Dear Editor,
The sub-endothelial retention of lipoproteins is one of the key events that trigger the atherosclerosis process. Low-density lipoprotein (LDL) particles trapped within the arterial wall are prone to progressive oxidation by monocytes/macrophages. Oxidized LDL (oxLDL) is present in atherosclerotic lesions, and has been suggested to play a significant role in atherogenesis (Nishi et al., 2002). The pathophysiology of atherosclerosis involves both apoptosis and proliferation at different stages of the vessel lesion. In advanced atherosclerotic plaques, up to 50% of the apoptotic cells are macrophages, which may promote core expansion and plaque instability (Tabas et al., 2009).
In our study, we defined the mechanism by which oxLDL induces apoptotic cell death in the J774A.1 macrophage cell line, a widely used in vitro model for evaluating the mechanisms underlying the adverse effects of oxidized lipids and oxLDL with native LDL (nLDL) as the control (Supplementary Materials and methods).
First, we demonstrated that oxLDL induced apoptosis in J774A.1 by activating caspase-9 and caspase-3, but not caspase-8 using the specific inhibitors, and determining the active form of caspases (Supplementary Figure S1). The first toxic effect of oxLDL was an early and progressive increase in reactive oxygen species (ROSs) production, showing a biphasic trend with the highest, non-reversible increase after 6 h, which corresponded to a significant depletion of glutathione (GSH) (Supplementary Figure S2). The onset of oxidative stress clearly indicated that the antioxidant system could no longer buffer the overproduction of ROSs. ROSs are involved in various biological events that are mediated by different signaling pathways, such as MAPK, NF-κB, Akt, and PKC (Irani, 2000; Cichon and Radisky, 2010), some controlling the level and nuclear accumulation of the tumor suppressor protein p53. In the presence of oxidative stress and DNA damage, p53 arrests the cell cycle, allowing time for cells to repair damaged DNA. If the damage cannot be successfully repaired, p53 acts as an apoptotic signal.
To unravel the molecular mechanisms responsible for the ROS-mediated, oxLDL-induced apoptosis, we investigated both the content of total and nuclear p53, and its serine-phosphorylated active form. The results clearly showed that oxLDL induced the activation of p53. The expression of p53 rapidly increased, becoming significantly higher than the controls (Figure 1A, a); after 1 h, the phosphorylated form showed an early significant increase, while the nuclear translocation increased after 3 h (Figure 1A, b and c). We may thus hypothesize that oxLDL initiated an early ROS overproduction, most probably followed by genotoxic stress which, in turn, activated the transcriptional dependent/independent activities of p53 by up-regulating its intracellular level and activation. It has to be considered that p53 is a zinc-binding, redox-sensitive protein, thus, the balance of reductants to oxidants within the cell must be critical in determining the structure and the function of p53 which, in turn, determines cell fate via selection of different target genes (Hainaut and Mann, 2001).
Role of p53, p66Shc, p38MAPK, and PKCδ in oxLDL-induced apoptosis. (A) The immunoblotting determination of p53 in J774A.1 cells treated with 0.1 mg/ml oxLDL or nLDL considered as the control. Total p53 protein content normalized to cyclophilin (a); serine-phosphorylated p53 detected in anti-p53 immunoprecipitate, and normalized to the non-phosphorylated protein (b); p53 nuclear content determined in nuclear extract, and normalized to lamin B protein (c). (B) The expression of p66Shc protein assessed by immunoblotting in cells exposed to oxLDL or control nLDL. Data were normalized to cyclophilin protein content. (C) Effects of the p53 inhibitor pifithrin-α on the expression of p66Shc (a), Bax (b), and caspase-3 (c) after oxLDL exposure compared with nLDL-treated control cells. Data were normalized to cyclophilin protein content. (D) Effects of p38MAPK and PKCδ gene silencing on the expression of serine-phosphorylated p53 (a) and caspase-3 (b) normalized to the non-phosphorylated p53 and to cyclophilin protein content, respectively, in oxLDL-treated cells transfected with anti-p38MAPK siRNA (oxLDL + si-p38), with the anti-PKCδ siRNA (oxLDL + si-PKCδ), or with the corresponding scrambled RNAs (oxLDL + scrambled-p38 and oxLDL + scrambled-PKCδ). Untransfected oxLDL-treated cells (oxLDL) were considered as the control. (E) PKCδ activation depended on p38 activity: (a) effect of 10 μM SB203580 and p38MAPK gene silencing on the expression of phosphorylated PKCδ, (b) effect of 10 μM rottlerin and PKCδ gene silencing on the expression of phosphorylated p38MAPK. Each panel shows representative blots from at least three independent experiments. Values are the mean ± SEM. *P< 0.05 vs controls. Scram-p38, scrambled-p38; scram-PKCδ, scrambled-PKCδ; SB, SB203580; rottl, rottlerin.
We further investigated the targets of p53, p66Shc (Giorgio et al., 2005), and Bax. p66Shc was significantly higher at 6 h after oxLDL exposure, then dropped to control level (Figure 1B). It is worth noting that the up-regulation of p66Shc was before the massive overproduction of ROSs and the dramatic depletion of GSH, strongly suggesting that p66Shc might act as a temporary limited intracellular signal that can regulate cell response to ROS overproduction, and give rise to oxidative stress. The oxidoreductase activity of p66Shc is probably due to the increase in the p66Shc mitochondrial fraction (Giovannini et al., 2008). Furthermore, we observed the up-regulation of Bax, which corresponded to the down-regulation of the anti-apoptotic Bcl-XL, 18 and 24 h after treatment as a final effect during the apoptotic execution phase (data not shown). We also observed the inhibitory effect of pifithrin-α, a p53 inhibitor, on the up-regulation of p66Shc, Bax, and activated caspase-3 (Figure 1C, a–c), indicating the central role of p53 in triggering oxLDL-mediated apoptosis.
Finally, we provided evidence of the mechanism responsible for the activation of p53. The increase in p53 protein level is primarily an effect due to the post-translational rather than the transcriptional level (Lavin and Gueven, 2006), predominantly induced by the multiple-site phosphorylation of p53 (Wu, 2004). The treatment of macrophages with PD98059 (ERK1/2 inhibitor) or SP600125 (JNK inhibitor) did not alter phosphorylated p53 level in oxLDL-treated cells. Conversely, the p38MAPK inhibitor SB203580 and the PKCδ inhibitor rottlerin significantly attenuated (∼40% reduction) oxLDL-induced p53 phosphorylation (Supplementary Figure S3A). In addition, only the pre-treatment with SB203580 or rottlerin did reduce the level of active caspase-3 to control values (∼65% inhibition) in oxLDL-treated cells (Supplementary Figure S3B); furthermore, we demonstrated that both p38MAPK and PKCδ were immediately activated (after 15 min of treatment) in cells exposed to oxLDL (Supplementary Figure S3C and Supplementary Data). All these data confirmed the role of both p38MAPK and PKCδ in oxLDL-induced and p53-mediated apoptosis.
To provide conclusive evidence of the causal relationship between p38MAPK and PKCδ activities and the activation of p53 and caspase-3, we silenced p38MAPK/PKCδ gene expression in oxLDL-treated cells. First, we evaluated the effectiveness of gene silencing in oxLDL-treated macrophages, and showed that the cells transfected with the anti-p38 siRNA and the anti-PKCδ siRNA reduced the expression of the proteins by 64% and 80%, respectively. Next we showed that the silencing of a single kinase gene resulted in the inhibition of p53 phosphorylation and pro-caspase-3 cleavage, which provides conclusive evidence that the activities of both kinases were necessary for oxLDL to trigger macrophage apoptosis (Figure 1D, a and b). We can hypothesize that the phosphorylation of specific combined amino acids by both p38MAPK and PKCδ was necessary for p53 stabilization.
Further experiments aimed at unraveling the relationship between p38MAPK and PKCδ. We alternately inhibited each kinase using specific inhibitors. We found that, by inhibiting p38MAPK, PKCδ serine phosphorylation was strongly reduced (Figure 1E, a); conversely, the inhibition of PKCδ did not influence p38MAPK serine phosphorylation (Figure 1E, b). This relationship was further confirmed by silencing the corresponding genes (Figure 1E). In fact, while phosphorylated PKCδ was inhibited in oxLDL-treated cells transfected with anti-p38MAPK siRNA, the treatment with the anti-PKCδ siRNA had no effect on the phosphorylation of p38MAPK. These results suggested that p38MAPK was the upstream kinase in the induction of oxLDL-induced apoptosis in J774A.1 macrophage cells.
Our findings point to the likely conclusion that oxLDL early activated kinase pathways responsible for p53 stabilization, phosphorylation, and nuclear translocation. The combined activities of p38MAPK and PKCδ were necessary for p53 stabilization and, worthy of note, they were closely connected to each other, as PKCδ activation depended on p38MAPK activity. Furthermore, the p53-induced temporary overexpression of p66Shc seemed to be the intracellular signal responsible for triggering the execution of the cell death internal program. This study provides a novel, in-depth view of the events underlying the action of oxLDL that leads to macrophage death, and points out the key role played by p38MAPK, PKCδ, and p66Shc in the p53-signaling pathway.
[Supplementary material is available at Journal of Molecular Cell Biology online. We thank Prof. Gabriella Girelli, Director of Centro Trasfusionale, University of Rome ‘La Sapienza’, for providing human plasma, and Ms. Monica Brocco, Istituto Superiore di Sanità, Rome, Italy, for the English editing of the manuscript.]
