Oxidized LDL induces phosphorylation of non-muscle myosin IIA heavy chain in macrophages.

Oxidized LDL (oxLDL) performs critical roles in atherosclerosis by inducing macrophage foam cell formation and promoting inflammation. There have been reports showing that oxLDL modulates macrophage cytoskeletal functions for oxLDL uptake and trapping, however, the precise mechanism has not been clearly elucidated. Our study examined the effect of oxLDL on non-muscle myosin heavy chain IIA (MHC-IIA) in macrophages. We demonstrated that oxLDL induces phosphorylation of MHC-IIA (Ser1917) in peritoneal macrophages from wild-type mice and THP-1, a human monocytic cell line, but not in macrophages deficient for CD36, a scavenger receptor for oxLDL. Protein kinase C (PKC) inhibitor-treated macrophages did not undergo the oxLDL-induced MHC-IIA phosphorylation. Our immunoprecipitation revealed that oxLDL increased physical association between PKC and MHC-IIA, supporting the role of PKC in this process. We conclude that oxLDL via CD36 induces PKC-mediated MHC-IIA (Ser1917) phosphorylation and this may affect oxLDL-induced functions of macrophages involved in atherosclerosis.

INTRODUCTION

There have been extensive studies proving that low-density lipoprotein (LDL) is oxidatively modified in vivo and this modification provokes proinflammatory and proatherogenic responses (1-4). Oxidative modification increases atherogenicity of LDL by facilitating uptake and retention of circulating LDL in the arterial wall. Oxidized LDL (oxLDL) induces adhesion molecule expression in endothelial cells, cytokine secretion by monocytes/macrophages, and smooth muscle cell proliferation, which are hallmarks of the atherosclerotic process (5-9). Macrophages internalize oxLDL via scavenger receptors including CD36 and scavenger receptor A (SRA) and turn into foam cells that recruit inflammatory cell infiltrates into the arterial wall (10). However, the precise mechanism for oxLDL uptake has not been completely defined. OxLDL is also known to induce endothelial dysfunction (11). Alterations in the structural and functional integrity of the endothelial barrier allow a net influx of LDL from the circulation into the subendothelial space. However, the mechanism by which oxLDL induces endothelial dysfunction has not been fully defined.

The cytoskeleton is a cellular network of structural, signaling, and adaptor molecules that regulate most cellular functions such as migration, ligand recognition, signal activation and endocytosis/phagocytosis (12). There have been reports that oxLDL affects the cytoskeletal functions of various cell types involved in atherosclerosis. OxLDL drives endothelial cell stiffness and increases pinocytotic activity of endothelial cells through cytoskeletal reorganization (13-15). In macrophages, oxLDL facilitates actin polymerization and spreading and this process inhibits macrophage migration, promoting macrophage trapping (16, 17). In our recent report, oxLDL via CD36 was shown to inhibit macrophage migration through loss of cell polarity by inhibiting non-muscle myosin II (NMII) activity (18). Therefore, the cytoskeletal modulating activity of oxLDL should be a key mechanism that drives cellular dysfunction.

Myosins are molecular motor proteins that crosslink and translocate actin filaments, using energy from ATP hydrolysis. NMII molecules are composed of three pairs of peptides including two heavy chains of 230 kDa, two regulatory light chains (RLC) of 20 kDa, which are known to regulate NMII activity, and two 17 kDa essential light chains stabilizing the heavy chain structure (19, 20). NMII regulates cellular protrusion, polarity, migration and integrin-mediated adhesion (21-23). In addition to the motor function of NMII, the newly recognized functions of NMII involve internalization of epidermal growth factor receptor (EGFR) and the downstream signaling provoked by EGFR (24) and maturation of the immunological synapse (25), suggesting more diverse functions of NMII that warrant investigation.

NMII activity and the assembly of NMII filaments are known to be regulated by RLC phosphorylation, which is controlled by small molecular weight G proteins such as Rho and Rac (26). Whereas roles of RLC phosphorylation have been extensively studied, roles of phosphorylation events on non-muscle myosin heavy chain (MHC) have not been characterized.

In the current study, we demonstrate that interaction between oxLDL and CD36 induces MHC-IIA phosphorylation in macrophages and the MHC-IIA phosphorylation is mediated by protein kinase C (PKC). We expect that this finding may underly a crucial mechanism by which CD36 mediates uptake of oxLDL and provokes subsequent signaling for pro-inflammatory action. This also suggests an additional mechanism for macrophage trapping in atherosclerotic inflammatory lesions.

RESULTS

OxLDL induces MHC phosphorylation (Ser1917)

Based on our previous report showing that oxLDL inhibits NMII activity by dephosphorylating RLC (Thr18/Ser19), and that NMIIA is the dominant isoform expressed in macrophages (18), we tested if oxLDL affects MHC phosphorylation. Western blot detection of phosphorylated MHC-IIA (Ser1917) showed that oxLDL induced a significant increase in MHC-IIA phosphorylation (Ser1917), while native LDL (nLDL) did not induce an effect in murine peritoneal macrophages (Fig. 1). Another MHC-IIA residue (Ser1948) known to be phosphorylated by casein kinase 2 (CK2) (27) was not affected by oxLDL in our western blot analysis (data not shown).

Fig. 1

OxLDL induces phosphorylation of MHC-IIA (Ser1917) in macrophages. (A) Murine peritoneal macrophages were treated with or without nLDL (50 μg/ml) or oxLDL (50 μg/ml) for the indicated times and the cell lysates were analyzed by western blotting to detect phospho-MHC-IIA (Ser1917). Levels of p-MHC-IIA (Ser1917) were normalized to β-actin and are expressed as fold increases relative to the control (untreated macrophages) arbitrarily set at 1 (100%). (B) Quantitative analysis for the Western blots in (A). Scale bars are mean ± S.E.M. *P ＜ 0.05, **P ＜ 0.01, (Student’s t-test). The western blot was repeated three times.

OxLDL-induced MHC phosphorylation (Ser1917) depends on CD36

We tested if oxLDL-induced MHC phosphorylation (Ser1917) is mediated by CD36. We incubated peritoneal macrophages isolated from wild-type and Cd36 null mice with oxLDL and analyzed the cell lysates by western blotting for phosphorylated MHC-IIA (Ser1917). Our Western blot analysis showed that oxLDL did not induce MHC-IIA phosphorylation (Ser1917) in Cd36 null macrophages, while there was MHC-IIA phosphorylation (Ser1917) in wild-type macrophages in response to oxLDL (Fig. 2). The results suggest that oxLDL-induced phosphorylation of MHC is mediated by CD36.

Fig. 2.

OxLDL-induced phosphorylation of MHC-IIA (Ser1917) depends on CD36. (A) Peritoneal macrophages from wild-type and Cd36 null mice were treated with oxLDL (50 μg/ml) for indicated times and the cell lysates were analyzed by western blotting to detect phospho-MHC-IIA (Ser1917). Levels of p-MHC-IIA (Ser1917) were normalized to β-actin and are expressed as fold increases relative to the control (untreated macrophages) arbitrarily set at 1 (100%). (B) Quantitative analysis of the Western blots in (A). Scale bars are mean ± S.E.M. *P ＜ 0.05, **P ＜ 0.01, (Student’s t-test). The western blot was repeated three times.

OxLDL-induced MHC phosphorylation (Ser1917) is mediated by protein kinase C (PKC)

Previous reports have shown that MHC-IIA (Ser1917) phosphorylation is mediated by PKC (28, 29) and oxLDL activates PKC in macrophages (30, 31). To test if PKC is involved in MHC-IIA phosphorylation (Ser1917) by oxLDL, we blocked PKC activation by using bisindolylmaleimide-I (GF109203X), which is known to inhibit PKC. Our data revealed that PKC inhibitor-treated macrophages did not show oxLDL-induced MHC phosphorylation (Ser1917), while untreated macrophages had an oxLDL-induced increase in phospho-MHC-IIA (Ser1917) (Fig. 3A).

Fig. 3.

OxLDL-induced phosphorylation of MHC-IIA (Ser1917) is mediated by PKC. (A) Murine peritoneal macrophages untreated or pre-treated with bisindolylmaleimide-I (10 μM) were treated with or without oxLDL (50 μg/ml) for the indicated times. The cell lysates were analyzed by westen blot to detect phospho-MHC-IIA (Ser1917). Levels of p-MHC-IIA (Ser1917) were normalized to β-actin and are expressed as fold increases relative to the control (untreated macrophages)arbitrarily set at 1 (100%). (B) MHC-IIA protein was pulled down from murine macrophages incubated with or without oxLDL (50 μg/ml) for the indicated times using anti-MHC-IIA antibody. The immunoprecipitates were separated by SDS-PAGE gel electrophoresis and blotted for PKC-βII detection. (C) Quantitative analysis of the immunoprecipitation experiments performed as in (B). Affinity between MHC-IIA and PKC-βII was calculated and plotted in the graph. Scale bars are mean ± S.E.M. *P ＜ 0.05 (Student’s t-test). The experiment was repeated three times.

Based on the report showing that MHC IIA is phosphorylated (Ser 1917) by PKC-βII in mast cells (29), we tested if PKC-βII mediates MHC-IIA phosphorylation in macrophages. We used immunoprecipitation to evaluate changes in the physical association between MHC-IIA and PKC-βII. Our data revealed that oxLDL increased the physical association between MHC-IIA and PKC-βII by 9.8-fold (Fig. 3B) suggesting the involvement of PKC-βII in the MHC-IIA phosphorylation.

OxLDL induced MHC phosphorylation (Ser 1916) in THP-1 cells

We tested if oxLDL-induced MHC-IIA phosphorylation occurs in the human monocytic cell line, THP-1. THP-1 cells that had differentiated into macrophage-like cells were incubated with oxLDL and the cell lysates were analyzed by western blot for phospho-MHC IIA (Ser1916), which is the Ser1917-accordant site for human MHC-IIA. The Western blot analysis confirmed that oxLDL induced a 3-fold increase in phospho-MHC-IIA (Ser 1916) in THP-1-derived macrophage-like cells (Fig. 4A). As shown in murine macrophages, bisindolylmaleimide-I treatment blocked the oxLDL-induced phosphorylation of MHC-IIA (Ser 1916) in THP-1-derived macrophage-like cells (Fig. 4B), confirming the involvement of PKC.

Fig. 4.

OxLDL induced phosphorylation of MHC-IIA (Ser1917) in THP-1 cells. (A) THP-1 cells differentiated into macrophage-like cells were untreated or treated with oxLDL (50 μg/ml) for indicated times and the cell lysates were analyzed by western blotting to detect phospho-MHC-IIA (Ser1916). Levels of p-MHC-IIA (Ser1916) were normalized to β-actin and are expressed as fold increases relative to the control (untreated macrophages) arbitrarily set at 1 (100%). (B) THP-1 cells differentiated into macrophage-like cells were pre-treated (left panels) and untreated (right panels) with bisindolylmaleimide-I (10 μM) and exposed to oxLDL (50 μg/ml) for indicated times. Western blotting for phospho-MHC-IIA (Ser1916) was performed with the cell lysates. Levels of p-MHC-IIA (Ser1916) were normalized to β-actin and are expressed as fold increases relative to the control (untreated macrophages) arbitrarily set at 1 (100%).

DISCUSSION

OxLDL resides in the subendothelial space of atherosclerotic arteries and in plasma of patients with metabolic disorders (32-35), performing various pathological functions. Therefore, understanding the role of oxLDL may help define the pathophysiology of the metabolic diseases including atherosclerosis, and provide a new therapeutic strategy.

CD36 is a class B scavenger receptor expressed in various cell types including monocyte/macrophages, endothelial cells, platelets and adipocytes (36, 37). In atherosclerosis, CD36 mediates macrophage oxLDL uptake and promotes pro-inflammatory functions. CD36-mediated signaling events, including activation of src kinases and mitogen-activated protein kinases, mediate the internalization of oxLDL (38, 39). However, the question of which molecule is used as the cargo for the oxLDL transport, remains unclear. There have been obvious discrepancies in studies of the oxLDL uptake mechanism. One of the widely accepted views for the oxLDL uptake mechanism is that oxLDL is internalized through macropinocytosis. This is based on a previous observation that modified LDL is associated with ruffles and resides in macropinosomes of macrophages (40). However, Zeng et al. (41) showed that internalized oxLDL and CD36 were found in moderately-sized cytoplasmic structures co-localizing with a glycosylphosphatidylinositol-anchored protein, suggestive of uptake via lipid raft endocytosis. Moreover, Sun et al. (42) reported that oxLDL uptake by CD36 is independent of actin, but depends on dynamin. However, Trimble et al. demonstrated an actin dependency of the oxLDL uptake, which is distinct from macropinocytosis (39). Regarding the critical functions of foam cells in the atherogenic process, elucidation of the oxLDL uptake mechanism should be a primary focus of future studies.

Our current study reveals that oxLDL via CD36 induces phosphorylation of MHC-IIA (Ser1917). Unlike RLC phosphorylation, there have been few reports elucidating the roles of MHC phosphorylation. Various phosphorylation sites have been identified near the C-termini of MHCs, in both the coiled-coil domain and the non-helical tail, including sites that are phosphorylated by PKC (43), CK2 (27, 44) and transient receptor potential melastatin 7 (TRPM7) (45). MHC phosphorylation in each residue has been shown to dissociate myosin filaments or prevent filament formation in vitro. MHC phosphorylation (Ser1917) by PKC inhibits the assembly of NMIIA rods into filaments (44). Intriguingly, PKC-βII-induced MHC phosphorylation (Ser 1917) contributes to exocytosis in mast cells (29). In our data, oxLDL increases physical association between MHC-IIA and PKC-βII (Fig. 3B), suggesting involvement of a PKC-βII. As mast cells, macrophages secrete various cytokines in response to oxLDL, and thus, it may be worthwhile to test possible roles of MHC phosphorylation in macrophage exocytosis.

Recent studies of NMII have revealed that NMII is involved in various cellular functions beyond serving as a structural protein in cytoskeletal arrangement. Kim et al. showed that NMII mediates internalization of EGFR and the downstream signaling including ERK and Akt activation (24). In addition, experimental results have revealed that NMII interacts with several kinds of receptors including CXCR4 (chemokine receptor), N-methyl-D-aspartate (NMDA) receptor, DDR1 (a collagen receptor), and the inositol 1,4,5, triphosphate (P3) receptor (46-49). Therefore, interaction between CD36 and NMII and the role of NMII in CD36-mediated oxLDL uptake and downstream signaling are intriguing topics to study in the future. Roles of MHC-IIA phosphorylation in macrophages should be explored due to the potential involvement in various oxLDL-derived functions, however, transfection of primary macrophages with mutant MHC-IIA and suppression of endogenous MHC-IIA expression appear to be technically challenging at this time.

We recently reported that oxLDL via CD36 inhibits macrophage migration by modulating cytoskeletal function (17, 18), and suggested that oxLDL facilitates macrophage spreading and actin polymerization but inhibits migration via dysregulated activation of focal adhesion kinase (FAK) (17). In addition, oxLDL via CD36 induces loss of macrophage cell polarity by inhibiting the activity of NMIIA (18). OxLDL-induced inhibition of macrophage migration may explain the mechanism of macrophage trapping in atherosclerotic inflammation. Defining the mechanism of macrophage trapping and promoting macrophage egress may be a new therapeutic strategy for the treatment of atherosclerosis. Therefore, we suggest an additional mechanism for macrophage trapping involving MHC-IIA phosphorylation, another way for oxLDL to induce disassembly of myosin filaments.

Our current study demonstrates oxLDL-induced MHC-IIA phosphorylation in murine and human macrophages, suggesting a new function for the cytoskeletal modulating effect of oxLDL and providing new insights into the pathology of atherosclerosis.

MATERIALS AND METHODS

Reagents

LDL was prepared from human plasma by density gradient ultracentrifugation (50). Oxidatively modified LDL (Cu2+ oxLDL) was generated by dialysis of LDL with 5 μM CuSO4 in PBS for 6 hours at 37℃. Oxidation was terminated by dialysis against PBS containing 100 μM EDTA. Bisindolylmaleimide-I (GF 109203X) was purchased from Sigma. Antibodies for MHC-IIA, PKC-βII and β-actin were purchased from SantaCruz Biotechnology. Antibodies for the detection of phospho-MHC-IIA (Ser1917, in human, Ser1916) and MHC-IIA (Ser1948) were provided by Dr. Thomas Egelhoff at the Lerner Research Institute, Cleveland Clinic.

Cell culture

We collected peritoneal macrophages from C57BL/6 mice and Cd36 null mice provided by Dr. Roy L. Silverstein at the Medical College of Wisconsin, by peritoneal lavage 4 days after intraperitoneal injection of 4% thioglycolate (1 ml). Cells were cultured in RPMI containing 10% fetal bovine serum. The human monocyte cell line THP-1 was obtained from ATCC and were treated with phorbol 12-myristate 13-acetate (PMA) (200 nM) for 1 day and then cultured in RPMI containing fetal bovine serum (10%) for 5 days for macrophage differentiation.

Western blot analysis and immunoprecipitation

Mouse peritoneal macrophages incubated with 50 μg/ml oxLDL for the indicated times were lysed in a buffer containing 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP-40, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, and 1 mM sodium orthovanadate. Clarified lysates were separated by SDS-PAGE and transferred to PVDF membranes (Millipore). Membranes were probed with antibodies against p-MHC-IIA (Ser1917). After chemiluminescence detection, membranes were stripped with 0.2 M sodium hydroxide and reprobed with antibodies against β-actin for normalization. Band intensities were quantified by ImageJ (http://rsbweb.nih.gov/ij/) and Gel-Pro Analyzer (MediaCybernetics). For immunoprecipitation, macrophages incubated with or without oxLDL (50 μg/ml) were lysed using the buffer described above and the cell lysate containing 500 μg of protein was incubated with 3 μg of anti-MHC-IIA antibody immobilized on agarose beads overnight at 4℃. Beads were extensively washed and boiled in SDS-PAGE loading buffer and the bound material was analyzed by immunoblot using an antibody for PKC-βII. The affinity between MHC-IIA and PKC-βII was calculated by the band intensity of PKC-βII divided by the band intensity of MHC-IIA. All Western blot analyses were repeated 3 times.

Statistics

We performed ANOVA followed by Bonferroni’s multiple comparison test. A P value less than 0.05 was considered significant. Data are expressed as the mean ± SD. Analyses were formed using GraphPad Prism Software.