Phosphatidylinositol 3-kinase is involved in the induction of macrophage growth by oxidized low density lipoprotein.

Early atherosclerotic lesions are characterized by the presence of cholesterol-rich, macrophage-derived foam cells. It has recently been shown that macrophage proliferation occurs during the development of early lesions and that oxidized low density lipoprotein (LDL) stimulates macrophage growth. Possible mechanisms for this induction of macrophage growth include potentiation of mitogenic signal transduction by a component of oxidized LDL following internalization and degradation, interaction with integral plasma membrane proteins coupled to signaling pathways, or direct or indirect activation of growth factor receptors on the cell surface (e.g. GM-CSF receptor) through an autocrine/paracrine mechanism. The present study was undertaken to characterize some of the early intracellular signaling events by which oxidized LDL mediates macrophage cell growth. Extensively oxidized LDL increased protein-tyrosine phosphorylation and caused a 2-fold increase in phosphatidylinositol (PI) 3-kinase activity in phorbol ester-pretreated THP-1 cells (a human monocyte-like cell line). Similar concentrations of native LDL had no effect. Oxidized LDL also stimulated growth of resident mouse peritoneal macrophages, and this effect was reduced by 40-50% in cells treated with PI 3-kinase inhibitors (100 nM wortmannin or 20 microM LY294002). These results suggest that PI 3-kinase mediates part of the mitogenic effect of oxidized LDL, but parallel pathways involving other receptors and signal transduction pathways are likely also involved.

Several lines of evidence have implicated oxidized low density lipoprotein (LDL) 1 in the pathogenesis of atherosclerosis, including the demonstration that oxidatively modified LDL exists in the arterial intima in vivo (1,2) and that antioxidant drugs can retard atherogenesis in some animal models (3)(4)(5)(6)(7). Oxidized LDL has been shown to have many potentially atherogenic actions in vitro (8), but its role in foam cell formation is of particular relevance to early stages of atherogenesis (9 -12).
Foam cells are lipid-laden macrophages and are derived from blood-borne monocytes that have been recruited to sites of predilection of atherosclerosis by overexpression of endothelial adhesion molecules and by local release of chemotactic factors (13,14). Both of these effects have been associated with oxidized LDL (15)(16)(17)(18)(19)(20)(21).
An additional mechanism that would increase the number of macrophages in the arterial intima at sites of lesion formation is cell proliferation. Immunocytochemical studies of human atherosclerotic lesions have shown that macrophages are the predominant cell type expressing proliferating cell nuclear antigen in lesions, even in lesions containing cells derived mainly from smooth muscle cells (22,23). Oxidized LDL has recently been shown to be mitogenic for mouse peritoneal macrophages as well as human monocyte-derived macrophages (24,25). In these studies, the mitogenic effect of oxidized-LDL was attributed to lysophosphatidylcholine and was dependent on scavenger receptor-mediated uptake of the oxidized LDL particles (26). Macrophage growth was inhibited with anti-GM-CSF antibody, suggesting that oxidized LDL may induce secretion of GM-CSF, leading to autocrine or paracrine growth stimulation (24). Oxidized LDL is also mitogenic for bovine vascular smooth muscle cells (27). In these cells, the mitogenic effect was blocked with platelet-activating factor receptor antagonists, suggesting that oxidized LDL may also stimulate growth directly through platelet-activating factor receptor activation. Oxidized LDL has been found to increase phosphoinositide turnover in vascular smooth muscle cells (28), which would be consistent with an effect mediated by the platelet-activating factor receptor, a G-protein-linked receptor that activates phosphatidylinositol-specific phospholipase C (29).
Growth factor receptor-mediated signaling commonly involves the activation of mitogen-activated protein (MAP) kinase, protein-tyrosine kinases, and phosphatidylinositol (PI) 3-kinase (30,31). Exposure of transformed macrophage cell lines to oxidized LDL or acetyl LDL has been reported to lead to the activation of MAP kinase (32,33), protein kinase C (PKC) (30), and the cytoplasmic tyrosine kinase p53/p56 Lyn (31). The objective of the present study was to determine if oxidized LDL leads to increased protein-tyrosine phosphorylation and PI 3-kinase activation in macrophages and to define the role of PI 3-kinase in the mitogenic activity of oxidized LDL.
Lipoprotein Isolation and Oxidation-LDL (d ϭ 1.019 -1.063) was isolated by sequential ultracentrifugation of EDTA-anticoagulated plasma obtained from healthy normolipidemic volunteers (34). For lipoprotein modification, the concentration of EDTA in the isolated LDL preparation was reduced prior to oxidation by dialysis against Dulbecco's phosphate-buffered saline (PBS) containing 10 M EDTA. Standard conditions for LDL oxidation were incubation of 200 g/ml LDL in PBS with 5 M CuSO 4 at 37°C for 22 h (35,36). These conditions typically resulted in a 4-fold increase in electrophoretic mobility relative to unmodified native LDL on agarose gels. Oxidized LDL was concentrated to about 750 g/ml using ultrafiltration membrane cones (Centricon CF 25, Amicon) and passed through a 0.2-m filter. Oxidized LDL was stored at 4°C and was used within 2 weeks of preparation. Lipopolysaccharide levels in oxidized LDL were less than 100 pg/mg LDL protein.
Cell Culture-The THP-1 cell line was obtained from American Type Culture Collection (Rockville, MD). Cells from this human monocytic leukemia-derived line can be induced to differentiate to a macrophagelike phenotype upon exposure to phorbol ester, including the expression of Fc, C3b, and scavenger receptors (37,38). For analysis of tyrosine phosphorylation and PI 3-kinase activity, THP-1 cells were grown in RPMI 1640 medium supplemented with 10% fetal calf serum and 20 mM HEPES. Five million cells were dispensed into each culture dish, and treated with 10 ng/ml PMA for 24 h at 37°C. Cells were then washed three times with serum-free medium and preincubated for 3 h in serumfree medium. Different concentrations of oxidized or native LDL or an equivalent volume of PBS were then added. In each experiment, 1 g/ml LPS was used as a positive control for PI 3-kinase activation. In experiments to test the effect of oxidized LDL on macrophage growth, resident macrophages were collected from male CD-1 mice (25-30 g) by peritoneal lavage with ice-cold Ca 2ϩ -free Dulbecco's PBS. Cells were resuspended in RPMI 1640 supplemented with 10% FBS, and 0.5 g/ml gentamicin. Ten thousand macrophages were added to each well of 96-well tissue culture plates (Falcon, Lincoln Park, NJ). After incubation for 6 h at 37°C, nonadherent cells were removed by gentle washing with medium. Adherent macrophages were then cultured in 0.1 ml of RPMI medium containing 5% FBS and varying concentrations of LDL and PI 3-kinase inhibitors for 4 days without a medium change.
Anti-phosphotyrosine Immunoblotting-Cells were rinsed three times with Hanks' buffered saline solution and incubated for 15 min on ice with lysis buffer (20 mM Tris-HCl buffer, pH 8.0, containing 137 mM NaCl, 10% (v/v) glycerol, 2 mM EDTA, 1% (v/v) Triton X-100, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 1 mM molybdate, 10 g/ml aprotinin, and 10 g/ml leupeptin). Cell lysates were then spun in a microfuge at 14,000 rpm for 20 min at 4°C, and the supernatants containing soluble proteins were collected. For analysis of proteintyrosine phosphorylation in whole cell lysates, 50 g of protein from each sample was loaded and then separated by SDS-polyacrylamide gel electrophoresis (7.5%). Gels were calibrated using prestained high range molecular weight markers (Amersham). Proteins were then transferred to nitrocellulose paper, incubated overnight with 2% bovine serum albumin, 0.01% NaN 3 , and then washed three times with Tween buffer (20 mM Tris HCl, pH 7.6, 137 mM NaCl, and 0.1% Tween 20). Blots were probed with 4G10 murine anti-phosphotyrosine antibody for 2 h and washed three times with Tween buffer. Bound primary antibody was then visualized with horseradish peroxidase-conjugated goat anti-mouse IgG for 1 h at 20°C and imaged on Kodak XR film using an enhanced chemiluminescence system (Amersham) with 5 min of exposure time.
Immunoprecipitation and in Vitro PI 3-Kinase Assay-Aliquots of cell lysates each containing 350 g of protein were incubated overnight at 4°C with monoclonal antibody to PI 3-kinase. The immune complexes were then collected on protein A-agarose beads, washed twice with lysis buffer containing 50 M vanadate, and three times with 10 mM Tris, pH 7.4. Immunoprecipitates were then resuspended in 100 l of 20 mM HEPES, 1 mM EDTA, pH 7.4, and placed on ice. To assay PI 3-kinase activity, 20 g of phosphatidylinositol was dried under N 2 , resuspended in 20 l of 30 mM HEPES, pH 7.4, and dispersed by sonication. 20 l of sonicated lipid was then added to the immunoprecipitate, followed by 30 l of buffer containing 10 Ci of [␥-32 P]ATP, 200 M adenosine, and 50 M ATP. Reactions were stopped by the addition of 0.1 ml of 1 N HCl, and 0.2 ml of chloroform:methanol (1:1, v/v). Lipids were separated on oxalate-treated silica thin-layer chromatography plates using a solvent system of chloroform:methanol:water:28% ammonia (45:35:7.5:2.5, v/v/v/v). Plates were then exposed to x-ray film for 18 h at Ϫ70°C, and incorporation of radioactivity into the lipids was quantified by excising the corresponding portions of the plate followed by liquid scintillation counting.
XTT Growth Assay-Mouse peritoneal macrophage growth was determined by the XTT formazan method. This assay is based on the cellular reduction of XTT by mitochondrial dehydrogenase to an orange formazan product that can be measured spectrophotometrically and correlates well with cell number under these conditions (39). Briefly, 50 l of XTT solution (1 mg/ml XTT, 25 M N-methyldibenzopyrazine methyl sulfate salt) was added to each well and incubated for 4.5 h at 37°C. Absorbance at 450 nm was then measured with a multiwell spectrophotometer.

Protein-tyrosine Phosphorylation in Macrophages Treated
with Oxidized LDL-Differentiated THP-1 cells were exposed to 40 g/ml oxidized LDL, and cell lysates were assayed for total tyrosine-directed protein phosphorylation by immunoblotting with 4G10 anti-phosphotyrosine antibody. As shown in Fig. 1A, phosphorylation of cellular proteins was increased with exposure of cells to oxidized LDL, becoming maximal after 10 min, and returning to basal levels by 15 min. Stimulation of cells with LPS is shown as a positive control. In response to oxidized LDL, increased phosphorylation was evident in several bands with molecular masses of 85, 100, and 110 kDa. Densitometric quantification of the bands (Fig. 1B) indicates similar patterns of phosphorylation with oxidized LDL and LPS.
Phosphatidylinositol 3-Kinase Activity in Macrophages Treated with Oxidized LDL-Differentiated THP-1 cells were incubated with 40 g/ml oxidized LDL, and whole cell lysates were examined for PI 3-kinase activity. The time course of activation of PI 3-kinase by oxidized LDL is shown in Fig. 2. After 15 min of incubation with oxidized LDL, PI 3-kinase activity increased to a maximum of approximately 2-fold that of untreated controls and returned to basal levels by 30 min. Incubation of cells with native LDL for 15 min had no effect. Fig. 3 shows the effects of different concentrations of oxidized LDL on activation of PI 3-kinase. Increased PI 3-kinase activity was detectable at oxidized LDL concentrations as low as 10 g/ml and was maximal with 30 -40 g/ml oxidized LDL. To determine if the increased activity of PI 3-kinase might be accompanied by phosphorylation of the p85 subunit, cell lysates were immunoprecipitated with antibody to PI 3-kinase, and then parallel immunoblots were performed with anti-phosphotyrosine and anti-PI 3-kinase antibodies. The intensity of the p85 band on anti-PI 3-kinase immunostaining was constant, but no detectable phosphorylation of the p85 subunit in response to oxidized LDL could be demonstrated (data not shown). To exclude the possibility that the effect of oxidized LDL might simply be an artifact of LPS contamination of oxidized LDL preparations, endotoxin levels in oxidized LDL were determined with a sensitive chromogenic limulus lysate assay. Endotoxin levels were consistently less than 100 pg/ml. Control experiments indicated that 10-fold higher concentrations of LPS had no effect on PI 3-kinase activity either alone or in the presence of oxidized LDL (data not shown).
Oxidized LDL Promotes Macrophage Cell Growth-The effects of native or oxidized LDL on growth of mouse resident peritoneal macrophages were examined using the XTT reduction assay. This assay is sensitive enough to be applied to cells grown in 96-well plates and was found to correlate well with cell number (Fig. 4). Stimulation of macrophage growth was observed with concentrations of oxidized LDL as low as 5 g/ml and was maximal at 40 g/ml (Fig. 5). Incubation of cells with the same concentrations of native LDL or acetyl LDL had no effect. Cells incubated with oxidized LDL showed a substantial increase in size and change in morphology, leading to the formation of elongated, spindle-shaped cells. Concentrations of oxidized LDL greater than 60 g/ml were associated with blebbing and cell detachment and a decrease in the proportion of cells that excluded trypan blue. To determine if the increased mitochondrial reduction of XTT was accompanied by DNA replication and cell division, we performed cell counts and measured thymidine incorporation into DNA. Oxidized LDL caused a 4 -5-fold increase in thymidine incorporation into macrophages following 6 days of incubation and a 2-3-fold increase in cell number (data not shown).
Effect of PI 3-Kinase Inhibitors on Oxidized LDL-induced Macrophage Growth-To determine if PI 3-kinase activation was required for the induction of macrophage cell growth by oxidized LDL, mouse peritoneal macrophages were pretreated with the PI 3-kinase inhibitors wortmannin and LY294002 and then incubated with oxidized LDL. As shown in Fig. 6, pretreatment of cells with 100 nM wortmannin or 20 M LY294002 inhibited oxidized LDL-induced macrophage growth by 40 -50%. There was no further inhibition at 2-fold higher concentrations of inhibitors. Although we did not measure PI 3-kinase activity throughout the 4-day incubation, these concentrations of inhibitors completely block PI 3-kinase activity in human FIG. 1. Effects of oxidized-LDL on protein-tyrosine phosphorylation in THP-1 cells. THP-1 cells (5 ϫ 10 6 cells/dish) were induced to differentiate by incubation for 24 h with 10 nM PMA. Cells were incubated in medium alone (control), with 1 g/ml LPS for 15 min, or with 40 g/ml oxidized LDL for the indicated times. Cell lysates (50 g of protein) were then analyzed by immunoblot using phosphotyrosinespecific monoclonal antibody. The results shown are from one of three experiments that yielded similar results. A, increased tyrosine-directed phosphorylation is evident in proteins with corresponding molecular masses of approximately 55-65, 85, 100, and 110 kDa when compared with lysates from untreated controls. Native LDL (40 g/ml) had no effect on protein tyrosine phosphorylation (not shown). B, densitometric analysis of bands in the 55-65-, 85-, 100-, and 110-kDa regions showing similar patterns of tyrosine phosphorylation by LPS and oxidized LDL.

FIG. 2. Time course of increased PI 3-kinase activation in
THP-1 macrophages treated with oxidized LDL. THP-1 cells (5 ϫ 10 6 cells/dish) were induced to differentiate by incubation for 24 h with 10 nM PMA and were then treated with PBS alone, with 1 g/ml LPS for 15 min, or with 40 g/ml oxidized LDL or native LDL (N-LDL) for the indicated times. PI 3-kinase was immunoprecipitated from equivalent amounts of cell lysates and incubated with [␥-32 P]ATP and phosphatidylinositol, and labeled PI 3-phosphate (PI3-P) was detected by thinlayer chromatography and autoradiography (top panel). To quantify PI 3-kinase activity, spots corresponding to PI 3-phosphate were scraped and counted in a liquid scintillation spectrometer (lower panel). Results are expressed as a percentage of activity in untreated cells. Each value in the graph represents the mean Ϯ S.E. of three independent experiments, each using fresh LDL samples from different donors. Compared with control (Con), p Յ 0.05 for oxidized LDL (Ox-LDL) incubated for 10 or 15 min.

FIG. 3. Effect of oxidized LDL concentration on PI 3-kinase activation in THP-1 macrophages.
PMA-differentiated THP-1 cells were treated with PBS alone (Con), with 1 g/ml LPS for 15 min, or with the indicated concentrations of oxidized LDL for 15 min. Equivalent amounts of cell lysates were then assayed for PI 3-kinase activity, as described in the legend to Fig. 2. Each value represents the mean Ϯ S.E. of three independent experiments, each using fresh LDL samples from different donors. Compared with control, p Յ 0.05 for all concentrations of oxidized LDL (Ox-LDL, Student's unpaired t test). PI3-P, PI 3-phosphate. monocyte-derived macrophages (40) in human and murine B cells (41) and in several other other cell types (42)(43)(44). Therefore, it seems unlikely that incomplete inhibition of growth is due to incomplete inhibition of PI 3-kinase. Wortmannin was without toxic effects to cells at concentrations up to 1 M, as judged by trypan blue dye exclusion assay. DISCUSSION Oxidized LDL has recently been shown to be mitogenic toward several types of cells including murine peritoneal macrophages, human monocyte-derived macrophages, and vascular smooth muscle cells (24,25,27). In macrophages, the mitogenic effect has been attributed to lysophosphatidylcholine and is reported to require scavenger receptor-mediated internalization of oxidized LDL (26). In smooth muscle cells, the mitogenic effect was found to be mediated by oxidized phospholipids that interacted with the platelet-activating factor receptor (27). It is not clear whether the divergent conclusions from these studies can be explained by oxidized LDL acting through different pathways in smooth muscle cells and macrophages, or if both lysophosphatidylcholine and oxidized phospholipids can activate similar signaling pathways leading to mitogenesis in these cells.
The downstream signaling pathways involved in the mitogenic effect of oxidized LDL have not yet been defined. The objective of the present study was to examine the intracellular signaling pathways that may be involved in the induction of macrophage growth by oxidized LDL. The results show that oxidized LDL induces tyrosine phosphorylation of several different proteins in THP-1 macrophages. The main substrates detected have apparent molecular masses of 55-65, 85, 100, and 110 kDa. Maximal phosphorylation of the 55-65-kDa proteins was noted 5 min after addition of oxidized LDL, whereas the higher molecular mass proteins showed maximal phosphorylation after 10 min. The activation of PI 3-kinase by oxidized LDL was concentration-and time-dependent. Moreover, there was a 40 -50% decrease in oxidized LDL-induced growth in macrophages pretreated with PI 3-kinase inhibitors, indicating that pathways dependent on PI 3-kinase account for at least 40% of this effect.
The findings of this study are consistent with previous evidence supporting a role for PI 3-kinase in mitogenic signaling. For example, Iwama et al. have shown that association of PI 3-kinase with the platelet-derived growth factor receptor of vascular smooth muscle cells appears to be necessary for platelet-derived growth factor-induced cellular mitogenesis (45). Yusoff et al. concluded that PI 3-kinase activation may be involved in growth stimulation of bone marrow-derived macrophage by hematopoietic growth factors CSF-1 and GM-CSF (46). In these studies, CSF-1 and GM-CSF stimulated a dosedependent activation of PI 3-kinase, whereas concanavalin A, PMA, and formyl-methionyl-leucyl-phenylalanine had no mitogenic activity and did not significantly increase PI 3-kinase activity.
Several other studies have also shown that incubation of macrophages with modified LDL leads to rapid tyrosine phosphorylation of several intracellular proteins, including a member of the Src tyrosine kinase family, p53/p56 Lyn (31). Activated tyrosine kinases such as p53/p56 Lyn have been shown to physically interact with and activate PI 3-kinase in human monocytes and B-lymphocytes (40,47). Association of intracellular PI 3-kinase with activated tyrosine kinases is thought to be mediated by either SH3 domains of tyrosine kinases (48) or via SH2 domains of the p85 regulatory subunit of PI 3-kinase (49), leading to increased PI 3-kinase activity. It has been suggested that association with tyrosine kinases may account for PI 3-kinase activation (40) ation of the p85 subunit of PI 3-kinase is required for its activation. Several potential phosphorylation sites are present in the p85 subunit (50), but PI 3-kinase activation can occur without SH2 domain phosphorylation of this subunit (49). The results in Fig. 1 indicate an apparent increase in proteintyrosine phosphorylation in the 85-kDa region of whole cell lysates after exposure of cells to oxidized LDL, but there was no evidence of phosphorylation of the p85 band in immunoprecipitated PI 3-kinase (data not shown).
Although the findings of this study indicate that the growthpromoting effects of oxidized LDL are PI 3-kinase-dependent, it is likely that additional pathways independent of PI 3-kinase are also involved. One potential target for oxidized LDL-induced mitogenic signaling is PKC. Sakai and colleagues have suggested a role for PKC in oxidized LDL-induced mitogenesis because they found that exposure of murine resident peritoneal macrophages to oxidized LDL results in rapid calcium influx and a sustained increase in intracellular calcium concentrations (24). Exposure of vascular smooth muscle cells to oxidized LDL has also been shown to increase PKC activity as well as to enhance platelet-derived growth factor-AA production, platelet-derived growth factor receptor mRNA expression, and DNA synthesis (51). In contrast, inhibition of PKC with staurosporine decreased oxidized LDL-induced DNA synthesis (51). These findings suggest that PKC activation may contribute to the growth stimulation of both murine macrophages and vascular smooth muscle cells by oxidized LDL.
The effects of oxidized LDL on other enzymes known to be involved in mitogenesis has also been investigated. Recently, Deigner et al. demonstrated MAP kinase activation in U937 cells stimulated with oxidized LDL, independent of PKC activation (32). However, incubation of cells with native LDL induced an even greater increase in MAP kinase activity. Kusuhara and colleagues examined the effects of native or oxidized LDL on MAP kinase activity in smooth muscle cells, endothelial cells, and macrophages and found that both oxidized LDL and native LDL stimulated MAP kinase in a PKCdependent manner (33). However, in human monocyte-derived macrophages and in rat vascular smooth muscle cells the effect of oxidized LDL was greater than that of native LDL. Because native LDL does not induce growth of nontransformed macrophages or vascular smooth muscle cells (24,25), it does not seem likely that MAP kinases are directly involved in the mitogenic effect of oxidized LDL. Phospholipase D activation has also been observed in vascular smooth muscle cells stimulated with oxidized LDL (52). Phospholipase D activation results in the generation of phosphatidic acid and lysophosphatidic acid, both of which are known to be mitogenic (53,54). However, at present there is no direct evidence linking phospholipase D activation to the mitogenic effect of oxidized LDL.
Previous studies on growth stimulation by oxidized LDL suggested that the growth stimulation was mediated by a phospholipid component of oxidized LDL (24,27). However, the possibility that the modified apoB of oxidized LDL may interact with membrane proteins and stimulate growth by a process analogous to integrin-mediated signaling (55) has not been excluded. Oxidized LDL binds to the scavenger receptor class A, type I/II, and it has been suggested that this receptor leads to tyrosine phosphorylation (31). However, acetyl LDL, which is an excellent ligand for the scavenger receptor class A, type I/II, does not stimulate macrophage growth, indicating that mere ligation of this receptor is not sufficient to induce growth. Oxidized LDL also binds to membrane proteins that do not interact well with acetyl LDL, including CD36, Fc␥RII, and macrosialin/CD68 (56). The possibility that selective binding of oxidized LDL to these or other plasma membrane proteins may initiate the activation of mitogenic signal transduction cascades warrants further consideration.
In conclusion, the data reported in this study provide evidence for a direct link between oxidized LDL-induced PI 3-kinase activation and macrophage growth. These findings not only increase our understanding of how oxidized LDL transmits its proliferative signal in macrophages but also provide insight into the signaling pathways that may be involved in some of the other biological functions of oxidized LDL. Further studies are required to identify the components of oxidized LDL that are responsible for growth stimulation and to define interactions with or effects mediated by other signaling pathways.