Macrophage angiotensin-converting enzyme reduces atherosclerosis by increasing peroxisome proliferator-activated receptor α and fundamentally changing lipid metabolism

angiotensin converting enzyme, PPARα, atherosclerosis, macrophages, lipid metabolism

Time of primary review: 26 days

1. Introduction

Atherosclerosis is a complex process in which lipid deposition in arteries promotes an inflammatory response that itself contributes to lesion progression and ultimately lesion rupture. While atherosclerosis is associated with older age, the entry of lipids into blood vessel walls and the removal of this lipid by macrophages happens throughout life. Indeed, the failure of lipid removal and the formation of macrophage foam cells are major pathologic features of atherosclerotic lesions.¹ While lipid lowering is a standard therapeutic approach for dyslipidemia and preventing atherosclerosis, one may envision an approach to atherosclerosis where the capacity of macrophages to mobilize and metabolize vascular lipids is increased.²

Angiotensin-converting enzyme (ACE) is best known for its effects on blood pressure since it produces the potent vasoconstrictor angiotensin II, and ACE inhibitors (ACEi) are widely used in treating hypertension and other cardiovascular diseases.³ While ACE and the renin-angiotensin system are important in blood pressure control, recent studies have demonstrated that ACE also plays a regulatory role in macrophage function.^4–6 This was observed in a mouse line called ACE 10/10 in which a genetic approach was used to elevate the expression of ACE in monocytes and macrophages.^7,8 Such mice have an enhanced immune response against models of infection, tumour, and chronic diseases such as Alzheimer’s disease.^9,10 In contrast, the lack of ACE activity induces myeloid dysfunction in mice and probably also in humans.^5,11 In human atherosclerosis, lesional macrophages produce notable amounts of ACE but the function of this enzyme in atherosclerosis-associated inflammation is not well understood.^12,13 Here, using PCSK9-induced atherosclerosis, we find much less vascular disease in ACE 10/10 vs. WT mice. A key finding is that macrophages with increased ACE activity have elevated expression of the transcription factor peroxisome proliferator-activated receptor α (PPARα). This transcription factor is known to be a major regulator of lipid metabolism and increased PPARα in ACE 10/10 macrophages is associated with increased macrophage lipid uptake, increased lipid catabolism via β-oxidation, increased cell ATP, and increased cholesterol efflux. Reduced atherosclerosis in ACE 10/10 mice underscores the absolutely central role of macrophages in atherosclerosis and that orchestrating a better macrophage response to lipids is a potentially powerful strategy for reducing atherosclerotic disease. Increasing macrophage ACE is one way to enhance macrophage function. Further, the discovery here that macrophage ACE increases the effectiveness of macrophage lipid metabolism has implications for the treatment of cardiovascular disease with angiotensin II receptor antagonists (ARBs) in place of ACEi.

2. Methods

2.1 Cell lines

THP-1 cells (human monocytic cell line) were obtained from ATCC and cultured in RPMI-1640 (Corning) supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (Omega), 10 U/mL penicillin and 100 μg/mL streptomycin (Gemini Bio-Products). To create a human ACE-THP-1 cell line, THP-1 cells were stably transfected with human ACE in the SBI CD710B-1 vector having an MSCV promoter and are now termed THP-1^ACE. Cells were also prepared stably transfected with an empty vector (THP-1^Vector). Cells were selected with 1 μg/mL puromycin (Sigma) in a complete RPMI-1640 medium. HL-60 cells (human lymphoblast-like cell line) were obtained from ATCC and cultured in Iscove’s Modified Dulbecco’s Medium (ATCC) with 15% (vol/vol) FBS at 37°C in 5% CO₂.

2.2 Primary cell cultures

Mice were intraperitoneally injected with thioglycolate 4 days before sacrifice to induce peritoneal macrophages (TPM).¹⁴ Macrophages were isolated following lavage of the peritoneum with medium B (HBSS containing penicillin (100 units), streptomycin (100 μg/mL), 10 mm HEPES, pH 7.4, BSA (0.06%), and NaHCO3 (0.0375%)).¹⁴ Macrophages were cultured in media C (RPM1 1640 without L-glutamine, containing 10% FBS, penicillin/streptomycin (1 U and 1 ug/mL, Gemini Bio-Products), sodium pyruvate (0.5 mM), HEPES (10 mM), 2-mercaptoethanol 50 μM), and 1 ng/mL of M-CSF (Peprotech) for 4 h. After 4 h, media was aspirated, and the plate was washed three times with PBS. We then added a new modified medium B with 1% fatty acid-free-BSA at 37°C without glucose and FBS. For some experiments, oleic acid (OA, 200 μM) was added to the cultured cells at time 0. After 48 h, cells were harvested and analysed. For ox-LDL treatment, cells were treated for 12 h with 30 μg/mL of either DiI-ox-LDL (Thermofisher L34358) or ox-LDL (Thermofisher L34357) derived from human plasma.

3. Animal experiments

All mice were described previously.¹⁴ For the in vivo atherosclerosis experiments, 5 × 10¹⁰ vector genome copies (VG) of AAV-PCSK-9 or null-AAV were used in 6-week-old WT and ACE 10/10 mice on a C57BL6/J background.^15–17 The mice were then fed with an atherogenic rodent diet (Research Diets, no. D12336i) for 18 weeks unless otherwise indicated. Systolic blood pressure was measured every 2 weeks in trained conscious mice using a tail-cuff apparatus (Visitech BP2000 system; Visitech Systems Inc., Apex, NC) as previously described.¹⁸ Blood pressure was measured 20 times for each time point. Food intake was measured using Phenomaster cages (TSE system).

For ACEi and ARB experiments, mice were treated with the atherosclerotic diet for two extra weeks during which they received either ramipril (36.3 mg/L) or losartan (600 mg/L) in water. In some experiments, mice were injected with thioglycolate 4 days before sacrifice to induce peritoneal macrophages (TPM).¹⁴ All procedures in mice, including euthanasia, were approved by the Cedars-Sinai Institutional Animal Care and Usage Committee and conform to the guidelines from the NIH Guide for the Care and Use of Laboratory Animals. Mice were killed by cervical dislocation after an overdose of isoﬂurane.

3.1 Systemic lipid measurement

We followed the procedure described by Temel et al. ¹⁹ for determining the lipid content of liver tissue (detailed protocol in https://www.protocols.io/view/extraction-and-analysis-of-liver-lipids-n92ld7ool5br/v1) and faeces (https://www.protocols.io/view/measurement-of-fecal-neutral-sterol-fns-excretion-dm6gpxe8vzpn/v1). The total plasma lipid (total cholesterol, triglyceride, HDL-cholesterol, LDL/VLDL cholesterol) profiling was performed as previously described.²⁰

The macrophages cholesterol efflux capacity was performed by a fluorescent-based kit from Abcam (ab196985) and following the manual’s instruction. Briefly, macrophages were pretreated with fluorescent-labelled cholesterol for 6 h. The specific inhibitors ramipril or GW6471 were included in the experiment to neutralize the effect of ACE and PPARα. Fluorescence was measured in media using the FLUOstar Omega microplate reader (BMG Labtech). We used LDL/VLDL-depleted mouse serum (prepared as described per the Abcam kit protocol) as the cholesterol acceptor in our experiments. Total cellular efflux was calculated as the ratio between the fluorescence intensity of media and fluorescence intensity of cell lysate plus media after subtraction of non-specific efflux background and presented as %.²¹

3.2 Histological and immunofluorescence staining of aortic atherosclerotic lesions

The whole length of the aorta was isolated and fixed in 10% buffered formalin for 12 h. Fixation specimens were rinsed for 5 min in 70% ethanol, then stained with Sudan IV for 5 min. After staining, samples were washed in 80% ethanol. The whole length of the artery was captured by a digital camera for analysis. For histological analysis, 8-μm-thick frozen sections of the aortic root were cut and stained with oil red O, H&E, and trichrome. QuPath (0.2.3) software was used to analyse lipid-stained lesional area.

For immunohistochemistry, aortic root cryosections were blocked by 1% BSA containing 0.1% Triton X-100 for 1 h at RT. Then sections were incubated with antibody cocktails overnight at 4°C [goat-anti-ACE (Invitrogen), mouse-anti-PPARα (Invitrogen), and APC-anti-CD206 (Biolegend)]. After primary antibody staining, the sections were washed and incubated with a secondary antibody cocktail at 4°C for 2 h, and then 5 μL Prolong Gold-DAPI was applied to the sections for 10 min. Fluorescent slides were scanned with an Axio Scan.Z1 (Zeiss).

3.3 PPRE-luciferase reporter gene assay

Activation of PPRE was measured by a luciferase assay. THP-1^Vector and THP-1^ACE cells were plated for 4 h before transfection at a density of 1 × 10⁵ cells/well in a 96-well plate using 100 µL glucose free-RPMI 1640 medium. For transfection complex preparation, we added 2.2 µg of PPRE-pNL1.3 DNA into 103 µL of Opti-MEM I Medium (Thermo Fisher). We then added 6.6 µL of FuGENE HD reagent (Promega), mixed gently by pipetting, and incubated for 10 min at RT. 5 µL of the PPRE-plasmid cocktail mix was added per well and mixed thoroughly. After a 12 h transfection, the medium was replaced with OA saturated culture medium. To make this media, OA was dissolved in ethanol at 1 mM and then added to media to make 200 μM. PPRE-pNL1.3 was a gift from Drs. Severine Degrelle & Thierry Fournier (Addgene plasmid Cat. 84394).²² After 48 h, a luciferase assay was performed according to the manufacturer’s instructions (Promega, Nano-Glo Luciferase Assay System).

3.4 Flow cytometry

Aortic cells were isolated by digestion as described previously and then washed three times with FACS buffer (PBS with 2% FBS, 0.1% sodium azide, 1 mm EDTA).²³ Cells were suspended in FACS staining buffer at a density of 0.5 × 10⁶ cells/100 μL and incubated with Fc block (Biolegend) for 10 min on ice. Then cells were stained at 4°C for 1 h with anti-mouse antibodies: Pacific Blue- CD45 (Biolegend), PE/cy7-CD11b (Biolegend), APC-CD206 (Biolegend), APC-CD86 (Biolegend), Alexa Fluor-700-F4/80 (Biolegend), Alexa Fluor-488-CD36 (Biolegend), Goat-anti-ACE (R&D), Donkey-anti-goat-PE (Thermo). Cells were also stained with PE-IL-10 (Biolegend) or FITC-c1q (Thermo). After washing three times with FACS buffer, cells were measured or sorted by flow cytometry performed with a Sony SA38000 instrument or a BDAria III Cell Sorter, and data were analysed with FlowJo version 10.4 (FlowJo, LLC, Ashland, OR). Macrophages were defined as CD45⁺CD11b⁺F4/80⁺.

3.4.1 Apoptotic and necrotic staining

An Apoptosis/Necrosis assay kit from Abcam (ab176749) was used to detect apoptotic and necrotic cells in whole aortas. After single cells were prepared from whole aortas as described above, cells were stained with Apopxin Green Indicator, 7-AAD, and CytoCalcein 450 to detect the apoptotic cell, necrotic cells, and health cells separately. After a 40-min room temperature incubation, cell fluorescence intensity was measured using a Sony SA38000 flow cytometry (Apopxin Green Ex/Em 490/525 nm; 7-AAD Ex/Em 550/650; CytoCalcein Violet 450 Ex/Em 405/450).

3.4.2 Macrophage phenotype

TPMs were cultured in a glucose-free medium containing 200 μM OA for 48 h. Cells were washed three times with PBS and then gently scraped from the plate. Cells were stained with different antibodies from Biolegend including Violet450-F4/80, PE/cy7-CD11b, APC- CD206, FITC-CD80, PE-CD163, AlexaFluor-700-CD86, and FITC-MerTK; FITC-c1q (Thermo), AlexaFluor-488-Ym-1 (Abcam), PE-Fizz-1 (Thermo), APC-Arginase-1 (Thermo). After 1 h, cells were washed and measured by flow cytometry performed with a Sony SA38000 instrument. Data were analysed with FlowJo version 10.4 (FlowJo, LLC, Ashland, OR).

3.4.3 Efferocytosis assay

Mouse macrophages were treated with CytoTell^TMBlue (Cayman) at 37°C for 30 min. The stained macrophages were plated and cultured with or without OA (200 μM)- modified medium B for 48 h without FBS and glucose. Bone marrow neutrophils (bait cells) were isolated and purified by Percol gradient.¹⁴ Neutrophils were treated with CFSE (Cayman) for 30 min and then cultured in serum-free RPMI-1640 overnight at 37°C to induce apoptosis. Bait cells were fed to the macrophages at a 4:1 ratio and, after 16 h, macrophages were washed three times with PBS to remove unbound neutrophils. THP-1 cells were pretreated with 250 nM PMA for 48 h and then stained with CytoTell^TMBlue as above. HL-60 cells were stained as above and were used as bait cells. A Sony-SA3800 flow cytometry (Sony) was used to measure efferocytosis (% macrophages taking-up bait cells). Data were analysed with FlowJo version 10.4 (FlowJo, LLC, Ashland, OR).

3.5 Western blotting

For aortic macrophages, we harvested and pooled six aortas from WT^PCSK9 and 10/10^PCSK9 mice. Tissue was digested with collagenase type I, hyaluronidase, and DNase as previously described.²³ Cells were selected with anti-mouse F4/80 MicroBeads (Miltenyi Biotec) to purify macrophages. The enriched aortic macrophages were lysed in RIPA buffer with protease and phosphatase inhibitors (Thermo), and protein concentration was measured using a Pierce BCA kit (Thermo). Then, the protein extracts were resolved by SDS–PAGE electrophoresis and transferred onto polyvinylidene difluoride (PVDF) membranes. The PVDF membranes were incubated with PPARα mouse mAb (Invitrogen, Cat. MA1-822), ACE mouse mAb (R&D, Cat. AF1513), CPT1A mouse antibody (Abcam, Cat. ab128568), CPT2 rabbit mAb (Abcam, Cat. ab231162), CPT1B rabbit antibody (Abcam, Cat. ab134988), acyl-CoA dehydrogenase long chain (ACADL) rabbit antibody (Thermo, Cat. PA5-82450), RXRa rabbit antibody (Abcam, Cat. ab125001), adipose triglyceride lipase (ATGL) rabbit antibody (CST, Cat.2138 s), β-actin mAb (Sigma Cat. A3854), ABCA1, and ABCG1 rabbit antibody (Novus, NB400-105 and NB400-132). Membranes were blocked with Li-COR blocking buffer for 1 h at room temperature and then incubated with primary antibodies and IRDye^® Secondary Antibodies (Li-COR). The blots were visualized with chemiluminescence, and the densitometry of the blots was detected using an LI-COR Odyssey Fc imager (LI-COR Biosciences, Lincoln, NE, USA). The fluorescence intensity was evaluated using Image Studio Lite Ver 5.2.

3.6 Quantitative Rt-PCR analysis

RNA was extracted from samples using DirectZol miniprep kits (Zymo Research). The purity of the RNA was assessed by absorbance at 260/280 nm using a NanoDrop spectrophotometer (Thermo Scientific). Complementary DNA was synthesized from 250 ng of RNA, which had a 260/280 ratio of >1.8, using oligo (dT) and Superscript II (Applied Biosystems). Quantitative RT-PCR was performed using a 7500 real-time PCR system (Applied Biosystems) and SYBR Green Master Mix reagents (Applied Biosystems).

3.7 Isotope tracing

TPM was isolated from atherosclerotic mice and 2 × 10⁶ cells were seeded into six wells plates in 2 mL medium containing 10% FBS for 4 h. Then, unattached cells were removed, and the cells were washed three times with PBS before adding the labelled compound to 2 mL prewarmed medium. For glucose isotope medium, we dissolved uniformly labelled ¹³C-glucose (Cambridge Isotope Laboratories) into glucose-free RPMI-1640 medium (Thermo) with fatty acid-free-BSA (Sigma), and OA to get 1% BSA, 10 mM ¹³C-glucose, 200 μM OA. To prepare media for uniformly labelled ¹³C-OA, we use normal glucose (10 mM) and 200 μM ¹³C-OA (Cambridge Isotope Laboratories). Cells were then incubated for 48 h at 37°C and 5% CO₂ in a humidified incubator. To process, media was removed, and wells were washed once with iced 0.9% NaCl. Then, 500 μL of cold MeOH was added to each well and the plate swirled to ensure coverage of all cells. The contents of the well were collected, transferred to a tube on ice, and 500 μL of cold chloroform was added to each tube. After vortexing for 5 min at 4°C, the tubes were centrifuged for 6 min at 10 000 g at 4°C. 400 ul of the upper aqueous phase containing polar metabolites including citrate was collected into GC/MS sample vials (LabConco) and dried using a SpeedVac. The tubes were then capped tightly and stored at −80°C. When taking samples from cold storage (−80°C), they were re-dried for 20–30 min.²⁴ Dried samples were resuspended in 20 µL of 2% (w/v) methoxyamine in pyridine and incubated at 37°C for 45 min followed by the addition of 20 µL of N-methyl-N-tert-butyldimethylsilyltrifluoroacetamide (MtBSTFA) + 1% TBDMSCl, silylating mixture VI, N-tert-butyldimethylsilyl-N-methyltrifluoroacetamide with 1% tert-butyldimethylchlorosilane (Sigma 00942), and incubation for another 45 min. at 37°C. Samples were analysed by mass spectrometry as previously described using Agilent MassHunter software.^25,26 For stable isotope tracing data, natural heavy isotope abundance was corrected against a reference set of unlabelled metabolite standards with FluxFix software.²⁷

3.8 Intracellular lipid droplets staining

Macrophages were seeded in 96-well plates. After 48 h of 200 μM OA treatment, culture media was gently aspirated, and cells were washed with PBS. We then added Droplite^TM red staining solution (ATT Bioquest) and incubated cells at 37°C, 5% CO2 for 30 min. To check lipid usage, cells were cultured with 200 μM OA for 48 h as described above. The OA-medium was removed and replaced with media free of OA. To check the number of cell lipid droplets at time points up to 48 h, cell fluorescence intensity was read at Ex/Em 550/640 nm. Images were taken using a fluorescence microscope with a TRITC filter.

3.9 Lipolysis

To test macrophages lipolysis efficiency, we used a lipolysis kit (BioVision) to evaluate cell glycerol content. Macrophages were seeded on 96 well plates and incubated with OA for 48 h. Cells were gently washed two times with 100 µL of lipolysis wash buffer. This was then removed and replaced with 150 µL lipolysis assay buffer containing 1.5 µL of 10 µM isoproterenol (final concentration 100 nM) to stimulate lipolysis for 4 h. Glycerol release was measured according to the manufacturer’s protocol at OD 570 nm.

3.10 Electron microscopy (EM)

The cellular lipid droplet content was measured by EM. Macrophages were fixed in 2.5% glutaraldehyde and 2% formaldehyde in 0.1 M sodium cacodylate at 4°C. After fixation, samples were dehydrated and embedded in epon resin (Polysciences). The ultrathin sections were stained with 1% uranyl acetate and 3% lead citrate. Ultimately, the images were taken using HT7700 transmission electron microscope (HITACHI) at 2,500x to 4,000x direct magnification.

3.11 Confocal analysis

For confocal imaging, TPM was treated with 200 μM OA for 48 h after which cells were carefully washed with PBS and fixed with 4% formaldehyde for 30 min. After rinsing, the cells were permeabilized with 0.1% Triton X-100 at room temperature for 20 min and washed three times with PBS. They were then incubated with a primary antibody cocktail (goat-anti-ACE and mouse- anti-PPARα) overnight at 4°C. Cells were washed with TBST, then treated with a secondary antibody cocktail containing anti-goat AlexaFlour 568 and anti-mouse Alexa 488 for 2 h at RT. After washing, 5 μL Prolong Gold-DAPI was applied to the sections for 10 min. The slides were scanned with a Zeiss LSM 780 confocal microscope (Zeiss).

3.12 Oxygen consumption and intracellular ATP content

3.12.1 Seahorse analysis

Atherosclerotic lesional macrophages (CD45⁺CD11b⁺F4/80⁺) were sorted by flow cytometry. The oxygen consumption rates were measured using an Agilent Seahorse XFe96 Analyzer as previously described.¹⁴ Briefly, macrophages were centrifuged onto Seahorse XF96 plates (600 × g for 5 min) coated with Cell-Tak (Corning) at 1.5 × 10⁵ cells/well. Measurements were conducted in DMEM (Sigma, catalogue no. D5030) supplemented with 5 mm HEPES, 8 mm glucose, 2 mm glutamine, and 2 mm pyruvate. Oligomycin (2 μm), FCCP (500 nm), and rotenone (200 nm) with antimycin A (1 μm) were added acutely to the wells, and respiratory parameters were calculated as described.^14,28

3.12.2 MitoXpress extra oxygen consumption assay

TPM was cultured in modified medium B (1% fatty acid-free-BSA without glucose and FBS) at 5% CO₂, 37°C. Cells were treated with different reagents for 48 h. Before the assay, media, reagents, and the plate-reader were prewarmed to 37°C. Agilent MitoXpress Xtra (MX-200-4, Agilent) reagent was reconstituted in 1 mL growth medium and then diluted in 10 mL prewarmed growth medium as previously described.²⁹ Cell growth medium was replaced in each well with 100 µL of the MitoXpress Xtra reagent in medium, then sealed by overlaying with 100 µL prewarmed HS oil (MX-200-4, Agilent). The plates were then immediately measured kinetically on a FLUORstar (BMG Labtech) plate reader (prewarmed to 37°C, Ex 380 nm, Em 650 nm) for 120 min. Kinetic data analysis was performed by performing linear regression over the linear part of the kinetic data.

3.12.3 ATP content

TPM was cultured with 200 μM OA along with etomoxir 5 μM (CPT1 inhibitor) or GW6471 10 μM (PPARα antagonist). 3 × 10⁵ cells were put in wells of a 96-well plate, and ATP was measured using the CellTiter-Glo 2.0 kit (Promega) as previously described.¹⁴

3.12.4 Acetyl-CoA concentration

5 × 10⁶ TPM were collected and deproteinized by a perchloric acid/KOH protocol (BioVision). After removing the protein, we added 10 µL of sample into duplicate wells (sample and background) of a 96-well plate and the volume was brought to 50 µL with Acetyl- CoA Assay Buffer (BioVision). Fluorescence was measured using Ex/Em = 535/587 nm.

3.13 Quantification and statistical analysis

All data are shown as the mean ± SEM. GraphPad Prism 9.0 software (GraphPad Software, San Diego, CA, USA) was used to analyse and present the data. Sample size was estimated based on previous experience, sample availability, and previously reported studies. No data were excluded from the data analysis. The normal distribution of data was determined by the Shapiro–Wilk normality test. For statistical comparisons, unpaired two-tailed t-tests were used to compare two groups; two-way ANOVA followed by Sidak’s multiple comparisons test was used to compare more than two groups). P < 0.05 was considered significant. Numbers per group in the figure legends refer to the number of mice per group.

4. Results

4.1 Less atherosclerosis in ACE 10/10^PCSK9 mice

To study atherosclerosis, 6-week-old wild type (WT) or ACE 10/10 mice were injected with PCSK9-AAV and fed an atherogenic rodent diet for 18 weeks.^16,17 Control mice received AAV lacking the PCSK9 gene. As expected, PCSK9 reduced liver expression of the LDL receptor (see Supplementary material online, Figure S1A). Both ACE 10/10 and WT mice injected with PCSK9 (termed ACE 10/10^PCSK9 and WT^PCSK9) had equivalent blood pressures, food intake, and body weights (see Supplementary material online, Figure S1B–D). Further, both groups of animals showed hypercholesterolaemia. However, total plasma cholesterol, triglycerides, HDL-cholesterol, and LDL/VLDL-cholesterol, measured in collaboration with Dr. Aldons Jake Lusis, were not significantly different between the two groups (see Supplementary material online, Figure S1E–H). In addition, liver total cholesterol, free cholesterol, and triglycerides, measured in collaboration with Dr. Ryan Temel, also showed no differences (see Supplementary material online, Figure S1I–K). Finally, faecal cholesterol was not different between 10/10^PCSK9 and WT^PCSK9 mice (see Supplementary material online, Figure S1L).

What was different was that ACE 10/10^PCSK9 mice have significantly less aortic atherosclerosis vs. WT^PCSK9 mice as measured by Sudan IV staining (25.0% vs. 44.7% of aortic area, Figure 1A and B). Aortic lipid content was also less in ACE 10/10^PCSK9 mice than in WT^PCSK9 (Figure 1C and D), consistent with previous analysis.¹⁸ Histologic examination of the aortic ring and valves also showed striking differences, with ACE 10/10^PCSK9 mice having significantly thinner valve cusps and rings, markedly diminished atherosclerosis, with fewer macrophages, cholesterol clefts, and calcium deposits, and much less collagen as compared to equivalent tissue from WT^PCSK9 mice (Figure 1E, quantitated in Supplementary material online, Table.S1). Aortas from ACE 10/10^PCSK9 mice have a lesser percentage of apoptotic cells (11.9% vs. 21.6%) and necrotic cells (2.8% vs. 4.6%) compared to WT^PCSK9 mice (Figure 1F and G).

Figure 1

Decreased atherosclerosis in ACE 10/10^PCSK9 mice. (A) Aortas from WT and ACE 10/10 mice (10/10) treated with PCSK9-AVV or control mice treated with null-AAV. Aortas were stained with Sudan IV; red indicates lipid-enriched plaque. Scale bar: 500 µm. (B) Percentage of aortic surface area staining for lipid-enriched plaque. (C and D) Histology and quantitation of aortic root sections stained with Oil Red O show decreased wall lipid in ACE 10/10^PCSK9 aortas compared to WT^PCSK9 (W: wall; L: lumen. Bar: 200 µm). (E) Aortic root sections from WT^PCSK9 mice (panels 1 & 3) vs. ACE 10/10^PCSK9 mice (2 & 4) stained with haematoxylin and eosin (panels 1 and 2) or the trichrome stain (3 and 4) (Bar: 200 µm). 10/10^PCSK9 mice have thinner valve cusps (thin arrows) and valve rings (VR) with less atherosclerotic plaque, less cholesterol crystal deposits (bold arrows), calcium deposits (*), and collagen (blue on the trichrome stain). Flow analysis of apoptotic cells (F) and necrotic cells (G) from the aortas of WT^PCSK9 and ACE 10/10^PCSK9 mice show more pathology in the tissues of WT^PCSK9 mice. Data are presented as the mean ± standard error of the mean (SEM). P-values are shown in the figure (B, D, F, and G) by a two-tailed t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). Each point is one mouse (n ≥ 8).

As expected, PCSK9-induced atherosclerosis was accompanied by increased percentages of F4/80⁺ macrophages among all CD11b⁺ cells within the aortic wall (Figure 2A, Supplementary material online, Figure S2A). Among all aortic CD11b⁺ F480⁺ cells, flow cytometry showed a higher percentage of CD206⁺ macrophages in ACE 10/10^PCSK9 mice vs. WT^PCSK9 (Figure 2B and C). CD206⁺ is often considered a reparative M2 marker and such cells in ACE 10/10^PCSK9 also expressed higher levels of the M2 markers CD36 (a scavenger receptor important in lipid uptake) and IL-10 (an anti-inflammatory cytokine) vs. equivalent WT^PCSK9 aortic cells (Figure 2D and E). To examine the role of ACE activity, some of the ACE 10/10^PCSK9 and WT^PCSK9 mice were treated with an ACEi or an ARB for two weeks in drinking water. The ACEi reduced the F4/80⁺CD206⁺ population in the aortas of both ACE 10/10^PCSK9 and WT^PCSK9 mice and eliminated expression differences of CD36 and IL-10 (Figure 2B–E). In contrast, the ARB had much less effect with differences in expression remaining. Additional flow analysis of ACE 10/10^PCSK9 mice found that the aortic CD11b⁺F4/80⁺CD206⁺ macrophages also have higher expression of the reparative markers Ym- 1 (Figure 2F), Arginase-1 (Figure 2G), and Fizz-1 (see Supplementary material online, Figure S2B), as compared to equivalent cells in WT^PCSK9 mice. In contrast, WT^PCSK9 mice had a higher percentage of aortic CD11b⁺F4/80⁺CD86⁺ cells (an M1 marker) (see Supplementary material online, Figure S2C and D). Thus, the macrophage phenotype observed in the aortas of ACE 10/10^PCSK9 mice shows more differentiation towards an M2- like phenotype than cells in WT^PCSK9 mice.

Figure 2

Reparative CD206⁺ macrophages in atherosclerosis. (A) Percent of CD11b⁺/F4/80⁺ cells in the CD11b⁺ aortic wall population as determined by flow cytometry. (B, C) Percent of F4/80⁺CD206⁺ macrophages in the aortas of WT^control, WT^PCSK9, ACE 10/10^control, and ACE 10/10^PCSK9 mice ± ramipril (an ACEi) or losartan (an ARB) assessed by flow cytometry. (D–G) Mean fluorescent intensity (MFI) of CD36 (D), IL-10 (E) Ym-1(F), and Arginase-1 (G). (H–K) MFI in F4/80⁺CD206⁺ aortic macrophages of MerTK (H), C1q (I), ABCA1 (J), and ABCG1 (K). Where indicated, mice were treated with either ramipril or losartan. Data are presented as the mean ± standard error of the mean (SEM). P-values are shown in the figure (A, F, G, J, and K) by two-tailed t-test or two-way ANOVA (C, D, E, H, and I). *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). n ≥ 6 mice.

MerTK (a receptor tyrosine kinase) and C1q (a complement component) are two molecules critical for macrophage efferocytosis. The expression of these molecules was studied by flow analysis and, like CD36, atherosclerotic ACE 10/10^PCSK9 aortic macrophages have more MerTK and C1q than equivalent WT^PCSK9 mice (Figure 2H and I). Further, the ACE inhibitor ramipril reduced MerTK and C1q expression far more than the ARB losartan treatment. In mice not treated with PCSK9, no significant differences were observed in the expression of these molecules. (Figure 2H and I).

As part of our flow analysis of aortic macrophages, we also examined the expression of the cholesterol transporters ATP-binding cassette A1 and G1 (ABCA1 and ABCG1) which are important in cholesterol efflux (Figure 2J and K). Expression of both transporters was significantly increased in aortic ACE 10/10^PCSK9 macrophages. Macrophage cholesterol efflux and efferocytosis are discussed below.

4.2 Increased ACE in atherosclerosis

Aortic tissue sections of ACE 10/10^PCSK9 and WT^PCSK9 mice were examined by confocal microscopy after staining with anti-ACE antibody. While more ACE is made by ACE 10/10^PCSK9 cells, significant ACE was also found in the lesional macrophages of WT^PCSK9 mice (Figure 3A and B, Supplementary material online, Figure S2E). The atherosclerotic environment increases ACE expression by macrophages as indicated by flow cytometric analysis of aortic wall F4/80⁺CD206⁺ macrophages from ACE 10/10^PCSK9 and WT^PCSK9 mice; cells from both groups made more ACE vs. equivalent cells from non-atherosclerotic ACE 10/10^control and WT^control mice (Figure 3C and D). Thus, this mouse model recapitulates the published histologic appearance of human atherosclerotic lesions which consistently show high expression of ACE by lesional macrophages.^12,13

Figure 3

Increased ACE in atherosclerotic macrophages. (A) Confocal microscopic pictures of the aortic roots from WT^PCSK9 and ACE 10/10^PCSK9 mice stained for expression of DAPI, CD206, PPARα, and ACE. Bar indicates 200 µm; enlarged bar 50 µm. L, lumen; W, wall. Arrows in the Enlarged panel indicate groups of cells. (B) In the histologic sections from panel A, CD206⁺ cells were quantitated for ACE expression. (C and D) In cells isolated from the aortas of WT^PCSK9 and ACE 10/10^PCSK9 mice, ACE expression by CD206⁺ macrophages was measured by flow cytometry. Panel C shows raw data while D shows mean fluorescence intensity (MFI). Data are presented as the mean ± standard error of the mean (SEM). P-values are shown in the figure (B and D) by a two-tailed t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). n ≥ 6 mice.

4.3 Increased peroxisome proliferator-activated receptor α (PPARα) in ACE 10/10^PCSK9 macrophages

Macrophages are a critical part of atherosclerosis lesion development and given the known role of PPARα in cellular handling of lipid, we next examined PPARα expression in aortic inflammatory cells by microscopy, flow analysis, and Western blot. Confocal microscopy after anti-PPARα staining showed that ACE 10/10^PCSK9 mice have more cells expressing PPARα around atherosclerotic lesions than WT^PCSK9 mice (Figures 3A and 4A). Higher PPARα expression in CD206 macrophages was also observed in the abdominal aorta (see Supplementary material online, Figure S2F). Flow analysis of aortic F4/80⁺CD206⁺ cells for PPARα also showed higher levels in ACE 10/10^PCSK9 cells vs. WT^PCSK9 cells (Figure 4B). Again, levels were reduced more with an ACEi than an ARB. Finally, PPARα levels were assessed by Western blot analysis of aortic inflammatory cells at the end of 24 weeks. And again, this showed higher levels of PPARα (1.9-fold WT^PCSK9 levels) and of effector molecules downstream of PPAR including carnitine palmitoyltransferase 1B (CPT1B), CPT1A, CPT2, and ACADL in ACE 10/10^PCSK9 cells vs. WT^PCSK9 cells (Figure 4C). Retinoid X receptor α (RXRα) was also increased. In contrast, Western blot analysis of PPARγ and PPARδ showed no difference between the two groups (see Supplementary material online, Figure S2G).

Figure 4

Increased ACE elevates PPARα in aortic macrophages. (A) In the histologic sections of the aortic root presented in Figure 3A, CD206⁺ cells were quantitated for PPARα expression by confocal microscopy. (B) The mean fluorescent intensity (MFI) of PPARα in CD206⁺ cells isolated from aortas of WT^PCSK9 and 10/10^PCSK9 mice was measured by flow analysis. Some mice were treated with either ramipril or losartan. (C) Western blot analysis of ACE and lipid metabolism-related markers CPT1A/B, CPT2, ACADL, PPARα, and RXRα made by aortic macrophages from WT^PCSK9 and 10/10^PCSK9 mice. (D–F) Oxygen consumption rates (OCR), calculated basal respiration, and maximal oxygen consumption rates of macrophages from atherosclerotic aortas determined by Seahorse analysis. (G–I) Seahorse analysis of basal respiration and maximal oxygen consumption by WT^PCSK9 and 10/10^PCSK9 thioglycollate elicited peritoneal macrophages (TPM). (J) Western blot analyses of CPT1A/B, CPT2, ACADL, PPARα, and RXRα in TPM from WT and ACE 10/10 mice treated with either control AAV (null-AAV) or PCSK9-AAV. Some of the PCSK9-AAV treated mice (i.e. WT^PCSK9 and ACE 10/10^PCSK9 mice) were also treated with ramipril or losartan. (K) ATP levels in OA-treated TPM from WT^PCSK9 and ACE 10/10^PCSK9 mice in the presence of ramipril or losartan. Data are presented as the mean ± standard error of the mean (SEM). P-values are shown in the figure (A, E, F, H, and I) by two-tailed t-test or two-way ANOVA (B and K) *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). Each point is one mouse (n ≥ 5).

To understand the effects of PPARα, the respiration of sorted F4/80⁺ aortic cells from ACE 10/10^PCSK9 and WT^PCSK9 mice were studied using the Seahorse machine. ACE 10/10^PCSK9 cells had a significant increase in both basal oxygen consumption (1.4-fold) and maximal oxygen consumption (1.6-fold) compared to WT^PCSK9 cells (Figure 4D–F).

The observation that cells expressing increased ACE also produce more PPARα, a transcription factor responsible for lipid metabolism, led us to ask if such cells use lipid differently from macrophages found in WT^PCSK9-treated mice. Due to the insufficient yield of aortic macrophages for these studies, lipid metabolism was studied using thioglycollate-elicited peritoneal macrophages (TPM). As in macrophages from the aorta, ACE 10/10^PCSK9 TPM had higher basal respiration, and higher maximal oxygen consumption vs. WT^PCSK9 cells (Figure 4G–I). Western blot analysis also showed a similar pattern to aortic macrophages: increased PPARα, CPT1B, CPT1A, CPT2, ACADL, and RXRα in 10/10^PCSK9 TPM (Figure 4J, Supplementary material online, Figure S3A–F). Cellular ATP was higher in ACE 10/10^PCSK9 TPM as compared to equivalent WT^PCSK9 TPM (Figure 4K). There was no significant change in fatty acid synthase (see Supplementary material online, Figure S3G). Treatment of animals with the ACE inhibitor ramipril reduced oxidative capacity, PPARα, and downstream effectors of fatty acid oxidation (Figure 4J and K). Thus, these cells are phenotypically similar to cells found in the aortas of ACE 10/10^PCSK9 and WT^PCSK9 mice.

4.4 Enhanced β-oxidation in 10/10^PCSK9 macrophages

TPM was cultured with either uniformly labelled ¹³C-glucose or ¹³C-oleic acid (OA) for 48 h and then analysed by GC/MS for the percent of citrate incorporating ¹³C.²⁶ This assessment of metabolism showed a striking difference between macrophages from ACE 10/10^PCSK9 and WT^PCSK9 mice, particularly for the handling of lipid. The incorporated ¹³C ratio for ACE 10/10 ÷ WT is 2.1 for glucose and 8.5 for OA (Figure 5A). Acetyl-CoA is also increased in ACE 10/10 cells (Figure 5B) which is typical of increased lipid metabolism.^30–32 The increased utilization and metabolism (β-oxidation) of a long-chain fatty acid by the ACE 10/10^PCSK9 cells is consistent with the increased oxygen consumption of these cells.

Figure 5

PPARα activation in macrophages with increased ACE expression. (A) Peritoneal macrophages from WT^PCSK9 and ACE 10/10^PCSK9 mice were incubated with ¹³C-glucose/unlabelled OA or ¹³C-OA/unlabelled glucose to compare ¹³C incorporation into citric acid as measured by mass spectrometry. This assay measures cell metabolic preference. (B) Acetyl-CoA levels in WT^PCSK9 and ACE 10/10^PCSK9 TPM. (C and D) The oxygen consumption of WT^PCSK9 TPM and ACE 10/10^PCSK9 TPM treated in vitro with OA was measured using the MitoXpress assay system. Basal respiration is shown in C and the maximal respiration rate in D. Data from TPM treated as above but incubated with ramipril, losartan, or the CPT1 etomoxir are also shown. (E) Cell expression of ACE and nuclear PPARα after OA treatment by WT, ACE 10/10, and ACE KO TPM were studied by immunofluorescence. The TPM is from mice without atherosclerosis or a high-fat diet. Bar indicates 10 µm. (F) Quantitation of PPARα fluorescence intensity of cells in panel E. (G and H) Western blot analysis of untreated (control) and OA treated TPM for CPT1A/B, CPT2, ACADL, ATGL, PPARα, and RXRα. Each point represents data from a single mouse. Data are given as the mean ± standard error of the mean (SEM). P-values are shown in the figure (A and B) by two-tailed t-test or two-way ANOVA (C, D, F, and H). *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). n ≥ 6.

To determine the direct influence of ACE on fatty acid oxidation, we cultured TPM in one of four conditions: oleic acid (OA), OA plus ramipril (10 μM), OA plus the ARB losartan (100 μM), and OA plus the CPT1 inhibitor etomoxir (5 μM). After 48 h of culture, basal respiration rate and maximum oxygen consumption were higher in ACE 10/10^PCSK9 than in WT^PCSK9 cell (Figure 5C and D, Supplementary material online, Figure S4A and B). Again, an ACEi was much more effective in reducing these metabolic parameters than an ARB suggesting that a peptide other than angiotensin II induces the metabolic and phenotypic changes found in ACE 10/10^PCSK9 macrophages. Further, etomoxir markedly reduced respiration rates of both ACE 10/10^PCSK9 and WT^PCSK9 macrophages, eliminating the difference between the two groups and indicating that enhanced β-oxidation also plays a significant role in the ACE 10/10^PCSK9 phenotype.

4.5 ACE 10/10 macrophages cultured with oleic acid express more PPARα

To further investigate the role of ACE in lipid metabolism, TPMs from WT, ACE 10/10, and ACE KO mice (not treated with PCSK9 or an atherogenic diet) were harvested and cultured for 48 h with OA to measure lipid uptake. The use of OA in vitro simulates the high lipid conditions of in vivo disease.^33,34 Confocal microscopy clearly demonstrated that ACE 10/10 macrophages contained more nuclear PPARα than WT or ACE KO cells. (Figure 5E and F). Indeed, fatty acid oxidation-related proteins such as CPT1B, CPT1A, CPT2, ACADL, and ATGL, as well as PPARα and RXRα were higher in ACE 10/10 macrophages treated with OA compared to equivalently treated WT or ACE KO cells (Figure 5G and H). Non-OA-treated groups showed no such differences (see Supplementary material online, Figure S4C). In addition, reparative macrophage markers CD206 and CD163 were increased in the ACE 10/10 macrophages (Figure 6A and B) while the pro-inflammatory markers CD80 and CD86 were similar in the three groups of cells (Figure 6C and D).

Figure 6

ACE increases the reparative macrophage phenotype induced by oleic acid treatment. Expression of CD206 (A), CD163 (B), CD80 (C) CD86 (D), MerTK (E), C1q (F), and IL-10 (G) by OA treated TPM from WT, ACE 10/10, and ACE KO mice. Expression was measured by flow cytometry. (H) Real-time quantitative PCR of genes associated with β-oxidation (PPARα, RXRα, CPT1A, CPT1B, CPT2, and CD36), reparative (M2) genes (Arg-1, Ym-1, FIZZ-1), and genes associated with efferocytosis (C1q, IL-10, and MerTK) by OA treated TPM from ACE KO, ACE 10/10, and WT mice. Each point is from a single mouse. Data are given as the mean ± standard error of the mean (SEM). P-values are shown in the figure (A–G) by two-way ANOVA. n ≥ 5. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). Each point represents data from a single mouse.

In analysing cells from the atherosclerotic aortas of ACE 10/10^PCSK9 and WT^PCSK9 mice, we found that 10/10^PCSK9 macrophages expressed increased amounts of the efferocytotic genes MerTK and C1q, as well as increased amounts of the M2 cytokine IL-10 (Figure 2E, H and I). Similar flow-based analyses of OA-treated TPMs showed increased expression of these proteins by ACE 10/10 cells vs. WT, and more expression in WT cells vs. ACE KO (Figure 6E–G). In fact, RNA quantitation of several genes from OA-treated TPMs isolated from ACE 10/10, WT, and ACE KO mice showed a similar gradient where the highest expression of PPARα and downstream genes (CPT1, CPT2), reparative M2 genes (Arg1, Ym1), and efferocytotic genes (C1q, MerTK) is in the ACE 10/10 cells while the least expression is in the ACE KO cells (Figure 6H). Thus, in many ways, peritoneal macrophages treated with OA resemble aortic and peritoneal macrophages from atherosclerotic ACE 10/10^PCSK9 mice.

4.6 Increased lipid metabolism in macrophages with increased ACE depends on PPARα

To analyse the effect of ACE on lipid handling, TPMs from WT, ACE 10/10, and ACE KO mice (not treated with PCSK9 or an atherogenic diet) were harvested and cultured for 48 h with OA to measure lipid uptake by electron microscopy (Figure 7A) and by measurement of fluorescence (Figure 7B, Supplementary material online, Figure S5A). More lipid droplets accumulated in ACE 10/10 macrophages as compared to WT and ACE KO (1.4-fold WT; 2.2-fold KO, Figure 7B). This was blocked by the PPARα inhibitor GW6471.³⁵ To study droplet utilization, TPM were cultured for 48 h in OA. OA culture media was then replaced with fatty acid-free media and the cells were cultured for an additional 48 h with periodic assessment of intracellular lipid. The reduction of lipid droplets in ACE 10/10 macrophages was significantly greater after 18 h vs. WT or ACE KO cells (Figure 7C, P < 0.01). The efficiency of lipolysis was also higher in ACE 10/10 cells as compared to the other 2 groups (Figure 7D, P < 0.05) and was reduced with GW6471. These data suggest that macrophages expressing increased ACE have an increased capability to process lipids via a PPARα dependent pathway.

Figure 7

ACE expression increases macrophage lipid uptake, utilization, and cell efferocytosis via the PPARα pathway. (A) In vitro assessment by electron microscopy of lipid droplets in ACE WT, ACE 10/10, and ACE KO TPM after 48 h OA treatment with or without GW6471. Bar indicates 4 µm and 2 µm; enlarged bar 600 and 800 nm. (B) TPMs were cultured with OA as in panel A and then stained with Droplite^TM red. Lipid uptake was determined by fluorescence. (C) Reduction of intracellular lipid droplets over time in WT, ACE 10/10, and ACE KO TPM. (D) Lipolysis of TPM after OA treatment with or without GW6471. (E) Acetyl-CoA levels in WT, ACE 10/10, and KO TPM after OA with or without GW6471. (F and G) Basal (F) and maximal (G) respiration rates of TPM from WT, ACE 10/10, and ACE KO mice were measured using the MitoXpress assay. Cells were treated with OA for 48 h along with the FAO inhibitor etomoxir (ETO, 5 μM) or the PPARα antagonist GW6471 (10 μM). (H) The ATP content of WT, ACE 10/10, and ACE KO TPM was measured. Cells were treated as indicated in the panel. (I and J) Efferocytosis of CSFE-labelled apoptotic neutrophils by ACE 10/10, WT, and ACE KO TPM after OA treatment in the presence of etomoxir or GW6471. The ratio of macrophages/apoptotic cells was 1:4. Neutrophil uptake was measured by flow cytometry. Each point represents a single mouse. Data are presented as the mean ± standard error of the mean (SEM). P-values are shown in the figure (B–I) by two-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). (n ≥ 5).

Similar to the TPM of mice fed a high-fat, ACE 10/10 macrophages exposed to OA in vitro contained more acetyl-CoA than similarly treated WT cells, and this difference was eliminated with GW6471 (Figure 7E). Further, macrophage basal respiration rate, maximum oxygen consumption, and cellular ATP content were increased in ACE 10/10 cells as compared to WT or ACE KO following OA treatment (Figure 7F–H, Supplementary material online, Figure S5B, C); the PPARα inhibitor GW6471 (10 μM) and the CPT1 inhibitor etomoxir eliminated the ACE-induced differences in fatty acid oxidation and ATP production.

4.7 Increased efferocytosis in ACE 10/10 macrophages

A critical function of macrophages in atherosclerosis and other disease processes is efferocytosis; others have suggested impaired efferocytosis in atherosclerotic macrophages secondary to impaired metabolism.^36,37 To directly measure efferocytosis, TPM from ACE 10/10, WT, and ACE KO cells were co-cultured for 16 h with CSFE-labelled apoptotic neutrophils. In the absence of OA, efferocytosis was similar in all three groups of cells (see Supplementary material online, Figure S5D). However, in the high lipid environment modelled by 48 h of culture with OA, ACE 10/10 macrophages showed significantly increased efferocytosis compared to WT or ACE KO cells (Figure 7I and J). Pretreatment of macrophages with the CPT1 inhibitor etomoxir eliminated the difference in efferocytosis seen in the three groups of cells, indicating an important role of lipid β-oxidation in the enhanced efferocytotic capacity of ACE 10/10 macrophages present in a high lipid environment.

4.8 Increased cholesterol efflux in ACE 10/10 macrophages

To mimic in vivo conditions in atherosclerosis, TPMs from ACE 10/10, WT, and ACE KO mice were cultured with ox-LDL (30 μg/mL) for 12 h and then ox-LDL uptake was determined by flow cytometry. ACE 10/10 macrophages had an increased rate of ox-LDL uptake compared to WT cells while ACE KO cells had decreased uptake compared to WT (Figure 8A). PPARα expression was also higher in ACE 10/10 cells compared to WT or ACE KO cells (Figure 8B). The PPARα inhibitor GW6471 eliminated the differences in ox-LDL uptake and PPARα expression.

Figure 8

ACE 10/10 macrophages have increased cholesterol efflux. (A–E) ACE WT, ACE 10/10, and ACE KO TPMs were treated with ox-LDL with or without GW6471 (30 μg/mL) for 12 h. Flow analysis of ox-LDL uptake (A), PPARα expression (B), CD36 expression (C), ABCA1 expression (D), and ABCG1 expression (E). (F) Cholesterol efflux percent over a 4-h period of ACE WT, ACE 10/10, and ACE KO TPMs treated with ox-LDL with or without GW6471 (30 μg/mL) for 12 h. (G) Cholesterol efflux percent over a 4-h period of TPM from WT^PCSK9 and ACE 10/10^PCSK9 mice treated with ox-LDL (30 μg/mL) for 12 h in the presence of ramipril or GW6471. Cholesterol efflux was measured using the protocol and reagents from BioVision (Milpitas, CA). Each point is from a single mouse. Data are given as the mean ± standard error of the mean (SEM). P-values are shown in the figure (A–G) by two-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). n ≥ 6.

As mentioned earlier, aortic macrophages from atherosclerotic ACE 10/10^PCSK9 mice express more of the scavenger receptor CD36 and of the cholesterol transporters ABCA1 and ABCG1 than WT^PCSK9 mice (Figure 2D, J, and K). Following in vitro ox-LDL exposure, we found that ACE 10/10 macrophages also expressed more CD36, ABCA1, and ABCG1 which play key roles in LDL uptake and cholesterol efflux (Figure 8C–E). As above, these differences were eliminated with GW6471. To measure cholesterol efflux, TPMs from ACE 10/10, WT, and ACE KO mice were treated with ox-LDL for 12 h ± GW6471. Then, after washing, the cells were incubated with fluorescently labelled cholesterol and allowed to equilibrate. Finally, total cellular efflux was measured (Figure 8F). The cholesterol efflux data were similar to the expression data for ABCA1 and ABCG1: significantly more efflux was observed with the ACE 10/10 cells vs. WT or ACE KO. These differences were eliminated with GW6471. Finally, to investigate the ACE 10/10 effect in a system closer to atherosclerosis, TPMs from 18-week-old ACE 10/10^PCSK9 and WT^PCSK9 mice were treated in vitro either with or without ox-LDL for 12 h, and then cholesterol efflux was measured as in Figure 8F. Both with exposure to ox-LDL or without exposure, ACE 10/10^PCSK9 TPMs had a higher cholesterol efflux than WT^PCSK9 (Figure 8G). Differences between the two groups were eliminated with either treatment with the ACE inhibitor lisinopril or with the PPARα inhibitor GW6471. The consistency of GW6471 inhibition suggests that many aspects of the better lipid handling seen in ACE 10/10 macrophages appear mediated by the PPARα pathway.

4.9 Increased ACE promotes PPARα activation in the human THP-1 cell line

THP-1 is a human monocyte-like line that becomes adherent when differentiated into macrophages with phorbol 12-myristate-13-acetate (PMA). We studied differentiated THP-1 cells stably transfected with a lentivirus construct that increased cellular ACE expression (THP-1^ACE) and compared it to THP-1 cells stably transfected with empty lentivirus vector (THP-1^Vector). In the presence of OA, THP-1^ACE cells increased expression of PPARα, CPT1A, and CPT1B (Figure 9A). PPARα is a transcription factor; it binds to specific DNA sequences termed PPARα response elements (PPRE) to activate downstream signalling. THP-1^Vector and THP-1^ACE cells were transfected with a PPRE-luciferase plasmid and exposed to OA. Under these conditions, the THP-1^ACE cells expressed more luciferase activity (Figure 9B), an effect blocked by the ACE inhibitor lisinopril. ACE expression increases cellular ATP and acetyl-CoA similar to that observed in mouse TPM overexpressing ACE (Figure 9C and D). Finally, the efferocytotic ability of THP-1^ACE cells was 1.5-fold higher than THP-1^Vector cells following culture with OA (Figure 9E and F). This difference, and indeed the stimulation of efferocytosis by OA, was eliminated by the PPARα inhibitor GW6471.

Figure 9

Effect of ACE expression in the human THP-1 cell line. THP-1 cells were stably transfected with an ACE-expressing lentivirus vector (THP-1^ACE) or an empty vector (THP- 1^VECTOR). The THP-1 cells were primed with 250 nM PMA for 48 h to induce a macrophage phenotype and then treated with OA (200 µM) for 48 h. (A) ACE, PPARα, and the downstream metabolic target CPT1A and CPT1B were studied by Western blot analysis. (B) THP-1^ACE and THP-1^VECTOR cells were transfected with a PPRE driven-luciferase plasmid, induced to a macrophage phenotype, and treated with OA 200 µM ± the ACE inhibitor lisinopril for 48 h before luciferase activity was measured. ACE inhibition blocks luciferase expression. (C and D) Cellular ATP production and acetyl-CoA concentration in THP-1^vector and THP-1^ACE cells treated with OA. (E and F) Flow cytometric analysis of efferocytosis. THP-1^ACE and THP-1^VECTOR cells were induced to a macrophage phenotype and then co-cultured with apoptotic neutrophils as in Figure 7I and J and the percent of cells efferocytosing neutrophils was measured. Data are given as the mean ± standard error of the mean (SEM). P-values are shown in the figure (B–E) by two-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001, and NS (not significant). n ≥ 5.

5. Discussion

Modern thinking about atherosclerosis focuses on both vascular wall lipids and the accompanying inflammation, with particular emphasis on macrophage function. This framework reflects the finding that lesional macrophages are not only ineffective in removing vascular lipid but through cell dysfunction and death contribute to lesion formation. Previous observation showed that the macrophages present in human atherosclerotic lesions—areas of high lipid—express abundant ACE,^12,13,38 but little is known regarding the specific function of ACE in these cells. In fact, macrophages express abundant ACE in a variety of chronic granulomatous diseases where lipid must be abundant including lesional macrophages in sarcoid, histoplasmosis, leprosy, and miliary tuberculosis.^4–6 Indeed, very high levels of macrophage ACE are observed in Gaucher’s disease where lipid-laden macrophages form due to the absence of lysosomal glucocerebrosidase.³⁹

ACE cleaves angiotensin I, but the enzyme has many other substrates, and it is now known that macrophage expression of ACE has metabolic effects that are independent of angiotensin II, as best exemplified in ACE 10/10 mice. In these mice, genetic mutation has altered control of the ACE gene from the natural ACE promoter to the c-fms promoter, active in myeloid cells, with the result that the monocytes and macrophages of these animals constitutively express increased levels of ACE. The surprising phenotype is an enhanced immune response to several different immune challenges and substantially less atherosclerosis than WT mice in both the PCSK9/high-fat diet (present study) and the ApoE knockout mouse/high-fat diet models.¹⁸ Here, a key finding is that macrophages expressing increased ACE demonstrate a profoundly altered phenotype in terms of their ability to process lipids. In addition to reduced atherosclerosis, we find that macrophages with increased ACE expression have higher levels of surface scavenger receptors (Figure 2D), take up and process more lipid (Figure 5A and Figure 7A–C), have increased capacity to transport long-chain fatty acids into mitochondria (Figure 4C), have higher oxidative metabolism (Figure 4D–I) with increased cell ATP (Figure 4K), have increased cell concentrations of the lipid transporters ABCA1 and ABCG1 with increased cholesterol efflux (Figures 2J, K and 8F), and have increased capacity for efferocytosis (Figure 7I). In short, ACE 10/10 has many changes in the handling of lipids that might lead one to predict more resistance to the development of atherosclerosis and, in fact, this is what is seen. Indeed, we speculate that increased ACE made by macrophages in human atherosclerotic lesions, increased ACE made by macrophages in human granuloma, and increased ACE made by mouse macrophages with tissue-destroying bacterial infections are all linked to the presence of a high lipid environment. Apparently, the natural expression of macrophage ACE is an adaptive response to high lipids and, at least in mice, this alters macrophage metabolism.

While increased ACE changes many aspects of macrophage lipid handling, we would highlight three findings. First, 5A shows the results of feeding TPM from atherosclerotic mice ¹³C-labelled glucose or ¹³C-OA and then measuring the incorporation of the label into the downstream metabolic product citrate. This type of assay is integrative in that what is measured is both uptake of the labelled metabolite and then its metabolism. While the ACE 10/10^PCSK9 cells incorporate more ¹³C from both glucose and OA into citrate than WT^PCSK9, the magnitude of the difference between ACE 10/10^PCSK9 and WT^PCSK9 cells in handling the lipid OA (over 8-fold that of WT^PCSK9 cells) indicates the level of discrepancy in how these two macrophage populations react to and accommodate a high lipid environment.

As reviewed by Kojima et al., evidence suggests that phagocyte efferocytosis becomes less effective during atherogenesis and that this may be a driver of atherosclerotic lesion expansion.³⁶ While there are probably several reasons for this, one cited by these authors is the possible skewing in developing atherosclerotic plaques of macrophages towards M1 cells and away from a more phagocytically effective M2 phenotype. As shown here, increased expression of ACE in ACE 10/10^PCSK9 is associated with greater M2 macrophage development. Further, two critical molecules for efferocytosis are MerTK and C1q, expression of which are both increased in ACE 10/10^PCSK9. Also, a recent finding suggests that macrophages need β-oxidation to improve efferocytosis.⁴⁰ Whatever the exact biochemical cause(s), in a lipid-rich environment, ACE 10/10 macrophages demonstrate an increased ability for efferocytosis which probably plays a role in the reduction of atherosclerotic disease in ACE 10/10 mice.

A third finding is the increased activity of the cholesterol transporters ABCA1 and ABCG1 in the ACE 10/10^PCSK9 macrophages. These transporters mediate ATP-dependent cholesterol efflux to HDL. Mutation of the ABC1 gene is responsible for Tangiers disease and it appears very clear that the systemic lack of ABC1 function results in low plasma HDL levels and the accumulation of cholesterol in tissue macrophages.⁴¹ A number of studies in mice have examined the role of macrophage ABCA1 and ABCG1 as related to the formation of atherosclerotic lesions.^42–46 These studies have demonstrated convincingly that whether the entire bone marrow or just macrophages/neutrophils lack ABCA1 and ABCG1, there is foam cell accumulation and accelerated atherosclerosis in mice. These data are consistent with our observations that increased macrophage ACE elevates cholesterol efflux which contributes to the reduced atherosclerosis in ACE 10/10 mice.

While we posit that changes in lipid handling, efferocytosis, and cholesterol efflux contribute to reduced atherosclerosis, the key finding of our study is that ACE 10/10 macrophages, as a direct result of ACE catalytic function, express significantly more PPARα than WT cells in a high lipid environment. We believe that increased macrophage PPARα sets in motion many of the biochemical changes discussed above and is a critical molecule responsible for better macrophage function and reduced atherosclerosis in ACE 10/10 mice. PPARα is a transcription factor that regulates many aspects of lipid metabolism for cholesterol, triglycerides, phospholipid, and fatty acids.⁴⁷ PPARα affects lipid uptake, import into mitochondria, and fatty acid oxidation, a process that generates abundant ATP in cells, particularly M2-like macrophages.^48–51 Proteins mentioned here whose expression is reported to be at least partially regulated by PPARα include CD36, CPT1, C1q, ABCA1, and ABCG1.⁴⁷

Our studies were performed in mice, but we note that human studies in atherosclerosis have found that, even with statin treatment, cardiovascular disease is predicted by a dyslipidemic profile of high triglycerides and low HDL-cholesterol.^52,53 This phenotype can be treated by induction of PPARα and is somewhat similar to our model in which ACE 10/10^PCSK9 mice fed a high-fat diet increase macrophage PPARα. We do not know yet whether the entirety of protection against atherosclerosis in the ACE 10/10 mouse is secondary to elevated PPARα but what is clear is that this protection is due to macrophage ACE catalytic activity making a presently unidentified peptide that induces macrophage PPARα and perhaps other molecules with anti-atherosclerotic properties. Alkatiri et al. have discussed human clinical trials comparing the use of ACEi vs. ARBs to induce atheromatous plaque regression.⁵⁴ While statins and ARBs have been found capable of inducing plaque regression, such regression was not seen with ACEi. Alkatire suggests several possible reasons for this difference. We would add one additional possibility: if a natural function of ACE made by human macrophages is to increase the ability of these cells to handle lipid, then ARBs, which block the actions of angiotensin II, might be predicted to be better than ACEi, which block the action of ACE on all peptides, in regressing atherosclerotic plaques. Finally, studies here, in previous analyses of ACE 10/10 mice, and in mice over- expressing ACE in neutrophils have consistently indicated that the ACE effects on myeloid cells are not mediated by angiotensin II or bradykinin but rather by some currently unidentified ACE product. Identification of this product and how it works may give new insights into treating atherosclerosis.^4–6,11,55

Supplementary material

Supplementary material is available at Cardiovascular Research online.

Authors’ contributions

D.-Y.C., K.E.B., and Z.K. conceived and designed the experiments. D.-Y.C., Z.K., D.J.L., and X.M.L. acquired the data with assistance from A.R.V. and M.C. (tissue sections and RNA analysis sampling), L.C.V., E.A.B., A.O.H, D.O.-D., S.S., and F.A. (macrophages isolation, cell culture, animal husbandry, and mice genotyping), S. Charugundla, and A.J. Lusis (plasma lipids), and L.C., and R.E.T. (liver lipids). D.-Y.C., D.O.-D., S.S., T.S., and A.S.D. conducted macrophages sorting and metabolism study. D.-Y.C. and Z.K. analysed data. D.-Y.C., Z.K., and K.E.B. interpreted the results of experiments. D.-Y.C. and K.E.B. prepared figures and drafted the manuscript. D.O.-D., J.F.G., and K.E.B. edited the manuscript.

Acknowledgements

This study was supported by AHA grants 23CDA1052548 (D.Y.C.), AHA grants 19CDA34760010 (Z.K.) and 16SDG30130015 (J.F.G.), NIH grant P01HL129941 (K.E.B.), R01AI134714 (K.E.B.), R01AI164519 (K.E.B.), R01HL142672 (J.F.G.), P30DK063491 (J.F.G.), K99HL141638 (D.O.-D.), R35 GM138003 (A.S.D.) and P30 DK063491 (A.S.D.). We thank Dr. Prediman K. Shah for the discussions concerning atherosclerosis. We also thank Dr. Aldons Jake Lusis (UCLA) and Sharda Charugundla for their discussions and help in measuring plasma lipid levels. The authors would also like to thank Daniel N. Leal, BS, CEMT for excellent assistance in performing electron microscopy.

Data availability

All primary data is available from the 1st author and will be deposited in Harvard Dataverse upon publication.

References

1

Moore

KJ

,

Sheedy

FJ

,

Fisher

EA

.

Macrophages in atherosclerosis: a dynamic balance

.

Nat Rev Immunol

2013

;

13

:

709

–

721

.

2

Barrett

TJ

.

Macrophages in atherosclerosis regression

.

Arterioscler Thromb Vasc Biol

2020

;

40

:

20

–

33

3

Chen

R

,

Suchard

MA

,

Krumholz

HM

,

Schuemie

MJ

,

Shea

S

,

Duke

J

,

Pratt

N

,

Reich

CG

,

Madigan

D

,

You

SC

,

Ryan

PB

,

Hripcsak

G

.

Comparative first-line effectiveness and safety of ACE (angiotensin-converting enzyme) inhibitors and angiotensin receptor blockers: a multinational cohort study

.

Hypertension

2021

;

78

:

591

–

603

.

4

Cao

DY

,

Saito

S

,

Veiras

LC

,

Okwan-Duodu

D

,

Bernstein

EA

,

Giani

JF

,

Bernstein

KE

,

Khan

Z

.

Role of angiotensin-converting enzyme in myeloid cell immune responses

.

Cell Mol Biol Lett

2020

;

25

:

31

.

5

Bernstein

KE

,

Khan

Z

,

Giani

JF

,

Cao

DY

,

Bernstein

EA

,

Shen

XZ

.

Angiotensin-converting enzyme in innate and adaptive immunity

.

Nat Rev Nephrol

2018

;

14

:

325

–

336

.

6

Cao

D

,

Veiras

L

,

Ahmed

F

,

Shibata

T

,

Bernstein

EA

,

Okwan-Duodu

D

,

Giani

JF

,

Khan

Z

,

Bernstein

KE

.

The non-cardiovascular actions of ACE

.

Peptides

2022

;

152

:

170769

.

7

Okwan-Duodu

D

,

Datta

V

,

Shen

XZ

,

Goodridge

HS

,

Bernstein

EA

,

Fuchs

S

,

Liu

GY

,

Bernstein

KE

.

Angiotensin-converting enzyme overexpression in mouse myelomonocytic cells augments resistance to Listeria and methicillin-resistant Staphylococcus aureus

.

J Biol Chem

2010

;

285

:

39051

–

39060

.

8

Shen

XZ

,

Li

P

,

Weiss

D

,

Fuchs

S

,

Xiao

HD

,

Adams

JA

,

Williams

IR

,

Capecchi

MR

,

Taylor

WR

,

Bernstein

KE

.

Mice with enhanced macrophage angiotensin-converting enzyme are resistant to melanoma

.

Am J Pathol

2007

;

170

:

2122

–

2134

.

9

Koronyo-Hamaoui

M

,

Sheyn

J

,

Hayden

EY

,

Li

S

,

Fuchs

DT

,

Regis

GC

,

Lopes

DHJ

,

Black

KL

,

Bernstein

KE

,

Teplow

DB

,

Fuchs

S

,

Koronyo

Y

,

Rentsendorj

A

.

Peripherally derived angiotensin converting enzyme-enhanced macrophages alleviate Alzheimer-related disease

.

Brain

2020

;

143

:

336

–

358

.

10

Bernstein

KE

,

Koronyo

Y

,

Salumbides

BC

,

Sheyn

J

,

Pelissier

L

,

Lopes

DH

,

Shah

KH

,

Bernstein

EA

,

Fuchs

DT

,

Yu

JJ

,

Pham

M

,

Black

KL

,

Shen

XZ

,

Fuchs

S

,

Koronyo-Hamaoui

M

.

Angiotensin-converting enzyme overexpression in myelomonocytes prevents Alzheimer’s-like cognitive decline

.

J Clin Invest

2014

;

124

:

1000

–

1012

.

11

Cao

DY

,

Giani

JF

,

Veiras

LC

,

Bernstein

EA

,

Okwan-Duodu

D

,

Ahmed

F

,

Bresee

C

,

Tourtellotte

WG

,

Karumanchi

SA

,

Bernstein

KE

,

Khan

Z

.

An ACE inhibitor reduces bactericidal activity of human neutrophils in vitro and impairs mouse neutrophil activity in vivo

.

Sci Transl Med

2021

;

13

:

eabj2138

.

12

Diet

F

,

Pratt

RE

,

Berry

GJ

,

Momose

N

,

Gibbons

GH

,

Dzau

VJ

.

Increased accumulation of tissue ACE in human atherosclerotic coronary artery disease

.

Circulation

1996

;

94

:

2756

–

2767

.

13

Ohishi

M

,

Ueda

M

,

Rakugi

H

,

Naruko

T

,

Kojima

A

,

Okamura

A

,

Higaki

J

,

Ogihara

T

.

Enhanced expression of angiotensin-converting enzyme is associated with progression of coronary atherosclerosis in humans

.

J Hypertens

1997

;

15

:

1295

–

1302

.

14

Cao

DY

,

Spivia

WR

,

Veiras

LC

,

Khan

Z

,

Peng

Z

,

Jones

AE

,

Bernstein

EA

,

Saito

S

,

Okwan-Duodu

D

,

Parker

SJ

,

Giani

JF

,

Divakaruni

AS

,

Van Eyk

JE

,

Bernstein

KE

.

ACE Overexpression in myeloid cells increases oxidative metabolism and cellular

ATP.

J Biol Chem

2020

;

295

:

1369

–

1384

.

15

Oppi

S

,

Luscher

TF

,

Stein

S

.

Mouse models for atherosclerosis research-which is my line?

Front Cardiovasc Med

2019

;

6

:

46

.

16

Goettsch

C

,

Hutcheson

JD

,

Hagita

S

,

Rogers

MA

,

Creager

MD

,

Pham

T

,

Choi

J

,

Mlynarchik

AK

,

Pieper

B

,

Kjolby

M

,

Aikawa

M

,

Aikawa

E

.

A single injection of gain-of- function mutant PCSK9 adeno-associated virus vector induces cardiovascular calcification in mice with no genetic modification

.

Atherosclerosis

2016

;

251

:

109

–

118

.

17

Kumar

S

,

Kang

DW

,

Rezvan

A

,

Jo

H

.

Accelerated atherosclerosis development in C57Bl6 mice by overexpressing AAV-mediated PCSK9 and partial carotid ligation

.

Lab Invest

2017

;

97

:

935

–

945

.

18

Okwan-Duodu

D

,

Weiss

D

,

Peng

Z

,

Veiras

LC

,

Cao

DY

,

Saito

S

,

Khan

Z

,

Bernstein

EA

,

Giani

JF

,

Taylor

WR

,

Bernstein

KE

.

Overexpression of myeloid angiotensin-converting enzyme (ACE) reduces atherosclerosis

.

Biochem Biophys Res Commun

2019

;

520

:

573

–

579

.

19

Temel

RE

,

Lee

RG

,

Kelley

KL

,

Davis

MA

,

Shah

R

,

Sawyer

JK

,

Wilson

MD

,

Rudel

LL

.

Intestinal cholesterol absorption is substantially reduced in mice deficient in both ABCA1 and ACAT2

.

J Lipid Res

2005

;

46

:

2423

–

2431

.

20

Sinha

SK

,

Miikeda

A

,

Fouladian

Z

,

Mehrabian

M

,

Edillor

C

,

Shih

D

,

Zhou

Z

,

Paul

MK

,

Charugundla

S

,

Davis

RC

,

Rajavashisth

TB

,

Lusis

AJ

.

Local M-CSF (macrophage colony-stimulating factor) expression regulates macrophage proliferation and apoptosis in atherosclerosis

.

Arterioscler Thromb Vasc Biol

2021

;

41

:

220

–

233

.

PubMed

21

Syed

S

,

Nissila

E

,

Ruhanen

H

,

Fudo

S

,

Gaytan

MO

,

Sihvo

SP

,

Lorey

MB

,

Metso

J

,

Oorni

K

,

King

SJ

,

Oommen

OP

,

Jauhiainen

M

,

Meri

S

,

Kakela

R

,

Haapasalo

K

.

Streptococcus pneumoniae pneumolysin and neuraminidase A convert high-density lipoproteins into pro- atherogenic particles

.

iScience

2021

;

24

:

102535

.

22

Degrelle

SA

,

Shoaito

H

,

Fournier

T

.

New transcriptional reporters to quantify and monitor PPARgamma activity

.

PPAR Res

2017

;

2017

:

6139107

.

23

Hu

D

,

Yin

C

,

Mohanta

SK

,

Weber

C

,

Habenicht

AJ

.

Preparation of single cell suspensions from mouse aorta

.

Bio Protoc

2016

;

6

:

e1832

.

24

Tumanov

S

,

Bulusu

V

,

Kamphorst

JJ

.

Analysis of fatty acid metabolism using stable isotope tracers and mass spectrometry

.

Methods Enzymol

2015

;

561

:

197

–

217

.

25

Vacanti

NM

,

Divakaruni

AS

,

Green

CR

,

Parker

SJ

,

Henry

RR

,

Ciaraldi

TP

,

Murphy

AN

,

Metallo

CM

.

Regulation of substrate utilization by the mitochondrial pyruvate carrier

.

Mol Cell

2014

;

56

:

425

–

435

.

26

Jones

AE

,

Sheng

L

,

Acevedo

A

,

Veliova

M

,

Shirihai

OS

,

Stiles

L

,

Divakaruni

AS

.

Forces, fluxes, and fuels: tracking mitochondrial metabolism by integrating measurements of membrane potential, respiration, and metabolites

.

Am J Physiol Cell Physiol

2021

;

320

:

C80

–

C91

.

PubMed

27

Trefely

S

,

Ashwell

P

,

Snyder

NW

.

Fluxfix: automatic isotopologue normalization for metabolic tracer analysis

.

BMC Bioinformatics

2016

;

17

:

485

.

28

Divakaruni

AS

,

Paradyse

A

,

Ferrick

DA

,

Murphy

AN

,

Jastroch

M

.

Analysis and interpretation of microplate-based oxygen consumption and pH data

.

Methods Enzymol

2014

;

547

:

309

–

354

.

29

Hynes

J

,

Nadanaciva

S

,

Swiss

R

,

Carey

C

,

Kirwan

S

,

Will

Y

.

A high-throughput dual parameter assay for assessing drug-induced mitochondrial dysfunction provides additional predictivity over two established mitochondrial toxicity assays

.

Toxicol In Vitro

2013

;

27

:

560

–

569

.

30

Houten

SM

,

Violante

S

,

Ventura

FV

,

Wanders

RJ

.

The biochemistry and physiology of mitochondrial fatty acid beta-oxidation and its genetic disorders

.

Annu Rev Physiol

2016

;

78

:

23

–

44

.

31

Aon

MA

,

Bhatt

N

,

Cortassa

SC

.

Mitochondrial and cellular mechanisms for managing lipid excess

.

Front Physiol

2014

;

5

:

282

.

32

Jang

HS

,

Noh

MR

,

Kim

J

,

Padanilam

BJ

.

Defective mitochondrial fatty acid oxidation and lipotoxicity in kidney diseases

.

Front Med (Lausanne)

2020

;

7

:

65

.

33

Ma

S

,

Yang

D

,

Li

D

,

Tang

B

,

Yang

Y

.

Oleic acid induces smooth muscle foam cell formation and enhances atherosclerotic lesion development via CD36

.

Lipids Health Dis

2011

;

10

:

53

.

34

Agarwal

P

,

Combes

TW

,

Shojaee-Moradie

F

,

Fielding

B

,

Gordon

S

,

Mizrahi

V

,

Martinez

FO

.

Foam cells control mycobacterium tuberculosis infection

.

Front Microbiol

2020

;

11

:

1394

.

35

Xu

HE

,

Stanley

TB

,

Montana

VG

,

Lambert

MH

,

Shearer

BG

,

Cobb

JE

,

McKee

DD

,

Galardi

CM

,

Plunket

KD

,

Nolte

RT

,

Parks

DJ

,

Moore

JT

,

Kliewer

SA

,

Willson

TM

,

Stimmel

JB

.

Structural basis for antagonist-mediated recruitment of nuclear co-repressors by PPARalpha

.

Nature

2002

;

415

:

813

–

817

.

36

Kojima

Y

,

Weissman

IL

,

Leeper

NJ

.

The role of efferocytosis in atherosclerosis

.

Circulation

2017

;

135

:

476

–

489

.

37

Yurdagul A

J

,

Subramanian

M

,

Wang

X

,

Crown

SB

,

Ilkayeva

OR

,

Darville

L

,

Kolluru

GK

,

Rymond

CC

,

Gerlach

BD

,

Zheng

Z

,

Kuriakose

G

,

Kevil

CG

,

Koomen

JM

,

Cleveland

JL

,

Muoio

DM

,

Tabas

I

.

Macrophage metabolism of apoptotic cell-derived arginine promotes continual efferocytosis and resolution of injury

.

Cell Metab

2020

;

31

:

518

–

533.e10

.

38

Fukuhara

M

,

Geary

RL

,

Diz

DI

,

Gallagher

PE

,

Wilson

JA

,

Glazier

SS

,

Dean

RH

,

Ferrario

CM

.

Angiotensin-converting enzyme expression in human carotid artery atherosclerosis

.

Hypertension

2000

;

35

:

353

–

359

.

39

Danilov

SM

,

Tikhomirova

VE

,

Metzger

R

,

Naperova

IA

,

Bukina

TM

,

Goker-Alpan

O

,

Tayebi

N

,

Gayfullin

NM

,

Schwartz

DE

,

Samokhodskaya

LM

,

Kost

OA

,

Sidransky

E

.

ACE Phenotyping in Gaucher disease

.

Mol Genet Metab

2018

;

123

:

501

–

510

.

40

Zhang

S

,

Weinberg

S

,

DeBerge

M

,

Gainullina

A

,

Schipma

M

,

Kinchen

JM

,

Ben-Sahra

I

,

Gius

DR

,

Yvan-Charvet

L

,

Chandel

NS

,

Schumacker

PT

,

Thorp

EB

.

Efferocytosis fuels requirements of fatty acid oxidation and the electron transport chain to polarize macrophages for tissue repair

.

Cell Metab

2019

;

29

:

443

–

456.e5

.

41

Oram

JF

.

Molecular basis of cholesterol homeostasis: lessons from Tangier disease and ABCA1

.

Trends Mol Med

2002

;

8

:

168

–

173

.

42

van Eck

M

,

Bos

IS

,

Kaminski

WE

,

Orso

E

,

Rothe

G

,

Twisk

J

,

Bottcher

A

,

Van Amersfoort

ES

,

Christiansen-Weber

TA

,

Fung-Leung

WP

,

Van Berkel

TJ

,

Schmitz

G

.

Leukocyte ABCA1 controls susceptibility to atherosclerosis and macrophage recruitment into tissues

.

Proc Natl Acad Sci U S A

2002

;

99

:

6298

–

6303

.

43

Yvan-Charvet

L

,

Ranalletta

M

,

Wang

N

,

Han

S

,

Terasaka

N

,

Li

R

,

Welch

C

,

Tall

AR

.

Combined deficiency of ABCA1 and ABCG1 promotes foam cell accumulation and accelerates atherosclerosis in mice

.

J Clin Invest

2007

;

117

:

3900

–

3908

.

PubMed

44

Out

R

,

Hoekstra

M

,

Habets

K

,

Meurs

I

,

de Waard

V

,

Hildebrand

RB

,

Wang

Y

,

Chimini

G

,

Kuiper

J

,

Van Berkel

TJ

,

Van Eck

M

.

Combined deletion of macrophage ABCA1 and ABCG1 leads to massive lipid accumulation in tissue macrophages and distinct atherosclerosis at relatively low plasma cholesterol levels

.

Arterioscler Thromb Vasc Biol

2008

;

28

:

258

–

264

.

45

Westerterp

M

,

Murphy

AJ

,

Wang

M

,

Pagler

TA

,

Vengrenyuk

Y

,

Kappus

MS

,

Gorman

DJ

,

Nagareddy

PR

,

Zhu

X

,

Abramowicz

S

,

Parks

JS

,

Welch

C

,

Fisher

EA

,

Wang

N

,

Yvan- Charvet

L

,

Tall

AR

.

Deficiency of ATP-binding cassette transporters A1 and G1 in macrophages increases inflammation and accelerates atherosclerosis in mice

.

Circ Res

2013

;

112

:

1456

–

1465

.

46

Wang

J

,

Xiao

Q

,

Wang

L

,

Wang

Y

,

Wang

D

,

Ding

H

.

Role of ABCA1 in cardiovascular disease

.

J Pers Med

2022

;

12

:

1010

.

47

Yu

XH

,

Zheng

XL

,

Tang

CK

.

Peroxisome proliferator-activated receptor alpha in lipid metabolism and atherosclerosis

.

Adv Clin Chem

2015

;

71

:

171

–

203

.

48

Zandbergen

F

,

Plutzky

J

.

PPARalpha in atherosclerosis and inflammation

.

Biochim Biophys Acta

2007

;

1771

:

972

–

982

.

49

Li

AC

,

Binder

CJ

,

Gutierrez

A

,

Brown

KK

,

Plotkin

CR

,

Pattison

JW

,

Valledor

AF

,

Davis

RA

,

Willson

TM

,

Witztum

JL

,

Palinski

W

,

Glass

CK

.

Differential inhibition of macrophage foam-cell formation and atherosclerosis in mice by PPARalpha, beta/delta, and gamma

.

J Clin Invest

2004

;

114

:

1564

–

1576

.

50

Martin-Nizard

F

,

Furman

C

,

Delerive

P

,

Kandoussi

A

,

Fruchart

JC

,

Staels

B

,

Duriez

P

.

Peroxisome proliferator-activated receptor activators inhibit oxidized low-density lipoprotein-induced endothelin-1 secretion in endothelial cells

.

J Cardiovasc Pharmacol

2002

;

40

:

822

–

831

.

51

Kim

YS

,

Lee

HM

,

Kim

JK

,

Yang

CS

,

Kim

TS

,

Jung

M

,

Jin

HS

,

Kim

S

,

Jang

J

,

Oh

GT

,

Kim

JM

,

Jo

EK

.

PPAR-alpha activation mediates innate host defense through induction of TFEB and lipid catabolism

.

J Immunol

2017

;

198

:

3283

–

3295

.

52

Kim

NH

,

Kim

SG

.

Fibrates revisited: potential role in cardiovascular risk reduction

.

Diabetes Metab J

2020

;

44

:

213

–

221

.

53

Rosenson

RS

.

Field of confusion: future prospects for fibrate therapy in cardiovascular disease

.

Curr Atheroscler Rep

2006

;

8

:

219

–

222

.

54

Alkatiri

AH

,

Firman

D

,

Alkatiri

AA

,

Suryajaya

PI

,

Sudharsono

A

.

The role of angiotensin antagonism in coronary plaque regression: insights from the glagovian model

.

Int J Vasc Med

2021

;

2021

:

8887248

.