rH,6

idaty,verershicUn

Opinionthe leading cause of death worldwide [1,2]. The mostimportant underlying cause of CHD is atherosclerosis ofthe coronary arteries, the vessels that supply blood to theheart and that are therefore essential for supporting car-diac function [1,3]. Atherosclerosis is a complex diseasethat results from the accumulation of cholesterol in mac-rophage cells within the arterial wall [4,5].

Cholesterol, therefore, is a central molecule in CHDpathogenesis, and strategies to modulate cholesterol levelsin the body occupy a central role in CHD therapy [6].Cholesterol is insoluble in aqueous solution and requireslipoprotein carriers for transport to peripheral tissues fromthe liver, where it is synthesized, or the gut, where it isabsorbed [3,7]. Lipoproteins are a class of naturally occur-ring nanoparticles that solubilize lipids, especially choles-terol, for systemic circulation [3,7]. Low-density lipoproteins(LDLs) carry cholesterol from the liver to peripheral tissues[3,7]. In excess, LDLs increasingly adhere to arterial walls,traverse the vascular endothelial cell barrier, depositwithin

transport [3]. Another major cholesterol-transporting mol-ecule is high-density lipoprotein (HDL) [3,7]. Multiplestudies show that levels of HDL are inversely correlatedwith incidence of CHD [3,17,18]. HDLs are thought to beinvolved in a process known as reverse cholesterol trans-port (RCT), removing cholesterol from the circulation anddeveloping atherosclerotic lesions, and transporting it tothe liver for biliary excretion [3,7]. In addition to removingcholesterol from the body, HDL exerts other atheroprotec-tive effects, such as reducing inflammation of the endothe-lium, preventing endothelial dysfunction, and acting as anantioxidant [17,19]. Accordingly, increasing levels of HDLin the blood represents a promising therapeutic strategy tocombat atherosclerosis [11,20].

In addition to therapeutic implications, HDL biologymight have diagnostic implications as a method for earlydetection of atherosclerosis. Many deaths due to CHDoccur without warning and with few prior symptoms[21,22]. Therefore, imaging and early identification ofatherosclerotic plaques are of particular interest[23,24]. Molecular imaging, or imaging carried out at

Corresponding authors: Mirkin, C.A. (chadnano@northwestern.edu); Thaxton,C.S. (cthaxton003@md.northwestern.edu).Nanotechnology fohigh-density lipoproAndrea J. Luthi1, Pinal C. Patel2, CarolineChad A. Mirkin1,4 and C. Shad Thaxton4,5

1Department of Chemistry, Northwestern University, 2145 Sher2 Interdepartmental Biological Sciences, Northwestern Universi3 Feinberg Cardiovascular Research Institute, Northwestern Uni4 International Institute for Nanotechnology, Northwestern Univ5 Feinberg School of Medicine, Department of Urology, 303 E. C6 Institute for BioNanotechnology and Medicine, Northwestern

Atherosclerosis is the disease mechanism responsiblefor coronary heart disease (CHD), the leading cause ofdeath worldwide. One strategy to combat atherosclero-sis is to increase the amount of circulating high-densitylipoproteins (HDL), which transport cholesterol fromperipheral tissues to the liver for excretion. The process,known as reverse cholesterol transport, is thought to beone of the main reasons for the significant inverse cor-relation observed between HDL blood levels and thedevelopment of CHD. This article highlights the mostcommon strategies for treating atherosclerosis usingHDL.We further detail potential treatment opportunitiesthat utilize nanotechnology to increase the amount ofHDL in circulation. The synthesis of biomimetic HDLnanostructures that replicate the chemical and physicalproperties of natural HDL provides novel materials forinvestigating the structure-function relationships of HDLand for potential new therapeutics to combat CHD.

Coronary heart disease and HDLCoronary heart disease (CHD) remains the leading cause ofdeath in the USA despite therapeutic advances, and is now1471-4914/$ see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2syntheticteins. Ko1, R. Kannan Mutharasan3,

n Road, Evanston, IL 60208, USA2145 Sheridan Road, Evanston, IL 60203, USAsity, 303 E. Chicago Avenue, Chicago, IL 60611, USAity, 2145 Sheridan Road, Evanston, IL 60208, USAago Avenue, Tarry 16-703, Chicago, IL 60611, USAiversity, 303 E. Superior, Suite 11-131, Chicago, IL 60611, USA

the intimal layers of the arterial wall, are oxidized, andinitiate atherosclerosis [4,5]. Oxidized LDL is taken up byinvading macrophages to form pathognomonic foam cells atarterial sites prone to lesion development [4,5]. Progressivevessel remodeling at these siteswithin the coronary arteriespromotes the development of CHD [5,8].

Many drugs have been developed to lower LDL-choles-terol (LDL-c) [9]. One particular class of LDL-loweringmedications, the hydroxymethylglutaryl CoA reductaseinhibitors (more commonly known as statins), have beenshown to decrease CHD-related deaths [10,11]. In oneinvestigation, the risk of coronary death was reduced by42% over the course of the study [12]. However, the benefitsof statin therapy must be weighed against potential sideeffects such asmuscle and liver toxicity that, in some cases,remove statins as a treatment option [1315]. Further-more, despite the success of statins and other pharamaco-logic and revascularization therapies designed to treat thisdevastating disease [16], CHD remains the most commoncause of death in the world [2]. Thus, a clear need exists foradditional strategies to combat atherosclerosis.

LDL is only one of the molecules involved in cholesterol010.10.006 Trends in Molecular Medicine, December 2010, Vol. 16, No. 12 553

the cellular level, is now possible owing to advances inimaging technology [23]. The development of novel com-puted tomography (CT) and magnetic resonance imaging(MRI) contrast agents has made possible the detection ofsmall changes in specific tissues using non-invasive imag-ing techniques [23]. Combining these imaging modalitieswith structures that mimic molecules involved in choles-terol transport, such as HDL, might facilitate detection ofchanges in the arteries due to atherosclerosis and thusearly identification and treatment of atherosclerosis[23,24].

Relationship between natural HDL structure andfunctionHDL is a dynamic natural nanoparticle that is constantlybeing remodeled as it transports and transfers cholesterolto cells and other lipoproteins [25]. Constant remodelingleads to HDL heterogeneity, which has been shown toinfluence its function (Box 1 and Figure 1) [17,19,26].Natural HDL ranges in size from 6 to 13 nm in diameterand has a density of 1.0631.210 g/mL [3,27].

It is being increasingly appreciated that all HDLs arenot created equal. Studies demonstrate that the size andcomposition of HDL can be determining factors in the roleHDL plays in the body and influence its ability to exertatheroprotective effects [17,19]. Research shows a strong

Box 1. Formation of different HDL species

Apo AI, the main protein component of HDL, modulates the form andfunction of HDL species [3,25] as supported by computational studiesof nanodisc structure [53]. Initially, Apo AI is secreted from cells of theliver and small intestine free of phospholipids and cholesterol [3,75].Apo AI acquires phospholipids and free cholesterol from cellmembranes through interaction with membrane-bound ATP bindingcassette A1 receptor (ABCA1) [3,7,17]. The HDL species formed iscalled nascent HDL and resembles a phospholipid bilayer disk withtwo molecules of Apo AI wrapped around the disk in a belt-likeconfiguration [3,17,63]. Molecules of free cholesterol interdigitate the160 phospholipids present in the disc core [3,76]. Nascent HDL is thesmallest, most cholesterol-poor HDL subspecies [3].Nascent HDL acquires free cholesterol through interaction with

ABCA1, another ATP binding cassette receptor (ABCG1) and thescavenger receptor B1 (SR-B1) [3,7,17]. Maturing HDLs activate theenzyme lecithin:cholesterol acyltransferase (LCAT), which esterifiesfree cholesterol bound by the growing HDL particle [3,17]. Esterifiedcholesterol migrates to the center of the disk, resulting in expansionand a more spherical shape [3,17,77]. Another enzyme, cholesterylester transfer protein (CETP), mediates the exchange of cholesterylesters from HDL for triglycerides from triglyceride-rich lipoproteins,such as LDL [17,78]. The hydrophobic triglycerides also move to thecenter of HDL, maintaining the spherical shape [17]. During theremodeling process, there can also be an exchange of Apo AIproteins, and HDL species contain between two and four proteinsper particle [3,25,73]. The differential acquisition and movement ofcholesterol into versus out of nascent HDL contributes to its growthinto spherical, mature HDL, and this growth process accounts for thedifferent subspecies of HDL (Figure 1) [3,17,77].

[()TD$FIG]

ee Apo AI acquires phospholipids and cholesterol from cells through ABCA1 to form

vides nascent HDL with additional free cholesterol. The protein LCAT esterifies free

l ester from HDL in exchange for triglycerides. The conversion of free cholesterol into

erical HDL species. Initially, spherical HDLs are small and lipid-poor. These are called

generated, which are termed HDL2a and HDL2b. When classified using two-dimensional

Opinion Trends in Molecular Medicine Vol.16 No.12Figure 1. Growth of HDL from lipid-free Apo AI to mature, spherical HDL. Lipid-fr

nascent HDL. Further interaction with ABCA1, as well as ABCG1 and SR-B1, pro

cholesterol, forming cholesteryl esters. Another protein, CETP, transfers cholestery

cholesteryl esters and the acquisition of triglycerides lead to the formation of sph

HDL3a, HDL3b and HDL3c. As HDL continues to grow, large cholesterol-rich HDLs aregel electrophoresis, nascent HDL occurs as pre-b HDL, whereas HDL3 and HDL2 are found as a or pre-a HDL [25,27]. Apo AI, red and orange; phospholipids, blue; cholesterol

and cholesteryl esters, green; and triglycerides, yellow.

554

correlation between different subpopulations of HDL andCHD [28]. Specifically, Asztalos et al. reported that higherlevels of a-3 and pre a-1 HDL particles are significantlyassociated with CHD and that individuals with CHD had asignificantly lower concentration of a-1 HDL (Figure 1)[28]. Furthermore, esterification of HDL-bound cholester-ol, which ensures that a free cholesterol gradient is main-tained between cells and HDL for RCT, occurs at differentrates for different sizes of HDL [19,29]. There is alsoevidence that large cholesterol rich HDL particles, suchas HDL2 (Figure 1), have higher levels of antioxidantactivity, another of the atheroprotective properties ofHDL [30]. These findings demonstrate the importance ofsize and composition in the ability of HDL to protectagainst atherosclerosis and the notion that the most effec-tive HDL therapeutic would ideally increase the amount ofthose HDL species that are most atheroprotective. To fullyunderstand the structurefunction relationship of HDLspecies and create novel HDL therapeutics, a syntheticstrategy for producing spherical HDLs is needed.

Current technologies for increasing HDLGenerally, two strategies for increasing HDL levels havebeen pursued. First, small-molecule drugs have been de-veloped that act on HDL metabolism and the RCT path-way to increase levels of circulating HDL [18,26]. Second,several forms of HDL replacement therapy have beenstudied [31,32]. Synthetic forms ofHDLhave been createdby combining known biological components of HDL, ApoAI and phospholipids, to create structures very similar tonascent HDL, termed reconstituted HDL (rHDL) [3234].In addition, administration of different forms of Apo AI,including Apo AI mutants (e.g. Apo AIMilano) and peptidesthatmimic the class A amphipathic helices of ApoAI, havebeen explored [31,35]. Administration of Apo AI alone hasalso been studied [36] but is only discussed in terms ofcomplexation with phospholipids in this article.

Small-molecule drugsOne of the most effective small-molecule drugs for increas-ing HDL levels is niacin [36]. Human studies show thatniacin reduces the advancement of atherosclerosis anddecreases cardiovascular events [36]. However, a signifi-cant side effect of niacin is dilation of the blood vessels inthe skin of the face and upper body, accompanied by aburning sensation [37]. This condition is known as flushingand limits tolerance and patient use [36,37].

Another class of small-molecule drugs being investigat-ed is inhibitors of cholesteryl ester transfer protein(CETP). Inhibition of CETP prevents cholesterol transportfrom HDL to LDL, and thus elevates HDL and HDL-clevels [36]. The most well-known CETP inhibitor is torce-trapib [36,38]. Human studies with torcetrapib have dem-onstrated that the drug increases levels of HDL comparedto placebo [3943]. However, torcetrapib was found to havelittle to no ability to reduce atherosclerosis, caused unde-sirable side effects, including increased blood pressure, andmost importantly increased the risk of death in treatedpatients [4345].

OpinionTwo other CETP inhibitors are being studied, JTT-705and anacetrapib. In human trials, both increase levels ofHDL while decreasing levels of LDL compared to placeboand do not seem to have an effect on blood pressure [4649].The ability of these molecules to inhibit atherosclerosis inhumans is still unknown, although studies with JTT-705 inrabbits exhibitedmixed results for limiting the progressionof atherosclerosis [36,50,51]. The long-term effects of JTT-705 and anacetrapib, as well as the potential of CETPinhibitors to decrease atherosclerosis in humans, remain tobe seen.

HDL replacement therapiesVarious methods exist for generating rHDL. These rely onthe amphipathic properties of Apo AI for self-assembly[17,35]. In all of these methods, Apo AI participates in aself-assembly process to form a disk-like structure verysimilar to nascentHDL [52,53]. There is some control of thediameter of the rHDL disks by varying the ratio of thephospholipids and Apo AI [54,55]. rHDL particles range insize from approximately 7 to 13 nm, and monodisperserHDL can be obtained by chromatography or gradientultracentrifugation [17,33,54].

Numerous studies in animal models demonstrate theanti-inflammatory and anti-atherogenic properties ofrHDL [17,32,56]. Two studies in humans with rHDL haveshown potential for rHDL use in treating atherosclerosis[57,58]. In the first study, no significant changes wereobserved in plaque volume compared to placebo for thelower dose of rHDL (40 mg/kg) [57]. However, the plaquecharacterization index, a composite measure of the char-acteristics of an atherosclerotic plaque as assessed byintravascular ultrasound, showed a modest but statisti-cally significant improvement in the rHDL group com-pared to placebo [57]. The highest dose of rHDL (80 mg/kg) was terminated owing to elevations in transaminaselevels, a sign of abnormal liver function and possible livertoxicity [57,59].

The second study examined plaque from the peripheralvasculature after a single injection of 80 mg/kg of rHDL[58]. In this study, administration of rHDL compared toplacebo resulted in a decrease in plaque lipid content, inthe size of macrophage cells in the plaque, and in theexpression of vascular cell adhesion molecule-1, indicatinga decrease in inflammation [58]. Furthermore, enhanced invitro cholesterol efflux to Apo B-depleted serum was ob-served for treated versus untreated patient serum [58].These clinical studies demonstrate the therapeutic poten-tial of increasing HDL levels through administration ofsynthetic HDL. Perceived drawbacks to rHDL therapiesare the requirement of intravenous administration, theexpense of producing rHDL and limitations in the massproduction of rHDL [32,38].

Amutant of Apo AI, called Apo AIMilano, has also shownmany beneficial anti-atherogenic properties, includingthe ability to efflux cholesterol from macrophage cellsand a greater capacity to protect phospholipids fromoxidation compared to wild-type Apo AI [56,60]. ApoAIMilano has been combined with phospholipids to gener-ate rHDL. In a human clinical trial, rHDLmade with ApoAIMilano (ETC-216) that was administered weekly for 5

Trends in Molecular Medicine Vol.16 No.12weeks resulted in a 4.2% decrease in plaque volumecompared to baseline values [61]. As with rHDL prepared

555

using wild-type Apo AI, studies of Apo AIMilano rHDLsuggest that administration of rHDL initiates plaqueregression and could be used to treat CHD [32]. Theuse of Apo AIMilano rHDL as a therapeutic also facessimilar disadvantages to rHDLs synthesized usingwild-type Apo AI [32]. Furthermore, rHDL species arerestricted in terms of ability to manipulate their chemicalcomposition, size and shape [17,54,55] factors thatmight well impact therapeutic efficacy.

As an alternative to Apo AI and rHDL, peptides thatmimic the class A amphipathic helices of Apo AI have beendeveloped [35]. The peptide subject to the greatest investi-gation is an 18-amino-acid peptide modified with differentnumbers of phenylalanine (F) residues on the hydrophobicface [35]. Studies have shown that the peptide 4F, whichcontains four phenylalanine residues, is the most biologi-cally active and reduces atherosclerotic plaque and inflam-mation in mice [31,35]. One study has been carried out inhumans. In this study, the peptide 4F made with D-aminoacids (D-4F) caused a significant improvement in the in-flammatory index of HDL compared to placebo and had afavorable side-effect profile [62]. This study shows that D-4F peptides might greatly enhance the anti-inflammatoryproperties of HDL, but does not provide any informationabout their ability to decrease atherosclerosis and treatCHD. Further studies are needed to determine the effec-tiveness of Apo AI mimetic peptides in treating atheroscle-rosis in humans.

Replacement HDL therapies have been largely restrict-ed to Apo AI alone, nascent rHDL species and Apo AImimetic peptides [31,38]. Less explored are more maturespherical forms of HDL. This is due, in part, to the com-plexity of the biological steps required for fabricatingmature spherical HDL species [52,63]. Biological matura-tion of discoidal to spherical rHDL species requiresenzymes and intermediates [52,54]. Current syntheticroutes provide no direct control over the final composition,and thus present difficulties in reproducibly synthesizingand purifying spherical HDL products [52,63]. With moredata emerging that indicate that HDL size, shape andcomposition largely dictate in vivo function [19,26], it isimperative to develop methods that provide exquisite con-trol over these parameters. Thus, new technologies areneeded to generate synthetic tailorable HDL structuresthat can accurately mimic specific species of HDL to iden-tify optimal HDL mimics for treating atherosclerosis. Inaddition, the ability to modify these structures for imagingpurposes might facilitate early detection of vulnerableplaques and decrease the occurrence of sudden coronaryevents.

Nanotechnology for synthetic HDLNanotechnology provides a new platform for the develop-ment of CHD therapeutics. Inorganic nanoparticles pro-vide a scaffold for creating synthetic structures thatmimicmature spherical forms of HDL [64,65]. This approachprovides a strategy for bypassing the biological and self-assembly methods currently used to produce sphericalHDL. One useful inorganic nanoparticle platform for

Opinionbiological application is colloidal gold nanoparticles (AuNPs).

556Monodisperse colloidal Au NPs can be synthesized withprecise control over size and shape [66,67]. In addition, AuNPs can be surface-functionalized with a variety of biolog-ical molecules, including nucleic acids, proteins, lipids andother small molecules [6770]. Au NPs generally seem tobe non-toxic [71,72]; however, surface functionalities sig-nificantly impact the in vitro and in vivo behavior ofnanomaterials [68]. Complete characterization of Au NPcytotoxicity, biodistribution and efficacy will be necessaryfor eachNP conjugate before an individual structure can beapproved for use in humans.

Replacing the hydrophobic core of natural HDL with anAu NP has the potential to generate a robust and stablestructure that accurately replicates the size, shape andsurface chemistry of mature spherical forms of HDL. Syn-thetic control over each of these parameters might facili-tate amore detailed investigation of the structurefunctionrelationship of HDLs and provide access to particularproperties of HDL. Furthermore, the Au NP core can bereplaced by other inorganic NPs, such as iron oxide (FeO)NPs or quantum dots (QDs), which would lend additionalcapabilities (i.e. imaging) [64,69]. Overall, NPs constitute amodular system with great potential to create an array ofHDL mimetics with multimodal capabilities for detectingand treating CHD.

Synthetic HDL based on inorganic NPsTherapeuticThe first biomimetic HDL using an Au NP core (HDL AuNP) for therapeutic purposes was synthesized and charac-terized by the Thaxton and Mirkin groups [65]. A 5-nm-diameter Au NP was first surface-functionalized with ApoAI and then two different phospholipids were added tocreate the lipid layer (Figure 2). This HDL AuNP has bothsurface-chemical and physical properties similar to thoseof natural HDL. The hydrodynamic diameter of the HDLAu NPs measured by dynamic light scattering (DLS) is183 nm. Furthermore, each HDL Au NP has two to fourApo AI proteins and approximately 7595 aminated phos-pholipids [65]. These values correlate well with those fornatural mature spherical HDL [3,27,73].

To quantify a functional baseline for cholesterolbinding to a synthetic HDL Au NP, a fluorescently labeledcholesterol molecule, 25-[N-[(7-nitro-2-1,3-benzoxadiazol-4-yl)methyl]amino]-27-norcholesterol (NBD cholesterol)that is minimally fluorescent in aqueous environmentsand becomes highly fluorescent in lipid membranes [74],was used to evaluate the ability of the HDL Au NP to bindcholesterol in solution. HDL Au NPs were titrated withincreasing concentrations of NBD cholesterol to develop abinding isotherm. The dissociation constant Kd calculatedfrom the binding isothermwas 3.80.8 nM for the HDLAuNPs [65]. Dissociation constants, even for natural HDL,have not been reported in the literature. Thus, this valuerepresents an initial baseline and serves as a point ofcomparison for future biomimetic HDL structures. Thisstudy demonstrates that biomimetic HDL Au NPs, withthe size, shape and surface chemistry of natural HDL andcholesterol binding capabilities, have the potential to be

Trends in Molecular Medicine Vol.16 No.12used to raise levels of circulating HDL as a novel thera-peutic to treat atherosclerosis.

. Citr

olam

ng wImaging agentAnother example of HDL prepared with inorganic NPshas been reported for imaging purposes. In fact, imagingwas the first application evaluated for inorganic nano-particle-based synthetic forms of HDL. Cormode et al.demonstrated the use of three different inorganic NPs toconstruct multi-modal imaging agents that mimic HDL(nanocrystal HDL) to image macrophages in atheroscle-rotic plaque [64]. Gold nanoparticles (Au HDL), ironoxide nanoparticles (FeO HDL) and quantum dots (QDHDL) were combined with Apo AI and various phospho-lipids, including fluorescent and paramagnetic phospho-lipids, to create HDL mimetic structures that can be

[()TD$FIG]

Figure 2. Synthesis of biomimetic HDL using a Au NP core for use as a therapeutic

and then with two phospholipids, 1,2-dipalmitoyl-sn-glycero-3-phosphoethan

phosphocholine (green). This HDL mimetic demonstrated tight cholesterol bindi

Society with permission from the American Chemical Society.

Opinionimaged by CT, MRI and fluorescence (Figure 3) [64].Control NPs were also prepared using a pegylated phos-pholipid (Au PEG, FeO PEG and QD PEG). Au HDL andQD HDL both had approximately three Apo AI proteinsper nanoparticle, whereas FeO HDL had approximatelyfour Apo AI proteins per nanoparticle. The size, asmeasured by TEM, was 9.7, 11.9, and 12.0 nm for AuHDL, FeO HDL and QD HDL, respectively [64]. DLS andgel electrophoresis data corroborated these values [64].These analyses demonstrate that nanocrystal HDLsclosely resemble spherical HDL in terms of size andcomposition.

Nanocrystal HDLs were studied with macrophage cellsand in the ApoE/ atherosclerotic mouse model. Imagingof macrophage cells exposed to nanocrystal HDLs demon-strated much greater uptake of nanocrystal HDL com-pared to PEG control particles [64]. In the ApoE/

mouse model, a significant change in MRI signal fromthe aortic wall was observed for nanocrystal HDLs com-pared to PEG controls [64]. Fluorescence and CT imagingof appropriate nanocrystal HDLs also verified absorptionof nanocrystal HDLs into the aorta wall compared to PEGcontrol particles [64]. This work establishes the ability touse different inorganic NPs to create structures that accu-rately mimic natural HDL and can be used to specificallyimage macrophage cells in the wall of a mouse aorta. Byaccurately mimicking HDL, nanocrystal HDLs can poten-tially be used for non-invasive diagnosis of atherosclerosisand CHD.

The studies reviewed above showcase the ability to useinorganic NPs to create synthetic structures that mimicnatural HDL in size, shape, composition, and function.They further highlight the importance and necessity ofaccurately mimicking natural HDL to obtain desired spec-ificity. The first example binds cholesterol in solution andhas the potential to be used as a therapeutic to combatatherosclerosis. The second example delivers imaging mo-dalities to macrophage cells for the purpose of imaging

ate-stabilized Au NPs of 5 nm in diameter were surface-functionalized with Apo AI

ine-N-[3-(2-pyridyldithio)propionate] (yellow) and 1-2-dipalmitoyl-sn-glycero-3-

ith a Kd of 3.80.8 nM [65]. Reprinted from Journal of the American Chemical

Trends in Molecular Medicine Vol.16 No.12atherosclerotic plaque. Without correctly mimicking natu-ral HDL, it would be difficult to obtain the desired speci-ficity and perform either of these functions. Theseinvestigations indicate that nanotechnology has the poten-tial to provide synthetic control not available with existingstrategies used to make replacement HDL and to generatesynthetic structures that accurately mimic natural spher-ical HDL.

Translational considerationsQuestions remain regarding the in vivo safety and efficacyof inorganic NPs for treating atherosclerosis, CHD andother diseases. Safety and efficacy concerns are not uniqueto novel therapeutic or imaging agents under considerationfor human use. Unique, however, is the relative lack ofexperience that pharmaceutical companies and regulatoryagencies have with regard to evaluating and regulatingNPs. Furthermore, the ensemble of NP platform, surfacefunctionality and internal functionality cooperates to dic-tate in vivo pharmacokinetics and efficacy such that eachagent will require individual detailed attention by inves-tigators, interested industrial partners and regulatoryagencies. Ultimately, successful application of nanotech-nology to atherosclerosis and CHD will require multidisci-plinary collaboration (Box 2).

557

[()TD$FIG]

OpinionConcluding remarksCHD is the leading cause of death in the world and newstrategies for its treatment are needed. Novel nanotech-nology constructs present promising therapeutic possibili-ties and offer advantages that are lacking in currenttherapeutic strategies. Although the most widely exploredmethods for augmenting circulating levels of HDL havedemonstrated some effectiveness clinically, administrationrequirements, synthetic challenges and undesirable sideeffects hinder their therapeutic potential. The use of nano-technology to accurately mimic natural HDL, therebyavoiding off-target effects and potentially imitating the

Figure 3. Synthetic HDLs with a nanocrystal core for imaging atherosclerotic plaque. (a

Apo AI on their surface to mimic natural HDL. The Au HDL has an Au NP core that i

rhodamine (Rhod) for MRI and fluorescence imaging, respectively. The FeO HDL con

phospholipids for fluorescence imaging. The QD-HDL has a quantum dot core for flu

nanocrystal HDLs. (i) The phospholipids myristoyl hydroxy phosphatidylcholine (MHP

acid (Gd-DTPA-DMPE) and dimyristoyl phosphoethanolamine-N-(lissamine rhodamine

Apo AI was added to the solution for incorporation into the phospholipid layer. (iii) U

demonstrated specificity for macrophage cells in vitro and in vivo and showed poten

Reprinted from Nano Letters with permission from the American Chemical Society.

558Trends in Molecular Medicine Vol.16 No.12atheroprotective abilities of natural HDL, holds greatpromise for creating a therapeutic to combat atherosclero-sis. The synthetic processes used to create synthetic HDLswith an inorganic nanoparticle core provide control thatmight allow investigation of the structurefunction rela-tionship of HDL and identify the synthetic HDL structuremost effective for reversing atherosclerosis and treatingCHD. In addition, the inclusion of imaging modalities orother molecules lends the ability to generate multimodalstructures with additional functionalities. The novel capa-bilities presented by nanotechnology hold great promise forimproving the treatment of atherosclerosis and CHD.

) Representations of the three nanocrystal HDLs. All three nanocrystal HDLs have

s suitable for CT imaging and phospholipids modified with gadolinium (Gd) and

tains an FeO nanoparticle core that is useful for MRI, as well as Rhod-modified

orescence imaging and Gd-modified phospholipids for MRI. (b) Synthesis of the

C), gadolinium dimyristoyl phosphoethanolamine diethylenetriamine penta-acetic

B sulfonyl) ammonium salt (Rhod-DMPE) were first added to the nanoparticles. (ii)

ltracentrifugation was used to purify the nanocrystal HDL. These HDL mimetics

tial for imaging these cells through either MRI, CT or fluorescence imaging [64].

Opinion Trends in Molecular Medicine Vol.16 No.12Conflict of interestCAM and CST are co-founders of AuraSense, LLC, which has licensedHDL AuNPs from Northwestern University.

References1 Lloyd-Jones, D. et al. (2010) Heart disease and stroke statistics 2010update: a report from the AmericanHeart Association.Circulation 121,E46E215

2 Mackay, J. and Mensah, G.A. (2004) Atlas of Heart Disease and Stroke,World Health Organization and Centers for Disease Control andPrevention

3 Libby, P. et al., eds (2007) Braunwalds Heart Disease: A Textbook ofCardiovascular Medicine (8th edn), WB Saunders

4 Lusis, A.J. (2000) Atherosclerosis. Nature 407, 2332415 Falk, E. (2006) Pathogenesis of atherosclerosis. J. Am. Coll. Cardiol.47, C7C12

6 Khera, A.V. and Rader, D.J. (2010) Future therapeutic directions inreverse cholesterol transport. Curr. Atheroscl. Rep. 12, 7381

7 Ikonen, E. (2008) Cellular cholesterol trafficking andcompartmentalization. Nat. Rev. Mol. Cell Biol. 9, 125138

8 Libby, P. and Theroux, P. (2005) Pathophysiology of coronary arterydisease. Circulation 111, 34813488

9 Ucar, M. et al. (2000) HMG-CoA reductase inhibitors and myotoxicity.Drug Saf. 22, 441457

10 Gotto, J.A.M. and Grundy, S.M. (1999) Lowering LDL cholesterol:questions from recent meta-analyses and subset analyses of clinicaltrial data Issues from the Interdisciplinary Council on Reducing theRisk for Coronary Heart Disease, Ninth Council Meeting. Circulation99, e1e7

11 Singh, I.M. (2007) High-density lipoprotein as a therapeutic target: asystematic review. J. Am. Med. Assoc. 298, 786798

12 Pedersen, T.R. et al. (1994) Randomized trial of cholesterol-lowering in4444 patients with coronary heart disease the ScandinavianSimvastatin Survival Study (4 s). Lancet 344, 13831389

13 Bruckert, E. et al. (2005) Mild to moderate muscular symptoms withhigh-dosage statin therapy in hyperlipidemic patients the PRIMOstudy. Cardiovasc. Drugs Ther. 19, 403414

Box 2. Outstanding Questions

Will inorganic nanoparticle-based HDL absorb cholesterol in vivoand transport it to the liver according to the RCT pathway?

If inorganic nanoparticle-based HDLs do transport cholesterol invivo, will they be effective at removing cholesterol from foam cellsand decreasing or reversing the formation of atheroscleroticplaque?

Do inorganic nanoparticle-based HDLs require intravenous injec-tion, or can they be dosed orally?

What is the stability and lifetime of inorganic nanoparticle-basedHDL in vivo? Does it accumulate in the body in a particular tissue?How is it degraded and/or excreted?

Are there any side effects or toxicity related to inorganicnanoparticle-based HDL?14 Bays, H. (2006) Statin safety: an overview and assessment of the data 2005. Am. J. Cardiol. 97, 6C26C

15 Jacobson, T.A. (2008) Toward pain-free statin prescribing: clinicalalgorithm for diagnosis and management of myalgia.Mayo Clin. Proc.83, 687700

16 Unal, B. et al. (2004) Explaining the decline in coronary heart diseasemortality in England and Wales between 1981 and 2000. Circulation109, 11011107

17 Calabresi, L. et al. (2006) Synthetic high density lipoproteins for thetreatment of myocardial ischemia/reperfusion injury. Pharmacol.Ther. 111, 836854

18 Joy, T. and Hegele, R.A. (2008) Is raising HDL a futile strategy foratheroprotection? Nat. Rev. Drug Discov. 7, 143155

19 Movva, R. and Rader, D.J. (2008) Laboratory assessment of HDLheterogeneity and function. Clin. Chem. 54, 788800

20 Duffy, D. and Rader, D.J. (2006) Emerging therapies targeting high-density lipoprotein metabolism and reverse cholesterol transport.Circulation 113, 1140115021 Naghavi, M. et al. (2003) From vulnerable plaque to vulnerable patient a call for new definitions and risk assessment strategies: part I.Circulation 108, 16641672

22 Zipes, D.P. and Wellens, H.J.J. (1998) Sudden cardiac death.Circulation 98, 23342351

23 Sanz, J. and Fayad, Z.A. (2008) Imaging of atheroscleroticcardiovascular disease. Nature 451, 953957

24 Fuster, V. et al. (2010) Early identification of atherosclerotic disease bynoninvasive imaging. Nat. Rev. Cardiol. 7, 327333

25 Jonas, A. and Phillips, M.C. (2008) Lipoprotein structure, InBiochemistry of Lipids, Lipoproteins and Membranes (5th edn)(Vance, D.E. and Vance, J.E., eds), pp. 485506, Elsevier

26 Degoma, E.M. et al. (2008) Beyond high-density lipoprotein cholesterollevels evaluating high-density lipoprotein function as influenced bynovel therapeutic approaches. J. Am. Coll. Cardiol. 51, 21992211

27 Warnick, G.R. et al. (2006) Polyacrylamide gradient gel electrophoresisof lipoprotein subclasses. Clin. Lab. Med. 26, 803846

28 Asztalos, B.F. et al. (2004) High-density lipoprotein subpopulationprofile and coronary heart disease prevalence in male participantsof the FraminghamOffspring Study.Arterioscl. Thromb. Vasc. Biol. 24,21812187

29 Dobiasova,M. et al. (2000) Effect of labeling of plasma lipoproteins with[3H]cholesterol on values of esterification rate of cholesterol inapolipoprotein B-depleted plasma. J. Lipid Res. 41, 13561357

30 Brites, F.D. et al. (2004) Paraoxonase 1 and platelet-activating factoracetylhydrolase activities in patients with low HDL-cholesterol levelswith or without primary hypertriglyceridemia. Arch. Med. Res. 35,235240

31 Navab, M. et al. (2010) Structure and function of HDL mimetics.Arterioscl. Thromb. Vasc. Biol. 30, 164168

32 Murphy, A.J. et al. (2009) Reconstituted HDL: a therapy foratherosclerosis and beyond. Clin. Lipidol. 4, 731739

33 Jonas, A. et al. (1989) Defined apolipoprotein A-I conformations inreconstituted high-density lipoprotein disks. J. Biol. Chem. 264, 48184824

34 Matz, C.E. and Jonas, A. (1982) Micellar complexes of humanapolipoprotein A-I with phosphatidylcholines and cholesterolprepared from cholate-lipid dispersions. J. Biol. Chem. 257, 45354540

35 Sherman, C.B. et al. (2010) Apolipoprotein A-I mimetic peptides: apotential new therapy for the prevention of atherosclerosis. Cardiol.Rev. 18, 141147

36 Duffy, D. and Rader, D.J. (2009) Update on strategies to increase HDLquantity and function. Nat. Rev. Cardiol. 6, 455463

37 Benyo, Z. et al. (2005) GPR109A (PUMA-G/HM74A) mediates nicotinicacid-induced flushing. J. Clin. Investig. 115, 36343640

38 Conca, P. and Franceschini, G. (2008) Synthetic HDL as a newtreatment for atherosclerosis regression: has the time come? Nutr.Metab. Cardiovasc. Dis. 18, 329335

39 Brousseau, M.E. et al. (2004) Effects of an inhibitor of cholesteryl estertransfer protein on HDL cholesterol. N. Engl. J. Med. 350, 15051515

40 Brousseau, M.E. et al. (2005) Effects of cholesteryl ester transferprotein inhibition on high-density lipoprotein subspecies,apolipoprotein A-I metabolism, and fecal sterol excretion. Arterioscl.Thromb. Vasc. Biol. 25, 10571064

41 McKenney, J.M. et al. (2006) Efficacy and safety of torcetrapib, a novelcholesteryl ester transfer protein inhibitor, in individuals with below-average high-density lipoprotein cholesterol levels on a background ofatorvastatin. J. Am. Coll. Cardiol. 48, 17821790

42 Davidson, M.H. et al. (2006) Efficacy and safety of torcetrapib, a novelcholesteryl ester transfer protein inhibitor, in individuals with below-average high-density lipoprotein cholesterol levels. J. Am. Coll.Cardiol. 48, 17741781

43 Barter, P.J. et al. (2007) Effects of torcetrapib in patients at high riskfor coronary events. N. Engl. J. Med. 357, 21092122

44 Bots, M.L. et al. (2007) Torcetrapib and carotid intima-media thicknessin mixed dyslipidaemia (RADIANCE 2 study): a randomised, double-blind trial. Lancet 370, 153160

45 Kastelein, J.J.P. et al. (2007) Effect of torcetrapib on carotidatherosclerosis in familial hypercholesterolemia. N. Engl. J. Med.356, 16201630

46 de Grooth, G.J. et al. (2002) Efficacy and safety of a novel cholesterylester transfer protein inhibitor, JTT-705, in humans a randomized

phase II doseresponse study. Circulation 105, 21592165

559

47 Krishna, R. et al. (2007) Effect of the cholesteryl ester transfer proteininhibitor, anacetrapib, on lipoproteins in patients with dyslipidaemiaand on 24-h ambulatory blood pressure in healthy individuals: twodouble-blind, randomised placebo-controlled phase I studies. Lancet370, 19071914

48 Bloomfield, D. et al. (2009) Efficacy and safety of the cholesteryl estertransfer protein inhibitor anacetrapib as monotherapy andcoadministered with atorvastatin in dyslipidemic patients. Am.Heart J. 157, 352360

49 Hermann, F. et al. (2009) Cholesterylestertransfer protein inhibitionand endothelial function in type II hyperlipidemia. Thromb. Res. 123,460465

50 Okamoto, H. et al. (2000) A cholesteryl ester transfer protein inhibitorattenuates atherosclerosis in rabbits. Nature 406, 203207

51 Huang, Z.P. et al. (2002) Cholesteryl ester transfer protein inhibitor(JTT-705) and the development of atherosclerosis in rabbits withsevere hypercholesterolaemia. Clin. Sci. 103, 587594

52 Jonas, A. (1986) Reconstitution of high-density lipoproteins. MethodsEnzymol. 128, 553582

53 Shih, A.Y. et al. (2007) Disassembly of nanodiscs with cholate. NanoLetters 7, 16921696

54 Calabresi, L. et al. (1993) Apolipoprotein-A-I conformation in discoidalparticles evidence for alternate structures. Biochemistry 32, 64776484

55 Silva, R.A.G.D. et al. (2005) A mass spectrometric determination of the

62 Bloedon, L.T. et al. (2008) Safety, pharmacokinetics, andpharmacodynamics of oral apoA-I mimetic peptide D-4F in high-riskcardiovascular patients. J. Lipid Res. 49, 13441352

63 Silva, R.A.G.D. et al. (2008) Structure of apolipoprotein A-I in sphericalhigh density lipoproteins of different sizes. Proc. Natl. Acad. Sci. U. S. A.105, 1217612181

64 Cormode, D.P. et al. (2008) Nanocrystal core high-densitylipoproteins: a multimodality contrast agent platform. Nano Letters8, 37153723

65 Thaxton, C.S. et al. (2009) Templated spherical high densitylipoprotein nanoparticles. J. Am. Chem. Soc. 131, 13841385

66 Millstone, J.E. et al. (2008) Iodide ions control seed-mediated growth ofanisotropic gold nanoparticles. Nano Letters 8, 25262529

67 Wang, Z.X. and Ma, L.N. (2009) Gold nanoparticle probes. Coord.Chem. Rev. 253, 16071618

68 Boisselier, E. and Astruc, D. (2009) Gold nanoparticles in nanomedicine:preparations, imaging, diagnostics, therapies and toxicity. Chem. Soc.Rev. 38, 17591782

69 Agasti, S.S. et al. (2010) Nanoparticles for detection and diagnosis.Adv.Drug Deliv. Rev. 62, 316328

70 Mirkin, C.A. (2010) The polyvalent gold nanoparticle conjugate materials synthesis, biodiagnostics, and intracellular generegulation. MRS Bull. 35, 532539

71 Esther, R.J. et al. (2005) Gold nanoparticles do not affect theglobal transcriptional program of human umbilical vein endothelialcells: aDNA-microarray analysis. J.Biomed.Nanotechnol. 1, 328335

72 Fadeel, B. and Garcia-Bennett, A.E. (2010) Better safe than sorry:

Opinion Trends in Molecular Medicine Vol.16 No.12conformation of dimeric apolipoprotein A-I in discoidal high densitylipoproteins. Biochemistry 44, 86008607

56 Chiesa, G. et al. (2008) Acute effects of high-density lipoproteins:biochemical basis and clinical findings. Curr. Opin. Cardiol. 23,379385

57 Tardif, J.C. et al. (2007) Effects of reconstituted high-densitylipoprotein infusions on coronary atherosclerosis a randomizedcontrolled trial. J. Am. Med. Assoc. 297, 16751682

58 Shaw, J.A. et al. (2008) Infusion of reconstituted high-densitylipoprotein leads to acute changes in human atherosclerotic plaque.Circ. Res. 103, 10841091

59 Bai, J. and Cederbaum, A.I. (2006) Adenovirus-mediated expression ofCYP2E1 produces liver toxicity in mice. Toxicol. Sci. 91, 365371

60 Favari, E. et al. (2007) A unique protease-sensitive high densitylipoprotein particle containing the apolipoprotein A-I-Milano dimereffectively promotes ATP-binding cassette A1-mediated cell cholesterolefflux. J. Biol. Chem. 282, 51255132

61 Nissen, S.E. et al. (2003) Effect of recombinant ApoA-I Milano oncoronary atherosclerosis in patients with acute coronary syndromes a randomized controlled trial. J. Am. Med. Assoc. 290, 22922300560understanding the toxicological properties of inorganic nanoparticlesmanufactured for biomedical applications. Adv. Drug Deliv. Rev. 62,362374

73 Segrest, J.P. et al. (2000) Structure and function of apolipoprotein A-Iand high-density lipoprotein. Curr. Opin. Lipidol. 11, 105115

74 Chattopadhyay, A. (1990) Chemistry and biology of N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)-labeled lipids fluorescent-probes of biological andmodel membranes. Chem. Phys. Lipids 53, 115

75 Rader, D.J. (2006) Molecular regulation of HDL metabolism andfunction: implications for novel therapies. J. Clin. Investig. 116,30903100

76 Segrest, J.P. et al. (1999) A detailed molecular belt model forapolipoprotein A-I in discoidal high density lipoprotein. J. Biol.Chem. 274, 3175531758

77 Shih, A.Y. et al. (2009) Maturation of high-density lipoproteins. J. R.Soc. Interface 6, 863871

78 Barkowski, R.S. and Frishman, W.H. (2008) HDL metabolism andCETP inhibition. Cardiol. Rev. 16, 154162

Nanotechnology for synthetic high-density lipoproteinsCoronary heart disease and HDLRelationship between natural HDL structure and functionCurrent technologies for increasing HDLSmall-molecule drugsHDL replacement therapies

Nanotechnology for synthetic HDLSynthetic HDL based on inorganic NPsTherapeuticImaging agentTranslational considerations

Concluding remarksReferences