LDL Receptor-Related Protein 1 Prevents Early Atherosclerosis by Limiting Lesional Apoptosis and Inflammatory Ly-6C^high Monocytosis: Evidence that the Effects Are Not ApoE-Dependent

Abstract

Background

We previously demonstrated that macrophage LRP1 deficiency increases atherosclerosis despite anti-atherogenic changes including decreased uptake of remnants and increased secretion of apoE. Thus, our objective was to determine whether the atheroprotective effects of LRP1 require interaction with apoE, one of its ligands with multiple beneficial effects.

Methods and Results

We examined atherosclerosis development in mice with specific deletion of macrophage LRP1 (apoE^−/−MΦLRP1^−/−) and in LDLR^−/− mice reconstituted with apoE^−/−MΦLRP1^−/− bone marrow. The combined absence of apoE and LRP1 promoted atherogenesis more than did macrophage apoE deletion alone in both apoE-producing LDLR^−/− (+88%) and apoE^−/− mice (+163%). The lesions of both mouse models with apoE^−/−LRP1^−/− macrophages had increased macrophage content. In vitro, apoE and LRP1 additively inhibit macrophage apoptosis. Furthermore, there was excessive accumulation of apoptotic cells in lesions of both LDLR^−/− (+110%) and apoE^−/−MΦLRP1^−/− mice (+252%). The apoptotic cell accumulation was partially due to decreased efferocytosis as the ratio of free to cell-associated apoptotic nuclei was 3.5-fold higher in lesions of apoE^−/−MΦLRP1^−/− versus apoE^−/− mice. Lesion necrosis was also increased (6-fold) in apoE^−/−MΦLRP1^−/− versus apoE^−/− mice. Compared to apoE^−/− mice, the spleens of apoE^−/−MΦLRP1^−/− mice contained 1.6 – and 2.4 –fold more total and Ly6-C^high monocytes. Finally, there were 3.6- and 2.4-fold increases in Ly6-C^high and CCR2+ cells in lesions of apoE^−/−MΦLRP1^−/− versus apoE^−/− mice, suggesting that accumulation of apoptotic cells enhances lesion development and macrophage content by promoting the recruitment of inflammatory monocytes.

Conclusion

LRP1 exerts anti-atherogenic effects via pathways independent of apoE involving macrophage apoptosis and monocyte recruitment.

Macrophages play a critical role in atherogenesis by taking up lipoprotein particles trapped in the arterial intima and activating the inflammatory response, or turning into apoptosis-prone foam cells. Apolipoprotein E (apoE) and the LDL receptor-related protein 1 (LRP1) are critical for these processes^1–3. Plasma apoE associates with lipoproteins and causes bulk clearance of remnants⁴. Macrophages contribute to less than 10% of plasma apoE⁵, but produce most of the apoE in the atheroma. Evidence supports an anti-atherogenic role for both systemic and macrophage apoE^{1, 2, 6}. LRP1 is both a cargo-transporter and a signaling receptor. It binds over 30 distinct extracellular ligands and various cytoplasmic adaptor proteins⁷. LRP1 protects against atherosclerosis when expressed by vascular smooth muscle cells (SMCs)⁸, hepatocytes⁹, or macrophages^{3, 10}.

LRP1 and apoE may work in common pathways in preventing atherosclerosis. ApoE on remnant lipoproteins serves as a ligand for their LRP1-mediated internalization⁹, and is one of several triggers of LRP1-activated downstream signaling⁷. ApoE signaling via LRP1 has been shown to promote cell survival¹¹. Interaction of apoE with macrophage LRP1 initiates calcium mobilization and both apoE and LRP1 limit the macrophage inflammatory response by preventing NF-κβ activation^{12, 13}. In addition, our previous studies demonstrated that macrophages lacking LRP1 cause increased atherosclerosis development in LDLR^−/− mice despite up-regulation of apoE secretion,^{3, 14} suggesting that LRP1 mediates the beneficial effects of apoE. In these studies, we also showed that the LRP1/apoE axis limits macrophage inflammation and regulates the rate of efferocytosis of apoptotic macrophages¹⁴.

The objective of the current studies was to investigate whether LRP1 modulates atherosclerosis independently of apoE. We examined atherosclerosis development in mice with specific deletion of macrophage LRP1 (apoE^−/−MΦLRP1^−/−) and in LDLR^−/− mice reconstituted with apoE^−/−MΦLRP1^−/− bone marrow. Our studies demonstrate that the pro-atherogenic effects of macrophage LRP1 deletion are enhanced in the absence of macrophage apoE and further enhanced in the absence of systemic apoE. The atherogenic effects of macrophage LRP1 deletion in the setting of complete apoE deficiency include enhanced intimal macrophage accumulation, excessive macrophage apoptosis, decreased efferocytosis, increased necrosis, and enhanced content of pro-infammatory Ly6-C^high monocytes and CCR2+ macrophages^{15, 16}.

Methods

A detailed description of all methods is available in the supplemental materials (available online at http://circ.ahajournals.org).

Mice

Mice lacking macrophage LRP1 (MΦLRP1^−/−) were developed as described,³ and mated with apoE^−/− mice to obtain apoE^−/−MΦLRP1^−/− mice. All mice were on C57BL/6 background. Animal procedures were performed in accordance with the Institutional Animal Care and Usage Committee of Vanderbilt University.

Atherosclerosis Analyses

LDLR^−/− mice (12 week old) were lethally irradiated (9.5 Gy) using a cesium gamma source and transplanted with 5×10⁶ bone marrow (BM) cells from female apoE^−/−(n=10) or apoE^−/−MΦLRP1^−/− (DKO) mice (n=10) as described^{1, 2}. After four weeks, the mice were placed on western-type diet for 8 weeks, and then the extent of atherosclerosis was examined. For experiments comparing atherosclerosis in apoE^−/− (n=12) or apoE^−/−MΦLRP1^−/− mice (n=12), eight to ten week old female mice were fed a western-type diet for 8 weeks. The extent of atherosclerosis was examined both in oil red O–stained cross-sections of the proximal aorta (15 alternate 10-μm cryosections) and by en face analysis using the KS300 imaging system (Kontron Elektronik GmbH) as described ¹.

Analyses of Macrophage Apoptosis

Peritoneal macrophages were seeded in chamber slides (Nalge Nunc International) and incubated for 16h in serum-free DMEM alone or containing lipopolysaccharide (LPS, 50ng/ml). Cell death was determined by TUNEL (Tdt-mediated dUTP nick end labeling) staining using the in situ detection kit (Roche).

Analyses of Lesion Apoptosis, Efferocytosis, and Necrosis

TUNEL staining was performed on 5-μm proximal aortic cryosections using the in situ cell death detection kit (Roche). The TUNEL positive cells were counted in four sections per mouse and normalized to Oil-red-O lesion area. For analysis of efferocytosis, 5-μm proximal aortic cryosections were stained with TUNEL using the TMR red (Roche) detection kit. Nuclei were counterstained with Hoechst and macrophage cytoplasm was detected using rabbit anti-macrophage antibody (Accurate Chemical and Scientific Corp.), goat anti-rabbit Alexa Fluor 488 conjugated secondary antibody (Molecular Probes, Inc.). The free versus macrophage associated apoptotic bodies were then counted in 4 sections per mouse. Necrosis was detected using Harris’s hematoxylin and eosin (H&E), and quantitated by measuring the H&E negative acellular area in the intima versus total intimal area.

Analyses of Lesion MOMA-2, Ly-6C, and CCR2

Macrophage content was detected using rabbit antibody against mouse MOMA-2 (Accurate Chemical & Scientific Corp.) as described³. Ly-6Cand CCR2 positive cells were detected in 4 cryosections per mouse using rat anti-mouse Ly6-C biotin (BD Pharmingen #557359), rabbit anti-mouse CCR2 monoclonal (AbCam #ab32144), either Streptavadin-AlexaFluor 488 (Invitrogen #S11223) or goat anti-rabbit AlexaFluor 647 (Invitrogen #A21244).

Flow Cytometry Analyses of Blood and Spleen Ly-6C^high Monocytes

Wild type (WT, n=10), apoE^−/− (n=10), MΦLRP1^−/− (n=8), or apoE^−/−MΦLRP1^−/− (n=10) mice were fed a western diet for 8 weeks, and then blood was collected via cardiac puncture in sodium citrate (10mM sodium citrate, 13mM glucose, pH 6.5). Spleens were homogenized by disruption in sterile PBS pH 7.4 through a 70μM mesh screen. To distinguish monocytes from other blood and spleen cells, FITC fluorochrome tagged rat anti-mouse CD90.2, B220, GR1 (Pharmingen) and NK cells (Caltag) were used. Total monocytes were detected using rat anti-mouse CD11b-PE (Pharmingen) and Ly6-C^high monocytes were quantitated using rat anti-mouse Ly6-C labeled with biotin (Pharmingen) and streptavadin-linked AlexaFluor 647 (Invitrogen).

Statistical Analysis

Differences between mean values were determined by one-way ANOVA (Bonferroni’s post test) and Mann-Whitney test using GraphPad PRISM. Prior to using one-way ANOVA (Bonferroni’s post test) to test for significance, the normality of the sample populations was tested by the Kolmogorov-Smirnov test. Significance was set for p<0.05.

Results

Impact of macrophage LRP1 deletion on atherosclerosis and intimal macrophage accumulation

Our previous studies have shown that LDLR^−/− mice recipient of apoE^+/+MΦLRP1^−/− or apoE^−/−MΦLRP1^+/+ BM develop 40% and 75% larger lesions in the aortic sinus, respectively, compared to LDLR^−/− mice transplanted with WT BM^{2, 3}. In addition, we have shown that LRP1^−/−macrophages express and secrete more apoE compared to WT cells¹⁴. Since macrophage apoE is a potent anti-atherogenic agent², the failure of increased expression of apoE in LRP1^−/−macrophages to decrease atherosclerosis³ is consistent with the possibility that the atherogenic phenotype caused by LRP1^−/− macrophages is due to the abrogation of apoE-mediated benefits. Therefore, we set out to study the role of an apoE/LRP1 interaction in atherogenesis by looking at the cellular and vascular consequences of removing either macrophage apoE alone or both systemic and macrophage apoE from mice with LRP1^−/− macrophages. We first examined the role of macrophage apoE/LRP1 interaction by transplanting BM from either apoE^−/− or apoE^−/−MΦLRP1^−/− mice into LDLR^−/− recipients. We reasoned that, if the atheroprotective effects of macrophage LRP1 were mediated, exclusively or to a large extent, by its interaction with macrophage apoE, then atherosclerosis development would be similar in LDLR^−/− mice transplanted with apoE^−/− or apoE^−/−/MΦLRP1^−/− bone marrow. Four weeks after BM transplantation, recipient mice were placed on a Western diet for 8 weeks to raise plasma cholesterol levels and induce atherosclerosis. After 8 weeks on western diet, the plasma cholesterol and triglyceride levels as well as the lipoprotein cholesterol profiles were not different between the two groups (Supplemental Figure 1A). Compared to LDLR^−/− mice receiving apoE^−/− BM, those that received apoE^−/−MΦLRP1^−/− marrow had 88% more lipid-stainable lesion area in the proximal aorta (Figures 1A and 1C), and a 138% increase in MOMA-2 stainable intimal macrophages (Figures 1B and 1C). Aorta en face lesions were minimal and not significantly different between apoE^−/−MΦLRP1^−/− and apoE^−/− BM recipient mice (0.91% vs 0.73%, p=0.36). These results are compatible with independent and additive effects of LRP1 and apoE on atherogenesis despite their cooperation in functional networks^{7, 9, 11}.

Figure 1. Atherosclerosis in LDLR^−/− mice transplanted with BM from apoE^−/− or apoE^−/−MΦLRP^−/− mice.

Quantitation of the mean Oil-red-O (A) and MOMA-2 (B) stained cross-sectional area of proximal aortas from LDLR^−/− mice transplanted with either apoE^−/− (n=10) or apoE^−/−MΦLRP^−/− (n=10) BM. *p<0.05, Mann-Whitney test. C. Representative images show Oil-red-O and MOMA-2 stain in aortic root sections.

Studies have shown that even small amounts of circulating apoE can decrease atherosclerosis without changing plasma cholesterol¹⁷. Therefore, it is important to bear in mind that in the LDLR^−/− mice receiving apoE^−/− marrow, systemic apoE can navigate into the vessel wall and affect lesional macrophage function, therefore influencing atherogenesis. To eliminate the effects of plasma apoE, we compared plaque development in apoE^−/− and apoE^−/−/MΦLRP1^−/−mice (n=12 each) fed a Western diet for 8 weeks. There were no differences in plasma cholesterol, triglyceride, and lipoprotein FPLC profiles between the two groups (Supplemental Figure 1B). ApoE^−/−MΦLRP1^−/− mice had 163% more Oil-red-O and 133% more MOMA-2 staining in the proximal aorta compared to apoE^−/− mice (Figures 2A, 2B, and 2C). The fact that LRP1 modulates atherogenesis in the complete absence of apoE proves that LRP1 has apoE-independent effects in the vessel wall.

Figure 2. Atherosclerosis in apoE^−/− and apoE^−/−MΦLRP^−/− mice.

Quantitation of the mean Oil-red-O (A) and MOMA-2 (B) stained cross-sectional area of proximal aortas from apoE^−/− and apoE^−/−MΦLRP^−/− mice (n=12 per group). *p<0.05, Mann-Whitney test. C. Images show Oil-redO and MOMA-2 stain in aortic root sections.

Effects of apoE and LRP1 deletion on lesion and macrophage apoptosis

Macrophage apoptosis influences atherosclerosis development¹⁸, and both LRP1 and apoE play roles in cell survival and in efferocytosis of apoptotic macrophages^{11, 14, 19}. Therefore, we next compared the level of apoptosis induction in peritoneal macrophages of the four genotypes under conditions of stress. To induce apoptosis, the cells were incubated for 16h in serum-free DMEM (nutritional stressor), and cell death was determined by TUNEL staining (Figures 3Aand 3B). Compared to wild-type macrophages, significantly more apoptotic cells were seen in cultures of apoE^−/− (3.7-fold), LRP1^−/− (2.5 fold), and apoE^−/−LRP1^−/− (6.7-fold) macrophages, suggesting additive effects of LRP1 and apoE on apoptotic susceptibility. Similar differences were observed when apoptosis was stimulated with LPS, where TUNEL positive cells were 10%, 21%, 24%, and 48% of lesions in wild-type, apoE^−/−, LRP1^−/−, and apoE^−/−LRP1^−/−macrophage cultures, respectively.

Figure 3. In vitro macrophage apoptosis and atherosclerotic lesion apoptosis.

A and B. Images show TUNEL positive (red) and negatively stained macrophages after 16h with DMEM (A). Quantitation of the percent TUNEL positive cells in WT, apoE^−/−, LRP^−/−, and apoE^−/−LRP^−/−cultures (n=6 per group). (B). *, #, and & denotes statistically significances compared to WT, apoE^−/−, and LRP^−/− cells, respectively, p<0.05, ANOVA with Bonferroni’s post test. C. Quantitation of apoptotic cells in proximal aorta sections from apoE^−/− (n=6) or apoE^−/−MΦLRP^−/−(n=7) BM recipient LDLr^−/− mice. *p<0.05, Mann-Whitney test. D. Images show TUNEL staining (dark red) of nuclei in aortic root sections.

We next examined whether the enhanced atherosclerosis that occurs in the absence of macrophage LRP1 can be attributed to increased macrophage apoptosis. TUNEL staining analyses show that lesions of apoE^−/−MΦLRP1^−/−→LDLR^−/− mice contained more than twice as many apoptotic cells per mm² lesion area compared to the lesions of apoE^−/−→LDLR^−/− mice (Figures 3C and 3D). Compared to lesions of apoE^−/− mice, there was a 3.5-fold increase in apoptotic cells in lesions of apoE^−/−MΦLRP1^−/− mice (Figures 4A and 4B), demonstrating that the effects of macrophage LRP1 deletion on lesional apoptotic cell accumulation are more exacerbated in the complete absence of apoE. The apoptotic cells co-localized with macrophage enriched areas of lesions in both apoE^−/− and apoE^−/− MΦLRP1^−/− mice (Figure 4A). Staining of lesion apoptosis, nuclei, and macrophage cytoplasm enabled the quantitation of free versus macrophage associated apoptotic bodies as previously described in detail¹⁴ (also see Supplemental Methods). The ratio of free to macrophage associated apoptotic nuclei/bodies was 3.8-fold higher in lesions of apoE^−/− MΦLRP1^−/− mice compared to apoE^−/− mice (Figure 4C), indicating that deletion of macrophage LRP1 impairs efferocytosis independently of apoE. Consistent with the excessive accumulation of apoptotic cells and defective lesional efferocytosis, the percent necrotic area was 6-fold greater in lesions of apoE^−/− MΦLRP1^−/− mice (Figures 5A and 5B).

Figure 4. Macrophage apoptosis and efferocytosis in lesions of apoE^−/− and apoE^−/− MΦLRP^−/− mice.

A Micrographs show nuclei (Hoechst, blue), nuclei + TUNEL positive staining (red), and merged images of macrophage cytoplasm (green), nuclei, and TUNEL in aortic root sections from apoE^−/− and apoE^−/−MΦLRP^−/− mice. Quantitation of the number of apoptotic cells (B) and of the ratio of free versus macrophage associated TUNEL positive cells (C) in proximal aorta sections of lesions from apoE^−/− and apoE^−/−MΦLRP^−/− mice (n=6 per group). *p<0.05, Mann-Whitney test.

Figure 5. Necrosis in lesions of apoE^−/− and apoE^−/−MΦLRP^−/− mice.

A Images show hematoxylin and eosin staining of aortic root sections from apoE^−/− and apoE^−/−MΦLRP^−/− mice. NEC denotes necrotic area. B. Quantitation of the percent necrotic area in the aortic lesions from apoE^−/− and apoE^−/−MΦLRP^−/− mice (n=6 per group). *p<0.05, Mann-Whitney test.

Effects of apoE and LRP1 deletion on circulating, spleen, and lesion Ly-6C^high monocytes

Studies have shown that apoE^−/− mice on a western diet accumulate more pro-inflammatory Ly-6C^high monocytes (Gr1+CCR2+CX₃CR1^low) in blood, spleen, and atherosclerotic lesions¹⁵. In addition, our previous studies have shown that deletion of macrophage LRP1 increases cell migration in response to the CCR2 ligand, MCP-1. Furthermore, studies suggest that apoptotic macrophages can enhance lesion development by promoting endothelial cell inflammation and the recruitment of monocytes^{18, 20, 21}. Therefore, we investigated the impact of LRP1 expression on the accumulation of Ly-6C^high monocytes. We first examined the effects of either single or combined deletion of apoE and LRP1 on circulating and splenic Ly-6C^high monocytes in WT, MΦLRP1^−/−, apoE^−/−, and apoE^−/−MΦLRP1^−/− mice fed a western diet for 8 weeks, as measured by flow cytometry. The total number of blood monocytes was not significantly different among the four strains of mice (Figure 6A). Similar to what was previously reported,¹⁵ Ly-6C^high monocyte levels in apoE^−/− mice were 2.9-fold higher compared to WT mice (Figure 6B). Interestingly, Ly-6C^high monocyte levels were equally increased in MΦLRP1^−/− and apoE^−/− MΦLRP1^−/− mice (Figure 6B) suggesting that an apoE-LRP1 interaction regulates blood Ly-6C^high monocytosis. Compared to WT spleens (Figure 6C), the number of total monocytes was 88%, 52%, and 150% higher in MΦLRP1^−/−, apoE^−/−, and apoE^−/−MΦLRP1^−/−spleens, respectively, indicating independent roles for apoE and LRP1 in spleen monocytosis. The number of spleen Ly-6C^high monocytes were markedly increased in MΦLRP1^−/−, apoE^−/−, and apoE^−/−MΦLRP1^−/− mice compared to WT mice (Figure 6D) suggesting in large part that the increased monocytosis was due to accumulation of Ly-6C^high monocytes. Consistent with independent roles of LRP1 and apoE in spleen Ly-6C^high monocytosis, the number of Ly-6C^high monocytes in apoE^−/−MΦLRP1^−/− spleens was 97% higher compared to MΦLRP1^−/− spleens and 141% higher compared to apoE^−/− spleens (Figure 6D). We next examined the distribution of Ly-6C positive cells in lesions from apoE^−/− and apoE^−/−MΦLRP1^−/− mice (Figures 7A and 7B). Interestingly, the lesions of apoE^−/−MΦLRP1^−/− mice contained 3.6-fold more Ly-6C positive cells compared to lesions of apoE^−/− mice, suggesting that LRP1 acts independently of apoE in controlling the accumulation of pro-inflammatory Ly-6C^high monoctyes in the plaque. CCR2 is also a marker for pro-inflammatory monocytes and macrophages¹⁶ and is critical for the recruitment of Ly-6C^high monocytes into the intima²². Consistent with the increased Ly-6C^high monocyte content in apoE^−/−MΦLRP1^−/− lesions, these plaques also contained 2.4-fold more CCR2 positive cells compared to lesions of apoE^−/− mice (Figures 7C and 7D).

Figure 6. Total and Ly-6C^high monocytes in blood and spleens of WT, MΦLRP^−/−, apoE^−/−, and apoE^−/−MΦLRP^−/− mice.

A-D WT (n=10), MΦLRP^−/− (n=8), apoE^−/− (n=10), and apoE^−/−MΦLRP^−/− (n=10) mice were fed a western type diet for 8 weeks and then blood (A-B) and spleen (C-D) total (A,C) and Ly-6C^high monocytes (B,D) were measured by flow cytometry analysis. A-D. *, #, and & denotes statistically significances compared to WT, apoE^−/−, and MΦLRP^−/− mice, respectively, p<0.05, ANOVA with Bonferroni’s post test.

Figure 7. Ly-6C and CCR2 positive cells in lesions of apoE^−/− and apoE^−/−MΦLRP^−/− mice.

A and D Representative images show nuclei (blue, Dapi) and merged images of Ly-6C (A, green) and CCR positive staining (D, red) and nuclei. B and C. Quantitation of the percent Ly-6C (B) and CCR2 (C) positive cells in aortic root sections from apoE^−/− and apoE^−/−MΦLRP^−/− mice (n=6 per group). *p<0.05, Mann-Whitney test.

Discussion

We previously demonstrated that macrophages lacking LRP1 enhance atherogenesis in the setting of increased apoE production^{3, 14}. These results suggested that the established protective effects of apoE are mediated by its interaction with LRP1. The present studies were designed to determine the effects of macrophage LRP1 deletion in the absence of apoE, either from macrophages only or from all cells, with the aim of evaluating whether the atheroprotective effects of LRP1 are independent of its interaction with apoE. Our results show that LRP1 operates independently of apoE in regulating apoptosis, necrosis, and inflammatory Ly-6C^high monocytosis, and that deletion of apoE, locally or systemically, amplifies the atherogenic effects of macrophage LRP1 deletion.

The atheroprotective effects of macrophage LRP1 are independent of apoE

We previously showed that MΦLRP1^−/− BM recipient LDLR^−/− mice develop 40% more atherosclerosis in the proximal aorta compared to LDLR^−/− mice receiving control BM³. Prior to that, we had shown that deletion of macrophage apoE in LDLR^−/− mice increases aortic sinus plaque burden by 75%². These data demonstrate that LRP1 and apoE both exert atheroprotective functions. Given that apoE is a ligand for LRP1, we set out to study whether these proteins work exclusively in an interdependent pathway to maintain vascular integrity. Thus, we examined the effects of combined apoE and LRP1 deletion in macrophages. Our findings that apoE^−/− MΦLRP1^−/−BM→LDLR^−/− mice develop 88% more atherosclerosis than recipients of apoE^−/− BM proves that the combined deletion of these proteins has additive effects, thus favoring the hypothesis that the atheroprotective effects of macrophage LRP1 are largely independent of its interaction with apoE. In addition, the increased atherosclerotic effect of combined macrophage apoE/LRP deletion relative to that of single deletion of macrophage LRP1^{3, 14} may be attributed to the fact that apoE^−/−LRP^−/− macrophages lack the compensatory benefit of increased macrophage apoE expression caused by isolated LRP1 deletion.

Because in LDLR^−/− mice systemic apoE may interact with macrophage LRP1, we also compared atherosclerosis development in apoE^−/− and apoE^−/−MΦLRP1^−/− mice, the latter lacking both systemic and macrophage apoE. Interestingly, apoE^−/−MΦLRP1^−/− mice developed 163% larger lesions than apoE^−/− only mice. This result is consistent with recent studies of Hu and colleagues¹⁰ that showed a 114% increase in apoE^−/−LDLR^−/−MΦLRP1^−/− mice compared to apoE^−/−LDLR^−/− mice. Based on all these studies, we conclude that macrophage LRP1 is atheroprotective regardless of the absence or presence of systemic or macrophage apoE and that other LRP1 ligands and/or signaling events must in part mediate the anti-atherogenic functions of macrophage LRP1. Conversely, apoE influences the process of plaque formation even in the absence of macrophage LRP1, as evidenced by the observations that the effects of LRP deletion on atherosclerosis and lesion apoptosis were much more pronounced in the setting of complete apoE deficiency. This is consistent with studies showing that even small amounts of circulating apoE can decrease atherosclerosis development¹⁷ and that exogenous apoE limits inflammation, apoptosis, and foam cell formation^{19, 23, 24}.

LRP1 expression limits Ly-6C^high monocytosis

Studies of Swirski and colleagues have shown that apoE^−/− mice on a western diet develop enhanced Ly-6C^high monocytosis that results from increased survival and proliferation as well as lower conversion of Ly-6C^high monocytes to Ly-6C^low monocytes.²⁵ A role for lipid loading in this process was suggested by the fact that statins prevented the Ly-6C^high monocytosis whereas incubation with LDL promoted Ly-6C^high monocyte survival²⁵. Our studies demonstrate that deletion of macrophage LRP1 enhances circulating, splenic, and atheroma Ly-6C^high monocytosis (Figures 6B, 6D, and 7B). The observation that Ly-6C^high monocytosis in blood and spleen is similar in MΦLRP1^−/− versus apoE^−/− mice suggests that LRP1 plays a direct role in limiting Ly-6C^high monocytosis as plasma cholesterol levels were much lower in MΦLRP1^−/− mice compared to apoE^−/− mice (101±12 vs. 817±40 mg/dl). Consistent with this concept is the observation that blood and spleen Ly-6C^high monocytosis was markedly enhanced in MΦLRP1^−/− mice compared to WT mice, which had similar plasma cholesterol levels (Figures 6B and 6D). The finding that blood Ly-6C^high monocytosis was similar in MΦLRP1^−/−, apoE^−/−, and apoE^−/−MΦLRP1^−/− mice (Figure 6B) could indicate that an interdependent LRP1/apoE interaction regulates circulating Ly-6C^high monocyte levels, where the absence of the ligand causes an equivalent effect to the absence of its receptor. However, it is worth noting that circulating Ly-6C^high monocytes are maintained at low levels due to their rapid deployment to sites of inflammation^26–28 and their shuttling back to the bone marrow compartment for conversion to Ly-6C^low monocytes²⁶, raising the possibility that the lack of a difference in blood Ly-6C^high monocyte levels is merely a reflection of increased recruitment of monocytes to inflammatory areas in apoE^−/−MΦLRP1^−/− mice. Spleen and lesion Ly-6C^high monocyte levels were indeed markedly increased in apoE^−/−MΦLRP1^−/− mice compared to apoE^−/− mice (Figure 6B), supporting an independent role for LRP1 in controlling Ly-6C^high monocytosis. Splenic Ly-6C^high monocytosis is increased during inflammation,²⁹ and the spleen has been recently proven to serve as the source of Ly-6C^high monocytes for recruitment to acute inflammatory sites, such as ischemic heart muscle²⁸. Our previous studies showed that deletion of macrophage LRP1 enhances the expression of inflammatory cytokines including MCP-1, TNF-α, and IL-1β¹⁴, which could enhance recruitment of Ly-6C^high monocytes to the spleen. In addition, the enhanced Ly-6C^high monocytosis is likely due in part to activation of NF-κβ as increases in monocyte/macrophage survival and proliferation result from NF-κβ activation ^30–32 and our studies demonstrated that expression of LRP1 reduces NF-κβ activation¹². Another possibility is that LRP1 regulates Ly-6C^high monocytosis by influencing cholesterol homeostasis, as cholesterol efflux reduces monocytosis³³ and deletion of LRP1 decreases ABCA1 expression and cholesterol efflux in smooth muscle cells³⁴.

LRP1 prevents macrophage apoptosis independently of apoE

Macrophage apoptosis plays a role in atherosclerosis development¹⁸. Studies have shown that LRP1 mediates the phagocytosis of apoptotic macrophages and promotes cell survival^{11, 14}. ApoE has also been implicated in mediating phagocytic^{14, 19} and anti-apoptotic pathways in cells³⁵. In some cell types, the anti-apoptotic effects of apoE are mediated by LRP1/apoE interaction¹¹. Our in vitro studies show that both LRP1 and apoE deletion increase macrophage apoptosis, and that the combined deletion of LRP1 and apoE produces an additive effect in increasing the number of apoptotic macrophages (Figures 3A and 3B). This observation suggests that apoE and LRP1 have independent roles in macrophage apoptosis. Thus, apoE may protect macrophages from apoptosis through interacting with proteins other than LRP1. As an example, SR-BI has been shown to mediate the phagocytic clearance of apoptotic cells and bind with high affinity to apoE/phospholipid complexes³⁶. However, our recent studies demonstrate that macrophage LRP1 deletion in LDLR^−/− mice promotes the accumulation of lesion apoptotic macrophages by decreasing efferocytosis despite increased apoE secretion compared to WT cells¹⁴. This suggests that in vivo the independent antiapoptotic mechanisms of apoE cannot compensate for the absence of LRP1. Consistent with the in vitro studies, we demonstrate that combined deletion of macrophage LRP1 and apoE in vivo results in markedly increased numbers of lesion apoptotic cells both in the presence of systemic apoE (Figure 3C) and in the complete absence of apoE (Figure 4B). Furthermore, the accumulation of apoptotic macrophages resulting from combined deletion of LRP1 and apoE was associated with impaired efferocytosis (Figure 4C). LRP1 ligands that can regulate apoptotic cell homeostasis in vivo include prosaposin³⁷, α₂-macroglobulin³⁸, β₂ glycoprotein-1³⁹, and calreticulin⁴⁰.

In recent years, it has become clear that macrophage apoptosis has divergent effects on atherosclerosis^{18, 41} depending on the balance between generation of apoptotic cells and efficient phagocytic clearance. Studies by our group and others have shown that in the early fatty streak phase^{18, 42, 43}, modest increases in apoptotic macrophages coincide with decreased atherosclerosis due to efficient efferocytosis. In later lesion stages, the accumulation of apoptotic cells has been proposed to promote plaque instability⁴¹. The non-internalized apoptotic cells secrete inflammatory cytokines^{14, 44}, driving more post-apoptotic necrosis resulting in plaque instability^{14, 41}. Consistent with the accumulation of inflammatory apoptotic macrophages promoting plaque instability, the necrotic area was markedly increased in lesions of apoE^−/−MΦLRP1^−/− mice compared to lesions in apoE^−/− mice (Figure 5) after only 8 weeks of consuming a western diet.

Besides impacting plaque stability, the excessive accumulation of apoptotic macrophages in lesions of mice with combined deficiency of apoE and LRP likely contributes to the accelerated atherosclerosis development by facilitating the recruitment of monocytes. Studies have demonstrated that apoptotic macrophages contain oxidized phospholipid, which promote secretion of inflammatory cytokines, endothelial cell activation, and recruitment of monocytes^18,21. The current studies show that the combined deficiency of macrophage apoE and LRP1 enhance the macrophage content in lesions of both LDLR^−/− and apoE^−/− mice. Consistent with this possibility also is the finding that the lesions of apoE^−/−MΦLRP1^−/− contained more Ly-6C^high monocytes (Figure 7B). Compared to Ly-6C^low monocytes, Ly-6C^high monocytes preferentially bind to activated endothelium and migrate into atherosclertic lesions¹⁵. CCR2 is a marker for pro-inflammatory monocytes and macrophages¹⁶ and is critical for the recruitment of Ly-6C^high monocytes into the intima¹⁵. Thus, the increased content of CCR2 positive cells (Figure 7C) in apoE^−/−MΦLRP1^−/− lesions is also consistent with enhanced recruitment of pro-inflammatory Ly-6C^high monocytes. Other studies have shown that accumulation of lesion apoptotic macrophages is associated with enhanced atherosclerosis progression and monocyte recruitment^{18, 45, 46}, particularly in the presence of inflammation resulting from defective efferocytosis^{20, 47, 48}.

In conclusion, our study demonstrates that macrophage LRP1 slows atherosclerosis development by limiting lesion macrophage death and Ly-6C^high monocytosis. These atheroprotective functions occur via mechanisms that are independent of its interaction with apoE. Thus, apoE and LRP1, two abundant atheroma proteins that physically interact with each other to regulate cell survivial, remnant uptake, and activate downstream signaling,^{7, 9, 11} actually influence atherogenesis through non-redundant pathways.

Clinical Perspective.

ApoE is a plasma protein that regulates both clearance of VLDL and maturation of HDL. It is also expressed at high levels by macrophages, and found to have strong anti-atherogenic effects in mouse models. In humans, HDL-associated apoE correlates with presence of CAD and may become a biomarker for this common disease. ApoE binds to multiple receptors, including LRP1, a member of the LDL receptor family. LRP1 binds multiple ligands and can both internalize cargo and trigger signaling-mediated downstream effects. Both proteins control cellular cholesterol trafficking and plaque volume via regulation of cell death. These functions are key targets for the development of therapeutic strategies aiming at inducing plaque regression, an elusive and highly prized objective. We previously determined that macrophages lacking LRP1 cause accelerated atherosclerosis, a paradoxical finding given that in these cells: a) atherogenic lipoproteins are internalized with reduced efficiency; b) secretion of apoE is significantly up-regulated. Since this observation was made in mice expressing normal amounts of systemic and macrophage apoE, it was not possible to determine whether the negative effect of LRP1 removal was either caused by the interruption of an “apoE-LRP1 axis” or attenuated by the over-expression of apoE. The current studies clearly show that most functions of apoE and LRP1 in the artery wall occur through mutually independent pathways, and that the absence of apoE greatly magnifies the effects of LRP1 deficiency on cell death. Our results help understand the forces controlling plaque volume expansion or contraction, and may inform development of regression-inducing agents.