Magnetic resonance molecular imaging of thrombosis in an arachidonic acid mouse model using an activated platelet targeted probe

Abstract

Objective: Atherosclerotic plaque rupture leads to acute thrombus formation and may trigger serious clinical events such as myocardial infarction or stroke. Therefore, it would be valuable to identify atherothrombosis and vulnerable plaques before the onset of such clinical events. We sought to determine whether the noninvasive in vivo visualization of activated platelets was effective when using a target-specific MRI contrast agent to identify thrombi, hallmarks of vulnerable or high-risk atherosclerotic plaques.

Methods and results: Inflammatory thrombi were induced in mice via topical application of arachidonic acid on the carotid. Thrombus formation was imaged with intravital fluorescence microscopy and molecular MRI. To accomplish the latter, a paramagnetic contrast agent (P975) that targets the glycoprotein alpha(IIb)beta(3), expressed on activated platelets, was investigated. The specificity of P975 for activated platelets was studied in vitro. In vivo, high spatial-resolution MRI was performed at baseline and longitudinally over 2 hours after injecting P975 or a nonspecific agent. The contralateral carotid, a sham surgery group, and a competitive inhibition experiment served as controls. P975 showed a good affinity for activated platelets, with an IC(50) (concentration of dose that produces 50% inhibition) value of 2.6 micromol/L. In thrombosed animals, P975 produced an immediate and sustained increase in MRI signal, whereas none of the control groups revealed any significant increase in MRI signal 2 hours after injection. More important, the competitive inhibition experiment with an alpha(IIb)beta(3) antagonist suppressed the MRI signal enhancement, which is indicative for the specificity of P975 for the activated platelets.

Conclusions: P975 allowed in vivo target-specific noninvasive MRI of activated platelets.

FIGURE 1

Quantification of FITC-Fibrinogen specific binding to TRAP-simulated platelets as a function of increasing concentration of Tirofiban (positive control), P977 (α_IIbβ₃ targeted peptide) and P975 (α_IIbβ₃ targeted peptide attached to a Gd-DOTA moeity)

FIGURE 2

A-B. Fluorescent intravital microscopy of an arachidonic acid (AA) induced model of thrombosis at two different time points. Platelets were isolated, labeled with an intracellular fluorophore (Calcein AM) and injected in a receiving mouse that underwent AA thrombosis. The vessel was placed under a fluorescent microscope and platelet aggregation was monitored. The platelets started adhering onto the vessel wall (A) and as the process of thrombosis continued, they aggregated towards the inside forming an intra-luminal thrombus (B). C. Histological section of the intra-arterial thrombus induced by periadventitial delivery of AA stained with H&E.

FIGURE 3

In vivo MRI after AA-induced arterial thrombosis. The Trachea (white arrow), the non-injured left carotid (control, green arrow) and the injured right carotid (red arrow), are shown. Inserts in the upper left angle of each image represent magnification images of the injured carotid section. The panel A displays the behavior of P975 over 2 hours after its injection in thrombus-induced animals versus Gadolinium shown below in panel B. Before contrast agent injection, cross sections depict an occluded right carotid artery while the left carotid is visible on a T1-weighted black blood sequence. After injection of P975, signal intensity in the lumen of the thrombosed carotid increased at 30 minutes post-injection and persisted at 120 minutes. Conversely, Gd-DOTA only showed signal enhancement 30 minutes after injection and rapidly cleared from the circulation. Some non-specific perivascular enhancement was observed in both cases at early time points. No signal enhancement was observed in the contralateral carotid (negative control).

FIGURE 4

MRI Cross sections of the non-injured (green arrow) and the injured (red arrow) carotid arteries. The trachea is shown (white arrows). Inserts in the upper left angle of each image represent magnification images of the injured carotid section. (A) Platelet inhibitor Eptifibatide was injected at a saturating dose after thrombus induction and before administration of P975. Although the presence of intra-luminal thrombosis was confirmed by histology using a modified Masson’s Trichrome staining, the results showed no signal enhancement at 2 hours when compared to baseline in the injured carotid. (B) No MRI signal enhancement or thrombosis was visible in the injured carotid of animals who received the sham surgery i.e. delivery of EtOH and slight reduction of carotid diameter. Absence of intraluminal thrombosis was confirmed by histology in these animals.

FIGURE 5

Quantification of signal increases in the carotid artery lumens. (A) ΔCNRs were calculated at each imaging time point and plotted (0 represents the baseline scan). After Gd-DOTA injection, an initial signal increase occurred at 30 minutes post-injection (red line) possibly due to contrast agent molecules being transiently trapped within the mesh of the thrombus. Eventually, the values returned close to baseline over time. In contrast, P975 injection caused a signal increase that was still significantly higher compared to Gd-DOTA at 120 minutes post-injection (dark blue line). Animals injected with Eptifibatide prior to P975 (purple line) exhibited a similar behavior compared to animals injected with Gd-DOTA. The control groups (turquoise blue and green lines) did not show any enhancement. (B) At 2 hours post injection P975 exhibited a significant increase in ΔCNRs compared to the rest of the groups.