Introduction

The cardiovascular system is responsible for the distribution of oxygen, nutrients, cells, and signaling molecules in the human body, and the removal of waste products to maintain cellular homeostasis and fight disease. The cardiovascular system consists of the blood distribution network (the heart and the blood vessels), but in a broader view the lymphatic system can also be included. With every beat, the heart pumps blood through the pulmonary circulation to pick up oxygen and remove gaseous waste products, and into the systemic circulation to provide oxygen and pick up and deliver nutrients to the body. The heart is the strongest muscle in the body and an apparently tireless pump, typically beating 60–70 times per minute and accumulating more than 2 billion heartbeats over a lifespan of 70 years. The cardiovascular system is of primary importance to maintaining a healthy condition. Unfortunately however, the heart and blood vessels are not immune to disease. Both are susceptible to breakdown and attacks from various diseases, which are collectively referred to as cardiovascular disease (CVD).

There exist a large number of diseases that affect the cardiovascular system at various levels, leading to disability and death. Those which occur most frequently include rheumatic heart disease, congenital heart disease, aneurisms, peripheral artery disease, thrombosis, pulmonary embolism, stroke, and coronary heart disease (CHD) [1]. Taken together, an estimated 17.3 million people worldwide died from CVD in 2008—a number that is expected to increase to more than 23 million in 2030. More people die annually of CVD than from any other cause. Of all CVDs, stroke and coronary artery disease account for more than 70 % of all deaths. Risk factors for CVD are well known and involve smoking, obesity, diabetes, hyperlipidemia, physical inactivity, and hypertension. Nevertheless, it is unrealistic to expect that the CVD epidemic can be halted by prevention, raising public awareness and changes in lifestyle alone. Hence, there is great need for new treatment strategies and tools to determine the condition of the individual patient at risk for or suffering from CVD, in order to predict and improve clinical outcome and to accurately monitor and screen response to new therapies.

Molecular imaging holds the promise to revolutionize the understanding of the etiology of CVD, thereby profoundly impacting future clinical CVD care [2–5]. The use of targeted molecular imaging probes enables the identification of key pathophysiological events, including necrosis, inflammation, thrombosis, necrosis, apoptosis, remodeling, and revascularization. Novel therapies for CVD, such as those based on cellular therapy and using innovative therapeutic compounds may greatly benefit from the insights obtained by molecular imaging. In the future, the use of nanoparticles may enable combined (targeted) imaging and drug delivery tailored to the patient’s specific CVD molecular fingerprint.

In this review, recent promising developments in the use of nanoparticles for molecular imaging of CVD are highlighted and discussed. For reason of focus, this review mainly covers the molecular imaging of CHD.

Coronary Heart Disease

Coronary heart disease can be defined as diseases of the blood vessels supplying the heart muscle and resulting clinical complications. Most coronary circulation problems are related to the formation of atherosclerotic plaques in the major branches of the coronary arteries [6–8]. Although many aspects of the etiology of atherosclerosis remain unknown, the formation of an atherosclerotic plaque is generally accepted to be an inflammatory response to buildup of low-density lipoprotein (LDL) in the vessel wall. This early plaque is referred to as a fatty-streak. LDL uptake in the sub-intima is followed by LDL oxidation by free radicals, originating from myeloperoxidase (MPO) produced by macrophages and neutrophils. The oxidized-LDL in turn triggers an aggravated inflammatory reaction and increased endothelial dysfunction, leading to intensified recruitment and activation of cells of the immune system. Macrophages absorb the oxidized-LDL to become foam cells, which eventually rupture and thereby gradually accumulate lipids in the plaque core. Recruited macrophages also release enzymes, such as matrix metalloproteinase (MMP), that degrade the extracellular matrix and lead to necrosis and apoptosis of endothelial cells, smooth muscle cells and macrophages. Dying cells deposit calcium leading to extracellular calcification of the plaque. New fragile blood vessels may be formed in the plaque, which are leaky and lead to intraplaque hemorrhage. Eventually, this slow-progressing cascade of self-sustaining events may lead to artery occlusion or to vulnerable plaque rupture, causing an acute thrombotic event. In other cases, the process may linger over a lifetime and the plaque may stay clinically silent. Clinical risk assessment by prediction of plaque vulnerability using noninvasive (molecular) imaging technology remains a great, unsolved challenge.

Clinical complications resulting from atherosclerosis in the coronary artery tree include angina pectoris, as a consequence of a partly obstructed coronary artery, and myocardial infarction due to full occlusion of a coronary artery after a thrombotic event. The molecular processes following myocardial infarction have many parallels with atherosclerotic disease [9, 10]. Cell death proceeds principally via apoptosis in the early hours after infarction and through necrosis thereafter. Reperfusion of the affected coronary artery by aggressive thrombolytic therapy to dissolve the occlusive cloth may limit the amount of necrosis but causes additional apoptosis and inflammation. Leukocytes, carried to the infarct by the blood, release a number of inflammatory factors as well as free radicals in response to tissue damage. Later, granulation tissue is produced by fibroblasts and contractile myofibroblasts, providing new tensile strength to the infarct area.

Imaging Inflammation in Cardiovascular Disease

Inflammation plays a key role in CHD and is considered an important sign of atherosclerotic plaque vulnerability. In recent years, a number of targeted agents have been employed to visualize plaque inflammation using various imaging modalities. The nuclear techniques (e.g. PET) have superior sensitivity for the detection of low concentrations of tracer material. Several years ago, it was found that FDG uptake in plaques correlates with macrophage content and is considerably higher in inflamed plaques than in control vessels [11–13]. FDG signal in plaques responds to anti-inflammatory treatment [14]. PET-FDG may therefore be an attractive imaging modality to study and through which to evaluate novel therapeutics aimed at silencing plaque inflammation and reducing rupture risk. On the other hand, FDG may not be specific to plaque inflammation alone since other factors, such as hypoxia and neovascularization, may increase uptake as well. Also, PET-FDG of coronary vessels is problematic due to the close proximity of the myocardium, which avidly takes up FDG. Other PET tracers for imaging inflammation are therefore under development [15].

The limited spatial resolution of PET hampers accurate localization of signals in small plaques in moving vessels. For this purpose, magnetic resonance imaging (MRI) and computed tomography (CT) have distinct advantages. MRI combines high spatial and temporal resolution with a variety of readouts of plaque and vessel morphology [16]. CT angiography is considered the most clinically robust and accurate method for grading coronary artery stenosis and determining plaque calcification [17]. On the downside, MRI and CT have limited sensitivity for the detection of molecular tracers and therefore require a suitable amplification strategy.

One of the first type of nanoparticles applied to study plaque inflammation by MRI were the iron oxide nanoparticles, including superparamagnetic iron oxide (SPIO) and ultrasmall superparamagnetic iron oxide particles (USPIO). Following the finding that carbohydrate-coated iron oxide nanoparticles are rapidly phagocytosed by macrophages and accumulate in spleen, liver, bone marrow and lymph nodes, thus enabling MR imaging of macrophage activity in these organs, the particles were also found to specifically accumulate in plaque resident macrophages. In a rabbit model of atherosclerosis, Ruehm et al. [18] demonstrated specific signal loss in T2*-weighted images of the aorta originating from USPIO accumulation in plaque macrophages. In patients originally scheduled to receive MRI for staging lymph node metastases, accumulation of SPIOs was observed as pronounced signal loss in T2*-weighted images in aortic and arterial wall segments [19]. Kooi et al. [20] demonstrated significant signal loss in T2*-weighted images of human carotid artery plaques 24 h after injection of USPIOs. After surgical resection of the lesion, histological and electron microscopy confirmed the presence of the USPIOs in the plaque primarily located in macrophages. In patients, aggressive lipid-lowering therapy over a 3-month period resulted in significant reduction of USPIO-detected inflammation, demonstrating that inflammation imaging by MRI could develop into a useful clinical tool to evaluate anti-inflammatory treatment [21]. A possible complication for the use of these types of particles in imaging plaque inflammation is that accumulation of iron oxides in perivascular lymph nodes may lead to false positive findings depending on the choice of imaging sequence and image resolution [22].

Iron oxide nanoparticles have also been extensively investigated to assess inflammation in the heart. Kanno et al. [23] used USPIOs to study acute allograft rejection in rat after heart and lung transplantation. In rats that received a syngeneic transplant, injection of USPIOs 6 days after surgery did not result in signal changes on gradient-echo MRI at day 7. In rats that received an allotransplant leading to acute rejection, significant signal losses were observed, which could be attributed to USPIO-laden infiltrating macrophages. The response could be modulated by anti-inflammatory treatment. Sosnovik et al. [24] used iron oxides to visualize macrophages in the infarcted mouse myocardium. The uptake of the iron oxides by macrophages infiltrating the infarcted myocardium was confirmed by fluorescence microscopy and immunohistochemistry. In mice, injection of micron-sized particles of iron oxide (MPIO) 7 days before infarct surgery resulted in a gradual signal decrease over time in T2*-weighted images of the infarcts [25]. Signal loss could be attributed to the influx of pre-labeled macrophages that were recruited to the infarct.

The iron oxides are under investigation for tracking inflammatory cells in humans after recent myocardial infarction. Figure 1 highlights a recent study, in which patients with acute myocardial infarction were imaged at baseline and at 24 and 48 h after injection of USPIOs (ferumoxytol; AMAG Pharmaceuticals, Lexington, MA) [26••]. Injection of USPIO in healthy subjects did not result in R2* (=1/T2*) enhancement in the myocardium, and without injection of contrast agent the R2* values in patients with recent myocardial infarction were stable at these time points (Fig. 1a–c). Application of the USPIOs gave rise to a strong increase in R2* in the infarcts, peaking at 24 h after injection (Fig. 1d, e). Although no histological validations were performed in this study, these observations together with previous knowledge obtained from animal studies strongly suggest that iron oxide-laden macrophages infiltrated the infarcts and that the strong R2* increase therefore indicated the inflammatory response after infarction. Interestingly, an R2* increase was also seen in remote myocardium, although to a lesser extent, indicating post-infarct inflammatory cell infiltration in remote myocardium as well. This technique may find clinical application in the assessment inflammatory response after myocardial infarction and in the evaluation of treatment aimed at controlling the inflammatory response. It could also be of interest for other inflammatory diseases of the heart.

Fig. 1
figure 1

Use of iron oxide nanoparticles to study the inflammatory response after myocardial infarction in humans. R2* maps of the hearts of patients with recent myocardial infarction at a, d baseline and b, e 24 h after a, b no infusion or d, e infusion of iron oxides. c R2* values in the heart without injection remained stable, whereas f infusion of iron oxides resulted in highly elevated R2* values in the infarct and to a lesser extent in the remote myocardium. R2* mapping after infusion of iron oxide nanoparticles therefore shows great promise for clinical MR imaging of inflammation in the heart. Reproduced from [26••], with permission

Full size image

A potential drawback of the use of iron oxides is that T2*-weighted MRI and its associated R2* mapping is artifact prone. Particularly in the assessment of small atherosclerotic plaques and to a lesser degree in the heart, partial volume effects, susceptibility mismatch, blood flow and movement artifacts may be mistaken for nanoparticle-induced signal loss. Several alternative approaches therefore were developed. Nanoparticles suitable as a T1 lowering agent allow for the use of fast T1-weighted MRI methods and inversion recovery imaging techniques resulting in a positive signal change, which can be attributed less ambiguously to areas of nanoparticles accumulation.

In one such approach, Naresh et al. [27, 28] employed Gd-labeled liposomes for targeting macrophages in a mouse model of myocardial infarction. Figure 2 shows three inversion recovery T1-weighted MR images (Fig. 2a–c) of a mouse heart at day 2 and 4 after induction of myocardial infarction. At day 2 (Fig. 2a), traditional late gadolinium enhancement (LGE) using Gd-DTPA contrast agent confirmed the presence of a large infarct. After injection of Gd-liposomes at day 2 (Fig. 2b), T1 contrast in the myocardium was not immediately observed. At day 4 (Fig. 2c), however, a bright signal was apparent particularly in the infarct core, which was attributed to the presence of macrophages labeled with Gd-liposomes. In a slightly different study setup, the Gd-liposomes were injected 2 days before the infarct surgery to pre-label monocytes. Serial measurements of the difference in the longitudinal relaxation rate R1 (=1/T1) of the myocardium between non-injected and injected mice (Fig. 2d) displayed a significant increase that peaked at day 4 after infarct induction, revealing the time course of macrophage infiltration into the infarct zone. A small increase in R1 was also observed in the remote zone, which indicates inflammatory cell infiltration in this region in agreement with observations using iron oxides described above.

Fig. 2
figure 2

Application of Gd-liposomes for imaging of inflammation after myocardial infarction in mice. Inversion recovery MR image of an infarcted mouse heart a at day 2 using Gd-DTPA late gadolinium enhancement (LGE) to depict location and size of the infarction, b at day 2 after injection of Gd-labeled liposomes, and c at day 4, 2 days after injection of the liposomes. Enhancement of the infarct core resulted from liposome labeled macrophages that infiltrated in the infarct. d R1 measurements in infarct zone and remote zone. Gd-liposome accumulation in the infarcts reached maximal levels on day 4 after myocardial infarction. Reproduced from [27, 28], with permission

Full size image

Although T1-weighted imaging is less prone to artifacts than T2*-weighted imaging, the use of Gd-labeled nanoparticles is not without complications. First of all, affinity of the Gd-labeled liposomes for inflammatory cells proves highly dependent upon the (surface) composition of the particles. Paulis et al. [29] and Geelen et al. [30•] have recently shown that pegylated Gd-liposomes and smaller Gd-micelles readily accumulate in the myocardium of mice with an infarction induced by either permanent or transient occlusion of a coronary artery. In both studies, only minor association of nanoparticles with inflammatory cells was observed, indicating that the contrast enhancement in the infarcts observed on T1-weighted imaging reported on the perfusion status of the myocardium and vascular hyperpermeability rather than on the presence of inflammatory cells. Previously, such passive accumulation of Gd-labeled liposomes and micelles was also observed in mouse atherosclerotic plaques [31, 32]. Passive accumulation of nanoparticles through damaged vasculature in diseased tissues leads to background signal that reduces the contrast to background ratio for imaging with targeted nanoparticles. Passive accumulation must be considered not only for the Gd-containing liposomes, but also for other types of nanoparticles such as the iron oxides. On the other side, massive accumulation of nanoparticles in the myocardium opens new opportunities for delivering large payloads of therapeutics [29, 30•]. Secondly, T1 and contrast changes induced by Gd-containing nanoparticles may be less appropriate for absolute quantification purposes, since several studies have shown that the relaxivity of the nanoparticles is strongly reduced upon internalization of the nanoparticles in cells [33–37].

A powerful alternative to iron oxide and Gd-labeled nanoparticles for imaging inflammation are 19F-containing nanoparticles. The 19F atom can be imaged directly with 19F-MRI and since the human body contains very little native fluorine, the nanoparticles can be imaged in an MRI silent background enabling unambiguous identification of inflammatory ‘hot spots’. A recent example of such an approach applied to the heart is illustrated in Fig. 3 [38, 39••]. Here, perfluorocarbon (PFC) nanoparticles containing a high payload of 19F atoms were injected in mice shortly after infarct surgery. Flow cytometry revealed that approximately 50 % of the blood monocytes and macrophages were labeled with PFC 2 h after injection. Labeled inflammatory cells subsequently infiltrated the infarct myocardium, where they could be imaged with 19F-MRI (Fig. 3). Signal intensities from the 19F-density MRI sequence are proportional to the amount of 19F and there is no background signal other than noise, and therefore the 19F-MR images report on the severity of the inflammation. The approach may therefore be applicable to monitoring the efficacy of therapeutic interventions. PFCs are nontoxic and biochemically inert and some PFC-emulsions already have been clinically applied as an artificial blood substitute [40]. We may therefore expect a rapid clinical translation of this approach to inflammation imaging.

Fig. 3
figure 3

Imaging of perfluorocarbon (PFC) emulsions after myocardial infarction in mice. Injection of the PFCs was done 2 h after infarct surgery. MRI was performed after 4 days. a 1H-MRI of the mouse heart for anatomical reference. b 19F-MRI of the same slice and c a 19F/1H overlay image showing ‘hot spots’ in the infarct (indicated by I) and in the wound where the thorax was opened for surgery (indicated by T). d Bright-field microscopy confirmed the location of the infarct. e PFC rhodamine fluorescence and f image overlay revealed colocalization of PFCs with monocytes and macrophages in the infarct. Reproduced from [38, 39••], with permission

Full size image

Various Targets

To improve specificity, nanoparticles may be equipped with moieties that target inflammation or other processes involved in the CVD pathways. Maiseyeu et al. [41] reported on the use of phosophatidylserine-containing liposomes for specific targeting of macrophages in atherosclerotic plaque. Phosphatidylserine (PS) is known to promote phagocytosis by inflammatory cells. In an ApoE−/− mouse model of atherosclerosis, injection of Gd- and PS-containing liposomes resulted in distinct signal enhancement in T1-weighted images of plaques in the aorta (Fig. 4). Liposomes in the plaques colocalized with macrophages. In a similar approach, Geelen et al. [42, 43] developed a Gd-containing liposome with PS for imaging of myocardial infarction and showed that liposomes with 6 mol % PS were maximally internalized by murine macrophages.

Fig. 4
figure 4

Liposomes containing Gd and phosphatidylserine (PS) for specific recognition by macrophages in mouse atherosclerotic plaque. After injection of PS-containing liposomes, specific signal enhancement could be observed in plaques in the aorta of ApoE−/− mice (white arrows in the inset images). Maximum contrast was observed 2 h after injection. IVC inferior vena cava, L lumen, F abdominal fat. The bottom-right image is a histological section stained with Hematoxylin-Eosin. Reproduced from [41], with permission

Full size image

Nanoparticles may be equipped with peptides, proteins or antibodies for targeting monocytes/macrophages [22, 44–49]. However, use of such ligands may be less suitable for clinical translation because of costs, toxicity and problems with immunogenicity. Nevertheless, the use of recognition ligands opens the opportunities for molecular MR imaging of other components of the coronary artery disease pathways, including dysfunctional endothelium, enzymatic activity, cell death, neovascularization, thrombosis, and extracellular matrix production or breakdown.

Targeted nanoparticles for MR imaging of inflamed endothelium were developed and applied in proof of concept studies in murine models of atherosclerosis [50–54], stroke [55–58], renal injury [59] and myocardial infarction [60, 61]. MMP expression was imaged using a small peptide with affinity for MMPs covalently bound to a small Gd-chelate in atherosclerosis [62–64]. Te Boekhorst et al. [65] developed Gd-containing micelles targeted towards neutrophil gelatinase-associated lipocalin-2 (NGAL), a protein that is associated with adverse cardiovascular events in patients after carotid endarterectomy. A small Gd-based contrast agent (Gd-MPO) was developed to image the enzyme MPO, excreted by monocytes, macrophages and neutrophils in inflamed tissues [66, 67]. The enzymatic activity of MPO leads to polymerization of Gd-MPO, which due to its increased size, gets subsequently trapped in the tissue. The Gd-MPO probe was validated in animal models of inflammation in atherosclerosis [68] and myocardial infarction [69•]. A number of targeted Gd- and iron oxide-based nanoparticles were developed for imaging apoptotic cell death in atherosclerotic plaque and infarcted myocardium [70–78]. Necrosis could be imaged with MRI using a Gd-based contrast agent that binds to exposed DNA fragments after necrotic cell death [79, 80]. The role of neovascularization in atherosclerosis was extensively studied using Gd- and 19F-containing nanoparticles [81–86]. Various products in the coagulation cascade, including fibrin, have been exploited for molecular MRI of thrombus using targeted Gd-chelates and nanoparticles [87–95]. Fibrosis plays a major role in atherosclerosis and a thick fibrous cap shielding the blood from thrombogenic material is considered to be a sign of a stable plaque. Excessive collagen formation in myocardial disease reduces cardiac function and may induce cardiac arrhythmia. Various targeted approaches for imaging fibrosis with MRI have been described in literature, using collagen- and elastin-binding peptides [96–100••] and proteins [101–104].

Molecular MRI using nanoparticles may develop into a powerful asset for studying CVD, particularly when combined with the more traditional functional and physiological MRI methods that have long been available for assessing the heart and blood vessels. There is certainly no lack of new ideas and many new clever concepts for molecular imaging of CVD beyond the strategies outlined in this review are continuously being developed. Unfortunately, many of the new concepts have not survived beyond proof of principle studies in (mostly murine) animal models of CVD. Imaging studies in small animal models of disease are certainly valuable for gaining increased understanding in disease mechanisms and for developing new treatment strategies with emerging therapeutics. Nevertheless, ultimately the goal is to translate successful concepts to clinical research and routine patient care. The reasons for lacking or slow translation towards human application are manifold. First of all, there are practical hurdles considering costs and industrial return on investment for a widespread use of targeted nanoparticles in diagnostic routine. Also, there are serious concerns about the safe application of nanotechnology in humans [105, 106]. The use of proteins, peptides or antibodies may lead to immunogenicity problems. Additionally, there are complications concerning slow blood clearance kinetics of certain nanoparticles and unfavorable clearance pathways via the liver and spleen, which may lead to accumulation of metals and other nanoparticle material in the body with uncertain and yet unknown long-term health consequences. To address some of these issues,interest has grown in the use of natural nanoparticles such as viruses [107], (apo-) ferritin [108–110] and lipoproteins [111], suitably modified for molecular imaging purposes. The use of modified high-density-lipoprotein (HDL) nanoparticles is specifically relevant for CVD and seems particularly promising, and therefore will be discussed in more detail in the following section.

HDL Nanoparticles

HDL is composed of a hydrophobic core of mainly triglycerides and cholesterol esters surrounded by a phospholipid monolayer. The enclosing shell contains the apolipoprotein A-1 (apoA-I). HDL transports cholesterol in the body and enables reverse cholesterol transport from atherosclerotic plaque. HDL has anti-thrombotic properties and modulates plaque inflammation and oxidation [112]. The HDL particle therefore plays a key role in plaque etiology, and application of HDL can serve both imaging [113] as well as therapeutic purposes [114]. HDL can be modified to include contrast-generating material. Incorporation of Gd in the phospholipid shell facilitates MR imaging of atherosclerotic plaque [115–117]. Various versions of modified HDL particles were designed that enabled imaging not only with MRI, but also with CT and optical imaging [113]. Moreover, the modified HDL particles were used to study lipoprotein–lipoprotein interactions [118].

Gold containing HDL (Au-HDL) nanoparticles were employed to visualize atherosclerotic plaque in mice using spectral CT imaging [119••]. Spectral CT enables identification of materials by taking advantage of the polychromatic nature of the X-ray spectrum when creating CT images. Atherosclerotic plaques in the abdominal aorta of ApoE−/− mice were studied (Fig. 5). Au-HDL was injected 24 h prior to imaging. Shortly before imaging, a conventional iodine-based CT contrast agent was also injected for visualization of the vasculature. As shown in Fig. 5, spectral CT images enabled differentiation of the X-ray spectrum into gold, iodine, tissue and calcium (bone). High attenuation of X-rays in the atherosclerotic plaque could be attributed to accumulation of Au-HDL, which was found to largely colocalize with macrophages. This study demonstrates that molecular imaging with CT and biologically relevant nanoparticles is possible. A limitation to the study was that CT scanning times were impractically long because of low detector efficiency. Imaging therefore had to be performed ex vivo. However, this limitation may be overcome by improved detector technology.

Fig. 5
figure 5

Ex vivo spectral CT images of the thorax and abdomen of an ApoE−/− mouse with atherosclerotic plaque in the abdominal aorta. Mice were injected with Au-HDL nanoparticles and traditional iodinated CT contrast, 24 h prior to and just before imaging, respectively. a Conventional CT image in sagittal orientation. b Spectral CT image of gold. c Overlay showing Au-HDL accumulation in the aortic plaques. d Conventional CT image in transverse orientation. e Overlay image of gold (yellow), iodine (red; vasculature), photoelectric effect (blue-white; bone/calcium-rich), and Compton effect (gray; soft tissue). Reproduced from [119••], with permission (Color figure online)

Full size image

Recently, a collagen-specific peptide conjugated HDL nanoparticle was developed and used for MR imaging of compositional changes in atherosclerotic plaque regression [100••]. Pre-clinical studies were performed in the so-called Reversa mouse model, in which plaque formation can be reversed by switching from a high-fat diet to normal chow, resulting in a decrease in the number of plaque macrophages and an increase in plaque collagen content [120]. HDL collagen specificity was introduced by attaching a collagen-specific peptide (EP-3553). HDL conjugated to non-specific peptide (EP-3612) and non-conjugated HDL were used as a control. Figure 6 summarizes the main findings of the study. At day 0, high macrophage and low collagen content in the plaques resulted in significant enhancement in T1-weighted MRI of the aortic plaques after injection of non-specific controls, whereas only minor signal increase was observed with collagen-specific HDL. After 28 days of plaque regression in the Reversa mice, enhancement was primarily observed after injection of collagen-specific HDL, indicating increased collagen and reduced macrophage content in the stabilized plaques. This study elegantly showed that the HDL nanoparticles can be used for in vivo MRI monitoring of plaque composition and changes therein after treatment.

Fig. 6
figure 6

MR imaging of atherosclerotic plaque regression with HDL and collagen-specific HDL nanoparticles. a Schematic drawing of high-density lipoprotein (HDL) nanoparticles functionalized with collagen-specific EP3553 peptides. b Representative transversal MR images of a Reversa mouse pre and 24 h post-injection of EP3553-HDL with enhanced signals in the regressing atherosclerotic plaques (arrows). c Normalized enhancement ratio (NER; plaque signal normalized to muscle) before (day 0) and after regression (day 28) resulting from injection of HDL, collagen-specific EP3553-HDL, or non-specific EP3612-HDL. Increased NER at day 28 for EP3553-HDL is an imaging readout for increased collagen content and decreased macrophage activity in the regressed plaques. Reproduced from [100••], with permission

Full size image

Conclusions and Outlook

The last decade has witnessed an increased interest in the use of targeted nanoparticles as signal beacons for cardiovascular molecular imaging. The probes hold great promise for the facilitation of more specific patient diagnosis and treatment follow-up in the clinical management of atherosclerosis and myocardial infarction. Already a huge library of nanoparticles is available for specific imaging of various important hallmarks of CVD, including inflammation, dysfunctional endothelium, enzymatic activity, cell death, neovascularization, thrombosis, and extracellular matrix production or breakdown. Nevertheless, the field is eagerly awaiting clinical translation of some of these new disease-specific imaging technologies. The use of the body’s natural nanoparticles, such as HDL, for molecular imaging may accelerate such clinical translation. Generally, more knowledge is needed on the long-term fate of nanoparticles in the body and on the organ-clearance kinetics and pathways in view of possible toxicity.

The cardiovascular molecular imaging technology that we reviewed here can be employed for diagnosis or individualized evaluation of therapeutic interventions. Alternatively, nanoparticles can be equipped both with an imaging agent as well as with a therapeutic compound for a combined imaging and treatment protocol. This concept belongs to the broader field of theragnostics and personalized medicine [121, 122]. Treatment of myocardial infarction by stem cells to restore cardiac function and improve myocardial perfusion has shown promising results in preclinical models. However, the rapid translation of stem cell treatment into the clinic has not yet yielded satisfactory results. Labeling of cells with nanoparticles for (cellular) molecular imaging of stem cell fate, migration, survival and engraftment gives better insights into therapeutic benefits and eventually may lead to improved treatment strategies [123]. Finally, regeneration of human cardiac tissue after myocardial infarction is actively explored in the field of cardiac tissue engineering, which aims to re-establish the structural and functional features of native myocardium, in a predictable way and on a long-term basis [124]. The development of cardiac tissue engineering is slowed down by the lack of suitable imaging technology for in vivo monitoring of implantation, cell proliferation, and tissue development. Here, nanoparticle-based molecular imaging could also prove to be of significant benefit.