Attenuated Atherosclerotic Lesions in apoE-Fcγ–Chain-Deficient Hyperlipidemic Mouse Model Is Associated with Inhibition of Th17 Cells and Promotion of Regulatory T Cells

Though the presence of antioxidized low-density lipoprotein IgG is well documented in clinical and animal studies, the role for FcγRs to the progression of atherosclerosis has not been studied in detail. In the current study, we investigated the role for activating FcγR in the progression of atherosclerosis using apolipoprotein E (apoE)-Fcγ-chain double-knockout (DKO) mice. Relative to apoE knockout (KO) mice, arterial lesion formation was significantly decreased in apoE-Fcγ-chain DKO mice. Bone marrow chimera studies showed reduced lesions in apoE KO mice receiving the bone marrow of apoE-Fcγ-chain DKO mice. Compared to apoE KO mice, antioxidized low-density lipoprotein IgG1 (Th2) and IgG2a (Th1), IL-10, and IFN-γ secretion by activated T cells was increased in apoE-Fcγ-chain DKO mice. These findings suggest that reduced atherosclerotic lesion in apoE-Fcγ-chain DKO mice is not due to a Th1/Th2 imbalance. Interestingly, the number of Th17 cells and the secretion of IL-17 by activated CD4+ cells were decreased in apoE-Fcγ-chain DKO mice. Notably, the number of regulatory T cells, expression of mRNA, and secretion of TGF-β and IL-10 were increased in apoE-Fcγ-chain DKO mice. Furthermore, secretions of IL-6 and STAT-3 phosphorylation essential for Th17 cell genesis were reduced in apoE-Fcγ-chain DKO mice. Importantly, decrease in Th17 cells in apoE-Fcγ-chain DKO mice was due to reduced IL-6 release by APC of apoE-Fcγ-chain DKO mice. Collectively, our data suggest that activating FcγR promotes atherosclerosis by inducing a Th17 response in the hyperlipidemic apoE KO mouse model.

O ne of the risk factors implicated in the pathogenesis of atherogenesis is an elevated level of low-density lipoprotein (LDL) that leads to the generation of oxidized LDL (oxLDL) (1). OxLDL induces an autoimmune response as evidenced by the presence of anti-oxLDL IgG in atherosclerotic lesions in the hyperlipidemic mouse model (2,3) and in humans (4)(5)(6). These studies have suggested that the titer of autoantibodies against oxLDL correlates with the progression of atherosclerosis. Epidemiological studies have shown that plasma CRP, another FcgR ligand (7), is a marker of progression of atherosclerosis (8,9). However, recent studies using human CRP overexpression in hyperlipidemic mouse model showed there was no difference in atherosclerotic lesions. Very recent studies using mouse CRP deficiency in atherosclerosis-susceptible hyperlipidemic mouse models showed no reduction in atherosclerosis in mice (10), suggesting there is no direct link between CRP levels and progression of atherosclerosis.
FcgR plays an important role in inflammatory cell activation, clearance, and presentation of Ag and also in maintaining Ig homeostasis (11)(12)(13). In mice, four different classes of FcgRs have been recognized: FcgRI, FcgRII, FcgRIII, and FcgRIV (11)(12)(13). Functionally, FcgRs can be classified into the activating (FcgRI, III, and IV) and inhibitory (FcgRII) receptors (11)(12)(13). Fcg-chain is the signaling subunit that coassociates with the activating FcgRs, and assembly and cell-surface expression of the activating FcgRs (FcgRI, III, and IV) require the coexpression of Fcg-chain (14,15). Immune complex (IC) binding to the extracellular domain of the ligand binding subunit of the activating FcgRs results in phosphorylation of the ITAM motifs resides in the cytoplasmic domain of Fcg-chain subunit (11)(12)(13). On the contrary, FcgRII, an inhibitory FcgR, is a single subunit protein, and IC binding to FcgRII induces a negative signal through its ITIM in the cytoplasmic domain (11)(12)(13). Earlier studies have presented evidence that mice deficient in Fcg-chain are resistant to the onset of ICmediated chronic inflammatory diseases (16,17).
Activated T cells specific for oxLDL are present in human atherosclerotic plaques, suggesting the involvement of adaptive immune response (18) in the initiation and progression of atherosclerosis. Elevated levels of anti-oxLDL IgG, particularly IgG1 and IgG2a, have been observed in apolipoprotein E single knockout (ApoE KO) mice fed a hyperlipidemic diet (19). The binding of anti-oxLDL IgG to oxLDL can result in the formation of soluble oxLDL IC (oxLDL-IC). Using an in vitro cell-culture

Tissue preparation and morphometric determination of atherosclerosis
Mice were anesthetized with isoflurane before euthanization. Animals were sacrificed at 15 (high-fat diet) or 25 wk (chow diet) of age, and blood was collected by the cardiac puncture into heparin-coated tubes. Plasma was separated and stored at 280˚C until further analysis. The heart and descending aorta were excised and fixed in amphibian PBS/4% formalin/30% sucrose overnight before mounted in OCT medium and frozen at 270˚C. Aortic sinus cryosections (10 mm) were stained with Oil Red O (28). For quantitative analysis of atherosclerosis, the percent lesion area in each of five sections from each mouse was obtained. En face preparations of the descending aorta were washed in distilled water; en face analysis of the descending aorta was performed after staining descending aorta with Sudan IV as previously described (28).

Plasma lipid analyses
Concentrations of plasma total and high-density lipoprotein (HDL) cholesterol were determined by enzymatic methods using kits from BioVision (Mountain View, CA) as described earlier (28).

Preparation of pb-IC
Plate-bound OVA-IC and malondialdehyde-modified LDL IC (MDALDL-IC) were prepared incubating OVA or MDALDL with a saturating concentration of affinity-purified rabbit anti-OVA IgG and rabbit antimalondialdehyde (MDA) IgG, respectively. Plate-bound IC were prepared by coating OVA or MDALDL at 15 mg/ml in PBS for 1 h at 37˚C, followed by blocking with 10% FCS (Hyclone) in PBS for 1 h and a further 1-h incubation with saturating concentration of rabbit anti-OVA or rabbit anti-MDA IgG (10 mg/ml) in PBS/10% FCS. Parallel wells prepared with OVA/anti-MDA IgG and MDALDL/anti-OVA IgG were used as controls. Formation of pb-OVA-IC or pb-MDALDL-IC was confirmed by ELISA using HRP-conjugated anti-rabbit IgG.
Bone marrow-derived macrophage and bone marrow-derived dendritic cell preparation and IC activation Bone marrow-derived dendritic cells (BMDC) were prepared from bone marrow cells (2 3 10 6 cells/100-mm dish) after incubation with RPMI 1640/10% FBS/murine GM-CSF (20 ng/ml; PeproTech) for 6 d (29). Fresh medium with GM-CSF was replaced after 3 d. After 6 d, nonadherent BMDC was collected and used for IC-mediated activation. Bone marrowderived macrophages (BMDM) were prepared by incubating bone marrow cells in RPMI 1640/10% FBS supplemented with 30% L929 culture supernatant (source of M-CSF) for 6 d (30). BMDM were gently detached using ice-cold PBS. For IC activation, BMDC or BMDM was added to pb-IC and incubated for 24 h. Supernatant was collected to determine ICmediated secretion of TNF-a, IL-6, IL-12, and IFN-g. Cells added to wells with Ag alone or Ag with control rabbit IgG were used as controls.

Quantitative RT-PCR analysis
The aorta was perfused with nuclease-free PBS. Total RNA was isolated from a proximal portion of the descending aorta (aortic arch and aorta proximal to the subclavian artery) and spleen using TRIzol (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. RNA was quantified using Nanodrop spectrophotometer (Nanodrop Products, Wilmington, DE), and its integrity was determined using Bio-Rad Experion RNA analyzer (Bio-Rad, Hercules, CA). Gene expression was measured by real-time RT-PCR after reverse transcription of total RNA (0.5 mg) as described earlier (28). PCR primer pairs were purchased from SA-Biosciences (Frederick, MD). Two-step PCR with denaturation at 95˚C for 15 s and annealing and extension at 60˚C for 1 min for 40 cycles was conducted in an iCycler (Bio-Rad). Expression of target genes was calculated by the DD Ct method using threshold cycles for b-actin as a normalization reference. All real-time PCR reactions were carried out at least twice from independent cDNA preparations. RNA without reverse transcriptase served as a negative control.

Determination of MDALDL Ab
MDALDL (Academy Biomedical, Houston, TX) and anti-MDALDL IgG and IgM responses were determined by methods described earlier (26). Briefly, MDALDL (10 mg/ml in PBS/1 mM EDTA) was coated on Nunc Microfluor white plates (Thermoscientific, Pittsburgh, PA) overnight at 4˚C and blocked with PBS/1% BSA/1 mM EDTA for 2 h at room temperature. Sera (1:50 diluted) was added to the plate and incubated at room temperature for 2 h. Plates were washed, and then alkaline phosphatase (AP)conjugated anti-mouse IgG1, anti-mouse IgG2a, anti-mouse IgG2b, antimouse IgG3, or anti-mouse IgM (Southern Biotechnology Associates, Birmingham, AL) was added and incubated for 1 h at room temperature. The AP enzymatic activity was determined using luminescence substrate Lumi-Phos530 (Lumigen Southfield, MI), and luminescence was read at 530 nm in a Polarstar plate reader (BMG Labtech, Cary, NC).

Flow cytometric analysis of T cell subsets
Splenocytes (3 3 10 5 cells) were stained with T lymphocyte subset Ab mixture (BD Biosciences), and different T cell subsets were analyzed in an FACSCalibur flow cytometer followed by data analysis using FlowJo analytical software (Tree Star, Ashland, OR). T lymphocyte subset Ab mixture consists of the PE-Cy7-anti-CD3e (clone 145-2C11), PE-anti-CD4 (clone RM4-5), and APC-anti-CD8a (clone 53-6.7). Cells stained with corresponding isotype IgG were used as controls. To determine CD152 + /CD4 + T cells, splenocytes were stained with APC-anti-CD4, PE-Cy5-CD25, and PE-CTLA4 and were analyzed by flow cytometry. For Treg analysis, splenocytes were incubated with APC-anti-CD4 and PE-Cy5-anti-CD25. After surface staining, cells were fixed, permeabilized, and stained with PE-anti-Foxp3 according to the manufacturer's instructions (BD Biosciences). For Th17 cell analysis, purified CD4 + cells (10 6 cells/ml) were incubated with 50 ng/ml PMA and 1 mM ionomycin for 6 h in the presence of monensin. Then, cells were stained with APC-anti-CD4 at 4˚C for 20 min. Cells were fixed, permeabilized, and stained with PEanti-IL-17A mAb. Cells stained with isotype control IgG were used as controls. Cells were analyzed by flow cytometric analysis using an FACSCalibur flow cytometer equipped with CellQuest software (BD Biosciences). Percent positive cells were determined using FlowJo software (Tree Star).

Assay for p-Stat3
CD4 + T cells were stimulated with anti-CD3/CD28 mAb as described earlier and lysed in lysis buffer containing protease and phosphatase inhibitors (Cell Signaling Technology, Danvers, MA). To determine IL-6dependent STAT3 phosphorylation, CD4 + T cells were stimulated with IL-6 (20 ng/ml) for 30 min. Levels of p-STAT3 and total STAT3 in the cell lysates were determined using Pathscan ELISA kits specific for each one (Cell Signaling Technology). Color was developed using TMB-1 substrate, and the reaction was stopped with 2 N sulfuric acid. Absorbance was read at 450 nm in a Polarstar multiplate reader (BMG Labtech). Phospho-STAT3 levels were normalized to total STAT3 levels in each sample.

Statistical analyses
Values are expressed as mean 6 SD. Differences between the groups were considered significant at p , 0.05 using the two-tailed Student t test. All data were analyzed using GraphPad InStat version 3.1a for Macintosh (GraphPad, San Diego, CA).

ApoE-Fcg-chain DKO mice show reduced atherosclerotic lesions
To determine the role of activating FcgRs in the progression of atherosclerosis, total activating FcgR deficiency in apoE KO background was generated. Genotype analyses showed complete knockout for Fcg-chain in the apoE KO background (Fig. 1A). Spontaneous atherosclerotic lesions in apoE KO and apoE-Fcgchain DKO mice fed a chow diet (contains 5% fat with no cholesterol) were determined by staining aortic sinus cryosections (10 mm) with Oil Red O (Fig. 1B). Atherosclerotic lesions in the aortic tree were performed by staining descending aorta from the aortic arch to the iliac bifurcation with Sudan IV (Fig. 1C). Both analyses showed an ∼50% reduction in lesions (p , 0.01, Student t test) in apoE-Fcg-chain DKO mice compared with apoE KO mice (Fig. 1D). We then determined whether Fcg-chain deficiency also had an effect on accelerated atherosclerosis in mice fed a high-fat diet (21% fat and 0.15% cholesterol). ApoE KO mice showed more lesions in the aortic sinus ( Fig. 1E, 1F), where Fcg-chain deficiency was associated with a significant decrease in lesion area in the aortic root and aortic arch relative to apoE KO male mice (Fig. 1E, 1F). Female apoE-Fcg-chain DKO mice also showed a similar reduction in lesions, suggesting there is no effect of gender on the extent of lesions (data not shown). Because Fcgchain is essential for expression and assembly of FcgRI, FcgRIII, and FcgRIV (activating FcgRs), these findings suggest activating FcgRs contribute to the atherosclerotic lesion formation.

Activating FcgR expressed on hematopoietic cells is sufficient to contribute to the atherosclerotic lesions
Studies using human cell lines have shown the expression of FcgR in vascular endothelial and smooth muscle cells (31,32). Next, we investigated whether activating FcgRs expressed on cells of hematopoietic origin contribute to the progression of atherosclerosis by bone marrow chimera. Bone marrow cells from apoE KO or apoE-Fcg-chain DKO mice were transplanted to apoE KO recipient mice. Deletion of the Fcg-chain in hematopoietic cells was confirmed by the genomic PCR analyses ( Fig. 2A). ApoE KO mice transplanted with apoE KO bone marrow cells showed significantly increased atherosclerotic lesions in the aortic root (Fig. 2B, 2D). However, apoE KO mice transplanted with apoE-Fcg-chain DKO bone marrow cells revealed a reduction of .50% (p , 0.01) in atherosclerotic lesions (Fig. 2B, 2D). Immunohistochemical analyses showed reduced macrophage at the lesions in apoE-Fcgchain DKO chimera than in apoE KO chimera mice, suggesting reduced monocyte migration and subsequent transformation to macrophages at the lesion site. Moreover, analyses of mRNA expression of macrophage FcgRIIB (inhibitory) and the activating FcgR (FcgRI, III, and IV) did not reveal any difference in the ratio between the activating and inhibitory FcgR expression in inflammatory cells (data not shown). These findings suggest that activating FcgR expression on hematopoietic cells is sufficient to contribute the atherosclerotic lesions.

Fcg-chain deficiency does not alter plasma cholesterol levels
To determine the molecular mechanisms contributing to the decreased number of atherosclerotic lesions in apoE-Fcg-chain DKO mice, plasma total and HDL cholesterol levels were determined.
There were no differences in plasma lipid profiles in apoE-Fcgchain DKO mice compared with apoE KO controls fed a chow or high-fat diet (Table I). Similar findings were observed in bone marrow chimera mice (Table I). These data suggest that the reduction in lesion formation in apoE-Fcg DKO mice compared with apoE KO mice was not due to changes in plasma lipid levels.

Anti-oxLDL IgG levels were elevated in apoE-Fcg-chain DKO mice
Anti-oxLDL Ab (IgG and IgM) responses were determined for each isotype at 10 wk of the high-fat diet. The apoE KO mice lacking Fcg-chain expression had significantly higher levels of oxLDL-specific IgG response (Fig. 3A), whereas no difference was found in anti-oxLDL IgM response (Fig. 3B). Anti-oxLDL IgG isotype analyses showed anti-oxLDL IgG2a Abs were elevated several fold, as well as the anti-oxLDL IgG1, IgG2b, and IgG3 (Fig. 3C). Ratio of IgG2a/IgG1 was not different between apoE KO and apoE-Fcg-chain DKO mice (Fig. 3D). Similar results were obtained when oxLDL was used in this assay (data not shown). Total plasma IgG and IgM were not different between apoE KO mice and apoE-Fcg-chain DKO mice (data not shown). These findings suggest there was no Th1/Th2 bias in apoE-Fcg-chain DKO mice. These findings suggest there is no apparent increase in Th1/Th2 shift in apoE-Fcg-chain DKO mice.
Attenuated atherosclerotic lesion in apoE-Fcg-chain DKO mice is not due to Th1/Th2 shift Earlier studies have suggested that CD4 + T cells specific to oxLDL promote lesions by increasing Th1 cell responses (26,27). To investigate if the Th1/Th2 shift could have contributed to the attenuated lesions in the activating FcgR-deficient mice, cytokine response by activated CD4 + T cells (from high fat-fed mice) and anti-oxLDL IgG2a (Th1) and IgG1 (Th2) levels were determined. Analysis of CD4 + and CD8 + T cell populations showed there was no difference in CD4 + /CD8 + ratio in apoE-Fcg-chain DKO mice and apoE KO mice cells (data not shown). Activated CD4 + T-cells secreted high levels of IL-10 with no difference in IL-4 levels in apoE-Fcg-chain DKO mice compared with apoE KO mice (Fig.  4A, 4B). Surprisingly, a 2-fold increase in the secretion of IFN-g was observed from activated CD4 + T cells of apoE-Fcg-chain DKO mice relative to apoE KO controls (Fig. 4C). Quantitative RT-PCR analyses of CD4 + cells showed mRNA expression of IL-10 and IL-4 was ∼3and 1.5-fold high in apoE-Fcg-chain DKO mice (Fig. 4D). Notably, IFN-g mRNA levels were also ∼3- fold higher in apoE-Fcg-chain DKO compared with apoE KO mice (Fig. 4D). Zhou et al. (19) have shown that IFN-g mRNA expression was high in the arterial lesion site, suggesting that local Th1 response at the lesion site may be contributing to the progression of atherosclerosis. We then investigated whether the change in Th1/Th2 immune responses at the lesion sites may have contributed to the reduction in lesion formation in the apoE-Fcgchain DKO mice. Using aorta from high fat-fed mice, we determined the mRNA levels of cytokines in apoE KO and apoE-Fcg-chain DKO mice. Consistent with the changes in cytokine production by activated CD4 + T lymphocytes, mRNA levels for IL-10 in apoE-Fcg-chain DKO were significantly higher relative to apoE KO mice (Fig. 4E). In contrast to the earlier observation (19), IFN-g mRNA expression in the aorta (at the lesion site) was also higher in apoE-Fcg-chain DKO mice versus apoE KO mice (Fig. 4E). These findings suggest that Th1/Th2 imbalance may not be contributing to the attenuated atherosclerotic lesions in apoE-Fcg-chain DKO mice. To address whether CD4 + T cells from apoE-Fcg-chain DKO show a bias in Th1/Th2 differentiation intrinsically, basal expression of transcription factors T-bet (Th1) and GATA3 (Th2), which initiate Th1 and Th2 cell development, respectively (33,34), was analyzed. The mRNA expression of T-bet and GATA-3 in CD4 + T cells of apoE-Fcg DKO and apoE KO mice was not different (Fig. 4F).

Th17 cells were reduced in apoE-Fcg DKO mice
We then investigated alternative mechanism(s) that may be contributing to the reduced lesions in apoE-Fcg-chain DKO mice. CD4 + T cells of apoE-Fcg-chain DKO mice showed ∼50% lower IL-17 mRNA expression compared with apoE KO mice (Fig. 5A).   Moreover, compared with CD4 + cells of apoE KO, IL-17 secretion by CD4 + T cells of apoE-Fcg-chain DKO mice was significantly reduced after stimulation with anti-CD3 and anti-CD28 (Fig. 5B). Then we determined whether Fcg-chain deficiency in hyperlipidemic apoE KO mice resulted in a reduced number of Th17 cells. Intracellular staining for IL-17 in PMA/ionomycin-activated CD4 + cells revealed apoE KO mice have ∼2% Th17 + cells (Fig.  5C, 5D). However, the number of Th17 + cells was reduced to ,0.5% of CD4 + cells in apoE-Fcg-chain DKO mice (Fig. 5C,  5D). Then we determined the expression of IL-17 in the aorta samples to address whether IL-17 expression is reduced at the lesion site. IL-17A mRNA expression was lower in aorta sample from apoE-Fcg-chain DKO mice (Fig. 5E). These findings suggest that decreased generation of Th17 cells and IL-17 expression in apoE-Fcg-chain DKO mice could in part contribute to the attenuated atherosclerotic lesions in these mice.

More Tregs in apoE-Fcg-chain DKO-deficient mice
Recent studies using apoE KO mice have shown that there is an inverse relationship to the number of Th17 and CD4 + CD25 + Foxp3 + Tregs (35)(36)(37). Hence, we determined the number of Tregs in apoE KO and apoE-Fcg-chain DKO mice fed a high-fat diet. The deficiency of Fcg-chain in hyperlipidemic apoE KO mice was associated with an increase in the number of Tregs (Fig. 6A). Moreover, an ∼2-fold increase in Foxp3 mRNA expression was observed in apoE-Fcg DKO mice compared with apoE KO mice (Fig. 6B). As CD152 have been shown to be upregulated in Tregs, we determined the mRNA expression of CD152 in the CD4 + T lymphocytes of apoE KO and apoE-Fcg-chain DKO mice by real-time RT-PCR. Interestingly, CD152 mRNA expression was increased significantly in the spleen of apoE-Fcg-chain DKO mice compared with apoE KO mice, whereas expression of CD28, another T cell costimulatory molecule, was similar in apoE KO and apoE-Fcg-chain DKO mice (Fig. 6C). These findings suggest that Fcg-chain deficiency in a hyperlipidemic mouse background increased the regulatory CD4 + T cell population, causing an imbalance between Th17 and Tregs.

Mechanism contributing to the Th17/Treg shift in apoE-Fcgchain DKO mice
We then determined mechanisms contributing to the reduced Th17 cells in apoE-Fcg-chain DKO mice. Generation of Th17 cells in mice requires the presence of TGF-b and IL-6 (38-40). Hence, the levels of TGF-b and IL-6 in anti-CD3/CD28 activated CD4 + cells. Activated CD4 + cells of the apoE-Fcg-chain secrete 3-fold higher levels of TGF-b than in activated CD4 + cells of apoE KO mice (Fig. 7A). The decrease in TGF-b secretion also paralleled a .3fold increase in TGF-b mRNA expression in apoE-Fcg-chain DKO mice (Fig. 7B). TGF-b can prime the naive CD4 + T cells to differentiate into Tregs or Th17 cells depending on the presence of IL-6 (39). Hence, IL-6 secretion by activated CD4 + cells was determined. IL-6 mRNA (Fig. 7C) and IL-6 secretion by activated CD4 + cells (Fig. 7D) was significantly reduced in apoE-Fcg-chain DKO mice. Recent studies showed that the IL-6/STAT3 signaling pathway plays a predominant role in the induction of Th17 response in autoimmune disease (41,42). Because IL-6 and IL-17 secretion was reduced in anti-CD3-stimulated CD4 + cells from apoE-Fcg-chain DKO mice, we investigated whether STAT3 activation is reduced in apoE-Fcg-chain DKO mice. Total and phospho-STAT3 levels in activated CD4 + T cells were determined by ELISA-based assay. Total STAT3 levels were not different between the two strains of mice (Fig. 7E). However, less p-STAT3 was detected in CD4 + T cells from apoE-Fcg-chain DKO mice when compared with that from apoE KO mice (Fig. 7F). IL-21 selectively produced by Th17 cells has been shown to serve as an autocrine factor for promoting and sustaining Th17 lineage commitment and drive IL-17 production in a STAT3-dependent manner (43). Hence, we determined whether lower p-STAT3 in apoE-Fcg-chain DKO mice affects IL-21 secreted by activated T cells. IL-21 secretion in activated T cells was significantly lower Each value indicates mean 6 SD results from five mice. C, Anti-oxLDL IgG isotype analysis was determined in sera from apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk. Sera was diluted 1:50, and anti-oxLDL IgG1, IgG2a, IgG2b, and IgG3 levels were determined using isotype-specific secondary Abs. In all of these assays, lumophos530, a luminescence AP substrate, was used, and relative light units (RLU) were measured in a luminescence plate reader. Each group contained five mice. D, IgG1/IgG2a ratio was determined arbitrarily by taking a ratio between the RLU values obtained for each animal. Each group contained five mice. *p , 0.05 and compared with apoE KO mice.
in apoE-Fcg-chain DKO mice (Fig. 7G) compared with T cells from apoE KO mice. Because we observed both reduced IL-6 and STAT3 phosphorylation in apoE-Fcg-chain mice, we then determined whether Fcg-chain deficiency affects IL-6 signaling pathway, which may lead to an IL-17 response. CD4 + T cells were treated with IL-6, and p-STAT3 was detected. Total STAT3 levels were also not different between both strains (data not shown). Basal and IL-6-mediated p-STAT3 levels were not different between apoE KO and apoE-Fcg-chain DKO mice (Fig. 7H), suggesting that Fcg-chain deficiency does not affect IL-6 signaling pathway. These findings indicate that less Th17 response in apoE-Fcg-chain DKO mice may result from lack of optimal cytokine responses, such as IL-6 resulting in less activation of STAT3 signaling pathway.
Reduced IL-6 secretion by BMDM and BMDC of apoE-Fcg-chain DKO mice IL-6 release by APC such as DC has been shown to be critical for the generation and differentiation of Th17 cells (44,45). Hence, to identify the FcgR-dependent factors for Th17 skewing in vivo, OVA-IC-and MDALDL-IC-mediated IL-6 release by BMDM and BMDC of apoE-Fcg-chain DKO mice was determined. Platebound OVA-IC or MDALDL-IC was confirmed by ELISA assay prior to adding the cells on IC-coated plates (data not shown). OVA-or MDALDL-treated apoE KO BMDM and BMDC did not induce IL-6 ( Fig. 8A, 8B) secretion. However, OVA-IC or MDALDL-IC activation of apoE KO BMDM and BMDC showed very high levels of IL-6 secretion, ∼30-fold for BMDM and 8-fold for BMDC (Fig. 8A, 8B). Interestingly, both MDALDL-IC and OVA-IC activation of apoE-Fcg-chain-deficient BMDM and BMDC did not induce IL-6 secretion (Fig. 8A, 8B). Similar findings were observed for IC-mediated TNF-a (Fig. 8C, 8D) and IL-12 (Fig. 8E, 8F) secretion by BMDM and BMDC from apoE-Fcg-chain DKO mice. Under similar conditions, INF-g secretion was too low to be detected (data not shown). These findings suggest that more attenuated Th17 cells in apoE-Fcg DKO mice may be due to reduced levels of IL-6 secretion by APC deficient in the Fcg-chain.

Discussion
In the current study, we tested the hypothesis that oxLDL-IC binding to the activating FcgR promotes the progression of atherosclerosis. We showed a significant reduction in the formation of arterial lesions in Fcg-chain-deficient mice after a high-fat diet. Interestingly, bone marrow chimera studies showed attenuated lesions in apoE KO mice transplanted with bone marrow from apoE-Fcg-chain DKO mice. These findings suggest that the activating FcgR expressed on hematopoietic cells is sufficient for the progression of atherosclerosis. Remarkably, the reduction in atherosclerotic lesion progression in the activating FcgR deficiency in hyperlipidemic mouse model was associated with the increased generation of CD4 + Tregs with concomitant decrease in CD4 + Th17 cells.
Earlier studies have shown expression of activating FcgR in vascular endothelial and smooth muscle cells (31,32). Moreover, studies using apoE-Fcg-chain DKO mice revealed that expression of the inhibitory FcgRIIB was elevated in vascular smooth muscle cells (19). These studies have suggested that the ratio between inhibitory/activating FcgRs in vascular smooth muscle FIGURE 6. Number of Tregs was high in apoE-Fcg-chain DKO-deficient mice. A, Flow cytometric analysis of CD4 + CD25 + Foxp3 + cells was determined by staining purified CD4 cells from apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk. Number of Tregs was determined by analyzing the FACS data using FlowJo software (Tree Star). B, Foxp3 mRNA expression was high in CD4 + cells from apoE-Fcg-chain DKO mice. RNA was isolated from purified CD4 + cells of apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk. Foxp3 mRNA levels were determined by quantitative RT-PCR assays and normalized relative to housekeeping gene b-actin transcripts. Each value indicates mean 6 SD results from five mice. C, CD152 mRNA expression was high in CD4 + cells from apoE-Fcg-chain DKO mice. RNA was isolated from purified CD4 + cells of apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk. CTLA4 and CD28 mRNA levels were determined by quantitative RT-PCR assays and normalized relative to housekeeping gene b-actin transcripts. Each value indicates mean 6 SD results from five mice. *p , 0.05, ***p , 0.0001 compared with apoE KO mice. cells could be responsible for the observed reduced atherosclerotic lesions in apoE-Fcg-chain DKO mice (22). It is well established that both activating and inhibitory FcgRs are constitutively expressed on cells of hematopoietic origin, which include monocytes/macrophages, neutrophils, and NK cells (11)(12)(13). Notably, using the bone marrow chimera approach, we showed that the deficiency of activating FcgR expression on hematopoietic cells is sufficient to inhibit the progression of atherosclerotic lesions. Thus, the findings from the current study indicate the possibility that the activating FcgR expressed on hematopoietic cells is the major contributor in the progression of atherosclerosis in apoE KO mice. FIGURE 7. Reduced Th17 and higher Tregs are due to attenuated IL-6 expression. Purified CD4 + lymphocytes from apoE KO and apoE-Fcg-chain DKO (fed high-fat diet for 10 wk) were stimulated with pb-anti-CD3 and soluble CD28 for 72 h. The concentration of TGF-b (A) and IL-6 (C) was measured by cytokine bead array. Each value indicates mean 6 SD results from five mice. TGF-b (B) and IL-6 (D) mRNA expression in CD4 + cells from apoE-Fcg-chain DKO mice. RNA was isolated from purified CD4 + cells of apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk. TGF-b and IL-6 mRNA levels were determined by quantitative RT-PCR assays and normalized relative to housekeeping gene b-actin transcripts. Each value indicates mean 6 SD results from five mice. E and F, Reduced STAT3 phosphorylation in apoE-Fcg-chain DKO mice. Purified CD4 cells were stimulated with pb-CD3, and cell lysates were prepared. Total STAT (E) and p-STAT3 (F) were determined by Pathscan ELISA kit (Cell Signaling Technology) as described in Materials and Methods. G, IL-21 secretion by activated CD4 + cells from apoE KO and apoE-Fcg-chain DKO mice fed high-fat diet for 10 wk was determined by ELISA. Each value indicates mean 6 SD results from five mice. H, IL-6 induced STAT3 phosphorylation in apoE-Fcg-chain DKO mice. Purified CD4 cells of apoE KO and apoE-Fcg-chain DKO mice were stimulated without or with IL-6, and phospho-STAT3 was determined by Pathscan ELISA kit as described in Materials and Methods. Values are expressed as means 6 SD; n = 5/group. **p , 0.01, ***p , 0.0001 compared with apoE KO mice. Th1 response has long been recognized as having a proatherogenic potential and an important role in the development of atherosclerosis (19,26). Plasma anti-oxLDL IgG analyses revealed there is no clear distinction of Th1 (IgG2a) and Th2 (IgG1) type of Ab response in apoE-Fcg-chain DKO mice. Notably, we showed anti-oxLDL IgG1 and IgG2a levels were higher in apoE-Fcgchain DKO mice compared with apoE KO mice, despite lesions being lower in apoE-Fcg-chain DKO mice. Earlier studies have shown that in moderate hypercholesterolemia, anti-oxLDL Ab response is predominantly the IgG2a isotype, correlating with increased IFN-g-producing T cells (19). However, severe hypercholesterolemia was shown to be associated with a shift from Th1 to Th2 response as evident from elevated anti-oxLDL IgG1 isotype and appearance of IL-4, Th2 cytokine, producing cells in atherosclerotic lesions (19). These findings suggest that antigenic load may determine the Th1/Th2 autoimmune responses in atherosclerosis. Immunization of apoE KO with MDALDL resulting in elevated plasma anti-oxLDL IgG was atheroprotective (46)(47)(48). This raises an interesting possibility that higher anti-oxLDL levels in apoE-Fcg-chain DKO mice may be atheroprotective. However, this possibility needs to be resolved by passive transfer of anti-oxLDL IgG.
Th1 response has long been recognized having a proatherogenic potential and an important role in the development of atherosclerosis (19,26). Our findings showed no difference in IL-4 levels in apoE-Fcg-chain DKO compared with apoE KO mice, ruling out a role for IL-4 in reduced lesions in apoE-Fcg-chain DKO mice. Surprisingly, IFN-g levels were elevated in activated CD4 + T cells of apoE-Fcg-chain DKO mice though the lesions were fewer in these mice. This finding is in agreement with earlier report showing elevated IFN-g production with attenuated lesion in LDLR-FcgRIII DKO mice (23). The mechanism(s) contributing to the elevated IFN-g in the total activating FcgR deficiency (our report) and FcgRIII deficiency (23) needs to be explored. To address whether Fcg-chain deficiency inherently influences different T cell subsets, we determined basal level expression of T-bet and GATA-3, transcription factors essential for Th1 and Th2 cell differentiation. We did not detect the difference in T-bet and GATA-3 expression in CD4 + T cells from apoE-Fcg-chain DKO mice, ruling out that there is an intrinsic effect of Fcg-chain deficiency on Th1 and Th2 differentiation. Nevertheless, these findings suggest that the imbalance in Th1/Th2 may not be contributing to the attenuated lesions seen in apoE-Fcg-chain DKO mice. Notably, our findings also showed activated CD4 + T cells produced elevated IL-10 in apoE-Fcg-chain DKO compared with apoE KO mice, suggesting that IL-10 secreted by other T cell subsets such as Tregs may be contributing to the reduced lesions in apoE-Fcg-chain DKO mice.
Th17 cells, a subset of CD4 cells secreting IL-17, have been implicated in proinflammatory responses (41), suggesting a role for IL-17 in the progression of atherosclerosis. However, the role of IL-17 and Th17 in atherosclerosis is still evolving. The proatherogenic role of IL-17 was demonstrated in apoE KO mice using anti-IL-17 mAb (35), recombinant soluble IL-17R-A (36), or LDLR KO recipient mice transplanted with IL-17R-deficient bone marrow cells (49) or IL17 KO mice (50). On the contrary, two studies from the Mallat group (51, 52) have suggested that the elevated level of IL-17 is atheroprotective. In the first study, LDLR KO mice received suppressor of cytokine signaling 3-deficient bone marrow cells showing reduced atherosclerosis with elevated IL-17 and IL-10 levels (51). In the second study, the atheroprotective role of IL-17 was suggested in B cell-depleted mice showing elevated IL-17 levels (52). It should be pointed out that in all of the studies examining the direct role for IL-17 using apoE KO, the hyperlipidemic mouse model showed IL-17 is proatherogenic. Our findings showing reduced lesions, number of Th17 cells, and secretion of IL-17 by activated CD4 + lymphocytes in apoE-Fcg-chain DKO compared with apoE KO mice suggest a proatherogenic role for IL-17. Moreover, IL-21 selectively produced by Th17 cells has been shown to serve as an autocrine factor for promoting and sustaining Th17 lineage commitment and to drive IL-17 production (43). Our findings showing reduced IL-21 secretion by activated T cells further confirmed that Th17 response is lower in apoE-Fcg DKO mice. However, more studies may be needed to resolve the cloud around the role for IL-17 in atherosclerosis.
Finally, we determined the FcgR-dependent factors that might be a possible mechanism for the reduced Th17 cells in apoE-Fcgchain DKO mice. IL-6 and IL-6-dependent STAT3 signaling pathways are essential for Th17 generation (42,44). We showed IL-6 secretion and STAT3 phosphorylation in activated CD4 + cells was reduced in apoE-Fcg-chain DKO mice, suggesting less IL-6 secretion and STAT3 phosphorylation may relate to reduction of Th17 response in apoE-Fcg-chain DKO mice. This raises an interesting link between the Fcg-chain and IL-6 signaling pathway. However, exogenous addition of IL-6 did not show a difference in p-STAT3 levels in CD4 cells of apoE KO and apoE-Fcg-chain DKO mice (Fig. 7H), ruling out a direct link between the Fcgchain and IL-6 signaling pathway. Previous studies showing DCderived IL-6 is critical for differentiation (45) raise the possibility that lack of IL-6 secretion by APCs from Fcg-chain-deficient mice may affect Th17 differentiation. We showed IC-mediated IL-6 and TNF-a secretion was impaired in BMDM and BMDC from apoE-Fcg-chain DKO mice, but not from apoE KO mice. Hence, it is possible that in the absence of IL-6, the Treg differentiation pathway may be activated rather than Th17 differentiation (53,54). Recent studies have also shown an inverse relationship between Th17 and Tregs in the progression of atherosclerosis in hyperlipidemic mouse models (37) and human studies (55)(56)(57). We showed that deficiency of Fcg-chain resulted in expansion of Tregs producing high levels of IL-10 and TGF-b, which have been implicated in antiatherogenic effects (58)(59)(60). Collectively, our findings suggest that an anti-inflammatory response by CD4 + CD25 + Tregs may inhibit Th17 differentiation and inhibit the atherosclerosis in hyperlipidemic mouse models.
In summary, our investigation demonstrated that Fcg-chain deficiency leading to the impaired functions of the activating FcgR (FcgRI, III, and IV) has resulted in reduced atherosclerotic lesions in hyperlipidemic apoE KO mouse model. The attenuated lesions in apoE-Fcg-chain DKO mice are not due to the imbalance in Th1/Th2 shift. On the contrary, our findings showed higher Tregs with a concomitant decrease in Th17 cells in part contributed to the abridged lesions in Fcg-chain-deficient mice in hyperlipidemic conditions. Importantly, reduced FcgR-mediated IL-6 secretion may contribute to attenuated Th17 response and subsequently attenuated atherosclerosis in apoE-Fcg-chain DKO mice. These studies collectively suggest that the lack of IL-6 secretion in Fcg-chain-deficient mice may be contributing to the reduced number of Th17 cells in apoE-Fcg-chain DKO mice.