PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Circ Res. Author manuscript; available in PMC 2016 July 17.
Published in final edited form as:
PMCID: PMC4578799
NIHMSID: NIHMS697006

Macrophage Mitochondrial Energy Status Regulates Cholesterol Efflux and is Enhanced by Anti-miR33 in Atherosclerosis

Abstract

Rationale

Therapeutically targeting macrophage reverse cholesterol transport is a promising approach to treat atherosclerosis. Macrophage energy metabolism can significantly influence macrophage phenotype, but how this is controlled in foam cells is not known. Bioinformatic pathway analysis predicts that miR-33 represses a cluster of genes controlling cellular energy metabolism that may be important in macrophage cholesterol efflux.

Objective

We hypothesized that cellular energy status can influence cholesterol efflux from macrophages, and that miR-33 reduces cholesterol efflux via repression of mitochondrial energy metabolism pathways.

Methods and Results

In this study, we demonstrated that macrophage cholesterol efflux is regulated by mitochondrial ATP production, and that miR-33 controls a network of genes that synchronize mitochondrial function. Inhibition of mitochondrial ATP synthase markedly reduces macrophage cholesterol efflux capacity, and anti-miR33 required fully functional mitochondria to enhance ABCA1-mediated cholesterol efflux. Specifically, anti-miR33 de-repressed the novel target genes PGC-1α, PDK4 and SLC25A25 and boosted mitochondrial respiration and production of ATP. Treatment of atherosclerotic Apoe-/- mice with anti-miR33 oligonucleotides reduced aortic sinus lesion area compared to controls, despite no changes in HDL-C or other circulating lipids. Expression of miR-33a/b was markedly increased in human carotid atherosclerotic plaques compared to normal arteries, and there was a concomitant decrease in mitochondrial regulatory genes PGC-1α, SLC25A25, NRF1 and TFAM, suggesting these genes are associated with advanced atherosclerosis in humans.

Conclusions

This study demonstrates that anti-miR33 therapy de-represses genes that enhance mitochondrial respiration and ATP production, which in conjunction with increased ABCA1 expression, works to promote macrophage cholesterol efflux and reduce atherosclerosis.

Keywords: Mitochondria, macrophages, efflux, microRNA-33, PGC-1α, atherosclerosis, cholesterol

Introduction

The accrual of modified lipoproteins and macrophages in the vessel wall drives the progression of atherosclerosis1. Excess circulating lipoproteins, in particular LDL, become trapped beneath the protective endothelial layer and become modified in the oxidant-rich environment, recruiting monocytes that differentiate into inflammatory macrophages. In an attempt to restore the lipid-balance within the vessel wall, excess intracellular cholesterol is removed from macrophage foam cells into the reverse cholesterol transport (RCT) pathway by the interaction of ABC cholesterol transporter proteins (e.g. ABCA1) and apolipoprotein A-I (apoA1), a component of high-density lipoprotein (HDL). Although HDL-C has been widely used as a surrogate for HDL function, it is now appreciated that that capacity for HDL to promote cholesterol efflux from macrophages is a more predictive measure of the anti-atherosclerotic abilities of HDL2-4. However, to date, there are no therapies that specifically enhance macrophage RCT in the vessel wall.

Mitochondria are widely recognized as the powerhouses of the cell, yet, relatively little is known about how mitochondrial function is regulated in macrophages, especially as it relates to foam cell formation in atherosclerosis5. Mitochondrial metabolism encompasses a complex series of oxidizing processes that ultimately produce the cell's energy currency, ATP, which are tightly regulated by both nuclear and mitochondrial genes6. Macrophages can have a high demand for cellular energy, and require the efficient use of either glycolysis and/or fatty acid metabolism to maintain necessary levels of ATP to meet the demands during inflammation and, in the atherosclerotic plaque, increased rates of cholesterol efflux7. While the plasticity of macrophage energy metabolism is becoming evident during acute inflammation8, less is known about how macrophage mitochondrial energy metabolism is controlled during atherosclerosis.

Excitement has grown for the potential of microRNA-based therapeutics for preventing and regressing atherosclerosis9-12. In particular, strategies that inhibit miR-33 have shown promise in both small and large pre-clinical models to raise HDL-C and promote RCT and the regression of existing atherosclerotic plaques13, 14. miR-33 (and its human equivalents miR-33a and miR-33b) is an intronic miRNA located within the gene coding for the cholesterol transcription factor SREBP-2 (and in humans, also SREBP-1) where they cooperate to increase cholesterol synthesis and decrease cholesterol elimination, in part through targeting cholesterol transport proteins ABCA1 and ABCG115-17. In addition, miR-33 represses genes involved in fatty acid oxidation (i.e. HADHB, CROT, Cpt1α) and high levels of miR-33 activity can limit the oxidation of fatty acids18. Thus, although miR-33 has been primarily studied for its role in regulating HDL biogenesis and circulating HDL levels, it is becoming increasingly evident that miR-33 post-transcriptionally modifies a number of genes that control cellular energy status.

Given the importance of mitochondrial metabolism in maintaining cellular energy status, we hypothesized that mitochondrial function can regulate cholesterol efflux from foam cells, and that miR-33 can fine-tune cellular energetic pathways to regulate this process. Microarray analysis of plaque macrophages from anti-miR33 treated Ldlr-/- mice revealed a series of novel miR33 targets genes that were de-repressed upon miR-33 inhibition14, including PPARγ coactivator-1 α (PPARGC1A, or PGC-1 α), a gene that plays a central role in regulating cellular energy homeostasis and mitochondrial metabolism. As miRNAs are known to target entire functional pathways, we tested whether miR-33 targets a network of genes controlling mitochondrial metabolism and identified additional novel targets in this pathway. Furthermore, using gain and loss of function approaches, we show that this regulation contributes to the regulation of cholesterol efflux in atherosclerosis in mice, and is associated with the presence of advanced atherosclerosis in humans.

Methods

An expanded Materials and Methods section is included in the Online Supplement, and provides details on the identification of novel miR-33 target genes using bioinformatics, the delivery of anti-microRNAs in vitro and in vivo, assays measuring mitochondrial respiration and ATP production, as well as details regarding the analysis of human atherosclerotic plaque tissue for miR-33a/b and target gene expression.

Results

Mitochondrial activity controls cholesterol efflux in macrophages

Excess cholesterol removal from macrophage foam cells is mediated through ABCA1 transporters on the cell surface, which account for a significant portion of both phospholipid and cholesterol efflux from the cell. Although the exact mechanism by which these lipids are transported across the cell membrane onto apoA1 remains incompletely understood, it is known to be an ATP-dependent process19. Therefore we hypothesized that mitochondrial function was important for efficient cholesterol removal from macrophages to apoA1. To test this hypothesis, we treated THP-1 macrophages with the mitochondrial respiration inhibitor oligomycin, which blocks the production of ATP from oxidative phosphorylation in the mitochondria. This resulted in a significant reduction in cholesterol efflux to apoA1, diminishing the percentage of cholesterol efflux back to those found the absence of an acceptor, similar to what had been observed previously in mouse macrophages7 (Figure 1A). Similarly, macrophages from Pgc-1α-/- mice, which have reduced mitochondrial function and a reduced capacity for oxidative phosphorylation, showed impaired cholesterol efflux to apoA1, in both cholesterol-loaded and unloaded conditions (Figure 1B). Taken together, these results confirm that mitochondrial production of ATP via oxidative phosphorylation is important for efficient cholesterol efflux from macrophages, and confirms the notion that enhancing mitochondrial function may serve to enhance cholesterol removal from foam cells.

Figure 1Figure 1
Mitochondria are required for cholesterol efflux in macrophages and are predicted to be regulated by miR-33

Cholesterol efflux is tightly controlled by both transcriptional and post-transcriptional mechanisms. miR-33 has been shown to modulate cholesterol efflux pathways by reducing the expression of the cholesterol transporters ABCA1 and ABCG1, however, relatively little is known about its impact on other energy metabolism pathways. As mitochondria are central regulators of cellular energy homeostasis, we sought to determine whether miR-33 targets genes involved in maintaining mitochondrial function. We interrogated a robust list of miR-33 predicted target genes, as determined using 5 prediction algorithms, and performed bioinformatic pathway analysis using the DAVID tool. In addition to PGC-1α, we identified a number of other genes encoding mitochondrial proteins predicted to be targeted by miR-33, including genes involved in oxidation of pyruvate (pyruvate dehydrogenase kinase 4, or PDK4), solute carrier proteins (SLC25A25, SLC25A23) and previously confirmed targets involved in fatty acid oxidation (HADHB, CROT)18 (Table 1). Molecular interaction analysis using Cytoscape revealed that many of the miR-33 targets, both predicted and validated, interact with other mitochondrial genes, suggesting that miR-33 may regulate mitochondrial function by both direct and indirect mechanisms (Figure 1C).

Table 1
GO analysis of predicted miR-33 target genes involved in mitochondrial function. (Bold – previously identified miR-33 targets, Italics – miR-33 targets of interest)

Increasing mitochondrial gene expression via miR-33 pathway inhibition

Given their established role in metabolism in vivo20-22, we next sought to confirm the putative target genes PGC-1α, SLC25A25, and PDK4 as direct miR-33 targets. Over-expression of miR-33 mimics in conjunction with candidate 3′UTR-luciferase reporter constructs confirmed that miR-33 directly represses the 3′UTR of human PGC-1α 23, PDK4 and SLC25A25 and disrupting the miR-33 binding sites in these genes by site-directed mutagenesis with these sites abolishes the inhibitory effects of miR-33 on these genes (Figure 2A, Supplemental Figure I). miR-33 binding sites are also conserved in the 3′UTR of these genes in mice, indicating that miR-33 can repress gene expression in both species (Supplemental Figure I). To confirm whether miR-33 endogenously regulates mitochondrial gene expression in macrophages, we transfected mouse peritoneal and human THP-1 macrophages with anti-miR33 or control anti-miRs and examined the expression of target genes. We observed a significant de-repression of PGC-1α, PDK4, SLC25A25 and SLC25A23, as well as the previously established miR-33 target genes ABCA1 and HADHB, at the mRNA level (Figure 2B). Moreover, anti-miR33 treatment increased the protein expression of PGC-1α, PDK4 and SLC25A25 as observed by western blot analysis, in both mouse and human macrophages (Figure 2C). These data confirm that in macrophages, miR-33 directly regulates the expression of PGC-1α, SLC25A25 and PDK4 via binding to complementary sites in the 3′UTR, in addition to its previously described targets ABCA1 and HADHB.

Figure 2
Anti-miR33 treatment increases the expression of novel mitochondrial genes in macrophages

Our pathway interaction analysis suggests that miR-33 has the capacity to regulate the expression of multiple mitochondrial genes, both directly (i.e. by 3′UTR binding) and indirectly (i.e. via interaction with direct miR-33 targets) (Figure 1C). In particular, PGC-1α directly activates important activators of mitochondrial biogenesis, including nuclear respiratory factor 1 (NRF1) a transcription factor that activates expression of nuclear-encoded mitochondrial genes and is essential for mitochondrial respiration. We therefore measured the expression of NRF1 and show that indeed miR-33 inhibition leads to an upregulation of NRF1 mRNA (Figure 3A) and protein expression (Figure 3B). As the 3′UTR of NRF1 in both human and mouse does not contain any predicted miR-33 binding sites, we conclude that the upregulation of NRF1 upon miR-33 inhibition is an indirect consequence of the de-repression of other direct miR-33 target genes that control mitochondrial biogenesis, primarily PGC-1α. The increase in PGC-1α and NRF1 resulted in a significant increase in mitochondrial DNA copy number, a readout of mitochondrial biogenesis, in macrophages treated with anti-miR33 compared to controls (1.7-fold anti-miR33 vs. cont anti-miR, p≤0.01, Figure 3C). To further explore the indirect effects of anti-miR33 on macrophage mitochondrial gene expression, we used a pathway-focused PCR arrays, which contain 84 genes involved in mitochondrial metabolism, and compared control anti-miR and anti-miR33 treated macrophages. In agreement with our network interaction analysis, mitochondrial pathway arrays demonstrated that miR-33 regulates the expression of several genes involved in mitochondrial function (Supplemental Figure IIB and Table 1). Of the genes analyzed in the mitochondrial pathway, some genes containing miR-33 binding sites in their 3′UTR showed de-repression upon miR-33 inhibition (i.e. Slc25a25, Slc25a23) while others lacking miR-33 binding sites also showed up- and down-regulated expression (i.e. Ucp-2, Bcl-2). We next measured levels of the mitochondrial oxidative phosphorylation machinery (OXPHOS), a complex of proteins that are found on the inner membrane of the mitochondria and together produce the majority of cellular ATP. The OXPHOS complex is comprised of complexes I-V, which together use the reducing equivalents from oxidized fuels to produce the protonmotive force across the membrane, which then drives the conversion of ADP to ATP by ATP synthase. We quantified the levels of the OXPHOS machinery in macrophages transfected with anti-miR33 or control oligonucleotides, and show that complexes I, III, IV and V are significantly up-regulated with miR-33 inhibition (Figure 3D). Together, these data reveal that in addition to directly targeting mitochondrial genes PGC-1α, PDK4 and SLC25A25, anti-miR33 indirectly modifies the expression of additional mitochondrial metabolism genes that may have important functional consequences on mitochondrial metabolism.

Figure 3
Indirect regulation of mitochondrial gene expression by miR-33

Anti-miR33 treatment increases mitochondrial respiration and ATP production

Given its ability to alter mitochondrial gene expression and augment oxidative phosphorylation machinery, we tested whether anti-miR33 can specifically enhance mitochondrial function. We measured the outcome of miR-33 inhibition on oxygen consumption rates (OCR) in macrophages, under ADP phosphorylating and non-phosphorylating conditions. Cellular respiration was quantified in macrophages transfected with control anti-miR or anti-miR33 using the Seahorse XF Extracellular Flux Analyzer. Under basal conditions, anti-miR33 treatment increased OCR compared to control cells (Figure 4A). When oxidative phosphorylation is blocked using oligomycin to induce the non-phosphorylating (proton leak) condition, anti-miR33 had no effect on OCR. Finally, anti-miR33 enhanced maximal cellular respiration rates after treatment with uncoupling agent, FCCP, suggesting inhibition of miR-33 promotes electron transport chain activity (and/or substrate delivery) in the mitochondria (Figure 4A). To further confirm that miR-33 alters mitochondrial energy metabolism, we measured the intracellular ATP production as a measure of optimal mitochondrial function and activity. As shown in Figure 4B, macrophages over-expressing miR-33 had decreased ATP production relative to control (-45.83%, p≤0.05). In contrast, the inhibition of miR-33 resulted in increased production of ATP relative to controls (+28.01%, p≤0.05). Collectively, these data demonstrate that the inhibition of miR-33 positively drives mitochondrial aerobic respiration and activity, which in turn results in an increase in ATP production, likely through its direct and indirect modulation of multiple mitochondrial genes.

Figure 4
Inhibition of miR-33 increases mitochondrial respiration and ATP production that contributes to macrophage cholesterol efflux potential

Given the previously established role of miR-33 in controlling cholesterol efflux and RCT13, 14, 16, we postulated that miR-33 could regulate cholesterol efflux activity via its effects on mitochondria gene expression. To test this, we measured cholesterol efflux to apoA1 in macrophages transfected with control anti-miR or anti-miR33 in the presence or absence of oligomycin, an inhibitor of mitochondrial respiration. As in previous studies, macrophages transfected with anti-miR33 had increased cholesterol efflux to apoA1 compared to controls (Figure 4C). Inhibition of mitochondrial ATP production blocked the ability of anti-miR33 to promote cholesterol efflux. In the absence of optimally respiring mitochondria, macrophages treated with anti-miR33 had reduced cholesterol efflux compared to controls, suggesting that the salutary effects of miR-33 inhibition absolutely require mitochondrial respiration and energy production (Figure 4C). These experiments highlight an essential role for mitochondrial respiration and ATP production in miR-33 regulation of cholesterol homeostasis in macrophages, and represent a novel mechanism by which cholesterol efflux can be augmented in atherosclerosis.

Given the ability of miR-33 to control mitochondrial gene expression and function, we next asked whether the regulation of cholesterol efflux by anti-miR33 was dependent on PGC-1α and other newly identified mitochondrial target genes. Using mice deficient in Pgc-1α, which have with impaired mitochondrial metabolism, we measured ATP production upon inhibition of miR-33. In wild-type (WT) macrophages, anti-miR33 robustly augments ATP production, but this effect is lost in the absence of Pgc-1α (Figure 4D). We next tested the dependency of PGC-1α on anti-miR33 regulation of cholesterol efflux. While anti-miR33 treatment results in a robust 50% increase in cholesterol efflux in macrophages from WT mice, anti-miR33 has no effect on cholesterol efflux in macrophages from Pgc-1α-/- mice (Figure 4E). In contrast, anti-miR33 could still augment efflux in Pdk4-/- macrophages, albeit to a lesser extent that WT macrophages (Figure 4E). Anti-miR33 could equally increase cholesterol efflux in both Slc25a25-/- and WT cells (Figure 4E). Taken together, these data reveal that miR-33 inhibition depends on functional Pgc-1α to regulate ATP production as well as cholesterol efflux from macrophages, highlighting the essential role for mitochondrial metabolism in the salutary effects of anti-miR33.

Anti-miR33 protects from atherosclerosis and increases mitochondrial gene expression in Apoe-/- mice

Previous studies using miR-33 inhibition have shown beneficial effects on RCT and atherosclerosis progression and regression14,24. Part of the athero-protective mechanism of miR-33 inhibition may be attributable to raising of HDL-C14, 15, 25, however, studies in Ldlr-/- mice on a western diet have shown reductions in atherosclerosis in the absence of HDL-raising24. We wondered if the mitochondrial pathways regulated by anti-miR33 could promote HDL-independent cholesterol efflux pathways in lesions from the highly inflamed Apoe-/- mice, which have little to no circulating HDL. Eight week old Apoe-/- mice were simultaneously fed a western diet (WD) and administered control anti-miR or anti-miR33 oligonucleotides via weekly injections for 8 weeks. Quantification of the aortic sinus lesion area shows a reduction in lesion burden in mice treated with anti-miR33 compared to their controls (Figure 5A, p≤0.05). However, the anti-miR33 dose used in this study did not affect circulating HDL levels (Figure 5B), nor did it alter total plasma cholesterol or LDL cholesterol (Table 2). Visualization of lipid droplets, elastin, and collagen was performed used coherent anti-Stokes Raman scattering (CARS), two-photon fluorescence (TPF), and second harmonic generation (SHG), respectively (Figure 5C). Anti-miR33 treated mice have reduced lipid aggregation and overall lesion area (Figure 5D), both of which are consistent with an increased macrophage cholesterol efflux capacity induced by anti-miR33 therapy. Given that miR-33 inhibition resulted in increased expression of PGC-1α and other mitochondrial target genes in vitro, we measured the expression of these targets in the lesions of Apoe-/- treated mice in vivo. Immunohistochemical staining of aortic sinus lesions demonstrate that Apoe-/- mice treated with anti-miR33 have increased expression of Pgc-1α and Pdk4 in the plaque compared to control treated mice (Figure 5D).

Figure 5
Anti-miR33 therapy decreases atherosclerosis in Apoe-/- mice independently of HDL cholesterol
Table 2
Body weight and serum cholesterol measurements in ApoE-/- mice

To understand whether the expression of novel miR-33 controlling mitochondrial metabolism are dysregulated in the pro-atherogenic milieu, we isolated peritoneal macrophages from hypercholesterolemic mice, which are considered a surrogate for plaque macrophages26. In vivo formed foam cells from Apoe-/- mice treated with anti-miR33 treated mice showed significantly increased expression of target genes Abca1, Pgc-1α and Slc25a25 (Figure 6A, p≤0.05), in parallel with the observed increases in of Pgc-1α and Pdk4 protein in the vessel wall. We also measured the corresponding levels of ATP in the aortas of Apoe-/- mice, and mice treated with anti-miR33 tended to have increased levels of aortic ATP compared to controls (968.8 ± 225.5μM ATP versus 2930 ± 1367μM ATP/total lesion area, p=0.15; Supplemental Figure IIIA). We analyzed plaque macrophage gene expression using laser-capture microdissection in a related model of atherosclerosis progression, WD-fed Ldlr-/- mice treated with anti-miR33 or control anti-miR for 8 weeks. Quantitative PCR on LCM-isolated macrophages revealed a trend towards increased expression of miR-33 target genes Pgc-1α and Slc25a23, and a significant increase in Pdk4 mRNA (Supplemental Figure IIIC). Nrf1, a Pgc-1α downstream target and a marker of mitochondrial biogenesis, was very robustly increased in lesional macrophages from anti-miR33 treated mice compared to controls (Supplemental Figure IIIC). Finally, in an attempt to translate our findings to humans, we examined whether miR-33 and its mitochondrial target genes are dysregulated in human atherosclerosis. Indeed, miR-33 (both copies, miR-33a and miR-33b) are significantly elevated in atherosclerotic plaques from patients with carotid atherosclerosis compared to control arteries (Figure 6B). This was associated with a parallel decrease in miR-33 target gene expression of PGC-1α, SLC25A25 and SLC25A23, as well as in the indirect markers of mitochondrial biogenesis NRF1 and TFAM (Figure 6B). These data are the first to show that miR-33a/b expression is dysregulated in atherosclerosis in humans, and that genes with known roles regulating mitochondrial biogenesis (i.e. PGC-1α, NRF1, TFAM) are significantly lower in atherosclerotic versus healthy arteries. Together these data suggest that anti-miR33 therapy would be predicted to increase the expression of mitochondrial gene expression in the plaque to promote mitochondrial respiration and cholesterol efflux capacity, which may contribute to the ability of miR-33 inhibitors to reduce atherosclerotic lesion size in the absence of changes in HDL-C, and that this pathway may be active in human atherosclerosis development.

Figure 6
Novel miR-33 mitochondrial target gene expression is regulated in vivo

Discussion

The metabolic status of a cell is a strictly regulated process. In cells of the innate immune system, whose main purpose is to offer rapid protection against invading non-self and modified-self antigens, the availability of energy substrates is vital to mount an appropriate response in times of stress. For macrophages in the atherosclerotic plaque, this stress comes in the form of excess cholesterol accumulation, and macrophages need to boost the removal and detoxification of cholesterol via efflux pathways. We demonstrate that macrophage cholesterol efflux is dependent upon functionally respiring mitochondria, as inhibiting mitochondrial ATP production and function by pharmacologic (i.e. oligomycin) or genetic (i.e. deletion of Pgc-1α) means significantly blunts the ability of cells to efflux cholesterol to apoA1. Moreover, we show that miR-33, a microRNA with reported roles in controlling cholesterol efflux, can also regulate production of ATP by the mitochondria and can limit ATP availability for ABCA1-dependent cholesterol efflux. These data indicate that boosting energy metabolism pathways may be a novel method to enhance cholesterol efflux in macrophages and positively impact atherosclerosis development.

miRNAs are small but potent post-transcriptional modulators of gene expression that can regulate entire genetic networks. Previous studies from our group and others have shown that miR-33 (miR-33a/b in humans) controls the expression of genes that influence cellular metabolism, including cholesterol transport (e.g. ABCA1, ABCG1) and fatty acid β-oxidation (e.g. CPT1α, CROT, HADHB)17, 18 pathways. Using bioinformatic analysis, we have now identified another network of genes targeted by miR-33 to regulate mitochondrial respiration and metabolism. We confirmed several novel mitochondrial genes (i.e. PGC-1 α 23, PDK4 and SLC25A25) that are direct and specific targets of miR-33, with conserved binding sites in the 3′UTR of both human and mouse transcripts. Importantly, inhibition of endogenous miR-33 in mouse and human macrophages increases the protein expression of PGC-1α, PDK4 and SLC25A25, as well as the mitochondrial fatty acid oxidation genes HADHB and CROT, which are also direct targets of miR-3318, 27. In addition to the direct mRNA targets of miR-33, we also found that anti-miR33 can indirectly increase the expression of NRF1 and OXPHOS complexes- key factors that promote mitochondrial biogenesis and efficient production of ATP, respectively. Both NRF1 and OXPHOS are down-stream of PGC-1α, and miR-33 inhibition could result in an increase in mitochondrial biogenesis when PGC-1α levels are de-repressed28. Indeed, we observed a significant increase in mitochondrial DNA copy number, a readout of mitochondrial biogenesis, in anti-miR33 treated macrophages. Although the changes in expression exerted by miR-33 on any one target gene may be small, the cumulative functional outcome of fine-tuning many genes in the same pathway or network can be large, as evidenced by the significant changes in ATP synthesis upon manipulation of miR-33 levels. Thus, these data support the notion that miR-33 exerts control over genetic networks that regulate cellular energy homeostasis (i.e. lipid, fatty acid, mitochondrial), with miR-33 serving as a regulatory hub of energy metabolism.

ABCA1 is a member of the ABC transporter protein family, with two ATP-binding cassettes in its inner membrane domain, which hydrolyze and transport ATP molecules during efflux29, 30. During the transport of cholesterol and phospholipids, a steady supply of ATP is required for ABCA1 (and other ABC transporters, such as ABCG1) to move substrate across the membrane to an acceptor, making ATP central to the efficient removal of cholesterol. In addition to directly exerting post-transcriptional control over ABCA1 expression, we now show that miR-33 regulates the availability of ATP for cholesterol efflux via repressing a network of genes that control mitochondrial respiration and ATP production. miR-33 expression is activated during times of nutrient depletion (i.e. sterol, fatty acid) where it serves to repress the expression of cholesterol efflux and fatty acid oxidation pathways, in order to conserve energy31. Our data is consistent with miR-33 conserving cellular energy by regulating multiple pathways, eg. repression of sterol and fatty acid utilization, and concomitantly dampening of ATP production. Further, our data also supports the hypothesis that under conditions of excess cholesterol (i.e. in atherogenic macrophage foam cells), enhancing mitochondrial function can promote cholesterol efflux in part by increasing ATP production. Indeed, de-repression of PGC-1α, PDK4 and SLC25A25 upon inhibition of miR-33 corresponds to an increase in mitochondrial respiration and ATP production – both of which contribute to the increased cholesterol efflux capacity seen in anti-miR33 transfected macrophages. We found that inhibition of mitochondrial function (i.e. upon treatment with oligomycin) markedly reduced macrophage cholesterol efflux to apoA1, and blocking normal mitochondrial function blunted the beneficial effects of anti-miR33 on macrophage cholesterol efflux, suggesting respiring mitochondria are necessary for the favourable effects of anti-miR33 on efflux. Similarly, in the absence of PGC-1α, anti-miR33 can no longer exert its effects on ATP production or cholesterol efflux, once again confirming that intact mitochondrial ATP production enhances anti-miR33 regulation of cholesterol efflux. Mitochondrial dysfunction has been mechanistically linked to the progression of atherosclerosis in Apoe-/- mice, independent of reactive oxygen species production, due in part to defects in oxidative phosphorylation32. Furthermore, it has been suggested that mitochondrial distress can reduce the efficiency of cholesterol efflux from macrophages, which could be a consequence of reduced ATP production by the mitochondria5. Here, we show enhanced mitochondrial respiration in anti-miR33 treated macrophages by de-repression of key mitochondrial genes to promote ATP production and in conjunction with the major miR-33 target gene ABCA1, enhances cholesterol efflux, indicating that miR-33 dampens cholesterol efflux pathways via its direct effects on the mitochondria. Interestingly, we observed the greatest increase in mitochondrial respiration by anti-miR33 in the presence of the uncoupling agent, FCCP. In uncoupled states, when the mitochondria are in overdrive to respire, we believe that miR33 represses mitochondrial biogenesis and respiration through its multitude of mitochondrial gene targets. Anti-miR33 de-represses many of these gene targets simultaneously, amplifying their additive effects on an overdriven respiratory system, which when translated in vivo, could have beneficial outcomes in the plaque where defective oxidative phosphorylation and mitochondrial function is present. Similarly, although we measured ATP production at one fixed time point, blocking miR-33 continuously over time would expect to have a sustained effect on mitochondrial respiration and thus contribute to increased cholesterol efflux and reduced atherosclerotic lesion area in vivo. These data support the concept that improving mitochondrial function in the vessel wall could be a novel therapeutic approach for reducing atherosclerosis.

PGC-1α is a well-characterized master regulator of mitochondrial biogenesis and metabolism, and promotes fatty acid oxidation, gluconeogenesis and browning of white adipose tissue33, 34. In macrophages, anti-miR33 therapy increases PGC-1α expression, which in turn increases cellular respiration and ATP production- two fundamental roles of the mitochondria. Clinical evidence suggests that promoting the activity of PGC-1α has positive outcomes on metrics of the metabolic syndrome, including insulin resistance and obesity35, 36. However until recently, PGC-1α activity in macrophages had not been well studied, and its role in atherogenesis was unknown. To investigate the function of PGC-1α in atherogenesis, McCarthy et al recently reported that deletion of Pgc-1α in hematopoietic cells accelerated the progression of atherosclerosis in Ldlr-/- mice37. Our current study supports the idea that increasing PGC-1α expression has salutary effects on the development of atherosclerosis, as mice treated with anti-miR33 oligonucleotides had increased vascular Pgc-1α protein expression, increased foam cell expression of Pgc-1α transcripts, and a concomitant decrease in aortic lesion area. Indeed in human atherosclerotic lesions, the expression of PGC-1 α in macrophages was inversely correlated to the severity of disease, and over-expression of macrophage PGC-1 α in vitro reduced foam cell formation37. In agreement with these observations, we show that in arteries from patients suffering from atherosclerosis, PGC-1α levels are significantly downregulated compared to healthy arteries, and there is a concomitant increase in miR-33a/b expression. Similarly, the downstream target of PGC-1α, NRF1, is also decreased in atherosclerotic lesions, suggesting an impairment of mitochondrial biogenesis in atherosclerosis. Thus, augmenting PGC-1α expression and function in macrophages could have protective effects on the progression of atherosclerosis in both mice and humans.

In addition to PGC-1α, this study also identifies a number of novel mitochondrial targets of miR-33, and confirms that PDK4 and SLC25A25 are indeed directly repressed by miR-33 via binding to the 3′UTR. In keeping with its role as a regulator of energy homeostasis, these novel targets PDK4 and SLC25A25 have known roles in regulating mitochondrial energy supply. PDK4 phosphorylates and deactivates the pyruvate dehydrogenase complex, switching the primary ATP-generating oxidative reactions in the mitochondria from relying on glucose (via glycolysis) to fatty acid, and the expression of PDK4 pushes the cell's energy supply from relying on the energy-poor glycolytic to the energy-rich fatty acid β-oxidation. SLC25A25 (and its related gene, SLC25A23, also a target of miR-33) is a member of the SLC25 mitochondrial solute carrier protein family20. Both SLC25A25 and SLC25A23 are nuclear encoded ATP-Mg2+/Pi carriers that drive the reversible exchange of substrates such as ATP-Mg2+, Pi, ATP and ADP across the mitochondrial inner membrane, tightly regulating mitochondrial metabolic pathways20. However, like PDK4, the exact role of SLC25A25 in macrophages is unclear. Here, we present the first evidence of that miR-33 controls PDK4 and SLC25A25 in macrophages, which may potentially regulate mitochondrial respiration, ATP production and cholesterol efflux. In agreement with its known ability to regulate fatty acid oxidation genes our data suggests that anti-miR33 therapy targets additional macrophage energy metabolism pathways to re-direct energy utilization and production to fatty acid oxidation pathways, contributing to the increased energy demand during the energy-intensive process of macrophage cholesterol efflux. While it is difficult to decipher the specific contribution of any one of the miR-33 target genes on mitochondrial output and cholesterol efflux, miRNAs are designed to work on entire pathways and thus it is likely not due to a single target gene isolation. Indeed, our data suggests that inhibition of endogenous miR-33 is able to augment mitochondrial activity via direct de-repression of target genes (PGC-1α, PDK4, SLC25A25) and indirect activation of other key mitochondrial genes (i.e. NRF1, OXPHOS), cumulatively resulting in increased ATP production and, acting in concert with the central miR-33 target gene ABCA1, enhancing overall cholesterol efflux.

Despite the excitement for HDL-C raising agents as a means to reduce atherosclerosis, their clinical utility to treat vascular disease has been called into question. On the other hand, agents that promote cholesterol efflux in lesional macrophages are believed to hold tremendous promise, yet no therapies exist that can specifically enhance macrophage efflux and RCT. We and others have previously shown that miR-33 controls the expression of ABCA1, the terminal step in cholesterol efflux to apoA1 and the formation of HDL, making the inhibition of miR-33 an exciting therapeutic strategy16, 17, 27, 38. Indeed, in most pre-clinical models, blocking miR-33 activity results in a decrease in lesion size, with an accompanying increase in HDL-C14, 15. We show that anti-miR33 treatment of atherosclerotic Apoe-/- mice can reduce atherosclerotic lesion burden even in the absence of increased HDL-C. This corresponds to decreased lipid content in lesions, and increased expression of the potent mitochondrial regulators Pgc-1α and Pdk4. In plaque macrophages from similarly-treated Ldlr-/- mice, anti-miR33 also results in de-repression of mitochondrial genes and a very robust increase in Nrf1 expression, indicating that anti-miR33 promotes mitochondrial biogenesis in vivo in the plaque. The regulation of mitochondrial energy metabolic pathways corresponds to an increase in cellular respiration and ATP production in vitro, and a trend toward increases aortic ATP production in vivo, all of which work in concert with the known roles of anti-miR33 in promoting macrophage cholesterol efflux via de-repression of ABCA1. Despite recent evidence suggests that long-term inhibition of miR-33 leads to increases in VLDL cholesterol and can promote obesity, we did not observe any changes in apoB-containing lipoproteins (VLDL and LDL) or body weight with anti-miR33 treatment compared to controls in this 8-week study39, 40. Thus, in the setting of short-term therapeutic inhibition, anti-miR33 can promote overall macrophage mitochondrial metabolism to enhance efflux pathways, protecting from the development of atherosclerosis. This occurs independently from any effects on circulating lipoproteins, and could have important therapeutic value for the treatment of cardiovascular disease. In the clinical setting, a short-term regimen of anti-miR33 treatment in combination with other established lipid-lowering agents (i.e. statins) could promote significant cholesterol removal from advanced lesions, promoting the stabilization and regression of atherosclerotic disease. This work highlights the potential of miRNA-based therapeutics as agents that promote cholesterol efflux via targeting novel mitochondrial genetic networks to reduce atherosclerotic lesion burden.

Novelty and Significance

What Is Known?

  • microRNA-33 (miR-33) has been identified as a post-transcriptional regulator of cholesterol efflux and circulating high-density lipoprotein (HDL) levels in mice and non-human primates.
  • Although miR-33 antagonism alters HDL levels, it is becoming increasingly evident that miR-33 regulates a number of genes that control cellular metabolism.
  • The complete repertoire of miR-33 target genes is unknown.

What New Information Does This Article Contribute?

  • miR-33 controls a network of mitochondrial genes and coordinates mitochondrial energy metabolism.
  • To promote cholesterol efflux, anti-miR33 augments ATP production via de-repression of mitochondrial target genes and enhanced mitochondrial respiration.
  • In plaques from patients with atherosclerosis, miR-33a/b levels are elevated in comparison with normal arteries, whereas mitochondrial metabolic genes are downregulated, suggesting an impairment of the miR-33/mitochondrial axis in disease

The cholesterol efflux capacity of HDL is an important metric of HDL function, associated with the protection from atherosclerosis. Thus, there is an urgent need to improve our understanding of the mechanisms that govern cholesterol efflux and HDL biogenesis. Our study provides insight into how miR-33 regulates cholesterol efflux capacity beyond its previously-described role in targeting ABCA1. We found that miR-33 controls energy metabolism via repression of key mitochondrial genes to limit ATP production and dampen cholesterol efflux, and therefore inhibiting miR-33 enhances mitochondrial respiration, ATP production and cholesterol efflux, which is largely dependent upon the miR-33 target gene PGC-1α. In vivo, miR-33 inhibitors reduced atherosclerotic lesion burden independently of HDL cholesterol levels, an effect that was associated with increased mitochondrial gene expression within atherosclerotic plaques. Importantly, the expression of key mitochondrial metabolic genes is downregulated in human atherosclerotic plaques, while levels of miR-33a/b are elevated. Therefore, it appears that miR-33 antagonism exerts its anti-atherogenic effects directly in macrophages by coordinating a network of metabolic processes that boosts ATP production to feed the ATP-dependent cholesterol efflux, underscoring a novel therapeutic avenue for promoting cholesterol efflux and reducing atherosclerosis beyond raising HDL cholesterol.

Supplementary Material

305624DR2 Online Data Supplement

Acknowledgments

We would like to acknowledge Regulus Therapeutics for providing the control anti-miR and anti-miR33 oligonucleotides; the Animal Care and Veterinary Services at the University of Ottawa Heart Institute for their expertise; Vivian Franklin and Chantal Gaudet for technical assistance; and Majid Nikay for assistance with bioinformatic pathway analysis.

Sources Of Funding: This work was supported by the Canadian Institutes for Health Research (CIHR) operating grants MOP130365 and OCN126572 (KJR). D.K. was supported by an Endowed Cardiovascular Genetics Postdoctoral Fellowship from the University of Ottawa Heart Institute.The BiKE study received support from the Swedish Heart and Lung Foundation, the Swedish Research Council (K2009-65X-2233-01-3, K2013-65X-06816-30-4 and 349-2007-8703), Uppdrag Besegra Stroke (P581/2011-123), the Strategic Cardiovascular Programs of Karolinska Institutet and Stockholm County Council, the Stockholm County Council (ALF2011-0260 and ALF-2011-0279).

Nonstandard Abbreviations and Acronyms

miRNA
microRNA
HD
high-density lipoprotein
ATP
adenosine triphosphate
Apoe
apolipoprotein E
ABCA1
ATP-binding cassette transporter 1
RCT
reverse cholesterol transport

Footnotes

Disclosures: The authors declare no conflicts of interest.

Author Contributions: DK, ER, MN, LR and PS performed the in vitro protein and RNA analyses. DK performed all other functional in vitro assays. DK, MG and KJR performed the animal studies. ABT performed the Seahorse analysis. RS performed the CARS microscopy. MO performed LCM expression analysis. LP, LM and UH analyzed human atherosclerosis samples from the BiKE Biobank. MEH, JPP, and KJM provided critical input into design and interpretation of experiments. DK and KJR wrote the manuscript.

References

1. Moore KJ, Tabas I. Macrophages in the pathogenesis of atherosclerosis. Cell. 2011;145:341–55. [PMC free article] [PubMed]
2. Rader DJ, Tall AR. The not-so-simple HDL story: Is it time to revise the HDL cholesterol hypothesis? Nat Med. 2012;18:1344–6. [PubMed]
3. Rosenson RS, Brewer HB, Jr, Davidson WS, Fayad ZA, Fuster V, Goldstein J, Hellerstein M, Jiang XC, Phillips MC, Rader DJ, Remaley AT, Rothblat GH, Tall AR, Yvan-Charvet L. Cholesterol efflux and atheroprotection: advancing the concept of reverse cholesterol transport. Circulation. 2012;125:1905–19. [PMC free article] [PubMed]
4. Rohatgi A, Khera A, Berry JD, Givens EG, Ayers CR, Wedin KE, Neeland IJ, Yuhanna IS, Rader DR, de Lemos JA, Shaul PW. HDL Cholesterol Efflux Capacity and Incident Cardiovascular Events. N Engl J Med. 2014 [PMC free article] [PubMed]
5. Allen AM, Taylor JM, Graham A. Mitochondrial (dys)function and regulation of macrophage cholesterol efflux. Clin Sci (Lond) 2013;124:509–15. [PubMed]
6. Nunnari J, Suomalainen A. Mitochondria: in sickness and in health. Cell. 2012;148:1145–59. [PubMed]
7. Allen AM, Graham A. Mitochondrial function is involved in regulation of cholesterol efflux to apolipoprotein (apo)A-I from murine RAW 264.7 macrophages. Lipids in health and disease. 2012;11:169. [PMC free article] [PubMed]
8. Vats D, Mukundan L, Odegaard JI, Zhang L, Smith KL, Morel CR, Wagner RA, Greaves DR, Murray PJ, Chawla A. Oxidative metabolism and PGC-1beta attenuate macrophage-mediated inflammation. Cell Metab. 2006;4:13–24. [PMC free article] [PubMed]
9. van Rooij E, Olson EN. MicroRNA therapeutics for cardiovascular disease: opportunities and obstacles. Nature reviews Drug discovery. 2012;11:860–72. [PubMed]
10. Thum T. MicroRNA therapeutics in cardiovascular medicine. EMBO molecular medicine. 2012;4:3–14. [PMC free article] [PubMed]
11. Rayner KJ, Moore KJ. MicroRNA control of high-density lipoprotein metabolism and function. Circulation research. 2014;114:183–92. [PMC free article] [PubMed]
12. Nguyen MA, Karunakaran D, Rayner KJ. Unlocking the door to new therapies in cardiovascular disease: microRNAs hold the key. Current cardiology reports. 2014;16:539. [PubMed]
13. Allen RM, Marquart TJ, Albert CJ, Suchy FJ, Wang DQ, Ananthanarayanan M, Ford DA, Baldan A. miR-33 controls the expression of biliary transporters, and mediates statin- and diet-induced hepatotoxicity. EMBO Mol Med. 2012;4:882–95. [PMC free article] [PubMed]
14. Rayner KJ, Sheedy FJ, Esau CC, Hussain FN, Temel RE, Parathath S, van Gils JM, Rayner AJ, Chang AN, Suarez Y, Fernandez-Hernando C, Fisher EA, Moore KJ. Antagonism of miR-33 in mice promotes reverse cholesterol transport and regression of atherosclerosis. Journal of Clinical Investigation. 2011;121:2921–2931. [PMC free article] [PubMed]
15. Horie T, Baba O, Kuwabara Y, Chujo Y, Watanabe S, Kinoshita M, Horiguchi M, Nakamura T, Chonabayashi K, Hishizawa M, Hasegawa K, Kume N, Yokode M, Kita T, Kimura T, Ono K. MicroRNA-33 deficiency reduces the progression of atherosclerotic plaque in ApoE-/- mice. J Am Heart Assoc. 2012;1:e003376. [PMC free article] [PubMed]
16. Najafi-Shoushtari SH, Kristo F, Li Y, Shioda T, Cohen DE, Gerszten RE, Naar AM. MicroRNA-33 and the SREBP host genes cooperate to control cholesterol homeostasis. Science. 2010;328:1566–9. [PMC free article] [PubMed]
17. Rayner KJ, Suarez Y, Davalos A, Parathath S, Fitzgerald ML, Tamehiro N, Fisher EA, Moore KJ, Fernandez-Hernando C. MiR-33 contributes to the regulation of cholesterol homeostasis. Science. 2010;328:1570–3. [PMC free article] [PubMed]
18. Davalos A, Goedeke L, Smibert P, Ramirez CM, Warrier NP, Andreo U, Cirera-Salinas D, Rayner K, Suresh U, Pastor-Pareja JC, Esplugues E, Fisher EA, Penalva LO, Moore KJ, Suarez Y, Lai EC, Fernandez-Hernando C. miR-33a/b contribute to the regulation of fatty acid metabolism and insulin signaling. Proc Natl Acad Sci U S A. 2011;108:9232–7. [PubMed]
19. Wang N, Silver DL, Thiele C, Tall AR. ATP-binding cassette transporter A1 (ABCA1) functions as a cholesterol efflux regulatory protein. J Biol Chem. 2001;276:23742–7. [PubMed]
20. Anunciado-Koza RP, Zhang J, Ukropec J, Bajpeyi S, Koza RA, Rogers RC, Cefalu WT, Mynatt RL, Kozak LP. Inactivation of the mitochondrial carrier SLC25A25 (ATP-Mg2+/Pi transporter) reduces physical endurance and metabolic efficiency in mice. J Biol Chem. 2011;286:11659–71. [PMC free article] [PubMed]
21. Finck BN, Kelly DP. PGC-1 coactivators: inducible regulators of energy metabolism in health and disease. J Clin Invest. 2006;116:615–22. [PMC free article] [PubMed]
22. Majer M, Popov KM, Harris RA, Bogardus C, Prochazka M. Insulin downregulates pyruvate dehydrogenase kinase (PDK) mRNA: potential mechanism contributing to increased lipid oxidation in insulin-resistant subjects. Molecular genetics and metabolism. 1998;65:181–6. [PubMed]
23. Ramirez CM, Goedeke L, Rotllan N, Yoon JH, Cirera-Salinas D, Mattison JA, Suarez Y, de Cabo R, Gorospe M, Fernandez-Hernando C. MicroRNA 33 regulates glucose metabolism. Mol Cell Biol. 2013;33:2891–902. [PMC free article] [PubMed]
24. Rotllan N, Ramirez CM, Aryal B, Esau CC, Fernandez-Hernando C. Therapeutic silencing of microRNA-33 inhibits the progression of atherosclerosis in Ldlr-/- mice--brief report. Arterioscler Thromb Vasc Biol. 2013;33:1973–7. [PMC free article] [PubMed]
25. Rayner KJ, Esau CC, Hussain FN, McDaniel AL, Marshall SM, van Gils JM, Ray TD, Sheedy FJ, Goedeke L, Liu X, Khatsenko OG, Kaimal V, Lees CJ, Fernandez-Hernando C, Fisher EA, Temel RE, Moore KJ. Inhibition of miR-33a/b in non-human primates raises plasma HDL and lowers VLDL triglycerides. Nature. 2011;478:404–7. [PMC free article] [PubMed]
26. Li AC, Binder CJ, Gutierrez A, Brown KK, Plotkin CR, Pattison JW, Valledor AF, Davis RA, Willson TM, Witztum JL, Palinski W, Glass CK. Differential inhibition of macrophage foam-cell formation and atherosclerosis in mice by PPARalpha, beta/delta, and gamma. J Clin Invest. 2004;114:1564–76. [PMC free article] [PubMed]
27. Gerin I, Clerbaux LA, Haumont O, Lanthier N, Das AK, Burant CF, Leclercq IA, MacDougald OA, Bommer GT. Expression of miR-33 from an SREBP2 intron inhibits cholesterol export and fatty acid oxidation. J Biol Chem. 2010;285:33652–61. [PMC free article] [PubMed]
28. Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, Mootha V, Troy A, Cinti S, Lowell B, Scarpulla RC, Spiegelman BM. Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell. 1999;98:115–24. [PubMed]
29. Higgins CF. ABC transporters: physiology, structure and mechanism--an overview. Research in microbiology. 2001;152:205–10. [PubMed]
30. Lee JY, Karwatsky J, Ma L, Zha X. ABCA1 increases extracellular ATP to mediate cholesterol efflux to ApoA-I. American journal of physiology Cell physiology. 2011;301:C886–94. [PubMed]
31. Fernandez-Hernando C, Suarez Y, Rayner KJ, Moore KJ. MicroRNAs in lipid metabolism. Current Opinion in Lipidology. 2011;22:86–92. [PMC free article] [PubMed]
32. Yu E, Calvert PA, Mercer JR, Harrison J, Baker L, Figg NL, Kumar S, Wang JC, Hurst LA, Obaid DR, Logan A, West NE, Clarke MC, Vidal-Puig A, Murphy MP, Bennett MR. Mitochondrial DNA damage can promote atherosclerosis independently of reactive oxygen species through effects on smooth muscle cells and monocytes and correlates with higher-risk plaques in humans. Circulation. 2013;128:702–12. [PubMed]
33. Handschin C, Spiegelman BM. Peroxisome proliferator-activated receptor gamma coactivator 1 coactivators, energy homeostasis, and metabolism. Endocrine reviews. 2006;27:728–35. [PubMed]
34. Handschin C, Spiegelman BM. The role of exercise and PGC1alpha in inflammation and chronic disease. Nature. 2008;454:463–9. [PMC free article] [PubMed]
35. Kleiner S, Mepani RJ, Laznik D, Ye L, Jurczak MJ, Jornayvaz FR, Estall JL, Chatterjee Bhowmick D, Shulman GI, Spiegelman BM. Development of insulin resistance in mice lacking PGC-1alpha in adipose tissues. Proc Natl Acad Sci U S A. 2012;109:9635–40. [PubMed]
36. Ling C, Del Guerra S, Lupi R, Ronn T, Granhall C, Luthman H, Masiello P, Marchetti P, Groop L, Del Prato S. Epigenetic regulation of PPARGC1A in human type 2 diabetic islets and effect on insulin secretion. Diabetologia. 2008;51:615–22. [PMC free article] [PubMed]
37. McCarthy C, Lieggi NT, Barry D, Mooney D, de Gaetano M, James WG, McClelland S, Barry MC, Escoubet-Lozach L, Li AC, Glass CK, Fitzgerald DJ, Belton O. Macrophage PPAR gamma Co-activator-1 alpha participates in repressing foam cell formation and atherosclerosis in response to conjugated linoleic acid. EMBO Mol Med. 2013;5:1443–57. [PMC free article] [PubMed]
38. Marquart TJ, Allen RM, Ory DS, Baldan A. miR-33 links SREBP-2 induction to repression of sterol transporters. Proc Natl Acad Sci U S A. 2010;107:12228–32. [PubMed]
39. Goedeke L, Salerno A, Ramirez CM, Guo L, Allen RM, Yin X, Langley SR, Esau C, Wanschel A, Fisher EA, Suarez Y, Baldan A, Mayr M, Fernandez-Hernando C. Long-term therapeutic silencing of miR-33 increases circulating triglyceride levels and hepatic lipid accumulation in mice. EMBO Mol Med. 2014;6:1133–41. [PMC free article] [PubMed]
40. Horie T, Nishino T, Baba O, Kuwabara Y, Nakao T, Nishiga M, Usami S, Izuhara M, Nakazeki F, Ide Y, Koyama S, Sowa N, Yahagi N, Shimano H, Nakamura T, Hasegawa K, Kume N, Yokode M, Kita T, Kimura T, Ono K. MicroRNA-33b knock-in mice for an intron of sterol regulatory element-binding factor 1 (Srebf1) exhibit reduced HDL-C in vivo. Scientific reports. 2014;4:5312. [PMC free article] [PubMed]
41. Jeoung NH, Wu P, Joshi MA, Jaskiewicz J, Bock CB, Depaoli-Roach AA, Harris RA. Role of pyruvate dehydrogenase kinase isoenzyme 4 (PDHK4) in glucose homoeostasis during starvation. Biochem J. 2006;397:417–25. [PubMed]
42. Venegas V, Wang J, Dimmock D, Wong LJ. Real-Time Quantitative PCR Analysis of Mitochondrial DNA Content Current Protocols in Human Genetics. John Wiley & Sons, Inc.; 2001. [PubMed]
43. Pezacki JP, Blake JA, Danielson DC, Kennedy DC, Lyn RK, Singaravelu R. Chemical contrast for imaging living systems: molecular vibrations drive CARS microscopy. Nature chemical biology. 2011;7:137–45. [PubMed]
44. Pegoraro AF, Ridsdale A, Moffatt DJ, Jia Y, Pezacki JP, Stolow A. Optimally chirped multimodal CARS microscopy based on a single Ti:sapphire oscillator. Optics express. 2009;17:2984–96. [PubMed]
45. Razuvaev A, Ekstrand J, Folkersen L, Agardh H, Markus D, Swedenborg J, Hansson GK, Gabrielsen A, Paulsson-Berne G, Roy J, Hedin U. Correlations between clinical variables and gene-expression profiles in carotid plaque instability. Eur J Vasc Endovasc Surg. 2011;42:722–30. [PubMed]
46. Perisic L, Hedin E, Razuvaev A, Lengquist M, Osterholm C, Folkersen L, Gillgren P, Paulsson-Berne G, Ponten F, Odeberg J, Hedin U. Profiling of atherosclerotic lesions by gene and tissue microarrays reveals PCSK6 as a novel protease in unstable carotid atherosclerosis. Arterioscler Thromb Vasc Biol. 2013;33:2432–43. [PubMed]