As an early event, endothelial dysfunction is involved in the progression of atherosclerosis (AS). Gadolinium based blood pool agents were successfully applied in the past for fibrous plaque detection and to gain insight into the leakiness of the endothelial layer via the altered vascular permeability and albumin as vehicle into the lesion.^1,2

In this work we set out to measure the endothelial barrier function of atherosclerotic rabbits using B-22956/1 an albumin binding agent in combination with NO donor S-nitroso human serum albumin (S-NO-HSA) and DCE-MRI. Furthermore, we tested whether the long-term S-NO-HSA-B22956/1 infusion led to a NO induced vasodilation effect.

B22956/1: Gadocoletic acid trisodium salt (Bracco Imaging SpA) is an intravascular high affinity serum albumin binding blood pool agent already applied to image neovessel- and macrophage- density in atherosclerotic rabbits.³

S-nitroso human serum albumin: A nitric oxide donor used for NO supplementation shown to prevent endothelial NO synthase (eNOS) uncoupling in rabbits.⁴

Animal model: A total of three male New Zealand White (NZW) rabbits (age:10-11months; body weight:3,5-4,5kg) were used repeatedly in this study. Abdominal aortic atherosclerosis was induced in two rabbits by a combination of balloon catheter denudation and a two month high cholesterol diet. The injury was performed from the iliac bifurcation to the renal artery bifurcation introducing a 4F wedge pressure balloon catheter through the carotid artery. One non-treated rabbit fed a normal chow diet was used as control.

S-NO-HSA(-B22956/1) infusion strategy: Either 18,5mL of B22956/1(2mM) pre-incubated with S-NO-HSA(1mM) or B22956/1(2mM) diluted in sodium chloride were administered intravenously during DCE-MRI. A slow infusion rate of 20mL/h was chosen for all experiments to prevent uncontrolled NO mediated hypotension. To assure contrast agent wash out from the vessel wall, experiments were separated by a whole week. For the injured rabbits the experiments were repeated two-to-three times ending up in a total of five-to-six imaging sessions.

Imaging: A dark blood double inversion recovery (DIR) T₁-weighted turbo spin echo (T_1w-TSE) sequence was used on one selected axial slice for DCE measurements. Scan parameter were TR/TE:600/12ms; NEX/ETL:4/7; matrix:256x176; FOV:120x82,5mm; resolution:0.46x0.46x3mm. To avoid chemical shift artifacts caused by peri-adventitial fat, a spectral fat saturation pulse was applied. A total of 60 images were acquired with a temporal resolution of 1min1sec. Before long-term infusion a total of five precontrast images were taken for baseline signal averaging. All experiments were performed on a human 3Tesla platform (Siemens Magnetom Prisma) using a 15Ch TxRx knee coil.

Data analysis: Inner and outer vessel wall boundaries of DCE images were manually segmented using ITK-SNAP. To uncover possible NO mediated wall relaxation effects boundaries were used to calculate lumen areas over time. Average vessel wall signal intensities (SI) were calculated for each DCE time point. In addition, the normalized wall index (NWI=wall area/total vessel area) was calculated as a measure of wall thickening. The relative signal enhancement c(t) was calculated by subtracting the signal intensity SI(t) at a given time point from the baseline tissue signal intensity SI(0) according to equation⁵: c(t)=(SI(0)-SI(t))/SI(0).

Figure 1 shows the relative vessel wall signal enhancement c(t) during the long-term infusion of S-NO-HSA-B22956/1 (green line) and B22956/1 (blue line), respectively. The relative change in lumen area is outlined in Figure 2. A representative T_1w image together with the NWI is illustrated in Figure 3. Given that a delayed wall enhancement (decreased vascular permeability) and vasodilation of the NO donor carrying compound can be observed, a regulation of the endothelial barrier function as well as an NO induced endothelium-independent vasomotor response of the control (CTRL) rabbits’ vessel is assumable. For the balloon treated NZW I (BT-NZW I) no differences in wall enhancement but an NO triggered relaxation of the artery can be observed, indicating that a vascular damage with endothelial dysfunction but existing vasorelaxation of smooth muscle cells is assumable. BT-NZW II with a prominent thickening of the wall (NWI=0.56) shows a reconstituted delayed endothelial function and contraction of the artery, assuming advanced plaque burden with possible endothelial regrowth and paradoxical NO induced vasoconstriction. In this work, we demonstrate for the first time the feasibility to investigate vascular permeability and NO induced endothelium-independent vasomotor response of the abdominal aorta in control and balloon denuded rabbits simultaneously. The role of NO in controlling the interendothelial junctions and therefore to regulate the endothelial barrier is discussed quite controversial in literature but our data assume specific patterns depending on the degree of plaque burden. Future work will investigate the influence of plaque progression on endothelial barrier regulation and vasomotion which may help to predict the formation of atherosclerotic plaques.