Reconstituted High-Density Lipoproteins Inhibit the Acute Pro-Oxidant and Proinflammatory Vascular Changes Induced by a Periarterial Collar
http://www.100md.com
循环学杂志 2005年第3期
the Heart Research Institute, Sydney (S.J.N., B.C., K.-A.R., P.J.B., S.B.)
the Department of Medicine, University of Adelaide, Adelaide (S.J.N.)
the Howard Florey Institute, University of Melbourne, Melbourne (G.J.D., G.R.D.)
the Department of Medicine, University of Sydney, Sydney (P.J.B.)
the Department of Pathology, University of Sydney, Sydney (S.B.), Australia.
Abstract
Background— HDLs have antiinflammatory and antioxidant properties in vitro. This study investigates these properties in vivo.
Methods and Results— Chow-fed, normocholesterolemic New Zealand White rabbits received a daily infusion of (1) saline, (2) reconstituted HDL (rHDL) containing 25 mg apolipoprotein (apo) A-I and 50 mg of either 1-palmitoyl-2-linoleoyl phosphatidylcholine (PLPC) or 1,2-dipalmitoyl phosphatidylcholine (DPPC), (3) 25 mg lipid-free apoA-I, or (4) 50 mg of either PLPC-small unilamellar vesicles (SUVs) or DPPC-SUVs on each of 3 consecutive days. Nonocclusive carotid periarterial collars were implanted after the second dose of treatment. Forty-eight hours after insertion of the collars, the arteries were removed and analyzed for the presence of reactive oxygen species, the infiltration of neutrophils, and the expression of adhesion proteins and chemokines. Insertion of the periarterial collar induced a 4.1-fold increase in presence of vascular wall reactive oxygen species. This effect was completely abolished in the animals infused with rHDL. The periarterial collar was associated with a dense infiltration of the arterial wall by polymorphonuclear leukocytes. This infiltration was inhibited by 73% to 94% in the animals infused with rHDL, by 75% in the animals infused with lipid-free apoA-I, and by 51% to 65% in animals infused with SUVs. There were no significant differences between the effects of PLPC and DPPC in either the rHDL or SUVs. Endothelial expression of vascular cell adhesion molecule-1, intercellular adhesion molecule-1, and monocyte chemoattractant protein-1 was also increased by the collar insertion and inhibited by rHDL, lipid-free apoA-I, and, to a lesser extent, also by the SUVs.
Conclusions— Infusion of rHDL, apoA-I, and phospholipid-SUVs inhibits the early pro-oxidant and proinflammatory changes induced by a periarterial collar in normocholesterolemic rabbits.
Key Words: lipoproteins ; inflammation ; cholesterol ; antioxidants ; endothelium
Introduction
High-density lipoproteins (HDLs) have several functions that may contribute to their ability to protect against atherosclerosis. The best known of these relates to their role in promoting the efflux of cholesterol from macrophages in the artery wall. However, HDLs also have antioxidant, antithrombotic, and antiinflammatory functions that may contribute to their protective properties. The present study is concerned with the antiinflammatory effects of HDL in vivo.
It has been shown previously in studies conducted in vitro that HDLs inhibit the cytokine-induced expression of vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and E-selectin in human umbilical vein endothelial cells (HUVECs) growing in tissue culture.1 HDLs also inhibit binding of monocytes2 and neutrophils3 to endothelial cells growing in culture. There is evidence that the antiinflammatory properties of HDL also operate in vivo, although in most cases this has been demonstrated in a setting of hypercholesterolemia and atherosclerosis. For example, intravenous infusion of reconstituted HDL (rHDL) reduces the vivo expression of endothelial adhesion molecules induced by insertion of carotid periarterial cuffs in cholesterol-fed, apoE-knockout mice.4 In another study of apoE-knockout mice, the increase in HDL concentration accompanying an overexpression of the human apoA-I gene reduced macrophage accumulation in the aortic root by more than 3-fold.5 This was associated with a reduced in vivo oxidation of ;-VLDL, lower ICAM-1 and VCAM-1 expression, and diminished ex vivo leukocyte adhesion.
However, there are also examples of in vivo antiinflammatory effects of HDL in the absence of hypercholesterolemia and atherosclerosis. For example, injection of rHDL inhibits the development of a local inflammatory infiltrate after the subcutaneous administration of interleukin-1 in a porcine model.6 Also, in studies of experimental stroke in rats, pretreatment with rHDL significantly and substantially reduced the brain necrotic area in a process possibly related to an rHDL-induced reduction in levels of reactive oxygen species (ROS).7 Furthermore, in a study of hemorrhagic shock in rats, the resulting multiple organ dysfunction syndrome was largely abolished by a single injection of human HDL given 90 minutes after the hemorrhage and 1 minute before resuscitation. In that model, injection of HDL prevented both the severe disruption of tissue architecture and the extensive cellular infiltration into the affected tissues.8
There is also circumstantial evidence that HDLs have acute antiinflammatory effects in vivo in humans. In one study, a single intravenous infusion of rHDL into hypercholesterolemic humans normalized endothelium-dependent vasodilation, possibly by increasing NO bioavailability.9 In a second human study, a single injection of rHDL corrected the endothelial dysfunction associated with low levels of HDL in ABCA1 heterozygotes.10
Thus, there is mounting evidence that HDLs have acute antiinflammatory effects that are extremely rapid and possibly unrelated to their cholesterol-transporting function. In the present study, we have used a model of acute arterial inflammation in normocholesterolemic, nonatherosclerotic rabbits to demonstrate in vivo for the first time that injection of rHDL markedly inhibits both the infiltration of neutrophils into the artery wall and the subsequent generation of ROS.
Methods
Animals
Male New Zealand White rabbits (Nanowie Small Animal Production Unit, Modewarre, Australia) weighing approximately 3 kg were maintained on a normal laboratory chow diet throughout the study. All procedures were approved by the Howard Florey Institute Animal Ethics Committee (protocol 02-009).
Isolation of Lipid-Free ApoA-I
HDLs were isolated from pooled samples of rabbit plasma (Quality Farms of Australia, Lara, Australia) by sequential ultracentrifugation in the 1.063- to 1.21-g/mL density range. The HDLs were lyophilized and delipidated.11 ApoA-I was isolated by chromatography12 on a Q-Sepharose Fast Flow column (Amersham Biosciences) attached to a fast performance liquid chromatography system (Amersham Biosciences). The purified apoA-I was lyophilized and stored at –20°C until used. Lyophilized apoA-I was reconstituted in 3 mol/L guanidine hydrochloride and dialyzed against endotoxin-free PBS (pH 7.4; Sigma) containing 0.2 g/L KH2PO4, 0.2 g/L KCl, 8 g/L NaCl, and 1.15 g/L Na2PO4 before being infused or used to prepare rHDL. The concentration of apoA-I was determined by use of an immunoturbidimetric assay.13
Preparation of Reconstituted HDL
Discoidal rHDL containing apoA-I complexed to either 1-palmitoyl-2-linoleoyl phosphatidylcholine (PC) (PLPC, Sigma) or 1,2-dipalmitoyl PC (DPPC, Sigma), in a molar ratio of PC to apoA-I of 200:1, were prepared by use of the cholate dialysis method.14 The resulting rHDLs were dialyzed extensively against endotoxin-free PBS before use. Protein15 and phospholipid16 concentrations were determined by enzymatic assay. The rHDLs were subjected to agarose gel electrophoresis17 and 3% to 40% nondenaturing gradient gel electrophoresis to determine their surface charge and Stokes’ diameter, respectively.
Preparation of Phospholipid Small Unilamellar Vesicles
Small unilamellar vesicles (SUVs) containing either PLPC or DPPC and BHT (molar ratio of PC to BHT, 10:1) were prepared in PBS as described.18
Administration of Reconstituted HDL
Twenty-nine rabbits were randomly allocated to receive treatment with saline (n=6), PLPC-rHDL (n=4), DPPC-rHDL (n=4), lipid-free apoA-I (n=5), PLPC SUVs (n=5), or DPPC (n=5). Each treatment of rHDL and lipid-free apoA-I contained 25 mg of apoA-I (8 mg/kg). Each treatment of rHDL and SUVs contained 50 mg of PC (16 mg/kg). The treatments were administered via a marginal ear vein on each of 3 occasions: 24 hours before collar implantation, directly before the surgical procedure, and 24 hours after the collar implantation. The animals were euthanized 48 hours after the collar implantation.
Collar Implantation
The rabbits were anesthetized with intravenous propofol (5 mg/kg) followed by intramuscular ketamine/xylazine (50/10 mg/kg). Both carotid arteries were exposed surgically and cleared of connective tissue along a 30-mm length. Hollow, nonocclusive Silastic collars (length, 20 mm; internal diameter along bore, 4 mm; internal diameter at ends, 1 mm) were then placed around each artery and held in place with a nylon sleeve. The space inside the collar was filled with sterile saline (0.9%). Muscle, fat, and skin layers were sutured, the wound was dressed with antibiotic, and animals were allowed to recover for 48 hours before being euthanized.
Tissue Harvesting
Forty-eight hours after insertion of the collar, blood was sampled from a marginal ear vein. Animals were then heparinized (1000 U IV) before euthanasia with an overdose of sodium pentobarbitone (90 mg/kg IV). The collared segment of the carotid artery and approximately 10 mm of noncollared artery proximal to the collar were excised and placed in ice-cold Krebs-HEPES buffer (composition, in mmol/L: NaCl 99.0, KCl 4.7, KH2PO4 1.0, MgSO4 1.2, CaCl2 2.5, NaHCO3 25.0, Na-HEPES 20.0, and glucose 11.0, pH 7.4). Collars were removed and arteries cleaned of fat and connective tissue. Three ring sections (3 mm) were cut both from the area enclosed by the collar and from the proximal noncollared segment of artery. One section from each segment was snap-frozen in Tissue-Tek OCT (Sakura) fixative for immunohistochemical analysis. The other sections were used to determine generation of ROS.
Detection of ROS
Levels of ROS in the carotid arteries were measured by 5 μmol/L lucigenin-enhanced chemiluminescence.19 The increased chemiluminescence signal, probably reflecting superoxide, is mediated by increased activity of NADPH oxidase.20 To minimize the possibility that the lucigenin used in the assay may have contributed to the production of superoxide, we used a concentration of lucigenin that was not sufficient to participate in superoxide production by redox cycling.21 Ring segments were incubated for 45 minutes at 37°C in Krebs-HEPES buffer containing diethyldithiocarbamate (DETCA, 3 mmol/L) to irreversibly inactivate endogenous Cu2+/Zn2+ superoxide dismutase. The possibility that HDL may have interfered with the ability of DETCA to inactivate superoxide dismutase was excluded on the grounds that HDLs were not present in the wells during the incubation. Some rings were further treated with NADPH (10 μmol/L), the preferred substrate of NADPH oxidase, which is, in turn, the predominant source of superoxide in rabbit carotid arteries.20 Each segment was then transferred to a separate well of a white, opaque 96-well plate containing 300 μL of 5-μmol/L lucigenin in Krebs-HEPES buffer as well as the appropriate drug treatment. DETCA was excluded from the lucigenin assay solution. The 96-well plate was loaded into a TopCount Single Photon Counter (Packard Bioscience), and photon emission per second was measured (6-second count time per cycle, 12 cycles, 1-minute delay between cycles). Ring segments were dried for 2 to 3 days in a 65°C oven, and the production of ROS was normalized to dry tissue weight (counts per second per milligram).
Immunohistochemistry
Frozen tissues were sectioned in 5-μm slices. A section from the middle of each of the collared and noncollared segments of both carotid arteries was then subjected to immunohistochemical staining. Sections were fixed with methanol/acetone (1:1, vol/vol) at room temperature for 5 minutes. Arterial wall infiltration by inflammatory cells was determined by use of mouse monoclonal antibodies against macrophages (RAM11, DakoCytomation), neutrophils (CD18, Serotag), and lymphocytes (CD43). In addition, the endothelial expression of adhesion proteins and chemokines was determined by use of murine monoclonal antibodies against rabbit VCAM-1 and ICAM-1 (gifts from Dr. M. Cybulsky, University of Toronto), monocyte chemotactic protein (MCP)-1 (a gift from Dr. A. Matsukowa, Kumamoto University), a goat polyclonal antibody against human E-selectin (R&D), endothelial nitric oxide synthase (eNOS) (BD Biosciences), and inducible nitric oxide synthase (iNOS) (BD Biosciences). These were applied and incubated overnight at room temperature. Endothelial integrity was determined by use of a mouse monoclonal antibody against human CD31 (DakoCytomation). Sections were incubated with biotinylated anti-mouse or anti-goat immunoglobulins for 30 minutes and then incubated with alkaline phosphatase-labeled streptavidin solution for 60 minutes. Slides were rinsed in PBS (pH 7.4) after each incubation. Peroxidase activity was revealed by diaminobenzamine. Slides were counterstained with hematoxylin and mounted. These slides were subsequently graded independently by 4 pathologists who were blinded to the treatment status of the animal. The degree of endothelial staining was graded22,23 by use of a scale that incorporated both the strength of staining and the amount of endothelial surface involved (0=no staining, 1=weak staining of less than 50% of endothelium, 2=strong staining of less than 50% or weak staining of more than 50% of endothelium, 3=strong staining between 50% and 99% of endothelium, and 4=strong staining of 100% of endothelium). The collar-induced change in expression was calculated as the difference in mean score between noncollared and collared segments of each individual artery. Infiltration of the arterial wall by neutrophils, demonstrated by CD18 staining, was determined by quantitative immunohistochemistry. Digital micrographs were acquired with an Olympus BX40 microscope, and the percentage of total vessel wall area occupied by positive staining was determined by use of ImagePro Plus 4 (Cybernetics).24
Plasma Analyses
Plasma collected at the commencement of the study and before euthanasia of the animal was stored at –80°C in EDTA until required for analysis. All chemical analyses were performed on a Roche Diagnostics/Hitachi 902 autoanalyzer (Roche Diagnostics GmbH). Triglyceride25 was determined enzymatically. Total cholesterol was determined by use of a Roche Diagnostics kit. HDL cholesterol was determined by enzymatic assay after precipitation of apoB containing lipoproteins with polyethylene glycol.26 ApoA-I concentrations were determined by an immunoturbidimetric assay using a sheep anti-rabbit apoA-I immunoglobulin.13
Data Analysis
All results are expressed as mean±SEM. Statistical comparisons were made by Student t tests and 1-way ANOVA using the statistical program in GraphPad Prism Version 4.0. A value of P<0.05 was considered significant.
Results
Plasma Lipid Concentrations
Plasma concentrations of triglyceride, total and free cholesterol, HDL cholesterol, and apoA-I at the commencement of the study and at the time of euthanasia of the animals are presented in the Table. No significant differences were found between the groups, nor were there significant differences in lipid levels resulting from the infusions of saline, rHDL, lipid-free apoA-I, or PC-SUVs.
Plasma Lipid Profiles From Animals Infused With Saline (n=6), PLPC-rHDL (n=4), DPPC-rHDL (n=4), Lipid-Free apoA-I (n=5), PLPC-SUV (n=5), and DPPC-SUV (n=5) at Baseline and Immediately Before Death
Generation of ROS
Application of the periarterial collar increased the production of ROS in the vessel wall both in the unstimulated state without NADPH (126.9±48.1 and 6.6±2.9 counts · s–1 · mg–1 in the collared and noncollared segments, respectively, P<0.005) and after stimulating by incubation with NADPH (540.3±169.1 and 130.3±42.1 counts · s–1 · mg–1 in the collared and noncollared segments, respectively, P<0.01) (Figure 1). The collar-induced increase in vascular ROS that was observed in both the unstimulated (Figure 1A) and the stimulated (Figure 1B) states was significantly reduced in the vessels isolated from animals infused with rHDL. Infusion of PLPC-rHDL inhibited the collar-induced increase in vascular ROS in the unstimulated state by 92% (10±4.1 and 126.9±48.1 counts · s–1 · mg–1 in the PLPC-rHDL and saline-infused animals, respectively, P<0.03) and in the stimulated state by 85% (82.3±34.4 and 540.3±169.1 counts · s–1 · mg–1 in the PLPC-rHDL and the saline-infused animals, respectively, P<0.001). The collar-induced increase in ROS in the stimulated state was also inhibited by 90% by infusion of DPPC-rHDL (56.7±29.8 and 540.3±169.1 counts · s–1 · mg–1 in the DPPC-rHDL- and saline-treated animals, respectively, P<0.001) (Figure 1B). The effects of infusing PLPC-rHDL and DPPC-rHDL were not significantly different.
Neutrophil Infiltration
The collar-induced neutrophil recruitment was also inhibited 73% by infusion of lipid-free apoA-I (30±9% of the total vessel wall area in the saline-infused rabbits reduced to 8±2.9% in the animals infused with lipid-free apoA-I, P<0.02). This effect of infusing lipid-free apoA-I was not significantly different from that of rHDL (Figure 3).
Infusion of SUVs also reduced the infiltration of neutrophils but appeared to be less effective than either rHDL or lipid-free apoA-I (Figure 3). The collar-induced neutrophil recruitment was 30±9% of the total vessel wall area in the saline-infused rabbits, 14.6±4.2% in the animals infused with the PLPC-SUVs (P=NS compared with saline) and 10.4% in the animals infused with the DPPC-SUVs (P<0.05 compared with saline) (Figure 3). The accumulation of neutrophils in the collared arteries of animals infused with PLPC-SUVs and DPPC-SUVs was not significantly different.
Vascular Adhesion Molecules and Chemokines
Compared with saline, infusion of rHDL inhibited the collar-induced increase in endothelial expression of VCAM-1 (by 53.9% and 54.4% with PLPC-rHDL and DPPC-rHDL, respectively, P<0.005), ICAM-1 (by 50% and 74.6% with PLPC-rHDL, P<0.01, and DPPC-rHDL, P<0.005, respectively), and MCP-1 (by 76.5% with PLPC-rHDL, P<0.01). Although DPPC-rHDL inhibited the collar-induced change in MCP-1 by 50% compared with saline, this did not reach statistical significance (P=0.09). The collar-induced change in MCP-1 did not differ significantly between the PLPC-rHDL and DPPC-rHDL groups. The infusion of rHDL, combining the PLPC and DPPC groups, significantly inhibited the collar-induced change in MCP-1 staining (by 63%, P<0.05). Infusion of rHDL had no measurable effect on the expression of either iNOS or eNOS (results not shown).
Infusion of lipid-free apoA-I had effects similar to those of the rHDL infusions. Compared with saline, infusion of lipid-free apoA-I inhibited the collar-induced increase in expression of VCAM-1 by 40.5% (P<0.01), ICAM-1 by 94.1% (P<0.0005), and MCP-1 by 76.5% (P<0.05).
Infusion of PLPC-SUVs inhibited the collar-induced increase in expression of VCAM-1 by 47.2% (P<0.01) and ICAM-1 by 79.7% (P<0.01), but the 55.6% reduction in MCP-1 expression was not statistically significant. The effects of infusing DPPC-SUVs on the expression of VCAM-1 (–17%), ICAM-1 (–23%), and MCP-1 (–64%) were not statistically significant.
Discussion
This is the first study to demonstrate that HDLs act in vivo in inflamed, nonatherosclerotic arteries to reduce the acute infiltration of neutrophils and inhibit the generation of ROS. The antiinflammatory properties of HDL have been demonstrated previously both in vitro1,2,27 and in vivo4–6 in several, but not in all,28,29 previous studies. In vivo studies have generally been conducted in animals with hypercholesterolemia and atherosclerosis, although in vivo antiinflammatory effects of rHDL have also been observed in the absence of atherosclerosis in a porcine model of acute inflammation induced by the intradermal injection of interleukin-1.6 In the porcine model, a single infusion of rHDL inhibited the expression of endothelial E-selectin in the intradermal vessels, but effects on neutrophil recruitment and generation of ROS were not reported. The present studies demonstrate a profound ability of rHDL to inhibit the acute inflammatory changes induced by insertion of a nonocclusive Silastic collar around carotid arteries of normocholesterolemic, nonatherosclerotic rabbits.
Periarterial collars have been used previously in rabbits,30 in monkeys,31 and in both wild-type31 and hypercholesterolemic apoE-knockout4 mice to induce acute inflammatory changes in the artery wall. Early effects (within the first 3 days) include upregulation of cellular adhesion molecules and MCP-1, with an influx of neutrophils being apparent as early as 24 hours after the collar was inserted.32 Collar insertion has also been shown to increase iNOS and reduce eNOS in rabbit arteries, but only after about 5 days.30 Monocytes first appear in the artery wall after 5 days, and by 7 to 21 days, there is demonstrable neointimal hyperplasia.4,33 The present study was designed to assess the antiinflammatory effects of rHDL during the very early stages of the inflammatory response before there was observable infiltration of monocytes and before development of neointimal hyperplasia. The substantial HDL-mediated reduction in neutrophil infiltration was probably secondary to the known ability of HDL to inhibit endothelial cell ICAM-1,1 a factor that promotes neutrophil infiltration into the artery wall.34 The fact that neutrophils are an important component of many acute inflammatory conditions, including myocardial ischemia-reperfusion injury35 and stroke,36 raises the possibility that HDL may have an important protective role to play in these conditions.
Neutrophils are a major source of ROS.37 In turn, ROS increase the expression of endothelial ICAM-138 and thus recruit additional neutrophils into the artery wall. The vicious circle that results may be broken at several points by HDL. For example, HDLs inhibit the expression of ICAM-1 in activated endothelial cells by a mechanism involving the inhibition of endothelial cell sphingosine kinase.39 Such inhibition of ICAM-1 would reduce the infiltration of neutrophils into the artery wall and, as a consequence, reduce the generation of ROS, thus breaking the cycle. A reduction in the generation of ROS may also be achieved directly as a consequence of the antioxidant properties of HDL.40,41
In the present study, the possibility was considered that a reduced availability of nitric oxide may have contributed to the acute inflammatory changes observed in collared segments of artery and that some of the beneficial effects of HDL may have been secondary to their ability to generate nitric oxide by upregulating eNOS.42 However, this was not supported by the observation that iNOS expression was undetectable in both the collared and noncollared artery segments, whether or not rHDLs were infused. In addition, neither collar insertion nor rHDL infusion had a statistically significant effect on the minimal constitutive expression of eNOS. These observations are consistent with a previous report showing that collar-induced effects on iNOS and eNOS become apparent only after 5 days,30 and not at the 48-hour time point used in the present study. It is possible, however, that infusion of rHDL increased the bioavailability of nitric oxide by reducing collar-induced superoxide production (which would otherwise have inactivated nitric oxide). Such an effect would be independent of changes in NOS expression.
It was of interest to note that infusion of lipid-free apoA-I had an antiinflammatory effect comparable to that of rHDL. This contrasted with observations made in vitro27 in which lipid-free apoA-I was found to have no antiinflammatory activity. It is possible, however, that the positive effects of lipid-free apoA-I in the present study reflected no more than a rapid in vivo lipidation of the lipid-free apoA-I to form apoA-I/PC complexes with a composition similar to that of the rHDLs that were infused. Another difference between the results of these in vivo studies and those previously obtained in vitro relates to the effects of PC composition. In the present in vivo studies, there were no differences in the antiinflammatory effects of PLPC and DPPC, whether administered as components of rHDL or SUVs, whereas in previous studies conducted in vitro, PLPC was found to be much more effective than DPPC in terms of inhibiting VCAM-1 expression in activated HUVECs.43 The explanation for these differences is unclear.
The antiinflammatory effects of infusing rHDL and lipid-free apoA-I in the present study were achieved with remarkably small amounts of apoA-I. Each infusion contained only 25 mg of apoA-I. This amount, when given to a 3-kg rabbit (plasma volume approximately 120 mL), would have increased the plasma apoA-I concentration in the immediate postinfusion period by 0.2 mg/mL, an increase of only 40% over the preinfusion concentration of 0.5 mg/mL. Furthermore, with a fractional catabolic rate for rabbit plasma apoA-I of 0.8 pools/d,44 it would be predicted (as observed) that the apoA-I concentration would have returned to preinfusion levels by the time of euthanasia 24 hours after the last infusion. Yet despite what was no more than a minor and transient increase in the concentration of plasma HDL, the infusions resulted in an almost complete inhibition of the collar-induced arterial inflammation. This suggests that the infused rHDL, apoA-I, and SUVs had profound antiinflammatory effects that extended well beyond those resulting from a simple increase in the concentration of plasma HDL.
The mechanism underlying these potent antiinflammatory effects of infused rHDL, lipid-free apoA-I, and SUVs is unknown. Despite this, the therapeutic implications of the present findings are substantial, providing strong support for a proposition that the infusion of rHDL (or indeed lipid-free apoA-I or PC SUVs) should be investigated as first-line therapy to minimize tissue damage in devastating acute inflammatory states such as acute coronary syndromes, stroke, and ischemia-reperfusion injury.
Acknowledgments
This work was supported by a grant from the Pfizer International HDL Awards. Dr Nicholls was supported by a postgraduate research scholarship from the National Heart Foundation of Australia. Dr Drummond is supported by a Peter Doherty Fellowship from the National Health and Medical Research Council of Australia (007044). Dr Rye is a National Heart Foundation of Australia Principal Research Fellow.
References
Cockerill GW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol. 1995; 15: 1987–1994.
Navab M, Imes SS, Hama SY, Hough GP, Ross LA, Bork RW, Valente AJ, Berliner JA, Drinkwater DC, Laks H, Fogelman AM. Monocyte transmigration induced by modification of low density lipoprotein in cocultures of human aortic wall cells is due to induction of monocyte chemotactic protein 1 synthesis and is abolished by high density lipoprotein. J Clin Invest. 1991; 88: 2039–2046.
Moudry R, Spycher MO, Doran JE. Reconstituted high density lipoprotein modulates adherence of polymorphonuclear leukocytes to human endothelial cells. Shock. 1997; 7: 175–181.
Dimayuga P, Zhu J, Oguchi S, Chyu K-Y, Xu X-OH, Yano J, Shah PK, Nilsson J, Cercek B. Reconstituted HDL containing human apolipoprotein A-I reduces VCAM-1 expression and neointima formation following periadventitial cuff-induced carotid injury in apoE null mice. Biochem Biophys Res Commun. 1999; 264: 465–468.
Rong JX, Li J, Reis ED, Choudhury RP, Dansky HM, Elmalem VI, Fallon JT, Breslow JL, Fisher EA. Elevating high-density lipoprotein cholesterol in apolipoprotein E-deficient mice remodels advanced atherosclerotic lesions by decreasing macrophage and increasing smooth muscle cell content. Circulation. 2001; 104: 2447–2452.
Cockerill GW, Huehns TY, Weerasinghe A, Stocker C, Lerch PG, Miller NE, Haskard DO. Elevation of plasma high-density lipoprotein concentration reduces interleukin-1-induced expression of E-selectin in an in vivo model of acute inflammation. Circulation. 2001; 103: 108–112.
Paterno R, Ruocco A, Postiglione A, Hubsch A, Anderson I, Lang MG. Reconstituted high-density lipoprotein exhibits neuroprotection in two rat models of stroke. Cerebrovasc Dis. 2004; 17: 204–211.
Cockerill GW, McDonald MC, Mota-Filipe H, Cuzzocrea S, Miller NE, Thiemermann C. High density lipoproteins reduce organ injury and organ dysfunction in a rat model of hemorrhagic shock. FASEB J. 2001; 15: 1941–1952.
Spieker LE, Sudano I, Hurlimann D, Lerch PG, Lang MG, Binggeli C, Corti R, Ruschitzka F, Luscher TF, Noll G. High-density lipoprotein restores endothelial function in hypercholesterolemic men. Circulation. 2002; 105: 1399–1402.
Bisoendial RJ, Hovingh GK, Levels JHM, Lerch PG, Andresen I, Hayden MR, Kastelein JJP, Stroes ESG. Restoration of endothelial function by increasing high-density lipoprotein in subjects with isolated low high-density lipoprotein. Circulation. 2003; 107: 2944–2948.
Osborne J. Delipidation of plasma lipoproteins. Methods Enzymol. 1986; 128: 213–222.
Weisweiler P. Isolation and quantitation of apolipoproteins A-I and A-II from human high-density lipoproteins by fast-protein liquid chromatography. Clin Chim Acta. 1987; 169: 249–254.
Clay M, Rye KA, Barter PJ. Evidence in vitro that hepatic lipase reduces the concentration of apolipoprotein A-I in rabbit high-density lipoproteins. Biochim Biophys Acta. 1990; 1044: 50–56.
Matz CE, Jonas A. Micellar complexes of human apolipoprotein A-I with phosphatidylcholines and cholesterol prepared from cholate-lipid dispersions. J Biol Chem. 1982; 257: 4535.
Smith P, Krohn R, Hermanson G. Measurement of protein using bicinchonic acid. Anal Biochem. 1985; 150: 76–85.
Takayama M, Itoh S, Nagasaki T. A new enzymatic method for determination of serum choline-containing phospholipids. Clin Chim Acta. 1977; 79: 93–98.
Rainwater D, Andres D, Ford A. Production of polyacrylamide gradient gels for the electrophoretic resolution of lipoproteins. J Lipid Res. 1992; 33.
Wetterau JR, Jonas A. Effect of dipalmitoylphosphatidylcholine vesicle curvature on the reaction with human apolipoprotein A-I. J Biol Chem. 1982; 257: 10961–10966.
Laursen J, Somers M, Kurz S. Endothelial regulation of vasomation in ApoE-deficient mice: implications for interactions between peroxynitrite and tetrahydrobiopterin. Circulation. 2001; 103: 1282–1288.
Paravicini TM, Gulluyan LM, Dusting GJ, Drummond GR. Increased NADPH oxidase activity, gp91phox expression, and endothelium-dependent vasorelaxation during neointima formation in rabbits. Circ Res. 2002; 91: 54–61.
Munzel T. Afanas’ev IB, Kleschyov AL, Harrison DG. Detection of superoxide in vascular tissue. Arterioscler Thromb Vasc Biol. 2002; 22: 1761–1768.
Sukhova G, Williams J, Libby P. Statins reduce inflammation in atheroma of nonhuman primates independent of effects of serum cholesterol. Arterioscler Thromb Vasc Biol. 2002; 22: 1452001458.
Herdeg C, Oberhoff M, Baumbah A. Effects of local all-trans-retinoic acid delivery on experimental atherosclerosis in the rabbit carotid artery. Cardiovasc Res. 2003; 57: 544–553.
Shen J, Bao S, Reeve VE. Modulation of IL-10, IL-12, and IFN-gamma in the epidermis of hairless mice by UVA (320–400 nm) and UVB (280–320 nm) radiation. J Invest Dermatol. 1999; 113: 1059–1064.
Wahlefeld A. Triglycerides: determination after enzymatic hydrolysis. In: Bergmeyer H, ed. Methods of Enzymatic Analysis. 2nd ed. New York, NY: Academic Press; 1974.
Allen JK, Hensley WJ, Nicholls AV, Whitfield JB. An enzymatic and centrifugal method for estimating high-density lipoprotein cholesterol. Clin Chem. 1979; 25: 325–327.
Baker PW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. Ability of reconstituted high density lipoproteins to inhibit cytokine-induced expression of vascular cell adhesion molecule-1 in human umbilical vein endothelial cells. J Lipid Res. 1999; 40: 345–353.
Stannard AK, Khan S, Graham A, Owen JS, Allen SP. Inability of plasma high-density lipoproteins to inhibit cell adhesion molecule expression in human coronary artery endothelial cells. Atherosclerosis. 2001; 154: 31–38.
Dansky HM, Charlton SA, Barlow CB, Tamminen M, Smith JD, Frank JS, Breslow JL. Apo A-I inhibits foam cell formation in apoE-deficient mice after monocyte adherence to endothelium. J Clin Invest. 1999; 104: 31–39.
Arthur JF, Yin ZL, Young HM, Dusting GJ. Induction of nitric oxide synthase in the neointima induced by a periarterial collar in rabbits. Arterioscler Thromb Vasc Biol. 1997; 17: 737–740.
Egashira K, Zhao Q, Kataoka C, Ohtani K, Usui M, Charo IF, Nishida K, Inoue S, Katoh M, Ichiki T, Takeshita A. Importance of monocyte chemoattractant protein-1 pathway in neointimal hyperplasia after periarterial injury in mice and monkeys. Circ Res. 2002; 90: 1167–1172.
Donetti E, Baetta R, Comparato C, Altana C, Sartore S, Paoletti R, Castano P, Gabbiani G, Corsini A. Polymorphonuclear leukocyte-myocyte interaction: an early event in collar-induced rabbit carotid intimal thickening. Exp Cell Res. 2002; 274: 197–206.
Dusting GJ, Curcio A, Harris PJ, Lima B, Zambetis M, Martin JF. Supersensitivity to vasoconstrictor action of serotonin precedes the development of atheroma-like lesions in the rabbit. J Cardiovasc Pharmacol. 1990; 16: 667–674.
Justicia C, Panes J, Sole S, Cervera A, Deulofeu R, Chamorro A, Planas AM. Neutrophil infiltration increases matrix metalloproteinase-9 in the ischemic brain after occlusion/reperfusion of the middle cerebral artery in rats. J Cereb Blood Flow Metab. 2003; 23: 1430–1440.
Jordan JE, Zhao ZQ, Vinten-Johansen J. The role of neutrophils in myocardial ischemia-reperfusion injury. Cardiovasc Res. 1999; 43: 860–878.
Prestigiacomo CJ, Kim SC, Connolly ES Jr, Liao H, Yan SF, Pinsky DJ. CD18-mediated neutrophil recruitment contributes to the pathogenesis of reperfused but not nonreperfused stroke. Stroke. 1999; 30: 1110–1117.
Korthuis RJ, Granger DN. Reactive oxygen metabolites, neutrophils, and the pathogenesis of ischemic-tissue-reperfusion. Clin Cardiol. 1993; 16: I19–I26.
Sellak H, Franzini E, Hakim J, Pasquier C. Reactive oxygen species rapidly increase endothelial ICAM-1 ability to bind neutrophils without detectable upregulation. Blood. 1994; 83: 2669–2677.
Xia P, Vadas MA, Rye K-A, Barter PJ, Gamble JR. High density lipoproteins (HDL) interrupt the sphingosine kinase signaling pathway. A possible mechanism for protection against atherosclerosis by HDL. J Biol Chem. 1999; 274: 33143–33147.
Mackness MI, Durrington PN, Mackness B. How high-density lipoprotein protects against the effects of lipid peroxidation. Curr Opin Lipidol. 2000; 11: 383–388.
Navab M, Hama SY, Anantharamaiah GM, Hassan K, Hough GP, Watson AD, Reddy ST, Sevanian A, Fonarow GC, Fogelman AM. Normal high density lipoprotein inhibits three steps in the formation of mildly oxidized low density lipoprotein: steps 2 and 3. J Lipid Res. 2000; 41: 1495–1508.
Yuhanna IS, Zhu Y, Cox BE, Hahner LD, Osborne-Lawrence S, Lu P, Marcel YL, Anderson RG, Mendelsohn ME, Hobbs HH, Shaul PW. High-density lipoprotein binding to scavenger receptor-BI activates endothelial nitric oxide synthase. Nat Med. 2001; 7: 853–857.
Baker PW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. Phopholipid composition of reconstituted high density lipoproteins influences their ability to inhibit endothelial cell adhesion molecule expression. J Lipid Res. 2000; 41: 1261–1267.
Kee P, Rye K-A, Taylor JL, Barrett PH, Barter PJ. Metabolism of ApoA-I as lipid-free protein or as component of discoidal and spherical reconstituted HDLs: studies in wild-type and hepatic lipase transgenic rabbits. Arterioscler Thromb Vasc Biol. 2002; 22: 1912–1917.(Stephen J. Nicholls, MBBS)
the Department of Medicine, University of Adelaide, Adelaide (S.J.N.)
the Howard Florey Institute, University of Melbourne, Melbourne (G.J.D., G.R.D.)
the Department of Medicine, University of Sydney, Sydney (P.J.B.)
the Department of Pathology, University of Sydney, Sydney (S.B.), Australia.
Abstract
Background— HDLs have antiinflammatory and antioxidant properties in vitro. This study investigates these properties in vivo.
Methods and Results— Chow-fed, normocholesterolemic New Zealand White rabbits received a daily infusion of (1) saline, (2) reconstituted HDL (rHDL) containing 25 mg apolipoprotein (apo) A-I and 50 mg of either 1-palmitoyl-2-linoleoyl phosphatidylcholine (PLPC) or 1,2-dipalmitoyl phosphatidylcholine (DPPC), (3) 25 mg lipid-free apoA-I, or (4) 50 mg of either PLPC-small unilamellar vesicles (SUVs) or DPPC-SUVs on each of 3 consecutive days. Nonocclusive carotid periarterial collars were implanted after the second dose of treatment. Forty-eight hours after insertion of the collars, the arteries were removed and analyzed for the presence of reactive oxygen species, the infiltration of neutrophils, and the expression of adhesion proteins and chemokines. Insertion of the periarterial collar induced a 4.1-fold increase in presence of vascular wall reactive oxygen species. This effect was completely abolished in the animals infused with rHDL. The periarterial collar was associated with a dense infiltration of the arterial wall by polymorphonuclear leukocytes. This infiltration was inhibited by 73% to 94% in the animals infused with rHDL, by 75% in the animals infused with lipid-free apoA-I, and by 51% to 65% in animals infused with SUVs. There were no significant differences between the effects of PLPC and DPPC in either the rHDL or SUVs. Endothelial expression of vascular cell adhesion molecule-1, intercellular adhesion molecule-1, and monocyte chemoattractant protein-1 was also increased by the collar insertion and inhibited by rHDL, lipid-free apoA-I, and, to a lesser extent, also by the SUVs.
Conclusions— Infusion of rHDL, apoA-I, and phospholipid-SUVs inhibits the early pro-oxidant and proinflammatory changes induced by a periarterial collar in normocholesterolemic rabbits.
Key Words: lipoproteins ; inflammation ; cholesterol ; antioxidants ; endothelium
Introduction
High-density lipoproteins (HDLs) have several functions that may contribute to their ability to protect against atherosclerosis. The best known of these relates to their role in promoting the efflux of cholesterol from macrophages in the artery wall. However, HDLs also have antioxidant, antithrombotic, and antiinflammatory functions that may contribute to their protective properties. The present study is concerned with the antiinflammatory effects of HDL in vivo.
It has been shown previously in studies conducted in vitro that HDLs inhibit the cytokine-induced expression of vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and E-selectin in human umbilical vein endothelial cells (HUVECs) growing in tissue culture.1 HDLs also inhibit binding of monocytes2 and neutrophils3 to endothelial cells growing in culture. There is evidence that the antiinflammatory properties of HDL also operate in vivo, although in most cases this has been demonstrated in a setting of hypercholesterolemia and atherosclerosis. For example, intravenous infusion of reconstituted HDL (rHDL) reduces the vivo expression of endothelial adhesion molecules induced by insertion of carotid periarterial cuffs in cholesterol-fed, apoE-knockout mice.4 In another study of apoE-knockout mice, the increase in HDL concentration accompanying an overexpression of the human apoA-I gene reduced macrophage accumulation in the aortic root by more than 3-fold.5 This was associated with a reduced in vivo oxidation of ;-VLDL, lower ICAM-1 and VCAM-1 expression, and diminished ex vivo leukocyte adhesion.
However, there are also examples of in vivo antiinflammatory effects of HDL in the absence of hypercholesterolemia and atherosclerosis. For example, injection of rHDL inhibits the development of a local inflammatory infiltrate after the subcutaneous administration of interleukin-1 in a porcine model.6 Also, in studies of experimental stroke in rats, pretreatment with rHDL significantly and substantially reduced the brain necrotic area in a process possibly related to an rHDL-induced reduction in levels of reactive oxygen species (ROS).7 Furthermore, in a study of hemorrhagic shock in rats, the resulting multiple organ dysfunction syndrome was largely abolished by a single injection of human HDL given 90 minutes after the hemorrhage and 1 minute before resuscitation. In that model, injection of HDL prevented both the severe disruption of tissue architecture and the extensive cellular infiltration into the affected tissues.8
There is also circumstantial evidence that HDLs have acute antiinflammatory effects in vivo in humans. In one study, a single intravenous infusion of rHDL into hypercholesterolemic humans normalized endothelium-dependent vasodilation, possibly by increasing NO bioavailability.9 In a second human study, a single injection of rHDL corrected the endothelial dysfunction associated with low levels of HDL in ABCA1 heterozygotes.10
Thus, there is mounting evidence that HDLs have acute antiinflammatory effects that are extremely rapid and possibly unrelated to their cholesterol-transporting function. In the present study, we have used a model of acute arterial inflammation in normocholesterolemic, nonatherosclerotic rabbits to demonstrate in vivo for the first time that injection of rHDL markedly inhibits both the infiltration of neutrophils into the artery wall and the subsequent generation of ROS.
Methods
Animals
Male New Zealand White rabbits (Nanowie Small Animal Production Unit, Modewarre, Australia) weighing approximately 3 kg were maintained on a normal laboratory chow diet throughout the study. All procedures were approved by the Howard Florey Institute Animal Ethics Committee (protocol 02-009).
Isolation of Lipid-Free ApoA-I
HDLs were isolated from pooled samples of rabbit plasma (Quality Farms of Australia, Lara, Australia) by sequential ultracentrifugation in the 1.063- to 1.21-g/mL density range. The HDLs were lyophilized and delipidated.11 ApoA-I was isolated by chromatography12 on a Q-Sepharose Fast Flow column (Amersham Biosciences) attached to a fast performance liquid chromatography system (Amersham Biosciences). The purified apoA-I was lyophilized and stored at –20°C until used. Lyophilized apoA-I was reconstituted in 3 mol/L guanidine hydrochloride and dialyzed against endotoxin-free PBS (pH 7.4; Sigma) containing 0.2 g/L KH2PO4, 0.2 g/L KCl, 8 g/L NaCl, and 1.15 g/L Na2PO4 before being infused or used to prepare rHDL. The concentration of apoA-I was determined by use of an immunoturbidimetric assay.13
Preparation of Reconstituted HDL
Discoidal rHDL containing apoA-I complexed to either 1-palmitoyl-2-linoleoyl phosphatidylcholine (PC) (PLPC, Sigma) or 1,2-dipalmitoyl PC (DPPC, Sigma), in a molar ratio of PC to apoA-I of 200:1, were prepared by use of the cholate dialysis method.14 The resulting rHDLs were dialyzed extensively against endotoxin-free PBS before use. Protein15 and phospholipid16 concentrations were determined by enzymatic assay. The rHDLs were subjected to agarose gel electrophoresis17 and 3% to 40% nondenaturing gradient gel electrophoresis to determine their surface charge and Stokes’ diameter, respectively.
Preparation of Phospholipid Small Unilamellar Vesicles
Small unilamellar vesicles (SUVs) containing either PLPC or DPPC and BHT (molar ratio of PC to BHT, 10:1) were prepared in PBS as described.18
Administration of Reconstituted HDL
Twenty-nine rabbits were randomly allocated to receive treatment with saline (n=6), PLPC-rHDL (n=4), DPPC-rHDL (n=4), lipid-free apoA-I (n=5), PLPC SUVs (n=5), or DPPC (n=5). Each treatment of rHDL and lipid-free apoA-I contained 25 mg of apoA-I (8 mg/kg). Each treatment of rHDL and SUVs contained 50 mg of PC (16 mg/kg). The treatments were administered via a marginal ear vein on each of 3 occasions: 24 hours before collar implantation, directly before the surgical procedure, and 24 hours after the collar implantation. The animals were euthanized 48 hours after the collar implantation.
Collar Implantation
The rabbits were anesthetized with intravenous propofol (5 mg/kg) followed by intramuscular ketamine/xylazine (50/10 mg/kg). Both carotid arteries were exposed surgically and cleared of connective tissue along a 30-mm length. Hollow, nonocclusive Silastic collars (length, 20 mm; internal diameter along bore, 4 mm; internal diameter at ends, 1 mm) were then placed around each artery and held in place with a nylon sleeve. The space inside the collar was filled with sterile saline (0.9%). Muscle, fat, and skin layers were sutured, the wound was dressed with antibiotic, and animals were allowed to recover for 48 hours before being euthanized.
Tissue Harvesting
Forty-eight hours after insertion of the collar, blood was sampled from a marginal ear vein. Animals were then heparinized (1000 U IV) before euthanasia with an overdose of sodium pentobarbitone (90 mg/kg IV). The collared segment of the carotid artery and approximately 10 mm of noncollared artery proximal to the collar were excised and placed in ice-cold Krebs-HEPES buffer (composition, in mmol/L: NaCl 99.0, KCl 4.7, KH2PO4 1.0, MgSO4 1.2, CaCl2 2.5, NaHCO3 25.0, Na-HEPES 20.0, and glucose 11.0, pH 7.4). Collars were removed and arteries cleaned of fat and connective tissue. Three ring sections (3 mm) were cut both from the area enclosed by the collar and from the proximal noncollared segment of artery. One section from each segment was snap-frozen in Tissue-Tek OCT (Sakura) fixative for immunohistochemical analysis. The other sections were used to determine generation of ROS.
Detection of ROS
Levels of ROS in the carotid arteries were measured by 5 μmol/L lucigenin-enhanced chemiluminescence.19 The increased chemiluminescence signal, probably reflecting superoxide, is mediated by increased activity of NADPH oxidase.20 To minimize the possibility that the lucigenin used in the assay may have contributed to the production of superoxide, we used a concentration of lucigenin that was not sufficient to participate in superoxide production by redox cycling.21 Ring segments were incubated for 45 minutes at 37°C in Krebs-HEPES buffer containing diethyldithiocarbamate (DETCA, 3 mmol/L) to irreversibly inactivate endogenous Cu2+/Zn2+ superoxide dismutase. The possibility that HDL may have interfered with the ability of DETCA to inactivate superoxide dismutase was excluded on the grounds that HDLs were not present in the wells during the incubation. Some rings were further treated with NADPH (10 μmol/L), the preferred substrate of NADPH oxidase, which is, in turn, the predominant source of superoxide in rabbit carotid arteries.20 Each segment was then transferred to a separate well of a white, opaque 96-well plate containing 300 μL of 5-μmol/L lucigenin in Krebs-HEPES buffer as well as the appropriate drug treatment. DETCA was excluded from the lucigenin assay solution. The 96-well plate was loaded into a TopCount Single Photon Counter (Packard Bioscience), and photon emission per second was measured (6-second count time per cycle, 12 cycles, 1-minute delay between cycles). Ring segments were dried for 2 to 3 days in a 65°C oven, and the production of ROS was normalized to dry tissue weight (counts per second per milligram).
Immunohistochemistry
Frozen tissues were sectioned in 5-μm slices. A section from the middle of each of the collared and noncollared segments of both carotid arteries was then subjected to immunohistochemical staining. Sections were fixed with methanol/acetone (1:1, vol/vol) at room temperature for 5 minutes. Arterial wall infiltration by inflammatory cells was determined by use of mouse monoclonal antibodies against macrophages (RAM11, DakoCytomation), neutrophils (CD18, Serotag), and lymphocytes (CD43). In addition, the endothelial expression of adhesion proteins and chemokines was determined by use of murine monoclonal antibodies against rabbit VCAM-1 and ICAM-1 (gifts from Dr. M. Cybulsky, University of Toronto), monocyte chemotactic protein (MCP)-1 (a gift from Dr. A. Matsukowa, Kumamoto University), a goat polyclonal antibody against human E-selectin (R&D), endothelial nitric oxide synthase (eNOS) (BD Biosciences), and inducible nitric oxide synthase (iNOS) (BD Biosciences). These were applied and incubated overnight at room temperature. Endothelial integrity was determined by use of a mouse monoclonal antibody against human CD31 (DakoCytomation). Sections were incubated with biotinylated anti-mouse or anti-goat immunoglobulins for 30 minutes and then incubated with alkaline phosphatase-labeled streptavidin solution for 60 minutes. Slides were rinsed in PBS (pH 7.4) after each incubation. Peroxidase activity was revealed by diaminobenzamine. Slides were counterstained with hematoxylin and mounted. These slides were subsequently graded independently by 4 pathologists who were blinded to the treatment status of the animal. The degree of endothelial staining was graded22,23 by use of a scale that incorporated both the strength of staining and the amount of endothelial surface involved (0=no staining, 1=weak staining of less than 50% of endothelium, 2=strong staining of less than 50% or weak staining of more than 50% of endothelium, 3=strong staining between 50% and 99% of endothelium, and 4=strong staining of 100% of endothelium). The collar-induced change in expression was calculated as the difference in mean score between noncollared and collared segments of each individual artery. Infiltration of the arterial wall by neutrophils, demonstrated by CD18 staining, was determined by quantitative immunohistochemistry. Digital micrographs were acquired with an Olympus BX40 microscope, and the percentage of total vessel wall area occupied by positive staining was determined by use of ImagePro Plus 4 (Cybernetics).24
Plasma Analyses
Plasma collected at the commencement of the study and before euthanasia of the animal was stored at –80°C in EDTA until required for analysis. All chemical analyses were performed on a Roche Diagnostics/Hitachi 902 autoanalyzer (Roche Diagnostics GmbH). Triglyceride25 was determined enzymatically. Total cholesterol was determined by use of a Roche Diagnostics kit. HDL cholesterol was determined by enzymatic assay after precipitation of apoB containing lipoproteins with polyethylene glycol.26 ApoA-I concentrations were determined by an immunoturbidimetric assay using a sheep anti-rabbit apoA-I immunoglobulin.13
Data Analysis
All results are expressed as mean±SEM. Statistical comparisons were made by Student t tests and 1-way ANOVA using the statistical program in GraphPad Prism Version 4.0. A value of P<0.05 was considered significant.
Results
Plasma Lipid Concentrations
Plasma concentrations of triglyceride, total and free cholesterol, HDL cholesterol, and apoA-I at the commencement of the study and at the time of euthanasia of the animals are presented in the Table. No significant differences were found between the groups, nor were there significant differences in lipid levels resulting from the infusions of saline, rHDL, lipid-free apoA-I, or PC-SUVs.
Plasma Lipid Profiles From Animals Infused With Saline (n=6), PLPC-rHDL (n=4), DPPC-rHDL (n=4), Lipid-Free apoA-I (n=5), PLPC-SUV (n=5), and DPPC-SUV (n=5) at Baseline and Immediately Before Death
Generation of ROS
Application of the periarterial collar increased the production of ROS in the vessel wall both in the unstimulated state without NADPH (126.9±48.1 and 6.6±2.9 counts · s–1 · mg–1 in the collared and noncollared segments, respectively, P<0.005) and after stimulating by incubation with NADPH (540.3±169.1 and 130.3±42.1 counts · s–1 · mg–1 in the collared and noncollared segments, respectively, P<0.01) (Figure 1). The collar-induced increase in vascular ROS that was observed in both the unstimulated (Figure 1A) and the stimulated (Figure 1B) states was significantly reduced in the vessels isolated from animals infused with rHDL. Infusion of PLPC-rHDL inhibited the collar-induced increase in vascular ROS in the unstimulated state by 92% (10±4.1 and 126.9±48.1 counts · s–1 · mg–1 in the PLPC-rHDL and saline-infused animals, respectively, P<0.03) and in the stimulated state by 85% (82.3±34.4 and 540.3±169.1 counts · s–1 · mg–1 in the PLPC-rHDL and the saline-infused animals, respectively, P<0.001). The collar-induced increase in ROS in the stimulated state was also inhibited by 90% by infusion of DPPC-rHDL (56.7±29.8 and 540.3±169.1 counts · s–1 · mg–1 in the DPPC-rHDL- and saline-treated animals, respectively, P<0.001) (Figure 1B). The effects of infusing PLPC-rHDL and DPPC-rHDL were not significantly different.
Neutrophil Infiltration
The collar-induced neutrophil recruitment was also inhibited 73% by infusion of lipid-free apoA-I (30±9% of the total vessel wall area in the saline-infused rabbits reduced to 8±2.9% in the animals infused with lipid-free apoA-I, P<0.02). This effect of infusing lipid-free apoA-I was not significantly different from that of rHDL (Figure 3).
Infusion of SUVs also reduced the infiltration of neutrophils but appeared to be less effective than either rHDL or lipid-free apoA-I (Figure 3). The collar-induced neutrophil recruitment was 30±9% of the total vessel wall area in the saline-infused rabbits, 14.6±4.2% in the animals infused with the PLPC-SUVs (P=NS compared with saline) and 10.4% in the animals infused with the DPPC-SUVs (P<0.05 compared with saline) (Figure 3). The accumulation of neutrophils in the collared arteries of animals infused with PLPC-SUVs and DPPC-SUVs was not significantly different.
Vascular Adhesion Molecules and Chemokines
Compared with saline, infusion of rHDL inhibited the collar-induced increase in endothelial expression of VCAM-1 (by 53.9% and 54.4% with PLPC-rHDL and DPPC-rHDL, respectively, P<0.005), ICAM-1 (by 50% and 74.6% with PLPC-rHDL, P<0.01, and DPPC-rHDL, P<0.005, respectively), and MCP-1 (by 76.5% with PLPC-rHDL, P<0.01). Although DPPC-rHDL inhibited the collar-induced change in MCP-1 by 50% compared with saline, this did not reach statistical significance (P=0.09). The collar-induced change in MCP-1 did not differ significantly between the PLPC-rHDL and DPPC-rHDL groups. The infusion of rHDL, combining the PLPC and DPPC groups, significantly inhibited the collar-induced change in MCP-1 staining (by 63%, P<0.05). Infusion of rHDL had no measurable effect on the expression of either iNOS or eNOS (results not shown).
Infusion of lipid-free apoA-I had effects similar to those of the rHDL infusions. Compared with saline, infusion of lipid-free apoA-I inhibited the collar-induced increase in expression of VCAM-1 by 40.5% (P<0.01), ICAM-1 by 94.1% (P<0.0005), and MCP-1 by 76.5% (P<0.05).
Infusion of PLPC-SUVs inhibited the collar-induced increase in expression of VCAM-1 by 47.2% (P<0.01) and ICAM-1 by 79.7% (P<0.01), but the 55.6% reduction in MCP-1 expression was not statistically significant. The effects of infusing DPPC-SUVs on the expression of VCAM-1 (–17%), ICAM-1 (–23%), and MCP-1 (–64%) were not statistically significant.
Discussion
This is the first study to demonstrate that HDLs act in vivo in inflamed, nonatherosclerotic arteries to reduce the acute infiltration of neutrophils and inhibit the generation of ROS. The antiinflammatory properties of HDL have been demonstrated previously both in vitro1,2,27 and in vivo4–6 in several, but not in all,28,29 previous studies. In vivo studies have generally been conducted in animals with hypercholesterolemia and atherosclerosis, although in vivo antiinflammatory effects of rHDL have also been observed in the absence of atherosclerosis in a porcine model of acute inflammation induced by the intradermal injection of interleukin-1.6 In the porcine model, a single infusion of rHDL inhibited the expression of endothelial E-selectin in the intradermal vessels, but effects on neutrophil recruitment and generation of ROS were not reported. The present studies demonstrate a profound ability of rHDL to inhibit the acute inflammatory changes induced by insertion of a nonocclusive Silastic collar around carotid arteries of normocholesterolemic, nonatherosclerotic rabbits.
Periarterial collars have been used previously in rabbits,30 in monkeys,31 and in both wild-type31 and hypercholesterolemic apoE-knockout4 mice to induce acute inflammatory changes in the artery wall. Early effects (within the first 3 days) include upregulation of cellular adhesion molecules and MCP-1, with an influx of neutrophils being apparent as early as 24 hours after the collar was inserted.32 Collar insertion has also been shown to increase iNOS and reduce eNOS in rabbit arteries, but only after about 5 days.30 Monocytes first appear in the artery wall after 5 days, and by 7 to 21 days, there is demonstrable neointimal hyperplasia.4,33 The present study was designed to assess the antiinflammatory effects of rHDL during the very early stages of the inflammatory response before there was observable infiltration of monocytes and before development of neointimal hyperplasia. The substantial HDL-mediated reduction in neutrophil infiltration was probably secondary to the known ability of HDL to inhibit endothelial cell ICAM-1,1 a factor that promotes neutrophil infiltration into the artery wall.34 The fact that neutrophils are an important component of many acute inflammatory conditions, including myocardial ischemia-reperfusion injury35 and stroke,36 raises the possibility that HDL may have an important protective role to play in these conditions.
Neutrophils are a major source of ROS.37 In turn, ROS increase the expression of endothelial ICAM-138 and thus recruit additional neutrophils into the artery wall. The vicious circle that results may be broken at several points by HDL. For example, HDLs inhibit the expression of ICAM-1 in activated endothelial cells by a mechanism involving the inhibition of endothelial cell sphingosine kinase.39 Such inhibition of ICAM-1 would reduce the infiltration of neutrophils into the artery wall and, as a consequence, reduce the generation of ROS, thus breaking the cycle. A reduction in the generation of ROS may also be achieved directly as a consequence of the antioxidant properties of HDL.40,41
In the present study, the possibility was considered that a reduced availability of nitric oxide may have contributed to the acute inflammatory changes observed in collared segments of artery and that some of the beneficial effects of HDL may have been secondary to their ability to generate nitric oxide by upregulating eNOS.42 However, this was not supported by the observation that iNOS expression was undetectable in both the collared and noncollared artery segments, whether or not rHDLs were infused. In addition, neither collar insertion nor rHDL infusion had a statistically significant effect on the minimal constitutive expression of eNOS. These observations are consistent with a previous report showing that collar-induced effects on iNOS and eNOS become apparent only after 5 days,30 and not at the 48-hour time point used in the present study. It is possible, however, that infusion of rHDL increased the bioavailability of nitric oxide by reducing collar-induced superoxide production (which would otherwise have inactivated nitric oxide). Such an effect would be independent of changes in NOS expression.
It was of interest to note that infusion of lipid-free apoA-I had an antiinflammatory effect comparable to that of rHDL. This contrasted with observations made in vitro27 in which lipid-free apoA-I was found to have no antiinflammatory activity. It is possible, however, that the positive effects of lipid-free apoA-I in the present study reflected no more than a rapid in vivo lipidation of the lipid-free apoA-I to form apoA-I/PC complexes with a composition similar to that of the rHDLs that were infused. Another difference between the results of these in vivo studies and those previously obtained in vitro relates to the effects of PC composition. In the present in vivo studies, there were no differences in the antiinflammatory effects of PLPC and DPPC, whether administered as components of rHDL or SUVs, whereas in previous studies conducted in vitro, PLPC was found to be much more effective than DPPC in terms of inhibiting VCAM-1 expression in activated HUVECs.43 The explanation for these differences is unclear.
The antiinflammatory effects of infusing rHDL and lipid-free apoA-I in the present study were achieved with remarkably small amounts of apoA-I. Each infusion contained only 25 mg of apoA-I. This amount, when given to a 3-kg rabbit (plasma volume approximately 120 mL), would have increased the plasma apoA-I concentration in the immediate postinfusion period by 0.2 mg/mL, an increase of only 40% over the preinfusion concentration of 0.5 mg/mL. Furthermore, with a fractional catabolic rate for rabbit plasma apoA-I of 0.8 pools/d,44 it would be predicted (as observed) that the apoA-I concentration would have returned to preinfusion levels by the time of euthanasia 24 hours after the last infusion. Yet despite what was no more than a minor and transient increase in the concentration of plasma HDL, the infusions resulted in an almost complete inhibition of the collar-induced arterial inflammation. This suggests that the infused rHDL, apoA-I, and SUVs had profound antiinflammatory effects that extended well beyond those resulting from a simple increase in the concentration of plasma HDL.
The mechanism underlying these potent antiinflammatory effects of infused rHDL, lipid-free apoA-I, and SUVs is unknown. Despite this, the therapeutic implications of the present findings are substantial, providing strong support for a proposition that the infusion of rHDL (or indeed lipid-free apoA-I or PC SUVs) should be investigated as first-line therapy to minimize tissue damage in devastating acute inflammatory states such as acute coronary syndromes, stroke, and ischemia-reperfusion injury.
Acknowledgments
This work was supported by a grant from the Pfizer International HDL Awards. Dr Nicholls was supported by a postgraduate research scholarship from the National Heart Foundation of Australia. Dr Drummond is supported by a Peter Doherty Fellowship from the National Health and Medical Research Council of Australia (007044). Dr Rye is a National Heart Foundation of Australia Principal Research Fellow.
References
Cockerill GW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. High-density lipoproteins inhibit cytokine-induced expression of endothelial cell adhesion molecules. Arterioscler Thromb Vasc Biol. 1995; 15: 1987–1994.
Navab M, Imes SS, Hama SY, Hough GP, Ross LA, Bork RW, Valente AJ, Berliner JA, Drinkwater DC, Laks H, Fogelman AM. Monocyte transmigration induced by modification of low density lipoprotein in cocultures of human aortic wall cells is due to induction of monocyte chemotactic protein 1 synthesis and is abolished by high density lipoprotein. J Clin Invest. 1991; 88: 2039–2046.
Moudry R, Spycher MO, Doran JE. Reconstituted high density lipoprotein modulates adherence of polymorphonuclear leukocytes to human endothelial cells. Shock. 1997; 7: 175–181.
Dimayuga P, Zhu J, Oguchi S, Chyu K-Y, Xu X-OH, Yano J, Shah PK, Nilsson J, Cercek B. Reconstituted HDL containing human apolipoprotein A-I reduces VCAM-1 expression and neointima formation following periadventitial cuff-induced carotid injury in apoE null mice. Biochem Biophys Res Commun. 1999; 264: 465–468.
Rong JX, Li J, Reis ED, Choudhury RP, Dansky HM, Elmalem VI, Fallon JT, Breslow JL, Fisher EA. Elevating high-density lipoprotein cholesterol in apolipoprotein E-deficient mice remodels advanced atherosclerotic lesions by decreasing macrophage and increasing smooth muscle cell content. Circulation. 2001; 104: 2447–2452.
Cockerill GW, Huehns TY, Weerasinghe A, Stocker C, Lerch PG, Miller NE, Haskard DO. Elevation of plasma high-density lipoprotein concentration reduces interleukin-1-induced expression of E-selectin in an in vivo model of acute inflammation. Circulation. 2001; 103: 108–112.
Paterno R, Ruocco A, Postiglione A, Hubsch A, Anderson I, Lang MG. Reconstituted high-density lipoprotein exhibits neuroprotection in two rat models of stroke. Cerebrovasc Dis. 2004; 17: 204–211.
Cockerill GW, McDonald MC, Mota-Filipe H, Cuzzocrea S, Miller NE, Thiemermann C. High density lipoproteins reduce organ injury and organ dysfunction in a rat model of hemorrhagic shock. FASEB J. 2001; 15: 1941–1952.
Spieker LE, Sudano I, Hurlimann D, Lerch PG, Lang MG, Binggeli C, Corti R, Ruschitzka F, Luscher TF, Noll G. High-density lipoprotein restores endothelial function in hypercholesterolemic men. Circulation. 2002; 105: 1399–1402.
Bisoendial RJ, Hovingh GK, Levels JHM, Lerch PG, Andresen I, Hayden MR, Kastelein JJP, Stroes ESG. Restoration of endothelial function by increasing high-density lipoprotein in subjects with isolated low high-density lipoprotein. Circulation. 2003; 107: 2944–2948.
Osborne J. Delipidation of plasma lipoproteins. Methods Enzymol. 1986; 128: 213–222.
Weisweiler P. Isolation and quantitation of apolipoproteins A-I and A-II from human high-density lipoproteins by fast-protein liquid chromatography. Clin Chim Acta. 1987; 169: 249–254.
Clay M, Rye KA, Barter PJ. Evidence in vitro that hepatic lipase reduces the concentration of apolipoprotein A-I in rabbit high-density lipoproteins. Biochim Biophys Acta. 1990; 1044: 50–56.
Matz CE, Jonas A. Micellar complexes of human apolipoprotein A-I with phosphatidylcholines and cholesterol prepared from cholate-lipid dispersions. J Biol Chem. 1982; 257: 4535.
Smith P, Krohn R, Hermanson G. Measurement of protein using bicinchonic acid. Anal Biochem. 1985; 150: 76–85.
Takayama M, Itoh S, Nagasaki T. A new enzymatic method for determination of serum choline-containing phospholipids. Clin Chim Acta. 1977; 79: 93–98.
Rainwater D, Andres D, Ford A. Production of polyacrylamide gradient gels for the electrophoretic resolution of lipoproteins. J Lipid Res. 1992; 33.
Wetterau JR, Jonas A. Effect of dipalmitoylphosphatidylcholine vesicle curvature on the reaction with human apolipoprotein A-I. J Biol Chem. 1982; 257: 10961–10966.
Laursen J, Somers M, Kurz S. Endothelial regulation of vasomation in ApoE-deficient mice: implications for interactions between peroxynitrite and tetrahydrobiopterin. Circulation. 2001; 103: 1282–1288.
Paravicini TM, Gulluyan LM, Dusting GJ, Drummond GR. Increased NADPH oxidase activity, gp91phox expression, and endothelium-dependent vasorelaxation during neointima formation in rabbits. Circ Res. 2002; 91: 54–61.
Munzel T. Afanas’ev IB, Kleschyov AL, Harrison DG. Detection of superoxide in vascular tissue. Arterioscler Thromb Vasc Biol. 2002; 22: 1761–1768.
Sukhova G, Williams J, Libby P. Statins reduce inflammation in atheroma of nonhuman primates independent of effects of serum cholesterol. Arterioscler Thromb Vasc Biol. 2002; 22: 1452001458.
Herdeg C, Oberhoff M, Baumbah A. Effects of local all-trans-retinoic acid delivery on experimental atherosclerosis in the rabbit carotid artery. Cardiovasc Res. 2003; 57: 544–553.
Shen J, Bao S, Reeve VE. Modulation of IL-10, IL-12, and IFN-gamma in the epidermis of hairless mice by UVA (320–400 nm) and UVB (280–320 nm) radiation. J Invest Dermatol. 1999; 113: 1059–1064.
Wahlefeld A. Triglycerides: determination after enzymatic hydrolysis. In: Bergmeyer H, ed. Methods of Enzymatic Analysis. 2nd ed. New York, NY: Academic Press; 1974.
Allen JK, Hensley WJ, Nicholls AV, Whitfield JB. An enzymatic and centrifugal method for estimating high-density lipoprotein cholesterol. Clin Chem. 1979; 25: 325–327.
Baker PW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. Ability of reconstituted high density lipoproteins to inhibit cytokine-induced expression of vascular cell adhesion molecule-1 in human umbilical vein endothelial cells. J Lipid Res. 1999; 40: 345–353.
Stannard AK, Khan S, Graham A, Owen JS, Allen SP. Inability of plasma high-density lipoproteins to inhibit cell adhesion molecule expression in human coronary artery endothelial cells. Atherosclerosis. 2001; 154: 31–38.
Dansky HM, Charlton SA, Barlow CB, Tamminen M, Smith JD, Frank JS, Breslow JL. Apo A-I inhibits foam cell formation in apoE-deficient mice after monocyte adherence to endothelium. J Clin Invest. 1999; 104: 31–39.
Arthur JF, Yin ZL, Young HM, Dusting GJ. Induction of nitric oxide synthase in the neointima induced by a periarterial collar in rabbits. Arterioscler Thromb Vasc Biol. 1997; 17: 737–740.
Egashira K, Zhao Q, Kataoka C, Ohtani K, Usui M, Charo IF, Nishida K, Inoue S, Katoh M, Ichiki T, Takeshita A. Importance of monocyte chemoattractant protein-1 pathway in neointimal hyperplasia after periarterial injury in mice and monkeys. Circ Res. 2002; 90: 1167–1172.
Donetti E, Baetta R, Comparato C, Altana C, Sartore S, Paoletti R, Castano P, Gabbiani G, Corsini A. Polymorphonuclear leukocyte-myocyte interaction: an early event in collar-induced rabbit carotid intimal thickening. Exp Cell Res. 2002; 274: 197–206.
Dusting GJ, Curcio A, Harris PJ, Lima B, Zambetis M, Martin JF. Supersensitivity to vasoconstrictor action of serotonin precedes the development of atheroma-like lesions in the rabbit. J Cardiovasc Pharmacol. 1990; 16: 667–674.
Justicia C, Panes J, Sole S, Cervera A, Deulofeu R, Chamorro A, Planas AM. Neutrophil infiltration increases matrix metalloproteinase-9 in the ischemic brain after occlusion/reperfusion of the middle cerebral artery in rats. J Cereb Blood Flow Metab. 2003; 23: 1430–1440.
Jordan JE, Zhao ZQ, Vinten-Johansen J. The role of neutrophils in myocardial ischemia-reperfusion injury. Cardiovasc Res. 1999; 43: 860–878.
Prestigiacomo CJ, Kim SC, Connolly ES Jr, Liao H, Yan SF, Pinsky DJ. CD18-mediated neutrophil recruitment contributes to the pathogenesis of reperfused but not nonreperfused stroke. Stroke. 1999; 30: 1110–1117.
Korthuis RJ, Granger DN. Reactive oxygen metabolites, neutrophils, and the pathogenesis of ischemic-tissue-reperfusion. Clin Cardiol. 1993; 16: I19–I26.
Sellak H, Franzini E, Hakim J, Pasquier C. Reactive oxygen species rapidly increase endothelial ICAM-1 ability to bind neutrophils without detectable upregulation. Blood. 1994; 83: 2669–2677.
Xia P, Vadas MA, Rye K-A, Barter PJ, Gamble JR. High density lipoproteins (HDL) interrupt the sphingosine kinase signaling pathway. A possible mechanism for protection against atherosclerosis by HDL. J Biol Chem. 1999; 274: 33143–33147.
Mackness MI, Durrington PN, Mackness B. How high-density lipoprotein protects against the effects of lipid peroxidation. Curr Opin Lipidol. 2000; 11: 383–388.
Navab M, Hama SY, Anantharamaiah GM, Hassan K, Hough GP, Watson AD, Reddy ST, Sevanian A, Fonarow GC, Fogelman AM. Normal high density lipoprotein inhibits three steps in the formation of mildly oxidized low density lipoprotein: steps 2 and 3. J Lipid Res. 2000; 41: 1495–1508.
Yuhanna IS, Zhu Y, Cox BE, Hahner LD, Osborne-Lawrence S, Lu P, Marcel YL, Anderson RG, Mendelsohn ME, Hobbs HH, Shaul PW. High-density lipoprotein binding to scavenger receptor-BI activates endothelial nitric oxide synthase. Nat Med. 2001; 7: 853–857.
Baker PW, Rye K-A, Gamble JR, Vadas MA, Barter PJ. Phopholipid composition of reconstituted high density lipoproteins influences their ability to inhibit endothelial cell adhesion molecule expression. J Lipid Res. 2000; 41: 1261–1267.
Kee P, Rye K-A, Taylor JL, Barrett PH, Barter PJ. Metabolism of ApoA-I as lipid-free protein or as component of discoidal and spherical reconstituted HDLs: studies in wild-type and hepatic lipase transgenic rabbits. Arterioscler Thromb Vasc Biol. 2002; 22: 1912–1917.(Stephen J. Nicholls, MBBS)