15-Lipoxygenase-mediated modification of high-density lipoproteins impairs SR-BI- and ABCA1-dependent cholesterol efflux from macrophages

Biochimica et Biophysica Acta 1761 (2006) 292 – 300 http://www.elsevier.com/locate/bba

15-Lipoxygenase-mediated modification of high-density lipoproteins impairs SR-BI- and ABCA1-dependent cholesterol efflux from macrophages Angela Pirillo a,⁎, Patrizia Uboldi a , Hartmut Kuhn b , Alberico L. Catapano a a

b

Department of Pharmacological Sciences, University of Milan, Milan, Italy Institute of Biochemistry, University Medicine Berlin–Charité, Berlin, Germany

Received 20 September 2005; received in revised form 14 March 2006; accepted 14 March 2006 Available online 7 April 2006

Abstract Elevated plasma levels of high-density lipoprotein cholesterol (HDL-C) are atheroprotective and HDL-dependent reverse cholesterol transport has been related to this effect. HDL particles may, however, undergo modifications that affect their biological activities. Lipoxygenases (LOs) belong to a family of lipid peroxidizing enzymes; among them, reticulocyte-type 15-lipoxygenase (15-LO-1) appears to play a pathophysiological role in atherosclerosis, as its expression is increased in atherosclerotic plaques and it has been shown to oxidize low-density lipoproteins to an atherogenic form. In this work we investigated the impact of in vitro 15-lipoxygenase-catalyzed modification of HDL3 on their ability to act as cholesterol acceptor and found that 15-LO-modified HDL3 were less effective in mediating cholesterol efflux from lipid-laden J774 cells. A reduced binding of 15-LO-modified HDL3 to scavenger receptor class B, type I (SR-BI), due to HDL apoproteins cross-linking, explained, at least in part, the observed reduction of cholesterol efflux. In addition, ATP-binding cassette transporter A1 (ABCA1)-mediated cholesterol efflux was also reduced, as a consequence of pre-β-particles loss after HDL3 modification. These results suggest that 15-lipoxygenase might induce structural alterations of HDL3 particles that impair their capability of triggering reverse cholesterol transport. © 2006 Elsevier B.V. All rights reserved. Keywords: High-density lipoprotein; 15-lipoxygenase; Reverse cholesterol transport; Scavenger receptor class B, type I; ATP-binding cassette transporter A1

1. Introduction The uptake of modified lipoproteins by macrophages in the arterial wall is one of the early processes in atherosclerosis [1]. Lipid accumulation triggers activation of macrophages to produce inflammatory mediators (cytokines, eicosanoids, growth factors) and matrix decomposing metalloproteases [2]. On the other hand, HDL-mediated reverse cholesterol transport (RCT) represents a counteracting process that may slow down the progression of foam cell formation or even induce lesion regression [3–6]. Reverse cholesterol transport involves mainly two membrane proteins: i) scavenger receptor class B, type I (SR-BI) which promotes a bi-directional cholesterol movement by facilitated

⁎ Corresponding author. Tel.: +39 02 50318293; fax: +39 02 50318386. E-mail address: [email protected] (A. Pirillo). 1388-1981/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbalip.2006.03.009

diffusion following the interaction with mature α-HDL particles [7]; ii) ATP-binding cassette transporter A1 (ABCA1), which promotes phospholipid and consequent cholesterol efflux following interaction with lipid-free or lipid-poor apoAI, namely pre-β-HDL [8]. The integrity of both HDL subfractions is a basic requirement for the proper function of HDL. However, HDL is susceptible to oxidation [9–11]. In fact, different types of modification trigger lipoprotein structural alterations, which impact its biological activities. Oxidatively modified HDL particles increase the free cholesterol intracellular content [12] and are cytotoxic to macrophages [13]. Moreover, in most cases, they lose their capability of triggering cholesterol efflux from foam cells [14–18]. Short-term feeding of an atherogenic diet to LDL receptor knockout mice induces a remarkable decrease in circulating HDL apolipoproteins, apoAI and apoAII, with a dramatic increase of higher molecular weight forms of apoAI [19]. Recently, it has been shown that apoAI colocalizes with hypochlorous acidmodified proteins in human atherosclerotic lesions [20], and that

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

the levels of specific markers for HDL oxidation are significantly higher in HDL isolated from subjects with cardiovascular disease than in HDL from healthy donors [20,21]. Moreover, considerable amounts of higher molecular weight forms of apoAI were detected in HDL isolated from human carotid atherosclerotic tissue [20]. These observations raise the possibility that HDL might be oxidized in vivo, in arterial wall microenvironments in which an unbalance between antioxidants and pro-oxidant species is generated. Among the enzymes whose expression is increased in the atherosclerotic lesions, 12/15-lipoxygenase has received attention. Lipoxygenases (LOs) are non-heme iron-containing enzymes which catalyze the insertion of molecular oxygen into polyunsaturated fatty acids forming hydroperoxy derivatives. Human and rabbit 15-LOs share a high degree of homology with mouse and rat 12-LOs [22–24]; both 15-LO and 12-LO induce the formation of similar products and possess similar substrate specificity. 15-LO, which is absent in normal arterial wall, is expressed at high levels in macrophages of human and animal early atherosclerotic lesions [25,26]. Several studies have been published indicating that specific lipoxygenase products can be detected in young developing human atherosclerotic lesions, thus suggesting that 15-lipoxygenase is enzymatically active [27,28]. Advanced human plaques, however, do not express significant amounts of the enzyme [29], suggesting that 15-LO might play a pathophysiological role during the early phases of atherogenesis [27,28]. In vivo, 15-LO overexpression accelerates lesion formation in LDLreceptor deficient mice [30]; on the other hand, the functional inactivation of 15-LO gene [31,32] or the inhibition of its enzymatic activity [33] reduced atherosclerosis in animal models. In vitro, purified 15-LO oxygenates isolated LDL to an atherogenic form [23] and cells overexpressing the enzyme oxidize LDL more efficiently than corresponding controls lacking 15-LO [34]. Macrophage 12/15-LO increases the expression of adhesion molecules in endothelial cells in the presence of LDL, while no activation was observed in the absence of lipoprotein [31]; this finding suggests that the activity of this enzyme is, at least in part, related to its ability to modify LDL. In this study, we investigated the ability of purified rabbit 15-LO to modify isolated HDL3 in an in vitro system and found that HDL represents a suitable LO substrate. 15-LO-treatment induced structural alterations of HDL particles; both SR-BIand ABCA1-mediated cholesterol efflux to 15-LO-modified HDL3 resulted significantly impaired. These results suggest that 15-lipoxygenase might contribute to the generation of dysfunctional HDL, and that, besides its role in LDL oxidation, 15-LO may play a role also in HDL modification. 2. Materials and methods 2.1. Materials The materials used were obtained from the following sources: tissue culture media, media supplements, fetal bovine serum (FBS), essentially fatty acid free albumin (EFAF-BSA), rabbit serum, 22R-OH-cholesterol (22-OH), 9-cis retinoic acid (9cis), DiO (3,3′-dioctadecyloxacarbocyanine perchlorate)

293

from Sigma (St. Louis, MO, USA); PD10 columns, 3H-cholesterol and 3Hcholesteryl oleyl ether from Amersham Biosciences (Uppsala, Sweden); antihuman apoAI antiserum from Dade Behring (Marburg, Germany); anti-human apoAII antiserum from Daiichi Pure Chemicals (Tokyo, Japan); rabbit anti– SR-BI for blocking experiments (extracellular domain), rabbit anti-SR-BI for Western blotting (aa 496–509) and anti-ABCA1 (aa 1201–1211) from Abcam (Cambridge, UK). The native reticulocyte-type 15-LO-1 was prepared as previously described [23].

2.2. Cell culture J774 macrophages were cultured in MEM containing 10% fetal bovine serum. ldlA7 (a low-density lipoprotein receptor deficient CHO cell line) was cultured in Ham's F12 medium containing 5% FBS, 2 mM glutamine, 50 U/ml penicillin, 50 μg/ml streptomycin. Stable transfectants expressing murine SR-BI (ldlA[mSR-BI]) were cultured in the same medium in the presence of 300 μg/ml G418. ldlA7 and ldlA[mSR-BI] cell lines were kindly provided by Prof. Monty Krieger, MIT.

2.3. Isolation and modification of lipoproteins Human LDL (d = 1.019–1.063 g/ml) and HDL3 (d = 1.125–1.21 g/ml) were isolated from healthy volunteers [35]. The protein content was determined by the method of Lowry [36]. LDL acetylation was carried out using the method described by Basu [37]; oxidation of LDL was performed as described [38]. HDL3 modification by 15-LO was performed at 1 mg protein/ml in the presence of 2 μl 15-LO/mg HDL protein at 37 °C for 24 h or 72 h (referred to as 15-LO24h and 15-LO72h). HDL3 was oxidized at 1 mg protein/ml with CuSO4 (5 μM) at 37 °C for 2 h or 24 h (referred to as Cu2h and Cu24h). Oxidation was stopped with 40 μM BHT and 0.01% EDTA. HDL3 oxidation extent was evaluated as the thiobarbituric-acid reactive substances (TBARS) content by a colorimetric method by the assay described by Schuh et al. [39]. Lipids from native and modified HDL3 were extracted under reducing conditions (presence of triphenylphosphine), for the determination of hydroxy PUFA/PUFA ratio (%), as described [23]. For the lipid labeling, native HDL3 was incubated with the fluorescent dye DiO (300 μg DiO/mg HDL protein) for 18 h at 4 °C and then purified on a PD10 column. DiO-labeled HDL3 was modified with 15-LO for 72 h or with CuSO4 for 2 h and 24 h. Here again, the labeled lipoproteins were purified on a PD10 column. Comparable labeling ratios (ng DiO/μg protein) were achieved for native and modified HDL3.

2.4. Cholesterol efflux experiments J774 cells were grown to subconfluence, then incubated for 24 h with MEM containing 0.2% BSA, 3H-cholesterol (1 μCi/ml) and 50 μg/ml AcLDL. After washing, cells were incubated overnight in fresh medium containing 0.2% BSA, then native or modified HDL3 were added for 6 h. The media were collected, centrifuged and aliquots were used for liquid scintillation counting. Cell monolayers were lyzed with 0.1 N NaOH and aliquots were used for protein determination. To study the role of SR-BI, cells were pre-incubated for 45′ with an anti-SR-BI blocking antibody (1:1.000 dilution) before the incubation with lipoproteins; alternatively, cells were pre-incubated with OxLDL (10 μg/ml) for 24 h to reduce SR-BI expression [40], then media were removed, cell were washed twice with PBS and native or modified HDL3 were added as described above. To investigate the role of ABCA1 in cholesterol efflux to 15-LO-modified HDL3, after labeling, J774 were incubated for 24 h in MEM + 0.2% BSA in the presence of 22-OH/9cis (10 μM and 1 μM, respectively) to increase ABCA1 expression. Native and modified HDL3 (10 μg/ml) were added to the cells for 6 h and the 3H-cholesterol efflux was determined as described above. CHO cells (ldlA7 and ldlA[mSR-BI]) were labeled in HAM's F12 medium containing 10% FBS and 1 μCi/ml 3H-cholesterol. After 48 h the medium was replaced with Ham's F12 containing 1% EFAF-BSA for further 24 h. Cells were then incubated with 100 μg protein/ml of native and modified HDL3 and 3Hcholesterol efflux was determined.

294

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

2.5. 3H-cholesteryl oleyl ether cell uptake Aliquots of 3H-cholesteryl oleyl ether were dried under N2 stream, then native and modified HDL3 were added to the tube for 24 h at 37 °C at the ratio 1 μCi/mg HDL protein. The labeling efficiency was evaluated by liquid scintillation counting. ldlA7 and ldlA[mSR-BI] cells were incubated for 6 h at 37 °C in medium containing 200 μg/ml of 3H-CE-labeled native or modified HDL3. Cells were then washed thrice with cold PBS and lyzed with 0.1 N NaOH. For each well radioactivity was evaluated, and expressed as percentage of internalized 3H-CE over total 3H-CE added to each well.

2.6. Flow cytometry J774 were loaded with AcLDL (50 μg/ml) in MEM + 0.2% BSA, followed by a 24 h wash out in MEM + 0.2% BSA. Lipid loaded J774, ldlA7 and ldlA[mSR-BI] cells were incubated at 37 °C for 1 h with increasing concentrations of native or modified DiO-labeled HDL3. Cells were then washed thrice with cold PBS, detached by scraping, fixed in 1% paraformaldehyde and immediately subjected to fluorescence flow cytometry using a FACScan (Becton Dickinson). For each sample, 10,000 events were analyzed. To study the role of SR-BI, J774 were pre-incubated with a 1:1.000 dilution of anti-SR-BI for 45′ at 37 °C before the addition of DiO-labeled lipoproteins.

2.7. Immunoblotting Apoproteins of native and modified HDL3 were separated by 10% SDSPAGE. After electrophoresis, proteins were transferred onto a nitrocellulose membrane; immunoblotting was performed utilizing anti-apoAI (1:50.000) or anti-apoAII (1:50.000) polyclonal antibodies. Immunocomplexes were detected by ECL followed by autoradiography. To analyze the expression of SR-BI and ABCA1 cells were lyzed with a buffer containing 50 mM TRIS, 50 mM EDTA, 2% SDS and 0.1 M βmercaptoethanol; cell proteins were electophoresed on a 6% SDS-PAGE, then transferred onto a nitrocellulose membrane. Protein expression was

analyzed by immunoblotting using either a rabbit anti-SR-BI antibody (1:2000) or a rabbit anti-ABCA1 antibody (1:500); a mouse anti-ß-actin antibody (1:5000) was used to normalize the protein loading. After incubation with an anti-rabbit or an anti-mouse IgG peroxidase-conjugated as secondary antibodies, immunocomplexes were detected by ECL followed by autoradiography. To determine the distribution of apoAI in HDL3 particles, aliquots of native and modified HDL3 were electrophoresed on a 0.8% agarose gel, then transferred onto a nitrocellulose membrane. ApoAI was identified using an anti serum anti-human apoA-I from rabbit (1:20,000) followed by a goat anti-rabbit IgG peroxidase-conjugated (1:20,000). Immunocomplexes were detected by ECL followed by autoradiography.

2.8. Statistical analysis Differences were analyzed by the Student's t-test and values of P b 0.05 were considered to be significant.

3. Results 3.1. Characteristics of native and modified HDL3 Structural modifications of HDL3 induced by 15-LO were analyzed and compared to the modifications induced by copper ions. As expected, when HDL3 was incubated with Cu++ large amounts of TBARS were formed (Fig. 1A) and the electrophoretic mobility was increased (Fig. 1B). Oxidation of HDL3 with 15-LO did also induce TBARS formation (Fig. 1A), but no changes in the electrophoretic mobility were observed (Fig. 1B). To evaluate more precisely the extent of HDL3 modification, the hydroxy PUFA/PUFA ratio (%) was determined [23]. The modification of HDL3 with 15-LO for 72 h resulted in a slightly elevated content of oxygenated fatty acid (4.7%), while the

Fig. 1. Characterization of enzymatically and non-enzymatically modified HDL3. (A) TBARS content of native and modified HDL3. (B) Electrophoretic mobility of native and modified HDL3 was evaluated by agarose gel electrophoresis. (C, D) HDL3 were modified with Cu++ or 15-LO for different period times, then subjected to a 10% SDS-PAGE followed by Western blotting with anti-human apoAI serum (C) or anti-human apoAII-serum (D). Lanes are as follows: 1, native HDL3; 2, Cu2h; 3, Cu24h; 4, 15-LO24h; 5, 15-LO72h.

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

Fig. 2. Cholesterol efflux from J774 to native or modified HDL3. J774 cells were incubated with 3H-cholesterol and AcLDL, as described in Materials and methods. After an 18 h equilibration period, cells were incubated for 6 h with increasing concentrations of native or modified HDL3 (50–400 μg/ml). Radioactivity content in the extracellular medium was evaluated by liquid scintillation counting. Results are expressed as dpm corrected for cellular protein content. Data are mean ± S.D. of 5 experiments performed in duplicate. Student's t-test: *Pb0.05; **P b 0.005; ***P b 0.0005 vs. native HDL3.

modification with Cu++ for 2 h induced a strong increase in hydroxy PUFA content (71%). As expected, Cu++-catalyzed oxidative modification induced profound changes in the apolipoprotein structure, with the appearance of cross-linked forms of apoAI and apoAII, and a concomitant decrease of the monomeric form (Fig. 1C and D). Enzymatic modification did also induce time-dependent apolipoprotein cross-linking, although to a lower extent (Fig. 1C and D). 3.2. Impact of 15-LO-induced HDL3 modification on cholesterol efflux The modification of HDL3 with 15-LO reduced the lipoprotein ability to promote cholesterol efflux, and the

295

Fig. 4. Effect of HDL3 modification on SR-BI-dependent cholesterol efflux in SR-BI-overexpressing cells. (A) Immunoblot analysis of SR-BI protein levels in J774, ldlA7 and ldlA[mSR-BI] cells. (B) Cholesterol efflux from ldlA[mSR-BI] and ldlA7 cells. Cells were labeled with 3H-cholesterol (see Materials and methods). After an 18-h equilibration period, cells were incubated for 6 h with native or modified HDL3 (100 μg/ml). Radioactivity content in the extracellular medium was measured by liquid scintillation counting. Results, expressed as dpm corrected for the cellular protein content, are given as means ± S.D. of triplicate determinations. Student's t-test, *P b 0.05; **P b 0.01 vs. native HDL3.

extent of reduction depended on the duration of the modification period (Fig. 2). Unexpectedly, we found that modification of HDL3 with 15-LO for 72 h or with Cu++ for 2 h, although generating lipoproteins with considerably different oxidation degrees, produced similar functional effects, as the cholesterol efflux mediated by this two lipoprotein species was comparable at all concentrations tested. The cholesterol efflux mediated by HDL3 modified with Cu++ for 24 h was, as expected, lower.

Fig. 3. Effect of HDL3 modification on SR-BI-dependent cholesterol efflux. (A) 3H-cholesterol labeled J774 were pre-incubated in the absence or the presence of a 1:1.000 dilution of a SR-BI blocking antibody for 45′, then incubated for 6 h at 37 °C with 200 μg/ml of native or modified HDL3 (a); alternatively, cells were preincubated for 24 h with 10 μg/ml OxLDL, then incubated for 6 h with 200 μg/ml of native or modified HDL3 (b). Radioactivity in the extracellular medium was measured by liquid scintillation counting. SR-BI-dependent efflux was calculated as the difference between total cholesterol efflux and the efflux obtained after cell pre-treatment (SR-BI-blocking antibody (a) or OxLDL(b)). Data are mean ± S.D. of 3 experiments performed in duplicate. Student's t-test: *P b 0.05; **P b 0.01 vs. SR-BI-dependent cholesterol efflux to native HDL3. (B) ABCA1 and SR-BI protein expression in OxLDL treated J774 cells. J774 were lipid loaded with 50 μg/ml AcLDL in 0.2% BSA, then incubated for 24 h in the absence or the presence of 10 μg/ml OxLDL. Cells were solubilized in lysis buffer for Western blotting analysis as described in Materials and methods.

296

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

Fig. 5. Effect of HDL3 modification on ABCA1-dependent cholesterol efflux. (A) J774 were incubated with 10% FBS, or AcLDL (50 μg/ml), or AcLDL followed by 22OH/9cis (10 μM, 1 μM). ABCA1 and SR-BI protein expression were determined by Western blotting. (B) 3H-cholesterol-labeled J774 cells were incubated with 22-OH/9cis (10 μM and 1 μM, respectively) to increase ABCA1 expression. Native and modified HDL3 (10 μg/ml) were added to the cells for 6 h and the 3Hcholesterol efflux was determined. Data are mean ± S.D. of 3 experiments performed in duplicate. *P b 0.01 vs. native HDL3. (C) Distribution of apoAI in subfractions of native and modified HDL3. was analyzed by 0,8% agarose gel electrophoresis, followed by immunoblotting using an anti-apoAI antibody.

3.3. Effect of 15-LO-induced HDL3 modification on SR-BI-dependent cholesterol efflux To investigate the consequences of 15-LO-mediated oxidation of HDL3 on their ability to remove cholesterol through SR-BI receptor, cholesterol efflux experiments were performed in the presence of a SR-BI blocking antibody, which impairs the receptor function but not its expression (not shown). Assuming that antibody treatment completely abolished SR-BI-dependent cholesterol efflux, ∼50% of native HDL3-mediated cholesterol efflux was SR-BI-dependent (Fig. 3A); non-immune serum or irrelevant IgG had no effect on cholesterol efflux to HDL3 (not shown). After modification of HDL3 with 15-LO for 72 h or with Cu++ for 2 h, the fraction of SR-BI-dependent cholesterol efflux was reduced to ∼35% (Fig. 3A). As expected, modification of HDL3 with Cu++ for 24 h further reduced the SR-BI-mediated cholesterol efflux (28%) (Fig. 3A). These results were confirmed in experiments in which J774 cells were preincubated with OxLDL to reduce the SR-BI expression before the exposure to native or modified HDL3 (Fig. 3B); in addition, OxLDL pre-incubation reduced ABCA1 expression (Fig. 3B). Under these experimental conditions, 56% of HDL3-mediated cholesterol efflux was found to be SR-BIdependent (Fig. 3A); modification of HDL3 with 15-LO for 72 h or with Cu++ for 2 h significantly reduced this percentage (Fig. 3A). As lipid loading of macrophages with acetylated LDL decreases the expression of SR-BI protein [41] (Fig. 3B), cholesterol efflux was determined in ldlA[mSR-BI], a cell line stably transfected with SR-BI [42]. Cells overexpressing SR-BI (Fig. 4A) showed a higher cholesterol efflux in

response to native HDL3, compared to untransfected control cells (ldlA7). HDL3 modification with 15-LO for 72 h or with Cu++ for 2 h significantly reduced the efflux rate (Fig. 4B). Using the SR-BI blocking antibody, we could establish that ∼70% of cholesterol efflux to native HDL3 was SR-BIdependent; this percentage was significantly reduced after HDL3 modification (∼30%) (data not shown). 3.4. Effect of 15-LO-induced HDL3 modification on ABCA1-dependent cholesterol efflux Next, we aimed at investigating whether exposure of HDL3 to 15-LO might also impair ABCA1-mediated cholesterol efflux. To up-regulate the expression of ABCA1, lipid-laden J774 were incubated with 22-OH/9cis (Fig. 5A); under these experimental conditions ABCA1 expression (negligible under basal conditions) was significantly increased, while SR-BI expression was further reduced. Modification of HDL3 with 15-LO for 72 h or with Cu++ (2 h or 24 h) significantly decreased ABCA1-mediated cholesterol efflux (Fig. 5B). Western blotting analysis revealed that modification of HDL3 with 15-LO or with Cu++ caused the disappearance of pre-β-HDL3 particles (Fig. 5C), that may explain the decreased ABCA1-dependent cholesterol efflux. 3.5. Impact of 15-LO-induced HDL3 modification on lipoprotein–cell association Modification of HDL3 with 15-LO for 72 h or with Cu++ for 2 h significantly impaired lipoprotein association to both J774 (Fig. 6A) or ldlA[mSR-BI] cells (Fig. 6B); HDL3 modified with

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

297

Fig. 6. Interaction of HDL3 species with cells. (A) J774 cells were loaded with AcLDL, as described in Materials and methods. After an 18-h equilibration period, cells were incubated with increasing concentrations of DiO-labeled lipoproteins (10–75 μg/ml) for 1 h at 37°C. Cell-associated fluorescence was evaluated by flow cytometry. Data are mean ± S.D. of 3 independent experiments performed in duplicate. Student's t-test, *P b 0.05; **P b 0.005 vs. native HDL3. (B) ldlA[mSR-BI] cells were incubated for 1 h at 37°C in the presence of increasing concentrations of DiO-labeled native or modified HDL3. Cell-associated fluorescence was determined by flow cytometry. Data are mean ± S.D. of 3 independent experiments performed in duplicate. Student′s t-test: *P b 0.005; **P b 0.0005; ***P b 0.00005 vs. native HDL3. (C) Lipid-laden J774 were preincubated (45’) with a SR-BI blocking antibody, then incubated for 2 h with 10 μg/ml of DiO-labeled native or modified HDL3. Cell-associated fluorescence was assayed by flow cytometry. SR-BI-dependent association was calculated as the difference between lipoprotein–cell association in the absence and in the presence of SR-BI blocking antibody. Data are mean ± S.D. of 3 independent experiments performed in duplicate. Student's t-test: *P b 0.01; **P b 0.005 vs. native HDL3.

Cu++ for 24 h showed a negligible cell-association (not shown). The association of native or modified HDL3 to control ldlA7 cells was much lower (not shown). Using the SR-BI blocking antibody, we established that 62% of native HDL3 association to J774 cells is SR-BI-dependent (Fig. 6C); exposure of HDL3 to 15-LO for 72 h or to Cu++ for 2 h significantly reduced the relevance of SR-BI in determining the association of HDL to the cells (Fig. 6C).

in ldlA7 cells incubated with native or 15-LO72h modified HDL3, whereas the 3 H-CE uptake from Cu24h modified HDL3 was lower.

3.6. 3H-cholesteryl oleyl ether cell uptake Binding of HDL3 to SR-BI is required for cholesterol efflux but this interaction may also initiate uptake of cholesteryl esters (CE) from HDL. To obtain independent evidence for an impaired binding of 15-LO-modified HDL3 to SR-BI, we determined cellular CE uptake in ldlA[mSR-BI] cells. Similar reductions of 3 H-CE uptake in cells exposed to HDL3 modified with 15-LO for 72 h or with Cu++ for 24 h were observed, compared with native lipoprotein (Fig. 7). Less significant differences of 3 H-CE uptake were observed

Fig. 7. 3H-cholesteryl ether uptake from native or modified HDL3. ldlA7 and ldlA[mSR-BI] were incubated for 18 h with 200 μg/ml of 3H-cholesteryl etherlabeled native, 15-LO72h or Cu24h modified HDL3. After washing, cells were lyzed in 0.1 M NaOH and 3H-CE uptake was determined by radioactivity measurement. Data are mean ± S.D. of 3 independent experiments in duplicate. Student's t-test: *P b 0.05; **P b 0.00001 vs. native HDL3.

298

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

4. Discussion HDL is susceptible to oxidation and this reaction may be induced non-enzymatically (transition metals) or enzymatically, as consequence of metabolic formation of oxidizing compounds [43,44]. In most cases oxidative modification impairs the intrinsic capacity of HDL to promote cholesterol efflux from lipidladen cells [14–17]. Here, we explored whether 15-lipoxygenase, an enzyme capable of converting LDL to an atherogenic form, could modify HDL, thus affecting its cholesterol effluxing properties. Lipoxygenases constitute a family of enzymes, which catalyze a selective lipid oxidation [45]. Providing hydroperoxy lipids, these enzymes may initiate an oxidative cascade, which involves radical-mediated propagation reactions. Therefore, though 12/15-lipoxygenases undergo irreversible suicide inactivation during fatty acid oxygenation, a process that may be relevant for down-regulation of enzyme activity, the formed radicals may, in turn, lead to the formation of an unspecific product pattern [28] and to a co-oxidative modification of the apolipoproteins [22]. Further, it has been shown that rabbit and human 15-lipoxygenase induce LDL oxidation mostly by nonenzymatic reactions [46]. Accordingly, with these observations, we found that 15-LO induced a time-dependent cross-linking of the apoproteins, with the appearance of homo- and heterodimers as well as apoAI trimers. More importantly, however, the incubation of cells with 15-LO modified HDL3 resulted in an impaired cholesterol efflux. Previous experiments suggested that HDL binding to SR-BI is involved in the cholesterol efflux from macrophages [47] and recent studies indicated that modification of HDL with hypochlorous acid impairs reverse cholesterol transport while increasing the lipoprotein affinity for SR-BI [15]. This finding proposes a “non-productive” binding of HDL to SR-BI resulting in a reduced cholesterol efflux. Our data suggest that modification of HDL3 induced by 15-LO impairs its ability to interact with SR-BI, which is consistent with the observed reduction of cholesterol efflux. One of the main factors regulating SR-BI-mediated cholesterol efflux is the phospholipid composition of HDL, mainly its content in phosphatidylcholine and sphingomyelin [48]. Factors modulating HDL phospholipid composition may have relevant effects on HDL/SR-BI interaction and, consequently, on SR-BImediated cholesterol efflux [49,50]. However, we did not detect changes in the phosphatidylcholine or sphingomyelin content of 15-LO-modified HDL3 (not shown), in line with previous observations [15,51]. From these data one might conclude that the differences in the ability of 15-LO-modified HDL3 to interact with cell surface receptors should be attributed to the apolipoproteins modification. In fact, human apoAI binds murine SR-BI with high affinity independently of its lipidation state [52]; moreover, cross-linking studies demonstrated that SR-BI interacts with multiple sites of apoAI suggesting that direct protein–protein contact is the main feature of apoAI/SR-BI interaction [52]. On the basis of these observations, we hypothesize that the observed cross-links between HDL apoproteins (apoAI and apoAII) may hamper the interaction with SR-BI and thus, cholesterol efflux.

One of the major observations of this study is that a relatively low degree of oxidative modification of HDL may suffice to impair the cholesterol efflux properties. After a 72-h modification of HDL3 with 15-LO only a minor formation of hydroxy fatty acids could be observed. In contrast, more than 70% of HDL PUFAs were oxidized when Cu++ was used as catalyst (2 h). Nevertheless, comparable reductions in cholesterol efflux were observed, suggesting that the degree of lipid oxygenation does not parallel the capability of HDL to act as cholesterol acceptor. In fact, a relatively low degree of apoprotein modification may be sufficient to significantly alter the affinity of HDL3 for SR-BI receptor. This observation is in line with the earlier finding that HDL oxidized by gamma-radiolysisgenerated oxyradicals revealed comparable reduction in cholesterol effluxing activity compared to copper-modified HDL, although exhibiting a significantly lower TBARS content [53]. Another remarkable finding is that 15-LO-mediated modification of HDL3, besides the effects on SR-BI-mediated cholesterol efflux, induced also a significant impairment of ABCA1-mediated cholesterol efflux from macrophages, an effect associated to the loss of pre-ß-particles upon HDL3 exposure to the enzyme. Taken together, these results suggested that, besides the function in LDL modification, 15-LO could reveal a further proatherogenic role through the generation of a dysfunctional HDL which, by exhibiting impaired cholesterol effluxing properties, might contribute to the formation of foam cells in the atherosclerotic lesions. Acknowledgements We are grateful to Prof. Monty Krieger (MIT, Cambridge, MA) for providing ldlA7 and ldlA[mSR-BI] CHO cells, and to Prof. D. Caruso for mass spectrometry analysis. This work was supported in part by grants from FIRB (Fondo per gli Investimenti della Ricerca di Base) and PRIN (Programmi di Ricerca di Rilevante Interesse Nazionale) to A.L.C., in part by a grant (Ku 961/8-1) of Deutsche Forschungsgemeinschaft and the European Commission (FP6, LSHM-CT-2004-0050333) to H.K.

References [1] I. Tabas, Consequences of cellular cholesterol accumulation: basic concepts and physiological implications, J. Clin. Invest. 110 (2002) 905–911. [2] B. Osterud, E. Bjorklid, Role of monocytes in atherogenesis, Physiol. Rev. 83 (2003) 1069–1112. [3] A. von Eckardstein, J.R. Nofer, G. Assmann, Acceleration of reverse cholesterol transport, Curr. Opin. Cardiol. 15 (2000) 348–354. [4] D. Sviridov, P. Nestel, Dynamics of reverse cholesterol transport: protection against atherosclerosis, Atherosclerosis 161 (2002) 245–254. [5] A. von Eckardstein, J.R. Nofer, G. Assmann, High density lipoproteins and arteriosclerosis. Role of cholesterol efflux and reverse cholesterol transport, Arterioscler. Thromb. Vasc. Biol. 21 (2001) 13–27. [6] H.B. Brewer Jr., A.T. Remaley, E.B. Neufeld, F. Basso, C. Joyce, Regulation of plasma high-density lipoprotein levels by the ABCA1 transporter and the emerging role of high-density lipoprotein in the treatment of cardiovascular disease, Arterioscler. Thromb. Vasc. Biol. 24 (2004) 1755–1760.

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300 [7] P.G. Yancey, A.E. Bortnick, G. Kellner-Weibel, M. de la Llera-Moya, M.C. Phillips, G.H. Rothblat, Importance of different pathways of cellular cholesterol efflux, Arterioscler. Thromb. Vasc. Biol. 23 (2003) 712–719. [8] J.F. Oram, HDL apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol, Arterioscler. Thromb. Vasc. Biol. 23 (2003) 720–727. [9] J.P. Slotte, L. Gronberg, Oxidation of cholesterol in low density and high density lipoproteins by cholesterol oxidase, J. Lipid Res. 31 (1990) 2235–2242. [10] A. Lavy, G.J. Brook, G. Dankner, A. Ben Amotz, M. Aviram, Enhanced in vitro oxidation of plasma lipoproteins derived from hypercholesterolemic patients, Metabolism 40 (1991) 794–799. [11] S. Bradamante, L. Barenghi, G.A. Giudici, C. Vergani, Free radicals promote modifications in plasma high-density lipoprotein: nuclear magnetic resonance analysis, Free Radic. Biol. Med. 12 (1992) 193–203. [12] R. Musanti, G. Ghiselli, Interaction of oxidized HDLs with J774-A1 macrophages causes intracellular accumulation of unesterified cholesterol, Arterioscler. Thromb. 13 (1993) 1334–1345. [13] I. Hurtado, C. Fiol, V. Gracia, P. Caldu, In vitro oxidised HDL exerts a cytotoxic effect on macrophages, Atherosclerosis 125 (1996) 39–46. [14] I. Suc, S. Brunet, G. Mitchell, G.E. Rivard, E. Levy, Oxidative tyrosylation of high density lipoproteins impairs cholesterol efflux from mouse J774 macrophages: role of scavenger receptors, classes A and B, J. Cell Sci. 116 (2003) 89–99. [15] G. Marsche, A. Hammer, O. Oskolkova, K.F. Kozarsky, W. Sattler, E. Malle, Hypochlorite-modified high density lipoprotein, a high affinity ligand to scavenger receptor class B, type I, impairs high density lipoprotein-dependent selective lipid uptake and reverse cholesterol transport, J. Biol. Chem. 277 (2002) 32172–32179. [16] L. Gesquiere, N. Loreau, D. Blache, Impaired cellular cholesterol efflux by oxysterol-enriched high density lipoproteins, Free Radic. Biol. Med. 23 (1997) 541–547. [17] V.A. Rifici, A.K. Khachadurian, Oxidation of high density lipoproteins: characterization and effects on cholesterol efflux from J774 macrophages, Biochim. Biophys. Acta 1299 (1996) 87–94. [18] Y. Nagano, H. Arai, T. Kita, High density lipoprotein loses its effect to stimulate efflux of cholesterol from foam cells after oxidative modification, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 6457–6461. [19] C.C. Hedrick, K. Hassan, G.P. Hough, J.H. Yoo, S. Simzar, C.R. Quinto, S.M. Kim, A. Dooley, S. Langi, S.Y. Hama, M. Navab, J.L. Witztum, A.M. Fogelman, Short-term feeding of atherogenic diet to mice results in reduction of HDL and paraoxonase that may be mediated by an immune mechanism, Arterioscler. Thromb. Vasc. Biol. 20 (2000) 1946–1952. [20] C. Bergt, S. Pennathur, X. Fu, J. Byun, K. O'Brien, T.O. McDonald, P. Singh, G.M. Anantharamaiah, A. Chait, J. Brunzell, R.L. Geary, J.F. Oram, J.W. Heinecke, The myeloperoxidase product hypochlorous acid oxidizes HDL in the human artery wall and impairs ABCA1-dependent cholesterol transport, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 13032–13037. [21] L. Zheng, B. Nukuna, M.L. Brennan, M. Sun, M. Goormastic, M. Settle, D. Schmitt, X. Fu, L. Thomson, P.L. Fox, H. Ischiropoulos, J.D. Smith, M. Kinter, S.L. Hazen, Apolipoprotein A-I is a selective target for myeloperoxidase-catalyzed oxidation and functional impairment in subjects with cardiovascular disease, J. Clin. Invest. 114 (2004) 529–541. [22] H. Kuhn, J. Belkner, H. Suzuki, S. Yamamoto, Oxidative modification of human lipoproteins by lipoxygenases of different positional specificities, J. Lipid Res. 35 (1994) 1749–1759. [23] J. Belkner, H. Stender, H. Kuhn, The rabbit 15-lipoxygenase preferentially oxygenates LDL cholesterol esters, and this reaction does not require vitamin E, J. Biol. Chem. 273 (1998) 23225–23232. [24] Y. Takahashi, W.C. Glasgow, H. Suzuki, Y. Taketani, S. Yamamoto, M. Anton, H. Kuhn, A.R. Brash, Investigation of the oxygenation of phospholipids by the porcine leukocyte and human platelet arachidonate 12-lipoxygenases, Eur. J. Biochem. 218 (1993) 165–171. [25] S. Yla-Herttuala, M.E. Rosenfeld, S. Parthasarathy, C.K. Glass, E. Sigal, J.L. Witztum, D. Steinberg, Colocalization of 15-lipoxygenase mRNA and protein with epitopes of oxidized low density lipoprotein in macrophagerich areas of atherosclerotic lesions, Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 6959–6963.

299

[26] S. Yla-Herttuala, M.E. Rosenfeld, S. Parthasarathy, E. Sigal, T. Sarkioja, J.L. Witztum, D. Steinberg, Gene expression in macrophage-rich human atherosclerotic lesions. 15-lipoxygenase and acetyl low density lipoprotein receptor messenger RNA colocalize with oxidation specific lipid–protein adducts, J. Clin. Invest. 87 (1991) 1146–1152. [27] H. Kuhn, D. Heydeck, I. Hugou, C. Gniwotta, In vivo action of 15lipoxygenase in early stages of human atherogenesis, J. Clin. Invest. 99 (1997) 888–893. [28] H. Kuhn, J. Belkner, S. Zaiss, T. Fahrenklemper, S. Wohlfeil, Involvement of 15-lipoxygenase in early stages of atherogenesis, J. Exp. Med. 179 (1994) 1903–1911. [29] R. Spanbroek, R. Grabner, K. Lotzer, M. Hildner, A. Urbach, K. Ruhling, M.P. Moos, B. Kaiser, T.U. Cohnert, T. Wahlers, A. Zieske, G. Plenz, H. Robenek, P. Salbach, H. Kuhn, O. Radmark, B. Samuelsson, A.J. Habenicht, Expanding expression of the 5-lipoxygenase pathway within the arterial wall during human atherogenesis, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 1238–1243. [30] D. Harats, A. Shaish, J. George, M. Mulkins, H. Kurihara, H. Levkovitz, E. Sigal, Overexpression of 15-lipoxygenase in vascular endothelium accelerates early atherosclerosis in LDL receptor-deficient mice, Arterioscler. Thromb. Vasc. Biol. 20 (2000) 2100–2105. [31] Y. Huo, L. Zhao, M.C. Hyman, P. Shashkin, B.L. Harry, T. Burcin, S.B. Forlow, M.A. Stark, D.F. Smith, S. Clarke, S. Srinivasan, C.C. Hedrick, D. Pratico, J.L. Witztum, J.L. Nadler, C.D. Funk, K. Ley, Critical role of macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein Edeficient mice, Circulation 110 (2004) 2024–2031. [32] T. Cyrus, D. Pratico, L. Zhao, J.L. Witztum, D.J. Rader, J. Rokach, G.A. FitzGerald, C.D. Funk, Absence of 12/15-lipoxygenase expression decreases lipid peroxidation and atherogenesis in apolipoprotein edeficient mice, Circulation 103 (2001) 2277–2282. [33] T.M. Bocan, W.S. Rosebury, S.B. Mueller, S. Kuchera, K. Welch, A. Daugherty, J.A. Cornicelli, A specific 15-lipoxygenase inhibitor limits the progression and monocyte-macrophage enrichment of hypercholesterolemia-induced atherosclerosis in the rabbit, Atherosclerosis 136 (1998) 203–216. [34] D.J. Benz, M. Mol, M. Ezaki, N. Mori-Ito, I. Zelan, A. Miyanohara, T. Friedmann, S. Parthasarathy, D. Steinberg, J.L. Witztum, Enhanced levels of lipoperoxides in low density lipoprotein incubated with murine fibroblast expressing high levels of human 15-lipoxygenase, J. Biol. Chem. 270 (1995) 5191–5197. [35] R.J. Havel, H.A. Eder, J.H. Bragdon, The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum, J. Clin. Invest. 34 (1955) 1345–1353. [36] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193 (1951) 265–275. [37] S.K. Basu, J.L. Goldstein, G.W. Anderson, M.S. Brown, Degradation of cationized low density lipoprotein and regulation of cholesterol metabolism in homozygous familial hypercholesterolemia fibroblasts, Proc. Natl. Acad. Sci. U. S. A. 73 (1976) 3178–3182. [38] A. Pirillo, G.D. Norata, T. Zanelli, A.L. Catapano, Overexpression of inducible heat shock protein 70 in Cos-1 cells fails to protect from cytotoxicity of oxidized ldls, Arterioscler. Thromb. Vasc. Biol. 21 (2001) 348–354. [39] J. Schuh, G.F. Fairclough Jr., R.H. Haschemeyer, Oxygen-mediated heterogeneity of apo-low-density lipoprotein, Proc. Natl. Acad. Sci. U. S. A. 75 (1978) 3173–3177. [40] J. Han, A.C. Nicholson, X. Zhou, J. Feng, A.M. Gotto Jr., D.P. Hajjar, Oxidized low density lipoprotein decreases macrophage expression of scavenger receptor B-I, J. Biol. Chem. 276 (2001) 16567–16572. [41] L. Yu, G. Cao, J. Repa, H. Stangl, Sterol regulation of scavenger receptor class B type I in macrophages, J. Lipid Res. 45 (2004) 889–899. [42] S. Acton, A. Rigotti, K.T. Landschulz, S. Xu, H.H. Hobbs, M. Krieger, Identification of scavenger receptor SR-BI as a high density lipoprotein receptor, Science 271 (1996) 518–520. [43] V.W. Bowry, K.K. Stanley, R. Stocker, High density lipoprotein is the major carrier of lipid hydroperoxides in human blood plasma from fasting donors, Proc. Natl. Acad. Sci. U. S. A. 89 (1992) 10316–10320.

300

A. Pirillo et al. / Biochimica et Biophysica Acta 1761 (2006) 292–300

[44] X. Niu, V. Zammit, J.M. Upston, R.T. Dean, R. Stocker, Coexistence of oxidized lipids and alpha-tocopherol in all lipoprotein density fractions isolated from advanced human atherosclerotic plaques, Arterioscler. Thromb. Vasc. Biol. 19 (1999) 1708–1718. [45] M.K. Cathcart, V.A. Folcik, Lipoxygenases and atherosclerosis: protection versus pathogenesis, Free Radic. Biol. Med. 28 (2000) 1726–1734. [46] D. Heydeck, J.M. Upston, H. Viita, S. Yla-Herttuala, R. Stocker, Oxidation of LDL by rabbit and human 15-lipoxygenase: prevalence of nonenzymatic reactions, J. Lipid Res. 42 (2001) 1082–1088. [47] X. Gu, K. Kozarsky, M. Krieger, Scavenger receptor class B, type Imediated [3H]cholesterol efflux to high and low density lipoproteins is dependent on lipoprotein binding to the receptor, J. Biol. Chem. 275 (2000) 29993–30001. [48] P.G. Yancey, M. de la Llera-Moya, S. Swarnakar, P. Monzo, S.M. Klein, M.A. Connelly, W.J. Johnson, D.L. Williams, G.H. Rothblat, High density lipoprotein phospholipid composition is a major determinant of the bidirectional flux and net movement of cellular free cholesterol mediated by scavenger receptor BI, J. Biol. Chem. 275 (2000) 36596–36604. [49] F.C. de Beer, P.M. Connell, J. Yu, M.C. de Beer, N.R. Webb, D.R. van der Westhuyzen, HDL modification by secretory phospholipase A(2) promotes

[50]

[51]

[52]

[53]

scavenger receptor class B type I interaction and accelerates HDL catabolism, J. Lipid Res. 41 (2000) 1849–1857. P.G. Yancey, M.A. Kawashiri, R. Moore, J.M. Glick, D.L. Williams, M.A. Connelly, D.J. Rader, G.H. Rothblat, In vivo modulation of HDL phospholipid has opposing effects on SR-BI- and ABCA1-mediated cholesterol efflux, J. Lipid Res. 45 (2004) 337–346. C. Bergt, H. Reicher, E. Malle, W. Sattler, Hypochlorite modification of high density lipoprotein: effects on cholesterol efflux from J774 macrophages, FEBS Lett. 452 (1999) 295–300. D.L. Williams, M. de La Llera-Moya, S.T. Thuahnai, S. Lund-Katz, M.A. Connelly, S. Azhar, G.M. Anantharamaiah, M.C. Phillips, Binding and cross-linking studies show that scavenger receptor BI interacts with multiple sites in apolipoprotein A-I and identify the class A amphipathic alpha-helix as a recognition motif, J. Biol. Chem. 275 (2000) 18897–18904. D. Bonnefont-Rousselot, C. Motta, A.O. Khalil, R. Sola, A.E. La Ville, J. Delattre, M. Gardes-Albert, Physicochemical changes in human highdensity lipoproteins (HDL) oxidized by gamma radiolysis-generated oxyradicals. Effect on their cholesterol effluxing capacity, Biochim. Biophys. Acta 1255 (1995) 23–30.