Oxidized LDL attenuates apoptosis in monocytic cells by activating ERK signaling.

Low concentrations of oxidized low density lipoprotein (OxLDL) are cytoprotective for phagocytes, although the underlying mechanisms remain unclear. We investigated signaling pathways used by OxLDL to attenuate apoptosis in monocytic cells. OxLDL at 25–50 μg/ml inhibited staurosporine-induced apoptosis in THP-1 cells and mouse peritoneal macrophages, and it was cytoprotective in human primary monocytes upon serum withdrawal. Attenuated cell demise was reversed by blocking extracellular signal-regulated kinase (ERK) signaling. Translocation of cytochrome c to the cytosol was attenuated by OxLDL, which again demanded ERK signaling. Analysis of Bcl-2 family proteins revealed phosphorylation of Bad at serine 112 as well as ERK-dependent inhibition of Mcl-1 degradation. Although the formation of reactive oxygen species (ROS) is an established signal generated by OxLDL, ROS scavengers did not interfere with cell protection by OxLDL. Thus, activation of the ERK signaling pathway by OxLDL is important to protect phagocytes from apoptosis.

Abstract Low concentrations of oxidized low density lipoprotein (OxLDL) are cytoprotective for phagocytes, although the underlying mechanisms remain unclear. We investigated signaling pathways used by OxLDL to attenuate apoptosis in monocytic cells. OxLDL at 25-50 mg/ml inhibited staurosporine-induced apoptosis in THP-1 cells and mouse peritoneal macrophages, and it was cytoprotective in human primary monocytes upon serum withdrawal. Attenuated cell demise was reversed by blocking extracellular signal-regulated kinase (ERK) signaling. Translocation of cytochrome c to the cytosol was attenuated by OxLDL, which again demanded ERK signaling. Analysis of Bcl-2 family proteins revealed phosphorylation of Bad at serine 112 as well as ERK-dependent inhibition of Mcl-1 degradation. Although the formation of reactive oxygen species (ROS) is an established signal generated by OxLDL, ROS scavengers did not interfere with cell protection by OxLDL. Thus, activation of the ERK signaling pathway by OxLDL is important to protect phagocytes from apoptosis.-Namgaladze, D., A. Kollas, and B. Brü ne. Oxidized LDL attenuates apoptosis in monocytic cells by activating ERK signaling. J. Lipid Res. 2008. 49: 58-65.
Supplementary key words atherosclerosis & oxidized low density lipoprotein & reactive oxygen species & extracellular signal-regulated kinase Atherosclerotic plaque progression involves dynamic changes in cellular composition. These changes comprise the migration of cells from the bloodstream into the intima, their egress into the lymph system, and the proliferation and migration of smooth muscle cells from the media. Moreover, cell death affects cell numbers in atherosclerotic plaques. Several studies have established that phagocyte cell death diminished lesion progression during early stages of atherosclerosis (1,2), whereas an increased death of phagocytes and smooth muscle may contribute to plaque destabilization in advanced atherosclerosis (3,4).
Modified lipoproteins are important factors in lesion development, and the oxidation of lipoproteins is a characteristic feature of the plaque (5). Oxidized low density lipoproteins (OxLDLs) affect numerous phagocyte functions, including cell survival (6). Although high concentrations of OxLDL are cytotoxic for cells of the vascular wall, cytoprotective actions of OxLDL have been reported at lower doses in phagocytes (7,8). Recent studies suggest that cytoprotection by OxLDL depends on the degree of its oxidative modification as well as OxLDL inclusion in immune complexes (9,10).
Mechanistically, OxLDL-evoked cytoprotection was linked to activation of the phosphatidylinositol 3-kinase (PI3K)-Akt survival pathway (7,(9)(10)(11). However, little is known regarding other potential antiapoptotic mechanisms conveyed by OxLDL. Activation of extracellular signal-regulated kinase (ERK) constitutes a major mitogenic and antiapoptotic pathway activated by a variety of extracellular agonists, such as growth factors or hormones (12). ERK targets a number of apoptotic regulators, including proteins of the Bcl-2 family (13-15), caspases (16), and inhibitor of apoptosis proteins (17). Although some data suggest that ERK inhibition reduces the viability of OxLDL-treated phagocytes (18), other studies reported ERK not to be involved (7). Therefore, the role of ERK signaling in provoking antiapoptotic actions of OxLDL remains unclear.
In this study, we explored signaling mechanisms evoked by OxLDL in protecting THP-1 monocytic cells from staurosporine-induced cell death. The antiapoptotic effect of OxLDL involved the activation of ERK, whereas PI3K/Akt activation was not involved. Inhibition of apoptosis occurred upstream of cytochrome c trans-location, involved Bad phosphorylation, and impaired Mcl-1 degradation.

Cell culture
THP-1 monocytic cells were maintained in RPMI 1640 medium containing 100 U/ml penicillin, 100 mg/ml streptomycin, and 10% heat-inactivated fetal calf serum. Before the experiments, the medium was changed to fetal calf serum-free medium. Human monocytes were isolated from buffy coats (DRK-Blutspendedienst Baden-Wü rttemberg-Hessen, Institut fü r Transfusionsmedizin und Immunhämatologie Frankfurt am Main) using Ficoll density centrifugation (LSM 1077; PAA Laboratories) followed by magnetic separation with positive selection (CD14 MicroBeads; Miltenyi Biotec, Bergisch Gladbach, Germany). The purity of monocytes was assessed by flow cytometry using CD14 staining. Resident mouse peritoneal macrophages were obtained from C3H/HeJ mice and cultured in RPMI 1640 medium containing 100 U/ml penicillin, 100 mg/ml streptomycin, and 10% heat-inactivated fetal calf serum. Cells were incubated overnight in the same medium without serum before treatments. This study conformed with the principles outlined in the Declaration of Helsinki.

Western blot analysis
Cells were incubated overnight in serum-free medium, treated, collected, lysed in 200 ml of buffer A (50 mM Tris, 150 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, 10 mM NaF, 1 mM Na 3 VO 4 , 1 mM PMSF, and protease inhibitor mixture, pH 7.5), and sonified, followed by centrifugation (15,000 g, 10 min). A total of 30-60 mg of protein was loaded onto and resolved on SDS-polyacrylamide gels. Proteins were blotted onto nitrocellulose. Membranes were blocked and incubated with primary antibodies according to the manufacturer's instructions. Membranes were washed three times for 5 min each with TTBS (10 mM Tris, 150 mM NaCl, and 0.05% Tween 20, pH 7.5). For protein detection, blots were incubated with a horseradish peroxidase-labeled secondary antibody for 1 h and washed three times for 5 min each with TTBS, followed by ECL detection.

Cell death detection
Cells were incubated overnight in serum-free medium, treated, and processed for cell death detection by flow cytometry (FACS Canto; BD Biosciences, Heidelberg, Germany) using the Annexin V-FITC apoptosis detection kit (Beckman Coulter, Krefeld, Germany) according to the manufacturer's instructions. Apoptosis of mouse peritoneal macrophages was assessed by detecting cytoplasmic histone-associated DNA fragments using the cell death ELISA kit from Roche Diagnostics, according to the manufacturer's instructions.

Cytochrome c release assay
Cytochrome c translocation to the cytosol was monitored by fractionation of cell lysates according to Leist et al. (20). Cell pellets were resuspended in 100 ml of PBS. A total of 100 ml of digitonin/sucrose solution (0.16 mg/ml digitonin and 500 mM sucrose) was added for 30 s, followed by centrifugation at 10,000 g for 1 min. A total of 50 mg of cytosolic and 25 mg of mitochondrial fractions were analyzed by Western blotting for cytochrome c as well as ATP synthase versus ERK as controls for mitochondrial and cytosolic fractions, respectively.

Reactive oxygen species determination
Reactive oxygen species (ROS) generation in THP-1 cells was analyzed by flow cytometry monitoring the oxidation of 2 ¶,7 ¶dichlorodihydrofluorescein (DCFH). Cells were loaded with 5 mM DCFH-diacetate for 30 min, pelleted, resuspended in PBS, and treated for 1 h with OxLDL and/or antioxidants followed by flow cytometry.

Statistical analysis
Data are expressed as means 6 SEM. Two treatment groups were compared by independent Student's t-tests. Results were considered statistically significant at P , 0.05.

RESULTS
To investigate the influence of OxLDL on apoptosis, monocytic THP-1 cells were treated with 0.2 mg/ml staurosporine for 4 h to induce the mitochondrial apoptotic signaling pathway. Cell death was followed by Annexin V/ propidium iodide staining and flow cytometry. As shown in Fig. 1A, staurosporine provoked massive apoptosis and thus cells became Annexin V-positive but basically remained propidium iodide-negative, indicating that only a small fraction of cells was necrotic. Exposing cells to staurosporine in the presence of 50 mg/ml OxLDL drastically reduced the number of apoptotic cells with minor signs of necrosis. Quantification of Annexin V staining and thus apoptosis is shown in Fig. 1B. Inhibition of staurosporine-induced apoptosis was prominent already at 12.5 mg/ml OxLDL and reached maximal protection with 50 mg/ml. We noticed that at concentrations of 100 mg/ml and greater, OxLDL induced necrotic cell death (19.2 6 2.7% Annexin V/propidium iodide double-positive cells after 4 h with 150 mg/ml OxLDL), confirming previous data on OxLDL cytotoxicity (21). Native LDL was protective as well, but ?5-fold higher LDL concentrations were needed to achieve an antiapoptotic effect similar to that of OxLDL.
Staurosporine-induced apoptosis involves the activation of caspases, provoking the cleavage and activation of executioner caspases-3 and -7. To evaluate the impact of OxLDL on caspase-3/7, we measured the cleavage of Ac-DEVD-AMC in lysates of THP-1 cells. As shown in  Fig. 2A, activation of caspase-3/7 by staurosporine was dose-dependently antagonized by OxLDL. Western blot analysis of procaspase-3 and its processed fragments is presented in Fig. 2B. Staurosporine induced a timedependent decrease in the amount of procaspase-3 with a corresponding increase of its cleaved (i.e., 21 and/or 17 kDa) fragments. At 4 h of staurosporine treatment, caspase-3 was cleaved to the 21 kDa fragment, whereas at 8 h, the 17 kDa fragment predominated. Twenty-five as well as 50 mg/ml OxLDL, added together with staurosporine, blocked the processing of procaspase-3 completely, which corroborates caspase activity data.
OxLDL-induced survival of mouse bone marrow-derived macrophages was associated with activation of the PI3K-Akt pathway (7). In addition, activation of mitogen-activated protein kinases, particularly ERK, was linked to growthpromoting and prosurvival actions of OxLDL (18). Therefore, we set out to study these potential prosurvival pathways during protection from apoptosis in THP-1 cells by OxLDL.
In a first set of experiments shown in Fig. 3A, we explored the activation of ERK and Akt by OxLDL. OxLDL elicited a transient phosphorylation of both Akt and ERK. Maximal effects were seen at 2-10 min, with no changes in the expression of total Akt. Bad, a proapoptotic Bcl-2 family member, is known to be a target of phosphorylation at serine 112 or serine 136 after activation of ERK or Akt, respectively. OxLDL induced Bad phosphorylation at serine 112 within a time period overlapping with ERK activation (Fig. 3A). However, we could not detect phosphorylation of Bad at serine 136, suggesting that ERK activation dominates for Bad phosphorylation. To further elucidate the role of PI3K-Akt versus ERK signaling in protecting THP-1 cells from apoptosis, we pretreated cells with the PI3K inhibitor wortmannin or blocked mitogen-activated protein kinase kinase with PD98059 before cell stimulation with staurosporine in the presence/absence of OxLDL. As shown for Annexin V staining (Fig. 3B), the protection evoked by OxLDL against staurosporineinduced cell death was reversed by PD98059 but not by wortmannin. Although another PI3K inhibitor, LY294002, reversed cytoprotection by OxLDL, it also reduced OxLDL-induced ERK phosphorylation and thus appeared nonspecific for PI3K/Akt inhibition in our system (data not shown). We also tested inhibitors of two other mitogen-activated protein kinases: SP600125 to block JNK and SB203580 to antagonize p38. Neither of these inhibitors interfered with OxLDL-evoked protection (data not shown).
To explore whether the antiapoptotic effect of OxLDL is not limited only to cell lines, we examined its effect on apoptosis in resident mouse peritoneal macrophages. Apoptosis was assessed by measuring DNA fragmentation using the cell death ELISA kit (Roche Diagnostics). As shown in Fig. 3D, staurosporine treatment of macrophages induced DNA fragmentation, which was reduced significantly in the presence 50 mg/ml OxLDL. Cytoprotection by OxLDL was reversed if cells were preincubated with PD98059. Thus, OxLDL can exert antiapoptotic effects in an ERKdependent manner in primary macrophages. Additionally, OxLDL increased the viability of human primary monocytes cultured in serum-free medium, a condition known to induce apoptosis in monocytes. Incubations of monocytes with 50 mg/ml OxLDL for 24 h increased cell viability, as measured by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay (absorbance at 560 nm 5 0.117 6 0.007 for control and 0.224 6 0.031 for OxLDL), and reduced caspase-3/7 activity (4,310 6 1,120 arbitrary units for control and 960 6 200 arbitrary units for OxLDL).
LDL oxidation generates a diversity of biologically active substances, including oxysterols, lysophospholipids, aldehydes, and oxidized fatty acids. Oxidized phospholipids such as OxPAPC are thought to be generated early in the process of LDL oxidation and represent major components of minimally modified LDL, recently shown to be cytoprotective in macrophages (9). We observed that OxPAPC reduced staurosporine-induced apoptosis of THP-1 cells and that the protection attributed to OxPAPC was partly reversed by PD98059 (Fig. 3C). In contrast, 4-hydroxynonenal Assuming that ERK plays a dominant role in conveying protection by OxLDL raised the question regarding its potential targets. Several ERK targets have been identified in regulating mitochondrial pathways of apoptosis, including proteins of the Bcl-2 family, caspases, and inhibitor of apoptosis proteins. Cytochrome c translocation from mitochondria to the cytosol constitutes a central role in the intrinsic cell death signaling cascade. Therefore, we explored whether OxLDL interfered with cell death signaling upstream or downstream of cytochrome c release. Cell fractionation followed by Western analysis of cytochrome c revealed that staurosporine induced cytochrome c release to the cytosol compared with controls, an effect antagonized by OxLDL (Fig. 4A). The protection afforded by OxLDL with regard to cytochrome c translocation was reversed by PD98059, again suggesting the involvement of ERK signaling. At the same time, the general caspase in-hibitor z-VAD-fmk did not impair the release of cytochrome c from mitochondria but blocked staurosporineinduced apoptosis (data not shown). Cytosolic fractions were controlled for equal protein loading by probing the membranes with ERK antibodies, whereas the lack of contamination with the mitochondrial fraction was confirmed using antibodies against ATP synthase (data not shown). Conclusively, OxLDL interfered with staurosporine-induced apoptosis upstream of cytochrome c release.
Cytochrome c release is controlled by interactions of proapoptotic versus antiapoptotic members of the Bcl-2 protein family. Several of these proteins constitute ERK targets, including Bad, Bim, or Mcl-1. Therefore, we analyzed the expression of several Bcl-2 family members after THP-1 treatments with staurosporine and/or OxLDL (Fig. 4B). Neither staurosporine nor OxLDL changed the amount of Bcl-2, Bcl-xL, or BimEL. In contrast, staurosporine induced a rapid disappearance of Mcl-1, which is known as a key antiapoptotic player in lymphoid and myeloid cells. Degradation of Mcl-1 was reduced when staurosporine was supplied in the presence of OxLDL. We went on to investigate whether the attenuation of Mcl-1 degradation by OxLDL was ERK-dependent. As shown in Fig. 4C, the expression of Mcl-1 that was preserved by OxLDL disappeared in the presence of the ERK inhibitor PD98059. Moreover, Mcl-1 degradation was not affected by caspase inhibition, verifying that regulation of Mcl-1 expression is not simply a consequence of reduced apoptosis.
The generation of ROS by OxLDL has been implicated in eliciting several of its biological actions, including cy-totoxicity (21) and nuclear factor-kB activation (22). Recently, it was suggested that ROS, particularly peroxynitrite, might be involved in OxLDL-induced ERK activation in endothelial cells (23). To determine whether ROS play a role in the inhibition of apoptosis by OxLDL, we measured DCFH oxidation to follow ROS production in OxLDL-treated THP-1 cells. Treatment of THP-1 cells with OxLDL for 1 h increased DCFH oxidation by 3-fold (Fig. 5A). DCFH oxidation was reduced significantly in the presence of the peroxyl radical scavenger N,N ¶-diphenyl-4phenylenediamine (DPPD) or the lipophylic antioxidant  2 [3]-t-butyl-4-hydroxyanisole (BHA). However, neither DPPD nor BHA affected apoptosis inhibition evoked by OxLDL (Fig. 5B), ruling out the possibility that ROS are important for blocking cell death. Interestingly, the same concentrations of DPPD and BHA suppressed toxicity achieved with high OxLDL concentrations (data not shown), thus confirming previous observations showing ROS-dependent OxLDL toxicity to macrophages (21).

DISCUSSION
Our data indicate that active ERK signaling is critical for the suppression of apoptosis by OxLDL in THP-1 cells, thus providing a mechanistic explanation for the antiapoptotic actions of oxidized lipoproteins. Considering that phagocyte apoptosis reduces cell numbers, thereby retarding lesion progression in the early phase of atherosclerosis, suppression of cell death may contribute to the proatherosclerotic actions of OxLDL.
LDL modifications are considered a hallmark during the development of atherosclerosis (5). However, native LDL can also support foam cell formation (24) and activate intracellular signaling pathways (25). We noticed that both OxLDL and native LDL attenuated apoptotic cell death, although the protection by LDL was minor compared with OxLDL. This implies that oxidation confers antiapoptotic properties to LDL. The nature of the OxLDL epitopes responsible for the antiapoptotic action as well as the receptor(s) mediating this effect remain unclear. Recently, it was shown that CD36, a major receptor for OxLDL on phagocytes, transmits only part of the OxLDL signals, with the ERK pathway remaining active in CD36 2/2 cells (26). Thus, other scavenger receptors, such as LOX-1, may be involved in stimulating ERK (27). Although LDL showed minor protection, the involvement of the LDL receptor can be ruled out. Treatment of LDL with proteinase K did not avert protection induced by LDL, although apoB, the component interacting with the LDL receptor, was lost during proteolytic digestion. Therefore, lipid components of LDL appear more important and lipids such as sphingolipids may contribute to protection (25). In addition, we showed that OxPAPC, an oxidized phospholipid thought to underlie the biological effects of minimally modified LDL, was cytoprotective. Our preparations resulted in LDL being less extensively oxidized compared with the mostly used copper-oxidized LDL. Therefore, oxidized phospholipids may be present in the OxLDL used by us and may be partly responsible for its cytoprotective effect.
We observed potent inhibition of staurosporine-induced phosphatidylserine exposure and caspase activation at 25-50 mg/ml OxLDL. However, at concentrations exceeding 100 mg/ml, OxLDL induced necrosis. A biphasic behavior, with protection dominating at low concentrations but a cytotoxic one at higher doses of OxLDL, is in agreement with published data (7,8). It is possible that the suppression of apoptotic mechanisms by OxLDL contributes to necrotic cell death at higher OxLDL concentrations and might also be relevant to necrotic core formation in advanced atherosclerosis. Similar observations have been made previously in human macrophages (21), in which OxLDL elicited a caspase-independent but ROS-dependent cell death.
PI3K-Akt and ERK signaling pathways are critical for the regulation of cell survival. We noticed that in addition to antiapoptotic actions attributed to Akt in lipoproteininduced survival (9,10,28), ERK can also be responsible for the cytoprotective effects of OxLDL, as shown by pharmacological inhibition experiments. Previous data and ours imply that both ERK and Akt signaling pathways may be critical for the prosurvival action of modified lipoproteins. Recently, we showed in the same THP-1 system that phospholipase A 2 -modified LDL activates Akt-dependent survival signaling when cell death was induced by H 2 O 2 or toxic concentrations of OxLDL (28). However, phospholipase A 2 -modified LDL at concentrations inducing prolonged Akt activation had no effect on staurosporineinduced apoptosis, supporting our notion that Akt activation alone is not sufficient to block staurosporine-induced apoptosis in this system (data not shown). The involvement of ERK signaling in cell protection was recently postulated when inhibition of ERK aggravated 7-ketocholesterol-induced apoptosis (29). As 7-ketocholesterol still induced caspase activation, despite ERK being active, this setup is different from ours, which shows a more prominent impact of ERK in protection.
The release of cytochrome c from mitochondria often is considered the "point of no return" in the cell death execution program. We show that OxLDL inhibits staurosporineinduced cytochrome c release in an ERK-dependent manner. This allows us to conclude that OxLDL interferes early in the apoptotic signaling cascade, most likely involving Bcl-2 family proteins. Indeed, one of the established ERK targets, Bad, was phosphorylated at serine 112, which constitutes the site of phosphorylation by ERK. Additionally, Mcl-1 degradation after staurosporine treatment was reduced by OxLDL, again with ERK inhibition reversing this effect. Mcl-1 is critical for the survival of myeloid cells (30), and interfering with its degradation appears to be an important antiapoptotic signal. Interestingly, BimEL, another ERK target within the Bcl-2 family (13), was not affected, suggesting some specificity in the OxLDL/ERK protective system. Intracellular ROS generated by OxLDL may participate in initiating various signaling events (31). Along that line, the activation of ERK by OxLDL in endothelial cells involved peroxynitrite-dependent Ras glutathionylation (23). Similarly, lysophosphatidylcholine, an OxLDL constituent, induced ERK activation in endothelial cells, which was dependent on mitochondrial ROS formation (32). Although we observed increased ROS generation by OxLDL that was antioxidant-sensitive, the same inhibitors failed to reverse protection by OxLDL. These results rule out ROS as participating messengers during the antiapoptotic action of OxLDL.
Together, our data provide evidence for the existence of an ERK-dependent antiapoptotic signaling pathway induced by OxLDL in phagocytes. The critical role of ERK in OxLDL-elicited protection suggests the possibility that blocking this pathway by pharmacological interference would enhance phagocyte apoptosis in the lesion, thus reducing cell numbers and disease progression.