Figure 1.
Born-normalized molecular optical projection tomography of inflamed hearts.
(A) Dual channel imaging. A cocktail containing two different molecular imaging agents is administered intravenously. (B) Dual channel Born-normalized near-infrared transillumination fluorescence image on day 5 after myocardial infarction (MI), 24 hours after probe injection. High signal in both the 680 nm (left) and 750 nm (right) fluorescence channels can be observed in the heart region, indicating elevated protease and phagocytic activity. (C) A narrow bandwidth light source excites the fluorophores attached to a molecular probe located within the sample, and fluorescent light is then emitted. Both excitation and emission light are, in part, absorbed by the sample on the way to and from the fluorophore, respectively. Using a Born-normalized approach, fluorescence tomographic reconstructions can be obtained, after correcting for the sample's absorption map. (D) Image processing algorithms are applied to tomographically obtain signal distribution reconstructions in whole hearts and to remove noise and artifacts contributions.
Figure 2.
Fluorescence Reflectance Imaging.
White light and ex-vivo fluorescence reflectance imaging (FRI) in the VT-680 and VT-750 fluorescence channels evidence Prosense-680 and Clio-750 activities on the hearts' surface at day 5 post-MI. Probes were injected 24 h prior to heart explantation. (A) Myocardial infarct (permanent occlusion of coronary artery), probes injected. FRI images reveal strong localized near infrared fluorescence (NIRF) signal. (B) Ischemia reperfusion injury, probes injected. FRI images indicate strong (NIRF) signal homogeneously distributed in the whole anterior left ventricle. (C) Myocardial infarct, no probes injected. Extremely weak autofluorescence signal is present in the FRI images. (D) Heart without myocardial infarction (sham surgery), probes injected. Weak NIRF signal homogenously distributed in the whole heart is present. For each fluorescence channel all FRI images are acquired using the same parameters and identically scaled. In each panel the lower row shows the corresponding Born normalized projection for the cleared sample.
Figure 3.
Coronal (top), saggital (middle), and axial (bottom) tomographic reconstructions of enzyme (A) and phagocytic cell (B) distributions. After permanent occlusion of the coronary artery, both probes are primarily confined to the infarct scar area. This area is associated with a strong localized near-infrared fluorescence (NIRF) signal, whereas only weak fluorescence intensity is observed in the basal part of the ventricle.
Figure 4.
Validation of tomographic reconstructions.
Comparison between OPT reconstructions and histological sections of molecular probe activity distribution (Prosense-680) in the inflamed heart after MI. Comparisons between reconstructions obtained with fluorescence OPT (A), Bornnormalized OPT (B), and the corresponding histological section (C). Born normalization preserves the molecular distributions in the reconstructed fluorescence channels. This is particularly evident for the papillary muscles located deep within the left ventricle, which appear less fluorescent without normalization, and the epicardium, which always shows bright fluorescent signal in absence of normalization. Born-normalized OPT reconstructions were obtained on the whole heart. The histological section (500 µm) belongs to the same specimen.
Figure 5.
Born-normalized fluorescence OPT reconstructions (A) of the two molecular imaging agents Prosense-680 and CLIO-750 are compared with immune-reactive staining of the imaging target protease Cathepsin B (B) and for the Mac3 macrophage antigen (C).
Figure 6.
Heart modeling and segmentation.
A. Ellipsoid heart modeling for left ventricle segmentation and American Heart Association (AHA)-conform segmentation. Bull's Eye (B) and cylindrical maps (C) are generated from the radial projection of the probes' tomographic reconstructions, over the surfaces of the concentric fitting ellipsoids. The correlation between the 17-sector Bull's Eye map and analogs on the unfolded cylindrical map is given. (D) Myocardial infarction with its corresponding occlusion and visible ligation thread, in the left anterior descending coronary artery (LAD). (E) Tomographic reconstruction of a non-injured heart. The reconstructed LAD (shown in red), courses to the apex of the heart and perfuses the basal anterior, the basal lateral and anterolateral, as well as the apical lateral walls. (F) A segmented left ventricle ellipsoid representation where different colors represent the wall thickness. Map projection (over a flat surface) of the ventricle wall thickness, according to both a Bull's Eye (G) and a cylindrical (H) map representation. Papillary muscles (black arrows) can be clearly identified by their thicknesses. The LAD is mapped and represented in black, while the black grid indicates the location of the 17 sectors. The red spot indicates the location of the ligation.
Figure 7.
Probe activity distribution maps in inflamed left ventricle.
(A) Bull's Eye (top) and cylindrical (right) maps of molecular probe activity at day 5 post myocardial infarction. Inflammation is accentuated along the border of the injured tissue, while the core of the infarct displays low signal. Continuous white line indicates the border of the interventricular septum. (B) One-dimensional plots showing the intensities of the probes activity, along the directions indicated in the maps (white faded bands). Signal plots indicate a high correlation in probe activities. (C) Axial reconstructions of cell-associated probe activity along the indicated plane, with their corresponding co-localization maps. (D) Overlapping cross correlation functions (CCF) are calculated for the two channels, taking into account both translational and rotational shifts. Overall, a high degree of correlation is present, however there are distinct differences in each channel reflecting the different spatial distribution of the molecular target.
Figure 8.
Quantification of probe activity in infarct area.
(A) Cylindrical (left) and Bull's Eye (right) maps of molecular probe and bead distribution of an excised heart at day 5 after myocardial infarction. High cathepsin B and phagocytic activity, are evident in the proximity of the injured area with a high degree of colocalization. Fluorescently labeled (FITC) microbeads, injected in vivo into the left ventricle before harvesting, are homogeneously distributed across the whole heart, with the exception of the non-perfused infarct area. The white dotted line indicates the border of the necrotic infarct scar, localized in apical and lateral segments. Continuous white line indicates the border of the interventricular septum. (B) Axial reconstructions of probe activities at the indicated plane. (C) Cylindrical (left) and Bull's Eye (right) representation of reparative (Ly6Clo, green) and inflammatory (Ly6Chi, red) monocyte subsets together with the an axial (D) and tomographic (E) reconstruction.