- Top of page
- MATERIALS AND METHODS
- Author contributions
- For more information
- Supporting Information
Conjugated linoleic acid (CLA) is a generic term denoting a group of naturally occurring isomers of linoleic acid (18:2, n6), which vary in the position and geometry of their double bonds (Mitchell & McLeod, 2008). CLA has anti-atherogenic properties, inhibiting the progression of atherosclerosis in animal models (Kritchevsky et al, 2004; Lee et al, 1994). We have previously shown that dietary administration of a 1% CLA blend (80:20, cis9,trans11-CLA:trans10,cis12-CLA) induces regression of pre-established atherosclerosis in the apoE−/− mouse model despite continuing a high cholesterol challenge (Toomey et al, 2006). Understanding the mechanisms involved through which CLA mediates regression may help identify endogenous pathways that limit or reverse human atherosclerosis.
Previously, we reported that macrophage cell accumulation was reduced in the atherosclerotic plaques of CLA-fed mice suggesting this cell as a potential target for CLA (Toomey et al, 2006). Other studies have shown that CLA isomers reduce cholesterol accumulation in macrophage-derived foam cells potentially by enhancing lipid acceptor-dependent cholesterol efflux (Ringseis et al, 2008). Thus, altered cholesterol handling by macrophages may in part explain how CLA mediates its atheroprotective effect. However, other evidence suggests that CLA modulates the macrophage pro-inflammatory phenotype, with suppression of pro-inflammatory genes such as VCAM-1, matrix metalloproteinase (MMP)-9, platelet endothelial cell adhesion molecule (PECAM)-1 and pro-inflammatory cytokines interleukin (IL)-1α, IL-1β and IL-6 (Lee et al, 2009; Toomey et al, 2006). We have also shown that CLA inhibits the inflammatory phenotype of the monocyte/macrophage with reduced COX-2 and cPLA2 expression and inhibition of PGE2 and MMP-9 generation (McClelland et al, 2010). The findings cumulatively point to the monocyte/macrophage as an important cellular target for CLA in atherosclerosis.
One of the proposed mechanisms for the effects of CLA is activation of peroxisome proliferator-activated receptors (PPARs). Activated PPARs heterodimerize with the retinoic X receptor (RXR) and bind to specific peroxisome proliferator response elements (PPREs) located in the promoter of target genes thus regulating their transcription. Activators of PPARα and PPARγ inhibit foam cell formation (Li et al, 2004). Inhibition of foam-cell formation by PPARα is dependent on regulation of the nuclear receptor, liver X receptor α (LXRα), which regulates genes involved in cholesterol homeostasis and lipid metabolism, while activation of PPARγ stimulates HDL-dependent cholesterol efflux and inhibition of foam cell formation independently of LXRα (Li et al, 2004). In vivo, PPARγ agonists inhibit the development of atherosclerosis in the LDLR−/− and apoE−/− murine models (Chen et al, 2001; Li et al, 2000). Importantly, while activation of PPARγ inhibits the development of atherosclerosis, regression of atherosclerosis is not induced (Nakaya et al, 2009), which is in contrast to the regression seen with CLA supplementation (Toomey et al, 2004). Thus, PPAR activation alone cannot explain the response to CLA. Indeed, PPAR agonists and CLA differ in their effects on monocytes. We recently reported that both CLA and the PPARγ agonist, troglitazone, inhibit MCP-1-induced monocyte migration. However, only CLA inhibited platelet-releasate-induced migration implying a PPARγ-independent mechanism for CLA-induced suppression of monocyte migration (McClelland et al, 2010).
In this study we employed a comprehensive transcriptomic approach to identify novel gene clusters regulated by CLA during atherosclerosis regression in vivo. The implications of this are important as to date there is limited knowledge of gene networks which explain how atherosclerosis could be reversed in vivo. In a screen of genes regulated during CLA-mediated regression of atherosclerosis, we identified several gene clusters, one of which contained the gene PGC-1α at its hub. Given the role of PGC-1α in regulating genes involved in lipid metabolism and its emerging role in vascular cell function we pursued PGC-1α as a potential target of CLA induced atherosclerosis regression.
PGC-1α is a transcriptional coactivator of several nuclear receptors that regulate key metabolic steps in energy homeostasis (Puigserver & Spiegelman 2003), including PPARγ. PGC-1α is expressed in tissues such as heart and skeletal muscle where it regulates mitochondrial biogenesis via the regulation of genes involved in fatty acid oxidation and oxidative phosphorylation. PGC1-α also plays a role in vascular biology. PGC-1α is expressed in VSMCs and endothelial cells (Kim et al, 2007). Overexpression of PGC-1α blocks oleic acid (OA)-induced VSMC proliferation and migration (Zhang et al, 2007) and PGC-1α has been shown to mediate the inhibitory effect of dexamethasone on PDGF-induced VSMC migration (Xu et al, 2010). PGC-1α also inhibits neointimal formation in the rat carotid artery (Qu et al, 2009). In endothelial cells, PGC-1α triggered an increase in the expression of mitochondrial antioxidative enzymes (Valle et al, 2005), and prevented linoleic acid-induced increases in ROS generation and apoptosis by increasing fatty acid oxidation. In addition, PGC-1α serves as a coactivator of LXRα (Oberkofler et al, 2003) and activates Cyp7A1 expression (Shin et al, 2003), a crucial enzyme in mammalian cholesterol metabolism in the liver, where it mediates the classical pathway of bile acid synthesis (Chiang, 2002). These findings point to PGC-1α as a potential therapeutic target in atherosclerosis (Won et al, 2010).
Here, we identify regulation of a PGC-1α network during regression of pre-established atherosclerosis. We further extended our studies to explore the expression of PGC-1α in murine models and human atherosclerosis. Importantly we describe the effect of CLA and PGC-1α on macrophage to foam cell transition. Individual CLA isomers as well as the 80:20 CLA blend, which is atheroprotective in vivo was examined for their effect on oxLDL uptake and macrophage foam cell formation and on the expression of PGC-1α⋅ Finally, we directly explored the role of PGC-1α in regulating foam cell formation, over-expressing the gene in vitro and using PGC-1α depleted macrophages in vivo.
Our data shows for the first time that CLA regulates PGC-1α expression in macrophages in vitro and in vivo in murine and human atherosclerosis. Furthermore, we show that PGC-1α expression inhibits macrophage- foam cell transition. Finally for the first time we provide evidence, that macrophage specific PGC-1α depletion accelerates atherosclerosis in vivo, strongly supporting a role for PGC-1α in atheroprotection.
- Top of page
- MATERIALS AND METHODS
- Author contributions
- For more information
- Supporting Information
A critical step in the development of atherosclerosis is the accumulation of cholesterol in macrophages, which leads to foam cell formation. Cellular cholesterol content in macrophages is determined by uptake (mediated by scavenger receptors) and efflux of cholesterol (mediated by cholesterol acceptors) (de Villiers & Smart, 1999), an imbalance of which results in the formation of foam cells, which in turn promote lipid deposition and lesion growth.
CLA both inhibits progression (Lee et al, 1994) and induces regression (Toomey et al, 2006) of atherosclerosis in animal models. Here we provide evidence that the c9,t11-CLA isomer and the atheroprotective CLA blend alter the phenotype of the macrophage cell by inhibiting oxLDL uptake and subsequent foam cell formation, which confirm previous findings showing that CLA treatment of RAW 264.7 cells regulates foam cell formation (Ringseis et al, 2008). Interestingly, t10,c12-CLA had no effect on foam cell formation, consistent with the divergent effects of the two CLA isomers in atherosclerosis (Arbonés-Mainar et al, 2006). To determine if CLA inhibited oxLDL uptake, increased cholesterol efflux or a combination of both, we examined cholesterol efflux in acLDL loaded macrophage cells. Only the CLA blend altered cholesterol efflux towards HDL suggesting that inhibition of cholesterol uptake is the primary mechanism of action of CLA isomers in inhibiting foam cell formation. However, the increased efflux is consistent with the increased expression of ABCA1 in murine aorta during CLA-induced regression and thus it is feasible to suggest that this may be also related to the increase in Cyp7b1 as a mechanism for hydroxylated sterol removal. SR-A1 and CD36 have been identified as the two major receptors responsible for lipoprotein uptake into macrophages and mice lacking either receptor show a reduction in atherosclerotic lesions (Suzuki et al, 1997; Febbraio et al, 2000). We show reduced expression of the scavenger receptors, confirming a role for CLA in inhibition of oxLDL uptake.
One possible mechanism for the effects of CLA is activation of PPARγ (Ringseis et al, 2006). In the PPARγ knockout model, there is a reduction in oxLDL accumulation in macrophages but no difference in uptake of oxLDL (Babaev et al, 2005). In parallel PPARγ agonists do not change SRA-1 expression but upregulate the expression of CD36 (Tontonoz et al, 1998). In contrast, we found that CLA inhibits both CD36 and SRA-1 suggesting that PPARγ activation does not fully explain the effects of CLA. Indeed, we have previously shown that CLA inhibits monocyte migration towards platelet releasate through a PPARγ- independent mechanism (McClelland et al, 2010).
Using a transcriptomic approach we looked for additional related or distinct genes/pathways regulated by CLA that may yield further information as to how CLA modulates atherosclerosis and/or macrophage function. In searching for potential targets we identified the nuclear receptor co-activator PGC-1α, as a nexus gene in a network regulated by CLA. Interestingly, CLA regulation of PGC-1α has previously been reported in chronic inflammation. Bassaganya-Riera et al showed that CLA ameoliorated colitis in an experimental model. Interestingly, CLA induced both PPARγ and PPARδ expression and transcriptionally regulated gene clusters involved in lipid metabolism including UCP1, PGC-1α and CD36 which is in keeping with our findings (Bassaganya-Riera et al, 2004). Indeed, several activities of PGC-1α suggest it could play a role in atherosclerosis (Oberkofler et al, 2003; Qu et al, 2009; Xu et al, 2010; Zhang et al, 2007).
To validate the CLA regulation of PGC-1α in vivo we investigated the expression of PGC-1α in the CLA-induced regression model of atherosclerosis (Toomey et al, 2006). There was increased PGC-1α mRNA and protein expression which was localized to the macrophage cell in the aorta of apoE−/− mice supplemented with CLA coincident with a significant increase in expression of the PGC-1 α target gene, UCP-1. Interestingly, there was also decreased macrophage PGC-1α expression in plaques from symptomatic compared to asymptomatic patients, consistent with a role for PGC-1α in regulating the severity of disease. This is the first report that has identified dysregulation of PGC-1α in macrophage/foam cell in murine and human atherosclerosis.
We investigated the expression of PGC-1α in CLA mediated inhibition of foam cell formation. It is important to note that to date there has been no direct evidence linking PGC-1α with foam cell formation, although recently Stein et al reported increased accumulation of CD68 positive macrophage cells and ICAM-1 expression in atherosclerotic plaques from apoE−/−/PGC-1α−/− DKO mice when compared with apoE−/− mice (Stein et al, 2010). The data presented here show that c9-t11 CLA and CLA blend, which inhibit foam cell formation, also induce the expression of PGC-1α in oxLDL loaded macrophages.
To further address a role of PGC-1α in atheroprotection we investigated oxLDL accumulation in macrophage cells transfected with PGC-1α. Although not all cells expressed the GFP plasmid, those that did maintained the macrophage phenotype and did not accumulate oxLDL. Furthermore, similar to what was observed in RAW macrophages treated with CLA, over expression of PGC-1α significantly increased the expression of Cyp7b1 and UCP-1 and induced a small but significant decrease in Cd36 expression although this was not of the same magnitude as that observed with CLA treatment likely due to the low transfection efficiency of the cells. We supported these data by examining the effect of PGC-1α deletion on oxLDL uptake in primary bone marrow derived macrophages from PGC-1α−/− animals. Our data shows that PGC-1α gene deletion increases foam cell formation. Importantly, treatment of BMDMs from PGC-1α KO animals treated with CLA isomers or CLA blend had no effect on foam cell formation. In the LDLR−/− model of atherosclerosis peritoneal macrophages, which display characteristics of foam cells, have decreased PGC-1α expression. Importantly, using bone marrow transplantation studies we show that macrophage deletion of PGC-1α accelerated atherosclerosis and increased lesion size in the LDLR−/− mouse. A limitation of our study is that the evidence for the role of macrophage PGC-1α in atherosclerosis is based primarily on bone marrow transfer studies. Therefore, additional studies for verification are required in a macrophage cell specific knockout to fully define the role of macrophage PGC-1α in this context. Nonetheless, these data for the first time describe a role for PGC-1α in macrophage function and regulation of foam cell formation and moreover, suggest that there is an endogenous pathway involving PGC-1α that can be activated, for example by CLA, to limit, or induce regression of, atherosclerosis.
PGC-1α expression and/or CLA may invoke additional anti-atherogenic pathways such as the metabolism of cholesterol to bile acids.
In both CLA treated macrophages and in murine aorta in the CLA-induced regression model, we have shown an increase in Cyp7b1 expression, which is located in extrahepatic tissue (Chiang, 2002) where it inactivates oxysterols, products generated during the breakdown of oxLDL (Setchell et al, 1998). Enzymes involved in cholesterol to bile acid synthesis, specifically Cyp27, have been implicated to play a role in atherosclerosis (Björkhem et al, 1994) and thus it is feasible to suggest that increased Cyp7b1 expression by CLA, increases the removal of cholesterol metabolites from foam cells resulting in a less lipid laden foam cell phenotype. In our transcriptomic analysis we identified the nuclear receptor RORα on the PGC1α network and confirmed its regulation in CLA-induced regression in vivo. It is likely that CLA mediated regulation of both PGC-1α and RORα may regulate Cyb7b1 since it has previously been shown that in RORα−/− mice, which develop severe atherosclerosis, there is suppression of Cyp7b1 expression suggesting RORα may be a positive regulator of Cyp7b1 (Wada et al, 2008).
Several studies provide clues to the mechanism of PGC-1α regulation by CLA. Activation of PPARγ in adipocyte cells induces PGC-1α expression due to a PPRE at – 2043–2055 in the distal promoter sequence of mouse PGC-1α (Hondares et al, 2006) and here we show increased PPARγ expression in CLA treated macrophages. An alternative pathway is through activation of AMPK. Treatment of primary myotubes with the AMPK agonist AICAR, results in direct phosphorylation of the PGC-1α protein at threonine 177 and serine 538 (Jäger et al, 2007). We have recently shown an increase in phosphorylated/activated AMPK in aortic samples from CLA fed mice compared to cholesterol fed mice (Supporting Information Fig S10), suggesting that activation of AMPK may underlie the CLA regulation of PGC-1α.
Our work with CLA is designed to identify pathways that may prevent the progression and/or induce the regression of atherosclerosis. In this study we demonstrate that expression of PGC-1α is increased during CLA-induced regression of atherosclerosis and that CLA inhibition of foam cell formation is linked to induction of PGC-1α and that the gene alone could inhibit foam cell formation. Furthermore, we show that macrophage specific deletion of PGC-1α increases atherosclerosis in vivo. Importantly, we showed that PGC1α is expressed in the plaques of patients, raising the possibility that this is a regulatory pathway in human atherosclerosis.