Received: 24 January 2020 | Revised: 8 August 2020 | Accepted: 27 August 2020 DOI: 10.1096/fj.202000177R RESEARCH ARTICLE CD163 deficiency increases foam cell formation and plaque progression in atherosclerotic mice Carmen Gutiérrez-Muñoz1 | Nerea Méndez-Barbero1,2 | Pia Svendsen3 | Cristina Sastre4 | Valvanera Fernández-Laso1,2 | Patricia Quesada1 | Jesús Egido5,6 | Joan C. Escolá-Gil6,7 | Jose L. Martín-Ventura1,2 | Soren K. Moestrup8,9 | Luis M. Blanco-Colio1,2 1 Vascular Research Laboratory, IIS- Fundación Jiménez Díaz University Abstract Hospital, Madrid, Spain Atherosclerosis is an inflammatory disease characterized by the accumulation of mac- 2 Centro de Investigación Biomédica en rophages in the vessel wall. Macrophages depend on their polarization to exert either Red de Enfermedades Cardiovasculares pro-inflammatory or anti-inflammatory effects. Macrophages of the anti-inflamma- (CIBERCV), Madrid, Spain 3 tory phenotype express high levels of CD163, a scavenger receptor for the hemo- Department of Clinical Medicine, Aarhus University Hospital, Aarhus, Denmark globin-haptoglobin complex. CD163 can also bind to the pro-inflammatory cytokine 4 Center for Genomic Medicine, TWEAK. Using ApoE-deficient or ApoE/CD163 double-deficient mice we aim to in- Massachusetts General Hospital, Boston, vestigate the involvement of CD163 in atherosclerosis development and its capacity MA, USA 5 to neutralize the TWEAK actions. ApoE/CD163 double-deficient mice displayed a Renal, Vascular and Diabetes Research Laboratory, IIS-Fundación Jiménez Díaz, more unstable plaque phenotype characterized by an increased lipid and macrophage Madrid, Spain content, plaque size, and pro-inflammatory cytokine expression. In vitro experiments 6 Centro de Investigación Biomédica en Red demonstrated that the absence of CD163 in M2-type macrophages-induced foam cell de Diabetes y Enfermedades Metabólicas (CIBERDEM), Madrid, Spain formation through upregulation of CD36 expression. Moreover, exogenous TWEAK 7 Institut d'Investigacions Biomèdiques (IIB) administration increased atherosclerotic lesion size, lipids, and macrophages content Sant Pau, Barcelona, Spain in ApoE−/−/CD163−/− compared with ApoE−/−/CD163+/+ mice. Treatment with re- 8 Department of Molecular Medicine, combinant CD163 was able to neutralize the proatherogenic effects of TWEAK in University of Southern Denmark, Oddense, ApoE/CD163 double-deficient mice. Recombinant CD163 abolished the pro-inflam- Denmark 9 Department of Biomedicine, Aarhus matory actions of TWEAK on vascular smooth muscle cells, decreasing NF-kB ac- University, Aarhus, Denmark tivation, cytokines and metalloproteinases expression, and macrophages migration. In conclusion, CD163-expressing macrophages serve as a protective mechanism to Correspondence Abbreviations: ABCA1, ATP-binding cassette subfamily; ABCG1, ATP-binding cassette subfamily G member 1; ADAM17, ADAM metallopeptidase domain 17; ApoE−/− , apolipoprotein E-deficient mice; Arg1, arginase 1; BCA, brachiocephalic arteries; CCL2, chemokine ligand 2; CCL5, chemokine ligand 5; CD36, cluster of differentiation 36; CD68, cluster of differentiation 68; CD163, 130nonbreakingspacekDa member of the scavenger receptor cysteine rich (SRCR) family; Fn14, fibroblast growth factor-inducible 14; ICAM-1, intercellular adhesion molecule 1; IL-4, interleukin 4; IL-10, interleukin 10; IL-13, interleukin 13; LDH, lactate dehydrogenase; LXR, liver X receptor; MMP-2, matrix metallopeptidase 2; MMP-9, matrix metallopeptidase 9; NF-kB, nuclear factor-kB; Nos2, nitric oxide synthase 2; ORO, Oil Red O; ox-LDL, oxidized low-density lipoproteins; rCD, recombinant 130nonbreakingspacekDa member of the scavenger receptor cysteine rich (SRCR) family; rTW, recombinant soluble tumor necrosis factor-like weak inducer of apoptosis; SR-A, scavenger receptor A; TWEAK, tumor necrosis factor-like weak inducer of apoptosis; VSMCs, vascular smooth muscle cells. Carmen Gutiérrez-Muñoz and Nerea Méndez-Barbero contributed equally to this work. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2020 The Authors. The FASEB Journal published by Wiley Periodicals LLC on behalf of Federation of American Societies for Experimental Biology 14960 |  wileyonlinelibrary.com/journal/fsb2 The FASEB Journal. 2020;34:14960–14976. GUTIÉRREZ-MUÑOZ et al.     | 14961 Luis M. Blanco-Colio, Vascular Research Laboratory, IIS-Fundación Jiménez Díaz prevent the deleterious effects of TWEAK on atherosclerotic plaque development University Hospital, Av. Reyes Católicos 2, and progression. 28040, Madrid, Spain. Email: lblanco@fjd.es KEYWORDS atherosclerosis, CD163, inflammation, TWEAK Funding information MINECO | Instituto de Salud Carlos III (ISCIII), Grant/Award Number: PI16/01419 and PI19/00128 1 | IN T RO D U C T IO N plaque angiogenesis and vascular permeability in human atherosclerosis.11 Nevertheless, the direct role of CD163 in Atherosclerosis is a lipid-driven inflammatory disease char- the development of atherosclerotic plaques deserves further acterized by monocyte recruitment and differentiation to studies. macrophages/foam cells.1 During the early stages of ath- CD163 is also a scavenger receptor for tumor necrosis erosclerosis, oxidized low-density lipoproteins (ox-LDL) factor-like weak inducer of apoptosis (TWEAK; Tnfsf12).12 accumulated in the intima, induce the activation of both en- TWEAK and its soluble form are mainly expressed and dothelial cells and vascular smooth muscle cells (VSMCs), secreted by inflammatory leukocytes.13 TWEAK acts by leading to the expression and secretion of several pro-inflam- binding to its sole functional receptor fibroblast growth fac- matory cytokines, chemokines, and adhesion molecules that tor-inducible 14 (Fn14; Tnfrsf12a) that is linked to several recruit monocytes within the arterial wall. Within the intima, intracellular and inflammatory signaling pathways.14 The monocytes differentiate into the macrophages to remove ox- signaling induced by TNF superfamily receptors involves LDL. Macrophages are heterogeneous immune cells that the presence of death domains in their cytoplasmic tail. respond to the pathophysiological signals such as lipopoly- However, the Fn14 cytoplasmic tail is too short to have a saccharide and interferon-γ to form pro-inflammatory M1 death domain, but it contains a TNF receptor-associated phenotype or to IL-4 and IL-13 leading to anti-inflammatory factor (TRAF)-binding site that could be potentially phos- M2 phenotype.2 Whereas the M1-type macrophages seem to phorylated to induce TRAF-binding and subsequent trans- have direct effect in pathogen killing, M2-type macrophage mission of TWEAK signaling.15 TWEAK trimerizes and express high levels of scavenging molecules and anti-inflam- binds to Fn14 monomers, promoting receptor trimerization matory cytokines, promoting tissue remodeling.3 However, and signal transduction.15 Both, TWEAK and Fn14 are M1/M2 nomenclature is a simplification of a continuum of closely related in humans and mice, being that their homol- complex macrophage phenotypes. ogy is higher than 90%.16 This detail is of importance since M2-type macrophages express high levels of CD163, TWEAK does not cross-react with any other members of a member of the scavenger receptor cysteine rich family.4 the TNF or TNFR superfamilies, being its interaction spe- CD163 was identified as an endocytic receptor for removal of cific for Fn14. hemoglobin:haptoglobin (Hb:Hp) complexes to avoid toxic TWEAK/Fn14 axis participates in several steps related effects of the oxidant Hb.5 Scavenging of Hb:Hp complexes to atherosclerotic plaque development and progression by CD163 is part of an adaptive process, linked to secretion of such as endothelial dysfunction, inflammation, prolif- the anti-inflammatory cytokine interleukin-10 and elevation eration, and thrombosis.17-19 In fact, while recombinant of heme oxygene-1.6 Shedding of the extracellular domain of TWEAK injection aggravates vascular lesion,20 TWEAK CD163 from macrophages by ADAM17 generates a soluble deficiency as well as Fn14-Fc or anti-TWEAK antibody form (sCD163) that is a normal constituent of plasma.7 CD163 treatment display reduced atherosclerotic lesion size and is considered an anti-inflammatory molecule involved in the increased plaque stability in ApoE-deficient mice.21,22 resolution of inflammation since pro-inflammatory stimuli However, little is known about the biological mechanisms reduce its expression.8,9 Decreased expression of CD163 on regulating TWEAK bioavailability and the downstream peripheral blood mononuclear cells has been observed in pathways by which TWEAK might participate in athero- patients with a specific Hp genotype (Hp 2 to 2), which is sclerotic progression. associated with an increased risk of cardiovascular disease.10 In the present study, we have now investigated how CD163 In addition, CD163 expression is associated to areas with in- deficiency affects TWEAK levels and atherosclerosis in mice flammation in injured tissues.6 However, it was reported that to further investigate the role of CD163 during atherosclero- the presence of CD163-expressing macrophages promotes sis progression. 14962 |    GUTIÉRREZ-MUÑOZ et al. 2 | M AT E R IA L S A N D ME T HODS performed for analysis of collagen content by measuring bi- refringence to plane-polarized light. Immunohistochemistry 2.1 | Animals was carried out as previously described.22 To analyze the effect of CD163 deficiency on the pro- Female ApoE−/− mice (#002052; Jackson Laboratory) were gression of the atherosclerotic plaques in mice, we used the crossed with male CD163−/− mice (both on the C57BL/6 Stary method, which classified the atherosclerotic lesions background), and the progeny bred back to ApoE−/− mice to based on their histological composition and structure23: obtain the double knockout (ApoE−/−CD163-/-) and their lit- grade I, early plaques containing only macrophages; grade termate control (ApoE−/−CD163-/-) mice. To study the effect II, lesions containing macrophages, VSMCs and a few cho- of CD163 deletion on atherosclerotic lesions, 12-week-old lesterol clefts; grade III, lesions containing macrophages, ApoE−/−CD163-/- (N = 10), and ApoE−/−CD163-/- (N = 10) VSMCs and numerous cholesterol clefts; and grade IV, were fed on a hyperlipidemic diet (21.1% fat [0.15% choles- advanced plaques containing macrophages, VSMCs and a terol] +16.7% proteins) during 8 weeks. In addition, TWEAK- large lipid core. accelerated atherosclerosis was induced in 12-week-old Brachiocephalic arteries were serially sectioned in 5 μm ApoE−/−CD163-/- (N = 10), and ApoE−/−CD163-/- (N = 10) thickness from the aortic root to the right subclavian ar- feeding a hyperlipidemic diet by murine TWEAK injection tery. For morphometric analysis, sections of each brachio- (5 μg/kg/three times per week; 1237-TW; R&D Systems) cephalic artery were stained with modified Russell-Movat during the last 4 weeks. A fifth group of ApoE−/−CD163-/- pentachrome (Movat) at 90 μm intervals from 0 to 450 μm (N = 8) mice feeding a hyperlipidemic diet were injected distal to the aortic root. The frequency of plaque instabil- with rCD163 (50 μg/kg/three times per week; 7435-CD; ity features in each Movat-stained section was evaluated R&D Systems) and 2 hours later with TWEAK (5 μg/kg/ (5-6 slides per animal, 40 slides per group), including the three times per week) during the last 4 weeks. following: necrotic core area, medial enlargement/erosion All mice were maintained under barrier conditions. Water (defined as the replacement of the normal aorta by plaque and diet were available ad libitum. At the end of the study, components), and the presence of buried caps (signature of 16-hour fasted mice were anesthetized and euthanized by silent plaque rupture and confirmed by α-SMA staining). overdose of 100 mg/kg ketamine and 15 mg/kg xylazine and These parameters were recorded as binary outcome, and saline perfused. Blood samples were collected for biochem- the frequency per lesion for each animal was determined istry. Total serum cholesterol, HDL-c, and triglycerides were with a 100% maximum possible. measured in our central laboratory (IIS-Fundación Jiménez Díaz). Serum Lactate Dehydrogenase (LDH) activity was measured with LDH colorimetric assay kit (ab102526; 2.3 | Immunohistochemical analysis Abcam) following the manufacturer's instructions. The housing and care of animals and all procedures carried Immunohistochemical analysis was done as previously out in this study were strictly in accordance with the Directive described.24 Primary Abs were the monocyte/macrophage 2010/63/EU of the European Parliament and were approved marker CD68 (Ab53444; Abcam), and the smooth mus- by the Institutional Animal Care and Use Committee of IIS- cle cell marker smooth muscle actin (Clone 1A4; Sigma). Fundación Jiménez Díaz. Goat anti-rat biotin (Amersham) was used as secondary antibodies. ABComplex/HRP was then added and sections were stained with AEC substrate-chromogen (DAKO), 2.2 | Aortic root and brachiocephalic artery counterstained with hematoxylin, and mounted in Pertex morphometric analysis (Medite). Incubation without primary Abs and/or irrele- vant species- and isotype-matched immunoglobulins was Brachiocephalic arteries (BCA) and hearts containing aor- used as a negative control for allimmunostainings. For tic roots were carefully dissected and frozen in OCT. Aortic colocalization studies anti-rat Alexa Fluor 568 was used roots were sectioned at 5 μm thickness beginning proximally as secondary Ab. at the first evidence of the aortic valves at their attachment Computer-assisted morphometric analysis was performed site of aorta. Sections were stained with Oil red O/hematoxy- with Image-Pro Plus software (version 1.0 for Windows). The lin at 100 μm intervals from 0 to the end of the aortic valves. threshold setting for area measurement was equal for all im- Maximal lesion area was calculated for each mouse by aver- ages. Samples from each animal were examined in a blinded aging the values for three sections. The individual maximal manner. Results were expressed as % positive area vs total lesion areas were further averaged to determine the maxi- area (collagen, macrophages and α-actin) or CD68+ cells vs mal lesion area for each group. Picrosirius red staining was total cells (BCA). GUTIÉRREZ-MUÑOZ et al.     | 14963 2.4 | Evans Blue permeability assays 2.7 | Flow cytometry ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice were in- For the uptake assays, peritoneal macrophages from ApoE- travenously injected with 20 mg/kg of 1% Evans blue dye deficient or ApoE/CD163 double deficient mice were washed (#E2129, Sigma-Aldrich), and perfused with saline after once in PBS and incubated in RPMI media with IL-10 (4 U/ 30 minutes. Aorta, lungs, kidney, liver, and front and back mL) during 48 hours. After that, cells were incubated in legs were removed for later analysis. Tissues were weighed fresh media containing Dil-ox-LDL (10 μg/mL; L34358; and Evans blue dye was extracted from different tissues by ThermoFisher Scientific) for 2 hours at 37°C. Then, cells formamide at 56°C overnight. Evans blue was quantified by were washed and resuspended in 0.5 mL of PBS and ana- spectrophotometry at 595 nm. Results were expressed in ng lyzed by flow cytometry. of Evans Blue per mg of tissue. For CD36 immunostaining, peritoneal macrophages from ApoE-deficient or ApoE/CD163 double deficient mice were washed once in PBS and incubated in RPMI 2.5 | Cell culture media with IL-10 (4 U/mL) during 48 hours. After that, cells were suspended in PBS containing 4% BSA, preincu- Aortic VSMCs were isolated from aorta of wild-type mice. bated with Mouse BD Fc Block (553141; BD Pharmingen) Briefly, mice were anesthetized (100 mg/kg ketamine and for 10 minutes and costained for 18 hours at 4°C with an- 15 mg/kg xylazine), saline perfused, and aortas were iso- ti-CD36 (Ab23680; Abcam). Alexa Fluor 488-conjugated lated. Adhering fat and connective tissue were removed goat anti-mouse (A11001; Life Technology) was used as and vessels were sectioned in small rings and incubated in secondary antibody. Cells were counted by flow cytom- DMEM (Sigma) containing 1 mg/mL collagenase (type II, etry using a BD FACS Canto II Flow Cytometer (BD 290 U/mg), penicillin (100 U/mL), streptomycin (100 μg/ Biosciences). mL), and glutamine (2 mmol/L) (Sigma) for 40 minutes at 37°C in 95% air/5% CO2. Then, the reaction was stopped with DMEM with 20% fetal bovine serum (FBS) and cells 2.8 | PCR were seeded in plastic culture flasks (Costar) in DMEM with 20% FBS. Cells were harvested for passaging at 2- Total RNA from VSMCs, peritoneal macrophages, or to 3-day intervals and used between the second and sev- total aorta of mice was obtained by TRIzol method (Life enth passages. For experimental analysis, cells were made Technologies) and quantified by absorbance at 260 nm in quiescent by 24-hour incubation in medium without FBS duplicate. Real-time PCR was performed on a TaqMan before stimulation with recombinant murine TWEAK ABI 7700 Sequence Detection System using heat-acti- (0.1 μg/mL; 1237-TW; R&D Systems) and preincubated vated TaqDNA polymerase (Amplitaq Gold). After an or not with different concentrations of rCD163 (7435-CD; initial hold of 2 minutes at 50°C and 10 minutes at 95°C, R&D Systems). the samples were cycled 40 times at 95°C for 15 seconds and 60°C for 60 seconds. A total of 18S rRNA served as housekeeping gene and was amplified in parallel with the 2.6 | Foam cell formation genes of interest. The expression of target genes was nor- malized to housekeeping transcripts. The following PCR Peritoneal macrophages were collected from ApoE-deficient primers and TaqMan probes were purchased from Applied or ApoE/CD163 double-deficient mice by peritoneal lavage Biosystems and optimized according to the manufacturer's 4 days after intraperitoneal injection of 3% (wt/vol) thiogly- protocol: 18S (4310893E), mouse CCL5 (Mm01302427), colate. Cells were cultured 24 hours in RPMI supplemented CCL2 (Mm00441242), MMP-2 (Mm00439498), MMP- with 10% FBS, L-glutamine, and antibiotics and then, stimu- 9 (Mm00442991), CD163 (Mm00474091), and ICAM-1 lated for 48 hours in medium containing 2% FBS and 4 U/mL (00516023). All measurements were performed in tripli- IL10 (11340103; Immunotools). After that, cells were incu- cate. The amount of target mRNA in samples was estimated bated with 10 μg/mL Dil-ox-LDL for different time points. In by the 2ΔCT relative quantification method. Values of each some cases, cells were also treated with anti-CD36 antibody sample were obtained as multiples of their baseline values. (2 μg/mL) (JC63.1; ab23680; Abcam) 1-hour prior Dil-ox- Mouse mRNA levels for Nos2, Arg1, SR-A, CD36, ABCG1, LDL incubation. Cells were fixed in 4% paraformaldehyde and ABCA1 were done by amplification of cDNA using for 15 minutes and stained with DAPI. Percentage of foam SYBR Premix Ex TaqTM (Takara Biotechnology). The cells were analyzed in 5-10 random different field per cell primer sequences are summarized in Table S1. The mRNA culture. levels were normalized to the GADPH mRNA content. 14964 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 1 CD163 reduces atherosclerotic lesion size in ApoE knockout mice. A, sTWEAK serum concentrations in ApoE−/−CD163+/+ (n = 10) or ApoE−/−CD163−/− (n = 10) mice at the end of the study. *P < .001, Student's t test. B, Representative Oil Red O/Hematoxylin and Sirius Red staining in ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice at the end of the study. Scale bars, 200 μm. C, Quantification of lesion area, (D) Oil Red O and (E) Sirius Red staining in the aortic root of ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice. Values shown are mean ± SEM of 10 animals per group. *P < .01, Student's t test. F, Representative photographs of Movat's pentachrome sections, Oil Red O/Hematoxylin and Sirius Red staining in the brachiocephalic artery from ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice. Scale bars, 200 μm. G, Quantification of lesion area, % stenosis, and (H) Oil Red O and Sirius Red staining in the brachiocephalic arteries of ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice. Values shown are mean ± SEM of 8-10 animals per group. *P < .005, Student's t test GUTIÉRREZ-MUÑOZ et al.     | 14965 2.9 | Migration assay chamber. The lower wells of the chemotaxis chambers were filled with the supernatant of VSMCs in the differ- Migration of murine macrophages Raw 264.7 were meas- ent experimental conditions. After 4 hours of nonmigrat- ured in 8 μm pore transwell 24-well cell culture inserts ing cells were removed from the upper part of the chamber (Costar). Cells were resuspended in DMEM with 0.2% and, the nuclei of migrated cells were fixated and stained BSA and seeded (70 × 104 per well) into the migration with DAPI. The number of migrated cells was counted in 14966 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 2 CD163 reduces plaque progression and diminishes features of plaque instability in ApoE knockout mice. A, Representative photographs of Oil Red O, Sirius red, α-SMA + DAPI, and CD68 stained sections classified according to the Stary method (grade I to IV). Scale bar, 100 μm. B, Percentage of each grade from Stary method between different groups. N = 25 plaques per group. C, Quantification of CD68 and α-SMA in aortic root from ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice. Values shown are mean ± SEM of 10 animals per group. *P < .005, Student's t test. D, Relative CCL2 and CCL5 mRNA expression levels normalized to 18S rRNA of aortas from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice. Data represent the mean ± SEM of 6 animals per group; *P < .001, Student's t test. E, Representative images of BCA cross-sections stained with MOVAT from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice. Dashed lines show the boundary of the developing necrotic area (NC). Quantification of the necrotic core area is shown in the below panel. Data represent the mean ± SEM of nine animals per group; *P < .005, Student's t test. Scale bar, 100 μm. F, LDH activity in serum from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice. Data represent the mean ± SEM of six animals per group; *P < .001, Student's t test. G, Representative images of BCA cross sections showing the presence of buried cap (α-SMA staining) and medial erosion (MOVAT staining) in ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice. Below panel: Frequency of morphological markers of plaque instability in the below panel. Data represent the mean ± SEM of nine animals per group; *P < .05, Student's t test. Scale bar, 100 μm (25 μm for medial erosion). H, Representative images of BCA cross sections stained with α-SMA or CD68 from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice. Quantification of CD68 positive cells is shown in the below panel. Data represent the mean ± SEM of eight animals per group; *P < .001, Student's t test. Scale bar, 100 μm 10-randomly selected fields by Nikon Eclipse E400 fluo- 2.12 | Statistical analysis rescence microscope. Data are expressed as mean ± the standard error of the mean (SEM). Kolmogorov-Smirnov test was used to assess normal 2.10 | Western blot distribution. Groups were compared with Student's t test or one-way ANOVA with Bonferroni's posttest. Significance Cultured murine VSMCs from different experimental con- was accepted at the level of P < .05. Statistical analysis was ditions were collected and lysated in lysis buffer containing performed using GraphPad Prism 6.0 (GraphPad, San Diego, 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 2 mM EDTA, 2 mM CA). EGTA, 0.2% Triton X-100, 0.3% NP-40, 0.2 mM PMSF, 0.2 mM Na3VO4, and 10 μL/mL of phosphatase inhibitor cocktail (P0044 Sigma) and pelleted. After normalizing for 3 | RESULTS equal protein concentration, cell lysates were resuspended in SDS sample buffer before separation by SDS-PAGE as 3.1 | CD163 promotes beneficial plaque previously described (23). Following transfer of the proteins remodeling in an atherogenic environment onto nitrocellulose membranes, membranes were probed using the following antibodies; NF-kB p65 (D14E12; Cell The main role of CD163 is to limit Hb toxicity within ather- Signaling), phospho-NF-kB p65 (S536; Cell Signaling), and osclerotic plaques.6 For that reason, we chose to investigate alpha-tubulin (T5168; Sigma). whether targeting CD163 in vivo would alter atherogenesis and plaque composition. To this end, 12-week-old mice lacking both CD163 and ApoE (ApoE−/−CD163−/−) and 2.11 | Cholesterol efflux their littermate ApoE−/−CD163+/+ single knockout mice were fed a hyperlipidemic diet for 8 weeks. No differences Peritoneal macrophages were isolated and labeled with were observed in body weight or lipid concentrations be- [1α,2α(n)-3H]cholesterol (2 million of cells per well, 1 µCi tween the different groups (Figure S1). Interestingly, per well) during 48 hours of incubation, as previously de- ApoE−/−CD163−/− mice presented increased sTWEAK scribed.25 Cells were washed and incubated with a serum-free plasma concentrations compared with ApoE−/−CD163+/+ medium (RPMI), which is supplemented with fatty acid-free mice (Figure 1A). BSA for 18 hours to allow equilibration of the radiolabeled ApoE−/−CD163−/− mice showed a non-significant cholesterol with the intracellular cholesterol pools. This ex- 21% increase plaque size at the aortic root compared with perimental step was carried out under basal conditions of after ApoE−/−CD163+/+ mice (Figure 1B,C). However, when we adding liver X receptor (LXR) agonist T0901317 (4 μM) to analyzed other vascular territory (brachiocephalic arteries; induce the expression of ABCA1 and ABCG1 transporters. BCA), a significant increment of 110% in plaque size and [3H]Cholesterol-labeled cells were then incubated for 4 hours 58% in stenosis was observed in ApoE−/−CD163−/− mice at 37°C with a 5% dilution of murine apoB-depleted serum. compared with their littermate mice (Figure 1F,G). Radioactivity was then measured in both medium and cells Whereas collagen fibers stabilize atherosclerotic plaques, and the percentage of cholesterol efflux calculated. lipid depositions make plaques more prone to rupture.26 GUTIÉRREZ-MUÑOZ et al.     | 14967 ApoE−/−CD163−/− mice exhibited increased Oil Red O To analyze the effect of CD163 deficiency in the pro- (ORO) content in atherosclerotic plaques present in both aor- gression of the atherosclerotic plaques in mice, we used the tic root (Figure 1B,D) and BCA (Figure 1F,H) compared with Stary method23 (Figure 2A). Approximately 40% plaques in ApoE−/−CD163+/+ mice. Collagen content was similar in ath- ApoE−/−CD163+/+ mice were early lesions (grade I) while erosclerotic lesions present in aortic root (Figure 1B,E) or only ≈13% plaques were advanced plaques (grade IV) BCA (Figure 1F,H) of both genotypes. (Figure 2B). In contrast, the percentage of advanced plaques 14968 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 3 Absence of CD163 promotes foam cell formation. A, Relative CD163 mRNA expression levels normalized to 18S rRNA of peritoneal macrophages from ApoE−/−CD163+/+ mice treated with IL-10 (4 ng/mL) during 48 hours. Data represent the mean ± SEM of three independent experiments; *P < .05; Student's t test. B, Representative images of ApoE−/−CD163+/+ or ApoE−/−CD163−/− peritoneal macrophages incubated with or without Dil-ox-LDL (10 μg/mL) for 4 hours. Quantification of Dil-ox-LDL positive cells incubated with or without Dil-ox-LDL for 2-6 hours in the right panel. Data represent the mean ± SEM of four independent experiments; *P < .05 and **P < .005; Student's t test. Scale bars, 50 μm (low magnification); 20 μm (High magnification). C, Flow cytometry analysis of Dil-ox-LDL uptake in peritoneal macrophages of ApoE−/−CD163+/+ or ApoE−/−CD163−/− incubated with Dil-ox-LDL (10 μg/mL) for 2 hours. The results are expressed in terms of geometric mean fluorescence intensity (MFI) after subtracting the autofluorescence of cells incubated in absence of Dil-ox-LDL. Data represent the mean ± SEM of four independent experiments. *P < .05; Student's t test. D, Cholesterol efflux to ApoB-depleted plasma in peritoneal macrophages from ApoE−/−CD163+/+ or ApoE−/−CD163−/− stimulated with or without LXR agonist. Data represent mean ± SEM of duplicate samples (n = 3 per group). E, Relative CD36, SR-A, ABCA1, and ABCG1 mRNA expression levels normalized to 18S rRNA of peritoneal macrophages from ApoE−/−CD163+/+ mice treated with IL-10 (4 ng/mL) during 48 hours. Data represent the mean ± SEM of five independent experiments; *P < .05; Student's t test. F, Flow cytometry analysis of CD36 expression in peritoneal macrophages of ApoE−/−CD163+/+ or ApoE−/−CD163−/− incubated with IL-10 (4 ng/mL) during 48 hours. The results are expressed in terms of geometric mean fluorescence intensity (MFI) after subtracting the autofluorescence of cells incubated in absence of CD36 antibody. Data represent the mean ± SEM of four independent experiments. *P < .05; Student's t test. G, Representative images of ApoE−/−CD163−/− peritoneal macrophages incubated with or without Dil-ox-LDL (10 μg/mL) for 4 hours. Anti-CD36 was preincubated 1 hour before Dil-ox-LDL treatment. IgG was used as control was ≈25% in ApoE−/−CD163−/− mice while ≈23% plaques Collectively, these results demonstrate that CD163 defi- were early lesions (Figure 2B). To further confirm that ciency increases the macrophages infiltration and plaque pro- CD163 is negatively associated to plaque progression, we gression in hyperlipidemic mice. analyzed the BCA for other features that both associate with and contribute to diminish plaque stability such as necrotic core size, and the presence of medial erosion and buried 3.2 | CD163 deficiency increases foam cell caps.27 These features are more evident in BCA than in the formation by CD36 upregulation aortic root. ApoE−/−CD163−/− mice exhibited an increased necrotic core size (Figure 2E). Accordingly, LDH activity, Macrophage foam cell formation as a result of the excess a marker of tissue damage, was increased in serum from of lipid deposition is an important step in atherosclerotic ApoE−/−CD163−/− mice compared to ApoE−/−CD163+/+ plaque development.30 Since we have observed that CD163- mice (Figure 2F). A higher frequency of medial erosion deficient mice showed atheroma plaques with higher number was also observed in ApoE−/−CD163−/− mice compared to of macrophages and higher intracellular lipids content, we ApoE−/−CD163+/+ mice (Figure 2G). analyzed the importance of CD163 to control macrophage Inflammation has a great impact on the vulnerability of functions. To this end, peritoneal macrophages from both atherosclerotic plaques.28 CD163-deficient atherosclerotic ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice were incu- lesions showed a 96% increase in the CD68 positive area bated with IL-10 (4 ng/mL; 48 hours), the best interleukin to in the aortic sinus (Figure 2C) and 74% increase in CD68 produce CD163-expressing macrophages31 (Figure 3A). positive cells in BCA cross-section analysis (Figure 2H). No After that, macrophages were incubated with fluores- changes were observed in the levels of smooth muscle actin cently labeled Dil-ox-LDL (10 μg/mL; 2-6 hours) and as- (Figure 2C,G, Figure S2). Monocytes infiltrate atheroscle- sessed lipid content by fluorescence incorporation. We found rotic lesions in response to pro-inflammatory cytokines such that a higher percentage of ApoE−/−CD163−/− macrophages as CCL2 and CCL5. CD163-deficient aortas showed an in- stimulated with Dil-ox-LDL accumulate more lipids than crease in CCL2 and CCL5 mRNA expression (Figure 2D), ApoE−/−CD163+/+ macrophages (Figure 3B), suggesting indicating that the presence of CD163 modulates monocyte that CD163 deficiency increases foam cell formation. The infiltration. No changes in vascular permeability were ob- increased lipid accumulation in ApoE−/−CD163−/− macro- served between ApoE-/-CD163+/+ and ApoE-/-CD163−/− mice phages could be an outcome of either increased uptake of (Figure S3). modified LDL or decreased cholesterol efflux. To analyze A continuum of pro- and anti-inflammatory macrophages, these possibilities, we performed uptake assays by incubating with extreme polarized phenotypes, can be found in athero- peritoneal macrophages isolated from ApoE−/−CD163+/+ and sclerotic lesions.29 Since CD163 is considered a marker of ApoE−/−CD163−/− mice with Dil-ox-LDL and quantified the M2-type macrophages, we tested whether CD163 deficiency uptake by flow cytometry. There was more uptake of Dil-ox- can influence macrophage polarization. No changes were LDL in CD163-deficient than in CD163-expressing macro- observed in the mRNA expression of Nos2-M1 or Arg1-M2 phages (Figure 3C). In addition, no differences were observed genetic markers in the aorta from ApoE−/−CD163+/+ or in cholesterol efflux to ApoB-depleted serum in macrophages ApoE−/−CD163−/− mice (Figure S4). under basal conditions or stimulated with LXR agonist GUTIÉRREZ-MUÑOZ et al.     | 14969 T0901217 (Figure 3D). In peritoneal macrophages, we ana- compared to ApoE−/−CD163+/+ macrophages (Figure 3E,F), lyzed the expression of SR-A and CD36 scavenger receptors, while SR-A, ABCG1, and ABCA1 mRNA expression was and ATP-binding cassette transporters ABCG1 and ABCA1, similar between both types of macrophages (Figure 3E). the main transporters responsible for cholesterol efflux in Preincubation with anti-CD36 inhibited Dil-ox-LDL uptake macrophages. We observed an increase in CD36 mRNA by ApoE−/−CD163−/− macrophages (Figure 3G). Collectively, and protein expression in ApoE−/−CD163−/− macrophages these results demonstrate that loss of CD163 increase the 14970 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 4 TWEAK increases atherosclerotic lesion size in CD163 deficient mice. A, sTWEAK serum concentrations in ApoE−/−CD163+/+ (n = 10) or ApoE−/−CD163−/− (n = 10) mice treated with recombinant TWEAK (rTWEAK; 5 μg/kg/three times per week) and ApoE−/−CD163−/− (N = 8) treated with recombinant CD163 (rCD; 50 μg/kg/three times per week) 2 hours before rTWEAK administration. *P < .01 vs ApoE−/−CD163+/+ + rTW; †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. B, Representative Oil Red O/ Hematoxylin and Sirius Red staining in ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice with the different treatments at the end of the study. Scale bars, 100 μm. C, Quantification of lesion area, (D) Oil Red O and Sirius Red staining in the aortic root of ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW and rCD. Values shown are mean ± SEM of 8-10 animals per group. *P < .01 vs ApoE−/−CD163+/+ + rTW. † P < .01 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. E, Representative photographs of Movat's pentachrome sections, Oil Red O/Hematoxylin and Sirius Red staining in the brachiocephalic artery from ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW and rCD. Scale bars, 200 μm. F, Quantification of lesion area and % stenosis in the BCA of ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW and rCD. Values shown are mean ± SEM of eight animals per group. *P < .01 vs ApoE−/−CD163+/+; †P < .001 vs ApoE−/−CD163−/− + rTW, One-way ANOVA with Bonferroni's posttest. G, Oil Red O and Sirius Red staining in the BCA of ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW and rCD. Values shown are mean ± SEM of 8 animals per group. *P < .01 vs ApoE−/−CD163+/+ + rTW; One-way ANOVA with Bonferroni's posttest CD36 expression and modified ox-LDL uptake promoting treated with rTW (Figure S4). In addition, no changes were macrophage foam cell formation. observed in the levels of α-SMA (Figure 5A,F, Figure S2). Finally, BCA from ApoE−/−CD163−/− mice treated with rTW exhibited an increased necrotic core size (Figure 5D), serum 3.3 | TWEAK aggravates atherosclerosis LDH activity (Figure 5E), and a higher frequency of buried progression in CD163-deficient mice caps and medial erosion (Figure 5F) than atherosclerotic plaques from ApoE−/−CD163+/+ mice. Next, we investigated the potential role of CD163 as a scav- In a second approach, we analyzed whether the systemic sol- enger receptor in TWEAK-induced atherosclerosis.20 First, uble CD163 can abolish the proatherogenic effects of TWEAK. we administered recombinant murine TWEAK (rTW; 5 μg/ For that purpose, a third group of animals were injected with kg/three times per week) into either ApoE−/−CD163+/+ or recombinant murine soluble CD163 (rCD; 50 μg/kg/three times ApoE−/−CD163−/− mice and examined atherosclerotic lesions per week) 2 hours before rTW administration. No differences 4 weeks later. No differences were observed in body weight were observed in body weight or lipid concentrations after rCD or lipid concentrations between the different groups (Figure administration (Figure S1). ApoE−/−CD163−/− mice injected S1). ApoE−/−CD163−/− mice presented increased sTWEAK with both rCD and rTW presented decreased sTWEAK plasma plasma concentrations compared with ApoE−/−CD163+/+ concentrations compared with ApoE−/−CD163−/− mice treated mice after rTW administration (Figure 4A). ApoE−/−CD163−/− with rTW alone (Figure 4A). In ApoE−/−CD163−/− mice, mice treated with rTW showed a significant 51% increase rCD163 administration diminished 50% and 76% the plaque size plaque size at the aortic root (Figure 4B,C) and in plaque in both, aortic root and BCA, respectively (Figure 4B,C,E,F). size (62%) and in stenosis (58%) at the BCA (Figure 4E-F) In addition, rCD decreased ORO content (Figure 4D,G) and compared with their littermate mice. ApoE−/−CD163−/− mice showed a 58% reduction in macrophage content in the aortic treated with rTW also exhibited increased ORO and reduced root (Figure 5A) and 64% reduction in CD68+ cells in BCA collagen content in atherosclerotic plaques present in both (Figure 5G), as well as aortic CCL2 and CCL5 mRNA ex- the aortic root (Figure 4B,D) and BCA (Figure 4E-G) com- pression (Figure 5C), compared with ApoE−/−CD163−/− mice pared with infused ApoE−/−CD163+/+ mice. treated with rTW alone. No changes were observed in the levels When plaque progression was analyzed, ≈17% plaques in of α-SMA (Figure 5A,F, Figure S2). When plaque progression ApoE−/−CD163+/+ mice treated with rTW were early lesions was analyzed, in contrast to the effect observed in rTW-in- (grade I) while ≈33% plaques were advanced plaques (grade jected ApoE−/−CD163−/− mice, in rCD and rTW-injected IV). In contrast, the percentage of advanced lesion was 46% ApoE−/−CD163−/− mice, ≈40% plaques were early lesions in ApoE−/−CD163−/− mice treated with rTW while only ≈8% (grade I) while ≈20% plaques were advanced plaques (grade plaques were early lesions (Figure 5B). ApoE−/−CD163−/− IV). In addition, BCA from rCD-injected ApoE−/−CD163−/− mice treated with rTW showed a 54% increase in macro- mice exhibit a decreased necrotic core size (Figure 5D), less phage content in the aortic root (Figure 5A) and 51% increase serum LDH activity (Figure 5E) and a lesser frequency of bur- in CD68+ cells in BCA (Figure 5G), and aortic CCL2 and ied caps and medial erosion (Figure 5F) than atherosclerotic CCL5 mRNA expression (Figure 5C), compared with plaques from rTW-treated ApoE−/−CD163−/− mice. ApoE−/−CD163+/+ mice. No changes were observed in the Overall, these in vivo experiments could indicate that de- mRNA expression of Nos2-M1 or Arg1-M2 genetic markers in ficiency of CD163 increases TWEAK availability enhancing the aorta from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice its proatherogenic effects. GUTIÉRREZ-MUÑOZ et al.     | 14971 3.4 | CD163 inhibits TWEAK-induced development, VSMCs and macrophages were studied in inflammatory responses in VSMCs and vitro. TWEAK regulates several pro-inflammatory mole- macrophage migration cules in VSMCs through nuclear factor kappa B activation.32 We have shown herein that addition of rCD163 to TWEAK- To determine how CD163 regulates the effect of TWEAK- stimulated cultured VSMCs can inhibit the biological effects induced inflammation during atherosclerotic plaque of TWEAK on NF-kB signaling, (Figure 6A). In addition, 14972 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 5 TWEAK enhances plaque instability in CD163 deficient mice. A, Representative photographs of α-SMA + DAPI and CD68 stained sections. Quantification of CD68 and α-SMA in aortic root from ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW plus rCD in the right panel. Values shown are mean ± SEM of 8-9 animals per group. *P < .05 vs ApoE−/−CD163+/+; †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. Scale bar, 100 μm. B, Percentage of each grade from Stary method between different groups. N = 20-25 plaques per group. C, Relative CCL2 and CCL5 mRNA expression levels normalized to 18S rRNA of aortas from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice treated with rTW alone or rTW plus rCD. Data represent the mean ± SEM of six animals per group; *P < .001 vs ApoE−/−CD163+/+ + rTW. †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. D, Representative images of BCA cross sections stained with MOVAT from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice treated with rTW alone or rTW plus rCD. Dashed lines show the boundary of the developing necrotic area (NC). Quantification of the necrotic core area is shown in the below panel. Data represent the mean ± SEM of eight animals per group; *P < .05 vs ApoE−/−CD163+/+; †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. Scale bar, 100 μm. E, LDH activity in serum from ApoE−/−CD163+/+ and ApoE−/−CD163−/− mice treated with rTW alone or rTW and rCD. Data represent the mean ± SEM of six animals per group; *P < .001, vs ApoE−/−CD163+/+ + rTW. †P < .001 vs ApoE−/−CD163−/− + rTW. One-way ANOVA with Bonferroni's posttest. F, Frequency of morphological markers of plaque instability in ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice treated with rTW alone or rTW plus rCD. Data represent the mean ± SEM of eight animals per group; *P < .05 vs ApoE−/−CD163+/+ + rTW. †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. G, Representative images of BCA cross sections stained with α-SMA or CD68 from ApoE−/−CD163+/+ or ApoE−/−CD163−/− mice treated with rTW alone or rTW plus rCD. Quantification CD68 positive cells is shown in the below panel. Data represent the mean ± SEM of eight animals per group; *P < .05 vs ApoE−/−CD163+/+ + rTW; †P < .001 vs ApoE−/−CD163−/− + rTW; One-way ANOVA with Bonferroni's posttest. Scale bar, 100 μm rTW increased CCL2, CCL5, ICAM-1, and MMP-9 mRNA by expressing high levels of CD163,5,35 a protein expressed expression and CCL2 and CCL5 secretion in VSMCs, which in human coronary, aortic and femoral lesions.6,10,36 A prin- was abolished when increased concentrations of rCD163 cipal function of CD163 in atherosclerotic plaques might be were present in the culture media (Figure 6B,C). to remove Hp-Hb complexes in intralesional hemorrhage. In The inflammatory environment mediated by VSMCs fact, hemoglobin uptake by CD163-expressing macrophages during atheroma plaques formation is also responsible for the leads to a distinct macrophage phenotype termed M(Hb) recruitment of the immune cells, especially monocyte/mac- or Mhem.37 M(Hb) macrophages are abundant in areas of rophages. As we have observed that the percentage of mac- neoangiogenesis and hemorrhage characterized by reduced rophages in plaques also depends on the presence of CD163, cytokine production and a lack of lipid retention.11,37,38 For we analyzed the macrophages migratory capacity upon dif- those reasons, M(Hb) macrophages have been considered ferent inflammatory scenarios. VSMCs were stimulated with atheroprotective. However, it has been recently reported that rTW in the absence or presence of increased amount of rCD M(Hb) macrophages promote angiogenesis and vascular per- for 24 hours. After that, the supernatants were collected and meability accompanied by inflammation in atherosclerosis.11 incubated with RAW 264.7 cell for 4 hours and cell migration Although Hb is a key inductor of CD163 expression, stimula- was tested using an in vitro transwell migration assay. As ex- tion of macrophages with other molecules such as pro- and pected, supernatant of rTW-stimulated VSMCs significantly anti-inflammatory cytokines can also induce the polariza- increased macrophage migration compared with medium tion to M2-type macrophages and modulates CD163 expres- alone (Figure 6D). In addition, RAW migration was inhibited sion in absence of Hb. In fact, INF-γ or TNF-α decrease and in a dose-dependent manner upon incubation with superna- IL-6 or IL-10 increase CD163 expression in macrophages.31 tant obtained from VSMCs treated with both, rTW and rCD It is important to note that contrary to human haptoglobin (Figure 6D). Overall, these data indicate that CD163 is able (Hp), mouse Hp does not promote high-affinity binding to to block the pro-inflammatory response induced by TWEAK CD163. Nevertheless, mouse hemoglobin (Hb) has a higher in VSMCs, decreasing macrophages recruitment. CD163 affinity than human Hb. The overall clearance of Hb is slower in CD163-deficient mice, and its accumulated in the liver. However, the ratio of mHb/mHpHb in serum is similar 4 | D IS C U S S ION in CD163+/+ and CD163−/− mice.39 In vitro experiments have demonstrated that CD163- Different populations of macrophages reside in human athero- expressing macrophages can also bind and internalize sclerotic plaques and microenvironment within plaques plays TWEAK.12,40 Now, we used CD163-deficient mice to a pivotal role in the differentiation program of macrophages.33 demonstrate an important role of CD163 in the setting M2-type macrophages promote the resolution of inflamma- of atherosclerotic plaque development and progression. tion and tissue repair. In fact, the conversion of Ly6Chigh We have observed that the loss of CD163 increases sol- monocytes in M2-type macrophages drives atherosclerosis uble TWEAK concentrations in ApoE-deficient mice. regression in mice.34 These macrophages are characterized According to the pro-inflammatory and proatherogenic GUTIÉRREZ-MUÑOZ et al.     | 14973 actions of TWEAK,20-22 CD163-deficient mice showed an a larger necrotic core and an increase in lipid content, mac- increase of plaque size in BCA. Clinically, lesion compo- rophage infiltration and pro-inflammatory cytokines ex- sition rather than size or degree of stenosis of the lesion pression in atherosclerotic lesions. Opposite to our results, determines the likehood of plaque rupture and subsequent it has been recently reported that ApoE-deficient mice thrombotic complications.41 CD163-deficient mice showed showed an increase in BCA plaque area compared with F I G U R E 6 CD163 prevents pro-inflammatory actions induced by TWEAK. A, Representative western blot analysis of p-p65 and p65 in VSMCs treated with 0% FBS (control), rTW (100 ng/mL) or rTW plus rCD (0.01-0.5 μg/mL). Right panel show the quantification of band densitometry values of p-p65 protein levels expressed in arbitrary units after correction for p65 (loading control). *P < .01 vs Control; †P < .01 vs rTW; Student's t test. B, Relative CCL2, CCL5, ICAM-1, MMP-2, and MMP-9 mRNA expression levels normalized to 18S rRNA of VSMCs treated with rTW (100 ng/mL) or rTW plus rCD (0.01-0.5 μg/mL) during 24 hours. The same cDNA was repeatedly tested for multiple genes. Data represent the mean ± SEM of three independent experiments. *P < .05 vs Control; †P < .05 vs rTW. Student's t test. C, Effect of CD163 on CCL2 and CCL5 secretion induced by TWEAK in VSMCs. VSMCs were incubated with rTW (100 ng/mL) or rTW plus rCD (0.01-0.5 μg/mL) during 24 hours and supernatants were tested by ELISA for CCL2 or CCL5. Data represent the mean ± SEM of six independent experiments; *P < .001 vs control; †P < .001 vs rTW; Student's t test. D, Peritoneal macrophages were seeded in the upper surface of chemotaxis chambers and stimulated with the supernatants of VSMCs treated with 0% FBS (control), rTWEAK (100 ng/mL) or rTW plus rCD (0.1-0.5 μg/mL). Quantification of migrated cells in 10 fields per condition is shown in the right panel. Data represent the mean ± SEM of four independent experiments; *P < .001 vs Control; †P < .001 vs rTW; Student's t test. Scale bars, 50 μm 14974 |    GUTIÉRREZ-MUÑOZ et al. F I G U R E 7 Model illustrating the potential mechanism of CD163. Cartoon depicting the role of CD163 as a receptor for TWEAK. In normal conditions, M2-type macrophages express CD163 and compete with Fn14 to bind TWEAK, limiting the pro-inflammatory effects of TWEAK/ Fn14 interaction. However, absence of CD163 in macrophages increases TWEAK availability, favoring its interaction with Fn14, and promoting atherosclerosis progression ApoE/CD163 double-deficient mice.11 However, the mice The underlying protective role of CD163 could be related used in the mentioned study are elderly (1-year old) and with at least two different mechanisms. First, we show that intraplaque hemorrhage is present in their BCA plaques. CD163 deficiency promotes foam cell formation through In this context, it is conceivable that Hb and TWEAK increased uptake of modified lipoproteins. The uptake of could compete to bind CD163 as demonstrated in vitro,12 ox-LDL occurs via different receptors such as CD36 and and that the M(Hb) macrophages display a pro-inflamma- SR-A.44 We demonstrated that CD36 is upregulated in ab- tory phenotype due to an increase in TWEAK availability. sence of CD163, without modifying cholesterol efflux and However, CD163 is also expressed in early lesions from the levels of the cellular transporters involved in this pro- mice and humans in which intraplaque hemorrhage is ab- cess. Accordingly, it has been previously shown that when sent11,42 and its role in the progression of those lesions have macrophages are polarized to an alternative phenotype, they not been previously assessed. lose their ability to retain lipids inside.37 In fact, CD163- CD163 and TWEAK colocalize in both human carotid expressing macrophages differentiated with IL-4 showed a and femoral atherosclerotic plaques, suggesting a potential diminished lipid uptake and differentiation of macrophages in vivo interaction.36,40 Accordingly, CD163 can interact with Hb:Hp totally abolished foam cell formation.37 with TWEAK to regulate tissue regeneration after ischemic Second, Fn14 is upregulated by ox-LDL or pro-inflam- injury.43 Now, our study demonstrates that CD163 could matory cytokines in VSMCs20,45 and, under those conditions, prevent the deleterious effects of TWEAK during the progres- TWEAK activates NF-kB signaling and different cytokines ex- sion of atherosclerosis. ApoE−/−CD163−/− mice injected with pression.15 Now, we demonstrated that CD163 can inhibit the TWEAK showed a marked increase in the progression of ath- pro-inflammatory response of TWEAK in VSMCs, indicating erosclerosis in both the aortic root and BCA by augmenting that CD163 and Fn14 compete to bind TWEAK. In addition, lesion size, lipid accumulation, and inflammation compared recombinant CD163 diminished migration of macrophages in- with ApoE−/−CD163+/+ mice. In addition, TWEAK-injected duced by supernatants from VSMCs stimulated with TWEAK. ApoE−/−CD163−/− mice showed an increase in plaque in- In conclusion, loss of CD163 in mice results in an in- stability represented by a larger necrotic core and a higher creased circulating TWEAK concentration that can acceler- presence of buried caps and medial erosion in BCA than ate atherosclerotic plaque progression through the increase ApoE−/−CD163+/+ mice. When ApoE−/−CD163−/− mice were in lipid content, inflammation, and plaque size (Figure 7). pretreated with recombinant CD163, all the proatherogenic In addition, treatment of CD163-deficient mice with re- effects of TWEAK were inhibited indicating that CD163 can combinant CD163 can ameliorate the pro-inflammatory block TWEAK signaling in vivo. and proatherogenic effects of TWEAK. Our results reveal GUTIÉRREZ-MUÑOZ et al.     | 14975 that CD163-expressing macrophages have a protective role 10. Levy AP, Purushothaman KR, Levy NS, et al. Downregulation during the progression of atherosclerosis. of the hemoglobin scavenger receptor in individuals with dia- betes and the Hp 2–2 genotype: implications for the response to intraplaque hemorrhage and plaque vulnerability. Circ Res. ACKNOWLEDGMENTS 2007;101(1):106-110. This work was supported by Fondo de Investigaciones 11. Guo L, Akahori H, Harari E, et al. CD163+ macrophages promote Sanitarias, Instituto de Salud Carlos III (ISCiii/FEDER angiogenesis and vascular permeability accompanied by inflam- PI16/01419, PI19/00128) and Spanish Biomedical Research mation in atherosclerosis. J Clin Invest. 2018;128(3):1106-1124. Centre in Diabetes and Associated Metabolic Disorders 12. Bover LC, Cardó-Vila M, Kuniyasu A, et al. A previously un- (CIBERDEM) and Cardiovascular Disease (CIBERCV), recognized protein-protein interaction between TWEAK and CD163: potential biological implications. J Immunol. Spain. 2007;178(12):8183-8194. 13. Zheng TS, Burkly LC. No end in site: TWEAK/Fn14 activation CONFLICT OF INTEREST and autoimmunity associated- end-organ pathologies. J Leukoc The authors declare that they have no conflict of interest. Biol. 2008;84(2):338-347. 14. Pamukcu B, Lip GYH, Shantsila E. The nuclear factor– AUTHOR CONTRIBUTIONS kappa B pathway in atherosclerosis: a potential therapeutic The study was conceived by J. Egido, J.L. Martín-Ventura, target for atherothrombotic vascular disease. Thromb Res. and L.M. Blanco-Colio. C. Gutiérrez-Muñoz, N. Méndez- 2011;128(2):117-123. 15. Winkles JA. The TWEAK-Fn14 cytokine-receptor axis: discov- Barbero, P. Quesada, C. Sastre, and V. Fernández-Laso ery, biology and therapeutic targeting. Nat Rev Drug Discov. performed the experiments. J.C. Escolá-Gil performed cho- 2008;7(5):411-425. lesterol efflux experiments. P. Svendsen and S.K. Moestrup 16. Bossen C, Ingold K, Tardivel A, et al. Interactions of tumor necro- generated CD163 deficient mice. C. Gutiérrez-Muñoz, N. sis factor (TNF) and TNF receptor family members in the mouse Méndez-Barbero, and L.M. Blanco-Colio analyzed the ob- and human. J Biol Chem. 2006;281(20):13964-13971. tained data. L.M. Blanco-Colio wrote the manuscript with 17. Harada N, Nakayama M, Nakano H, Fukuchi Y, Yagita H, contributions of N. Méndez-Barbero, J.L. Martín-Ventura, J. Okumura K. Pro-inflammatory effect of TWEAK/Fn14 interaction Egido, and S.K. Moestrup. All authors read and approved the on human umbilical vein endothelial cells. Biochem Biophys Res Commun. 2002;299(3):488-493. manuscript. 18. Lynch CN, Wang YC, Lund JK, Chen YW, Leal JA, Wiley SR. TWEAK induces angiogenesis and proliferation of endothelial R E F E R E NC E S cells. J Biol Chem. 1999;274(13):8455-8459. 1. Swirski FK, Nahrendorf M. Leukocyte behavior in athero- 19. Munoz-Garcia B, Madrigal-Matute J, Moreno JA, et al. sclerosis, myocardial infarction, and heart failure. Science. TWEAK-Fn14 interaction enhances plasminogen activator in- 2013;339(6116):161-166. hibitor 1 and tissue factor expression in atherosclerotic plaques 2. Gordon S, Martinez FO. Alternative activation of macrophages: and in cultured vascular smooth muscle cells. Cardiovasc Res. mechanism and functions. Immunity. 2010;32(5):593-604. 2011;89(1):225-233. 3. Gordon S. Alternative activation of macrophages. Nat Rev 20. Munoz-Garcia B, Moreno JA, Lopez-Franco O, et al. Tumor Immunol. 2003;3(1):23-35. necrosis factor-like weak inducer of apoptosis (TWEAK) 4. Pulford K, Micklem K, McCarthy S, Cordell J, Jones M, Mason enhances vascular and renal damage induced by hyperlipidemic DY. A monocyte/macrophage antigen recognized by the four an- diet in ApoE-knockout mice. Arterioscler Thromb Vasc Biol. tibodies GHI/61, Ber-MAC3, Ki-M8 and SM4. Immunology. 2009;29(12):2061-2068. 1992;75(4):588-595. 21. Schapira K, Burkly LC, Zheng TS, et al. Fn14-Fc fusion pro- 5. Kristiansen M, Graversen JH, Jacobsen C, et al. Identification of the tein regulates atherosclerosis in ApoE-/- mice and inhibits mac- haemoglobin scavenger receptor. Nature. 2001;409(6817):198-201. rophage lipid uptake in vitro. Arterioscler Thromb Vasc Biol. 6. Schaer CA, Schoedon G, Imhof A, Kurrer MO, Schaer DJ. 2009;29(12):2021-2027. Constitutive endocytosis of CD163 mediates hemoglobin-heme 22. Sastre C, Fernández-Laso V, Madrigal-Matute J, et al. Genetic de- uptake and determines the noninflammatory and protective tran- letion or TWEAK blocking antibody administration reduce athero- scriptional response of macrophages to hemoglobin. Circ Res. sclerosis and enhance plaque stability in mice. J Cell Mol Med. 2006;99(9):943-950. 2014;18(4):721-734. 7. Etzerodt A, Maniecki MB, Møller K, Møller HJ, Moestrup SK. 23. Stary HC, Chandler AB, Glagov S, et al. A definition of initial, Tumor necrosis factor α-converting enzyme (TACE/ADAM17) fatty streak, and intermediate lesions of atherosclerosis. A re- mediates ectodomain shedding of the scavenger receptor CD163. port from the Committee on Vascular Lesions of the Council J Leukoc Biol. 2010;88(6):1201-1205. on Arteriosclerosis, American Heart Association. Circulation. 8. Timmermann M, Högger P. Oxidative stress and 8-iso-prostaglandin 1994;89(5):2462-2478. F(2alpha) induce ectodomain shedding of CD163 and release of 24. Tarín C, Fernández-Laso V, Sastre C, et al. Tumor necrosis tumor necrosis factor-alpha from human monocytes. Free Radic factor-like weak inducer of apoptosis or Fn14 deficiency reduce Biol Med. 2005;39(1):98-107. elastase perfusion-induced aortic abdominal aneurysm in mice. 9. Møller HJ. Soluble CD163. Scand J Clin Lab Invest. 2012;72(1):1-13. J Am Heart Assoc. 2014;3(4):e000723. 14976 |    GUTIÉRREZ-MUÑOZ et al. 25. Escola-Gil JC, Chen X, Julve J, et al. Hepatic lipase- and endo- 38. Boyle JJ, Johns M, Kampfer T, et al. Activating transcription thelial lipase-deficiency in mice promotes macrophage-to-feces factor 1 directs Mhem atheroprotective macrophages through RCT and HDL antioxidant properties. Biochim Biophys Acta. coordinated iron handling and foam cell protection. Circ Res. 2013;1831(4):691-697. 2012;110(1):20-33. 26. Rekhter MD. How to evaluate plaque vulnerability in animal mod- 39. Etzerodt A, Kjolby M, Nielsen MJ, Maniecki M, Svendsen P, els of atherosclerosis? Cardiovasc Res. 2002;54(1):36-41. Moestrup SK. Plasma clearance of hemoglobin and haptoglobin 27. Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, in mice and effect of CD163 gene targeting disruption. Antioxid Schwartz SM. Advanced atherosclerotic lesions in the innominate Redox Signal. 2013;18(17):2254-2263. artery of the ApoE knockout mouse. Arterioscler Thromb Vasc 40. Moreno JA, Munoz-Garcia B, Martín-Ventura JL, et al. The CD163- Biol. 2000;20(12):2587-2592. expressing macrophages recognize and internalize TWEAK: 28. Stoll G, Bendszus M. Inflammation and atherosclerosis: novel potential consequences in atherosclerosis. Atherosclerosis. insights into plaque formation and destabilization. Stroke. 2009;207(1):103-110. 2006;37(7):1923-1932. 41. Libby P, Theroux P. Pathophysiology of coronary artery disease. 29. Leitinger N, Schulman IG. Phenotypic polarization of mac- Circulation. 2005;111(25):3481-3488. rophages in atherosclerosis. Arterioscler Thromb Vasc Biol. 42. Tarín C, Carril M, Martin-Ventura JL, et al. Targeted gold-coated 2013;33(6):1120-1126. iron oxide nanoparticles for CD163 detection in atherosclerosis by 30. Glass CK, Witztum JL. Atherosclerosis. The road ahead. Cell. MRI. Sci Rep. 2015;5:17135. 2001;104(4):503-516. 43. Akahori H, Karmali V, Polavarapu R, et al. CD163 interacts with 31. Buechler C, Ritter M, Orsó E, Langmann T, Klucken J, Schmitz TWEAK to regulate tissue regeneration after ischaemic injury. Nat G. Regulation of scavenger receptor CD163 expression in human Commun. 2015;6:7792. monocytes and macrophages by pro- and antiinflammatory stimuli. 44. Greaves DR, Gordon S. The macrophage scavenger receptor at 30 J Leukoc Biol. 2000;67(1):97-103. years of age: current knowledge and future challenges. J Lipid Res. 32. Blanco-Colio LM. TWEAK/Fn14 axis: a promising target for the 2009;50(Suppl):S282-S286. treatment of cardiovascular diseases. Front Immunol. 2014;5:3. 45. Munoz-Garcia B, Martin-Ventura JL, Martínez E, et al. Fn14 is up- 33. Bouhlel MA, Derudas B, Rigamonti E, et al. PPARgamma regulated in cytokine-stimulated vascular smooth muscle cells and activation primes human monocytes into alternative M2 macro- is expressed in human carotid atherosclerotic plaques: modulation phages with anti-inflammatory properties. Cell Metab. 2007;6(2): by atorvastatin. Stroke. 2006;37(8):2044-2053. 137-143. 34. Rahman K, Vengrenyuk Y, Ramsey SA, et al. Inflammatory SUPPORTING INFORMATION Ly6Chi monocytes and their conversion to M2 macrophages drive atherosclerosis regression. J Clin Invest. 2017;127(8): Additional Supporting Information may be found online in 2904-2915. the Supporting Information section. 35. Møller HJ, Nielsen MJ, Maniecki MB, Madsen M, Moestrup SK. Soluble macrophage-derived CD163: a homogenous ectodomain protein with a dissociable haptoglobin-hemoglobin binding. How to cite this article: Gutiérrez-Muñoz C, Immunobiology. 2010;215(5):406-412. Méndez-Barbero N, Svendsen P, et al. CD163 36. Moreno JA, Dejouvencel T, Labreuche J, et al. Peripheral artery deficiency increases foam cell formation and plaque disease is associated with a high CD163/TWEAK plasma ratio. progression in atherosclerotic mice. The FASEB Arterioscler Thromb Vasc Biol. 2010;30(6):1253-1262. Journal. 2020;34:14960–14976. https://doi. 37. Finn AV, Nakano M, Polavarapu R, et al. Hemoglobin directs mac- org/10.1096/fj.20200​0177R rophage differentiation and prevents foam cell formation in human atherosclerotic plaques. J Am Coll Cardiol. 2012;59(2):166-177.