- Email: [email protected]
High-density lipoprotein antagonizes oxidized low-density lipoprotein by suppressing oxygen free-radical formation and preserving nitric oxide bioactivity
Atherosclerosis 183 (2005) 251–258
High-density lipoprotein antagonizes oxidized low-density lipoprotein by suppressing oxygen free-radical formation and preserving nitric oxide bioactivity Chii-Ming Lee a , Chiang-Ting Chien b , Po-Yuan Chang a , Mo-Ying Hsieh a , Hsiang-Yiang Jui a , Chau-Song Liau a , Su-Ming Hsu c , Yuan-Teh Lee a,∗ a b
Departments of Internal Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan Departments of Medical Research, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan c Departments of Pathology, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan Received 6 December 2004; received in revised form 25 January 2005; accepted 2 March 2005 Available online 10 August 2005
Abstract The antiatherogenic role of high-density lipoprotein (HDL) has been related to its ability to increase the activity of endothelial nitric oxide synthase (eNOS) and to protect low-density lipoprotein (LDL) against oxidative modification. The present study was aimed to determine whether and how HDL antagonizes oxidized LDL (oxLDL) that has been formed and accumulated in circulation. Pre-infusion of rats with HDL effectively prevented oxLDL-induced renal vascular constriction. Consistently, pre-incubation of human saphenous vein endothelial cells with HDL (100 g/ml) reversed the oxLDL-induced suppression of endothelium-dependent cyclic-GMP production in co-cultured smooth muscle cells. However, the changes of Akt phosphorylation and eNOS activity in endothelial cells in response to lipoprotein treatments under our assay condition were not significant. Intriguingly, pretreatment of human umbilical vein endothelial cells with HDL (50 g/ml) for only 30 s effectively reduced the level of free radicals generated by oxLDL or H2 O2 . In kidneys of living rats, renal arterial infusion of oxLDL greatly enhanced ischemia/reperfusion-induced free radicals, which could be attenuated by HDL pretreatment. We conclude that HDL may antagonize oxLDL on endothelial function through an Akt-independent pathway in which HDL preserves nitric oxide bioactivity by attenuating oxLDL-triggered free radical generation. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: High-density lipoprotein; Oxidized low-density lipoprotein; Nitric oxide; Endothelial cells; Free radicals
1. Introduction Decreased high-density lipoprotein (HDL) plasma levels represent the major risk factor associated with coronary atherosclerosis progression [1,2]. Classically, the atheroprotective mechanism of HDL is thought to be related to its role in reverse transport of cholesterol from peripheral tissues to the liver [3]. Recent studies indicate that ∗
Corresponding author. Fax: +886 2 2395 9911. E-mail address: [email protected] (Y.-T. Lee).
0021-9150/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2005.03.029
endothelium is a key target of HDL action. It has been demonstrated that HDL stimulates endothelial nitric-oxide synthase (eNOS) activity through the scavenger receptor SR-BI in a ceramide-dependent manner [4,5] or through lysophospholipid receptor S1P3 binding and Akt activation [6,7]. NO produced by eNOS in the endothelium regulates blood flow, inflammation, and platelet aggregation, and consequently its disruption during endothelial dysfunction can impair endothelium-dependent vasorelaxation and encourage the formation of atherosclerotic lesions and thrombi.
252
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
The antiatherogenic action of HDL may also be partially ascribed to its effect to protect LDL from oxidation through HDL-associated enzymes, such as paraoxonase 1 (PON1), lecithin-cholesterol acryltransferase (LCAT), or platelet-activating factor acetylhydrolase (PAF-AH) [8]. Oxidized LDL (oxLDL) has been shown to be the active component in dyslipidemia that plays an important role in atherosclerosis [9]. Attenuation of endothelium-dependent NO-mediated vasodilation is one of the postulated atherogenesis mechanisms of oxLDL [10]. oxLDL per se contains lipid hydroperoxide [11] and is a potent inducer of reactive oxygen species (ROS) in smooth muscle cells (SMCs) [12]. On binding to oxLDL receptor on endothelial cells (ECs), oxLDL may reduce the intracellular concentration of NO via increased production of superoxide [13]. Because HDL exerts anti-oxidant effects and endothelium is a key target of HDL action, the present study was designed to determine, in addition to protect LDL against oxidative modification, whether and how HDL directly blocks the effects of oxLDL on vascular endothelium both in vivo and ex vivo.
2. Methods 2.1. Materials Tetrahydrobiopterin (BH4) was from RBI (Natick, MA), all other chemicals and lipoprotein-deficient human serum were from Sigma Chemical Co. (St. Louis, MO). Type I collagenase and elastase were from Worthington Biochemical Co. (Freehold, NJ, USA). M199 medium, fetal bovine serum, Earle’s balanced salt solution (EBSS) were from Invitrogen (Grand Island, NY). MCDB107 was from JRH Biosciences (Lenex, KS). Endothelial cell growth supplement (ECGS) was from Collaborative Research Inc. (Bedford, MA). Endothelial cell growth factor (ECGF), pronase, and complete protease inhibitors were from Roche (Indianapolis, IN). Cell culture wares were from Corning Inc. (Cambridge MA). Antibodies were from Cell Signaling Technology Inc. (Beverly, MA). Electrochemiluminescence (ECL) reagents and 3Harginine were from Amersham Bioscience (Piscataway, NJ). 2.2. Preparation of LDL, oxLDL, and HDL LDL (1.019–1.063 g/ml) and HDL (1.063–1.210 g/ml) were prepared from normal human plasma by sequential ultracentrifugation in a sodium bromide gradient. OxLDL was prepared by incubation of LDL (1.2 g/l) with 15 mol/l copper chloride at 37 ◦ C for 90–120 min. The extent of lipoprotein oxidation was measured by the thiobarbituric acid reactive substances (TBARS) assay and electrophoretic mobility (Lipogel, Beckman).
2.3. Measurement of regional blood flow rate in vivo The animal care and experimental protocol were in accordance with the guidelines of the National Science Council of the Republic of China. Regional blood flow rate was measured as described previously [14]. Female Wistar rats (200–250 g) were anesthetized with urethane (1 mg/1 g) and underwent tracheotomy. Lipoproteins were infused into the thoracic aorta by a syringe pump (KD Scientific, Holliston, MA). A pressure transducer (Gould-Statham, Quincy, MA) was placed in the left carotid artery for continuous monitoring of blood pressure (BP). Electromagnetic flow meter probes (Transonic System Inc., Ithaca, NY) were placed on the left renal artery and abdominal aorta for the detection of blood flow rates. Renal vascular resistance was generated via dividing BP by the renal artery flow rate. 2.4. Cell culture Human saphenous veins (HSVs) obtained from patients undergoing coronary artery bypass surgery were incubated with 0.1% collagenase at 37 ◦ C for 25 min [15]. After ECs were scratched from the luminal surface of the HSVs, the remaining tissues were torn in strips and further digested at 37 ◦ C with 0.2% elastase for 2 h followed with 0.3% collagenase for 3–5 h [16]. SMCs were collected by centrifugation. Both types of cells were cultured in M199 medium supplemented with 20% human serum, 5 U/ml heparin, 100 IU/ml penicillin, and 0.1 mg/ml streptomycin except that EC culture required addition of 100 g/ml ECGS. Cells from passages three to eight were used for subsequent experiments. Primary human umbilical vein endothelial cells (HUVECs) were cultured in medium MCDB107 containing 2% fetal bovine serum, a fibroblast growth factor—enriched fraction of porcine brain extract at 1 g/ml, 10 g/l ECGF, and 50 U/l heparin. Cells between passages two and four were used for subsequent experiments. 2.5. Measurement of cyclic-GMP production in SMCs co-cultured with ECs A co-culture model using buoyant chambers containing HSV ECs in Transwell® cell culture inserts (1 × 105 /4.9 cm2 /insert) and HSV SMCs in six-well plates (1.3 × 105 /9.4 cm2 /well) was included for the analysis of cyclic-GMP content. ECs were first incubated in EBSS containing 1 mmol/l l-arginine/62 mol/l BH4 , and treated with either oxLDL (100 g/ml) or HDL (100 g/ml), or both, at 37 ◦ C for 30 min. The treatment was stopped by washing with EBSS, and the Transwells® containing ECs were placed into the six-well plates containing SMCs. After cocultured with ECs in EBSS containing 1 mmol/l 3-isobutyl1-methlxanthine (IBMX) for 30 min, SMCs were collected and subjected to enzyme-linked immunoassay of cyclic-GMP
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
according to the manufacturer’s instructions (Cayman, Ann Arbor, MI).
253
pronase treatment were resolved by SDS-PAGE followed by coomassie blue staining to confirm the complete digestion of proteins.
2.6. Western blot analysis of Akt and eNOS HUVECs were treated with 50 g/ml of various lipoproteins for 5 and 30 min. Cells were lyzed in a buffer containing 0.5% Triton-100, 0.3 mol/l sodium chloride, 0.025 mol/l sodium phosphate pH 7.4, 20 mmol/l octyglucoside, complete protease inhibitors, 10 mmol/l sodium pyrophosphate, 100 mmol/l sodium fluoride, and 2 mmol/l sodium orthovanadate. Fifteen micrograms of cell lysate proteins were resolved on SDS-polyacrylamide gel electrophoresis followed by immunoblot analysis with antibodies to Akt, phospho-Ser473 Akt, eNOS, and phospho-Ser1177 -eNOS. The signal was detected by ECL. 2.7. Measurement of eNOS activity Confluent HUVECs cultured in six-well dishes were incubated in lipoprotein-deficient medium composed of medium M199 supplemented with 100 g/ml ECGS, 5 U/ml heparin, 62 mol/l BH4 , and 20% lipoprotein-deficient human serum for 24 h. After incubation with oxLDL (100 g/ml), HDL (100 g/ml), or both for 30 min, ECs were labeled with 3 H-arginine, and the radioactivity of 3 H-citrulline, the downstream product of arginine by eNOS, was detected by scintillation counting [17]. 2.8. Measurement of chemiluminescence (CL) in cultured cells HUVECs were cultured to 90% confluence in a 35 mm dish as described above and shift to serum-free, phenol redfree Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen 11054-020) immediately before CL measurements. HUVECs were treated with 50 g/ml of various lipoproteins, 10 mol/l of sphingosine-1-phosphate (S1P) or sphingosylphosphorylcholine (SPC), or H2 O2 (100 M) for 30 s, and then placed in a dark chamber of the Chemiluminescence Analyzing System (Tohoku Electronic Industrial Co., Sendai, Japan). This system contains a photon detector (Model CLD110) that is able to detect as low as 10−15 W of radiant energy and a CL counter (Model CLC-10). Photon emission from the sample was counted at 10 s intervals at 37 ◦ C and atmospheric conditions. At the 100 s time point, 50 mol/l luminol was injected into the medium and the CL in the sample was continuously measured for a total of 600 s. The total amount of CL was calculated by integrating the area under the curve and subtracting it from the background level. To inhibit Akt activity, HUVECs were pre-incubated with wortmannin (10−7 M) for 30 min prior to lipoprotein treatment. For breaking down protein components, HDL preparation was treated with 1 mg/ml of pronase, a nonspecific protease, at 37 ◦ C for 5 h according to manufacturer’s instruction. Aliquots of HDL preparation (1 g/ml) before and after
2.9. In vivo measurement of chemiluminescence (CL) in rat kidneys with ischemia/reperfusion injury CL measurement in rats was performed as described previously [18]. Briefly, after laparotomy, the left renal artery was cannulated for lipoprotein infusion by introduction of a length of stretched PE10 tubing from the left femoral artery via the aorta. A catheter was placed in the left femoral vein for continuous infusion of the CL probe, 2-methyl-6-[4-methoxyphenyl]-3,7-dihydroimidazo[1,2-a]-pyrazin-3-one hydrochloride (MCLA), which is a Cypridina luciferin analogue that is highly specific and sensitive to superoxide anion (O2 •− ) and singlet oxygen (1 O2 ). The rat was then placed on its right side in a modified dark chamber of the Chemiluminescence Analyzing System described in Section 2.8 with the left kidney positioned under the reflector. After lipoprotein infusion, the left renal artery was clamped for 15 min and then reperfused. The MCLAenhanced CL signal from the kidney surface was recorded before and after initiation of ischemia/reperfusion. Shamoperated (control) animals underwent similar operative procedures without occlusion of the renal artery. 2.10. Statistics Data are expressed as mean ± S.D. Data were analysed by Student’s t test or ANOVA. A value of P < 0.05 was considered statistically significant.
3. Results 3.1. HDL could prevent regional artery constriction induced by oxLDL in rats We first examined the effects of oxLDL and HDL on vascular tone in rats. As shown in Fig. 1A, continuous infusion of oxLDL into rat abdominal aorta at increasing rates from 0.1 to 1.0 mg/h gradually reduced renal artery blood flow by 22.5 ± 0.6% to 45.6 ± 14.3% (P < 0.05, Fig. 1C). Neither the blood flow in aorta nor of the systemic arterial pressure measured at carotid artery had remarkable change. The renal vascular resistance was thus raised significantly in response to oxLDL infusion in a dose dependent manner. On the contrary, the regional blood flow of renal artery was slightly elevated, although not statistically significant, in response to the increase of HDL infusion rates from 0.4 to 4.0 mg/h (Fig. 1B). More importantly, priming the vessel by HDL infusion at a concentration between 3.3 and 33 g/ml in renal artery (calculated by dividing HDL infusion rate with renal artery flow) prevented a subsequent oxLDL-induced reduction of renal artery blood flow (Fig. 1C).
254
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
HSV ECs with HDL (100 g/ml, 30 min) per se had no effect on the cyclic-GMP contents in SMCs. However, simultaneously treatment of ECs with HDL plus oxLDL partially reversed oxLDL induced cyclic-GMP reduction (Fig. 2). 3.3. Lipoproteins at low concentrations have no significant effects on Akt-mediated eNOS activation It has been demonstrated recently that HDL, at relatively high concentration (1 mg/ml), induces Akt phosphorylation at Ser473 [7], which in turn activates eNOS in HUVECs [19,20]. To elucidate whether HDL, at the concentration in our setting, antagonizes oxLDL through Akt activation, we examined phosphorylation of Akt and eNOS with phosphospecific antibodies. Treatment of HUVECs with HDL in the range of 0–100 g/ml for up to 30 min did not enhance Akt phosphorylation (Fig. 3A and B). Treatment of HUVECs with 50 g/ml of oxLDL, or HDL plus oxLDL for either 5 or 30 min did not result in significant changes in Akt and eNOS phosphorylation (Fig. 3C). Furthermore, the activity of eNOS in lipoprotein (100 g/ml) treated HUVECs, shown as the l-arginine to l-citrulline conversion rates (Fig. 3D), were similar (P > 0.05) in all groups. These data suggest that the effects of lipoproteins on EC-dependent cyclic-GMP production, shown in Fig. 2, could not be exclusively explained by either Akt activation or the substantial differences in eNOS activity. 3.4. HDL at low concentration partially reversed oxLDL- and H2 O2 -induced oxygen free radical generation in ECs It has been reported that oxLDL induced oxygen free radical formation in vascular endothelial cells may be one of the pathogenic mechanisms of atherosclerosis [13]. To test whether HDL attenuates oxLDL induced ROS in ECs, we Fig. 1. Effects of lipoproteins on renal artery flow. The rat thoracic aorta was infused sequentially with normal saline (0–20 min), LDL (20–35 min), and increasing amounts of oxLDL (35–100 min, expressed as infusion rate, mg/h) (A) or HDL (15–80 min) followed by oxLDL (80–150 min) (B) as indicated on the horizontal axis. Original recordings of renal artery flow, abdominal aortic flow, BP, and resistance were shown. (C) Percent change of renal artery flow from baseline following oxLDL infusion (n = 2 each). () No pre-treatment; () pretreatment with HDL, * P < 0.05 vs. baseline.
3.2. HDL reverses the suppressive effects of oxLDL on endothelium-dependent cyclic-GMP production in SMCs To exam whether the protective effect of HDL on oxLDL induced vessel constriction is NO-dependent, we measured the contents of cyclic-GMP, the second messenger of NO, in HSV SMCs co-cultured with HSV ECs. Incubation of ECs with 100 g/ml oxLDL for 30 min caused a two-thirds reduction in cyclic-GMP amounts in SMCs as compared with that in untreated ECs/SMCs co-culture (P < 0.05). Incubation of
Fig. 2. Effects of lipoproteins on EC-dependent cyclic-GMP production in SMCs. HSV ECs were pre-treated with various types of lipoproteins (100 g/ml) as indicated for 30 min, and then co-cultured with HSV SMCs for additional 30 min. The contents of cyclic-GMP in HSV SMCs were measured by enzyme-linked immunoassay and expressed as pmol/g of protein of SMC homogenate (n = 2). * P < 0.05 vs. untreated, # P < 0.05 vs. oxLDL.
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
255
Fig. 3. Effects of lipoproteins on Akt and eNOS activation. ECs were treated with the indicated lipoproteins (50 g/ml in C and 100 g/ml in D), and then subjected to Western blot analysis (A and C) or l-citrulline assay (D). (B) Densitometric analysis of Akt phosphorylation expressed as fold increase over the basal level in untreated cells (n = 3). In (D), enzyme activity is expressed as fold increase over the control (n = 2). * P < 0.05 vs. control.
analyzed the luminol-activated CL in HUVECs. Treatment of ECs with 50 g/ml oxLDL for 30 s remarkably increased CL signal as compared to that in cells treated with PBS. Neither native LDL nor HDL had a significant effect on ROS generation (Fig. 4A and C). Interestingly, pretreatment of ECs with HDL (50 g/ml) for only 30 s prior to oxLDL exposure reduced oxLDL augmented CL signal to 41.8 ± 13.0% (P < 0.01, Fig. 4C). HDL also reduced free radicals induced by H2 O2 , one of the representative oxidative reagents, to 45.6 ± 7.9% (Fig. 4D). To determine whether HDL-associated proteins are responsible for the attenuation of CL signals, we treated HDL with pronase (Fig. 4B). As shown in Fig. 4C, breaking down the protein components of HDL did not affect the inhibitory effect of HDL on oxLDL augmented ROS formation (45.9 ± 2.1%, P < 0.01). The attenuation of HDL on oxLDL induced CL signal was not blocked by wortmannin which inactivates Akt through inhibition of phosphatidylinositol 3-kinase (42.1 ± 1.5%, Fig. 4C). Furthermore, two HDL-associated lysophospholipids, S1P and SPC, which have been shown to activate Akt and eNOS [23], did not reduce oxLDL-induced free radicals. These data suggest that attenuation of free radicals by HDL may be independent of Akt. 3.5. HDL partially reversed oxLDL-enhanced free radical generation in rat kidneys with ischemia/reperfusion injury To determine whether HDL diminishes oxLDL-enhanced ROS in vivo, we investigated ischemia/reperfusion-induced
oxidative stress in rat kidneys. As shown in Fig. 5, preinfusion of rat kidneys with oxLDL prior to the induction of ischemia dramatically enhanced ROS formation in the reperfused kidney. The augmentation of ROS formation could be partially ameliorated by HDL treatment before oxLDL infusion.
4. Discussion In this report, we demonstrated the vascular protective effects of HDL in vivo. The increase of renal vascular resistance of living rats correlates with the oxLDL infusion rate (Fig. 1). By priming of the vascular beds with HDL (3.3–33 g/ml), the oxLDL-induced elevation of renal vascular resistance was abolished. This is comparable with the findings in organ culture that the inhibition of vasodilatation by oxLDL was prevented by the presence of HDL [21] and that the endothelin-induced vasoconstriction was reversed by HDL injection [22]. Various cholesterol-independent mechanisms of HDL on vascular protection have been postulated. HDL, on binding to SR-BI of ECs, may stimulate eNOS through a ceramidedependent pathway [5] or may prevent oxLDL from decreasing the capacity of eNOS activation by preserving the cholesterol concentration in caveolae [23]. The NO-dependent vasorelaxation effect of HDL has also been reported to be mediated via the lysophospholipid receptor S1P3 and subsequent activation of Akt and eNOS [22]. In this report, we demonstrated that oxLDL and HDL at the concentra-
256
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
Fig. 4. Effects of lipoproteins on luminol-enhanced chemiluminescence (CL) signals in HUVECs. HUVECs were treated with PBS, 50 g/ml of native LDL, HDL, or oxLDL, 10 mol/l of S1P or SPC, or 100 mol/l of H2 O2 for 30 s, and then subjected to CL measurement. In the experiments of HDL + oxLDL, S1P + oxLDL, SPC + oxLDL, and HDL + H2 O2 cells were treated with HDL, S1P, or SPC for 30 s followed by oxLDL or H2 O2 for additional 30 s before CL measurement. (A) Representative results of original recordings. (B) HDL preparation was treated with (+) or without (−) pronase (1 mg/ml) at 37 ◦ C for 5 h and resolved on 7.5% SDS-PAGE followed by coomassie blue staining. (C) Percent of oxLDL increased CL counts from baseline (n = 5). Total CL counts were calculated by integrating of the area under the curve. PBS-induced CL counts was considered as the baseline level of free radicals. (+) Wt, HUVECs were pre-incubated with wortmannin (10−7 mol/l) for 30 min. The increase of oxLDL-induced CL counts from baseline was arbitrarily set as 100%. * P < 0.01 vs. oxLDL. (D) Percent of H2 O2 -increased CL counts from baseline. # P < 0.01 vs. H2 O2 .
Fig. 5. Effects of lipoproteins on free radical generation in rat kidneys with ischemia/reperfusion injury. The left side of rat renal artery was infused with saline, oxLDL (1 mg/ml/h, 30 min), or HDL (2 mg/ml/h, 30 min) followed by oxLDL (1 mg/ml/h, 30 min). Free radical formation in response to ischemia/reperfusion injury was determined by MCLA-induced CL signal. Control, sham-operated rats without ischemia. Representative experiments were superimposed for comparison.
tions between 0 to 100 g/ml did not induce significant Akt or eNOS activation, based on either immunoblotting or lcitrulline formation assay (Fig. 3). However, at such low concentration, HDL effectively reverses the inhibitory effects of oxLDL on endothelium-dependent cyclic-GMP production in the co-cultured SMC (Fig. 2). These findings indicate that HDL may reverse oxLDL-induced impairment of NO signaling through an Akt-independent pathway. The rapidity of inhibition or protection of lipoproteins on NO signaling/cyclic-GMP production between ECs and SMCs precludes the possibility of transcriptional regulation, but favors a biochemical reaction. Recent data suggest that that oxLDL induces NF-B activation in SMCs and ECs, presumably through ROS generation [12,13]. In this report, free radicals were induced within seconds by addition of oxLDL to the cultured ECs and could be attenuated by HDL in a PI3-kinase independent manner. Because only the ECs were exposed to lipoproteins in either the co-culture system (Fig. 2) or CL detection (Fig. 4), it is unlikely that the guanylyl cyclase of vascular SMCs is the direct target of oxLDL or HDL; rather, NO released from ECs may be inactivated by
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
reacting with oxLDL-induced superoxide [13,24] and consequently, reduces downstream cGMP production in SMCs. Attenuation of oxLDL-enhanced free radical formation by HDL, therefore, preserves NO bioactivity. The hypothesis that HDL protects vessels through the attenuation of free radical formation is further supported in the ischemia-reperfusion model in vivo. It has been reported that HDL reduces renal ischemia-reperfusion injury [25]. The effectiveness of vitamin C in the prevention of postischemic inflammatory response indicates oxidative stress may be the pathogenesis mechanism [26]. In this report, pretreatment with HDL decreased the oxLDL-augmented free radical generation in the ischemia-reperfusion model of rat kidneys (Fig. 5). This finding may explain the prevention of oxLDL-induced renal arterial constriction by priming of the vascular beds with HDL. It is not clear what are the active components of HDL that attenuate oxLDL-augmented free radical. The HDLassociated enzymes, such as PON1, LCAT, and PAF-AH, exert potent antioxidative activity. However, it has been reported previously that small, dense HDL might inactivate oxidized LDL lipids by both enzymatic and nonenzymatic mechanisms [27]. Furthermore, a highly purified PON1 fraction from human serum did not protect LDL against oxidation [28]. In this report, we demonstrated that HDL reduced free radicals triggered by H2 O2 which is not a known substrate for LCAT and PAF-AH. More importantly, complete removal of the protein components of HDL preparation by pronase did not affect the antagonizing effect of HDL on oxLDLinduced free radicals (Fig. 4C). The lipid contents of HDL, such as S1P and SPC, that have been shown to activate Akt and eNOS [22], did not attenuate the oxLDL-induced free radicals (Fig. 4C). A recent report indicates that HDL is protected by its associated estradiol ester from oxidation in vitro [29]. A reconstituted HDL composed of aopA1 and phosphatidylcholine inhibits the expression of tissue factor, a thrombogenic molecule, in HUVECs without the activation of Akt/eNOS [30]. Whether these lipid components of HDL exert their antioxidant or antiatherothrombotic activities through scavenging free radicals in vivo remains to be elucidated. Therapeutic interventions in elevation of HDL plasma level or in mimicking its beneficial effects have become the promising strategies in the treatment and prevention of atherosclerosis [31,32]. An understanding of the atheroprotective mechanisms of HDL would be important for the rational utilization or development of pharmaceutic agents. The rapid attenuation of ROS by HDL, shown in this report, might be a surrogate marker of therapeutic efficacy in addition to the classical quantitative analysis of total HDL content or the subfractionation of HDL2 and HDL3 . We conclude that HDL decreases ROS formation augmented by oxLDL in vascular endothelial cells, and thus preserves NO bioactivity and reverses the inhibitory effects of oxLDL on endothelium-dependent cyclic-GMP production and vessel dilatation. This may be one of the
257
mechanisms by which HDL prevents the progression of atherosclerosis.
References [1] Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79:8–15. [2] Domanski MJ, Borkowf CB, Campeau L, et al. Prognostic factors for atherosclerosis progression in saphenous vein grafts: the postcoronary artery bypass graft (Post-CABG) trial. Post-CABG trial investigators. J Am Coll Cardiol 2000;36:1877–83. [3] Kwiterovich Jr PO. The antiatherogenic role of high-density lipoprotein cholesterol. Am J Cardiol 1998;82:13Q–21Q. [4] Yuhanna IS, Zhu Y, Cox BE, et al. High-density lipoprotein binding to scavenger receptor-BI activates endothelial nitric oxide synthase. Nat Med 2001;7:853–7. [5] Li XA, Titlow WB, Jackson BA, et al. High density lipoprotein binding to scavenger receptor, Class B, type I activates endothelial nitric-oxide synthase in a ceramide-dependent manner. J Biol Chem 2002;277:11058–63. [6] Mineo C, Yuhanna IS, Quon MJ, Shaul PW. High density lipoprotein-induced endothelial nitric-oxide synthase activation is mediated by Akt and MAP kinases. J Biol Chem 2003;278:9142–9. [7] Nofer JR, Levkau B, Wolinska I, et al. Suppression of endothelial cell apoptosis by high density lipoproteins (HDL) and HDL-associated lysosphingolipids. J Biol Chem 2001;276:34480–5. [8] Barter PJ, Nicholls S, Rye KA, et al. Antiinflammatory properties of HDL. Circ Res 2004;95:764–72. [9] Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low-density lipoprotein that increseits atherogenecity. N Engl J Med 1989;320:915–24. [10] Hein TW, Liao JC, Kuo L. oxLDL specifically impairs endotheliumdependent, NO-mediated dilation of coronary arterioles. Am J Physiol Heart Circ Physiol 2000;278:H175–83. [11] Kontush A, Chancharme L, Escargueil-Blanc I, et al. Mildly oxidized LDL particle subspecies are distinct in their capacity to induce apoptosis in endothelial cells: role of lipid hydroperoxides. FASEB J 2003;17:88–90. [12] Robbesyn F, Garcia V, Auge N, et al. HDL counterbalance the proinflammatory effect of oxidized LDL by inhibiting intracellular reactive oxygen species rise, proteasome activation, and subsequent NFkappaB activation in smooth muscle cells. FASEB J 2003;17:743–5. [13] Cominacini L, Rigoni A, Pasini AF, et al. The binding of oxidized low density lipoprotein (ox-LDL) to ox-LDL receptor-1 reduces the intracellular concentration of nitric oxide in endothelial cells through an increased production of superoxide. J Biol Chem 2001;276:13750–5. [14] Chien CT, Yu HJ, Cheng YJ, et al. Reduction in renal haemodynamics by exaggerated vesicovascular reflex in rats with acute urinary retention. J Physiol 2000;526:397–408. [15] Van Hinsbergh VWM, Scheffer MA, Langeler EG. Macro- and microvascular endothelial cells from human tissues. In: Piper HM, editor. Cell culture techniques in heart and vessel research. Berlin, Germany: Springer-Verlag; 1990. p. 178–204. [16] Smirnov VN, Orekhov AN. Smooth muscle cells from adult human aorta. In: Piper HM, editor. Cell culture techniques in heart and vessel research. Berlin, Germany: Springer-verlag; 1990. p. 271–89. [17] Liao JK, Shin WS, Lee WY, Clark SL. Oxidized low-density lipoprotein decreases the expression of endothelial nitric oxide synthase. J Biol Chem 1995;270:319–24. [18] Chien CT, Lee PH, Chen CF, et al. De novo demonstration and co-localization of free-radical production and apoptosis formation in rat kidney subjected to ischemia/reperfusion. J Am Soc Nephrol 2001;12:973–82.
258
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
[19] Fulton D, Gratton JP, McCabe TJ, et al. Regulation of endotheliumderived nitric oxide production by the protein kinase Akt. Nature 1999;399:597–601. [20] Dimmeler S, Fleming I, Fisslthaler B, et al. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature 1999;399:601–5. [21] Galle J, Ochslen M, Schollmeyer P, Wanner C. Oxidized lipoproteins inhibit endothelium-dependent vasodilation. Effects of pressure and high-density lipoprotein. Hypertension 1994;23:556–64. [22] Nofer JR, van der Giet M, Tolle M, et al. HDL induces NOdependent vasorelaxation via the lysophospholipid receptor S1P3. J Clin Invest 2004;113:569–81. [23] Shaul PW. Endothelial nitric oxide synthase, caveolae and the development of atherosclerosis. J Physiol 2003;547:21–33. [24] Beligni M, Lamattina L. Is nitric oxide toxic or protective? Trends Plant Sci 1999;4:299–300. [25] Thiemermann C, Patel NS, Kvale EO, et al. High density lipoprotein (HDL) reduces renal ischemia/reperfusion injury. J Am Soc Nephrol 2003;14:1833–43. [26] Lloberas N, Torras J, Herrero-Fresneda I, et al. Postischemic renal oxidative stress induces inflammatory response through PAF and oxidized phospholipids. Prevention by antioxidant treatment. Faseb J 2002;16:908–10.
[27] Kontush A, Chantepie S, Chapman MJ. Small, dense HDL particles exert potent protection of atherogenic LDL against oxidative stress. Arterioscler Thromb Vasc Biol 2003;23:1881–8. [28] Teiber JF, Draganov DI, La Du BN. Purified human serum PON1 does not protect LDL against oxidation in the in vitro assays initiated with copper or AAPH. J Lipid Res 2004;45:2260–8. [29] Hockerstedt A, Jauhiainen M, Tikkanen MJ. Lecithin/cholesterol acyltransferase induces estradiol esterification in high-density lipoprotein, increasing its antioxidant potential. J Clin Endocrinol Metab 2004;89:5088–93. [30] Viswambharan H, Ming XF, Zhu S, et al. Reconstituted highdensity lipoprotein inhibits thrombin-induced endothelial tissue factor expression through inhibition of RhoA and stimulation of phosphatidylinositol 3-kinase but not Akt/endothelial nitric oxide synthase. Circ Res 2004;94:918–25. [31] Bisoendial RJ, Hovingh GK, Levels JH, et al. Restoration of endothelial function by increasing high-density lipoprotein in subjects with isolated low high-density lipoprotein. Circulation 2003;107: 2944–8. [32] Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA 2003;290:2292–300.
High-density lipoprotein antagonizes oxidized low-density lipoprotein by suppressing oxygen free-radical formation and preserving nitric oxide bioactivity Chii-Ming Lee a , Chiang-Ting Chien b , Po-Yuan Chang a , Mo-Ying Hsieh a , Hsiang-Yiang Jui a , Chau-Song Liau a , Su-Ming Hsu c , Yuan-Teh Lee a,∗ a b
Departments of Internal Medicine, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan Departments of Medical Research, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan c Departments of Pathology, National Taiwan University Hospital, National Taiwan University College of Medicine, No. 7, Chung-Shan South Road, Taipei 100, Taiwan Received 6 December 2004; received in revised form 25 January 2005; accepted 2 March 2005 Available online 10 August 2005
Abstract The antiatherogenic role of high-density lipoprotein (HDL) has been related to its ability to increase the activity of endothelial nitric oxide synthase (eNOS) and to protect low-density lipoprotein (LDL) against oxidative modification. The present study was aimed to determine whether and how HDL antagonizes oxidized LDL (oxLDL) that has been formed and accumulated in circulation. Pre-infusion of rats with HDL effectively prevented oxLDL-induced renal vascular constriction. Consistently, pre-incubation of human saphenous vein endothelial cells with HDL (100 g/ml) reversed the oxLDL-induced suppression of endothelium-dependent cyclic-GMP production in co-cultured smooth muscle cells. However, the changes of Akt phosphorylation and eNOS activity in endothelial cells in response to lipoprotein treatments under our assay condition were not significant. Intriguingly, pretreatment of human umbilical vein endothelial cells with HDL (50 g/ml) for only 30 s effectively reduced the level of free radicals generated by oxLDL or H2 O2 . In kidneys of living rats, renal arterial infusion of oxLDL greatly enhanced ischemia/reperfusion-induced free radicals, which could be attenuated by HDL pretreatment. We conclude that HDL may antagonize oxLDL on endothelial function through an Akt-independent pathway in which HDL preserves nitric oxide bioactivity by attenuating oxLDL-triggered free radical generation. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: High-density lipoprotein; Oxidized low-density lipoprotein; Nitric oxide; Endothelial cells; Free radicals
1. Introduction Decreased high-density lipoprotein (HDL) plasma levels represent the major risk factor associated with coronary atherosclerosis progression [1,2]. Classically, the atheroprotective mechanism of HDL is thought to be related to its role in reverse transport of cholesterol from peripheral tissues to the liver [3]. Recent studies indicate that ∗
Corresponding author. Fax: +886 2 2395 9911. E-mail address: [email protected] (Y.-T. Lee).
0021-9150/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2005.03.029
endothelium is a key target of HDL action. It has been demonstrated that HDL stimulates endothelial nitric-oxide synthase (eNOS) activity through the scavenger receptor SR-BI in a ceramide-dependent manner [4,5] or through lysophospholipid receptor S1P3 binding and Akt activation [6,7]. NO produced by eNOS in the endothelium regulates blood flow, inflammation, and platelet aggregation, and consequently its disruption during endothelial dysfunction can impair endothelium-dependent vasorelaxation and encourage the formation of atherosclerotic lesions and thrombi.
252
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
The antiatherogenic action of HDL may also be partially ascribed to its effect to protect LDL from oxidation through HDL-associated enzymes, such as paraoxonase 1 (PON1), lecithin-cholesterol acryltransferase (LCAT), or platelet-activating factor acetylhydrolase (PAF-AH) [8]. Oxidized LDL (oxLDL) has been shown to be the active component in dyslipidemia that plays an important role in atherosclerosis [9]. Attenuation of endothelium-dependent NO-mediated vasodilation is one of the postulated atherogenesis mechanisms of oxLDL [10]. oxLDL per se contains lipid hydroperoxide [11] and is a potent inducer of reactive oxygen species (ROS) in smooth muscle cells (SMCs) [12]. On binding to oxLDL receptor on endothelial cells (ECs), oxLDL may reduce the intracellular concentration of NO via increased production of superoxide [13]. Because HDL exerts anti-oxidant effects and endothelium is a key target of HDL action, the present study was designed to determine, in addition to protect LDL against oxidative modification, whether and how HDL directly blocks the effects of oxLDL on vascular endothelium both in vivo and ex vivo.
2. Methods 2.1. Materials Tetrahydrobiopterin (BH4) was from RBI (Natick, MA), all other chemicals and lipoprotein-deficient human serum were from Sigma Chemical Co. (St. Louis, MO). Type I collagenase and elastase were from Worthington Biochemical Co. (Freehold, NJ, USA). M199 medium, fetal bovine serum, Earle’s balanced salt solution (EBSS) were from Invitrogen (Grand Island, NY). MCDB107 was from JRH Biosciences (Lenex, KS). Endothelial cell growth supplement (ECGS) was from Collaborative Research Inc. (Bedford, MA). Endothelial cell growth factor (ECGF), pronase, and complete protease inhibitors were from Roche (Indianapolis, IN). Cell culture wares were from Corning Inc. (Cambridge MA). Antibodies were from Cell Signaling Technology Inc. (Beverly, MA). Electrochemiluminescence (ECL) reagents and 3Harginine were from Amersham Bioscience (Piscataway, NJ). 2.2. Preparation of LDL, oxLDL, and HDL LDL (1.019–1.063 g/ml) and HDL (1.063–1.210 g/ml) were prepared from normal human plasma by sequential ultracentrifugation in a sodium bromide gradient. OxLDL was prepared by incubation of LDL (1.2 g/l) with 15 mol/l copper chloride at 37 ◦ C for 90–120 min. The extent of lipoprotein oxidation was measured by the thiobarbituric acid reactive substances (TBARS) assay and electrophoretic mobility (Lipogel, Beckman).
2.3. Measurement of regional blood flow rate in vivo The animal care and experimental protocol were in accordance with the guidelines of the National Science Council of the Republic of China. Regional blood flow rate was measured as described previously [14]. Female Wistar rats (200–250 g) were anesthetized with urethane (1 mg/1 g) and underwent tracheotomy. Lipoproteins were infused into the thoracic aorta by a syringe pump (KD Scientific, Holliston, MA). A pressure transducer (Gould-Statham, Quincy, MA) was placed in the left carotid artery for continuous monitoring of blood pressure (BP). Electromagnetic flow meter probes (Transonic System Inc., Ithaca, NY) were placed on the left renal artery and abdominal aorta for the detection of blood flow rates. Renal vascular resistance was generated via dividing BP by the renal artery flow rate. 2.4. Cell culture Human saphenous veins (HSVs) obtained from patients undergoing coronary artery bypass surgery were incubated with 0.1% collagenase at 37 ◦ C for 25 min [15]. After ECs were scratched from the luminal surface of the HSVs, the remaining tissues were torn in strips and further digested at 37 ◦ C with 0.2% elastase for 2 h followed with 0.3% collagenase for 3–5 h [16]. SMCs were collected by centrifugation. Both types of cells were cultured in M199 medium supplemented with 20% human serum, 5 U/ml heparin, 100 IU/ml penicillin, and 0.1 mg/ml streptomycin except that EC culture required addition of 100 g/ml ECGS. Cells from passages three to eight were used for subsequent experiments. Primary human umbilical vein endothelial cells (HUVECs) were cultured in medium MCDB107 containing 2% fetal bovine serum, a fibroblast growth factor—enriched fraction of porcine brain extract at 1 g/ml, 10 g/l ECGF, and 50 U/l heparin. Cells between passages two and four were used for subsequent experiments. 2.5. Measurement of cyclic-GMP production in SMCs co-cultured with ECs A co-culture model using buoyant chambers containing HSV ECs in Transwell® cell culture inserts (1 × 105 /4.9 cm2 /insert) and HSV SMCs in six-well plates (1.3 × 105 /9.4 cm2 /well) was included for the analysis of cyclic-GMP content. ECs were first incubated in EBSS containing 1 mmol/l l-arginine/62 mol/l BH4 , and treated with either oxLDL (100 g/ml) or HDL (100 g/ml), or both, at 37 ◦ C for 30 min. The treatment was stopped by washing with EBSS, and the Transwells® containing ECs were placed into the six-well plates containing SMCs. After cocultured with ECs in EBSS containing 1 mmol/l 3-isobutyl1-methlxanthine (IBMX) for 30 min, SMCs were collected and subjected to enzyme-linked immunoassay of cyclic-GMP
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
according to the manufacturer’s instructions (Cayman, Ann Arbor, MI).
253
pronase treatment were resolved by SDS-PAGE followed by coomassie blue staining to confirm the complete digestion of proteins.
2.6. Western blot analysis of Akt and eNOS HUVECs were treated with 50 g/ml of various lipoproteins for 5 and 30 min. Cells were lyzed in a buffer containing 0.5% Triton-100, 0.3 mol/l sodium chloride, 0.025 mol/l sodium phosphate pH 7.4, 20 mmol/l octyglucoside, complete protease inhibitors, 10 mmol/l sodium pyrophosphate, 100 mmol/l sodium fluoride, and 2 mmol/l sodium orthovanadate. Fifteen micrograms of cell lysate proteins were resolved on SDS-polyacrylamide gel electrophoresis followed by immunoblot analysis with antibodies to Akt, phospho-Ser473 Akt, eNOS, and phospho-Ser1177 -eNOS. The signal was detected by ECL. 2.7. Measurement of eNOS activity Confluent HUVECs cultured in six-well dishes were incubated in lipoprotein-deficient medium composed of medium M199 supplemented with 100 g/ml ECGS, 5 U/ml heparin, 62 mol/l BH4 , and 20% lipoprotein-deficient human serum for 24 h. After incubation with oxLDL (100 g/ml), HDL (100 g/ml), or both for 30 min, ECs were labeled with 3 H-arginine, and the radioactivity of 3 H-citrulline, the downstream product of arginine by eNOS, was detected by scintillation counting [17]. 2.8. Measurement of chemiluminescence (CL) in cultured cells HUVECs were cultured to 90% confluence in a 35 mm dish as described above and shift to serum-free, phenol redfree Dulbecco’s Modified Eagle Medium (DMEM, Invitrogen 11054-020) immediately before CL measurements. HUVECs were treated with 50 g/ml of various lipoproteins, 10 mol/l of sphingosine-1-phosphate (S1P) or sphingosylphosphorylcholine (SPC), or H2 O2 (100 M) for 30 s, and then placed in a dark chamber of the Chemiluminescence Analyzing System (Tohoku Electronic Industrial Co., Sendai, Japan). This system contains a photon detector (Model CLD110) that is able to detect as low as 10−15 W of radiant energy and a CL counter (Model CLC-10). Photon emission from the sample was counted at 10 s intervals at 37 ◦ C and atmospheric conditions. At the 100 s time point, 50 mol/l luminol was injected into the medium and the CL in the sample was continuously measured for a total of 600 s. The total amount of CL was calculated by integrating the area under the curve and subtracting it from the background level. To inhibit Akt activity, HUVECs were pre-incubated with wortmannin (10−7 M) for 30 min prior to lipoprotein treatment. For breaking down protein components, HDL preparation was treated with 1 mg/ml of pronase, a nonspecific protease, at 37 ◦ C for 5 h according to manufacturer’s instruction. Aliquots of HDL preparation (1 g/ml) before and after
2.9. In vivo measurement of chemiluminescence (CL) in rat kidneys with ischemia/reperfusion injury CL measurement in rats was performed as described previously [18]. Briefly, after laparotomy, the left renal artery was cannulated for lipoprotein infusion by introduction of a length of stretched PE10 tubing from the left femoral artery via the aorta. A catheter was placed in the left femoral vein for continuous infusion of the CL probe, 2-methyl-6-[4-methoxyphenyl]-3,7-dihydroimidazo[1,2-a]-pyrazin-3-one hydrochloride (MCLA), which is a Cypridina luciferin analogue that is highly specific and sensitive to superoxide anion (O2 •− ) and singlet oxygen (1 O2 ). The rat was then placed on its right side in a modified dark chamber of the Chemiluminescence Analyzing System described in Section 2.8 with the left kidney positioned under the reflector. After lipoprotein infusion, the left renal artery was clamped for 15 min and then reperfused. The MCLAenhanced CL signal from the kidney surface was recorded before and after initiation of ischemia/reperfusion. Shamoperated (control) animals underwent similar operative procedures without occlusion of the renal artery. 2.10. Statistics Data are expressed as mean ± S.D. Data were analysed by Student’s t test or ANOVA. A value of P < 0.05 was considered statistically significant.
3. Results 3.1. HDL could prevent regional artery constriction induced by oxLDL in rats We first examined the effects of oxLDL and HDL on vascular tone in rats. As shown in Fig. 1A, continuous infusion of oxLDL into rat abdominal aorta at increasing rates from 0.1 to 1.0 mg/h gradually reduced renal artery blood flow by 22.5 ± 0.6% to 45.6 ± 14.3% (P < 0.05, Fig. 1C). Neither the blood flow in aorta nor of the systemic arterial pressure measured at carotid artery had remarkable change. The renal vascular resistance was thus raised significantly in response to oxLDL infusion in a dose dependent manner. On the contrary, the regional blood flow of renal artery was slightly elevated, although not statistically significant, in response to the increase of HDL infusion rates from 0.4 to 4.0 mg/h (Fig. 1B). More importantly, priming the vessel by HDL infusion at a concentration between 3.3 and 33 g/ml in renal artery (calculated by dividing HDL infusion rate with renal artery flow) prevented a subsequent oxLDL-induced reduction of renal artery blood flow (Fig. 1C).
254
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
HSV ECs with HDL (100 g/ml, 30 min) per se had no effect on the cyclic-GMP contents in SMCs. However, simultaneously treatment of ECs with HDL plus oxLDL partially reversed oxLDL induced cyclic-GMP reduction (Fig. 2). 3.3. Lipoproteins at low concentrations have no significant effects on Akt-mediated eNOS activation It has been demonstrated recently that HDL, at relatively high concentration (1 mg/ml), induces Akt phosphorylation at Ser473 [7], which in turn activates eNOS in HUVECs [19,20]. To elucidate whether HDL, at the concentration in our setting, antagonizes oxLDL through Akt activation, we examined phosphorylation of Akt and eNOS with phosphospecific antibodies. Treatment of HUVECs with HDL in the range of 0–100 g/ml for up to 30 min did not enhance Akt phosphorylation (Fig. 3A and B). Treatment of HUVECs with 50 g/ml of oxLDL, or HDL plus oxLDL for either 5 or 30 min did not result in significant changes in Akt and eNOS phosphorylation (Fig. 3C). Furthermore, the activity of eNOS in lipoprotein (100 g/ml) treated HUVECs, shown as the l-arginine to l-citrulline conversion rates (Fig. 3D), were similar (P > 0.05) in all groups. These data suggest that the effects of lipoproteins on EC-dependent cyclic-GMP production, shown in Fig. 2, could not be exclusively explained by either Akt activation or the substantial differences in eNOS activity. 3.4. HDL at low concentration partially reversed oxLDL- and H2 O2 -induced oxygen free radical generation in ECs It has been reported that oxLDL induced oxygen free radical formation in vascular endothelial cells may be one of the pathogenic mechanisms of atherosclerosis [13]. To test whether HDL attenuates oxLDL induced ROS in ECs, we Fig. 1. Effects of lipoproteins on renal artery flow. The rat thoracic aorta was infused sequentially with normal saline (0–20 min), LDL (20–35 min), and increasing amounts of oxLDL (35–100 min, expressed as infusion rate, mg/h) (A) or HDL (15–80 min) followed by oxLDL (80–150 min) (B) as indicated on the horizontal axis. Original recordings of renal artery flow, abdominal aortic flow, BP, and resistance were shown. (C) Percent change of renal artery flow from baseline following oxLDL infusion (n = 2 each). () No pre-treatment; () pretreatment with HDL, * P < 0.05 vs. baseline.
3.2. HDL reverses the suppressive effects of oxLDL on endothelium-dependent cyclic-GMP production in SMCs To exam whether the protective effect of HDL on oxLDL induced vessel constriction is NO-dependent, we measured the contents of cyclic-GMP, the second messenger of NO, in HSV SMCs co-cultured with HSV ECs. Incubation of ECs with 100 g/ml oxLDL for 30 min caused a two-thirds reduction in cyclic-GMP amounts in SMCs as compared with that in untreated ECs/SMCs co-culture (P < 0.05). Incubation of
Fig. 2. Effects of lipoproteins on EC-dependent cyclic-GMP production in SMCs. HSV ECs were pre-treated with various types of lipoproteins (100 g/ml) as indicated for 30 min, and then co-cultured with HSV SMCs for additional 30 min. The contents of cyclic-GMP in HSV SMCs were measured by enzyme-linked immunoassay and expressed as pmol/g of protein of SMC homogenate (n = 2). * P < 0.05 vs. untreated, # P < 0.05 vs. oxLDL.
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
255
Fig. 3. Effects of lipoproteins on Akt and eNOS activation. ECs were treated with the indicated lipoproteins (50 g/ml in C and 100 g/ml in D), and then subjected to Western blot analysis (A and C) or l-citrulline assay (D). (B) Densitometric analysis of Akt phosphorylation expressed as fold increase over the basal level in untreated cells (n = 3). In (D), enzyme activity is expressed as fold increase over the control (n = 2). * P < 0.05 vs. control.
analyzed the luminol-activated CL in HUVECs. Treatment of ECs with 50 g/ml oxLDL for 30 s remarkably increased CL signal as compared to that in cells treated with PBS. Neither native LDL nor HDL had a significant effect on ROS generation (Fig. 4A and C). Interestingly, pretreatment of ECs with HDL (50 g/ml) for only 30 s prior to oxLDL exposure reduced oxLDL augmented CL signal to 41.8 ± 13.0% (P < 0.01, Fig. 4C). HDL also reduced free radicals induced by H2 O2 , one of the representative oxidative reagents, to 45.6 ± 7.9% (Fig. 4D). To determine whether HDL-associated proteins are responsible for the attenuation of CL signals, we treated HDL with pronase (Fig. 4B). As shown in Fig. 4C, breaking down the protein components of HDL did not affect the inhibitory effect of HDL on oxLDL augmented ROS formation (45.9 ± 2.1%, P < 0.01). The attenuation of HDL on oxLDL induced CL signal was not blocked by wortmannin which inactivates Akt through inhibition of phosphatidylinositol 3-kinase (42.1 ± 1.5%, Fig. 4C). Furthermore, two HDL-associated lysophospholipids, S1P and SPC, which have been shown to activate Akt and eNOS [23], did not reduce oxLDL-induced free radicals. These data suggest that attenuation of free radicals by HDL may be independent of Akt. 3.5. HDL partially reversed oxLDL-enhanced free radical generation in rat kidneys with ischemia/reperfusion injury To determine whether HDL diminishes oxLDL-enhanced ROS in vivo, we investigated ischemia/reperfusion-induced
oxidative stress in rat kidneys. As shown in Fig. 5, preinfusion of rat kidneys with oxLDL prior to the induction of ischemia dramatically enhanced ROS formation in the reperfused kidney. The augmentation of ROS formation could be partially ameliorated by HDL treatment before oxLDL infusion.
4. Discussion In this report, we demonstrated the vascular protective effects of HDL in vivo. The increase of renal vascular resistance of living rats correlates with the oxLDL infusion rate (Fig. 1). By priming of the vascular beds with HDL (3.3–33 g/ml), the oxLDL-induced elevation of renal vascular resistance was abolished. This is comparable with the findings in organ culture that the inhibition of vasodilatation by oxLDL was prevented by the presence of HDL [21] and that the endothelin-induced vasoconstriction was reversed by HDL injection [22]. Various cholesterol-independent mechanisms of HDL on vascular protection have been postulated. HDL, on binding to SR-BI of ECs, may stimulate eNOS through a ceramidedependent pathway [5] or may prevent oxLDL from decreasing the capacity of eNOS activation by preserving the cholesterol concentration in caveolae [23]. The NO-dependent vasorelaxation effect of HDL has also been reported to be mediated via the lysophospholipid receptor S1P3 and subsequent activation of Akt and eNOS [22]. In this report, we demonstrated that oxLDL and HDL at the concentra-
256
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
Fig. 4. Effects of lipoproteins on luminol-enhanced chemiluminescence (CL) signals in HUVECs. HUVECs were treated with PBS, 50 g/ml of native LDL, HDL, or oxLDL, 10 mol/l of S1P or SPC, or 100 mol/l of H2 O2 for 30 s, and then subjected to CL measurement. In the experiments of HDL + oxLDL, S1P + oxLDL, SPC + oxLDL, and HDL + H2 O2 cells were treated with HDL, S1P, or SPC for 30 s followed by oxLDL or H2 O2 for additional 30 s before CL measurement. (A) Representative results of original recordings. (B) HDL preparation was treated with (+) or without (−) pronase (1 mg/ml) at 37 ◦ C for 5 h and resolved on 7.5% SDS-PAGE followed by coomassie blue staining. (C) Percent of oxLDL increased CL counts from baseline (n = 5). Total CL counts were calculated by integrating of the area under the curve. PBS-induced CL counts was considered as the baseline level of free radicals. (+) Wt, HUVECs were pre-incubated with wortmannin (10−7 mol/l) for 30 min. The increase of oxLDL-induced CL counts from baseline was arbitrarily set as 100%. * P < 0.01 vs. oxLDL. (D) Percent of H2 O2 -increased CL counts from baseline. # P < 0.01 vs. H2 O2 .
Fig. 5. Effects of lipoproteins on free radical generation in rat kidneys with ischemia/reperfusion injury. The left side of rat renal artery was infused with saline, oxLDL (1 mg/ml/h, 30 min), or HDL (2 mg/ml/h, 30 min) followed by oxLDL (1 mg/ml/h, 30 min). Free radical formation in response to ischemia/reperfusion injury was determined by MCLA-induced CL signal. Control, sham-operated rats without ischemia. Representative experiments were superimposed for comparison.
tions between 0 to 100 g/ml did not induce significant Akt or eNOS activation, based on either immunoblotting or lcitrulline formation assay (Fig. 3). However, at such low concentration, HDL effectively reverses the inhibitory effects of oxLDL on endothelium-dependent cyclic-GMP production in the co-cultured SMC (Fig. 2). These findings indicate that HDL may reverse oxLDL-induced impairment of NO signaling through an Akt-independent pathway. The rapidity of inhibition or protection of lipoproteins on NO signaling/cyclic-GMP production between ECs and SMCs precludes the possibility of transcriptional regulation, but favors a biochemical reaction. Recent data suggest that that oxLDL induces NF-B activation in SMCs and ECs, presumably through ROS generation [12,13]. In this report, free radicals were induced within seconds by addition of oxLDL to the cultured ECs and could be attenuated by HDL in a PI3-kinase independent manner. Because only the ECs were exposed to lipoproteins in either the co-culture system (Fig. 2) or CL detection (Fig. 4), it is unlikely that the guanylyl cyclase of vascular SMCs is the direct target of oxLDL or HDL; rather, NO released from ECs may be inactivated by
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
reacting with oxLDL-induced superoxide [13,24] and consequently, reduces downstream cGMP production in SMCs. Attenuation of oxLDL-enhanced free radical formation by HDL, therefore, preserves NO bioactivity. The hypothesis that HDL protects vessels through the attenuation of free radical formation is further supported in the ischemia-reperfusion model in vivo. It has been reported that HDL reduces renal ischemia-reperfusion injury [25]. The effectiveness of vitamin C in the prevention of postischemic inflammatory response indicates oxidative stress may be the pathogenesis mechanism [26]. In this report, pretreatment with HDL decreased the oxLDL-augmented free radical generation in the ischemia-reperfusion model of rat kidneys (Fig. 5). This finding may explain the prevention of oxLDL-induced renal arterial constriction by priming of the vascular beds with HDL. It is not clear what are the active components of HDL that attenuate oxLDL-augmented free radical. The HDLassociated enzymes, such as PON1, LCAT, and PAF-AH, exert potent antioxidative activity. However, it has been reported previously that small, dense HDL might inactivate oxidized LDL lipids by both enzymatic and nonenzymatic mechanisms [27]. Furthermore, a highly purified PON1 fraction from human serum did not protect LDL against oxidation [28]. In this report, we demonstrated that HDL reduced free radicals triggered by H2 O2 which is not a known substrate for LCAT and PAF-AH. More importantly, complete removal of the protein components of HDL preparation by pronase did not affect the antagonizing effect of HDL on oxLDLinduced free radicals (Fig. 4C). The lipid contents of HDL, such as S1P and SPC, that have been shown to activate Akt and eNOS [22], did not attenuate the oxLDL-induced free radicals (Fig. 4C). A recent report indicates that HDL is protected by its associated estradiol ester from oxidation in vitro [29]. A reconstituted HDL composed of aopA1 and phosphatidylcholine inhibits the expression of tissue factor, a thrombogenic molecule, in HUVECs without the activation of Akt/eNOS [30]. Whether these lipid components of HDL exert their antioxidant or antiatherothrombotic activities through scavenging free radicals in vivo remains to be elucidated. Therapeutic interventions in elevation of HDL plasma level or in mimicking its beneficial effects have become the promising strategies in the treatment and prevention of atherosclerosis [31,32]. An understanding of the atheroprotective mechanisms of HDL would be important for the rational utilization or development of pharmaceutic agents. The rapid attenuation of ROS by HDL, shown in this report, might be a surrogate marker of therapeutic efficacy in addition to the classical quantitative analysis of total HDL content or the subfractionation of HDL2 and HDL3 . We conclude that HDL decreases ROS formation augmented by oxLDL in vascular endothelial cells, and thus preserves NO bioactivity and reverses the inhibitory effects of oxLDL on endothelium-dependent cyclic-GMP production and vessel dilatation. This may be one of the
257
mechanisms by which HDL prevents the progression of atherosclerosis.
References [1] Gordon DJ, Probstfield JL, Garrison RJ, et al. High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79:8–15. [2] Domanski MJ, Borkowf CB, Campeau L, et al. Prognostic factors for atherosclerosis progression in saphenous vein grafts: the postcoronary artery bypass graft (Post-CABG) trial. Post-CABG trial investigators. J Am Coll Cardiol 2000;36:1877–83. [3] Kwiterovich Jr PO. The antiatherogenic role of high-density lipoprotein cholesterol. Am J Cardiol 1998;82:13Q–21Q. [4] Yuhanna IS, Zhu Y, Cox BE, et al. High-density lipoprotein binding to scavenger receptor-BI activates endothelial nitric oxide synthase. Nat Med 2001;7:853–7. [5] Li XA, Titlow WB, Jackson BA, et al. High density lipoprotein binding to scavenger receptor, Class B, type I activates endothelial nitric-oxide synthase in a ceramide-dependent manner. J Biol Chem 2002;277:11058–63. [6] Mineo C, Yuhanna IS, Quon MJ, Shaul PW. High density lipoprotein-induced endothelial nitric-oxide synthase activation is mediated by Akt and MAP kinases. J Biol Chem 2003;278:9142–9. [7] Nofer JR, Levkau B, Wolinska I, et al. Suppression of endothelial cell apoptosis by high density lipoproteins (HDL) and HDL-associated lysosphingolipids. J Biol Chem 2001;276:34480–5. [8] Barter PJ, Nicholls S, Rye KA, et al. Antiinflammatory properties of HDL. Circ Res 2004;95:764–72. [9] Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low-density lipoprotein that increseits atherogenecity. N Engl J Med 1989;320:915–24. [10] Hein TW, Liao JC, Kuo L. oxLDL specifically impairs endotheliumdependent, NO-mediated dilation of coronary arterioles. Am J Physiol Heart Circ Physiol 2000;278:H175–83. [11] Kontush A, Chancharme L, Escargueil-Blanc I, et al. Mildly oxidized LDL particle subspecies are distinct in their capacity to induce apoptosis in endothelial cells: role of lipid hydroperoxides. FASEB J 2003;17:88–90. [12] Robbesyn F, Garcia V, Auge N, et al. HDL counterbalance the proinflammatory effect of oxidized LDL by inhibiting intracellular reactive oxygen species rise, proteasome activation, and subsequent NFkappaB activation in smooth muscle cells. FASEB J 2003;17:743–5. [13] Cominacini L, Rigoni A, Pasini AF, et al. The binding of oxidized low density lipoprotein (ox-LDL) to ox-LDL receptor-1 reduces the intracellular concentration of nitric oxide in endothelial cells through an increased production of superoxide. J Biol Chem 2001;276:13750–5. [14] Chien CT, Yu HJ, Cheng YJ, et al. Reduction in renal haemodynamics by exaggerated vesicovascular reflex in rats with acute urinary retention. J Physiol 2000;526:397–408. [15] Van Hinsbergh VWM, Scheffer MA, Langeler EG. Macro- and microvascular endothelial cells from human tissues. In: Piper HM, editor. Cell culture techniques in heart and vessel research. Berlin, Germany: Springer-Verlag; 1990. p. 178–204. [16] Smirnov VN, Orekhov AN. Smooth muscle cells from adult human aorta. In: Piper HM, editor. Cell culture techniques in heart and vessel research. Berlin, Germany: Springer-verlag; 1990. p. 271–89. [17] Liao JK, Shin WS, Lee WY, Clark SL. Oxidized low-density lipoprotein decreases the expression of endothelial nitric oxide synthase. J Biol Chem 1995;270:319–24. [18] Chien CT, Lee PH, Chen CF, et al. De novo demonstration and co-localization of free-radical production and apoptosis formation in rat kidney subjected to ischemia/reperfusion. J Am Soc Nephrol 2001;12:973–82.
258
C.-M. Lee et al. / Atherosclerosis 183 (2005) 251–258
[19] Fulton D, Gratton JP, McCabe TJ, et al. Regulation of endotheliumderived nitric oxide production by the protein kinase Akt. Nature 1999;399:597–601. [20] Dimmeler S, Fleming I, Fisslthaler B, et al. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature 1999;399:601–5. [21] Galle J, Ochslen M, Schollmeyer P, Wanner C. Oxidized lipoproteins inhibit endothelium-dependent vasodilation. Effects of pressure and high-density lipoprotein. Hypertension 1994;23:556–64. [22] Nofer JR, van der Giet M, Tolle M, et al. HDL induces NOdependent vasorelaxation via the lysophospholipid receptor S1P3. J Clin Invest 2004;113:569–81. [23] Shaul PW. Endothelial nitric oxide synthase, caveolae and the development of atherosclerosis. J Physiol 2003;547:21–33. [24] Beligni M, Lamattina L. Is nitric oxide toxic or protective? Trends Plant Sci 1999;4:299–300. [25] Thiemermann C, Patel NS, Kvale EO, et al. High density lipoprotein (HDL) reduces renal ischemia/reperfusion injury. J Am Soc Nephrol 2003;14:1833–43. [26] Lloberas N, Torras J, Herrero-Fresneda I, et al. Postischemic renal oxidative stress induces inflammatory response through PAF and oxidized phospholipids. Prevention by antioxidant treatment. Faseb J 2002;16:908–10.
[27] Kontush A, Chantepie S, Chapman MJ. Small, dense HDL particles exert potent protection of atherogenic LDL against oxidative stress. Arterioscler Thromb Vasc Biol 2003;23:1881–8. [28] Teiber JF, Draganov DI, La Du BN. Purified human serum PON1 does not protect LDL against oxidation in the in vitro assays initiated with copper or AAPH. J Lipid Res 2004;45:2260–8. [29] Hockerstedt A, Jauhiainen M, Tikkanen MJ. Lecithin/cholesterol acyltransferase induces estradiol esterification in high-density lipoprotein, increasing its antioxidant potential. J Clin Endocrinol Metab 2004;89:5088–93. [30] Viswambharan H, Ming XF, Zhu S, et al. Reconstituted highdensity lipoprotein inhibits thrombin-induced endothelial tissue factor expression through inhibition of RhoA and stimulation of phosphatidylinositol 3-kinase but not Akt/endothelial nitric oxide synthase. Circ Res 2004;94:918–25. [31] Bisoendial RJ, Hovingh GK, Levels JH, et al. Restoration of endothelial function by increasing high-density lipoprotein in subjects with isolated low high-density lipoprotein. Circulation 2003;107: 2944–8. [32] Nissen SE, Tsunoda T, Tuzcu EM, et al. Effect of recombinant ApoA-I Milano on coronary atherosclerosis in patients with acute coronary syndromes: a randomized controlled trial. JAMA 2003;290:2292–300.