Introduction

Atherosclerosis is a progressive pathophysiological phenomenon related to various cardiovascular diseases. Atherosclerotic plaques can be stable and unstable. The latter are also called vulnerable plaques since they easily rupture. Acute coronary syndrome (ACS) is typically caused by coronary atherosclerosis [1], and, specifically, by evolution of atherosclerotic plaque [2]. The pathophysiological mechanisms of atherosclerotic plaque rupture include vasospasm, plaque rupture, platelet adhesion, aggregation, and thrombus formation [3–5]. Vulnerable plaques have large lipid core, thin fibrous cap, intraplaque hemorrhage, and neo-vascularization of the vasa vasorum [6–9]. It is important to timely identity vulnerable plaques. To this end, several new techniques in molecular imaging were developed. Molecular imaging possesses high sensitivity, specificity, and signal-to-noise ratio [10]. High-resolution magnetic resonance imaging (MRI) is a non-invasive diagnostic method useful for plaque imaging [11, 12]. Recently, ultra-small superparamagnetic iron oxide (USPIO) was introduced as an MRI contrast agent. USPIO can be up-taken by macrophages which then become part of vulnerable plaques, facilitating accumulation of the agent in the shoulders or the necrotic lipid core of ruptured atherosclerotic plaques [13–15]. Thus, the USPIO-enhanced MRI may be useful for detection of macrophages within atherosclerotic plaques [16].

In this study, we assessed the feasibility of using the USPIO-enhanced MRI to detect macrophages and vulnerable plagues in experimental atherosclerosis.

Materials and Methods

Animal Model

Thirty male New Zealand white rabbits weighing 2.5–3.0 kg were obtained from Animal Center of Xuzhou Medical College. In all experimental studies, the Principles of Laboratory Animal Care [17] were followed. The experimental protocol was approved by the Animal Care Committee of Xuzhou Medical University.

Rabbits were randomly divided into two groups, Group A and B. The animals in both groups (n = 15 per group) were fed a high cholesterol diet (1 % cholesterol, 0.2 % bile salt, 10 % lard, 88.8 % normal diet; 150 g/kg/day) for 8 weeks and then subjected to balloon-induced endothelial injury of the abdominal aorta. All invasive procedures were performed under general anesthesia with an intravenous injection of 3 % sodium pentobarbital (30 mg/kg) via the rabbit ear vein. The endotheliocyte injury was performed from the aortic arch to the iliac bifurcation with a coronal artery sacculus tube of 4.0 Fr that was inserted at a length of 20 cm through the right femoral artery. This process was repeated three times, then the catheter was removed and incision closed.

At the end of 16 weeks, all rabbits of Group B received an injection of recombinant adenovirus carrying human wild-type p53 gene into their abdominal aorta plaques. Two weeks later, rabbits of both study groups were exposed to 0.15 mg/kg of Chinese Russell’s viper venom as a pharmacological stimulus (Guangdong Institute of Snake venom; Guangdong, China). The venom was injected intraperitoneally followed by injection of 0.02 mg/kg histamine (Shanghai Dongfeng Biological Technology, Shanghai, China) in the ear vein 30 min later.

MRI

Rabbits were anesthetized as above and imaged supine using 1.5T MRI system (PHILIPS, Gyroscan Intera 1.5T), synergy knee coil. The aorta USPIO-enhanced MRI imaging was conducted on all experimental rabbits at 0, 24, 48, 72, and 96 h after the medication (Chinese Russell’s viper venom and histamine). Sequential transverse images of the abdominal aorta were made from the renal arteries to the iliac bifurcation. The MR parameters were as follows. Image analysis: FSE (fast spin-echo) T1-weighted (repetition time TR = 470 ms, echo time TE = 5 ms, flip angle FA = 80°, slice thickness = 5 mm with no interval gap, matrix: 256 × 256), STIR (short time inversion recovery) T1-weighted (repetition time TR = 300 ms, echo time TE = 15 ms, slice thickness = 5 mm with no interval gap, matrix: 256 × 256). FSE T2-weighted (repetition time TR = 3000 ms, echo time TE = 85 ms, slice thickness = 5 mm with no interval gap, matrix: 256 × 256). FSE PDWI (repetition ime TR = 2500 ms, echo time TE = 11.5 ms, slice thickness = 5 mm with no interval gap, matrix: 256 × 256). FFE (fast field echo) T2-weighted (repetition time TR = 600 ms, echo time TE = 13.8 ms, FA = 30°, slice thickness = 5 mm with no interval gap, matrix: 192 × 256).

MRI Image Analysis

MR images were read independently by two cardiovascular specialists in separate reading sessions. Plaques were determined in different sequences as an increased signal intensity on T1-weighted images, and as a decreased signal intensity on T2-weighted and T2*-weighted images. The reading results from two specialists were compared. To decrease a bias in determining the plagues, two specialists reviewed the readings together during the third reading session. In addition, we calculated the number of lesions. A signal-to-noise ratio (SNR) was calculated as follows: SNR = SI/SD.

Biochemical Studies

Blood samples were collected for measurements of plasma total cholesterols (TC), triglycerides (TG), low density lipoprotein-cholesterol (LDL-C), high density lipoprotein-cholesterol (HDL-C) at 0 and 18 weeks. In parallel, serum samples were obtained by centrifugation (10 min, 3000 rpm), and serum was used for measurements of ICAM-1 and VCAM-1 by respective ELISAs (Shanghai Huyu Biotechnology, Shanghai, China).

Histopathology and Immunohistochemistry

Abdominal aorta was excised and cut into two segments. One segment was placed into cold phosphate-buffered saline, trimmed of adherent connective tissue, and fixed in 4 % formaldehyde for 24 h. After that, first segment was embedded in paraffin and cut into 5-µm sections for staining with hematoxylin/eosin, or Perls and Masson stains. The second segment was fixed in 3 % glutaraldehyde for at least 2 h for subsequent electron microscopy.

Statistical Analysis

All statistical analyses were carried out using SPSS16.0 (SPSS, Chicago, USA). Quantitative variables are expressed as mean ± SD. Statistical comparisons were made using the t and one-way ANOVA tests. The data were considered statistically different at p < 0.05.

Results

Animal Model

Animal model was successfully established in 26 out 30 animals (14 animals in Group A and 12 animals in Group B). One animal in Group A died of anesthetic accident (respiratory distress due to an overdose of an anesthetic), while three animals in Group B died of medication overdose. Vulnerable plaques were observed in 12 animals (two animals of Group A and ten animals of Group B). Thirty-five atherosclerotic plaques and six vulnerable plaques were identified by pre-contrast and contrast-enhanced MRI after pharmacological triggering in Group A, while 29 vulnerable plaques were identified in Group B. Among the latter, 22 vulnerable plaques lesions occurred where wild-type p53 gene was adenovirally transduced. Plaque rupture and thrombosis occurred in nine lesions.

MR Imaging of Vulnerable Plaques

The formations of plaques were visible in both Group A and B. The stenosis of the abdominal aortic lumen was more prominent in Group B. Thus, the image of plaques in T1WI, PDWI, and T2*WI shows a slightly higher signal and an equal signal in T2WI (Fig. 1).

Fig. 1
figure 1

Series MRI

Full size image

High resolution imaging using T1, T2 and proton density weighting allows for identification of the following substances: calcium (hypointense on all imaging sequences), lipids (T1- and PD-hyperintense, and T2-hypointense), and fibrous tissue (produces increased signal intensity on PD-weighted imaging, while isointense on T1 and isointense-to-hyperintense on T2-weighted imaging). These multi-contrast signal intensity “signatures” allow for characterization of various plaque components and plaque morphology on high-resolution imaging and, thereby, to assess plaque vulnerability (Fig. 2). The central signal of the plaques was significantly reduced after USPIO injection, especially in T2*WI (Fig. 3). The peak of negative reinforcement was reached 96 h after the injection (Table 1).

Fig. 2
figure 2

MRI of stable plaque a and vulnerable plaque b

Full size image
Fig. 3
figure 3

MRI of vulnerable plaques at 0 (a) and 96 h (b) after injection of USPIO

Full size image
Table 1 Kinetics of signal-to-noise ratio after injection of USPIO
Full size table

Biochemical Results

As expected, high cholesterol diet led to lipid disorders. Compared with levels before the study, blood lipids were significantly increased in both study groups at 18 weeks (Table 2). The levels of ICAM-1 and VCAM-1 were significantly higher in Group B (p < 0.05; Table 3).

Table 2 Blood lipid levels
Full size table
Table 3 ICAM-1 and VCAM-1 levels
Full size table

Histopathological Analysis of Abdominal Aorta

We found an even wall thickening in Group A (Fig. 4a) while there was an obvious thickening with a substantial foam cell formation and large lipid core plaque hemorrhage in Group B (Fig. 4b). The Masson staining demonstrated collagen fiber hyperplasia in the vessel wall. A marked collagen fiber proliferation was seen in Group B (Fig. 4c). Further, Prussian blue staining in this group indicated iron deposition inside macrophages (Fig. 4d).

Fig. 4
figure 4

Hematoxylin & eosin staining shows thickening of the wall in Group A (a), and obvious thickening with a significant foam cell formation and large lipid core plaque hemorrhage in Group B (b; both ×40). Masson staining: marked collagen fiber proliferation seen in Group B (c; ×100). Prussian blue staining in Group B indicates iron deposition inside macrophages (d; ×100)

Full size image

Discussion

Identification of atherosclerotic vulnerable plaques is in the focus of intensive research. In the current study, we tested a suitability of the USPIO-enhanced MRI to detect vulnerable plaques in an experimental model of atherosclerosis.

Most of the models for atherosclerotic vulnerable plaques are not reliable. High cholesterol diet only leads to establishment of stable plaques. To circumvent this, Russell viper venom and histamine were used to trigger vulnerable plaques [18]. Further, Cheng et al. [19] established an animal model of vulnerable plaques by overexpressing p53 gene in experimental plaque. p53 controls abnormal proliferation of smooth muscle cells and promotes apoptosis of vascular smooth muscle cells [20]. In the present study, we utilized both pharmacological induction and overexpression of p53 in local plaques to facilitate establishment of vulnerable plaques.

Of all non-invasive imaging techniques, MRI is shown to be superior in artery lumen imaging and detailed information about the artery wall. In addition, imaging techniques can quantify lumen plaque burden and plaque constituents [21]. We demonstrate here that USPIO-enhanced MRI is sensitive to detect inflamed plaques. Using MRI enhanced with Sinerem (type of USPIO), Trivedi et al. [22] reported that high-risk individuals with inflamed plaques may be identified on the basis of a “focal” area of signal loss visualized on MRI after Sinerem infusion. Vulnerable plaques contain large numbers of inflammatory cells, including macrophages. Therefore, number of macrophages can be an important marker of vulnerable plaques. In a recent study [23], it was confirmed that combination of SPIO-labeled endothelial cells and MRI can be used to detect atherosclerotic plaques. In particular, this approach may be more suitable for instable plaques and early plaques due to the presence of active macrophages in lesions [23]. Further, it was demonstrated that a susceptibility gradient mapping-MRI allows a sensitive detection and quantification of intraplaque iron oxide particles after injection of very small iron oxide particles [24].

USPIO is a new type of a magnetic MRI contrast agent which can be identified by reticuloendothelial system. Moreover, because of long plasma half-life of USPIO, it can accumulate in macrophages of reticuloendothelial system. When USPIO is absorbed by macrophages, superparamagnetic effect causes inhomogeneity of the local magnetic field. Then, different components of a plaque (fat, calcification, fibrous tissue and thrombosis) can be identified due to their specific unique characteristic MRI signals [25].

In our study, we found loss of signals in rabbits that had T2*-weighted spiral sequence with USPIO-enhanced MRI. This was mainly due to formation of fibrous plaque in a lipid core stage. Its center is the main component of cholesterol-containing solid crystalline or liquid crystal cholesterol ester. The signal of the cholesterol crystalline solid and liquid crystal cholesterol ester lipid was significantly reduced compared with normal lipid with T2*-weighted sequence. We found more plaque sections with signal decline. These changes were significant, and our results were comparable with those reported by Kooi et al. [26]. In a different study, it was reported that areas of signal intensity reduction that correspond to USPIO/macrophage-positive histological sections demonstrate a quantitative MR signal reduction [27]. Intravenously injected USPIO are up-taken by phagocytosis by plaque macrophages and are mainly deposited under the intima of foam cells, as was demonstrated by pathology [28]. Atherosclerosis is associated with foam cell fusion and necrotic material collapse. The shedding of endometrium and the rupture of a fibrous cap can be seen as deposition of blue-stained particles of iron after the USPIO injection. This contrast agent is mainly gathered in the macrophages that engulf a large amount of lipid deposited in the intima, causing changes in MRI signal. It was also shown that macrophages play an important role in the plaque lipid core formation, expansion, and a plaque rupture [29].

The degree of macrophage aggregation is an important indicator to determine stability and activity of atherosclerotic plaque. For up to 48 h after infusion, the imaging can identify macrophages in vivo. We further confirmed that USPIO have a long half-life in blood (30 h) that allows phagocytosis of the contrast agent by cells of the monocyte–macrophage system within the atheromatous plaque [27]. The T2*WI sequence of USPIO-enhanced plaque that shows a more sensitive negative reinforcement helps to estimate the number of macrophages in the plaque and gauge inflammation.

In conclusion, our results suggest that the USPIO-enhanced MRI combined with histopathological analysis is useful in detection of inflammatory tissues. The USPIO-enhanced MRI can identify vulnerable plaques by detecting USPIO in macrophages and show neovascularization within the plaque in the abdominal aorta. Further, this technique is superior to conventional contrast agents in the identification of different components within the plaque. In addition, it can also estimate inflammatory infiltration in plaque. Therefore, it can be used as a new technology in detection of vulnerable plaques [16].