR-125b. The not miR-125b, we focused our study on macrophages and VSMCs. Thus, percentage of cholesterol influx was similar in the CM and miR-125b our study extended on the regulating mechanism of miR-125b and transfected cells (4.87% ± 0.18 vs 5.10% ± 0.24, respectively). predicted the miRNA binding sites of the 3′ UTR SCARB1 gene, which Next, we investigated the role of the macrophage expression of miR- was validated via luciferase assays in macrophages. However, although 125b in RCT in vivo. [3H]cholesterol-labeled J774A.1 mouse macro­ hsa-miR-125b was inversely associated with SCARB1 expression in the phages transfected with CM or hsa-miR-125b were injected i.p. into human aortas, this pattern did not reach statistical significance. It is wild-type C57BL/6 mice fed a Western-type diet for four weeks. The possible that the aortic cell diversity and wide range of SCARB1 plasma [3H]tracer levels in the mice injected with the hsa-miR-125b expression in multiple cell types [3] could partly attenuate the inverse transfected macrophages were decreased at 4, 24 and 48 h after injec­ association between hsa-miR-125b and SCARB1, since SCARB1 is tion and these changes were mainly due to lower levels of HDL- expressed in ECs but not miR-125b. associated [3H]cholesterol (Fig. 4A–B). Importantly, the overall trans­ We further determined whether the increased expression of hsa-miR- fer of cholesterol from the miR-125b-transfected macrophages to the 125b resulted in lower endogenous SR-BI expression levels and the liver and feces was significantly impaired when compared to that of the altered cholesterol efflux in human and mouse macrophages, as well as CM-transfected macrophages (Fig. 4C–D). VSMCs. Since the cholesterol-laden macrophages showed lower SR-BI 5 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 Fig. 2. Regulation of miR-125b accessibility to the 3′UTRs of the different scarb1 transcript variants. (A) Structure of the 3′ terminal region of the human SCARB1 gene. Boxes identify exons 11–13, the two UTRs, and their relative positions in the original genomic sequence (NC_000012.12). Interrupted black lines indicate intronic sequences. Splicing variants are color-coded. Colored doted lines linking the exons show alternative splicing events. Vertical colored bars in exon 11 show the cryptic splice sites predicted in this exon. Colored solid horizontal lines in both 3′ UTRs show the lengths for each splicing variant of SCARB1. Red circles indicate the position of the functional stop-codons. Inverted blue triangles indicate the positions of the polyadenylation signals (PAS). Also shown is the binding site for hsa-miR-125b in the 3′ UTR of SCARB1 variant 1 (seed sequence in bold). This genomic map was drawn by blasting the following cDNA sequences—SCARB1 variant 1 (NM_005505/ENST00000261693), SCARB1 variant 2 (NM_001082959/ENST00000339570), SCARB1 variant 3 (NM_001367981/ENST00000415380), and SCARB1 variant 4 (NM_001367982/ENST00000541205)—against the GRCh38.p13 primary assembly of the human genome. Diagram not drawn to scale, although the relative lengths of the exons have been maintained. (B) Upper panel, schematic representation of the predicted miR-125b-5p binding site on SCARB1 3′ -UTR (bold and underlined) and the mutated SCARB1 3′ -UTR with the deletion of eight nucleotides in the predicted miR-125b-5p binding site (strikethrough). Normalized luciferase activity 48 h after transfection of RAW264 cells that were co-transfected with pMirTarget vector containing the wild type 3′ -UTR of SCARB-1 (p_SCARB1. WT) or the mutated 3′ -UTR of SCARB-1 (p_SCARB1.Mut) in the 3′ -UTR of the luciferase gene together with hsa-miR-125b-5p mimic. Luciferase values for the 3′ -UTRs are normalized to the p_SCARB1 control plasmids (non-transfected miRNA mimic). The mean from the three replicates shows significant differences between the has- miR-125b-5p transfected wild type 3′ -UTR SCARB1 plasmid and mutated 3′ -UTR SCARB1 plasmid. The error bars represent the standard error of the mean (SEM). Significance was accepted at the level of p < 0.05, * compare to WT and & compare to Mutated. (C) Negative correlation between the expression of hsa-miRNA-125b (y-axis) and SCARB1 (x-axis) in the arteries of the patients. Statistics: A Shapiro-Wilk normality test was performed to test for Gaussian distribution, and the dif­ ferences between the groups were determined by an unpaired t-test. Pearson correlation (R) and the p-value were calculated for the C panel. levels—whereas the ABCA1- and ABCG1-dependent pathways were be noted that the downregulation of both transporters in the upregulated [19,21],—we used cholesterol-unloaded macrophages. To non-cholesterol-loaded J774.1 mouse macrophages did not affect the best of our knowledge, we showed for the first time that miR-125b cholesterol efflux stimulated by HDL [19], thus highlighting the minor overexpression reduced the macrophage and VSMC SCARB1 mRNA contribution of ABCA1/G1-dependent pathways in and protein levels, leading to an impaired α-HDL-mediated macrophage non-cholesterol-loaded macrophages. Furthermore, the miR-125b cholesterol efflux in vitro. In contrast, we did not find significant alter­ expression did not regulate the ABCA1-mediated efflux to lipid-free ations in the ABCG1 levels, and macrophage ABCA1 was very lowly apoA-I in pCPT-cAMP-treated macrophages. In line with these find­ expressed in these non-loaded-cholesterol macrophages. It should also ings, the rate of cholesterol efflux to mature α-HDL particles—but not 6 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 Fig. 3. miR-125b downregulates macrophage SR-B1 and impairs HDL-mediated macrophage cholesterol efflux. (A–C) Protein (upper panels) and qRT-PCR analysis (left) of SCARB1 and cholesterol efflux (right) in (A) THP-1, (B) J774A.1, and (C) SMC cells transfected with CM or hsa-miR-125b. Western blot analysis of SCARB1 in the upper panels. The data normalization analysis was performed with stain free images. Representative experiment out of 2–3 experiments with similar results. qPCR data is presented as the mean ± SEM. Cholesterol efflux data is the mean ± SEM and representative of ≥ 2–3 experiments. Statistics: A Shapiro-Wilk normality test was performed for Gaussian distribution, and an unpaired t-test was performed to compare CM and miR-125b transfection. (D) Cholesterol efflux to ApoA1 from J774A.1 cells transfected with CM or miR-125b. (E) Cholesterol efflux to HDL from CHO cells expressing SCARB1 (CHO-SR-BI) transfected with CM or miR-125b. Data is the mean ± SEM and representative of ≥ 2–3 experiments in triplicate. that to lipid-free apoA1—was significantly lower in the peritoneal and it does not provide an assessment of cell cholesterol unloading or the macrophages isolated from SR-BI knockout mice [22,23]. Additionally, cholesterol influx rate. However, we performed cholesterol influx ex­ we also found that miR-125b-expressing mouse macrophages impaired periments with radiolabel HDL in vitro and there were no differences the overall macrophage-specific RCT in vivo in WT mice fed a between CM and miR-125b transfected cells. These in vivo models were Western-type diet; this was mainly due to a defective HDL-mediated mimicking a single—but significant—pathogenic aspect of HDL- macrophage cholesterol efflux at all time points that were studied. mediated ATH. Indeed, the macrophage SR-B1 function within ATH However, cholesterol efflux to either human HDL3 or mouse whole has been reported to be highly dependent on the lipid context, indicating serum was not affected in bone-marrow-derived macrophages from that cholesterol efflux may be more critical in advanced foam cell for­ SR-B1-deficient mice and, consequently, neither in vivo macrophage RCT mation [3]. was altered [21]. In contrast with this latter study, we used WT mice fed Overall, the results of our study support the concept that miRNA- a Western-type diet, which enhances both the diffusional component of 125b downregulates SR-B1 in human and murine macrophages, as macrophage cholesterol efflux and the whole macrophage-to feces RCT well as in VSMCs, thereby impairing macrophage cholesterol efflux in pathway [24]. Therefore, the relative importance of impaired vitro and the whole macrophage-specific RCT pathway in vivo. These SR-B1-mediated cholesterol efflux becomes quite apparent under con­ results also indicate a mechanism by which macrophage miR-125b may ditions where the overall RCT pathway is stimulated, as also occurred promote foam cell formation and ATH. with cholesterol efflux when the peritoneal macrophages of hyper­ Interestingly, miR-125a, which belongs to the same miRNA family ( lipidemic mice were defective for both SR-B1 and ABCA1 [23]. and has an identical seed sequence) as miR-125b but has a different It should be noted that this in vivo RCT method was designed to genomic location, is induced by oxLDL stimulation in monocyte/mac­ evaluate the unidirectional transport of cholesterol from macrophages rophages and inhibits lipid uptake and the inflammatory cytokine 7 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 Fig. 4. miR-125b impairs HDL-mediated macrophage cholesterol efflux and macrophage-to-feces RCT in vivo. (A) Individually-housed wild-type mice were i. p. injected with [3H]cholesterol-labeled mouse macrophages transfected with CM or hsa-miR-125, and serum [3H]cholesterol was determined at the indicated times. (B) HDL- and APOB-lipoproteins-associated [3H]cholesterol at 48 h (for clarity only, HDL-associated [3H]cholesterol points are shown). (C) Hepatic [3H]cholesterol at 48 h. (D) [3H]cholesterol and [3H]tracer from fecal bile acids of feces collected over 48 h. Values are expressed as the mean ± SEM of 7 and 5 wild-type mice injected with CM or hsa-miR-125 transfected [3H]cholesterol-labeled mouse macrophages, respectively. The amount of [3H]tracer was expressed as the fraction of the injected dose. Statistics: A Kolmogorov-Smirnov normality test was performed to test Gaussian distribution. (A-B) Repeated measures two-way ANOVA followed by a Bonferroni multiple-comparisons test was performed to compare the serum [3H]cholesterol levels between the two groups in each time point and HDL- and APOB-lipoproteins-associated [3H]cholesterol at 48 h. Two-way ANOVA results were: time, p < 1 × 10− 4; and macrophage type, p = 5.1 × 10− 3 and, at 48 h, li­ poprotein distribution, p < 1 × 10− 4; and macrophage type, p = 0.0165; APOB-lipoproteins-associated [3H]cholesterol at 48 h were similar in both groups, p = 0.6883. (C-D) An unpaired t-test was performed to compare the liver and total fecal [3H]cholesterol tracer (bile acids + cholesterol) between the two groups. Significance was accepted at the level of p < 0.05. secretion, suggesting its potential role in ATH inflammatory responses Santos: Methodology, Validation. Marina Canyelles: Methodology, [25]. Moreover, miR-125a and miR-125b have been also identified in Validation. Josep Julve: Writing – review & editing. Joan Carles VSMCs, as we confirmed in our patients, and has been suggested to play Escolà-Gil: Conceptualization, Methodology, Validation, Formal anal­ a role in vascular pathogenesis regulating vasoconstrictive gene ysis, Investigation, Resources, Writing – original draft, Writing – review expression and affecting the synthesis and secretion of many vasoactive & editing. Noemi Rotllan: Conceptualization, Methodology, Valida­ substances [20]. Thus, future experiments will be needed to elucidate tion, Formal analysis, Investigation, Resources, Writing – original draft, the role of miR-125a and miR-125b during the pathogenesis of ATH at Writing – review & editing. the inflammatory and vascular level and shed lights onto the therapeutic potential of suppressing miR-125 in vivo. Author Contributions CRediT authorship contribution statement MH, NR, and JC-E conceived and designed the study and wrote the manuscript. MH, RG, AM, EN, EP, MG, GR, DS, MC, JJ, JC-EG and NR Miguel Hueso: Conceptualization, Methodology, Validation, Formal performed experiments and analyzed data. FS and AB provided the aorta analysis, Investigation, Resources, Writing – review & editing, Funding samples. All authors have read and agreed to the published version of aquisition. Raquel Griñán: Methodology, Validation, Formal analysis. the manuscript. Adrián Mallen: Methodology, Validation, Formal analysis. Estanislao Navarro: Conceptualization, Methodology, Validation, Formal analysis, Funding Investigation. Elvira Purqueras: Methodology, Validation, Formal analysis. Montse Gomá: Methodology, Validation, Formal analysis. This work was funded by Ministerio de Ciencia, Innovación y Uni­ Fabrizio Sbraga: Resources. Arnau Blasco-Lucas: Resources. Gio­ versidades (PID2019-104367RB-100), as well as from the Agencia vanna Revilla: Methodology, Validation, Formal analysis. David Estatal de Investigación (AEI/10.13039/501100011033) within the 8 M. Hueso et al. Biomedicine & Pharmacotherapy 146 (2022) 112596 Subprograma Ramón y Cajal (RYC-201722879) to NR. This study has Lipids Genetics, Rare variant in scavenger receptor BI raises HDL cholesterol and increases risk of coronary heart disease, Science 351 (2016) 1166–1171. been funded by the Instituto de Salud Carlos III (Co-funded by the Eu­ [7] A. Helgadottir, P. Sulem, G. Thorgeirsson, S. Gretarsdottir, G. Thorleifsson, B. ropean Regional Development Fund (ERDF), a way to build Europe) O. Jensson, G.A. Arnadottir, I. Olafsson, G.I. Eyjolfsson, O. Sigurdardottir, through the project PI 18/01108 (to M.H), Rio Hortega contract CM20/ U. Thorsteinsdottir, D.F. Gudbjartsson, H. Holm, K. Stefansson, Rare SCARB1 00033 (to M.C) and Grant PI17/00232 (to JJ). This work was funded mutations associate with high-density lipoprotein cholesterol but not with coronary artery disease, Eur. Heart J. 39 (2018) 2172–2178. also by FEDER “Una manera de hacer Europa”, and by Fundació La [8] D.A. Chistiakov, A.A. Melnichenko, V.A. Myasoedova, A.V. Grechko, A.N. Orekhov, Marató de TV3 2016 (303/C/2016) (201602.30.31) (to JJ). JJ was Mechanisms of foam cell formation in atherosclerosis, J. Mol. Med. 95 (2017) recipient of a Miguel Servet Type 2 contract (CPII18/00004; ISCIII). 1153–1165. [9] K.H. Lao, L. Zeng, Q. Xu, Endothelial and smooth muscle cell transformation in CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM) atherosclerosis, Curr. Opin. Lipidol. 26 (2015) 449–456. is a project of Instituto de Salud Carlos III. Institut de Recerca de [10] M. Hueso, L. De Ramon, E. Navarro, E. Ripoll, J.M. Cruzado, J.M. Grinyo, J. Torras, l′ Hospital de la Santa Creu i Sant Pau and Institut d′ Investigació Bio­ Silencing of CD40 in vivo reduces progression of experimental atherogenesis through an NF-kappaB/miR-125b axis and reveals new potential mediators in the mèdica de Bellvitge-IDIBELL are accredited by the Generalitat de Cata­ pathogenesis of atherosclerosis, Atherosclerosis 255 (2016) 80–89. lunya as a Centre de Recerca de Catalunya (CERCA). [11] T. Huan, J. Rong, C. Liu, X. Zhang, K. Tanriverdi, R. Joehanes, B.H. Chen, J. M. Murabito, C. Yao, P. Courchesne, P.J. Munson, C.J. O’Donnell, N. Cox, A. D. Johnson, M.G. Larson, D. Levy, J.E. Freedman, Genome-wide identification of Declaration of Interest microRNA expression quantitative trait loci, Nat. Commun. 6 (2015) 6601. [12] Z. Hu, W.J. Shen, F.B. Kraemer, S. Azhar, MicroRNAs 125a and 455 repress All the authors declare that they do not have financial/personal in­ lipoprotein-supported steroidogenesis by targeting scavenger receptor class B type I in steroidogenic cells, Mol. Cell Biol. 32 (2012) 5035–5045. terest or belief that could affect their objectivity. [13] K. Hirano, S. Yamashita, Y. Nakagawa, T. Ohya, F. Matsuura, K. Tsukamoto, Y. Okamoto, A. Matsuyama, K. Matsumoto, J. Miyagawa, Y. Matsuzawa, Acknowledgments Expression of human scavenger receptor class B type I in cultured human monocyte-derived macrophages and atherosclerotic lesions, Circ. Res. 85 (1999) 108–116. We thank RICORS (RD21/0005/0001) and the CERCA Program/ [14] A. Casas, A. Mallen, A. Blasco-Lucas, F. Sbraga, J. Guiteras, N. Bolanos, E. Castano, Generalitat de Catalunya for their institutional support. We want to J. Torras, J.M. Cruzado, E. Navarro, M. Hueso, Chronic kidney disease-associated particularly acknowledge the patients and the Biobank HUB-ICO- inflammation increases the risks of acute kidney injury and mortality after cardiac surgery, Int. J. Mol. Sci. 21 (2020) 9689. IDIBELL (PT17/0015/0024) integrated in the Spanish National Bio­ [15] I. Bar, A. Merhi, F. Abdel-Sater, A. Ben Addi, S. Sollennita, J.L. Canon, P. Delree, banks Network for their collaboration. We thank Benjamín Torrejón The microRNA miR-210 is expressed by cancer cells but also by the tumor Escribano from the Advanced Light Microscopy Unit in the Scientific and microenvironment in triple-negative breast cancer, J. Histochem. Cytochem. 65 (2017) 335–346. Technical Facility for his help with the confocal images. We thank to Dr. [16] L. Goedeke, N. Rotllan, A. Canfran-Duque, J.F. Aranda, C.M. Ramirez, E. Araldi, C. Joerg Heere who kindly provide us with the CHO (Chinese-hamster S. Lin, N.N. Anderson, A. Wagschal, R. de Cabo, J.D. Horton, M.A. Lasuncion, A. ovary) cells. M. Naar, Y. Suarez, C. Fernandez-Hernando, MicroRNA-148a regulates LDL receptor and ABCA1 expression to control circulating lipoprotein levels, Nat. Med. 21 (2015) 1280–1289. Appendix A. Supporting information [17] J.C. Escola-Gil, M. Lee-Rueckert, D. Santos, L. Cedo, F. Blanco-Vaca, J. Julve, Quantification of in vitro macrophage cholesterol efflux and in vivo macrophage- specific reverse cholesterol transport, Methods Mol. Biol. 1339 (2015) 211–233. Supplementary data associated with this article can be found in the [18] L. Cedo, S. Fernandez-Castillejo, L. Rubio, J. Metso, D. Santos, D. Munoz-Aguayo, online version at doi:10.1016/j.biopha.2021.112596. A. Rivas-Urbina, M. Tondo, K.A. Mendez-Lara, M. Farras, M. Jauhiainen, M. J. Motilva, M. Fito, F. Blanco-Vaca, R. Sola, J.C. Escola-Gil, Phenol-enriched virgin References olive oil promotes macrophage-specific reverse cholesterol transport in vivo, Biomedicines 8 (2020) 266. [19] L. Cedo, J. Metso, D. Santos, A. Garcia-Leon, N. Plana, S. Sabate-Soler, N. Rotllan, [1] M.R.F. Linton, P.G. Yancey, S.S. Davies, W.G. Jerome, E.F. Linton, W.L. Song, A. A. Rivas-Urbina, K.A. Mendez-Lara, M. Tondo, J. Girona, J. Julve, V. Pallares, C. Doran, K.C. Vickers, in: K.R. Feingold, B. Anawalt, A. Boyce, G. Chrousos, W. A. Benitez-Amaro, V. Llorente-Cortes, A. Perez, D. Gomez-Coronado, A. W. de Herder, K. Dhatariya, K. Dungan, A. Grossman, J.M. Hershman, J. Hofland, K. Ruotsalainen, A.L. Levonen, J.L. Sanchez-Quesada, L. Masana, P.T. Kovanen, S. Kalra, G. Kaltsas, C. Koch, P. Kopp, M. Korbonits, C.S. Kovacs, W. Kuohung, M. Jauhiainen, M. Lee-Rueckert, F. Blanco-Vaca, J.C. Escola-Gil, LDL receptor B. Laferrere, E.A. McGee, R. McLachlan, J.E. Morley, M. New, J. Purnell, R. Sahay, regulates the reverse transport of macrophage-derived unesterified cholesterol via F. Singer, C.A. Stratakis, D.L. Trence, D.P. Wilson (Eds.), The Role of Lipids and concerted action of the HDL-LDL axis: insight from mouse models, Circ. Res. 127 Lipoproteins in Atherosclerosis, Endotext South Dartmouth, MA, 2000. (2020) 778–792. [2] M.C. Phillips, Molecular mechanisms of cellular cholesterol efflux, J. Biol. Chem. [20] D. Li, P. Yang, Q. Xiong, X. Song, X. Yang, L. Liu, W. Yuan, Y.C. Rui, MicroRNA- 289 (2014) 24020–24029. 125a/b-5p inhibits endothelin-1 expression in vascular endothelial cells, [3] M. Hoekstra, SR-BI as target in atherosclerosis and cardiovascular disease – a J. Hypertens. 28 (2010) 1646–1654. comprehensive appraisal of the cellular functions of SR-BI in physiology and [21] X. Wang, H.L. Collins, M. Ranalletta, I.V. Fuki, J.T. Billheimer, G.H. Rothblat, A. disease, Atherosclerosis 258 (2017) 153–161. R. Tall, D.J. Rader, Macrophage ABCA1 and ABCG1, but not SR-BI, promote [4] A. Braun, B.L. Trigatti, M.J. Post, K. Sato, M. Simons, J.M. Edelberg, R. macrophage reverse cholesterol transport in vivo, J. Clin. Invest. 117 (2007) D. Rosenberg, M. Schrenzel, M. Krieger, Loss of SR-BI expression leads to the early 2216–2224. onset of occlusive atherosclerotic coronary artery disease, spontaneous myocardial [22] M. Van Eck, I.S. Bos, R.B. Hildebrand, B.T. Van Rij, T.J. Van Berkel, Dual role for infarctions, severe cardiac dysfunction, and premature death in apolipoprotein E- scavenger receptor class B, type I on bone marrow-derived cells in atherosclerotic deficient mice, Circ. Res. 90 (2002) 270–276. lesion development, Am. J. Pathol. 165 (2004) 785–794. [5] S.N. Koenig, H.C. Sucharski, E.M. Jose, E.K. Dudley, F. Madiai, O. Cavus, A. [23] Y. Zhao, M. Pennings, R.B. Hildebrand, D. Ye, L. Calpe-Berdiel, R. Out, M. Kjerrulf, D. Argall, J.L. Williams, N.P. Murphy, C.B.R. Keith, M. El Refaey, R.J. Gumina, K. E. Hurt-Camejo, A.K. Groen, M. Hoekstra, W. Jessup, G. Chimini, T.J. Van Berkel, D. Boudoulas, M.W. Milks, G. Sofowora, S.A. Smith, T.J. Hund, N.T. Wright, E. Van, Eck M. Enhanced foam cell formation, atherosclerotic lesion development, A. Bradley, K.M. Zareba, L.E. Wold, E.L. Mazzaferri Jr., P.J. Mohler, Inherited and inflammation by combined deletion of ABCA1 and SR-BI in bone marrow- variants in SCARB1 cause severe early-onset coronary artery disease, Circ. Res. 129 derived cells in LDL receptor knockout mice on western-type diet, Circ. Res. 107 (2021) 296–307. (2010) e20–31. [6] P. Zanoni, S.A. Khetarpal, D.B. Larach, W.F. Hancock-Cerutti, J.S. Millar, [24] J.C. Escola-Gil, G. Llaverias, J. Julve, M. Jauhiainen, J. Mendez-Gonzalez, M. Cuchel, S. DerOhannessian, A. Kontush, P. Surendran, D. Saleheen, S. Trompet, F. Blanco-Vaca, The cholesterol content of Western diets plays a major role in the J.W. Jukema, A. De Craen, P. Deloukas, N. Sattar, I. Ford, C. Packard, A. Majumder, paradoxical increase in high-density lipoprotein cholesterol and upregulates the D.S. Alam, E. Di Angelantonio, G. Abecasis, R. Chowdhury, J. Erdmann, B. macrophage reverse cholesterol transport pathway, Arterioscler. Thromb. Vasc. G. Nordestgaard, S.F. Nielsen, A. Tybjaerg-Hansen, R.F. Schmidt, K. Kuulasmaa, D. Biol. 31 (2011) 2493–2499. J. Liu, M. Perola, S. Blankenberg, V. Salomaa, S. Mannisto, P. Amouyel, D. Arveiler, [25] T. Chen, Z. Huang, L. Wang, Y. Wang, F. Wu, S. Meng, C. Wang, MicroRNA-125a- J. Ferrieres, M. Muller-Nurasyid, M. Ferrario, F. Kee, C.J. Willer, N. Samani, 5p partly regulates the inflammatory response, lipid uptake, and ORP9 expression H. Schunkert, A.S. Butterworth, J.M. Packard, G.M. Peloso, N.O. Stitziel, J. Danesh, in oxLDL-stimulated monocyte/macrophages, Cardiovasc. Res. 83 (2009) 131–139. S. Kathiresan, D.J. Rader, C.H.D.E. Consortium, C.A.E. Consortium, C. Global 9