Myeloid β-catenin deficiency exacerbates atherosclerosis in low-density lipoprotein receptor-deficient mice

Abstract

Objective

The Wnt/β-catenin signaling is an ancient and evolutionarily conserved pathway that regulates essential aspects of cell differentiation, proliferation, migration and polarity. Canonical Wnt/β-catenin signaling has also been implicated in the pathogenesis of atherosclerosis. Macrophage is one of the major cell types involved in the initiation and progression of atherosclerosis, but the role of macrophage β-catenin in atherosclerosis remains elusive. This study aims to investigate the impact of β-catenin expression on macrophage functions and atherosclerosis development.

Approach and Results

To investigate the role of macrophage canonical Wnt/β-catenin signaling in atherogenesis, we generated low-density lipoprotein receptor-deficient mice with myeloid-specific β-catenin deficiency (β-catenin^ΔmyeLDLR^−/−). As expected, deletion of β-catenin decreased macrophage adhesion and migration properties in vitro. However, deficiency of β-catenin significantly increased atherosclerotic lesion areas in the aortic root of LDLR^−/− mice without affecting the plasma lipid levels and atherosclerotic plaque composition. Mechanistic studies revealed that β-catenin can regulate activation of signal transducer and activator of transcription (STAT) pathway in macrophages, and ablation of β-catenin resulted in STAT3 downregulation and STAT1 activation, leading to elevated macrophage inflammatory responses and increased atherosclerosis.

Conclusions

This study demonstrates a critical role of myeloid β-catenin expression in atherosclerosis by modulating macrophage inflammatory responses.

Introduction

The Wnt signaling pathway plays essential roles in the regulation of cell proliferation, migration, and differentiation^{1, 2}. Wnt signaling is traditionally classified into the canonical β-catenin-dependent pathway (Wnt/β-catenin pathway) and non-canonical β-catenin-independent pathway². For the canonical Wnt/β-catenin pathway, Wnt ligands bind to the membrane-spanning receptor proteins Frizzled and low-density lipoprotein receptor-related protein (LRP) family, and subsequently trigger the phosphorylation of the Dishevelled protein, which further activates a signaling cascade that results in the stabilization and nuclear localization of β-catenin². Nuclear localized β-catenin interacts with the T cell-specific factor/lymphoid enhancer-binding factor to induce the transcription of canonical Wnt signaling target genes. In the absence of Wnt ligands, several kinases including glycogen synthase kinase-3beta (GSK3β) and priming kinase casein kinase Iα (CKIα) can phosphorylate the degron motif of β-catenin to prime it for β-TrCP-mediated ubiquitination and degradation^{2, 3}.

Canonical Wnt/β-catenin signaling has been implicated in the pathogenesis of human atherosclerosis⁴. For example, the levels of active β-catenin were increased in disrupted atherosclerotic plaques as compared with stable plaques⁵. The expression levels of LRP6 were also reduced in carotid atherosclerotic lesions of human LRP6 1062V variant carriers, and 1062V variant of LRP6 and carotid atherosclerosis were strongly associated in hypertensive patients, indicating that reduced activation of canonical Wnt signaling may contribute to the increased atherogenesis⁶. Animal studies have also identified the activation of β-catenin in vascular endothelium before and during early atherogenesis⁷. Activation of β-catenin can increase vascular smooth muscle cell (SMC) proliferation and migration, and has been suggested as a key component of atherosclerotic physiology^8–11. By contrast, Wnt signaling inhibitor Sclerostin inhibited angiotensin II-induced aortic aneurysm and atherosclerosis formation in apolipoprotein E-deficient (ApoE^−/−) mice¹². However, other studies found that excess plasma lipid can alter LRP5 expression in vessel wall¹³ and LRP5 deficiency can down-regulate Wnt signaling and promote aortic lipid infiltration in mice¹⁴, suggesting that canonical Wnt signaling may exert a protective defense mechanism against atherosclerosis development.

Macrophage accumulation within the vascular wall is a hallmark of atherosclerosis, yet the role of macrophage Wnt/β-catenin signaling in atherosclerosis remains elusive. Wnt/β-catenin pathway has been known to regulate macrophage adhesion and migration properties¹⁵ which play an important role in the initiation and progression of atherosclerosis¹⁶. Activation of monocyte Wnt/β-catenin signaling increased its adhesion to endothelial cells¹⁷. By contrast, myeloid-derived β-catenin deletion decreased the migration and adhesion ability of macrophages and impaired the wound-healing process¹⁵. Thus, macrophage Wnt/β-catenin signaling may contribute to atherogenesis by promoting the cell migration and adhesion. However, inflammatory responses are also the driving force of atherosclerosis development, and macrophages are the major inflammatory cells involved in the progression of atherosclerosis^{16, 18}. Wnt/β-catenin signaling has been demonstrated to regulate both pro-inflammatory^19–21 and anti-inflammatory^22–24 responses in various cell types depending on the context or environment. For example, activation of canonical Wnt/β-catenin by Wnt5a and Wnt3a treatment significantly induced pro-inflammatory cytokine production secretion in murine macrophages²⁵. However, knockdown of β-catenin also increased the pro-inflammatory cytokine IL-6 expression in RAW264.7 macrophages, indicating the anti-inflammatory effects of β-catenin in those cells²⁶. Therefore, the functions of β-catenin in the regulation of macrophage functions related to atherosclerosis are complex. To our knowledge, the impact of macrophage β-catenin expression on atherosclerosis development in appropriate animal models has not been reported.

To explore the role of macrophage-derived β-catenin in atherogenesis, we developed a low-density lipoprotein receptor-deficient mouse model with myeloid-specific β-catenin deficiency (β-catenin^ΔmyeLDLR^−/−). Our results demonstrate that β-catenin deficiency exacerbate atherosclerosis in LDLR^−/− mice by elevating macrophage inflammatory responses.

Materials and Methods

Animals

Myeloid-specific β-catenin knockout (β-catenin^ΔMye) mice were generated by crossing mice carrying loxP-flanked β-catenin alleles (β-catenin^F/F)²⁷ with LysM-Cre transgenic mice²⁸. To increase susceptibility to atherosclerotic lesion development, the β-catenin^ΔMye mice were crossed to LDLR^−/− mice (The Jackson Laboratory) to generate β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice. All mice used in this study had β-catenin^F/FLDLR^−/− double-mutant background, and β-catenin^ΔMyeLDLR^−/− mice carried heterozygous knock-in for LysM-Cre. For atherosclerosis study, 4-week-old male β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− littermates were fed a high-fat Western diet (WD) (21.2% fat, 0.2% cholesterol; TD88137, Harlan Teklad) for 12 weeks until euthanization at 16 weeks old. Body weight was measured weekly and body composition was measured by EchoMRI (Echo Medical System). Intraperitoneal glucose tolerance test (GTT) was performed as described previously²⁹. All experimental mice used in this study were male, partially due to the crosstalk between Wnt/β-catenin and estrogen signaling^{30, 31}. However, studying a single sex has limitations since sex differences have been widely reported in mouse atherosclerosis studies³². All the procedures were approved by the University of Kentucky Institutional Animal Care and Use Committee.

Blood Analysis

Plasma total cholesterol and triglyceride concentrations were determined enzymatically by a colorimetric method¹⁸. Plasma from multiple mice (n=6) was pooled, and plasma lipoprotein cholesterol distributions were determined by fast-performance liquid chromatography (FPLC) ³³. Plasma cytokine levels were determined by a mouse cytokine multiplex assay kit and a BioPlex 200 system (Bio-Rad Laboratories)³⁴.

Atherosclerosis lesion analyses

The atherosclerosis study and analysis were performed following the American Heart Association statement on atherosclerosis³⁵. Optimum cutting temperature(OCT) compound-embedded hearts were sectioned and stained with oil red O (Sigma, O0625-100G), and atherosclerotic lesion sizes at aortic roots were quantified as previously described^{18, 29, 36}. The aortic root sections were also stained with hematoxylin & eosin (H&E) and necrotic core size were quantified³⁷. Lesional collagen content and SMC composition were determined with Masson’s trichrome staining and quantified by using image pro plus software as previously described³⁸. Immunohistochemical staining of atherosclerotic lesions were performed on 12-μm sections of aortic roots embedded in OCT^{18, 39}. Sections were first fixed in 100% ice-cold acetone for 15 min and then washed with PBS for 20 min. Sections were permeabilized with PBS + 0.1% Triton X-100 (PBST) for 10 min. Nonspecific binding was reduced by incubating slides in 10% rabbit sera diluted in PBST for 20 min at room temperature. Sections were then incubated with antibodies against MCP-1 (Abcam, ab7202), TNFα (Abcam, ab6671), STAT3, Ki67 (eBioscience, 465698), Ly6G(1A8) (BD Bioscience, 551459) at 4°C for 12–15 h. Sections were rinsed with PBS and incubated with fluorescein-labeled secondary antibodies (Invitrogen, A11072). The nuclei were stained by mounting the slides with DAPI medium (Vector Laboratories, H-1200). Normal IgG antibodies were used as negative controls. Dual staining of MCP-1, TNFα, STAT3 and MOMA2 (Bio-Rad, MCA519a) were performed to demonstrate the co-localization of these factors and macrophages. For p65 nuclear translocation analysis, bone marrow-derived macrophages (BMM) were seeded on to 4-well chamber, after treated with LPS, cells were fixed with 4% PFA at room temperature for 15 min, then washed with PBS for 15 min. Nonspecific binding was reduced by incubating cells with 10% rabbit sera diluted in PBST for 20 min at room temperature. Cells were then incubated with antibody against p65 at 4°C for 12–15 h. Sections were rinsed with PBS and incubated with fluorescein-labeled secondary antibodies (Invitrogen, A11072). The nuclei were stained by mounting the slides with DAPI medium. Images were acquired with a Nikon fluorescence microscopy (Nikon).

Macrophage isolation and function assays

BMM and peritoneal macrophages (PM) were isolated as previously described⁴⁰. Scratch wound healing assays were performed using 200 μl pipette tip to make a straight scratch on the 100% confluent BMM cell plate. Cells were starved for 24 hours and washed with PBS twice before scratching assay. During the assay culture medium was replaced with MEM containing 0.5% FBS to prevent proliferation^{41, 42}. Images were taken at day 0 and day 1 after scratching. For adhesion assay, calcein acetoxymethyl–labeled PM were incubated with primary porcine endothelial cells, and attached cells were fixed and counted⁴³. Migration assays were performed using transwells with 8.0-μm pore polycarbonate membrane inserts (Fisher Scientific, 07-200-165)⁴⁴. 200μL matrigel (Corning, 356327, final concentration: 300μg/ml) was added to a 24-well transwell insert and solidified in a 37 °C incubator for 2 hours to form a thin gel layer. Macrophages were seeded on top of the matrigel coated transwell filters, and the lower chambers were filled with non-FBS MEM media supplemented with or without 500ng/ml LPS (Sigma, L4391). After 24 hours, cells were removed from the upper surface of the insert by scraping using Q-Tips. The membranes were fixed with 100% cold methanol (Fisher Scientific, A412-4), stained with hematoxylin (Leica, 3801575) and mounted on the slides using glycerol gelatin. Hematoxylin-stained cells were counted under the microscope.

In vitro cell culture and treatment

The murine monocyte/macrophage cell line RAW264.7 was obtained from American Type Culture Collection and maintained in 10% FBS containing MEM medium supplemented with penicillin and streptomycin (Invitrogen, 15140122). The Clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9) method was used to delete β-catenin in RAW264.7 cells as we previously described⁴⁵. Briefly, the lentiCRISPR plasmid (Addgene plasmid #52961), containing hCas9 and single-guide RNA (sgRNA), was digested with BbsI, and a pair of annealed oligonucleotides for β-catenin (5′-CACCGAGCCAAGCGCTGGACATTAG-3′) for targeting site was cloned into plasmid. The Transgenomic Surveyor Mutation Detection Kit (Thermo Fisher Scientific #706020, Waltham, MA, USA) and genomic DNA sequencing were used to verify genomic mutations within the locus of β-catenin. To study the effects of Wnt3a treatment on STAT3 activation and target gene expression, 100ng/ml Wnt3a recombinant protein (R&D system, 5036-WN-010) were added to serum-free cultured RAW264.7 cells and BMM for 6 hr before analysis. To study the impact of β-catenin ablation on the nuclear translocation of p65, 500ng/ml LPS were added to the serum-free cultured BMM cells for 30 min before analysis. To knockdown STAT3 in macrophages, 100pmol siRNA targeting STAT3 (Sigma-Aldrich, SASI_Mm01_00041176) were transfected into BMM using Fugene 6 transfection reagent (VWR, E2692) according to manufacturer’s protocol. After transfection, cells were cultured for 48 hours before analysis.

RNA isolation and QPCR analysis

Total RNA was isolated from mouse tissues or cells using TRIzol Reagent (Life Technologies, 15596018), and QPCR were performed using gene-specific primers and SYBR Green PCR kit (BioRad, 1725125) as previously described⁴⁵. The sequences of primer sets used in this study are listed in Table in the online-only Data Supplement.

Immunoprecipitation (IP) and immunoblotting analysis

Proteins were isolated from cells or mouse tissues by homogenization in RIPA buffer with complete mini protease inhibitor cocktail (Roche, 11836170001). Protein concentrations were determined by the Pierce BCA protein assay kit (Thermo Fisher Scientific, 23225). 300 μg peritoneal macrophage proteins from three identical genotype mice pooled per group per IP, proteinase inhibitor cocktail, Protein A agarose beads (Roche, 11134515001), anti-STAT1 antibody (Cell Signaling Technology, 14994T) and Rabbit normal IgG (Cell Signaling Technology, 2729P) were used. For immunoblotting, equal amount of proteins (20 or 30 μg) from peritoneal macrophage per mouse were resolved by SDS-PAGE and transferred onto Immunobilon-P membranes (Millipore, IPVH00010). Precision Plus Protein Standards (Bio-Rad Laboratories, 161-0394) were loaded into one lane of the gel. Membranes were incubated in 5% nonfat milk for 45 min and then were incubated for 18 h at 4°C with the following primary antibodies in 5% nonfat milk: anti-IL-1β (Abcam, ab9722), anti-IL-6 (Biorad, MCA1490), anti-MCP1 (Abcam, ab7202), anti-phospho-STAT1 (9167S), anti-phospho-STAT3 (9145s), anti-STAT3 (12640S), anti-STAT1 (Cell Signaling Technology), anti-phospho-p65 (Cell Signaling Technology, 3033P), anti-p65 (Santa Cruz, SC-372), anti-β-catenin (Sigma, C2206) and anti-actin (Sigma, A2066). Detailed information of antibodies please see the Major Resources Table in the Supplemental Material. Signals were detected using the Pierce^™ ECL Western Blotting Substrate (32106).

Statistical analysis

All data are presented as the mean ± SEM. Statistically significant differences between 2 groups were analyzed by 2-tailed Student t test for data normally distributed and by the Mann-Whitney test for data not normally distributed using Prism software. A p value <0.05 was considered significant.

Results

Generation of LDLR^−/− mice with myeloid-specific β-catenin deficiency

To investigate the role of β-catenin in macrophage functions, we generated myeloid-specific β-catenin knockout mice (β-catenin^ΔMye) by breeding β-catenin flox mice (β-catenin^F/F)²⁷ with LysMCre transgenic mice^{18, 28}. To increase susceptibility to atherosclerotic lesion development, β-catenin^ΔMye mice were further crossed with LDLR^−/− mice to generate β-catenin^ΔMyeLDLR^−/− mice. PCR analysis of genomic DNA indicates that the recombination was specific to the PM and BMM of β-catenin^ΔMyeLDLR^−/− (Figure I in the online-only Data Supplement). As expected, the mRNA levels of β-catenin were significantly decreased in both PM and BMM but not in other major tissues of β-catenin^ΔMyeLDLR^−/− mice as compared with β-catenin^F/FLDLR^−/− mice (Figure 1A). Consistently, β-catenin protein levels were also substantially reduced in both PM and BMM of β-catenin^ΔMyeLDLR^−/− mice (Figure 1B). These results demonstrated the specific and efficient β-catenin deletion in macrophage of β-catenin^ΔMyeLDLR^−/− mice.

Figure 1. Generation of LDLR^−/− mice with myeloid-specific β-catenin deficiency.

A, mRNA levels of β-catenin in major tissues and macrophages of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (** p<0.01, n=5). B, Protein levels of β-catenin in heart, liver, spleen, kidney and macrophages of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice. C, Macrophage migration assay. PM isolated from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice were seeded on the matrigel-coated transwell filters for 24 hours. Cells that infiltrated and migrated to the underside of transwell were stained with hematoxylin and counted under the microscope (**p<0.01, n=5) (bar=200μm). D, Macrophage wound healing assay. BMM collected from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice underwent in vitro scratch wound healing assay. Representative images of each group were captured at day 0 and 1 after scratching. Number of cells migrated to the scratch gap was calculated and quantified (**p<0.01, n=5) (bar=500μm).

Canonical Wnt/β-catenin signaling has been well-established to play an important role in the regulation of cell migration^{15, 46, 47}. To determine the impact of deficiency of β-catenin on macrophage migration properties, we performed both transwell migration and scratch wound healing assays. As expected, β-catenin deficiency significantly reduced migration ability of PM as determined by transwell assays (Figure 1C). Scratch wound healing assays also demonstrated that deletion of β-catenin significantly decreases the number of migrated BMM into the scratching gap (Figure 1D). Next, we also incubated freshly isolated PM with primary endothelia cells and found that macrophages from β-catenin^ΔMyeLDLR^−/− mice had reduced adhesion ability to endothelia cells (Figure IIA in the online-only Data Supplement). β-Catenin has been shown to positively regulate the expression of genes that mediate macrophage migration and adhesion ¹⁵. Consistently, we also found that deficiency of β-catenin significantly reduced the expression of those genes including Itga4, Adam9, and Cdh1 (Figure IIB in the online-only Data Supplement). Taken together, these results confirmed that ablation of β-catenin can indeed reduce macrophages adhesion and migration abilities.

Deficiency of β-catenin in macrophages does not affect plasma lipid levels but increases atherosclerosis in LDLR^−/− mice

To determine the role of macrophage β-catenin in atherosclerosis development, 4-week-old male β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− littermates were fed with WD for 12 weeks. At 16 weeks of age, β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice had similar body weight, fat mass and lean mass (Figure IIIA–B in the online-only Data Supplement). Deficiency of myeloid β-catenin did not affect the fasting glucose levels and glucose tolerance of β-catenin^ΔMyeLDLR^−/− mice (Figure IIIC–D in the online-only Data Supplement). Both β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice had diet-induced hyperlipidemia, and myeloid β-catenin deletion did not alter plasma total cholesterol and triglyceride levels (Figure 2A). Further, FPLC analysis also showed similar cholesterol distribution pattern between β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice (Figure 2B).

Figure 2. Deficiency of myeloid β-catenin increases atherosclerosis of LDLR^−/− mice without affecting lipid profile.

Four-week-old male β-catenin^F/FLDLR^−/− and β-Catenin^ΔMyeLDLR^−/− littermates were fed a Western diet for 12 weeks. (A and B) The plasma levels of total cholesterol and triglyceride (A) were measured by standard method, and plasma cholesterol distribution was analyzed by fast-performance liquid chromatography (B). C, Quantitative analysis of atherosclerotic lesion areas in the aortic root of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (n=10–12 per group, *p<0.05). Representative Oil-red-O–stained sections from each genotypes are displayed next to the quantification data (bar=500μm). D, Quantitative analysis of necrotic core areas in the aortic root of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (n=10–12 per group). Representative images of hematoxylin & eosin staining at aortic root sections from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice are displayed next to the quantification data (bar=500μm).

Although deficiency of myeloid β-catenin did not affect metabolic phenotypes and plasma lipid levels, quantification of cross-sectional lesion areas at the aortic root revealed that β-catenin^ΔMyeLDLR^−/− mice had 26% increased lesion sizes (286590.4 ± 43656.63 μm²) as compared with β-catenin^F/FLDLR^−/− littermates (227247.1 ± 26061.89 μm²) (p=0.013) (Figure 2C). Immunostaining results revealed that deficiency of β-catenin did not significantly affect the proportion of macrophage content in atherosclerotic lesions of β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice (Figure IVA in the online-only Data Supplement). Further, Masson’s trichrome staining also demonstrated that the plaque composition was similar between β-catenin^ΔmyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice (Figure IVB in the online-only Data Supplement). Although β-catenin signaling can regulate cell proliferation, β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice had similar proliferating Ki67-positive cells in the lesions (Figure V in the online-only Data Supplement). In addition, deficiency of β-catenin did not affect the proportion of necrotic core areas in atherosclerotic lesions of β-catenin^ΔMyeLDLR^−/− mice (Figure 2D). Since LysM is also expressed by neutrophils^{48, 49} that contribute to the atherosclerotic lesion formation⁵⁰, we also performed immunofluorescence staining with antibodies against neutrophil marker Ly6G(1A8) and found that LysM-Cre-mediated β-catenin deletion did not affected lesional neutrophil numbers (Figure VI in the online-only Data Supplement). Taken together, deletion of β-catenin in macrophages exacerbates diet-induced atherosclerosis in LDLR^−/− mice without altering plasma lipid level and plaque composition.

Ablation of β-catenin elevates macrophage inflammatory responses without affecting NF-κB activities

Atherosclerosis is an inflammatory disease and macrophages are the major inflammatory cells contributing to atherosclerotic lesion formation and progression^{16, 51}. We next investigated whether deficiency of β-catenin affected macrophage inflammatory responses in β-catenin^ΔMyeLDLR^−/− mice. QPCR analyses demonstrated that the mRNA levels of pro-inflammatory cytokines including IL-6 and TNFα were significantly increased in freshly isolated PM of β-catenin^ΔMyeLDLR^−/− mice (Figure 3A). However, the expression levels of anti-inflammatory genes were not affected by β-catenin deficiency (Figure VII in the online-only Data Supplement). In addition, protein levels of IL-1β, MCP1 and IL-6 were also elevated in PM of β-catenin^ΔMyeLDLR^−/− mice as compared with littermate controls (Figure 3B). Consistent with gene and protein expression analyses, immunofluorescence staining showed that the expression of key inflammatory cytokines, TNFα and MCP1, was increased in the atherosclerotic lesional macrophages of β-catenin^ΔMyeLDLR^−/− mice (Figure 3C and Figure VIII in the online-only Data Supplement). We next measured plasma cytokine levels to determine whether β-catenin^ΔMyeLDLR^−/− mice have increased systemic inflammation. Indeed, the plasma levels of pro-inflammatory cytokines including TNFα, IL-6, and IFNγ were significantly increased in β-catenin^ΔMyeLDLR^−/− mice (Figure 3D). Taken together, deficiency of β-catenin elevated macrophage inflammatory responses and increased lesional and systemic inflammation in β-catenin^ΔMyeLDLR^−/− mice.

Figure 3. Ablation of myeloid β-catenin increases macrophage and systemic inflammation in β-catenin^ΔMyeLDLR^−/− mice.

A, QPCR analysis of pro-inflammatory cytokine expression in PM of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (** p<0.01, n=8). B, Immunoblotting images and quantification data of pro-inflammatory cytokines in PM of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (* p<0.05, ** p<0.01, n=3). C, Representative images and quantification data of TNFα and MCP1 immunofluorescent staining in the aortic root of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (**p<0.01, n=6) (bar=100 μm). D, Plasma protein levels of indicated cytokines from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (**p<0.01, n=11–14).

Wnt/β-catenin signaling has been previously shown to repress NF-κB activity in several cell types^{52, 53}. Since NF-κB plays an important role in mediating macrophage inflammatory responses during atherosclerosis development¹⁸, we then investigated whether β-catenin deletion can affect NF-κB activities in macrophages. The total active form of NF-κB subunit p65 (phosphorylated p65) were not affected by β-catenin deletion in PM of β-catenin^ΔMyeLDLR^−/− mice (Figure IXA in the online-only Data Supplement), indicating the unchanged NF-κB activity in those cells. We also isolated BMM from β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice and treated them with LPS to stimulate the nuclear translocation of p65. Consistently, deficiency of β-catenin did not affect LPS-elicited p65 translocation as determined by both immunoblotting and immunofluorescence staining (Figure IXB–C in the online-only Data Supplement). Therefore, the increased inflammatory response of macrophage from β-catenin^ΔMyeLDLR^−/− mice was unlikely through NF-κB signaling.

β-Catenin regulates JAK-STAT signaling and inflammatory responses in macrophages

One of the downstream target genes of canonical Wnt/β-catenin signaling is signal transducer and activator of transcription 3 (STAT3)⁵⁴ which is a part of the Janus kinase (JAK)-STAT signaling, another central pathway controlling the expression of pro-inflammatory and anti-inflammatory cytokines in macrophages⁵⁵. Canonical Wnt/β-catenin signaling has been reported to positively regulate STAT3 signaling and to repress inflammatory responses in several different murine and human cell types^{23, 44, 54, 56, 57}. To investigate whether the STAT signaling contributes to the increased inflammatory responses in β-catenin-depleted macrophages, PM were isolated from β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice, and the expression levels of STAT1, STAT3 and the downstream target genes⁵⁸ were analyzed. Consistent with previous studies^{44, 54, 59}, deficiency of β-catenin significantly reduced the mRNA levels of STAT3 and its targets such as Cited2, Sbno2, BCL3 and Nfil3, but did not affect the expression of STAT1 (Figure 4A). The protein levels of STAT3 were also substantially decreased in β-catenin-deficient macrophages (Figure 4B). Immunofluorescence staining confirmed the decreased STAT3 expression in the lesions of β-catenin^ΔMyeLDLR^−/− mice (Figure 4C).

Figure 4. β-Catenin deficiency affects STAT3 expression and STAT1 activity in macrophages.

A, QPCR analysis of STAT3, STAT1, and STAT3 target genes in PM of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (**p<0.01, n=8). B, Immunoblotting of STAT3, STAT1, phosphorylated STAT1 and phosphorylated STAT3 in PM. C, Representative images and quantification data of STAT3 immunofluorescent staining in the aortic root of β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice (*p<0.05, n=5) (bar=100 μm). D, PM isolated from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice were lysed and immunoprecipitated with antibodies against STAT1 and IgG. Total STAT1, phosphorylated STAT1 and STAT3 proteins were further immunoblotted with specific antibodies. 10% cell lysate were used as input for immunoblotting.

STAT3 has been demonstrated to suppress the activation of STAT1 by forming STAT3/STAT1 heterodimer and inhibiting its tyrosine phosphorylation in a cellular context- and signal-dependent manner^60–62. Indeed, β-catenin-deficient macrophages had unchanged total STAT1 protein levels but increased STAT1 phosphorylation(Figure 4B). Co-immunoprecipitation assays also confirmed a direct interaction between STAT1 and STAT3 in control macrophages but the interaction was substantially reduced in β-catenin-deficient macrophages (Figure 4D). Further, phosphorylated STAT1 proteins were increased in the protein complex pulled down by anti-STAT1 antibodies in β-catenin-deficient macrophages (Figure 4D). In addition, siRNA-mediated STAT3 knockdown also led to increased STAT1 phosphorylation without affecting the total STAT1 protein levels (Figure X in the online-only Data Supplement). Collectively, these results suggest that decreased STAT3 levels may contribute to the activation of STAT1 and increased inflammatory responses in β-catenin-deficient macrophages.

To further explore the understudied functions of canonical Wnt/β-catenin signaling in the regulation of macrophage inflammatory responses, BMM isolated from β-catenin^ΔMyeLDLR^−/− and β-catenin^F/FLDLR^−/− mice were treated with Wnt3a that activates canonical Wnt/β-catenin signaling⁴⁵. Wnt3a treatment significantly upregulated STAT3 and its downstream genes in control macrophages but not in β-catenin-deficient macrophages (Figure 5A). As expected, increased STAT3 expression was associated with decreased pro-inflammatory cytokine expression in control macrophages but deficiency of β-catenin abolished Wnt3a-elicited downregulation of inflammatory genes(Figure 5B). In addition to primary macrophages, we also used CRISPR/Cas9 approach to generate β-catenin-deficient murine RAW264.7 macrophage cell line (Figure XI in the in the online-only Data Supplement). As shown in Figure 5C and 5D, CRISPR/Cas9-mediated β-catenin deletion also inhibited Wnt3a-stimulated upregulation of STAT3 and inflammatory responses in RAW264.7 cells. Taken together, these results demonstrated an important role of β-catenin-STAT3/STAT1 signaling cascade in the regulation of macrophage inflammatory responses, which likely contributes to the increased atherosclerosis in β-catenin^ΔMyeLDLR^−/− mice.

Figure 5. Deletion of β-catenin abolishes the impact of Wnt3a on STAT3 and pro-inflammatory cytokine expression in macrophages.

A and B, QPCR analysis of STAT3, STAT1, and STAT3 target genes (A), and proinflammatory genes (B) in BMM isolated from β-catenin^F/FLDLR^−/− and β-catenin^ΔMyeLDLR^−/− mice treated with vehicle control or Wnt 3a (*p<0.05, **p<0.01, n=8). C and D, QPCR analysis of STAT3, STAT1, and STAT3 target genes (C) and pro-inflammatory genes (D) in control or β-catenin-deficient RAW264.7 macrophages treated with vehicle control or Wnt3a (*p<0.05, **p<0.01, n=8).

Discussion

Emerging evidence has demonstrated the potential role of Wnt/β-catenin signaling in atherogenesis by regulating endothelial functions^{7, 63}, and SMC proliferation and migration^9–11. Myeloid-derived β-catenin has also been shown to promote macrophage migration and adhesion¹⁵. Therefore, it has been suggested that attenuation of Wnt/β-catenin signaling may have beneficial effects for the cardiovascular system^{12, 64, 65}. However, there have been no studies on the impact of macrophage β-catenin expression on atherosclerosis development in hyperlipidemic mouse models. In the current study, we generated myeloid-specific β-catenin-deficient LDLR^−/− mice to investigate the role of macrophage β-catenin in atherogenesis. Consistent with previous studies¹⁵, we found that deficiency of β-catenin decreased macrophage migration and adhesion abilities. Therefore, one would expect that deficiency of myeloid β-catenin should decrease atherosclerosis development in vivo. Unexpectedly, atherosclerosis analyses revealed that β-catenin^ΔMyeLDLR^−/− mice had increased rather than decreased aortic root atherosclerotic lesion sizes as compared with control β-catenin^F/FLDLR^−/− mice. Interestingly, deletion of β-catenin increased macrophage inflammatory responses and elevated plasma pro-inflammatory cytokine levels, which may lead to the increased atherosclerosis. To our knowledge, this study is the first to demonstrate the impact of myeloid β-catenin expression on the atherosclerosis development in an appropriate small animal model.

Atherosclerosis is a chronic inflammatory disease, and the key regulator of the innate and adaptive immune responses, NF-κB signaling has been implicated in the atherosclerosis development^{16, 18, 66, 67}. We previously demonstrated that inhibition of myeloid NF-κB activation by deletion of IκB kinase β (IKKβ) attenuated macrophage inflammatory responses and decreased atherosclerosis development in LDLR^−/− mice¹⁸. In the current study, deficiency of β-catenin did not affect NF-κB translocation and activation in macrophages, which cannot explain the elevated inflammatory responses in those cells. In addition to NF-κB signaling, the JAK-STAT signaling is another key pathway regulating cellular inflammatory responses⁶⁸, and canonical Wnt/β-catenin signaling can transcriptionally regulate STAT3 expression^{54, 56}. Indeed, we found that the expression of STAT3 were decreased in both macrophages and atherosclerotic lesions of β-catenin^ΔMyeLDLR^−/− mice. The balance between activation of STAT1 and STAT3 coordinately regulates macrophage polarization and inflammation in responding to different stimuli^{69, 70}. While activation of STAT1 promotes macrophage pro-inflammatory responses, activation of STAT3 results in suppressed immune responses^{57, 69–72}. Further, studies have also demonstrated that STAT3 can attenuate STAT1-mediated pro-inflammatory cytokine expression through directly interaction with STAT1 to suppress its phosphorylation or the formation of DNA binding STAT1 homodimers^{60, 61, 69}. Consistent with those studies, we also detected a direct interaction between STAT3 and STAT1 in macrophages, and decreased STAT3 protein levels led to increased STAT1 phosphorylation without affecting total STAT1 protein levels in β-catenin-deficient macrophages. Further, the impact of Wnt3a treatment on STAT3 and inflammatory gene expression was abolished by β-catenin deficiency. Therefore, it is very likely that deficiency of β-catenin increases macrophage inflammatory responses through modulation of STAT3/STAT1 signaling cascade.

The role of Wnt/β-catenin signaling in the regulation of inflammatory responses in different cell types has not been completely understood. For example, some studies demonstrated that β-catenin is a negative regulator of inflammation in certain immune cells such as dendritic cells^{23, 24}. By contrast, other studies suggested that β-catenin can increase inflammatory cytokine production in different cell types^{19, 20}. Thus, the functions of β-catenin in immunity are complex and β-catenin may have both pro-inflammatory and anti-inflammatory effects depending on the cell type, stimulus, and cellular environment. Nevertheless, our studies demonstrated that canonical Wnt/β-catenin signaling has anti-inflammatory effects in macrophages of WD-fed LDLR^−/− mice, and deficiency of β-catenin led to increased macrophage inflammatory responses and systemic inflammation. Future studies will be required to determine the cell type-specific functions of Wnt signaling in inflammation and the detailed underlying mechanisms.

In summary, we demonstrate that β-catenin can regulate multiple macrophage functions related to atherosclerosis, and myeloid-specific β-catenin deficiency increased diet-induced atherosclerosis at aortic root of LDLR^−/− mice. Although deletion of β-catenin decreased macrophage adhesion and migration properties, β-catenin deficiency increased macrophage inflammatory responses, leading to exacerbated atherosclerosis in LDLR^−/− mice. Wnt/β-catenin signaling is an evolutionarily conserved pathway and findings from this study will hopefully stimulate further investigations of the function of Wnt/β-catenin signaling in atherogenesis.

Highlights.

• Canonical Wnt/β-catenin signaling has both pro-atherogenic and anti-atherogenic effects in macrophages.

• Deficiency of myeloid β-catenin decreases macrophage adhesion and migration properties but increases atherosclerosis in LDLR^−/− mice.

• Ablation of β-catenin elevates macrophage inflammatory responses and systemic inflammation in LDLR^−/− mice.

• β-Catenin regulates STAT3/STAT1 signaling cascade in macrophages and deficiency of β-catenin results in elevated STAT1 activation and increased macrophage inflammation.

Abbreviations

LDLR

low-density lipoprotein receptor

LRP

low-density lipoprotein receptor-related protein

PM

peritoneal macrophage

BMM

bone marrow-derived macrophage

SMC

smooth muscle cell

WD

Western diet

GTT

glucose tolerance test

WAT

white adipose tissue

BAT

brown adipose tissue

H&E

hematoxylin & eosin

OCT

optimum cutting temperature

IP

immunoprecipitation

QPCR

quantitative real-time PCR

FPLC

fast-performance liquid chromatography analysis

STAT

signal transducer and activator of transcription