Apolipoprotein A1 Protects Against Necrotic Core Development in Atherosclerotic Plaques: PDZK1-Dependent High-Density Lipoprotein Suppression of Necroptosis in Macrophages

Abstract

Background: Atherosclerosis is a chronic disease affecting artery wall and a major contributor to cardiovascular diseases. Large necrotic cores increase risk of plaque rupture leading to thrombus formation. Necrotic cores are rich in debris from dead macrophages. Programmed necrosis (necroptosis) contributes to necrotic core formation. HDL (high-density lipoprotein) exerts direct atheroprotective effects on different cells within atherosclerotic plaques. Some of these depend on the SR-B1 (scavenger receptor class B type I) and the adapter protein PDZK1 (postsynaptic density protein/Drosophila disc-large protein/Zonula occludens protein containing 1). However, a role for HDL in protecting against necroptosis and necrotic core formation in atherosclerosis is not completely understood.

Methods: Low-density lipoprotein receptor-deficient mice engineered to express different amounts of ApoA1 (apolipoprotein A1), or to lack PDZK1 were fed a high fat diet for 10 weeks. Atherosclerotic plaque areas, necrotic cores, and key necroptosis mediators, RIPK3 (receptor interacting protein kinase 3), and MLKL (mixed lineage kinase domain-like protein) were characterized. Cultured macrophages were treated with HDL to determine its effects, as well as the roles of SR-B1, PDZK1, and the PI3K (phosphoinositide 3-kinase) signaling pathway on necroptotic cell death.

Results: Genetic overexpression reduced, and ApoA1 knockout increased necrotic core formation and RIPK3 and MLKL within atherosclerotic plaques. Macrophages were protected against necroptosis by HDL and this protection required SR-B1, PDZK1, and PI3K/Akt pathway. PDZK1 knockout increased atherosclerosis in LDLR^KO mice, increasing necrotic cores and phospho-MLKL; both of which were reversed by restoring PDZK1 in BM-derived cells.

Conclusions: Our findings demonstrate that HDL in vitro and ApoA1, in vivo, protect against necroptosis in macrophages and necrotic core formation in atherosclerosis, suggesting a pathway that could be a target for the treatment of atherosclerosis.

Keywords: atherosclerosis; cardiovascular diseases; high-density lipoprotein; macrophages; necroptosis.

Figure 1.

Effects of genetically varying ApoA1 (apolipoprotein A1) levels in low-density lipoprotein receptor (LDLR) KO mice on atherosclerotic plaque and necrotic core sizes and levels of phosphorylated RIPK3 (receptor interacting protein kinase 3) and MLKL (mixed lineage kinase domain-like protein). Male and female ApoA1^WT/WT/LDLR^KO/KO, ApoA1^KO/KO/LDLR^KO/KO, or hApoA1^TG/TG/LDLR^KO/KO mice (10-week-old, n=10/group) were challenged with a high-fat diet for 10 weeks. A, Representative images of H&E-stained atherosclerotic plaques from ApoA1^KO/KOLDLR^KO/KO, ApoA1^WT/WTLDLR^KO/KO, and hApoA1^TG/TGLDLR^KO/KO males and females. B and C, Quantification of plaque volume for males and females. D and E, Quantification of necrotic core area for males and females. F, Representative images of phospho-RIPK3 (green), phospho-MLKL (red), and DAPI (blue)-stained atherosclerotic plaques of male (top 2 rows) and female (bottom 2 rows) ApoA1^KO/KOLDLR^KO/KO (left column), ApoA1^WT/WTLDLR^KO/KO (middle column), and hApoA1^TG/TGLDLR^KO/KO (right column). Yellow dashed lines mark atherosclerotic plaques. G and H, Quantification of phosphorylation levels of RIPK3 and MLKL in males and females (n=5 mice/group). Data represent mean±SEM, n=5–10 biological replicates. B through E, Statistical analysis was done using 1-way ANOVA followed by Sidak multiple comparisons test. G and H, Statistical analysis was done using 2-way ANOVA followed by Tukey multiple comparisons test.

Figure 2.

HDL (high-density lipoprotein) mediates protection of cultured peritoneal macrophages against TNFα (tumor necrosis factor alpha) and h-oxLDL (highly-oxidized low-density lipoprotein)-induced necroptosis and requires SR-B1 (scavenger receptor class B type 1), and the adapter protein PDZK1 (postsynaptic density protein/Drosophila disc-large protein/Zonula occludens protein containing 1). Thioglycollate-elicited peritoneal macrophages were prepared from wild-type (WT), SR-B1 KO, and PDZK1 KO mice and treated as indicated in culture. A–C, Cells were cultured in lipoprotein-deficient medium and treated with h-oxLDL (50μg/mL) and ZVAD (50 µM), Nec-1s (12.5–800 µM), HDL (6.25–100 µg protein/mL; gray bars) or DMSO as a vehicle (control-black bars) for 24 hours. A, Representative images of nuclei from necrotic macrophages detected by propidium iodide (PI) staining (red) before fixation; nuclei were counterstained with DAPI (blue) after fixation. B and C, Quantification of the percentage of PI positive macrophage nuclei after 24-hour treatment in culture with different concentrations of Nec-1s or HDL, respectively. Data are means±SEM, n=4 biological replicates (cells isolated from different mice). Cells were cultured in lipoprotein-deficient medium and treated with TNFα (100 nM), and ZVAD (50 µM), with or without HDL (50 µg protein/mL) or DMSO as a vehicle (control) for 24 hours. Representative images (D) and quantification (E) of propidium iodide. Representative images (F) and quantification (G) of FITC-Annexin V-stained cells. Data are means±SEM, n=4 biological replicates (cells isolated from different mice). F, Cells were cultured in lipoprotein-deficient medium and treated for 24 hours with HDL (50 µg protein/mL). Then, cells were washed with PBS and media was replaced without HDL. Cells were treated with TNFα (100 nM) and ZVAD (50 µM) or DMSO as a vehicle (control), and harvested 6, 9, and 24 hours after treatment and stained with PI, fixed and then stained with DAPI. H, Representative images and quantification (I) of propidium iodide after 24 hours of treatment. Data are means±SEM, n=4 biological replicates (cells isolated from different mice). (J-M) Peritoneal macrophages from WT, SR-B1 KO, SR-B1 deltaCT, and PDZK1 KO were treated with TNFα (100 nM), ZVAD (50 µM), in the presence or absence of HDL (50 µg protein/mL) for 24 hours. After treatment, the supernatant from cells were subjected to the LDH assay to evaluate cytotoxicity, following the manufacture’s protocol. Graphics shows the percentage of cytotoxicity. Data are means±SEM, n=4 biological replicates (cells isolated from different mice). Statistical analysis was done using 1-way ANOVA followed by Sidak multiple comparisons test.

Figure 3.

HDL (high-density lipoprotein) protection against TNFα (tumor necrosis factor alpha)-induced necroptosis in macrophages involves activation of the PI3K (phosphoinositide 3-kinase)/Akt pathway. Thioglycollate-elicited peritoneal macrophages prepared from wild-type (WT) mice and cultured in lipoprotein-deficient medium were treated for 24 hours with TNFα (100 nM), ZVAD (50 µM), in the presence or absence of HDL (50 µg protein/mL) and the inhibitors AktV (3 µM), bpV-pic (100 nM), or DMSO as a vehicle (control). Cells were lysed and subjected to SDS-PAGE for immunoblotting. A, Representative immunoblots of phospho and total Akt, TAK1 (transforming growth factor beta-activated kinase 1), TBK1 (TANK-binding kinase 1), RIPK3 (receptor interacting protein kinase 3), MLKL (mixed lineage kinase domain-like protein), and β actin. B–F, Quantification of the ratios of phospho-/total-Akt (B), -TAK1 (C), -TBK1 (D), -RIPK3 (E), -MLKL (F) band intensities. Quantification was done as relative to UT group (mean±SEM, n=3, biological replicates consisting of cells isolated from different mice). UT (untreated), TZ (TNFα+ZVAD), TZH (TNFα+ZVAD+HDL), TZA (TNFα+ZVAD+AktV), TZAH (TNFα+ZVAD+AktV+HDL), TZB (TNFα+ZVAD+bpV-PIC), TZBH (TNFα+ZVAD+bpV-PIC+HDL). G and H, Cells were cultured in lipoprotein-deficient medium and treated for 24 hours with TNFα (100 nM) and ZVAD (50 µM), in the absence or presence of HDL (50 µg protein/mL) and the inhibitors AktV (3 µM), LY294002 (10 µM), bpV-pic (100 nM), or DMSO as a vehicle (control). Cells were then stained with PI, fixed and stained with DAPI. G, Representative images and (H) quantification of cell death as the percentage of PI positive macrophage nuclei. Data are means±SEM, n=4, biological replicates consisting of cells isolated from different mice. Statistical analysis was done using 1-way ANOVA followed by Sidak multiple comparisons test.

Figure 4.

The absence of PDZK1 (postsynaptic density protein/Drosophila disc-large protein/Zonula occludens protein containing 1) increases atherosclerotic lesion and necrotic core area, as well as the phosphorylation levels of MLKL (mixed lineage kinase domain-like protein) in atherosclerotic plaques. Male and female PDZK1^WT/WT/LDLR^KO/KO and PDZK1^KO/KO/LDLR^KO/KO mice (10-week-old, n=10/group) were fed the atherogenic HFD for 10 weeks. A, Representative images of H&E-stained aortic sinus atherosclerotic plaques from males and females. Quantification of (B and C) plaque volume and (D and E) necrotic core area for males and females. F, Representative images of phospho-MLKL (red) and DAPI (blue)-stained atherosclerotic plaques from PDZK1^WT/WT/LDLR^KO/KO and PDZK1^KO/KO/LDLR^KO/KO males and females. Yellow dashed lines mark atherosclerotic plaques. G and H, Quantification of pMLKL immunofluorescence in atherosclerotic plaques in males and females, respectively. n=5 biological replicates (individual mice) per group. Data were first subjected to the D’Agostino-Pearson test for normality and for equal variances. Statistical analysis was done using unpaired Student t-test. Data are means±SEM.

Figure 5.

Restoring PDZK1 (postsynaptic density protein/Drosophila disc-large protein/Zonula occludens protein containing 1) in BM-derived cells decreases atherosclerotic lesion and necrotic core areas, and phosphorylation levels of MLKL (mixed lineage kinase domain-like protein) and RIPK3 (receptor interacting protein kinase 3) in atherosclerotic plaques of PDZK1^KO/KOLDLR^KO/KO mice. Male and female (10-week-old, n=6–7/group) PDZK1^KO/KOLDLR^KO/KO transplanted with PDZK1^WT/WTLDLR^KO/KO or PDZK1^KO/KOLDLR^KO/KO BM were fed the atherogenic HFD for 10 weeks. A, Representative images of H&E-stained aortic sinus atherosclerotic plaques. Quantification of (B and D) plaque volumes and (C and E) necrotic core areas for males (B and C) and females (D and E), respectively. F through J, Immunofluorescence staining for RIPK3 and MLKL phosphorylation in atherosclerotic plaques. F and I, Representative images of DAPI (blue), phospho-RIPK3 (green), phospho-MLKL (red), and merged images of atherosclerotic plaques from male (F) and female (I) PDZK1^KO/KOLDLR^KO/KO mice transplanted with PDZK1^WT/WTLDLR^KO/KO or PDZK1^KO/KOLDLR^KO/KO BM. Yellow dashed lines mark atherosclerotic plaques. G, H, J, and K, Quantification of pRIPK3 and pMLKL immunofluorescence. n=6–7 individual mice (biological replicates) per group. Data were first subjected to the D’Agostino-Pearson test for normality and for equal variances. Data are means±SEM. Statistical analysis was done using unpaired Student t-test.