The Scavenger Receptor Class B, Type I Is a Primary Determinant of Paraoxonase-1 Association With High-Density Lipoproteins
Objective—To examine the contribution of the scavenger receptor (SR) BI to the mechanism by which high-density lipoprotein (HDL) acquires paraoxonase-1 (PON1).
Methods and Results—Serum PON1 activity contributes to the antioxidant capacity of HDLs and is suggested to be an independent risk factor for atherosclerosis. The association of PON1 with HDL is a major determinant of its serum activity levels. PON1 secretion was studied in stably transfected Chinese hamster ovary and HepG2 models. Complementary analyses were performed in transgenic models. Modulation of SR-BI expression, by SR-BI small and interfering RNA knockdown and pharmacologically, correlated with significant changes (P<0.01) in PON1 secretion to HDLs and very-low-density lipoproteins. Block lipid transport-1 (BLT1), which increases the affinity of HDL for SR-BI without modulating its expression, was associated with significant increases in secretion. Downregulating postsynaptic density 95/disc-large/zona occludens kinase in HepG2 reduced cell SR-BI protein and lowered enzyme secretion. Serum PON1 activity was significantly reduced in postsynaptic density 95/disc-large/zona occludens kinase knockout mice.
Conclusion—The present study identifies SR-BI as a major determinant of the capacity of HDL to acquire PON1. It reinforces the concept of the receptor as a docking molecule, allowing communication between HDL and the cell, and extends the importance of SR-BI to HDL metabolism and function.
Accumulating evidence suggests a protective role for the high-density lipoprotein (HDL) bound enzyme paraoxonase-1 (PON1) in atherosclerosis, which is postulated to arise from its impact on oxidative stress. Reduced PON1 expression in animal models is associated with increased formation of atherosclerotic lesions, increased oxidative stress, and reduced antioxidant capacity of HDL.1,2 The latter can be restored by adding purified PON11,2 to the lipoprotein. Recent prospective studies3,4 have provided persuasive arguments for PON1 also being an independent cardiovascular risk factor in humans. Notably, Bhattacharyya et al3 showed serum PON1 activity to be negatively correlated with risk and with a panel of serum-oxidized lipids and derivatives, supporting proposals of links with oxidative stress. These data imply that factors influencing serum PON1 activity will affect the risk of atherosclerosis.
An important determinant of PON1 serum activity is HDL. It is the serum transport vector for PON1, concomitantly stabilizing and sustaining enzyme activity.5,6 Conversely, PON1 makes an important contribution to the antioxidant capacity of the lipoprotein, as suggested by animal studies. In humans, reduced serum PON1 activity is correlated to increased serum levels of oxidized lipids3 and may be 1 determinant of the recently described phenomenon of dysfunctional HDL.7 In this context, how HDL acquires PON1 should be a major contributing factor to serum activity levels of the enzyme. Recently, a model was proposed, whereby HDL accelerates removal of PON1 from the plasma membrane of cells, principally hepatocytes, which are the primary source of the enzyme.5 By using this model, researchers envisaged transient anchorage of the lipoprotein to the cell membrane and speculated on a role for the HDL scavenger receptor (SR) BI. We provide evidence suggesting that SR-BI is the primary determinant of the acquisition of PON1 by HDL. This finding has implications for the role that the receptor may play in defining serum activity of PON1 and, thus, the antioxidant capacity of HDL and further underlines the importance of SR-BI to HDL metabolism and function.
Cell Culture Models
Three cell culture models were exploited to analyze the relationship between SR-BI and PON1 secretion. A Chinese hamster ovary (CHO) model for PON1 secretion was previously described in detail.5 The cell line has a high PON1 secretory capacity and is referred to as CHO-hPON1hi. A second CHO-hPON1 line was prepared by the same procedure but secreted PON1 at a much lower capacity (3- to 5-fold). This line was labeled CHO-hPON1lo. The third cell culture model (HepG2-hPON1) was developed with the human hepatocyte line, HepG2, using the same transfection procedure and the same human PON1 construct.
PON1 secretion from cells was determined by analysis of arylesterase (ARE) activity.8 This is also used as a surrogate for PON1 peptide levels.9 We confirmed the association by showing a strong correlation (r=0.98) between PON1 activity and PON1 peptide (by immunoblot) in CHO-hPON1hi–conditioned medium (supplemental Figure I; available online at http://atvb.ahajournals.org).
Downregulation of Gene Expression
SR-BI in CHO and HepG2 cells and postsynaptic density 95/disc-large/zona occludens kinase (PDZK) 1 in HepG2 cells were downregulated using small and interfering RNA (siRNA). The siRNA sequences were established using computer software (BLOCK-iT designer; Invitrogen) and are given in the supplemental Table.
Gene expression was determined by analysis of mRNA levels using real-time PCR and a commercially available system (LightCycler; Roche, Basel, Switzerland), according to the manufacturer’s instructions. The mRNA levels were standardized to those of TATA box-binding protein.
The mice for the present study were raised and genotyped in the animal facility of Hannover Medical School, Hannover, Germany, exactly as described,10 except that they were bred on an FVB/N background. Heterozygous parents and littermates were used for this study. Blood was collected as previously described11 by intra-arterial cannulation after isoflurane anesthesia and immediately spun for serum collection. Serum was stored at −20°C until used.
Additional information on methods and materials is available in the supplemental data.
SR-BI siRNA Downregulates PON1 Secretion in CHO-hPON1 Cells
Transfection of CHO-hPONhi cells with SR-BI siRNA caused a 70% reduction in SR-BI mRNA expression (Figure 1A) and, correspondingly, a reduction in the expression of SR-BI protein (Figure 1B) compared with that observed in cells treated with scramble siRNA. It was accompanied by a highly significant reduction in specific binding of iodine 125–HDL (mean HDL binding [mean±SD], 54.2±2.0% of scramble-treated cells; n=6; P<0.001) to the cell membrane (Figure 1C). There was also a highly significant reduction in secretion of PON1. This is illustrated by the dose-response curve for enzyme secretion promoted by HDL (Figure 1D), with a mean 57.3±3.7% reduction (n=18, P<0.001) over the concentration range studied. The scrambled siRNA had no impact on PON1 secretion (Figure 1D). Because the CHO-hPON1hi clone is a strong PON1 secretor, we determined the effects of SR-BI siRNA in a clone with lower levels of enzyme secretion. A comparable reduction in SR-BI mRNA (Figure 1A) and peptide (supplemental Figure II) was observed with the CHO-hPON1lo clone. Correspondingly, secretion of PON1 activity (Figure 1E) and peptide (data not shown) was reduced in a highly significant manner (P<0.001 for all comparisons) using HDL (5 μg/mL, −69.7±4.8%; 20 μg/mL, −69.1±4.1%; n=9 for both) to promote secretion.
In a previous study,12 it was established that very-low-density lipoprotein (VLDL) was also able to promote PON1 secretion. When VLDL-stimulated secretion was analyzed in SR-BI siRNA–treated CHO-hPON1lo cells, there was a significant reduction in enzyme activity (P<0.001, Figure 1E) and PON1 peptide (data not shown). The analysis was extended to whole medium (DMEM plus FCS, 10% vol/vol) using CHO-hPON1hi cells to follow secretion. Again, a highly significant reduction in PON1 secretion after treatment of cells with SR-BI siRNA was observed (Figure 1E).
The reduced secretion of PON1 was not because of interference with its synthesis by siRNA treatment. As shown in Figure 2A, there were no significant differences in the PON1 content of whole cell extracts of SR-BI siRNA and scramble-treated CHO-hPON1 (relative staining intensities of PON1/GAPDH: scramble, 0.90±0.10; and siRNA, 0.89±0.11; n=3). We also examined a possible involvement of SR-BI in transfer of PON1 to the cell surface. Figure 2C gives the results for external cell membrane expression of PON1 protein detected immunochemically. For SR-BI siRNA–treated cells, there was a slightly, but significantly, higher level of PON1 present in the cell membrane.
Impact of Variations in SR-BI Expression on PON1 Release From CHO-hPON1 Cells
To confirm the influence of SR-BI on PON1 secretion, the coding region of human PON1 was transiently transfected into CHO cells stably expressing low or high levels of SR-BI (corresponding to CHO lines low density lipoprotein (ldl)A and ldlA(mSR-BI), respectively; data from Acton et al13). After correcting for transfection efficiency, PON1 activity in conditioned medium was significantly greater (P<0.001, Figure 3A) in conditioned medium from CHO cells expressing higher levels of SR-BI.
To determine if variations in SR-BI expression of a more physiological nature could influence PON1 metabolism, we examined the effect of cAMP14 on release of the enzyme from CHO-hPON1 cells. As shown in Figure 3B, preincubation of cells with forskolin to stimulate adenylate cyclase activity and increase cAMP concentrations was associated with a highly significant (P<0.001) increase in the release of PON1 to HDL. This was also accompanied by an increase in the expression of SR-BI peptide (Figure 3C).
Impact of Variations in SR-BI Expression on PON1 Release From HepG2-hPON1 Cells
We examined several different sources of hepatocyte for their PON1 secretory capacity, but neither transformed cells lines HepG2 or human hepatoma (Huh)-7 nor primary human or rat hepatocyte cultures released sufficient PON1 over a 24-hour period for valid analyses by enzyme activity or immunoblot. Consequently, we developed a model in which HepG2 cells were stably transfected with the coding region of the human PON1 gene (HepG2-hPON1), using the same construct that was used to develop the CHO-hPON1 model.5
Treatment of the HepG2-hPON1 cells with SR-BI siRNA caused a 73% reduction in the level of SR-BI mRNA (Figure 4A) and a significant reduction in the expression of SR-BI peptide (Figure 4B). Correspondingly, there was a highly significant reduction in PON1 release from siRNA- compared with scramble-treated cells (Figure 4C), whether using HDL or whole medium (P<0.001 for both) to promote secretion. Differences between these media for PON1 secretion were less marked than for CHO cells, which may reflect the more extensive impact of hepatocytes on lipoproteins and their metabolism. The treatments did not modify the expression of PON1 protein (relative staining intensities of PON1/GAPDH: scramble, 0.38±0.07; and siRNA, 0.37±0.09; n=3) (Figure 2B, bottom).
Increasing cAMP levels by treating HepG2-hPON1 cells with forskolin occasioned a significantly increased release of PON1 into conditioned medium (13.1±5.5%) (Figure 3B). This was associated with significantly higher expression of SR-BI protein in forskolin-treated cells (Figure 3C).
BLT-1 Increases PON1 Secretion
Several studies have shown that BLT-1 modulates SR-BI activity by blocking cholesterol exchange between the cell and HDL. SR-BI is the specific target of BLT-1.15 BLT-1 does not prevent HDL binding to SR-BI but increases the affinity of the lipoprotein for the receptor15 (supplemental Figure III). When HepG2-hPON1 cells were preincubated with BLT-1, 0.5 μmol/L, PON1 secretion was significantly increased compared with nontreated cells, whether using HDL, VLDL, or whole medium as the PON1 acceptor (P<0.001 for all) (Figure 4D). BLT-1 did not modify PON1 activity when added to serum or isolated HDL containing active PON1 (Sara Deakin, PhD, unpublished data, 2009). Also, there were no significant changes in the expression of SR-BI peptide observed by immunoblot analyses of whole cell extracts after BLT-1 treatment (Figure 4D). Corresponding studies with BLT-1 and CHO-hPON1hi cells gave comparable results, with highly significant increases in PON1 secretion to HDL, 20 μg/mL (3-hour incubation) (ARE activity in nontreated versus treated cells: 4.2±0.4 versus 5.7±0.4 optical density at 270 per milliliter/milligram of protein; P<0.01).
Our attempts to perform complementary studies using anti–SR-BI–blocking antibodies to prevent HDL binding were hampered by the use of rabbit serum. Such antibodies are only available as whole rabbit antiserum; rabbit serum is the richest known source of PON activity.
PDZK1 siRNA Reduces PON1 Secretion From HepG2 Cells
PDZK1 is implicated in the metabolism of hepatic SR-BI.16 We examined the consequences of downregulating PDZK1 expression on PON1 secretion from HepG2-hPON1 cells. As shown in Figure 5A, PDZK1 siRNA reduced the expression of PDZK1 mRNA by 89% (P<0.005) compared with the scrambled sequence, with a corresponding reduction in the expression of PDZK1 protein (Figure 5D). This was accompanied by a highly significant 56% reduction in PON1 secretion (Figure 5C). Conversely, treatment with PDZK1 siRNA alone was associated with a significant increase in the expression of SR-BI mRNA (Figure 5B). Treatment of HepG2-hPON1 cells with SR-BI siRNA caused significant reductions in the expression of SR-BI mRNA (Figure 5B), as previously demonstrated, and in PON1 secretion (−64%, Figure 5C). However, the expression of PDZK mRNA was significantly increased (Figure 5B).
Serum PON1 in Animal Models
Serum PON1 was analyzed in animal models in which genes implicated in HDL metabolism were lacking. The heterozygous PDZK1-knockout mice were bred and maintained in the laboratory of Dr. Seidler. As shown in the Table, analysis of serum samples from the knockout mice revealed significant reductions (P<0.01) in PON and ARE activities compared with the wild-type littermates. A second analysis examined enzyme activities from mice in which ATP binding cassette subclass G (ABCG1) gene expression was downregulated. The latter is thought to mediate the export of cholesterol to more mature lipid-rich HDL particles that correspond to the type of particle that promotes PON1 secretion.5 These were developed and characterized as previously described.17 ABCG1-knockout mice showed no significant reduction in ARE activity (Table) and a small reduction (P=0.04) in PON activity. In a previous study, Van Eck et al18 showed that in an SR-BI–knockout mouse model, there were significant reductions in both ARE and PON activities in serum samples.
There is growing evidence that the antioxidant/anti-inflammatory capacity of HDL is a feature of its beneficial impact on vascular health and, correspondingly, reductions in this capacity contribute to dysfunctional HDL and increased vascular risk.7 Defining dysfunctions of HDL and identifying mechanisms that can produce them are of particular interest for refining our understanding of the role of the lipoprotein in vascular disease. It has already been demonstrated by several in vitro and in vivo studies that the antiinflammatory capacity of HDL is reduced when the PON1 content of HDL is also reduced, an effect probably linked to the antioxidant potential of PON1.19 Moreover, reductions in the PON1 content of HDL are a contributory factor to dysfunctional HDL,20 whereas serum PON1 activity is suggested to be an independent prospective risk factor for vascular disease in humans.3
We investigated the factors that influence serum concentrations and activities of the enzyme. The present studies tested our hypothesis of a receptor-mediated process for PON1 secretion.5 These studies identify SR-BI as the principal mediator of the ability of HDL to acquire PON1 in the 2 in vitro models of PON1 secretion. In a validated CHO model of secretion, we show that variation in the expression of SR-BI was associated with modulated release of the enzyme to HDL and whole serum (Figure 1). The receptor appears to be a requirement for PON1 secretion in these models because VLDL-mediated stimulation of PON1 secretion, which was previously described,12 was also largely dependent on SR-BI. VLDL is a ligand for SR-BI.21 We then extended the studies to the hepatocyte, the physiological source of PON1 (Figure 4). The observations with the CHO model were replicated in HepG2 cells. PON1 secretion was greatly reduced when SR-BI expression was blocked by siRNA transfection. Less dramatic variations in the expression of the receptor, induced by modulating cAMP concentrations or cell cholesterol content, were also accompanied by reciprocal changes to PON1 secretion. Although the mechanism by which PON1 is transferred to the cell membrane remains to be elucidated, SR-BI does not appear to play a role. Indeed, there was greater expression of PON1 in cell membranes from SR-BI–knockout cells, consistent with the need for the receptor to facilitate secretion of membrane-bound PON1.
To support our hypothesis of the role of SR-BI, we used BLT1. We confirmed that it does not modulate SR-BI expression. However, it does increase the affinity of HDL for the receptor.22 We reasoned that coincubation of BLT1 and HDL with cells would increase secretion of PON1, which was confirmed for both models (Figure 4D). Further support for the hypothesized role of SR-BI was furnished by studies of PDZK1. This scaffolding protein is of major importance for liver SR-BI metabolism and function and for maintaining hepatic concentrations of the receptor.16 Reduced PDZK1 expression was associated with reduced SR-BI protein expression in HepG2 and reduced secretion of PON1 (Figure 5). Incidentally, SR-BI mRNA levels were increased in PDZK1 siRNA-treated cells. We also observed an increase in PDZK1 mRNA when SR-BI expression was downregulated, suggesting regulatory cross talk between the 2 proteins.
Finally, we examined PON1 activity in serum samples from animal models. In PDZK1-knockout mice, there was a significant reduction in serum enzyme activities of PON1, measured as ARE and PON activities, despite increases in serum concentrations of HDL.23 This parallels the observations made with the HepG2 cells, in which PDZK1 downregulation reduced secretion of PON1. Reduced gene expression of PDZK1 is known to reduce expression specifically of hepatic SR-BI peptide,16 which presumably explains the reduction in serum PON1 activity. We also examined PON1 activity in ABCG1-knockout mice. There was no significant reduction in ARE activity and a minor decrease in PON activity. Because the transporter is primarily expressed in macrophages, it suggests that ABCG1 does not have a major role to play in PON1 secretion. Modified expression of ABCG1 does not influence expression of SR-BI. For animal models of SR-BI knockout, in a recent study, Van Eck et al18 reported significantly reduced serum levels of PON1 and significantly higher systemic oxidative stress, despite greatly increased concentrations of HDL. Although suggestive of a role for SR-BI, there were 2 potentially confounding factors of such studies: (1) the considerable modification of HDL structure/composition, which can influence the capacity of HDL to stimulate PON1 secretion5; and (2) the increased oxidative stress observed in the mice. This could be a cause, as much as a consequence, of reduced serum PON1 because the enzyme is particularly sensitive to oxidation.24 Our present studies point to modifications of SR-BI expression as a primary cause of reduced PON1 in the knockout model. The observations are in agreement with other studies showing increased oxidative stress in PON1-deficient animal models1,2 and increased systemic oxidative stress in humans with a low serum PON1 level.3
We propose that SR-BI provides temporary anchorage of HDL to the cell surface, thus facilitating transfer of PON1 to the lipoprotein and allowing more efficient secretion of the enzyme. Increased secretion of the enzyme under conditions (BLT1 coincubation) in which affinity of HDL for its receptor is increased is consistent with this interpretation. Such a process would preclude exposure of the retained hydrophobic signal sequence of PON1 to an aqueous milieu6 during transfer of the enzyme to HDL. Previously, such a role for SR-BI was speculated,5 after showing that PON1 was inserted into the hepatocyte plasma membrane before secretion. For total PON1 secretion, this may not reflect a 1:1 molar relationship between SRBI:HDL because each HDL particle can bind a variable amount of PON1. Random interaction of HDL with the hepatocyte membrane may also allow a certain level of transfer of PON1 to the lipoprotein, but it appears less efficient than SR-BI– mediated transfer. This may explain the presence of PON1 in serum from the SR-BI–knockout mice. In this context, we examined the effect of heparin on PON1 secretion to HDL but observed no reduction in enzyme activity eliminating the cell wall matrix as a major alternative anchorage mechanism for HDL in our models.
The present study has several important implications for SR-BI and HDL metabolism. It reinforces the concept of SR-BI as a transient docking molecule, allowing communication between HDL and the cell. The receptor facilitates bidirectional transfer of cholesterol between the lipoprotein and cells, and it may also allow HDL to acquire certain components important for its function. In a similar context, anchorage of the lipoprotein to the cell membrane via SR-BI may facilitate the interaction of specific HDL components with cells, as suggested for the interaction of sphingosine-1-phosphate, a highly hydrophobic molecule also transported by HDL, with its receptor.25 Despite substantial increases in HDL cholesterol, SR-BI–knockout models are usually associated with increased atherosclerosis risk.26 This has primarily been attributed to less efficient reverse cholesterol transport. After their studies, Van Eck et al18 suggested that reduced PON1 activity may also be a contributory factor to increased risk. Our data provide a mechanistic explanation for reduced serum levels of PON1, indicating that increased oxidative stress in the model is a consequence, rather than a cause, of reduced serum PON1. Finally, the observations extend the importance of the SR to HDL metabolism and function.
Sources of Funding
This study was supported by grant 31-118418 from the Swiss National Research Foundation (Dr James) and grants DFG Se460/9-6, SFB621/C9 (US).
Received on: January 15, 2010; final version accepted on: August 4, 2010.
Shih DM, Xia YR, Wang XP, Miller E, Castellani LW, Subbanagounder G, Cheroutre H, Faull KF, Berliner JA, Witztum JL, Lusis AJ. Combined serum paraoxonase knockout/apolipoprotein E knockout mice exhibit increased lipoprotein oxidation and atherosclerosis. J Biol Chem. 2000; 275: 17527–17535.
Bhattacharyya T, Nicholls SJ, Topol EJ, Zhang R, Yang X, Schmitt D, Fu X, Shao M, Brennan DM, Ellis SG, Brennan ML, Allayee H, Lusis AJ, Hazen SL. Relationship of paraoxonase 1 (PON1) gene polymorphisms and functional activity with systemic oxidative stress and cardiovascular risk. JAMA. 2008; 299: 1265–1276.
Mackness B, Durrington P, McElduff P, Yarnell J, Azam N, Watt M, Mackness M. Low paraoxonase activity predicts coronary events in the Caerphilly Prospective Study. Circulation. 2003; 107: 2775–2779.
Deakin S, Leviev I, Gomaraschi M, Calabresi L, Franceschini G, James RW. Enzymatically active paraoxonase-1 is located at the external membrane of producing cells and released by a high affinity, saturable, desorption mechanism. J Biol Chem. 2002; 277: 4301–4308.
Sorenson RC, Bisgaier CL, Aviram M, Hsu C, Billecke S, La Du BN. Human serum paraoxonase/arylesterase’s retained hydrophobic N-terminal leader sequence associates with HDLs by binding phospholipids: apolipoprotein A-I stabilizes activity. Arterioscler Thromb Vasc Biol. 1999; 19: 2214–2225.
Navab M, Anantharamaiah GM, Reddy ST, Van Lenten BJ, Ansell BJ, Fogelman AM. Mechanisms of disease: proatherogenic HDL: an evolving field. Nature Clin Pract Endocrinol Metab. 2006; 2: 504–511.
Connelly PW, Maguire GF, Picardo CM, Teiber JF, Draganov D. Development of an immunoblot assay with infrared fluorescence to quantify paraoxonase 1 in serum and plasma. J Lipid Res. 2008; 49: 245–250.
Sjöblom M, Singh AK, Zheng W, Wang J, Tuo BG, Krabbenhöft A, Riederer B, Gros G, Seidler U. Duodenal acidity “sensing” but not epithelial HCO3- supply is critically dependent on carbonic anhydrase II expression. Proc Natl Acad Sci. 2009; 106: 13094–13099.
Acton S, Rigotti A, Landschulz KT, Xu S, Hobbs HH, Krieger M. Identification of scavenger receptor SR-BI as a high density lipoprotein receptor. Science. 1996; 271: 518–520.
Out R, Hoekstra M, Meurs I, de Vos P, Kuiper J, Van Eck M, Van Berkel TJ. Total body ABCG1 expression protects against early atherosclerotic lesion development in mice. Arterioscler Thromb Vasc Biol. 2007; 27: 594–599.
Van Eck M, Hoekstra M, Hildebrand RB, Yaong Y, Stengel D, Kruijt JK, Sattler W, Tietge UJ, Ninio E, Van Berkel TJ, Praticò D. Increased oxidative stress in scavenger receptor BI knockout mice with dysfunctional HDL. Arterioscler Thromb Vasc Biol. 2007; 27: 2413–2419.
Van Lenten BJ, Wagner AC, Nayak DP, Hama S, Navab M, Fogelman AM. High-density lipoprotein loses its anti-inflammatory properties during acute influenza A infection. Circulation. 2001; 103: 2283–2288.
Calvo D, Gómez-Coronado D, Lasunción MA, Vega MA. CLA-1 is an 85-kD plasma membrane glycoprotein that acts as a high-affinity receptor for both native (HDL, LDL, and VLDL) and modified (OxLDL and AcLDL) lipoproteins. Arterioscler Thromb Vasc Biol. 1997; 17: 2341–2349.
Nieland TJ, Penman M, Dori L, Krieger M, Kirchhausen T. Discovery of chemical inhibitors of the selective transfer of lipids mediated by the HDL receptor SR-BI. Proc Natl Acad Sci U S A. 2002; 99: 15422–15427.
Kocher O, Yesilaltay A, Cirovic C, Pal R, Rigotti A, Krieger M. Targeted disruption of the PDZK1 gene in mice causes tissue-specific depletion of the high density lipoprotein receptor scavenger receptor class B type I and altered lipoprotein metabolism. J Biol Chem. 2003; 278: 52820–52825.
Tölle M, Pawlak A, Schuchardt M, Kawamura A, Tietge UJ, Lorkowski S, Keul P, Assmann G, Chun J, Levkau B, van der Giet M, Nofer JR. HDL-associated lysosphingolipids inhibit NAD(P)H oxidase-dependent monocyte chemoattractant protein-1 production. Arterioscler Thromb Vasc Biol. 2008; 28: 1542–1548.
Trigatti BL, Krieger M, Rigotti A. Influence of the HDL receptor SR-BI on lipoprotein metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol. 2003; 23: 1732–1738.