The critical role of IL-1 receptor-associated kinase 4-mediated NF-κB activation in modified low-density lipoprotein-induced inflammatory gene expression and atherosclerosis

Abstract

Exciting discoveries related to IL-1R/TLR signaling in the development of atherosclerosis plaque have triggered intense interest in the molecular mechanisms by which innate immune signaling modulates the onset and development of atherosclerosis. Previous studies have clearly shown the definitive role of proinflammatory cytokine IL-1 in the development of atherosclerosis. Recent studies have provided direct evidence supporting a link between innate immunity and atherogenesis. Although it is still controversial about whether infectious pathogens contribute to cardiovascular diseases, direct genetic evidence indicates the importance of IL-1R/TLR signaling in atherogenesis. In this study, we examined the role of IL-1R-associated kinase 4 (IRAK4) kinase activity in modified low-density lipoprotein (LDL)-mediated signaling using bone marrow-derived macrophage as well as an in vivo model of atherosclerosis. First, we found that the IRAK4 kinase activity was required for modified LDL-induced NF-κB activation and expression of a subset of proinflammatory genes but not for the activation of MAPKs in bone marrow-derived macrophage. IRAK4 kinase-inactive knockin (IRAK4KI) mice were bred onto ApoE(-/-) mice to generate IRAK4KI/ApoE(-/-) mice. Importantly, the aortic sinus lesion formation was impaired in IRAK4KI/ApoE(-/-) mice compared with that in ApoE(-/-) mice. Furthermore, proinflammatory cytokine production was reduced in the aortic sinus region of IRAK4KI/ApoE(-/-) mice compared with that in ApoE(-/-) mice. Taken together, our results indicate that the IRAK4 kinase plays an important role in modified LDL-mediated signaling and the development of atherosclerosis, suggesting that pharmacological inhibition of IRAK4 kinase activity might be a feasible approach in the development of antiatherosclerosis drugs.

Figure 1. Total cholesterol and plasma lipoprotein distribution in 16-week-old apoE −/− or IRAK4 KI/ApoE −/− mice

A. Total cholesterol in 16-week-old apoE−/− or IRAK4KI/apoE−/− mice B. FPLC elution profiles of cholesterol from apoE −/− and IRAK4 KI/apoE−/− mice. Arrows indicate the elution peak for each lipoprotein class: very low density lipoprotein (VLDL), low density lipoprotein (LDL) and high density lipoprotein (HDL).

Figure 2. Deficiency of IRAK4 kinase activity greatly reduced aortic root lesion area in Apoe −/− mice

A. Representative photographs of aortic sinus plaques from IRAK4 KI/Apoe−/− (lower) compared with Apoe−/− (upper) mice. ApoE−/− and IRAK4 KI/ApoE−/− mice were fed with normal chow diet for 18 weeks after weaning. Cross sections were stained with Oil Red O for neutral lipid (4X objective lens). B. Lesion area in ApoE−/− and IRAK4 KI/IRAK4 KI mice on 18 weeks of chow diet after weaning. Mean ± S.D.; p<0.00005

Figure 3. Reduced inflammatory gene expression in arterial tissue of ApoE/IRAK4 KI

(A–E) At age 18 weeks, apoE−/− (n=6), IRAK4KI/apoE−/− (n=6), and control B6 mice (n=2) were sacrificed and aortas retrieved. Total RNA from aorta was analyzed by quantitative RT-PCR of specific primer pairs for MCP1 (A), CCL4 (B), IL-6 (C), IP-10 (D) and IL-1β (E). Shown is the expression of MCP1, CCL4, IL-6, IP-10 and IL-1β normalized to beta-actin expression. The -fold induction was calculated as compared with the expression of WT untreated cells (set as 1-fold).

Figure 4. Impaired acLDL-mediated gene expression in macrophages from IRAK4 kinase-inactive knock-in mice

A–B. Bonemarrow-derived macrophages of wild-type and IRAK4 kinase-inactive knock-in mice were untreated or treated with acLDL (100 μg/ml) (A) or NO2-LDL (B) for 30 and 120 min. Total RNA was subjected to RT-PCR in order to measure the relative expression of TNFα (A), MIP-2 (B) and KC (C). Shown is the expression of TNFα, MIP-2 and KC normalized to beta-actin expression. The fold induction was calculated as compared with the expression of WT untreated cells.

Figure 5. Illumina mRNA expression profiling of IRAK4 kinase activity-dependent genes

A. Heatmap of the genes that were induced only in wild-type bonemarrow-derived macrophage, but not in the IRAK4 KI macrophage upon 24 hours of acLDL (100 μg/ml) stimulation. B. Quantitative Real-time PCR of selected genes from Figure 6 A. Wild-type, IRAK4 kinase-inactive knock-in and IRAK4-deficient macrophages were either untreated or stimulated with acLDL for 24 hours. The fold change was calculated compared with the expression of untreated samples.

Figure 6. Illumina mRNA expression profiling of IRAK4 kinase activity-independent genes

A. Heatmap of the genes that were induced similarly (>1.5 fold) in both wild-type and IRAK4 kinase inactive knock-in macrophage upon 24 hours of acLDL (100 μg/ml) stimulation. B. Quantitative Real-time PCR of selected genes from Figure 6 A. Wild-type, IRAK4 kinase-inactive knock-in and IRAK4-deficient macrophages were either untreated or stimulated with acLDL for 24 hours. The fold change was calculated compared with the expression of untreated samples (set as value 1)

Figure 7. SR-A/CD36/TLR2/TLR4 is partially required for acLDL-mediated gene transcriptions

A. Bonemarrow-derived macrophages of wild-type, TLR2, and TLR4 deficient mice were untreated or treated with acLDL (100 μg/ml) for 30 and 120 min. Total RNA was subjected to RT-PCR in order to measure the relative expression of TLR2, TLR4, MCP-1, TNFα, KC and IL-1β. Shown is the expression of TLR2, TLR4, MCP-1, TNFα, KC and IL-1β normalized to beta-actin expression. The -fold induction was calculated as compared with the expression of WT untreated cells. B. Bonemarrow-derived macrophages of wild-type and SR-A deficient mice were untreated or treated with acLDL (100 μg/ml) for 30 and 120 min. Total RNA was subjected to RT-PCR in order to measure the relative expression of MCP-1, TNFα, KC and IL-1β. Shown is the expression of MCP-1, TNFα, KC and IL-1β normalized to beta-actin expression. The -fold induction was calculated as compared with the expression of WT untreated cells. C. Bonemarrow-derived macrophages of wild-type and CD36 deficient mice were untreated or treated with acLDL (100 μg/ml) for 30 and 120 min. Total RNA was subjected to RT-PCR in order to measure the relative expression of MCP-1, TNFα, KC and IL-1β. Shown is the expression of MCP-1, TNFα, KC and IL-1β normalized to beta-actin expression. The -fold induction was calculated as compared with the expression of WT untreated cells.

Figure 8. The requirement of IRAK4 kinase activity in NFkB activation

A. Cell lysates from WT and IRAK4 KI bone marrow-derived macrophages that were either untreated or treated with acLDL (100 μg/ml) for the indicated times were analyzed by Western analysis with antibodies against anti-IκBα and anti-p-IκBα B. Cell lysates from WT and IRAK4 KI bone marrow-derived macrophages that were either untreated or treated with acLDL (100 μg/ml) for the indicated times were analyzed by Western analysis with antibodies against anti-p-AKT, anti-p-JNK, anti-p-ERK, anti-p-p38, anti-IRAK2, and anti-β-actin. C. Cell lysates from WT and IRAK4 KI bone marrow-derived macrophages that were either untreated or treated with NO2-LDL (100 μg/ml) for the indicated times were analyzed by Western analysis with antibodies against anti-IκBα and anti-p-IκBα D. Cell lysates from WT and IRAK4 KI bone marrow-derived macrophages that were either untreated or treated with acLDL (100 μg/ml) for the indicated times were analyzed by Western analysis with antibodies against anti-IRAK1, anti-p-JNK, anti-p-ERK, anti-p-p38, anti-IRAK2, and anti-GAPDH. E. Cell lysates from WT and IRAK4 KI bone marrow-derived macrophages that were either untreated or treated with acLDL (100 μg/ml) for the indicated times were analyzed by electrophoretic mobility shift assay with an NFκB specific probe F. Densitometry of NFκB electrophoretic mobility shift assay results in C using the NIH Image software package. Similar results were obtained in three separate experiments.