Sage Journals: Discover world-class research

Abstract

Objective: Magnetofluorescent nanoparticles (MFNPs) offer the ability to image cellular inflammation in vivo. To better understand their cellular targeting and imaging capabilities in atherosclerosis, we investigated prototypical dextran-coated near-infrared fluorescent MFNPs in the apolipoprotein E-deficient (apo E−/−) mouse model.

Methods and Results: In vitro MFNP uptake was highest in activated murine macrophages (p < .001). Apo E−/− mice (n = 11) were next injected with the MFNP (15 mg/kg iron) or saline. In vivo magnetic resonance imaging (MRI) demonstrated strong plaque enhancement by the MFNPs (p < .001 vs. saline), which was confirmed by multimodality ex vivo MRI and fluorescence reflectance imaging. On fluorescence microscopy, MFNPs were found in cellular-rich areas of atheroma and colocalized with immunofluorescent macrophages over endothelial cells and smooth muscle cells (p < .001).

Conclusions: Here we show that (1) the in vitro and in vivo cellular distribution of atherosclerosis-targeted MFNPs can be quantified by using fluorescence imaging methods; (2) in atherosclerosis, dextranated MFNPs preferentially target macrophages; and (3) MFNP deposition in murine atheroma can be noninvasively detected by in vivo MRI. This study thus provides a foundation for using MFNPs to image genetic and/or pharmacological perturbations of cellular inflammation in experimental atherosclerosis and for the future development of novel targeted nanomaterials for atherosclerosis.

Keywords

Atherosclerosis MRI fluorescence nanoparticles inflammation

Introduction

Nanoparticulate imaging agents are often of similar dimensions (1–100 nm) as molecules that interface with the surface or interior of cells, and therefore offer a unique window into the nanobiology of cells [1,2]. Indeed, a wave of novel nanomaterials have recently emerged that will ultimately enable cellular and molecular imaging in cardiovascular disease [3]. Within atherosclerosis, a variety of magnetic nanoparticles (MNPs) have been employed to report on various aspects of inflammation [3 –5]. For example, up-regulation of endothelial adhesion molecules, proteases, and peroxidases play key roles in the inflammatory process and represent ideal imaging targets [6]. While MNPs specially targeted to these processes are currently being developed, a number of polymer-coated MNPs are in advanced clinical trials. MNP agents tested in atherosclerosis include carboxydextran-coated ferumoxtrans [7,8], short-circulating ferumoxides [9], and long-circulating dextran-coated ferumoxtrans [10 –13]. Although some MNPs are detectable by in vivo high-resolution magnetic resonance imaging (MRI), their reporter capabilities, fate, and specific cellular distribution in atherosclerosis are not entirely understood.

To further extend the imaging capabilities of MNPs, we have developed fluorescent moieties of MNPs, or magnetofluorescent nanoparticles (MFNPs). In addition to conferring strong, reversible relaxation (R2 > R1) effects on MRI, these multimodality nanomaterials also possess a distinct fluorescence signature that permits detection in a variety of fluorescence imaging systems such as flow cytometry, fluorescence reflectance imaging (FRI), and fluorescence microscopy. These dextranated MFNPs modified with near-infrared or visible fluorochromes have been useful in imaging cellular inflammation in models of cancer [14] and diabetes [15].

To better understand the cellular targeting and imaging capabilities of MNPs in atherosclerosis, we characterized the in vitro and in vivo cellular distribution of prototypical dextran-coated MFNPs by capitalizing on their fluorescence properties. We next examined the ability of in vivo MRI to noninvasively image these atherosclerosis-targeted MFNPs and confirmed the results with ex vivo multimodality MRI and FRI.

Materials and Methods

Synthesis of the Near-Infrared Fluorescent Magnetofluorescent Nanoparticles

The imaging agent used was a monocrystalline iron oxide magnetic nanoparticle (MION-47) [16], consisting of a 3-nm core of (Fe₂O₃) _n (Fe₃O₄) _m , similar to preparations investigated in clinical trials [17]. The core was covered with a layer of 10-kDa dextran, which was crosslinked with epichlorohydrin to form cross-linked iron oxide (CLIO)-47. CLIO-47 was then aminated by reacting with ammonia to provide primary amine groups crosslinked to CLIO-47 (CLIO–NH₂). Of note, CLIO has been shown to have nearly identical biological and physical properties as clinical analogues of long-circulating MNPs [18]. In mice, these MNPs have a blood half-life of approximately 11 hr [18], irrespective of fluorochrome conjugation (unpublished results). The nanoparticle had an overall size (volume weighted) in aqueous solution of 38 nm, an R1 of 21 mM sec⁻¹, an R2 of 62 mM sec⁻¹ (37°C, 0.5 T), and had an average of 62 primary amines available for conjugation. To render the MNP fluorescent in the near-infrared, 1 mg of the fluorochrome cyanine 5.5 (excitation/emission 673/692 nm, Amersham, Piscataway, NJ) was dissolved in DMSO and reacted with 1 mg iron CLIO–NH₂. We intentionally chose a reporter fluorochrome in the near-infrared window [19] to minimize autofluorescence from the elastin, collagen, and lipid components within atheroma [20]. The final near-infrared fluorescent (NIRF) MFNPs had an iron concentration of 6.5 mg/ml and yielded an average of 2.5 NIR fluorochromes per nanoparticle.

Quantification of MFNP Uptake by Murine Cells

MFNP uptake into activated macrophages, endothelial cells, and smooth muscle cells was quantified by using flow cytometry. Monocytes were isolated via density centrifugation and then plated onto cell-culture-treated dishes and cultured in complete monocyte medium containing human recombinant granulocyte–macrophage colony stimulating factor (GM-CSF, BD Biosciences, Rockville, MD) at a concentration of 12 ng/ml. The cells were maintained at this concentration of GM-CSF for 7 days before use in experiments. Murine heart endothelial cells were then isolated as previously described [21]. Activation was induced by incubation of 25 ng/ml of tumor necrosis factor alpha for 4 hr. For smooth muscle cell studies, NIH 3T3 cells were obtained from ATCC (Manassas, VA), passaged according to instructions, and activated with 50 μg/ml of lipopolysaccharide for 24 hr. Activated cells of each subtype were then incubated with the MFNP (0.1 mg iron/ml) for 4 hr and then analyzed via flow cytometry (10,000 cells per sample; FACSCalibur, Becton Dickinson, Franklin Lakes, NJ). The experiments were performed in duplicate or triplicate.

Apolipoprotein-E-Deficient Model of Atherosclerosis

We employed the apolipoprotein E gene deletion mouse (apo E−/−) as a model of atherosclerosis [22]. A total of 11 mice (average age 29 weeks) were used in the study. Mice received a high-cholesterol diet (TD 88137 adjusted-calories diet; Harlan Teklad, Indianapolis, IN) for an average of 19 weeks. Eight mice were injected with the MFNP (15 mg/kg iron), and 3 mice were injected with 0.9% saline. All animal studies were approved by the Subcommittee on Research Animal Care at Massachusetts General Hospital.

Magnetic Resonance Imaging

Imaging studies were performed 24 hr after MFNP or saline injection. A subset of mice (n = 4, MFNP; n = 3, saline) first underwent high-field MRI. In vivo MRI studies were performed under inhalational anesthesia (halothane, 1–2% vol/vol + 2 L O₂) on a 9.4-T horizontal-bore small-animal scanner (Bruker Biospec, Billerica, MA). Bright-blood cine images were obtained with electrocardiography (ECG) and respiratory gating (SA Instruments, Stony Brook, NY) using a gradient-recalled echo FLASH sequence to create bright-blood contrast. An MRI-compatible warm-air blower (SA Instruments) was used to maintain physiologic temperature. After a series of scout images, an oblique slice was defined to show the aortic root, the ascending aorta, and the arch. Imaging parameters were as follows: FOV 25 mm, slice thickness 1 mm, matrix 200 × 200 (in-plane resolution 125 × 125 µm), 16 frames per RR interval (TR 7.0–8.0 msec); TE 2.7 msec, flip angle 30°, eight signal averages. A second acquisition with TE 4.5 msec was obtained to further enhance susceptibility contrast and MNP detection. The total imaging time was <2 hr per animal. All images were linearly mapped to a 16-bit absolute intensity grayscale.

All animals were then sacrificed and perfused with saline. For high-resolution ex vivo MRI (n = 2 MFNP groups), NIR fluorescence imaging (see below) was performed first. The aorta was then perfused with liquid agar, immersed in liquid agar, and allowed to cool. This technique minimized aortic collapse, air–tissue interface susceptibility artifacts, and movement during imaging. Aortae were then placed in a 14-T vertical-bore system with microimaging capabilities (Bruker). Gradient-recalled echo images were obtained using the following parameters: in-plane resolution 70 × 70 m, slice thickness 0.5 mm, TR 200 msec, TE 3.0 and 6.0 msec. The total imaging time was <1 hr per aorta. After imaging, the aorta was removed from the agar and embedded in OCT.

NIR Fluorescence Reflectance Imaging

Resected aortae (n = 11) were imaged by using a custom-built FRI system [23]. Light and NIRF images were obtained with respective exposure times of 75 msec and 60 sec. NIR fluorescence was detected by a 12-bit CCD camera (Kodak, New Haven, CT) equipped with a special C-mount lens and an NIRF band-pass emission filter (680–720 nm, Omega Optical, Brattleboro, VT). Images were digitally acquired as 16-bit images.

Cellular Assessment of MFNP Deposition in Atheroma

The resected proximal aorta was frozen in OCT compound (Sakura Finetek, CA). The well-validated Paigen's approach was used to analyze the aortic root [24]. Cryostat sections (6 m thickness) were stained with hematoxylin and eosin (H&E) for general morphology. Cells were identified with cell-specific antibodies: anti-Mac-3 for macrophages, anti-CD31 for endothelial cells (both from BD Biosciences Pharmingen), and anti-alpha smooth muscle actin for smooth muscle cells (Lab Vision, Fremont, CA). Immunofluorescent staining was performed on serial sections of the mouse aortic root. Briefly, sections treated with 0.3% hydrogen peroxide were incubated for 60 min with primary or matched control antibodies, followed by respective biotinylated secondary antibody, and finally streptavidin conjugated with Texas Red (Amersham). Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI).

Plaque sections were imaged using an upright epifluorescence microscope (Eclipse 80i, Nikon, Melville, NY) with a cooled CCD camera (Cascade, Photometrics, Tucson, AZ). Fluorescence images were obtained in four spectrally resolved channels: 630 nm (for immunofluorescent cell-specific antibodies), 710 nm (for MFNP), 400 nm (for nuclear DAPI staining), and 535 nm (for autofluorescence). Exposure times ranged from 100 to 1000 msec depending on the channel. Fluorescence images were acquired as digital 12-bit images and then fused to form multicolor fusion images (IP Lab Spectrum version 3.9.3, Scanalytics, Rockville, MD). For each of the cell types detected by immunofluorescence and DAPI staining, the number of MFNP-positive cells was counted on 10 high-power fields (hpf) for each aortic root, a common approach to quantify cells in histological sections [25].

Image Analysis

MRI and NIRF aortic images were analyzed with a shareware software package (OsiriX). Within the aorta, plaque, and adjacent tissue, signal intensities were measured by using a manual region-of-interest (ROI) tracing tool. For MRI, the blood–plaque contrast-to-noise ratio (CNR) was calculated as follows: CNR = (blood signal – plaque signal) / standard deviation of the noise. The plaque ROI in the aortic root was manually defined from the strong negative contrast in the MFNP group; a similarly sized and located ROI was used for the saline group analysis. For FRI, the plaque target-to-background ratio (TBR) was calculated as TBR = (ROI average plaque signal) / (ROI average aortic background signal adjacent to the plaque). The peak plaque TBR was reported for each animal. After image analysis, images were further windowed by using Photoshop 7.0 software (Adobe Systems, San Jose, CA).

Statistical Analyses

Data are presented as mean ± SD. For differences between multiple groups, the one-way ANOVA test was used followed by a post hoc Tukey's test for multiple comparisons (Prism 4.0a, Graph Pad Software, San Diego, CA). For differences between two groups, the Student's t test was used. A value of p < .05 was considered statistically significant.

Results

Activated Macrophages Preferentially Bind MFNPs in Vitro

To quantitatively assess the cellular distribution of MFNPs in cells found in atherosclerotic lesions, we performed flow cytometry on activated murine macrophages, endothelial cells, and smooth muscle cells incubated with the MFNPs (0.1 mg iron/ml for 4 hr; Figure 1). MFNP uptake was 7.9-fold greater in activated macrophages (mean fluorescence intensity [MFI] 108.5 ± 1.6 arbitrary units [AU]) compared to activated endothelial cells (MFI, 13.7 ± 0.8 AU; p < .001) and 8.4-fold greater compared to activated smooth muscle cells (MFI, 12.9 ± 0.1 AU; p < .001). All of the cell lines examined exhibited only one Gaussian-distributed population by FACS with 98% of the macrophages labeled, 1.1% of the 3T3 cells labeled, and 8.1% of the endothelial cells labeled.

Figure 1.

In vitro cell uptake of prototypical dextranated MFNPs in mouse atherosclerotic cells under stimulated conditions. Fluorescence intensity (arbitrary units [AU]) was measured by flow cytometry. Activated macrophages demonstrated significantly more MFNP uptake than activated endothelial cells or smooth muscle cells. The dashed line represents the autofluorescence level (10 AU) for all experiments. MAC = macrophages, EC = endothelial cells, SMC = smooth muscle cells. ^*p < .001, MAC versus EC, MAC versus SMC.

MFNP Deposition in Murine Atheroma Can Be Detected by in Vivo MRI

A total of seven apo E−/− mice underwent high-field MRI (n = 4 MNP, n = 3 saline). In MFNP-injected mice, in vivo 9.4-T gradient-recalled echo images revealed strong focal signal loss in the aortic root as well as the lesser curvature of the transverse arch (Figure 2), common sites of aortic atherosclerosis in the apo E−/− mouse [22]. The in vivo signal-to-noise ratio (SNR) of atherosclerotic plaque in the aortic root was lower in the MFNP group (SNR 13.6 ± 2.4 vs. saline 44.9 ± 10.9, p < .05), consistent with nanoparticle-induced signal loss [7,8,10 –13]. A trend to lower blood SNR was noted in the MFNP group (SNR 37.6 ± 1.7 vs. saline 52.3 ± 11.0, p = .19), consistent with the long blood half-life of MNPs in mice [18]. The overall blood–plaque CNR increased by 310% in the MFNP group compared to the saline group (24.0 ± 0.7 vs. 7.8 ± 0.7, p < .001).

Figure 2.

(A) In vivo bright-blood aortic MR images of representative apolipoprotein E-deficient (apo E−/−) mice. Twenty-four hours following either saline or MFNP 15 mg/kg iron oxide injection, T2^*-weighted gradient-recalled echo images (TE 2.7 msec for all images) were obtained by ECG and respiratory gating by using a 9.4-T MR scanner. (Top) In apo E−/− mice that received saline, minimal signal loss was noted in the aorta (left, long axis view), aortic root (center, magnified view from dashed box in left image) and in the transverse aortic arch (right, magnified view from perpendicular slice defined by the dashed line in the left image). (Bottom) In contrast, apo E−/− mice that received MFNP demonstrated strong focal signal loss in the aortic root (left and middle, arrow demonstrating a “W” pattern of negative signal enhancement) and transverse aortic arch (right, arrow). In the MFNP group, the blood-to-plaque contrast-to-noise ratio (CNR) was 24.0 ± 0.7 (vs. 7.8 ± 0.7 in the saline group, p < .001). (B) Ex vivo 14-T high-resolution aortic MRI of representative apo E−/− mice that received MFNP. Focal MRI signal loss was noted in the aortic root and transverse arch, similar to the in vivo images (arrows). Furthermore, greater signal loss was noted with longer echo times (TE), consistent with a T2^*-dependent relaxation effect from iron oxide. The cartoon below identifies the major cardiovascular structures of interest. AR = aortic root, LV = left ventricle, TA = transverse arch, CA = carotid artery.

Figure 3.

Ex vivo macroscopic light (left) and near-infrared fluorescence (NIRF) reflectance images of representative apo E−/− mouse aortae. (Top) In control apo E−/− mice that received saline, minimal NIRF signal was noted within atherosclerotic plaque. (Bottom) In contrast, MFNP-injected mice demonstrated strong focal NIRF signal enhancement in plaques in the aortic root, transverse arch, and descending aorta, corresponding to visible plaques (arrows) on the light image (left). NIRF images were processed and windowed identically.

Ex vivo 14T confirmed the in vivo MRI findings by also demonstrating strong negative signal enhancement in aortic root and aortic arch plaques, as seen in vivo (see Figure 2). Furthermore, longer echo times substantially increased the MR signal loss in aortic plaques compared to the normal aorta, consistent with MNP deposition in plaque and a T2^*-dependent effect [26].

MFNP Deposition in Atheroma Can Be Detected by Fluorescence Reflectance Imaging

Aortae from all 11 apo E−/− mice underwent macroscopic NIR FRI. In the 8 mice that received MFNPs, strong NIRF signal enhancement was noted in atherosclerotic plaques in the aortic root and aortic arch (Figure 3), confirming the results seen during in vivo and ex vivo MRI. There was good correspondence between visible aortic plaques seen on light images and NIRF signal. No significant NIRF signal enhancement was seen in the 3 control apo E−/− mice that received saline. The peak plaque-to-background ratio in the NIRF MNP group was 423% higher than in the saline group (5.9 ± 2.5 vs. 1.6 ± 0.4, p < .01).

MFNPs Colocalize with Immunofluorescent Cells in Apo E−/− Atheroma

Focal NIRF signal was noted in aortic plaque sections of apo E−/− mice that received the MFNP imaging agent (Figure 4). Signal was detected in various cellular-rich areas of atherosclerotic lesions as well as the adventitia adjacent to atheroma. There was minimal NIRF signal in the cellular-poor areas, the normal aorta, or in aortic plaques of apo E−/− mice that received saline. On correlative fluorescence microscopy, MFNPs were readily identified within immunofluorescent cells in atherosclerotic lesions (Figure 5). Macrophages were the predominant MFNP-positive cell type (20.6 ± 7.3/hpf) compared to endothelial cells (6.6 ± 1.8/hpf) and smooth muscle cells (4.3 ± 1.8/hpf, p < .0001). Therefore, macrophages represented 65% of the MNFP-positive cells. MFNP-positive endothelial cells and smooth muscle cells were often located adjacent to MFNP-positive macrophages, suggesting the presence of focal zones of cellular inflammation within plaques.

Figure 4.

Representative microscopic NIRF and histopathological (H&E) images of adjacent sections of the aortic root of apo E−/− mice. (Top) Saline-injected mice did not demonstrate focal NIRF enhancement within atheroma (original magnification, ×25). (Middle) MFNP-injected mice, in contrast, demonstrated marked focal NIRF enhancement in atherosclerotic lesions (original magnification, ×25). (Bottom) Higher magnification image demonstrates MFNP deposition in areas of cellular infiltration (arrowheads) but not in relatively acellular areas (original magnification, ×100).

Figure 5.

Fluorescence microscopy of MFNP distribution in atherosclerotic lesions. Cell types were immunofluorescently labeled in a distinct fluorescent channel from the NIR. (Top left) Quantitative analysis demonstrated significantly higher number of MFNP-positive macrophages per high-power field (hpf) compared to endothelial cells and smooth muscle cells (p < .0001). (Top right, bottom left, bottom right) Representative fused fluorescence microscopic images of the MFNP (red) and immunofluorescently labeled cell antigens (green). (Top right) Macrophages represented the majority of MNFP-positive cells and demonstrated yellow-orange NIRF and immunofluorescent signal colocalization (arrowheads). (Bottom left) Modest MFNP uptake into endothelial cells (arrowheads) was also seen adjacent to strongly MFNP-positive subendothelial cells, presumably macrophages (arrows). (Bottom right) Similarly, MFNP uptake by smooth muscle cells of the fibrous cap or media (arrowheads) was found adjacent to MFNP-positive subendothelial areas of atherosclerotic lesions (^*, asterisk). Occasional uptake was also noted in the adventitial layer of the atherosclerotic aorta (arrow). Original magnification, ×200. MAC = macrophages, EC = endothelial cells, SMC = smooth muscle cells. ^*p < .001, MAC versus EC, MAC versus SMC.

Discussion

In this report, we determined the cellular targeting and multimodal imaging capabilities of clinical-type magnetofluorescent nanomaterials in experimental murine atherosclerosis. We found that both the in vitro and in vivo cellular uptake of MFNPs was significantly higher in macrophages compared to endothelial cells and smooth muscle cells. Second, we demonstrated that the MFNPs allow imaging of cellular inflammation in murine atherosclerosis by in vivo MRI, as well as ex vivo multimodality imaging. These results have implications for the development of novel atherosclerosis-targeted imaging nanomaterials and for the noninvasive assessment of inflammation in atherosclerosis.

A variety of nanosized materials has been used as imaging agents for in vivo MRI [3 –5,27 –29]. In the current investigation, we employed a dextran-coated MFNP with a hydrodynamic diameter of 38 nm. This size is similar to that of large molecular weight fluorescent dextrans [30] that bind to SIGNR1 and SIGNR3, recently described C-type lectins found on macrophages [31,32]. This finding provides a possible link between the MFNP's nanosize dimension, dextran coating, and cell-internalization properties. Another important aspect regarding the MFNP's size is that it can avoid detection by the reticuloendothelial system, resulting in a prolonged blood half-life and greater vascular bioavailability compared to larger sized MNP preparations [33]. Additional favorable features of the multimodal MFNP include its fluorescence readout capabilities, including FRI, fluorescence microscopy, and flow cytometry for quantitative studies.

In particular, the fluorescence properties of the MFNP allowed assessment of its cellular distribution in cell types found in atherosclerosis (Figures 1 and 5). We found that prototypical dextranated MFNPs accumulated predominantly in macrophages (65% of the MFNP-positive cells, p < .001), consistent with prior data derived from histopathological analyses [7 –13]. Interesting, however, was that endothelial cells and smooth muscle cells in atheroma also took up the MFNP, albeit to a lesser degree. These nanoparticle uptake findings are consistent with prior in vitro endothelial cell data [34] and with in vivo data on human atheroma [12]. In addition, MFNP uptake by these cells within atherosclerotic lesions, but not within the normal vessel wall, is consistent with their activated state within atheroma [6].

To first approximation, the developed MFNP uptake assay predicted the cellular distribution of the nanoparticles in vivo, as demonstrated in the current apo E−/− model. This assay can therefore provide an efficient method to characterize the cellular distribution and efficiency of novel atherosclerosis-targeted nanomaterials. What reasons might justify the development of new MFNP agents targeted to atheroma? Two observations may be relevant. First, although clinical-type MFNPs have a macrophage-dominant cellular distribution (as shown here and in other studies [7 –13]), we found that the background MFNP uptake by endothelial cells and smooth muscle cells was 35% of the MFNP-positive cells. This feature may diminish the ability of clinical-type MFNP to specifically image macrophages, the key effector cell in atherogenesis and plaque rupture [6]. Second, MFNP delivery into macrophages could be augmented, in theory, by surface-modified MFNPs containing ligands that avidly bind to macrophages (e.g., to scavenger receptor A [6]). Higher concentrations of MNPs in atheroma could greatly improve the sensitivity of this approach for imaging cellular inflammation in atherosclerosis both experimentally and clinically [17].

Using high-field cardiac and respiratory-gated MRI, we demonstrated that prototypical MNPs could be imaged in murine atheroma in vivo (Figure 2). MFNP deposition in atheroma resulted in a significant tripling of the blood–plaque CNR compared to saline-injected apo E−/− mice. The blood SNR was 28% lower in MFNP-injected animals compared to saline-injected animals. This finding demonstrates that the high blood–plaque CNR was due to nanoparticle-conferred R2 relaxation effects in plaque, rather than R1 positive signal enhancement effects in blood. This feature of MNP is expected at higher B₀ magnetic fields, where R1 relaxation rates fall significantly, resulting in net higher R2/R1 ratios [35]. MFNP-enhancement of atheroma was further confirmed on ex vivo MRI and NIRF imaging, demonstrating the multimodality imaging capability of the MFNP agent (Figures 2 and 3). Compared to the only other imaging study to demonstrate feasibility of MNP-enhanced MRI in murine atherosclerosis [34], there are several important differences in the present study. First, we investigated clinical-type (underivatized) MFNPs that primarily target macrophages in atherosclerosis, as opposed to a vascular cell adhesion molecule 1-targeted derivatized MFNPs. Second, we did not require the administration of an additional contrast agent (protected graft copolymer–gadolinium) to enhance the blood signal, likely because we employed powerful gradient coils, physiologic gating, and optimized MR pulse sequences. Third, we were able to image MFNP deposition specifically in the aortic root of the apo E−/− mouse, an area of particular biological interest.

There are limitations to the present study. Changes in the MFNP incubation time, iron concentration, and different cell types employed could modify the observed flow cytometry results. Similarly, the in vivo iron dose was relatively high (but less than in some other experimental studies [9,10]), which could influence its distribution in plaque. Time points other than 24 hr after MFNP injection were not investigated; other time points could result in a different cellular distribution of MFNP in apo E−/− atheroma. Lymphocyte and other minority cell populations in atheroma were not investigated in this report, but could also be part of the MFNP cellular profile. Finally, stereological or three-dimensional approaches (e.g., multiphoton fluorescence microscopy) to quantify MFNP-positive cells may improve accuracy compared to the employed high power field based method.

Collectively, these findings establish that prototypical MNFPs can report on and image cellular inflammation in murine atherosclerosis. The study provides a foundation for the noninvasive assessment of cellular inflammation in atherosclerosis, particularly in concert with structural MRI of atherosclerosis [4,5]. Potential applications in atherosclerosis include assessing the effects of genetic and/or pharmacological manipulations of the inflammatory process in vivo. In addition, the validated MFNP uptake assays are expected to be useful in defining the cellular distribution and imaging capabilities of novel atherosclerosis-targeted MNFPs.

Footnotes

Acknowledgments

The authors acknowledge Nikolai Sergeyev, PhD, for the synthesis of the magnetofluorescent nanoparticle, Karen Mendelson, BS, for histological assistance, and Mikael Pittet, PhD, for analysis of the flow cytometry data. This work was supported by NIH Grants P50-CA86355 (RW), R24-CA92782 (RW), UO1-HL080731 (RW, FJ) and by the Donald W. Reynolds Foundation (FJ, RW).

Abbreviations:

References

Whitesides

(2003). The ‘right’ size in nanobiotechnology. Nat Biotechnol. 21:1161–1165.

LaVan

McGuire

Langer

(2003). Small-scale systems for in vivo drug delivery. Nat Biotechnol. 21:1184–1191.

Jaffer

Weissleder

(2004). Seeing within: Molecular imaging of the cardiovascular system. Circ Res. 94:433–445.

Choudhury

Fuster

Badimon

Fisher

Fayad

(2002). MRI and characterization of atherosclerotic plaque: Emerging applications and molecular imaging. Arterioscler Thromb Vasc Biol. 22:1065–1074.

Choudhury

Fuster

Fayad

(2004). Molecular, cellular and functional imaging of atherothrombosis. Nat Rev Drug Discov. 3:913–925.

Libby

(2002). Inflammation in atherosclerosis. Nature. 420: 868–874.

Schmitz

Coupland

Gust

Winterhalter

Wagner

Kresse

Semmler

Wolf

(2000). Superparamagnetic iron oxide-enhanced MRI of atherosclerotic plaques in Watanabe hereditable hyperlipidemic rabbits. Invest Radiol. 35:460–471.

Schmitz

Taupitz

Wagner

Coupland

Gust

Nikolova

Wolf

(2002). Iron-oxide-enhanced magnetic resonance imaging of atherosclerotic plaques: Postmortem analysis of accuracy, inter-observer agreement, and pitfalls. Invest Radiol. 37: 405–411.

Litovsky

Madjid

Zarrabi

Casscells

Willerson

Naghavi

(2003). Superparamagnetic iron oxide-based method for quantifying recruitment of monocytes to mouse atherosclerotic lesions in vivo: Enhancement by tissue necrosis factor-alpha, interleukin-1beta, and interferon-gamma. Circulation. 107: 1545–1549.

10.

Ruehm

Corot

Vogt

Kolb

Debatin

(2001). Magnetic resonance imaging of atherosclerotic plaque with ultrasmall superparamagnetic particles of iron oxide in hyperlipidemic rabbits. Circulation. 103:415–422.

11.

Schmitz

Taupitz

Wagner

Wolf

Beyersdorff

Hamm

(2001). Magnetic resonance imaging of atherosclerotic plaques using superparamagnetic iron oxide particles. J Magn Reson Imaging. 14:355–361.

12.

Kooi

Cappendijk

Cleutjens

Kessels

Kitslaar

Borgers

Frederik

Daemen

van Engelshoven

(2003). Accumulation of ultrasmall superparamagnetic particles of iron oxide in human atherosclerotic plaques can be detected by in vivo magnetic resonance imaging. Circulation. 107: 2453–2458.

13.

Trivedi

UK-I

J.M.

Graves

Cross

Horsley

Goddard

Skepper

Quartey

Warburton

Joubert

Wang

Kirkpatrick

Brown

Gillard

(2004). In vivo detection of macrophages in human carotid atheroma: Temporal dependence of ultrasmall superparamagnetic particles of iron oxide-enhanced MRI. Stroke. 35:1631–1635.

14.

Kircher

Mahmood

King

Weissleder

Josephson

(2003). A multimodal nanoparticle for preoperative magnetic resonance imaging and intraoperative optical brain tumor delineation. Cancer Res. 63:8122–8125.

15.

Denis

Mahmood

Benoist

Mathis

Weissleder

(2004). Imaging inflammation of the pancreatic islets in type 1 diabetes. Proc Natl Acad Sci USA. 101:12634–12639.

16.

Shen

Weissleder

Papisov

Bogdanov

Jr. Brady

(1993). Monocrystalline iron oxide nanocompounds (MION): Physicochemical properties. Magn Reson Med. 29:599–604.

17.

Jaffer

Weissleder

(2005). Molecular imaging in the clinical arena. JAMA. 293:855–862.

18.

Wunderbaldinger

Josephson

Weissleder

(2002). Crosslinked iron oxides (CLIO): A new platform for the development of targeted MR contrast agents. Acad Radiol. 9:S304–S306.

19.

Weissleder

Ntziachristos

(2003). Shedding light onto live molecular targets. Nat Med. 9:123–128.

20.

Christov

Dai

Drangova

Liu

Abela

Nash

McFadden

Lucas

(2000). Optical detection of triggered atherosclerotic plaque disruption by fluorescence emission analysis. Photochem Photobiol. 72:242–252.

21.

Allport

Lim

Shipley

Senior

Shapiro

Matsuyoshi

Vestweber

Luscinskas

(2002). Neutrophils from MMP-9- or neutrophil elastase-deficient mice show no defect in transendothelial migration under flow in vitro. J Leukoc Biol. 71: 821–828.

22.

Meir

Leitersdorf

(2004). Atherosclerosis in the apolipoprotein E-deficient mouse: A decade of progress. Arterioscler Thromb Vasc Biol. 24:1006–1014.

23.

Mahmood

Tung

Bogdanov

Weissleder

(1999). Near infrared optical imaging system to detect tumor protease activity. Radiology. 213:866–870.

24.

Paigen

Morrow

Holmes

Mitchell

Williams

(1987). Quantitative assessment of atherosclerotic lesions in mice. Atherosclerosis. 68:231–240.

25.

Rabkin

Aikawa

Stone

Fukumoto

Libby

Schoen

(2001). Activated interstitial myofibroblasts express catabolic enzymes and mediate matrix remodeling in myxomatous heart valves. Circulation. 104:2525–2532.

26.

Bowen

Zhang

Saab

Gareau

Rutt

(2002). Application of the static dephasing regime theory to superparamagnetic iron-oxide loaded cells. Magn Reson Med. 48:52–61.

27.

Wickline

Lanza

(2003). Nanotechnology for molecular imaging and targeted therapy. Circulation. 107:1092–1095.

28.

Pandit

Guccione

Bednarski

(2004). Molecular imaging applications in nanomedicine. Biomed Microdevices. 6: 113–116.

29.

Bulte

Kraitchman

(2004). Iron oxide MR contrast agents for molecular and cellular imaging. NMR Biomed. 17:484–499.

30.

McNamee

(1982). Restricted dextran transport in the sheep lung lymph preparation. J Appl Physiol. 52:585–590.

31.

Kang

Kim

Bruening

Pack

Charalambous

Pritsker

Moran

Loeffler

Steinman

Park

(2004). The C-type lectin SIGN-R1 mediates uptake of the capsular polysaccharide of Streptococcus pneumoniae in the marginal zone of mouse spleen. Proc Natl Acad Sci USA. 101:215–220.

32.

Nagaoka

Takahara

Tanaka

Yoshida

Steinman

Saitoh

Akashi-Takamura

Miyake

Kang

Park

Inaba

(2005). Association of SIGNR1 with TLR4-MD-2 enhances signal transduction by recognition of LPS in gram-negative bacteria. Int Immunol. 17:827–836.

33.

Weissleder

Elizondo

Wittenberg

Rabito

Bengele

Josephson

(1990). Ultrasmall superparamagnetic iron oxide: Characterization of a new class of contrast agents for MR imaging. Radiology. 175:489–493.

34.

Kelly

Allport

Tsourkas

Shinde-Patil

Josephson

Weissleder

(2005). Detection of vascular adhesion molecule-1 expression using a novel multimodal nanoparticle. Circ Res. 96:327–336.

35.

Kellar

Fujii

Gunther

Briley-Saebo

Bjornerod

Spiller

Koenig

(2002). Important considerations in the design of iron oxide nanoparticles as contrast agents for TI-weighted MRI and MRA. Acad Radiol. 9:S34–S37.

Cellular Imaging of Inflammation in Atherosclerosis Using Magnetofluorescent Nanomaterials

Abstract

Keywords

Introduction

Materials and Methods

Synthesis of the Near-Infrared Fluorescent Magnetofluorescent Nanoparticles

Quantification of MFNP Uptake by Murine Cells

Apolipoprotein-E-Deficient Model of Atherosclerosis

Magnetic Resonance Imaging

NIR Fluorescence Reflectance Imaging

Cellular Assessment of MFNP Deposition in Atheroma

Image Analysis

Statistical Analyses

Results

Activated Macrophages Preferentially Bind MFNPs in Vitro

MFNP Deposition in Murine Atheroma Can Be Detected by in Vivo MRI

MFNP Deposition in Atheroma Can Be Detected by Fluorescence Reflectance Imaging

MFNPs Colocalize with Immunofluorescent Cells in Apo E−/− Atheroma

Discussion

Footnotes

Acknowledgments

Abbreviations:

References