Interventional real-time molecular MRI for targeting early myocardial injury in a pig model

Fig. 1. Ischemia and reperfusion injury. A Coronary angiography of the left coronary artery. Left Overview of the left anterior descending (LAD) and left circumflex artery (LCX). Middle A standard coronary wire (red arrow) and obstructive coronary balloon (black arrow) are introduced into the LCX. Then the balloon is inflated to block blood the supply to the distal myocardium. Right Red box indicates the blocked LCX artery during contrast deposition. After 40 min of ischemia the balloon is removed for reperfusion. B ECG showing inferior leads II and III with ST-segment elevations during ischemia. C T1 mapping 2–4 h after reperfusion indicates a significantly elevated T1 time in the “Lesion” area of the LCX as compared to “Remote” (LAD/RCA area); n = 72 values per group, p < 0.001. Note, that in each of the 7 animals 3–4 short axis views were acquired with multiple repetitions resulting in 72 measurements. The arrow indicated the area of the lesion. D T2 mapping 2–4 h after reperfusion shows no difference between “Lesion” and “Remote”; n = 72 values per group, p = 0.11. The arrow indicated the area of the lesion. E After 40 min of ischemia and 4 h of reperfusion, no LGE can be detected in…

Fig. 2. Post-ischemic inflammation. A Flow cytometry characterization of innate immune cells in the blood after I/R. Above FACS plots indicate the gating scheme for neutrophils and monocyte subsets. Red boxes indicate final cell populations, while black boxes were used for further subtype characterization. Below Timeline analysis of innate immune cells from baseline to 4 h after onset of ischemia for neutrophils, classical-typed CD14+CD163neg monocytes and nonclassical-typed CD14negC163+ monocytes. *p < 0.05 for the comparison of the time point at 3 h versus baseline; n = 3 per group, Welch’s unpaired t test. B Immunofluorescence staining of cardiac tissue from “Lesion” and “Remote” targeting P-selectin expression (red) and nuclei (dapi-blue). After injury, P-selectin expression is significantly increased in the “Lesion” area compared to “Remote”; n = 6 per group, p = 0.02. Yellow arrows highlight examples of positive P-selectin or DAPI staining. *p < 0.05; Mann-Whitney U test.

Fig. 3. Targeting P-selectin in vitro. Binding of functionalized MPIO P-selectin (Psel-MPIO) to platelets and endothelial cells was evaluated using an incubation assay for endothelial cells (n = 60 per group, above) or an in vitro flow chamber model for platelets (n = 6 for P-selectin MPIO versus n = 3 control MPIO, below) in vitro. For comparison we used functionalized MPIO targeting unspecific IgG (control MPIO). Both essays confirmed enhanced binding of P-selectin MPIO to platelets (p < 0.0001) or endothelial cells (p < 0.0001) in comparison to control MPIO. Yellow arrows indicate MPIO. *p < 0.05; Mann–Whitney U test.

Fig. 4. Targeting P-selectin in the ischemic lesion in vivo. A Interventional MRI setup for selective contrast injection into the left coronary artery. B Perfusion map of the porcine heart. Cine image (corner) shows a midventricular short axis slice of the heart. Gadolinium injection into the left coronary artery (LCA) depicts the supply area of the left circumflex (LCX) and left anterior descending (LAD) arteries as well as the non-perfused area of the right coronary artery (RCA). C, D T2* maps of the left ventricular short axis after injection of P-selectin MPIO or control MPIO (above: P-selectin MPIO; below: control MPIO). E, F Injection of P-selectin MPIO induced a significant increase of R2* in the LCX (lesion) compared to RCA (p = 0.03), whereas no difference was observed when comparing the equally perfused but non-injured LAD area with RCA area. The control MPIO did not lead to a significant increase in R2* in the LCX and LAD compared to the RCA areas.

Fig. 5. Ex vivo analysis of heart specimens. A Ex vivo, high resolution inverse T2* (R2*) map of the left ventricle after I/R of the LCX and injection of P-selectin MPIO perfusion via the left coronary artery. B P-selection MPIO selectively enriched in the area of the LCX as compared to control MPIO, n = 3 each, p = 0.03; standard t test, as seen in an increased count ratio which compares the R2* values in ischemic and normal myocardial tissue (see “Post-processing”). C We randomly assessed tissue sections taken from specimens resected from the LCX, LAD or RCA area. P-selectin MPIO was found most frequent in the area of the lesion compared to perfused, non-injured LAD area or non-perfused, non-injured RCA area, n = 540 HPF (high power fields) per totel area. Only minimal MPIO moieties were found in the tissue after injection of control-MPIO with unspecific binding, n = 518 HPF (high power fields) total per area, *p < 0.05 one-way ANOVA.