High‐Density Lipoprotein: A Novel Target for Antirestenosis Therapy

Abstract

Restenosis is an integral pathological process central to the recurrent vessel narrowing after interventional procedures. Although the mechanisms for restenosis are diverse in different pathological conditions, endothelial dysfunction, inflammation, vascular smooth muscle cell (SMC) proliferation, and myofibroblasts transition have been thought to play crucial role in the development of restenosis. Indeed, there is an inverse relationship between high‐density lipoprotein (HDL) levels and risk for coronary heart disease (CHD). However, relatively studies on the direct assessment of HDL effect on restenosis are limited. In addition to involvement in the cholesterol reverse transport, many vascular protective effects of HDL, including protection of endothelium, antiinflammation, antithrombus actions, inhibition of SMC proliferation, and regulation by adventitial effects may contribute to the inhibition of restenosis, though the exact relationships between HDL and restenosis remain to be elucidated. This review summarizes the vascular protective effects of HDL, emphasizing the potential role of HDL in intimal hyperplasia and vascular remodeling, which may provide novel prophylactic and therapeutic strategies for antirestenosis.

Introduction

Percutaneous transluminal coronary angioplasty (PTCA) and intravascular stenting are the most commonly used treatments for atherosclerosis.1 However, the rates of recurrent vessel narrowing are unacceptably high, occurring even with the use of drug‐eluting stents (DES) and thus imposing a major obstacle to the long‐term success of these therapeutic methods.2 The pathophysiology behind this may be restenosis, a process comprised of intimal hyperplasia and arterial remodeling. Many factors and mechanisms have been found to attribute to the development of restenosis.3 For example, endothelial dysfunction, lipid deposition, inflammatory reaction, and vascular smooth muscle cell (SMC) proliferation seem to play critical roles in the pathophysiology of intimal hyperplasia. While the inflammatory response, myofibroblasts differentiation, and collagen accumulation have been thought to account for the arterial remodeling, which includes constrictive and adaptive remodeling, both of these processes result in a change in the size of the vessel.1, 2, 3, 4

High‐density lipoprotein (HDL) is one of the major lipoproteins in the blood. Numerous studies have demonstrated that plasma HDL levels are inversely correlated with the incidence of coronary heart disease (CHD).5 Furthermore, there are evidence that the risk of restenosis following a vascular intervention is inversely related to HDL.6, 7 However, exact effect of HDL on the restenosis remains to be explored. The antiatherosclerotic effect of HDL mainly contributes to the cholesterol efflux, where accumulated cholesterol is transported by a variety of cholesterol transporters, including ATP‐binding cassette transporter A1 (ABCA1) or ABCG1 to HDL or its major component, apolipoprotein (apo)A‐I.8, 9 Besides its role in lipid transport, HDL has multiple biological functions including modulation of the function of endothelial cells,10 anti‐inflammation,8 antithrombosis,11 inhibition of the proliferation of SMCs.12 In this review, we critically evaluated the currently available evidence regarding the potential effect of HDL on the endothelial function, immune‐inflammation, thrombosis, SMC proliferation, fibroblasts transformation, and perivascular adipose tissue (PVAT) regulation with special emphasis on the potential role of HDL on the regulation of vascular injury, which may provide novel targets for antirestenosis therapy.

The pathophysiology of restenosis

Although the underlying mechanisms of restenosis following balloon angioplasty and stent placement may be different, the pathological characteristics responsible for arterial response to injury caused by these interventions may be similar.13 Briefly, injury of the coronary artery by balloon angioplasty or stent placement leads to endothelial injury and dysfunction, which immediately results in the accumulation and activation of platelets at the site of injury. Activated platelets release a multitude of mediators, including transforming growth factor‐β (TGF‐β), platelet derived growth factor (PDGF), chemokines, interleukin (IL)‐1β, IL‐6, which further promote the infiltration of inflammatory cells.14 The inflammatory cells infiltrating the subendothelial layer subsequently ingest cell debris and lipid deposits while simultaneously releasing more growth factors, inflammatory cytokines, and proteases, which lead to SMC migration, proliferation, and degradation of the basement membrane surrounding SMCs. These processes, taken together, lead to neointimal hyperplasia and constitute a major component of restenosis14, 15 (Figure 1 ).

Figure 1.

Schematic representation of the intimal hyperplasia and vascular remodeling in restenosis after balloon angioplasty and stent placement. (1) Injury of the coronary artery by balloon and/or stent placement leads to endothelial injury, which immediately results in the activation of platelets and thrombosis. (2) Activated platelets release a multitude of cytokines including PDGF and TGF‐β, which further promote the infiltration of monocytes/macrophages. (3) After ingesting cell debris, macrophages release more cytokines and lead to SMC migration, proliferation, and degradation of the basement membrane surrounding SMC. (4) After trans‐differentiating to myofibroblasts, adventitial fibroblasts promote the secretion of ROS, RNS, MMPs, and collagen, which play crucial role in the vascular remodeling.

Following degradation of the basement membrane, medial SMCs shift from a contractile to a synthetic phenotype, proliferation and migration, initiating the process of vascular remodeling, where inflammation, fibroblasts transformation, and extracellular matrix (ECM) formation have been shown to play crucial roles in the process.16 Although vascular remodeling has been traditionally considered as an “inside‐out” process in blood vessels, growing evidence has suggested an “outside‐in” hypothesis.16, 17 This proposes that neointima formation is partially characterized by the acquisition of migratory and proliferative ability of adventitial fibroblasts after their transformation to myofibroblasts, which further promotes the expression of growth factors and inflammatory cytokines, contributing to neointima formation and intimal thickening.16, 17 The ECM is composed of a variety of molecules, including collagen, elastin, glycoproteins, and proteoglycans.4 In a muscular coronary artery, type III collagen is the most abundant matrix protein.4, 18 Collagen is mainly synthesized in the fibroblasts and SMCs in the vascular wall.4 Many factors, including various inflammatory cytokines and growth factors, may regulate collagen synthesis and degradation.16, 19 For example, IL‐1β and TGF‐β promote fibroblast transformation, proliferation, and collagen production via small mothers against decapentaplegic (SMAD)‐dependent and ‐independent pathways.20, 21 Inhibition of matrix metalloproteinases (MMPs), a family of proteolytic enzymes strongly enhanced in atherosclerosis as well as in‐stent restenosis, prevents adventitial and constrictive arterial remodeling via reducing collagen accumulation following balloon dilation in swine.22, 23 There is a positive correlation between the excessive production of reactive nitrogen species (RNS) and reactive oxygen species (ROS) during vascular injury and the development of restenosis.24 Antioxidants can reduce restenosis, which is, in part, attributable to adaptive remodeling of the vessel wall24 (Figure 1 ).

HDL and Its Major Receptors

HDL is a lipid and protein‐containing complex with heterogeneity in size and density. ApoA‐I and apoA‐II are the most abundant apolipoproteins of HDL while phosphatidylcholine and sphingomyelin are considered as the major activated lipid components of HDL.25 Epidemiologic studies have shown that plasma levels of HDL are inversely associated with risk for cardiovascular disease.5

One of the most frequently investigated mechanisms for athero‐protective effect of HDL is the reverse cholesterol transport (RCT), in which accumulated cholesterol is transported by HDL from the vessel wall to the liver for excretion.26 In addition to its role in cholesterol transport, HDL has multiple biological functions including inhibition of the proliferation of stem cells27 and SMCs,12 modulation of the function of endothelial cells10 and fibroblasts,28 and protecting against thrombosis.11 Furthermore, HDL has been found to play crucial role in antiinflammation,29 antioxidation,30 and inhibition of matrix degradation,29, 31 suggesting a complex function for HDL in the regulation of vascular activity. Many membrane lipid transporters, especially ABCA1, ABCG1, scavenger receptor class B type 1 (SR‐B1), and sphingosine‐1‐phosphate receptor (S1PR) that bind to HDL or its major components, have been advanced as putative HDL receptors to mediate cellular responses to HDL not only through modulating intracellular contents of cholesterol, calcium, numerous kinases, and enzymes, but also through stimulating the intracellular signaling pathways9 (Table 1 ).

Table 1.

The components and major receptors of HDL in different cells to activate downstream signaling molecules. ? = unknown

Cell types/tissues	Components	Receptors	Signaling molecules	References
Endothelial cells
Human endothelial cells	HDL/apoA1	ABCA1/ABCG1	p38 MAPK, ERK1/2, JAK2	32, 33, 34, 35, 56
Human endothelial cells	HDL	SR‐B1	CaMKK/AMPK/AKT, PI3K/Akt/eNOS, Src/AKT/MAPK, ceramide	36, 37, 38, 126, 143
Human endothelial cells	HDL/S1P	S1PR	CaMKK/AMPK/AKT, S1P1‐PI3K‐Akt,	39, 80, 119
Human endothelial cells	HDL/apoA1	F(1)‐ATPase	ATP hydrolysis	40
Human endothelial cells	HDL	GPCRs	Ras/MAPK	41
Macrophages
Human macrophages	HDL/apoA1	ABCA1	JAK2/STAT3	58, 140
THP‐1 macrophages	apoA1	ABCA1	CDC42	42, 43
Human macrophages	HDL	ABCA1/ABCG1/SR‐B1	?	44
Smooth muscle cells
Human primary SMCs	HDL	SR‐B1	ERK/NF‐kB	12
Mice SMCs	ApoA‐I	?	?	45
Fibroblast
Human fibroblast	ApoA‐I	ABCA1	cAMP	46
Human fibroblast	ApoA‐I	ABCA1	CDC42	47
Human fibroblast	HDL	ABCA1/ABCG1	?	48
Adipocytes
Mice adipocytes	ApoA‐I	?	PI3K/AKT pathway	49
Mouse adipocytes	HDL/apoA1	ABCA1/SR‐B1	?	50
3T3‐L1 adipocytes	HDL	?	AMPK	51
Stem cells
Hematopoietic stem cells	HDL/apoA1	ABCA1/ABCG1	IL‐3/GM‐CSF receptor	27
Endothelial progenitor cell	HDL	?	?	88
Leukemic stem cells	HDL	ABCA1/ABCG1	?	52
Multipotential progenitor cell	HDL	ABCA1/ABCG1	Flt3‐ITD	53
Other cells
β‐cells	HDL	SR‐B1	Akt/protein kinase B	54
C2C12 myocytes	ApoA‐I	?	AMPK phosphorylation	55

ABCA1

The ATP‐binding cassette (ABC) transporters are ubiquitous membrane proteins that couple the transport of diverse substrates across cellular membranes to the hydrolysis of ATP.26 ABCA1 is a member of the ABCA subfamily, which regulates intercellular cholesterol efflux to lipid‐free apoA‐I.26 Recent studies have supported that ABCA1 plays a central role in the intercellular cholesterol clearance and the function of HDL and apoA‐I.26 Besides its role in lipid metabolism, ABCA1 mediates the direct effects of apoA‐I or HDL in endothelial cells,32 neutrophils,57 monocytes,34 fibroblasts,28 SMCs,59 hematopoietic stem cells,27 pancreatic β‐cells,60 adipocytes61 and participates in the regulation of endothelial function,62 adventitial response,63 foam cell formation,35 inflammation,34 insulin secretion,36 and platelet activation.64

Mutation of ABCA1 gene in humans causes Tangier disease (TD), a rare genetic disorder characterized by severe reductions in plasma HDL levels, accumulation of lipids in peripheral tissues, and an increased risk of cardiovascular disease.65 Knockout of ABCA1 in mice increases inflammatory cell infiltration and foam cell formation in a number of tissues, including the vessel wall, peritoneal cavity, and the blood circulation.26 Additionally, ABCA1 mRNA has been found to be strongly induced during adipose differentiation from preadipocytes to adipocytes, which accompanied with lowering in membrane cholesterol in adipocytes, suggests a necessary role of ABCA1 in preadipocytes differentiation.66 Alone or together with ABCG1 and SR‐BI, ABCA1 has recently been found to play a crucial role in stem and progenitor cell proliferation and differentiation.27 In fibroblasts, the expression of ABCA1 as well as its mediated cholesterol efflux is increased after stimulation of the TGFβ1, one of the key modulating cytokines for differentiation and proliferation of fibroblasts,67 though the detailed role of ABCA1 in fibroblasts function is poorly understood.

ABCG1

ABCG1, another ABC family member which is highly expressed in macrophage, SMC, and fibroblasts, also plays an important role in cholesterol efflux from cells.26 Overexpression of ABCG1 not only increases efflux of cellular cholesterol, but also to some extent, cellular sterols and phospholipids to HDL particles.26 Several studies have shown the crucial role of ABCG1 in the antiinflammatory effects of HDL via its predominant effect in modulating the accumulation of prominent inflammatory macrophage foam cells in various tissues such as lung, liver, spleen, or thymus.68, 69 Additionally, ABCG1 seems to have a more potent antiinflammatory effect than ABCA1 because of its predominant effect in modulating the lipid composition of membrane lipid rafts.70 Munch et al. reported that the overexpression of ABCG1 protein attenuates arteriosclerosis by markedly easing endothelial dysfunction and reducing macrophage and SMC invasion into the vascular wall.71 Nevertheless, knockout of ABCG1 was associated with significantly smaller lesions in apoE^−/− mice, which may be due to the accumulation of specific oxysterols in the brain and macrophages of ABCG1^−/−apoE^−/− mice.72 Accumulation of intercellular oxysterols is crucial for many antiatherosclerotic mechanisms, including activation of liver X receptor target genes and suppression of proinflammatory response.73, 74 These results suggest ABCG1‐mediated sterols’ metabolism plays a crucial role in the maintenance of vascular function, though the exact role of ABCG1 in occlusive vascular diseases, including atherosclerosis and potentially other cardiovascular diseases is still largely unknown.

SR‐BI

SR‐BI belongs to the class B scavenger receptor family of proteins, which are characterized with a structure similar to that of scavenger receptor CD36.75 SR‐BI is highly expressed in liver, ovary, and adrenal gland, mildly expressed in the testis and mammary gland, and is expressed only in trace amounts in the heart.51 In many cells, including macrophages, fibroblasts, hepatocytes, neutrophils, and endothelial cells, SR‐BI is required for the direct actions of apoA‐I and HDL, mediating both cholesterol efflux from cells and selective transfer of cholesteryl esters into cells.76 Recent studies have shown that SR‐BI was required for the intercellular signals induced by HDL.9 It has been shown that both the C‐terminal cytoplasmic postsynaptic density 95, PSD‐85, discs large, Dlg, zonula occludens‐1, ZO‐1 (PDZ)‐interacting domain and the C‐terminal transmembrane domain of SR‐BI are required for HDL‐induced signaling.77 Emiel et al. have recently found that reconstituted HDL (rHDL) containing apoA‐I and phosphatidylcholine strikingly suppressed primary human SMC proliferation via inhibition of Extracellular signal‐regulated kinases (ERK) phosphorylation and the upstream signaling proteins, PI3K and phosphorylated Akt (pAkt).78 Knockdown of the SR‐BI attenuated HDL‐induced inhibition of SMC chemokine expression and proliferation,54 suggesting that HDL has a potential role in regulating the restenosis after angioplasty and intravascular stenting via a SR‐BI‐ dependent manner.

S1PR

Sphingosine‐1‐phosphate (S1P) is a lysosphingolipid that has been shown to be a major lipid component of plasma HDL and linked to an array of biological functions, such as cell growth, survival, and suppression of programmed cell death.79 S1P is recognized with high affinity by S1PR, where five members of the S1PR family, including S1PR1–5, have been cloned, all of which can bind to and be activated specifically by S1P to initiate the persistent activation of downstream signaling molecules, including phospholipase C, Akt, ERK, and small G‐proteins.56 Recent studies indicate that S1P/S1PR pathway is critical in mediating many of the cardiovascular effects of HDL, including the ability to promote angiogenesis, protect against ischemia/reperfusion injury, and inhibit/reverse atherosclerosis.81 However, the synthesis and secretion of S1P requires its release from cells in order to bind with receptors by certain transporters and carriers. Some studies have shown ABCA1, ABCG1, and SR‐BI play pivotal roles in this process, suggesting the presence of a functional crosstalk among ABC transporters, SR‐BI, and SIPR.82

Potential Function of HDL in Antirestenosis

HDL regulation of endothelial function

Interventional procedures in coronary artery disease cause marked vascular injury, including de‐endothelialization, which is associated with thrombus formation, neointimal thickening, and abnormal responses to endothelium‐dependent agonists.83 Moreover, antiproliferative drugs used in commercially available DES have dramatically reduced the long‐term rate of restenosis, but, have been found to delay stent endothelialization and are associated with late thrombotic events.84 Although newer technological approaches, including modification of the stent surface with antibodies to selectively bind circulating endothelial progenitor cells (EPCs) or increase endothelial cell migration from adjacent healthy endothelium, have been used to enhance stent endothelialization,85 the low number of circulating EPCs and impaired endothelial function in CHD patients seem to be the major obstacle for these methods. Stimulating EPCs expansion and/or mobilization in CHD, therefore, provide novel therapeutic options for re‐endothelialization following angioplasty or stenting.86 Indeed, rHDL can stimulate the differentiation of EPCs and enhance EPCs‐mediated endothelium repair in mice.64, 87 Up‐regulation of HDL levels by transferring the adenoviral human apoA‐I (AdA‐I) has been found to increase the number and function of EPCs and promote incorporation of EPCs in Balb/c allografts transplanted paratopically in C57BL/6 ApoE^−/− mice, thus attenuating transplant arteriosclerosis.89 In human subjects, Oostrom et al. for the first time showed a beneficial effect of increasing HDL levels on EPC biology, and found that systemic infusion of rHDL significantly increased availability and function of EPCs in patients with type‐2 diabetes (DM2).90 Multiple mechanisms have been described for the effect of HDL on EPCs, such as stimulating EPCs differentiation, proliferation, and migration.91 Recently, HDL has been found to regulate the proliferation of hematopoietic stem and multipotential progenitor cells,27 revealing a novel mechanism for the effect of HDL on EPCs. However, the precise mechanism has yet to be elucidated.

Nitric oxide (NO), synthesized by endothelial nitric oxide synthase (eNOS), is a key molecule preventing the detrimental consequences of arterial injury on the vascular wall.92 NO stimulates endothelial cell migration and reorganization, inhibits vascular smooth muscle cell (VSMC) migration and proliferation, and platelet adhesion to the vessel wall, all of which leads to the inhibition of both vascular negative remodeling and neointimal formation after vascular injury.93 Many proatherogenic factors, including oxidized LDL (oxLDL) and inflammatory cytokines, have been found to impair eNOS activity in endothelium.94 OxLDL can cause eNOS redistribution to an intracellular locale, resulting in an attenuated capacity to activate the enzyme, through the CD36‐mediated depletion of caveolae cholesterol in endothelial cells,70 and this negative effect is reversed by HDL, which may be due to the membrane cholesterol translocation induced by HDL that prevents eNOS displacement from caveolae and restores acetylcholine‐induced stimulation of the enzyme.95 Sonika and co‐investigators found SR‐BI is a major receptor for sensing plasma membrane (PM) cholesterol and transporting cholesterol from the endothelium to HDL.96 Mutation of a highly conserved C‐terminal transmembrane domain of SR‐BI decreased the interaction of PM cholesterol with SR‐BI and impaired HDL‐induced signaling, resulting in the inhibition of HDL‐induced activation of eNOS.72 Recently, Christian et al. found HDL from patients with acute coronary syndrome (ACS) does not have endothelial antiinflammatory effects and cannot stimulate endothelial repair compared with HDL from healthy subjects, which is attributed to the impairment of HDL from ACS patients to induce eNOS‐dependent endothelial NO production.97

Evidence has accumulated that endothelial cell apoptosis after angioplasty contributes to the pathogenesis of restenosis.98 Several studies have shown HDL can inhibit inflammatory cytokines, such as tumor necrosis factor‐α (TNF‐α), and oxLDL‐induced endothelial cell apoptosis.99, 100 Such effects of HDL are mediated by its major components, such as apoA‐I and S1P via a variety of receptors.75, 76 Claudia and colleagues reported that the stimulation of cell surface F1‐ATPase activity by apoA‐I inhibits endothelial cell apoptosis and promotes its proliferation.101 Overexpression of human ABCA1 in endothelial cell decreases inflammation and apoptosis in atherosclerotic lesions.102 HDL‐associated lysosphingolipids inhibit endothelial cell apoptosis by triggering growth factor deprivation, which resulted from the HDL‐induced attenuation of the dissipation of mitochondrial potential, oxygen‐derived free radical generation, cytochrome C release to the cytoplasm, and activation of caspase 3 and caspase 9.75 The S1P can stimulate the migration and survival of human endothelial cells via S1P receptors S1P1 and S1P3‐dependent phosphoinositide‐3 (PI3) kinase and ERK pathway.103 These results suggest that signaling by lysophospholipid components of HDL may be important for the inhibition of endothelial cell apoptosis. Moreover, Smart and colleagues revealed a novel ligand‐independent apoptotic pathway in endothelial cells induced by SR‐BI, and regulated by eNOS and HDL.104 These investigators found that in healthy cells, the SR‐BI apoptotic pathway is turned off by eNOS and HDL which prevents inappropriate apoptotic damage to the vascular wall, suggesting a crucial role of HDL/SR‐BI in maintaining the endothelial function.80

Recently, an apoA‐I binding protein (AIBP) has been found to accelerate cholesterol efflux from endothelial cells to HDL and thereby regulates angiogenesis. In this process, ABCA1 and ABCG1 mediate cholesterol depletion stimulated by AIBP and HDL, which further reduces lipid rafts and interferes with vascular endothelial growth factor receptor 2 dimerization and signaling, inhibiting VEGF‐induced angiogenesis in vitro, and mouse aortic neovascularization ex vivo.10 These results suggest the essential role of ABCA1 and ABCG1 in endothelial function and angiogenesis (Figure 2 ).

Figure 2.

Regulation of endothelial function by HDL. (1) HDL stimulate the differentiation of endothelial progenitor cells (EPCs), up‐regulate the function of EPCs and enhance EPCs‐mediated endothelium repair; (2) Under hypercholesterolemia, impaired eNOS activity was repaired by HDL via SR‐B1‐dependent manner, which results in the increase of NO that leads to the reorganization of endothelial cell and inhibition the SMC proliferation and platelet activation; (3) HDL inhibit inflammatory cytokines and oxLDL‐induced endothelial cell apoptosis through ABCA1‐dependent pathway.

HDL regulation of platelet and thrombosis

Platelet activation has been reported after coronary angioplasty, especially in DES, and is correlated with restenosis and the risk of late stent thrombosis.105 Platelet activation is inversely correlated with HDL levels in human, suggesting that HDL has antiplatelet properties.106 In vitro, HDL can interfere with LDL and ox‐LDL‐induced platelet activation.107, 108 In animal models, infusion of apoA‐I mimetic peptide into rats or mice inhibits platelet aggregation.109, 110 Furthermore, the administration of rHDL in human reduced collagen‐induced platelet aggregation111 and platelet‐activating factor (PAF) acetylhydrolase activity,112 further supporting the concept that HDL inhibits platelet activation in vivo. Deletion of the HDL receptor, SR‐BI, in mice modulates thrombosis susceptibility.113 Lacking of the apoA‐I or SR‐BI in mice also had a prothrombotic phenotype,114 suggesting SR‐BI is essential for maintaining normal platelet function and prevention of thrombosis by HDL.11 ABCA1 is also expressed in human platelets, and impaired platelet activation has been found in TD, a disorder caused by ABCA1 defects.40 Scott syndrome, a rare congenital bleeding disorder, is due to the defect in platelet function characterized by a failure to expose phosphatidylserine (PS) to the outer leaflet of the platelet PM.115 The mutation in ABCA1 might contribute to the defective PS translocation,116 suggesting ABCA1 also plays a crucial role in modulating platelet activation.

Aside from the direct effect on platelet activation, HDL may also act indirectly on platelet activation. For example, HDL may regulate platelet function by down‐regulating the release of PAF or by up‐regulating NO synthesis and release from endothelial cells.117 Moreover, HDL down‐regulates the synthesis of endothelial cell thromboxane A2 (TXA₂) and up‐regulates endothelial cell prostacyclin (PGI₂) production, which can decrease platelet aggregation and blunt leukocyte‐endothelial cell interactions.118 Recently, Alexander et al. found intravenous apoA‐I infusion abolished histamine‐induced platelet‐endothelial interactions, which are associated with up‐regulation of eNOS.90 Moreover, HDL inhibits many of the adhesion factors expressed in the vascular endothelium, which is crucial for inhibition of the initiation of coagulation and platelet activation33, 119 (Figure 3 ).

Figure 3.

The effect and mechanisms of HDL‐induced platelet activation and thrombosis. HDL inhibits platelet activation by a variety of mechanisms (right). (1) HDL abolishes LDL or ox‐LDL‐mediated platelet activation via decreasing the expression of P‐selective receptor, ROS, and CD40 ligand (CD40L) signal. (2) After infusion of apoA‐I or apoA‐I mimetics, HDL decreases the level of thromboxane A2 (TXA₂), prostacyclin H₂ (PGH₂), PGE₂, and inhibits the activation of fibrinogen/GPIIa/IIIb signal pathway via SR‐B1‐dependent manner, which prevents the activation of platelet in hyperlipidemic mice. HDL can also act indirectly on platelet activation and regulating the function of endothelial cells (left). (1) HDL down‐regulates the release of PAF, VCAM‐1, and TXA₂ via inhibition of the expression of acetyl‐CoA: lysoPAF acetyltransferase (lyso‐PAF) via SR‐B1 or S1P3‐dependent signaling pathway, such as Src or PI3K. (2) HDL up‐regulates endothelial cell PGI₂ and NO production via SR‐B1, which can decrease platelet aggregation and activation.

HDL and regulation of inflammation

Inflammation is a host defense mechanism to remove the injurious stimuli. However, inflammatory response after angioplasty, where macrophages and platelets release lot of cytokines and growth factors that induce vascular SMC proliferation and accumulation within the intima, plays a crucial role in the progress of the restenosis.120 The epidemiological studies have shown an inverse association between HDL levels and C‐reactive protein (CRP), a prototypical marker of inflammation.121 In animal models, normal HDL, apoA‐I, and apoA‐I mimetic peptides, respectively, prevent the inflammatory reaction characteristic of atherosclerosis and intimal thickening after balloon injury.37, 122, 123, 124 In normocholesterolemic rabbit, lipid‐free apoA‐I, upon infusion, incorporates rapidly into the HDL fraction and inhibits the activation of inflammatory cytokine‐exposed human coronary artery endothelial cells in vitro.125 In hypercholesterolemic rabbits, recombinant apoA‐I Milano, a naturally occurring mutated variant of the apoA‐I, significantly reduced intimal thickening and macrophage content after balloon injury without a change in arterial total cholesterol content, suggesting a direct antiinflammatory effect of apoA‐I on vascular injury.100 In collar‐induced vascular injury of rabbits, intravenous infusion of lipid‐free apoA‐I significantly decreased endothelial vascular cell adhesion molecule‐1 (VCAM‐1) and intercellular adhesion molecule‐1 (ICAM‐1) expression, reduced intima/media neutrophil infiltration, and increased 3β‐hydroxysteroid‐δ24 reductase (DHCR24) and heme oxygenase‐1 (HO‐1) mRNA levels.102 In vitro, rHDL inhibited TNF‐α‐induced VCAM‐1 and ICAM‐1 expression in human coronary artery endothelial cells by increasing the expression of DHCR24 and HO‐1.102 Infusion of rHDLs, apoA‐I, or apoA‐I 5A peptide inhibits acute inflammation and oxidative stress induced by inserting a collar around the carotid artery in rabbit, displaying direct antiinflammatory and anti oxidant properties of HDL repletion.127

As an important determinant of the innate immune response to gram‐negative bacteria, lipopolysaccharide (LPS) has been verified to be associated with angiographic restenosis in different models of balloon and stent injury.128, 129 HDL and apoA‐I have been found to bind and neutralize the bioactivity of LPS and inhibit the LPS‐induced systemic inflammation and atherosclerosis.130, 131, 132 Also, apoA‐I mimetic peptide 4F alters the assembly of Toll‐like receptor (TLR) ligand complexes via disruption of rafts in cell membranes and inhibits proinflammatory gene expression in macrophages, thus attenuating the responsiveness of macrophages to LPS.133 Pathogen‐ or damage‐associated molecular patterns interact with receptors such as membrane‐bound TLRs or cytosolic Nod‐like receptors (NLRs) and initiate a signaling cascade that leads to the transcription of inflammatory genes in macrophages.134 In human subjects with low HDL levels, the activated proinflammatory state of monocytes/macrophages constitutes a novel parameter of risk associated with HDL deficiency.135 Additionally, rHDL attenuates neutrophil activation in peripheral vascular disease in humans.33 Both native HDL and apoA‐I in vitro abolished the activation of macrophages and neutrophils induced by proatherogenic factors, such as LPS, inflammatory cytokines, and oxLDL.26 Macrophage ABCA1 has been definitively confirmed to be involved in the HDL‐induced expressional down‐regulation of CD11b and the monocyte transmigration.136 ApoA‐I inhibits CD40 proinflammatory signaling via ABCA1‐mediated cholesterol efflux in macrophages by modulating membrane cholesterol.137 ABCG1 appears to have a more potent role in modulating macrophage inflammatory response than ABCA1, perhaps reflecting a predominant role of this transporter in modulating the lipid composition of PM lipid rafts.8 Yvan‐Charvet and colleagues reported the antiinflammatory effect of HDL and apoA‐1 by promoting cholesterol efflux via combination effects of ABCG1 and ABCA1 with consequent attenuation of TLR signaling.46 In ABCA1/ABCG1‐deficient macrophages, inflammatory genes are significantly increased via TLR4‐ and MyD88/TRIF‐mediated signal transduction for the membrane cholesterol accumulation.46, 138 These results suggest the cholesterol export activity of HDL, which regulates the membrane lipid raft‐related proinflammatory signal transduction, could account for its potent antiinflammatory properties. Beyond its cholesterol transport, HDL has been found to inhibit the proliferation of hematopoietic stem cells and multipotential progenitor cells, suggesting a novel antiinflammation mechanism of HDL acting at an earlier stage in the leukocyte life cycle than the subsequent antiinflammatory occurring in the vessel wall.27, 139 In addition, two candidate STAT3 docking sites have been found in cytosolic domains of ABCA1.116 After interaction with ABCA1, apoA‐I rapidly activates the JAK2/STAT3 signal pathway, which suppresses inflammation through a posttranscriptional regulation of inflammatory cytokine mRNA decay by tristetraprolin in macrophages, suggesting that the intercellular signal pathway stimulated by HDL or apoA‐I might be a novel mechanism for its potent antiinflammatory properties34, 116 (Figure 4 ).

Figure 4.

The primary mechanisms of HDL on the suppression of vascular intimal inflammatory response. (1) HDL inhibits the proliferation of hematopoietic stem cells (HSCs) via ABCA1/ABCG1 in hypercholesterolemia and reduces the accumulation of inflammatory cells such as macrophages in subintima. (2) ABCA1‐mediated cholesterol efflux decreases intercellular proinflammatory signal, such as Toll‐like receptor 4 (TLR4), which inhibits the secretion of inflammatory cytokines. (3) The interaction of HDL with ABCA1 dramatically increases phosphorylation of JAK2 and thus activates STAT3 to further up‐regulate tristetraprolin (TTP). This suppresses inflammation through a posttranscriptional regulation of inflammatory cytokine mRNA decay. (4) HDL inhibits the expression of multiple adhesion molecules in endothelial cells, and thus inhibits adhesion and infiltration of inflammatory cells via multiple receptors, including SR‐BI and S1PR.

In endothelial cells, HDL inhibits the expression of adhesion molecules, which are crucial for the adhesion and infiltration of inflammatory cells via multiple receptors, including SR‐BI,141 SR‐BI adaptor protein PDZK1,142 AMP‐activated protein kinase (AMPK),119 eNOS,119 and S1P receptors.117 SR‐BI has also been found to mediate the inhibitory effect of S1P on TNFα‐induced expression of VCAM‐1 in human umbilical venous endothelial cells.144 Endothelial expression of human ABCA1 using the endothelial‐specific Tie2 promoter in mice significantly decreased vascular endothelial inflammation and apoptosis, suggesting a novel role of ABCA1 in endothelial inflammatory response.78 Angela et al. found EC isolated from ABCG1‐deficient mice exhibited the increased endothelial activation such as increased secretion of KC, MCP‐1, IL‐6, and promoted monocyte‐endothelial interactions compared to C57BL/6 mice,145 indicating that ABCG1 can also exert a regulating role in the expression of endothelial inflammatory cytokines.

Besides mediating vascular inflammatory response related to vascular disease, HDL also play an important role in the regulation of inflammation of other tissues such as adipose tissues and small intestine, suggesting diverse antiinflammatory actions of HDL in different cell types.37, 146 The net effect of these antiinflammatory activities in atherosclerosis and restenosis, however, is poorly understood. To further investigate the cross talk of these processes may provide new insight into the relationship between inflammation, atherosclerosis, and other cardiovascular diseases.

HDL regulation of VSMC proliferation and migration

The diverse functions manifested by SMCs in the vessel wall have well been considered to reflect plasticity in the differentiation status of SMCs.12, 147 After responding to atherogenic lipoproteins, such as oxLDL, SMCs convert to foam cells and promote the expression of proinflammatory cytokines, proliferation and migration, which are necessary for the pathogenesis of neointimal hyperplasia.12, 123 Studies in cell culture have revealed that HDL has numerous actions on vascular SMCs in atherogenic conditions. To begin, HDL promotes the cholesterol efflux from SMCs and attenuates human SMC‐derived foam cell formation via an ABCA1‐dependent process.148 In human intimal and medial SMCs, ABCA1 expression and apoA‐I binding are impaired, suggesting ABCA1 activity is an important regulating target for abnormal SMCs function, especially in foam cell formation of the intima.35, 123 Second, HDL inhibits inflammatory chemokine‐stimulated proliferation of SMCs, which may be related to the SR‐BI‐dependent deactivation of PI3K/AKT pathway.12 Moreover, HDL impedes vascular SMCs proliferation and S‐phase entry, and this results from the up‐regulation of prostacyclin (PGI2) and cyclooxygenase‐2 (COX‐2) expression.149, 150 HDL has also been found to inhibit PDGF‐induced migration of vascular SMCs, which is attributed to the S1P2 receptor‐dependent manner.151, 152 Studies of vascular SMCs and rat aortic explants have revealed the HDL‐dependent inhibition of MCP‐1 production was accompanied by the suppression of ROS, which is associated with SR‐BI and S1P3 receptor in vascular SMCs, suggesting inhibition of NAD(P)H oxidase activity by HDL‐associated lysosphingolipids may constitute an important mechanism by which HDL exerts its potent antiatherogenic effects.30 ApoA‐II is the second major apolipoprotein of HDL, yet its pathophysiological roles in the development of atherosclerosis remain unknown. Transgenic human apoA‐II in rabbits has been demonstrated to reduce SMCs in atherosclerosis lesions.153 Other protein components of HDL, such as apoD and apoE, have also been found to inhibit the PDGF‐induced vascular SMCs migration and proliferation, suggesting a hypothesis that beneficial effects of HDL likely arise from a wide variety of components, possibly working synergistically.154, 155

HDL regulation of adventitia function

Although much emphasis has been placed on investigating atherosclerosis and restenosis through intimal mechanisms, recent research suggests the critical role of adventitia in coordinating the progression of the disease.156 In the setting of both percutaneous coronary angioplasty and the placement of a circumferential silastic collar in animal models, damage to the adventitia has been found to precede neointimal formation and lesion development.157, 158 Several studies have demonstrated the presence of inflammatory cells, including T lymphocytes and macrophages in the adventitia adjacent to atherosclerotic lesions, suggesting the adventitia may contribute to the adaptive and innate immune responses that regulate atherosclerosis.159, 160 Recently, adventitial fibroblasts have been found to differentiate into myofibroblasts in response to vascular injury and further promote the expression of growth factors and inflammatory cytokines, underscoring a direct correlation between the adventitial fibroblasts and inflammation.17 Although the direct evidence for HDL in the adventitia function is lacking at present, the reciprocation lies in the plasma HDL and adventitia.161, 162 Compared with wild‐type bone marrow (BM) transplanted mice, ABCA1/G1 deficiency BM transplanted mice showed increased atherosclerotic lesion complexity and inflammatory cell infiltration into the adventitia.39 Circulating HDL in abdominal aortic aneurysm patients displayed an impaired capacity to inhibit intraluminal thrombus and adventitia‐initiated ROS production, and restoring HDL functionality reduces the ROS production within adventitia.137 The presence of HDL in the adventitia surrounding atherosclerotic lesions has long been recognized.138 Indeed, the most important efflux route for HDL esterified cholesterol is through the vasa vasorum and lymphatics in the outer media and adventitia, whereas LDL esterified cholesterol predominantly leaves intima via the lumen of the artery.138 Christopher et al. has also found the accumulation of HDL cholesterol in adventitial lipid in hyperlipidemic diet‐feeding apoE^−/− mice.163 These results suggest potentially a crucial role of HDL/apoA1 in adventitia function.

Collagen accumulation in the vascular adventitial plays a major role in the development of restenosis, after both balloon angioplasty and stenting.164 The collagen molecule is mainly synthesized in the fibroblasts and SMCs under normal conditions.165 Nevertheless, fibroblasts are phenotypically and functionally heterogeneous and can be phenotypically transferred to myofibroblasts, which display more contractile, proliferative, migratory and secretory activities, and thus playing a key role in arterial remodeling.16 The transformation of fibroblasts to myofibroblasts is modulated by a variety of factors including transforming growth factor‐β1 (TGF‐β1), IL‐1β, cholesterol.166 Different signal pathways, such as SMAD and cAMP, have been implicated to regulate the differentiation of myofibroblasts.142 Despite extensive study of the cholesterol efflux to HDL in fibroblasts, relatively little is known about the role of HDL on the pleiotropic function of fibroblasts, such as differentiation, proliferation, and migration.167, 168 Indeed, a recent study has confirmed that D‐4F, an apoA‐1 mimetic, can decrease TGF‐β1, fibroblasts specific protein‐1 and collagen deposition in ovalbumin (OVA)‐sensitized mice.169 In addition, patients with TD, a HDL deficiency syndrome that is characterized by impairment of HDL3‐mediated lipid efflux and ABCA1 gene mutation, present with abnormal growth and cell cycle of fibroblasts indicating a potential regulating role for HDL in fibroblasts function and collagen accumulation.170 Investigation underlying HDL and apoA‐I modulation of intracellular signaling revealed that diverse intracellular signaling events can be modulated by HDL or apoA‐I in fibroblasts including mitogen‐activated protein kinase (MAPK) and cyclic adenosine monophosphate (cAMP),171, 172 suggesting the possibility that the HDL has multivariable regulatory functions in addition to its lipid export activity in the fibroblasts.

PVATs have been demonstrated to be powerful endocrine tissues, which could promote inflammatory changes in the peri‐adventitial fat, contributing to vascular restenosis.173 Recently, HDL and apoAI have been shown to inhibit inflammatory effect in adipocytes and adipose tissue via ABCA1, ABCG1, and SR‐B1‐dependent pathways, which is similar to their effects in other cell types.37 Moreover, lack of ABCA1 specifically in adipocytes (ABCA1 (‐ad/‐ad)) increases cholesterol and triglyceride stores in adipose tissue, promotes the development of enlarged fat pads, and increases body weight.174 These results suggest a role of HDL in PVAT function, though the exact physiological and pathophysiological roles of this function remain to be elucidated (Figure 5 ).

Figure 5.

The primary effect of HDL on adventitia function. Damaging the vascular by coronary angioplasty leads to the accumulation of inflammation cells, including macrophages and T‐lymphocytes, from the adventitial vasa vasorum and intima, which together with perivascular adipocytes secrete multiple factors such as TGFβ1 and ROS and induce fibroblasts differentiation to myofibroblasts. HDL originated from adventitial vasa vasorum or intimal has the potential to inhibit fibroblast differentiation to myofibroblasts via suppressing MAPK and TGFβ1‐dependent pathway or promoting cholesterol efflux, which may reduce the oversecretion of collagen, MMPs, ROS, and inhibit vascular remodeling.

Conclusion

Early epidemiological studies have identified the association of low HDL levels with the risk of restenosis following a vascular intervention. Therefore, raising HDL levels could provide a novel strategy to address the restenosis in patients with PTCA. HDL modulates endothelial function, platelet activation and thrombosis, inflammatory cytokines secretion, foam cells formation, and SMC proliferation in the intima, resulting in the inhibition of intimal hyperplasia. On the other hand, HDL has the potential to regulate the adventitial fibroblasts differentiation and perivascular adipocytes function, which may organize with cholesterol transport to reduce vascular remodeling and eventually inhibit restenosis. Both lifestyle changes and pharmacological agents, such as niacin, fibrates, thiazolidinediones, bile acid sequestrants, can raise HDL levels. Nevertheless, there is still a lack of direct evidence to support the causal relationship between raising HDL by these drugs and the slowing of restenosis progression in human. HDL from patients with coronary artery disease (CAD) is associated with many acute phase proteins, supporting the proposal that HDL is changed in the pathological state of CAD, and functional properties of HDL must be considered when choosing a therapeutic strategy for antiatherosclerosis. ApoA‐I is the most abundant and most extensively studied component of HDL. Besides its ability to undergo delipidation/relipidation both in vitro and in vivo via interacting with the ABCA1, ABCG1 and SR‐B1, apoA‐I has gained increasing recognition for its antioxidant and anti‐inflammatory properties and even as a modulator of innate immunity. Nevertheless, apoA‐I is a large protein that requires venous administration. The smaller apoA‐I mimetic peptides, such as 4F and 5A, have become targets for pharmacologic development in the therapeutic management of human atherosclerosis. ApoA‐I mimetic peptides have a class amphipathic helix found within the lipid‐binding domains of apoA‐I, accordingly have function and structure similar to apoA‐I. Some apoA‐I mimetic peptides, such as apoA‐I Milano, 5A, and 4F, have been shown to inhibit balloon or collar‐induced intimal thickening in animal models. Therefore, the generation of smaller, easier to manufacture apoA‐I mimetic peptides represents an important strategy for reducing restenosis in patients after angioplasty and should be further investigated for evaluation of clinical outcomes in these patients.

Financial and Competing Interest Disclosure

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.