Biomedicine & Pharmacotherapy 152 (2022) 113270 Contents lists available at ScienceDirect Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha Targeting cholesteryl ester accumulation in the heart improves cardiac insulin response Virginia Actis Dato a, b, 1, Aleyda Benitez-Amaro c, d, 1, Eduardo Garcia c, d, 1, Lene Claudi c, d, Maria Teresa LaChica Lhoëst c, d, Antoni Iborra e, Joan Carles Escola-Gil f, Jose Maria Guerra g, h, Valerie Samouillan i, Carlos Enrich j, k, Gustavo Chiabrando a, b, *, 2, Vicenta Llorente-Cortés c, d, h, **, 3 a Departamento de Bioquímica Clínica, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Argentina b Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Centro de Investigaciones en Bioquímica Clínica e Inmunología (CIBICI), Córdoba, Argentina c Institute of Biomedical Research of Barcelona (IIBB)-Spanish National Research Council (CSIC), Barcelona, Spain d Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain e SCAC, Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain f Metabolic Basis of Cardiovascular Risk, Biomedical Research Institute Sant Pau (IIB Sant Pau), Hospital de la Santa Creu i Sant Pau. CIBER de Diabetes y enfermedades Metabólicas Asociadas (CIBERDEM), Barcelona. Spain g Department of Cardiology, Hospital de la Santa Creu i Sant Pau, Biomedical Research Institute Sant Pau (IIB Sant Pau), Universitat Autonoma de Barcelona, Barcelona, Spain h CIBERCV, Institute of Health Carlos III, 28029 Madrid, Spain i CIRIMAT, Université de Toulouse, Université Paul Sabatier, Equipe PHYPOL, 31062 Toulouse, France j Unitat de Biologia Cel⋅lular, Departament de Biomedicina, Facultat de Medicina i Ciències de la Salut, Universitat de Barcelona, Barcelona, Spain k Centre de Recerca Biomèdica CELLEX, Institut d′ Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Background: Antibodies against the P3 sequence (Gly1127-Cys1140) of LRP1 (anti-P3 Abs) specifically block High-fat diet cholesteryl ester (CE) accumulation in vascular cells. LRP1 is a key regulator of insulin receptor (InsR) trafficking Lipid droplets in different cell types. The link between CE accumulation and the insulin response are largely unknown. Here, Cholesteryl esters the effects of P3 peptide immunization on the alterations induced by a high-fat diet (HFD) in cardiac insulin Heart Lipoprotein response were evaluated. Insulin Methods: Irrelevant (IrP)- or P3 peptide-immunized rabbits were randomized into groups fed either HFD or LRP1 normal chow. Cardiac lipid content was characterized by thin-layer chromatography, confocal microscopy, and electron microscopy. LRP1, InsR and glucose transporter type 4 (GLUT4) levels were determined in membranes and total lysates from rabbit heart. The interaction between InsR and LRP1 was analyzed by immunoprecipi­ tation and confocal microscopy. Insulin signaling activity and glucose uptake were evaluated in HL-1 cells exposed to rabbit serum from the different groups. Findings: HFD reduces cardiac InsR and GLUT4 membrane levels and the interactions between LRP1/InsR. Targeting the P3 sequence on LRP1 through anti-P3 Abs specifically reduces CE accumulation in the heart independently of changes in the circulating lipid profile. This restores InsR and GLUT4 levels in cardiac mem­ branes as well as the LRP1/InsR interactions of HFD-fed rabbits. In addition, anti-P3 Abs restores the insulin signaling cascade and glucose uptake in HL-1 cells exposed to hypercholesterolemic rabbit serum. * Correspondence to: Departamento de Bioquímica Clínica, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Haya de la Torre s/n, Ciudad Universitaria, 5000 Córdoba, Argentina. ** Correspondence to: Institute of Biomedical Research of Barcelona (IIBB)-Spanish National Research Council (CSIC), Biomedical Research Institute Sant Pau (IIB Sant Pau), Sant Quintí 77-79, 08041 Barcelona, Spain. E-mail addresses: gustavo.chiabrando@unc.edu.ar (G. Chiabrando), vicenta.llorente@iibb.csic.es, cllorente@santpau.cat (V. Llorente-Cortés). 1 These authors equally contributed to this work 2 ORCID: 0000-0001-8902-6693 3 ORCID: 0000-0001-8591-7632 https://doi.org/10.1016/j.biopha.2022.113270 Received 4 May 2022; Received in revised form 2 June 2022; Accepted 6 June 2022 Available online 13 June 2022 0753-3322/© 2022 The Authors. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Interpretation: LRP1-immunotargeting can block CE accumulation within the heart with specificity, selectivity, and efficacy, thereby improving the cardiac insulin response; this has important therapeutic implications for a wide range of cardiac diseases. Funding: Fundació MARATÓ TV3: grant 101521-10, Instiuto de Salud Carlos III (ISCIII) and ERDF PI18/01584, Fundación BBVA Ayudas a Equipos de Investigación 2019. SECyT-UNC grants PROYECTOS CONSOLIDAR 2018- 2021; FONCyT, Préstamo BID PICT grant 2015-0807 and grant 2017-4497. 1. Introduction pro-inflammatory signaling in rabbits, as well as in ex vivo human macrophages or human coronary vascular smooth muscle cells exposed Cardiovascular disease (CVD) is the leading cause of death world­ to rabbit hypercholesterolemic sera [29]. The CR9 domain (which binds wide and is frequently associated with metabolic syndrome (MS) and anti-P3 Abs) show scarce affinity for serpins, such as plasminogen type 2 diabetes mellitus (T2DM)[1–4]. Individuals with MS or T2DM activator inhibitor-1 and protease nexin 1, which induce a protective have a 2.5-times higher risk of developing heart failure than healthy signal in the heart [30]. This makes CR9 an ideal target for blocking individuals [5], and frequently present myocardium steatosis, which is LRP1 interactions with pathological ligands (atherogenic lipoproteins) associated with alterations in insulin signaling [6–8] and precedes dia­ but without altering interactions with protective ligands (protea­ stolic dysfunction and loss of metabolic flexibility [9–11]. Further, in­ se-inhibitor complexes) [31]. On the basis of these previous results, the sulin resistance is associated with pathological remodeling of the heart, main objectives of the present study were i) to evaluate the impact of which is characterized by concentric hypertrophy of the left ventricle, CE-enriched lipoproteins on cardiac lipid accumulation, and ii) to study interstitial fibrosis, and extracellular matrix (ECM) remodeling that the consequences of CE accumulation on cardiac insulin response. For leads to diastolic dysfunction [12–14]. Recently, left ventricle myocar­ this, we specifically blocked the interaction of CE-enriched lipoproteins dial remodeling has been linked to high levels of cholesteryl ester with LRP1 receptor via immunization with the peptide P3. (CE)-enriched lipoproteins, such as low-density lipoprotein (LDL), using paired lipid and cardiovascular magnetic resonance data, in a 2. Methods one-sample Mendelian randomization study that included 17 311 in­ dividuals [15]. However, the molecular mechanisms underlying the link 2.1. Animal model between lipoprotein-derived cardiac lipid accumulation, insulin action, and cardiac dysfunction are largely unknown. Experimental procedures were approved by the Ethics Committee of The susceptibility of LDL particles to aggregate is associated with Animal Experimentation of the Vall d′ Hebron Institute of Research with increased cardiovascular mortality [16]. The small, dense LDL particles registration number 46/17, and performed in accordance with Spanish (prevalent in MS and T2DM patients) have a higher tendency to be legislation and also with the European Union directives (2010/63/ EU). retained and aggregated (agLDL) in the extracellular spaces of several Female rabbits (thirty New Zealand White [NZW]) were fed i) a chow R- tissues, including the arterial intima [17–19]. AgLDL, in contrast to 01 diet from Granja San Bernardo with the following formulation: 17.3 native LDL, is taken up by the LDL receptor-related protein 1 (LRP1), a % protein, 16.7 % fiber, and 3 % fat, or ii) a high-fat diet (HFD) TD.88 receptor that is upregulated by hypercholesterolemic conditions in cells 137 (42 % fat, 0.2 % cholesterol) from ENVIGO, for 1 month (Fig. S1a). of the cardiovascular system (including smooth muscle cells, macro­ Animals were acclimated for one week before the first immunization and phages, and cardiomyocytes) [20–22]. The interaction of LRP1 with immunized with a primary injection and four reminder doses (R1-R4) agLDL facilitates the selective uptake and transfer of CEs from agLDL to every 21 days of irrelevant peptide (IrP group; n = 10) or P3 peptide (P3 intracellular lipid droplets of vascular and cardiac cells [21,22]. Of note, group; n = 10) conjugated to the carrier. The four doses of IrP or P3 LRP1-mediated agLDL uptake decreases insulin-induced intracellular antigen conjugated with keyhole limpet hemocyanin (KLH) were signaling, glucose transporter type 4 (GLUT4) translocation to the administered subcutaneously (138 μg/kg, maximum volume 150 µl). For plasma membrane, and glucose uptake by HL-1 cells [6]. In fat and the first immunization, IrP or P3-KLH peptides were emulsified in muscle cells, LRP1 regulates the intracellular trafficking of complete Freund’s adjuvant; the rest of the immunizations were done insulin-responsive GLUT4 storage vesicles (GSV) [23]. Under insulin using IrP or P3-KLH conjugated in incomplete Freund’s adjuvant (both stimulation, GSVs are trafficked and fused with the plasma membrane from Sigma Aldrich). During the immunization period, the animals were by a mechanism that is dependent on the activation of the fed a normal chow diet. Starting at the R4 time point, IrP- and P3- phosphatidylinositol-3-kinase (PI3K)/protein kinase B (Akt)/AS160 immunized rabbits were randomly divided into a chow-fed or a HFD- (Akt substrate of 160 kDa) pathway [24]. LRP1 interacts with insulin fed group. Ten rabbits (n = 5 IrP-injected, n = 5 P3-injected) receptor (InsR) to regulate InsR-associated intracellular signaling in continued the chow diet, whereas ten rabbits (n = 5 IrP-injected, n = neurons, hepatocytes, and cardiomyocytes [6,25,26]. In addition, in 5 P3-injected) received HFD for 30 days. Four groups of rabbits were retinal Müller glial cells, blocking LRP1 exocytosis towards the plasma included in this study: i) chow IrP-immunized, ii) chow P3-immunized, membrane disturbs intracellular signaling induced by insulin[27]. These iii) HFD IrP-immunized, and iv) HFD P3-immunized (Fig. S1a). Animals data suggest that LRP1 is critical for the link between lipid accumulation were weighed at pre- and post-diet time points, and serum levels of and insulin signaling in different cell types. specific anti-P3 antibodies were determined by ELISA. At the end of the The LRP1 P3 peptide sequence of H-GDNDSEDNSDEENC-NH2 study, animals were euthanized, and the hearts and livers were removed (Gly1127–Cys1140) is located on the CR9 domain of the extracellular aseptically and processed for molecular, confocal, electron microscopy, alpha chain, cluster II and is crucial for the interaction of this receptor biophysical, and immunohistochemical studies. with CE-enriched lipoproteins [28]. We previously showed that poly­ clonal antibodies against the P3 sequence (anti-P3 Abs) efficiently 2.2. Peptide synthesis and conjugation reduced foam cell formation and atherosclerosis in in vitro, ex vivo, and in vivo models [28,29]. In particular, purified anti-P3 Abs have a higher The P3 peptide used to immunize rabbits contained the sequence affinity for the P3 epitope than agLDL, blocking the binding of this GDNDSEDNSDEENC, which corresponds to the amino acids 1127–1140 atherogenic LDL to LRP1 and its subsequent uptake by vascular cells located in LRP1 cluster II (domain CR9) [28,29]. The P3 sequence cor­ [28]. In in vivo studies, immunization against LRP1 prevents intracel­ responds to an area of high homology between human and rabbit LRP1, lular CE accumulation in the vasculature and harmful coupling of with the difference that the asparagine (N) in humans was replaced by a 2 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 serine (S) in the rabbit protein. In addition, the amino acid C1140 in the were incubated with serum (0.25 %, 1.5 h) from the different rabbit rabbit sequence (GDNDCEDNSDEENC) was replaced by S to achieve groups i) chow IrP-immunized, ii) chow P3-immunized, iii) HFD greater peptide immunogenic effectiveness. The irrelevant peptide (IrP) IrP-immunized, and iv) HFD P3-immunized-rabbits. has the same sequence than P3 but with amino acids in D-enantiomer configuration. Both peptides were synthesized by the Laboratory of Proteomics & Protein Chemistry, Department of Experimental & Health 2.7. Determination of neutral lipid content in heart and liver Sciences, Pompeu Fabra University, by the solid-phase method using a Prelude peptide synthesizer (Protein Technologies, Inc.). Peptides were Myocardial, hepatic, and cellular lipids were extracted and parti­ purified by high-performance liquid chromatography (HPLC, Waters tioned by thin-layer chromatography. Lipids were extracted with 600) using UV detection at 254 nanometers (Waters 2487) and char­ dichloromethane/methanol [1:2] and cholesteryl esters (CE), free acterized by mass spectrometry (Applied Biosystems 4700 Proteomics cholesterol (FC), and triglycerides (TG) were analyzed by thin-layer Analyser). Peptide conjugation to KLH or BSA (Sigma, St. Louis, MO) chromatography (TLC) on silica G-24 plates as previously described was performed as previously described [32]. Peptide–KLH conjugates [35,36]. Different concentrations of standards (a mixture of cholesterol, were used for rabbit immunization, and peptide-BSA conjugates were cholesterol palmitate and triglycerides) were applied to each plate. The used as a substrate in the immunoassay ELISA to detect specific anti-P3 spots corresponding to CE, TG, and FC were quantified by densitometry Abs in the rabbit serum. against the standard curve of cholesterol palmitate, triglycerides and cholesterol, respectively, using a computing densitometer. 2.3. Detection of specific antibodies 2.8. Assessment of lipid droplet morphology and size ELISAs were standardized to detect specific antibodies against P3 peptides. Briefly, all sera were analyzed using 96-well polystyrene plates 2.8.1. Electron microscopy (442 404 Maxisorp, NUNC, Labclinics, Spain) coated with peptide-BSA Cells growing in 60 mm plates were washed in PBS 1X and then fixed or BSA as a control to detect unspecific antibodies. ELISA plates were with 2.5 % glutaraldehyde in 0.1 M phosphate buffer at room temper­ incubated with several serum dilutions for 90 min; after washes, anti- ature for 1 h. Next, cells were gently scraped and then pelleted in 1.5 ml rabbit IgGs conjugated to peroxidase (170–6515, BioRad, Spain) were tubes. Cardiac tissue was cut into small pieces, extensively washed in added to detect the antigen–antibody complexes. ELISA was revealed PBS 1X and fixed with 2.5 % glutaraldehyde, 2 % PFA in 0.1 M phos­ using OPD substrate (P9187, Sigma Aldrich, Spain), and the absorbance phate buffer overnight. Samples were dehydrated, embedded in Spurr was read in a Multilabel reader Victor3 (Perkin Elmer, Turku, Finland) and sectioned using Leica ultramicrotome (Leica Microsystems). Ultra­ at 450 nm. The measured absorbances were adjusted to a 4PL curve to thin sections (50–70 nm) were stained with 2 % uranyl acetate for 10 calculate the IC50. The parameter 1/IC50 was used as the antibody titer. min, a lead-staining solution for 5 min and then analyzed with a trans­ mission electron microscope, JEOL JEM-1010 fitted with a Gatan Orius 2.4. Biochemical analysis of serum and isolated lipoproteins SC1000 (model 832) digital camera at the Unit of Electron Microscopy, Scientific and Technological Centers of the University of Barcelona, Serum lipids and lipoproteins, including cholesterol and triglyceride School of Medicine and Health Sciences (Barcelona, Spain). (TG) (corrected from free glycerol), were enzymatically determined using commercial kits adapted to a COBAS 6000 autoanalyzer (Roche 2.8.2. Confocal microscopy Diagnostics) [33]. VLDL, LDL, and HDL lipoproteins were isolated by A stock solution of the fluorescent dye boron-dipyrromethene sequential ultracentrifugation at 100,000g for 24 h at a density of 1.006, (BODIPY) was diluted to a final concentration of 1 mg/ml in DMSO. 1.019–1.063 and 1.063–1.21 g/ml, respectively, using an analytical To study lipid droplets, heart slides were incubated for 30 min with 100 fixed angle rotor (50.3, Beckman Coulter). The composition of each li­ µl of BODIPY (1 mg/ml in DMSO). Images of immunostained cells were poprotein, including total and free cholesterol, triglycerides and phos­ analyzed on a Leica inverted fluorescence confocal microscope (Leica pholipids, was determined by commercial methods adapted to the TCS SP2-AOBS; excitation wavelength 480 nm, emission maximum 515 COBAS 501 autoanalyzer. Lipoprotein protein concentrations were nm). determined by the bicinchoninic acid method (Termo Scientific, Rock­ ford, IL). Lipid and protein concentrations were used to calculate the total mass of each lipoprotein. 2.9. FTIR, freeze-dried state 2.5. Glucose tolerance test (GTT) Fourier transform infrared spectroscopy/attenuated total reflectance (FTIR/ATR) spectra of the freeze-dried lipoproteins and tissues were The glucose tolerance test (GTT) was performed at week four of the acquired using a Nicolet 5700 FTIR instrument (Thermo Fisher Scien­ dietetic intervention under fasting conditions. Basal blood glucose levels tific, Waltham, MA) equipped with an ATR device with a KBr beam were measured from an ear nick through ACCU-CHEK® Aviva gluc­ splitter and a MCT/B detector as previously described [37,38]. The ATR ometer (Roche Molecular Systems). Rabbits were then intraperitoneally accessory used was a Smart Orbit with a type IIA diamond crystal injected with glucose (1.3 mg/g BW). Blood glucose was measured at 15 (refractive index 2.4). Freeze-dried samples (1 mg) were directly min, 30 min, 60 min, 120 min and 180 min after glucose injection. The deposited on the entire active surface of the crystal and gently pressed area under the curve (AUC) of the response curve was then calculated with a Teflon tip to assure good contact. For each sample, 80 in­ using the software Prism 4.0 [34]. terferograms were recorded in the 4000–450/cm-1 region, co-added and Fourier transformed to generate an average spectrum of the segmented 2.6. Cell culture heart part with a nominal resolution of 1 cm-1 using Omnic 8.0 (Thermo Fisher Scientific, Waltham, MA). A single-beam background spectrum The murine HL-1 cardiomyocyte-derived cell line was generated by was collected from the clean diamond crystal before each experiment, Dr. W.C. Claycomb (Louisiana State University Medical Centre, New and this background was subtracted from the spectra. Spectra were then Orleans, Louisiana, USA). HL-1 cells were maintained in Claycomb subjected to ATR and baseline corrections and normalized using the medium (Sigma-Aldrich) in plastic dishes [6], coated with 12.5 g/ml maximum of the Amide II peak. Second derivatives were used to fibronectin (Sigma-Aldrich) and 0.02 % gelatin, in a 5% CO2 atmo­ enhance the chemical information present in overlapping infrared ab­ sphere at 37 ◦ C. HL-1 cells were fasted with 0.2 % FBS for 24 h and then sorption bands of spectra. 3 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 2.10. Determination of InsR and LRP1 levels in membranes and total Abcam), rabbit anti-pAS160 (Thr642, ab65753, Abcam), and mouse cardiac lysates monoclonal anti-β-actin (A2228; Sigma-Aldrich). Signals were detected with the ECL immunoblotting detection system (GE Healthcare), and the 2.10.1. Tissue subfractionation results were quantitatively analyzed using Chemidoc (BioRad). Tissue was washed with PBS 1X and homogenized using a POLY­ TRON® Immersion Dispersers PT 2500 as previously described [35,38]. 2.12. Confocal microscopy In brief, to obtain membrane protein fraction, the homogenized tissue was filtered using QIAshredder spin columns (79 656, QIAGEN), incu­ The procedures were followed as previously described [6]. HL-1 bated with lysis buffer A (NaCl 150 mM, HEPES 50 mM, digitonin 25 cardiomyocyte were washed with PBS 1 × , fixed with 4 % para­ µg/ml, 1 M hexylene glycol, protease inhibitor cocktail 1 % v-v) using an formaldehyde (PFA), permeabilized for 30 min with 0.5 % (v/v) saponin end-over-end rotator. After centrifugation, the supernatant containing and blocked with 2 % bovine serum albumin (BSA). Primary antibodies mainly cytosolic proteins was removed and the the pellet was incubated were used for 1 h, using rabbit anti-LRP1 monoclonal antibody with lysis buffer B (NaCl 150 mM, HEPES 50 mM, igepal 1 % v-v, 1 M (ab92544, Abcam) or mouse anti-InsR antibody (ab69508; Abcam), and hexylene glycol, protease inhibitor cocktail 1 % v-v). After centrifuga­ secondary antibodies conjugated with Alexa Fluor (1/800) and Hoechst tion, the supernatant containing mainly plasma membrane proteins was 33 258 (1/2000) were used for 1 h. Finally, cells were mounted with collected. Proteins were analyzed by Western blot with antibodies Mowiol 4–88 (Calbiochem (Merck, Darmstadt). Images were obtained against rabbit LRP1 (ab92544 Abcam) or InsR (LS-C63091, LSBio) or with an Olympus FluoView FV1200 confocal microscope (Olympus), GLUT4 (LS-C123618, LSBio). Total cardiac lysates were obtained using processed using the FV10-ASW Viewer 3.1 (Olympus) and quantified by RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1 % Triton X-100, ImageJ software. Quantification of the colocalization level was analyzed 0.5 % sodium deoxycholate, 0.1 % SDS, 1 mM PMSF, 10 mM sodium by JACoP plug-in from ImageJ software [6]. ortho-vanadate, and protease inhibitor cocktails (Sigma-Aldrich, St. Heart slides were washed with PBS 1 × , permeabilized for 5 min Louis, MO, USA), and proteins were analyzed by Western blot. with PBS 1 × with 1 % Tween and blocked with PBS 1 × with 1 % BSA. Anti-ATP1A1 (MA3–928, Invitrogen) and anti-GAPDH (sc-81 545, Santa Primary antibodies were used for 1 h (mouse anti-GLUT4 monoclonal Cruz Biotechnology) were used as total and plasma membrane loading antibody LS-C123618. LS-Bio) (1/50) and secondary antibody conju­ controls, respectively. gated with alexa fluor (1/100) and Hoechst 33 258 (1/1000) were used for 1 h. Slides were then mounted with ProLong™ Gold Antifade 2.10.2. Biotin-labeling of cell surface proteins Mountant (P10144, Thermoscientific). Images were obtained with an HL-1 cells were exposed to rabbit’s serum (0.25 %) for 1.5 h and then Olympus FluoView FV1200 confocal microscope (Olympus), processed treated with insulin (100 nM) for 30 min. Biotin-labeling protein assay using the FV10-ASW Viewer 3.1 (Olympus) and quantified by ImageJ (EZ-Link Sulfo-NHS-SS-Biotin (21 331,Thermo Scientific) was used to software. determine the protein levels of GLUT4 at the cell surface. Briefly, cells were incubated first with a biotin solution 0.12 mg/ml for 2 h at 4 ◦ C 2.13. Immunoprecipitation (IP) assay and, then with 0.1 mM glycine solution for 30 min at 4 ◦ C and then washed three times with PBS 1X. Biotinylated proteins were pulled Cardiac tissue from the rabbit groups and HL-1 cells exposed to down by streptavidin-conjugated agarose beads (Pierce Streptavidin rabbit’s serum from each of the four groups were subjected to immu­ Agarose (20 353, Thermo Scientific) for 2 h at room temperature. Bio­ noprecipitation (IP) assays as previously described [6]. Briefly, protein tinylated and total proteins (10 % of proteins incubated with agarose extracts were incubated with rabbit anti-LRP1 monoclonal antibody beads) were analyzed by Western blot after incubated with rabbit anti- (ab92544, Abcam) or rabbit non-immune IgG as control (2 μg/200 μg of GLUT4 (sc-7938; Santa Cruz Biotechnology), mouse monoclonal anti- total proteins) for 2 h at 4 ◦ C. Samples were then incubated with protein ATP1A1 (MA3–928, Invitrogen), or mouse monoclonal anti-β-actin A–conjugated agarose agarose beads following the manufacturer’s pro­ (A2228; Sigma-Aldrich) antibodies overnight at 4 ◦ C, following by in­ cedure (sc-2001; Santa Cruz Biotechnology) overnight at 4 ◦ C. Immu­ cubation with secondary antibodies (goat anti-mouse and goat anti- noprecipitated proteins were characterized by Western blot using rabbit; Dako) diluted 1/10 000 for 1 h at room temperature. Signals anti-InsR (ab69508; Abcam) and anti-LRP1 (ab92544; Abcam) mono­ were detected with the ECL immunoblotting detection system (GE clonal antibodies, and secondary antibodies (goat anti-mouse or goat Healthcare) and results were quantitatively analyzed using Chemidoc anti-rabbit; Dako). Signals were detected using ECL immunoblotting (BioRad). Biotinylated-ATP1A1 and β-actin were used as loading control detection system (GE Healthcare) and the results were quantitatively of plasma membrane protein and total protein extracts, respectively. analyzed using Chemidoc (BioRad). 2.11. Western blot analysis 2.14. 2-NBDG uptake assay Protein extracts were obtained using RIPA buffer (50 mM Tris-HCl HL-1 cells were exposed to rabbit’s serum (0.25 %) for 1.5 h and then pH 8.0, 150 mM NaCl, 1 % Triton X-100, 0.5 % sodium deoxycholate, treated with insulin (100 nM) in the presence of 2-deoxy-2-[(7-nitro- 0.1 % SDS, 1 mM PMSF, 10 mM sodium ortho-vanadate, and protease 2,1,3-benzoxadiazol-4-yl) amino]-D-glucose (2-NBDG solution; Sigma- inhibitor cocktails) (Sigma-Aldrich, St. Louis, MO, USA). Cell protein Aldrich) (80 µM) for 30 min. Cells were then washed with 1X PBS, extracts (40 µg) were diluted in 5 × sample buffer with dithiothreitol fixed with 4 % paraformaldehyde, blocked with 2 % BSA, and incubated (DTT) and heated for 5 min at 95 ◦ C. Electrophoresis on 10 % SDS- with Hoechst 33 258 colorant (1/2000) for 1 h [6]. Fluorescence was polyacrylamide gels was performed, and proteins were transferred to detected with a Leica DMI8 biological microscope (Leica, Germany). nitrocellulose membrane (GE Healthcare Life Science, Amsterdam, The Total fluorescence in the whole cell area was quantified by ImageJ Netherlands). Nonspecific binding was blocked with 5 % non-fat dry software. milk in a Tris-HCl buffer containing 0.01 % Tween 20 (TBS-T) for 60 min at room temperature. Membranes were incubated overnight at 4 ◦ C with 2.15. Statistical analysis primary antibodies and then with secondary antibodies (goat anti- mouse and goat anti-rabbit; Dako) diluted 1/10 000 for 1 h at room Differences between study groups were analyzed using two-way temperature. For signaling activation analysis, the following primary analysis of variance (ANOVA) followed by a Tukey post-hoc test. The antibodies were used: rabbit anti-Akt (9272, Cell Signaling Technology), statistical software R (www.r-project.org) was used for all statistical rabbit anti-pAkt (Ser473, 07–789, Merck), rabbit anti-AS160 (ab24469, analyses. Differences were considered statistically significant when p < 4 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 0.05. 3.2. Lipid profile and peripheral insulin resistance induced by HFD are not modified by anti-P3 Abs 3. Results Cholesterol levels (Fig. 1a) and triglyceride (TG) levels (Fig. 1b) were 3.1. Immunization with the P3 peptide produces anti-P3 Abs in rabbits higher in the serum of HFD-fed rabbits than in serum of control diet rabbits, with no differences between P3- or IrP-immunized rabbits. Liver The experimental rabbit immunization procedure included the pri­ lipid extraction followed by TLC showed a significant upregulatory ef­ mary injection and four booster injections of either the P3 peptide or an fect of HFD on levels of hepatic CE, TG, and free cholesterol (FC), also irrelevant peptide (IrP), and two blood extractions at the checking with no differences between P3- or IrP-immunized rabbits (Fig. 1c-f). points. Rabbits were either fed a high-fat diet (HFD) or a control diet (i. Rabbits in all four groups (e.g., HFD P3, HFD IrP, chow P3, and chow e., normal chow) for 30 days (Fig. S1a). ELISA analysis showed a sig­ IrP) exhibited a similar increase in body weight over the study period nificant anti-P3 Abs production that was maintained throughout the (30 days) (Fig. S2a). However, the glucose tolerance test (GTT) showed entire dietetic intervention period in the serum of P3-immunized rab­ that HFD induced glucose intolerance in both IrP and P3 immunized bits, based on the levels of antibodies at the two checkpoints, Chk1 and groups (Fig. S2b, c), indicating that the impairment of peripheral insulin Chk2 (Fig. S1b, middle and left panel). Rabbits immunized with the IrP response induced by HFD was not affected by P3 immunization. peptide did not show any response against P3 in ELISA (Fig. S1c). Fig. 1. Immunization with P3 peptides does not alter serum or liver lipid profiles in rabbits. (a, b) Bar graphs showing serum cholesterol and triglyceride (TG) levels. (c) Representative thin-layer chromatography (TLC) of cholesterol ester (CE), TG, and free cholesterol (FC) in the liver. The CE/TG/FC bands from the standards (Std) are shown. (d, e, f) Bar graphs showing results as mean ± SD of CE, TG, or FC, quantified as μg/mg tissue protein. Statistical significance was determined by 2-way ANOVA with Tukey’s post-hoc test. n = 5/group. 5 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 3.3. Cholesteryl esters in HFD-fed rabbits are mainly carried by VLDL and 3.4. The spectral FTIR lipid signature of CE-enriched lipoproteins (VLDL LDL lipoproteins and LDL) is evident in rabbit heart The HFD-fed groups had much higher total mass of circulating LDL FTIR spectroscopy is based upon the interaction between the IR ra­ (IrP: 32.10 ± 3.53 vs 1.23 ± 0.68 g/L, p < 0.001) and, to a lesser extent, diation and the covalent bonds of molecules. IR spectroscopy exploits of VLDL (IrP: 5.90 ± 0.85 vs 0.35 ± 0.13 g/L, p < 0.001), than either of the fact that molecules have specific frequencies at which they vibrate the chow-fed groups (Fig. S3), with most of HFD-increased plasma corresponding to discrete energy levels (vibrational modes). C-H cholesterol carried by LDL and VLDL. Both VLDL and LDL had much stretching vibrations arise essentially from lipids and lead to absorption higher CE content in HFD rabbits (Fig. S3c,d) than in control rabbits bands in the 3050–2800 cm− 1. In this infrared zone [3050–2800 cm− 1] (Fig. S3a,b), with no differences in the lipid/protein profile of VLDL and of the FTIR spectra, human and rabbit lipoproteins showed the specific LDL lipoproteins between IrP- and P3-injected rabbits, irrespective of absorption of asymmetric and symmetric (CH3) and (CH2) stretching, diet. Despite the low total mass of circulating VLDL and LDL in serum mainly associated with the vibrational answer of the different classes of from control groups, these lipoproteins had higher proportion of TG – H) stretching at 3010 cm− 1 that corresponds to lipids, as well as the (C– (Fig. S3a,b) than the HFD groups (Fig. S3c,d), irrespective of IrP- or P3- unsaturated lipids (Fig. S4a). The band associated with unsaturated immunization. HDL was the minor lipoprotein class, with a similar lipids as well as the specific bands of CE in the [2900–2880 cm− 1] zone composition in the four groups. Overall, the lipoprotein fraction content were better detected on the second derivative spectra (Fig. S4b) that of VLDL, LDL, and HDL lipoprotein masses, as well as of TG, phospho­ enhances resolution for both lipoproteins and rabbit hearts. lipid (Ph), FC, CE, and protein levels, were similar between IrP- and P3- In the [3050–2800 cm− 1] zone, the spectral signatures of lipids from injected groups irrespective of diet. These results clearly indicate that P3 LDL and VLDL were identified in the FTIR profile of hearts from both immunization did not significantly alter the lipid/protein profile of chow and HFD rabbit groups (Fig. 4a, Fig. S4a,b). These absorption VLDL, LDL, or HDL in rabbits. bands can be detected at the same position both in lipoproteins and heart tissues. This fact indicates that these lipoproteins are one of the main sources of neutral lipids to the heart. Fig. 2. Anti-P3 Abs specifically block the selective uptake of CE from lipoproteins in the heart. (a) Confocal microscopy images showing BODIPY-stained lipid droplets (red dots) in the hearts from the distinct rabbit groups. Cell nuclei are stained with DAPI (blue). (b) Representative TLC of heart lipid extracts and bands corresponding to the cardiac content of cholesteryl esters (CE), triglyceride (TG), and free cholesterol (FC). The CE/TG/FC bands of standards (Std) are shown in duplicate. Bar graph represents the mean ± SD of CE (c), TG (d), or FC (e), expressed as μg/mg tissue protein. Statistical significance was determined by 2-way ANOVA with Tukey’s post-hoc test. n = 5/group. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.) 6 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 3.5. Lipoprotein-derived CE accumulation in heart and HL-1 (Fig. S7a, S7b). Lipid droplets were not detected in HL-1 cells exposed to cardiomyocytes is reduced by anti-P3 Abs rabbit serum from the other groups. Confocal microscopy analysis of BODIPY-stained transversal cross 3.7. Anti-P3 Abs increase membrane InsR levels and restore InsR/LRP1 sections from the heart showed lower levels of lipid droplets in the HFD interaction and insulin signaling impaired by HFD in cardiomyocytes P3 group as compared to HFD IrP group (Fig. 2a). To further explore the impact of P3 immunization of cardiac lipid content, we performed thin- We then evaluated InsR and LRP1 protein levels in the cardiac layer chromatography (TLC) of cardiac lipid extracts. TLC showed that membranes and total extracts from the heart of the four different groups. HFD strongly increased intramyocardial CE and TG content as compared Plasma membrane was disrupted by digitonin fractionation [36], and to chow diet (Fig. 2b–d). P3 immunization showed efficacy in reducing ATP1A1 (ATPaseNa/K) used as a specific marker for plasma membrane. exclusively intramyocardial CE (Fig. 2b,c) but not TG content (Fig. 2b, As shown in Fig. 4a,b, membrane to total InsR ratio was significantly d). Cardiac FC was similar for all four groups (Fig. 2b, e). Additionally, in lower in the heart from HFD rabbits compared to than from chow rab­ ex vivo experiments, we explored the effect of increased percentages of bits, however, in the presence of anti-P3 Abs, membrane to total InsR serum (0.25–1 %; 1.5 h) from all four groups in the intracellular lipid ratio was restored and even exceeded those observed in chow rabbits. As content of the HL-1 cardiomyocytes. At 1.5 h, HFD serum induced shown in Fig. 4c,d, membrane to total LRP1 ratio was much higher in intracellular CE/FC ratio in a dose-dependent manner but no changes in HFD than in control chow groups without differences between IrP and the TG/FC ratio. The anti-P3 Abs present in HFD P3 rabbit serum effi­ P3 groups. In addition, immunoprecipitation assays showed that ciently blocked the intracellular CE/FC ratio induced by the hypercho­ InsR/LRP1 interaction complexes were reduced in the heart of HFD lesterolemic serum in HL-1 cells at each tested dose (Fig. S5a,b). compared to chow groups; and that anti-P3 Abs not only restored but even surpassed the basal levels of InsR/LRP1 complexes observed in the heart of chow animals (Fig. 4e,f). Along the same line, ex-vivo experi­ 3.6. P3 immunization changes the ultrastructure and bioenergetics of lipid ments showed that treatment of HL-1 cardiomyocytes with HFD IrP droplets in the rabbit heart serum reduced InsR/LRP1 complex levels in the absence or presence of insulin while HFD P3 serum restored and even increased the levels of Transmission electron microscopy (TEM) of ultrathin sections from InsR/LRP1 complex in these cells (Fig. S8a,b). heart of HFD-fed rabbits revealed the presence of neutral lipids forming Confocal microscopy experiments showed that serum from HFD IrP lipid-filled vacuoles in cardiomyocytes of hearts from HFD groups animals, but not that from HFD P3 animals, reduced InsR and LRP1 (Fig. 3a,b and S6c,d). These lipid droplets were located between the colocalization in HL-1 cardiomyocytes (Fig. 5a,b). In addition, HFD IrP sarcomere fibers and surrounded by a membrane that seems to be in serum significantly reduced insulin-induced Akt and AS160 phosphor­ close contact with mitochondria, as well as by the sarcoplasmic reticu­ ylation, while HFD P3 serum restored insulin-induced p-Akt and p- lum membrane. An important finding from electron microscopy studies AS160 levels to those found in HL-1 cells exposed to chow serum was that lipid droplets in the heart of P3-immunized rabbits (Fig. 3b and (Fig. 5c-e). Taken together, these results indicate that anti-P3 Abs S6d) were scarcer, more electrodense and bigger than those found in the reduced the deleterious effect of HFD on the InsR/LRP1 association, InsR heart of IrP injected rabbits (Fig. 3a and S6c). Another important finding membrane translocation, and InsR signaling in cardiomyocytes. was that the inter-sarcomere cytoplasm contained abundant glycogen in proximity with lipid droplets in the heart of HFD IrP (Fig. 3a and S6c) but not in that from HFD P3 rabbits (Fig. 3b and S6d). Similar or even higher abundance of glycogen drops were observed in the surroundings of lipid droplets formed in HL-1 cells exposed to rabbit HFD IrP serum Fig. 3. Anti-P3 Abs alters lipid droplet size and electrodensity as well as the glycogen content induced by HDF. Representative electron microscopy of cardiomyocyte ultrastructure from the heart of HFD-fed rabbits showing differences between the size and electrodensity of lipid droplets (LD) (asterisks) in HFD IrP (a) and HFD P3 (b) rabbit groups. LD are located in the intersarcomeric cytoplasm between the sarcomeric fibers (SF) and in close connection with mitochondria (mit) and sarcoplasmic reticulum (SR). A high number of glycogen (electrodense granules, red arrows) surrounding the LD (asterisks) can also be observed in the heart of HFD IrP group. Representative Z-disc (Z) are indicated. Scale bar: 400 nm. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.) 7 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Fig. 4. Anti-P3 Abs prevent the deleterious effects of HFD on cardiac LRP1 and InsR membrane localization and interaction. Representative Western blot analysis of InsR (a) and LRP1 (c) in cardiac membranes and total cardiac lysates. Bar graph showing the InsR (b) and LRP1 (d) ratio between membrane and total levels normalized by ATP1A1 (membranes) or GAPDH (total extracts). n = 5/group. (e) Immunoprecipitation assay for LRP1. Myocardial lysates were immunoprecipitated with anti-LRP1 antibodies and analyzed by Western blot assays for InsR and total LRP1. NC, non-immune control without anti-LRP1 Ab. (f) Bar graph showing the densitometry quantification of Western blot bands. Values are expressed as mean ± SD of the relative intensity of LRP1 and InsR bands with respect to the endogenous control and are shown as a fold-change against the mean value of control IrP-immunized group. n = 3/group. Statistical significance was determined by 2-way ANOVA with Tukey’s post-hoc test. 3.8. Anti-P3 Abs restore GLUT4 membrane levels in HFD hearts and the plasma membrane and insulin-induced glucose uptake in these cells. All insulin response in HL-1 cardiomyocytes exposed to HFD serum these outputs were significantly reduced in HL-1 cells exposed to hy­ percholesterolemic serums containing anti-P3 Abs. These novel findings Western blot analysis (using ATP1A1 as a specific marker for plasma show that cardiomyocyte intracellular CE accumulation plays a crucial membrane) and confocal microscopy revealed that membrane to total role in impaired cardiac insulin response induced by diet, and that these GLUT4 ratio was much lower in the heart of HFD rabbits than in that of alterations are efficiently reduced by exposure to antibodies against the control chow rabbits, however, in the presence of anti-P3 Abs, GLUT4 P3 sequence of LRP1 CR9 domain. membrane levels were restored to those observed in control chow rab­ Intramyocardial CE accumulation in the heart is promoted by bits (Fig. 6a-c). In addition, we evaluated whether anti-P3 Abs could ischemia [35,37,39], hypercholesterolemia [40], and obesity [38]. restore GLUT4 trafficking to plasma membrane and glucose uptake in However, the clinical impact of intramyocardial CE accumulation HL-1 cardiomyocytes exposed to HFD serum. Biotin-labeling cell surface (independently of TG accumulation) has not yet been investigated, even protein assays showed that insulin failed to induce GLUT4 translocation though experimental studies have consistently shown that the causes to the plasma membrane in HL-1 cells exposed to HFD IrP but not in cells and consequences of intracellular CE and TG accumulation involve exposed to HFD P3 and chow serums (Fig. 6a,b). In line, insulin failed to differential mechanisms and pathological consequences. TG accumula­ induce 2-deoxyglucose (2-NBDG) uptake in HL-1 cells exposed to HFD tion is mainly caused by an imbalance between the rates of fatty acid IrP serum but not in cells exposed to HFD P3 and chow serums (Fig. 6c). (FA) entry and FA oxidation, and it usually occurs in parallel to accu­ These results suggest that anti-P3 Abs counteract the deleterious effect mulation of detrimental lipid intermediates, such as long-chain acyl-­ of HFD on GLUT4 trafficking to plasma membrane and glucose uptake in CoA, ceramides, and diacylglycerol, all of which impair insulin signaling cardiomyocytes Fig. 7. and glucose use in the heart [41]. We and others have proposed that intramyocardial CE accumulation is caused by the uptake of 4. Discussion CE-enriched lipoproteins through lipoprotein receptors [22,35,39]. Here, molecular studies revealed that intramyocardial CE accumulation Here we show that CE accumulation within the heart in response to is mainly caused by LRP1-mediated uptake of CE-enriched lipoproteins, HFD is driven by LRP1. Immunization of the rabbit model against the P3 since it is blocked in rabbit heart by anti-P3 Abs, similar to our previous sequence of LRP1 is able to reverse myocardial CE accumulation inde­ results in neonatal rat ventricular myocytes and cardiac-derived HL-1 pendently of changes in the circulating lipid profile. Here, we also show cells by RNA silencing approaches [22,35]. CE-enriched VLDLs and LDLs that intramyocardial CE accumulation is linked to a deficiency of InsR are large lipoproteins in rabbits that display lower affinity for the LDL and GLUT4 in cardiac membranes as well as to impaired LRP1/InsR receptor [42,43] but increased affinity for the LRP1 receptor [29]. interaction, both of which were recovered in hypercholesterolemic Our electron microscopy studies revealed that the lipid droplets in rabbits by inhibition of myocardial CE accumulation. In line, ex vivo cardiomyocytes of P3- injected rabbits, although lower in number, are experiments showed that rabbit hypercholesterolemic serum strongly bigger and more electrodense than those found in the heart of IrP- induces intracellular CE accumulation in HL-1 cardiomyocytes, injected rabbits. These imaging results combined with molecular and decreasing InsR signaling, insulin-induced GLUT4 translocation to biophysical studies suggest that lipid droplets in the heart of P3- 8 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Fig. 5. Anti-P3 Abs block the down-regulatory effects of intracellular CE on InsR signaling in HL-1 cells. (a) Confocal microscopy of HL-1 cells exposed to rabbit serum (0.25 %, 1.5 h) from the different groups, showing the co-localization between LRP1 (red) and InsR (green). The inset represents the 4× magnification of the framed regions in the dotted lines. The white arrowheads indicate co-localization vesicles. Images are representative of three independent experiments (n = 3). Scale bar, 10 µm. (b) Bar graph showing the mean ± SD of Manders´coefficient of the co-localization between InsR and LRP1, obtained from the quantification performed in 30 cells per condition. (c) Western blot assays of phosphorylated Akt (p-Akt) and AS160 (p-AS160), and total Akt (t-Akt) and AS160 (t-AS160) in HL-1 cells exposed to serum (0.25 %, 1.5 h) from the different groups of rabbits and then stimulated with insulin (100 nM, 10–20 min). β-actin was used as a protein loading control. Bar graphs showing the densitometry quantification of p-Akt (d) and p-AS160 (e) normalized to t-Akt and t-AS160 bands respectively, and represented as a fold-change as compared to the mean of control IrP group. *** p < 0.0001 versus indicated conditions. ** p < 0.001 versus indicated conditions. Results are shown as mean ± SD of three independent experiments (n = 3). Statistical significance was determined by one-way ANOVA followed by a Dunnett’s multiple comparison. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.) immunized rabbits contain almost exclusively TG, while those in the rabbit HFD IrP serum are surrounded by abundant glycogen drops, in heart of IrP rabbits are mixed (cholesteryl esters and triglycerides). In contrast to TG-enriched lipid droplets in the heart of HFD P3 rabbits line with our findings, lipid droplets containing exclusively CE have (which are not surrounded by glycogen). Our results suggest that P3- been observed to have a reduced size [44,45]. In addition, from the immunization counteracts the increased glycogen content induced by biophysical point of view, the liquid crystalline nature of CE mecha­ HFD in the heart of rabbits. Of note, previous studies have reported an nistically determines the intracellular phase behaviors of lipid droplets, increased glycogen content in the liver of fasted mice [48]. and thus the structure and organization of a TG/CE lipid droplet, the Our subfractionation studies combined with Western blot analysis lipid droplet nucleation, and the TG packaging [46,47]. These studies showed that the membrane/total InsR ratio was lower in hearts from suggest that presence of CE limits lipid droplet growth, which would HFD rabbits than in those from normal chow rabbits. Coherently, LRP1 explain the smaller size of CE/TG lipid droplets present in the heart of immunoprecipitation assays in rabbit heart tissue showed that HFD not the HFD IrP rabbits as compared to the TG-enriched lipid droplets pre­ only reduced cardiac membrane InsR levels but also disrupted InsR sent in the heart of HFD P3 rabbits. In addition, TLC results showed that interaction with LRP1, suggesting that InsR/LRP1 interactions mainly P3- immunization had a strong efficacy in exclusively reducing intra­ occurs at the cardiac membrane. These results in the in vivo model are in myocardial CE content. Together, these results demonstrate that P3- line with previous studies from our group showing a reduction in the immunization shows specificity for targeting CE accumulation in the InsR/LRP1 association in aggregated LDL-loaded HL-1 cells [6]. It re­ heart. mains to be explored whether this interaction is direct or mediated by Another key observation in this study is that mixed CE/TG lipid other intermediate proteins. A key result from the present study is that droplets formed in the heart of HFD rabbits and in HL-1 cells exposed to anti-P3 Abs, which blocked LRP1 interactions with CE-enriched 9 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Fig. 6. Anti-P3 Abs block the deleterious effects of HFD on cardiac GLUT4 membrane localization. (a) Representative Western blot analysis of GLUT4 levels in cardiac membrane and total rabbit heart extracts. ATP1A1 and GADPH were used as protein loading controls for plasma membrane and total cardiac protein extracts, respectively. (b) Bar graphs showing the GLUT4 ratio between membrane and total levels normalized by ATP1A1 (membranes) or GADPH (total extracts), respectively. n = 5/group. (c) Confocal microscopy of slides from the heart of the four groups. Images show the localization of GLUT4 (in red) in cardiomyocytes. An overlay of confocal and transmission images helps to visualize GLUT4 localization in the boundary of cardiomyocytes in the hearts from the chow IrP, chow P3, and HFD P3 groups but not from HFD IrP rabbit group. Asteriks indicate GLUT4 in the cytoplasm of cardiomyocytes. Nuclei are shown in blue. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.) lipoproteins and intramyocardial CE accumulation, facilitated the re­ recovered in HFD P3 rabbits. In line with our results, HFD-mediated covery of InsR levels and InsR/LRP1 complex in membranes in the heart membrane cholesterol loading seems to promote the loss of cortical of rabbits and HL-1 cells exposed to HFD serum. Unexpectedly, we have actin filaments required for GLUT4 trafficking to the membrane (a also found that HFD increased total InsR levels in rabbit heart, and that hallmark of skeletal muscle insulin resistance) [54]. We showed that anti-P3 Abs also blocked this effect. Interestingly, the presence of blocking intramyocardial CE accumulation with anti-P3 Abs was suffi­ excessive InsR levels in muscle tissues has been proposed to contribute cient to restore insulin-induced InsR signaling, GLUT4 trafficking to to dyslipemia and steatosis [49] and to be a compensatory mechanism in plasma membrane, and glucose uptake by HL-1 cardiomyocytes. Thus, a situation of the HFD-generated insulin resistance [50]. LRP1-mediated CE accumulation is an early and reversible mechanism In our model, both total and membrane LRP1 were higher in HFD-fed linking HFD to altered insulin response. Indeed, the multiple roles for groups as compared to standard chow control groups, and anti-P3 Abs LRP1 in mediating lipoprotein uptake [51,55], InsR signaling [6,25], reduced the LRP1 levels to near that found in the control group, as ex­ and GLUT4 trafficking [6,23] also underpin the link between the car­ pected by the positive transcriptional upregulation of LRP1 by hyper­ diomyocyte intracellular CE accumulation and cardiac insulin signaling cholesterolemia [51]. In the presence of anti-P3 Abs, the cardiac abnormalities. It is tempting to suggest that LRP1 binding to membrane InsR levels and the InsR/LRP1 interactions were even higher CE-enriched lipoproteins prevents the LRP1 receptor from interacting than those found in the control group, suggesting that blocking LRP1 with other receptors or transporters, such as InsR or GLUT4, respec­ interactions with lipoproteins could be a highly efficient process for tively, and this precludes their movement to plasma membrane. Further recovering a correct cardiac insulin response. Our ex vivo studies showed studies are required to explore these crucial mechanistic aspects con­ that reducing the levels of the InsR/LRP1 complexes by a HFD was necting LRP1-mediated intramyocardial CE accumulation and insulin directly responsible for alterations in InsR signaling, in agreement with resistance in peripheral tissues. As glucose uptake in the heart is largely previous studies showing the pivotal role of LRP1 in insulin-induced controlled by insulin-sensitive mechanisms, insulin resistance is an InsR phosphorylation [6,25,26]. Further studies are now required to obstacle for enhanced myocardial glucose utilization. In recent studies, determine how intramyocardial CE accumulation impairs InsR/LRP1 the use of insulin-sensitizing agents, such as metformin or SGLT2 complex formation and, how anti-P3 Abs overcome this effect. (sodium-glucose cotransporter 2) inhibitors, was shown to reduce the Insulin and glucose metabolism are tightly related, and glucose risk of cardiovascular death or hospitalizations for heart failure in pa­ transport across the cell membrane is strongly influenced by membrane tients with T2DM [56,57]. fluidity [52] and, modulated by cholesterol [53]. Our current results A crucial question that remains open in this study is why anti-P3 Abs from ex vivo studies showed that hypercholesterolemic serum reduced protected against HFD-induced cardiac but not peripheral insulin insulin response in terms of glucose uptake, and that anti-P3 Abs resistance. Peripheral insulin resistance is considered to be caused by restored this response. In rabbit hearts, we showed that GLUT4 traf­ hepatic steatosis [58], and HFD feeding dramatically increased accu­ ficking to cardiac membrane was reduced in HFD IrP rabbits but mulation of both CE and TG in the liver of this rabbit model. Of note, 10 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Fig. 7. Anti-P3 abs block the deleterious effects of hypercholesterolemic serum on GLUT4 traffic to the plasma membrane and glucose uptake by HL-1 cells. (a) Representative Western blot analysis from biotin-labeling protein assays of GLUT4 in the plasma membrane (upper panel) and in total cell extracts (lower panel) of HL-1 cells exposed to rabbit serum (0.25 %, 1.5 h) and stimulated with insulin (100 nM, 30 min). ATP1A1 and β-actin were used as protein loading controls for plasma membrane and total cell protein extracts, respectively. Line 1 represents control without biotin. (b) Bar graphs showing the quantification of Western blot bands for cell surface GLUT4 normalized to ATP1A1, and expressed as a fold-change as compared to the mean value of HL-1 exposed to serum from chow IrP group. Values are expressed as mean ± SD of three independent experiments (n = 3). (c) Bar graph showing the 2-NBDG uptake in HL-1 cells exposed to serum (0.25 %, 1.5 h) and then stimulated with insulin (100 nM, 30 min) in the presence of 2-NBDG (80 μM). The bar graph shows the mean ± SD of the fluorescence intensity of 2- NBDG per area of three independent experiments (n = 3), expressed as fold change as compared to the mean value of control IrP group. * p < 0.05; ** p < 0.01 insulin-stimulated versus basal. Statistical significance was determined by two-way ANOVA followed by a Student t-test. hepatic TGs are the main regulators of VLDL secretion by the liver [59]. the heart and restored insulin response. This aspect deserves to be Controversially, insulin resistance has been reported to decrease mem­ investigated indepth given the potential role of myocardial CE accu­ brane LRP1 levels in the liver, which could contribute to impaired he­ mulation on insulin resistance in a wide range of cardiac diseases. The patic clearance of triglyceride-rich lipoproteins. We now show that innovative immunization-based approach that we took here, which anti-P3 Abs had not protective effect against hepatic TG or CE accu­ differs from more traditional pharmacological or genetic techniques, mulation induced by HFD. Importantly, anti-P3 Abs blocked the inter­ showed both specificity and selectivity for targeting cardiovascular CE action of vascular LRP1 with ApoB100-enriched lipoproteins [29], the accumulation. These findings have important therapeutic implications main lipoprotein ligand of LRP1 in the cardiovascular system. However, in the management of CVD and shed new light on the complex role of in liver, the main lipoprotein ligands of LRP1 are ApoE-enriched lipo­ LRP1 in regulating cardiac insulin response associated with hypercho­ proteins, such as chylomicrons [60]. It is known that ApoB100 and ApoE lesterolemia, obesity, and T2DM. interact with the LRP1 receptor through domains localized in different clusters: CR9 in cluster II for ApoB100 [28] and CR17 in cluster IV for Research in context ApoE [61]. The inefficiency of anti-P3 Abs to block the interactions of LRP1 with ApoE-enriched lipoproteins could be the reason behind their Evidence before this study lack of effect on HFD-induced systemic insulin resistance in rabbits. In conclusion, our study now shows that i) diet-induced intra­ Cardiovascular disease, the leading cause of death worldwide, is myocardial CE accumulation impaired cardiac insulin response, and that frequently associated with increased levels of atherogenic lipoproteins. ii) anti-P3 Abs specifically and efficiently blocked intramyocardial CE Cholesteryl ester (CE)-enriched lipoproteins are an essential source of accumulation, thereby normalizing the cardiac insulin response (sum­ lipids to the heart; however, their dysregulated uptake and consequent marized in Fig. 8). Further studies are required to determine the lipid accumulation in the heart contributes to cardiac metabolic alter­ mechanisms underlying the link between reduced CE accumulation in ations and cardiac dysfunction. In particular, CE accumulation in the 11 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 Fig. 8. Schematic summary of the main findings. A high-fat diet (HFD) causes a strong intramyocardial accumulation of cholesteryl ester (CE) associated with altered cardiac insulin response. By specifically blocking intramyocardial CE accumulation, P3 immunization restores insulin signaling and glucose uptake in cardiomyocytes. heart causes calcium-handling alterations and cardiac remodeling in Acknowledgements experimental models. Recently, a clinical study with more than 15,000 individuals has revealed the link between high levels of CE-enriched The economic support to develop this project was received from lipoproteins and left ventricle myocardial remodeling. Antibodies Fundació MARATÓ TV3 with the grant 201521–10 (to V.Ll.-C), from the against the receptor LRP1 (the P3 sequence of CR9 domain of cluster II) Instituto de Salud Carlos III (ISCIII) and co-financed with ERDFsFIS specifically block intracellular CE accumulation in vascular cells by PI18/01584 (to V.Ll.-C), and from Fundación BBVA Ayudas a equipos de successful competition with aggregated LDL uptake. Moreover, P3 im­ investigación 2019. This work was also funded by Secretaría de Ciencia munization of rabbits inhibits vascular CE accumulation and y Tecnología de la Universidad Nacional de Córdoba (SECyT-UNC) atherosclerosis. grants PROYECTOS CONSOLIDAR 2018–2021 (to G.A.C); Fondo para la Investigación Científica y Tecnológica (FONCyT), Préstamo BID Proyecto de Investigación en Ciencia y Tecnología (PICT) grant Added value of this study 2015–0807 and grant 2017–4497 (to G.A.C). Support was received from Albert Renold Travel Fellowship Programme 2019 from the European In this study, we used an innovative immunization-based approach Foundation of the study of Diabetes (EFSD) and the Wood-Whelan to inhibit the CE uptake from lipoproteins into the heart. By comple­ fellowship Programme 2019 from the International Union of Biochem­ mentary molecular, biophysical, and imaging approaches, we show that istry and Molecular Biology (IUBMB) to cover the stay of V.A.D. in P3 immunization has specificity and selectivity for targeting intracel­ Institute of Biomedical Research of Barcelona (IIBB)-Spanish National lular CE accumulation within the heart of rabbits without modifying the Research Council (CSIC) and Biomedical Research Institute Sant Pau (IIB circulating lipid profile or the myocardial triglyceride accumulation. In Sant Pau), Barcelona, Spain. V.A.D is a postdoctoral fellow of Consejo addition, we found that the selective inhibition of CE accumulation is Nacional de Investigaciones Científicas y Técnicas (CONICET), Centro enough to restore both the levels of InsR and GLUT4 levels in the cardiac de Investigaciones en Bioquímica Clínica e Inmunología (CIBICI), membrane as well as the interaction between LRP1/InsR that is neces­ Córdoba, Argentina. A.B.A. is a predoctoral fellow (FI19/00205) gran­ sary for correct insulin signaling of hypercholesterolemic rabbits. ted by the Programme _Contratos predoctorales de formación de investigación en salud_ from the Instituto de Salud Carlos III (ISCIII) and Implications of all the available evidence co-financed with ERDFs. Our group is part of CIBER Enfermedades Cardiovasculares (CIBERCV; CB16/11/00276 to J.M.G and VLl-C) and This study provides new insight into lipid metabolism and its rela­ CIBER Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM; tionship to the insulin response in the heart and uncovers a novel and CB07/08/0016 to JCE-G), projects run by the Instituto de Salud Carlos different approach to specifically modifying cholesterol metabolism and III. Our group also participates in Redes de investigación (Enfermedades treat impaired cardiac insulin response. By providing a novel strategy of Metabóloicas y Cáncer RED2018–102799-T), a project run by MINECO. preventing the deleterious effect of hypercholesterolemia, it opens the VLl-C group recognized by Generalitat de Catalunya (2017 SGR 946). possiblity for new therapeutical strategies of clinical management of We are thankful to Eva Prats and the staff from Centres Científics i impaired insulin response. Strikingly, reducing LRP1-mediated CE Tecnològics de la Universitat de Barcelona (Campus Casanova) (CCiT/ accumulation can restore insulin signaling in the myocardium. This is an UB) for the processing of samples for electron microscopy. We also thank important finding given the role of cardiac lipid accumulation and in­ Veronica Raker for the scientitic and english editing of the manuscript. sulin response in a range of cardiac diseases and underscores the ther­ apeutic potential/utility of antibody therapies. Author contributions Conflict of interest statement V.A.D., and A.B.A contributed equally to this work. G.A.C. and V.Ll.- C. are both corresponding authors of this work. V.Ll.-C. and G.A.C. The authors declare no conflict of interest. designed the research; V.A.D., A.B.A., E.G.,L.C; and M.T.L.C-H 12 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 performed most of the experimental research; A.I. contributed with the LDL through LDL receptor–related protein and LDL receptors, Arterioscler. Thromb. Vasc. Biol. 22 (2002) 387–393, https://doi.org/10.1161/hq0302.105367. immunization procedure and ELISA design and setting to test anti-P3 Ab [21] V. Llorente-Cortés, J. Martínez-González, L. Badimon, LDL receptor–related protein production in rabbits; JC.E.G contributed with biochemical de­ mediates uptake of aggregated LDL in human vascular smooth muscle cells, terminations of lipoproteins; V.S. contributed with the biophysical Arterioscler. Thromb. Vasc. Biol. 20 (2000) 1572–1579, https://doi.org/10.1161/ analysis of lipoproteins and hearts; C.E. contributed with the electron 01.ATV.20.6.1572. [22] R. Cal, O. Juan-Babot, V. Brossa, S. Roura, C. Gálvez-Montón, M. Portoles, et al., microscopy experiments and images; V.A.D., A.B.A., J.M.G., G.A.C. and Low density lipoprotein receptor-related protein 1 expression correlates with V.Ll.-C. analyzed the data; and V.A.D., G.A.C. and V.Ll.-C. wrote the cholesteryl ester accumulation in the myocardium of ischemic cardiomyopathy paper. patients, J. Transl. Med 10 (2012) 160, https://doi.org/10.1186/1479-5876-10- 160. [23] M.P. Jedrychowski, C.A. Gartner, S.P. Gygi, L. Zhou, J. Herz, K.V. Kandror, et al., Appendix A. Supporting information Proteomic analysis of GLUT4 storage vesicles reveals LRP1 to Be an important vesicle component and target of insulin signaling, J. Biol. Chem. 285 (2010) 104–114, https://doi.org/10.1074/jbc.M109.040428. Supplementary data associated with this article can be found in the [24] D. Leto, A.R. Saltiel, Regulation of glucose transport by insulin: traffic control of online version at doi:10.1016/j.biopha.2022.113270. GLUT4, Nat. Rev. Mol. Cell Biol. 13 (2012) 383–396, https://doi.org/10.1038/ nrm3351. [25] C. Liu, J. Hu, C. Tsai, M. Yue, H.L. Melrose, T. Kanekiyo, et al., Neuronal LRP1 References regulates glucose metabolism and insulin signaling in the brain, J. Neurosci. 35 (2015) 5851–5859, https://doi.org/10.1523/JNEUROSCI.5180-14.2015. [1] H. Bugger, E. Abel, Molecular mechanisms of diabetic cardiomyopathy, [26] Y. Ding, X. Xian, W. Holland, S. Tsai, J. Herz, Low-density lipoprotein receptor- Diabetologia 57 (2014) 660–671, https://doi.org/10.1007/s00125-014-3171-6. related protein-1 protects against hepatic insulin resistance and hepatic steatosis, [2] C. Riehle, E. Abel, Insulin signaling and heart failure, Circ. Res. 118 (2016) EBioMedicine 7 (2016) 135–145, https://doi.org/10.1016/j.ebiom.2016.04.002. 1151–1169, https://doi.org/10.1161/CIRCRESAHA.116.306206. [27] V. Actis Dato, R. Grosso, M.C. Sánchez, C.F. Fader, G.A. Chiabrando, Insulin- [3] M. Tadic, C. Cuspidi, D. Vasic, P.L.M. Kerkhof, Cardiovascular implications of induced exocytosis regulates the cell surface level of low-density lipoprotein- diabetes, metabolic syndrome, thyroid disease, and cardio-oncology in women, related protein-1 in Müller Glial cells, Biochem J. 475 (2018) 1669–1685, https:// Adv. Exp. Med. Biol. 1065 (2018) 471–488, https://doi.org/10.1007/978-3-319- doi.org/10.1042/BCJ20170891. 77932-4_29. [28] P. Costales, P. Fuentes-Prior, J. Castellano, E. Revuelta-Lopez, M.A. Corral- [4] B. Bozkurt, D. Aguilar, A. Deswal, S.B. Dunbar, G.S. Francis, T. Horwich, et al., Rodríguez, L. Nasarre, et al., K domain CR9 of low density lipoprotein (LDL) Contributory risk and management of comorbidities of hypertension, obesity, receptor-related protein 1 (LRP1) is critical for aggregated LDL-induced foam cell diabetes mellitus, hyperlipidemia, and metabolic syndrome in chronic heart formation from human vascular smooth muscle cells, J. Biol. Chem. 290 (2015) failure: a scientific statement from the American Heart Association, Circulation 134 14852–14865, https://doi.org/10.1074/jbc.M115.638361. (2016) e535–e578, https://doi.org/10.1161/CIR.0000000000000450. [29] O. Bornachea, A. Benitez-Amaro, A. Vea, L. Nasarre, D. de Gonzalo-Calvo, J. [5] W.B. Kannel, M. Hjortland, W.P. Castelli, Role of diabetes in congestive heart C. Escola-Gil, et al., Immunization with the Gly 1127 -Cys 1140 amino acid failure: the framingham study, Am. J. Cardiol. 34 (1974) 29–34, https://doi.org/ sequence of the LRP1 receptor reduces atherosclerosis in rabbits. Molecular, 10.1016/0002-9149(74)90089-7. immunohistochemical and nuclear imaging studies, Theranostics 10 (2020) [6] V. Actis Dato, A. Benitez-Amaro, D. de Gonzalo-Calvo, M. Vazquez, G. Bonacci, 3263–3280, https://doi.org/10.7150/thno.37305. V. Llorente-Cortés, et al., LRP1-mediated AggLDL endocytosis promotes cholesteryl [30] S. Toldo, D. Austin, A.G. Mauro, E. Mezzaroma, B.W. Van Tassell, C. Marchetti, et ester accumulation and impairs insulin response in HL-1 Cells, Cells 9 (2020) 182, al., Low-density lipoprotein receptor–related protein-1 Is a therapeutic target in https://doi.org/10.3390/cells9010182. acute myocardial infarction, JACC: Basic Transl. Sci. 2 (2017) 561–574, https:// [7] G. Jia, M.A. Hill, J.R. Sowers, Diabetic cardiomyopathy: an update of mechanisms doi.org/10.1016/j.jacbts.2017.05.007. contributing to this clinical entity, Circ. Res 122 (2018) 624–638, https://doi.org/ [31] N. Potere, M.G. del Buono, G. Niccoli, F. Crea, S. Toldo, A. Abbate, Developing 10.1161/CIRCRESAHA.117.311586. LRP1 agonists into a therapeutic strategy in acute myocardial infarction, Int J. Mol. [8] Q. Hu, H. Zhang, N. Gutiérrez Cortés, D. Wu, P. Wang, J. Zhang, et al., Increased Sci. 20 (2019) 544, https://doi.org/10.3390/ijms20030544. Drp1 Acetylation by Lipid Overload Induces Cardiomyocyte Death and Heart [32] S. Avrameas, Coupling of enzymes to proteins with glutaraldehyde, Dysfunction, Circ. Res 126 (2020) 456–470, https://doi.org/10.1161/ Immunochemistry 6 (1969) 43–52, https://doi.org/10.1016/0019-2791(69) CIRCRESAHA.119.315252. 90177-3. [9] A. Ruiz-Ramírez, O. López-Acosta, M.A. Barrios-Maya, M. El-Hafidi, Cell death and [33] J. Julve, J.C. Escolà-Gil, N. Rotllan, C. Fiévet, E. Vallez, C. de la Torre, et al., heart failure in obesity: role of uncoupling proteins, Oxid. Med Cell Longev. 2016 Human apolipoprotein A-II determines plasma triglycerides by regulating (2016) 1–11, https://doi.org/10.1155/2016/9340654. lipoprotein lipase activity and high-density lipoprotein proteome, Arterioscler. [10] G. Jia, V.G. DeMarco, J.R. Sowers, Insulin resistance and hyperinsulinaemia in Thromb. Vasc. Biol. 30 (2010) 232–238, https://doi.org/10.1161/ diabetic cardiomyopathy, Nat. Rev. Endocrinol. 12 (2016) 144–153, https://doi. ATVBAHA.109.198226. org/10.1038/nrendo.2015.216. [34] J.C. Escolà-Gil, G. Llaverias, J. Julve, M. Jauhiainen, J. Méndez-González, [11] N. Frangogiannis, Cardiac fibrosis, Cardiovasc Res 117 (2021) 1450–1488, https:// F. Blanco-Vaca, The cholesterol content of western diets plays a major role in the doi.org/10.1093/cvr/cvaa324. paradoxical increase in high-density lipoprotein cholesterol and upregulates the [12] M. Tate, D.J. Grieve, R.H. Ritchie, Are targeted therapies for diabetic macrophage reverse cholesterol transport pathway, Arterioscler. Thromb. Vasc. cardiomyopathy on the horizon? Clin. Sci. (Lond. ) 131 (2017) 897–915, https:// Biol. 31 (2011) 2493–2499, https://doi.org/10.1161/ATVBAHA.111.236075. doi.org/10.1042/CS20160491. [35] R. Cal, J. Castellano, E. Revuelta-López, R. Aledo, M. Barriga, J. Farré, et al., Low- [13] N.G. Frangogiannis, The extracellular matrix in myocardial injury, repair, and density lipoprotein receptor-related protein 1 mediates hypoxia-induced very low remodeling, J. Clin. Invest 127 (2017) 1600–1612, https://doi.org/10.1172/ density lipoprotein-cholesteryl ester uptake and accumulation in cardiomyocytes, JCI87491. Cardiovasc Res 94 (2012) 469–479, https://doi.org/10.1093/cvr/cvs136. [14] M.R. Hayden, N. Chowdhury, G. Govindarajan, R. Poorna, P.R. Karuparthi, [36] I. Schulz, [19] Permeabilizing cells: Some methods and applications for the study J. Habibi, et al., Myocardial myocyte remodeling and fibrosis in the of intracellular processes, Methods Enzym. 192 (1990) 280–300, https://doi.org/ cardiometabolic syndrome, J. Cardiometab Syndr. 1 (2007) 326–333, https://doi. 10.1016/0076-6879(90)92077-Q. org/10.1111/j.1559-4564.2006.05626.x. [37] V. Samouillan, I.M. Martinez de Lejarza Samper, A. Benitez-Amaro, D. Vilades, [15] N. Aung, M.M. Sanghvi, S.K. Piechnik, S. Neubauer, P.B. Munroe, S.E. Petersen, J. Dandurand, J. Casas, et al., Biophysical and lipidomic biomarkers of cardiac The effect of blood lipids on the left ventricle, J. Am. Coll. Cardiol. 76 (2020) remodeling post-myocardial infarction in humans, Biomolecules 10 (2020) 1471, 2477–2488, https://doi.org/10.1016/j.jacc.2020.09.583. https://doi.org/10.3390/biom10111471. [16] M. Ruuth, S.D. Nguyen, T. Vihervaara, M. Hilvo, T.D. Laajala, P.K. Kondadi, et al., [38] V. Samouillan, E. Revuelta-López, J. Dandurand, L. Nasarre, L. Badimon, Susceptibility of low-density lipoprotein particles to aggregate depends on particle C. Lacabanne, et al., Cardiomyocyte intracellular cholesteryl ester accumulation lipidome, is modifiable, and associates with future cardiovascular deaths, Eur. promotes tropoelastin physical alteration and degradation, Int J. Biochem Cell Biol. Heart J. 39 (2018) 2562–2573, https://doi.org/10.1093/eurheartj/ehy319. 55 (2014) 209–219, https://doi.org/10.1016/j.biocel.2014.09.005. [17] C. Bancells, S. Villegas, F.J. Blanco, S. Benítez, I. Gállego, L. Beloki, et al., [39] C. Drevinge, L.O. Karlsson, M. Ståhlman, T. Larsson, J. Perman Sundelin, L. Grip, et Aggregated electronegative low density lipoprotein in human plasma shows a high al., Cholesteryl esters accumulate in the heart in a porcine model of ischemia and tendency toward phospholipolysis and particle fusion, J. Biol. Chem. 285 (2010) reperfusion, PLoS One 8 (2013), e61942, https://doi.org/10.1371/journal. 32425–32435, https://doi.org/10.1074/jbc.M110.139691. pone.0061942. [18] E. Hurt-Camejo, G. Camejo, P. Sartipy, Phospholipase A2 and small, dense low- [40] Y. Huang, K.E. Walker, F. Hanley, J. Narula, S.R. Houser, T.N. Tulenko, Cardiac density lipoprotein, Curr. Opin. Lipido 11 (2000) 465–471, https://doi.org/ systolic and diastolic dysfunction after a cholesterol-rich diet, Circulation 109 10.1097/00041433-200010000-00004. (2004) 97–102, https://doi.org/10.1161/01.CIR.0000109213.10461.F6. [19] G. Camejo, E. Hurt-Camejo, O. Wiklund, G. Bondjers, Association of apo B [41] A. Gastaldelli, M.A. Morales, P. Marraccini, R. Sicari, The role of cardiac fat in lipoproteins with arterial proteoglycans: pathological significance and molecular insulin resistance, Curr. Opin. Clin. Nutr. Metab. Care 15 (2012) 523–528, https:// basis, Atherosclerosis 139 (1998) 205–222, https://doi.org/10.1016/S0021-9150 doi.org/10.1097/MCO.0b013e328358be7b. (98)00107-5. [42] F. Nigon, P. Lesnik, M. Rouis, M.J. Chapman, Discrete subspecies of human low [20] V. Llorente-Cortés, M. Otero-Viñas, E. Hurt-Camejo, J. Martínez-González, density lipoproteins are heterogeneous in their interaction with the cellular LDL L. Badimon, Human coronary smooth muscle cells internalize versican-modified 13 V. Actis Dato et al. Biomedicine & Pharmacotherapy 152 (2022) 113270 receptor, J. Lipid Res 32 (1991) 1741–1753, https://doi.org/10.1016/S0022-2275 [53] N. Saxena, N.C. Chandra, Cholesterol: a prelate in cell nucleus and its serendipity, (20)41629-3. Curr. Mol. Med. 20 (2021) 692–707, https://doi.org/10.2174/ [43] J.R. McNamara, D.M. Small, Z. Li, E.J. Schaefer, Differences in LDL subspecies 1566524020666200413112030. involve alterations in lipid composition and conformational changes in [54] K.M. Habegger, B.A. Penque, W. Sealls, L. Tackett, L.N. Bell, E.K. Blue, et al., Fat- apolipoprotein B, J. Lipid Res 37 (1996) 1924–1935, https://doi.org/10.1016/ induced membrane cholesterol accrual provokes cortical filamentous actin S0022-2275(20)37557-X. destabilisation and glucose transport dysfunction in skeletal muscle, Diabetologia [44] W. Shen, S. Azhar, F. Kraemer, Lipid droplets and steroidogenic cells, Exp. Cell Res. 55 (2012) 457–467, https://doi.org/10.1007/s00125-011-2334-y. 340 (2016) 209–214, https://doi.org/10.1016/j.yexcr.2015.11.024. [55] V. Llorente-Cortés, M. Otero-Viñas, S. Camino-López, P. Costales, L. Badimon, [45] X. Zhou, Z. Mo, Y. Li, L. Huang, S. Yu, L. Ge, Y. Hu, S. Shi, L. Zhang, L. Wang, Cholesteryl esters of aggregated LDL are internalized by selective uptake in human L. Gao, G. Yang, G. Chu, Oleic acid reduces steroidogenesis by changing the lipid vascular smooth muscle cells, Arterioscler. Thromb. Vasc. Biol. 26 (2006) 117–123, type stored in lipid droplets of ovarian granulosa cells, J. Anim. Sci. Biotechnol. 13 https://doi.org/10.1161/01.ATV.0000193618.32611.8b. (2022) 27, https://doi.org/10.1186/s40104-021-00660-5. [56] J.M.M. Evans, A.S.F. Doney, M.A. AlZadjali, S.A. Ogston, J.R. Petrie, A.D. Morris, [46] V. Zoni, R. Khaddaj, P. Campomanes, A. Thiam, R. Schneiter, S. Vanni, Pre-existing et al., Effect of metformin on mortality in patients with heart failure and type 2 bilayer stresses modulate triglyceride accumulation in the ER versus lipid droplets, diabetes mellitus, Am. J. Cardiol. 106 (2010) 1006–1010, https://doi.org/ eLife (2021) 10, https://doi.org/10.7554/eLife.62886. 10.1016/j.amjcard.2010.05.031. [47] S.F. Shimobayashi, Y. Ohsaki, Universal phase behaviors of intracellular lipid [57] M. Packer, Reconceptualization of the molecular mechanism by which sodium- droplets, PNAS 116 (2019) 25440–25445, https://doi.org/10.1073/ glucose cotransporter 2 inhibitors reduce the risk of heart failure events, pnas.1916248116. Circulation 140 (2019) 443–445, https://doi.org/10.1161/ [48] B. Lu, D. Bridges, Y. Yang, K. Fisher, A. Cheng, L. Chang, et al., Metabolic crosstalk: CIRCULATIONAHA.119.040909. molecular links between glycogen and lipid metabolism in obesity, Diabetes 63 [58] C. Postic, R. Dentin, J. Girard, Role of the liver in the control of carbohydrate and (2014) 2935–2948, https://doi.org/10.2337/db13-1531. lipid homeostasis, Diabetes Metab. 30 (2004) 398–408, https://doi.org/10.1016/ [49] R.K. Semple, A. Sleigh, P.R. Murgatroyd, C.A. Adams, L. Bluck, S. Jackson, et al., S1262-3636(07)70133-7. Postreceptor insulin resistance contributes to human dyslipidemia and hepatic [59] B.C. Moon, A. Hernandez-Ono, B. Stiles, H. Wu, H.N. Ginsberg, Apolipoprotein B steatosis, J. Clin. Invest 119 (2009) 315, https://doi.org/10.1172/JCI37432. secretion is regulated by hepatic triglyceride, and not insulin, in a model of [50] G. Wang, Y. Yu, W. Cai, T.M. Batista, S. Suk, H.L. Noh, et al., Muscle-specific increased hepatic insulin signaling, Arterioscler. Thromb. Vasc. Biol. 32 (2012) insulin receptor overexpression protects mice from diet-induced glucose 236–246, https://doi.org/10.1161/ATVBAHA.111.241356. intolerance but leads to postreceptor insulin resistance, Diabetes 69 (2020) [60] A. Rohlmann, M. Gotthardt, R.E. Hammer, J. Herz, Inducible inactivation of 2294–2309, https://doi.org/10.2337/db20-0439. hepatic LRP gene by cre-mediated recombination confirms role of LRP in clearance [51] V. Llorente-Cortés, M. Otero-Viñas, S. Sánchez, C. Rodríguez, L. Badimon, Low- of chylomicron remnants, J. Clin. Invest 101 (1998) 689–695, https://doi.org/ density lipoprotein upregulates low-density lipoprotein receptor-related protein 10.1172/JCI1240. expression in vascular smooth muscle cells, Circulation 106 (2002) 3104–3110, [61] M. Guttman, J.H. Prieto, J.E. Croy, E.A. Komives, Decoding of https://doi.org/10.1161/01.CIR.0000041434.28573.0B. lipoprotein− receptor interactions: properties of ligand binding modules governing [52] N. Al-Makdissy, M. Younsi, S. Pierre, O. Ziegler, M. Donner, Sphingomyelin/ interactions with apolipoprotein E, Biochemistry 49 (2010) 1207–1216, https:// cholesterol ratio: an important determinant of glucose transport mediated by doi.org/10.1021/bi9017208. GLUT-1 in 3T3-L1 preadipocytes, Cell Signal 15 (2003) 1019–1030, https://doi. org/10.1016/s0898-6568(03)00070-6. 14