Magnetic Microbubbles Targeted To VCAM-1 For Ultrasound Molecular Imaging Of Early Stage Atherosclerosis Inflammation

ObjectiveTraditionally, the diagnosis of atherosclerosis is possible only at advanced stages of disease, either by direct detection of arterial luminal narrowing or by evaluating the effect of arterial stenosis on organ perfusion or function. Molecular imaging with contrast-enhanced ultrasound (CEU) and targeted microbubbles offers the possibility of real-time, noninvasive visualization of molecular markers of cardiovascular disease using clinical ultrasound systems. Pre-clinical studies have demonstrated great potential for detection of microvascular inflammation, such as occurs in ischemia-reperfusion and cardiac transplant rejection. While CEU molecular imaging of early inflammatory changes of atherosclerosis has been demonstrated, there remain issues over the technical difficulties of targeting microbubbles in the setting of high shear stress in larger arterial vessels.Ultrasound contrast microbubbles are traditionally used as blood flow tracers that exhibit rheological behavior similar to erythrocytes in vivo, and thus tend to remain close to the axial center of blood vessels. This behavior may disadvantage targeted microbubble agents used for molecular imaging of atherosclerosis in larger vessels, where early inflammatory changes occur within the vascular endothelium. Additionally, hemodynamic factors are very different in arteries as compared to the microvasculature, where endothelial shear stress is higher. In this setting, high shear stress forces may limit the rapid formation of adhesive bonds and hamper microbubble targeting, leading to a high rate of dislodgement of adhered microbubbles and a loss of targeting. To overcome this limitation, we have developed a microbubble agent that can be manipulated by a magnetic field (MF) to alter the axial distribution of microbubbles, increasing the number of circulating microbubbles that contact with the vascular endothelium, and potentially resulting in higher microbubble attachment.The purpose of our study was to evaluate the finding of microbubble targeted to vascular cell adhesion molecule-1 (VCAM-1) coupled with a magnetic-guidance system could improve the efficacy of CEU molecular imaging of atherosclerosis in the aorta.Methods1. Binding capability of microbubbles targeted to VCAM-1 under pulsatile high-shear flow conditions1.1. Microbubbles preparation and determine their biological properties.Biotinylated, lipid-shelled microbubbles were prepared by sonication of dipalmitoyl phosphatidylcholine, poly (ethylene glycol) 40-stearate and biotin-poly (ethylene glycol) 2000-distearoylphosphatidylethanolamine with sonicator 3000 at maximum power in an atmosphere of perfluoropropane.1×108 biotinylated microbubbles were conjugated to either rat anti-mouse VCAM-1 monoclonal antibody (MBv) or isotype control (MBi) via a streptavidin bridge. The size distributions and concentrations of MBv and MBi were used coulter counter to counte.1.2. Assessment of targeted microbubbles targeted to VCAM-1 with Parallel plate flow chamber.The binding and retention of targeted microbubblesb (MBv) to VCAM-1Fc immobilized on a culture dish were assessed in a flow chamber at variable shear stress (0.5-16.0 dyn/cm2). The pulsatile flow conditions were generated and compared to the continuous flow conditions. The retentive ability of MBv was evaluated by the detachment test.2. Ultrasound molecular imaging of early stage artherosclerosis with microbubbles targeted to VCAM-12.1. Animals Preparation and DietWe used 4 different mouse models:apolipoprotein E-deficient (APOE-/-) mice on a hypercholesterolemic diet, APOE-/-mice on a regular diet, wild-type mice (C57BL/6) on a hypercholesterolemic diet, and wild-type mice on a regular diet which served as normal controls. Animals were anaesthetized and a jugular vein was cannulated for intravenous administration of microbubbles.2.2. Contrast Enhanced Ultrasound (CEU) Molecular ImagingUltrasound imaging was performed with a high-frequency linear-array probe (17L5) held in place. The abdominal aorta was imaged with fundamental imaging at 15 MHz to optimize the imaging plane in the longitudinal axis and the aortic diameter was measured. The peak flow velocity and pulse rates at the aorta were measured by pulsed-wave Doppler with a gate size set at the minimum setting. CEU was performed with Contrast Pulse Sequencing, which detects the nonlinear fundamental signal component from microbubbles. Imaging was performed at a centerline frequency of 7 MHz and a mechanical index (MI) of 0.18. Real-time imaging at 0.18 MI was performed after intravenous bolus injection of 1×106 magnetic microbubble, After continuous imaging for 8 minutes. The mechanical index was transiently increased to 1.0 for 3 s, to destroy adhered microbubbles, and subsequent post-destruction images were acquired at 0.18 MI, to obtain background images. To determine signal from retained microbubbles alone,3 post-destruction contrast frames representing any freely circulating microbubbles were averaged and digitally subtracted from 3 averaged pre-destruction frames use the Yabko MCE2.7 software (University of Virginia, USA) and then color-coded. Background-subtracted signal intensity was measured from a region of interest placed over the abdominal aorta.2.3. ImmunohistochemistryImmunostaining for VCAM-1 was performed on frozen sections of the abdominal aorta after drying for 2 hours and fixing with paraformaldehyde for 15 minutes at room temperature. Rat anti-mouse VCAM-1 monoclonal antibody was used as a primary antibody with a secondary anti-rat antibody. Staining was performed with HRP substrate solution. Slides were counterstained with hematoxylin. Slides were visualized under microscope and photographed with a CCD camera.3. Preparation of magnetic microbubbles targeted to VCAM-1 and in-vitro assessment.3.1. Microbubble PreparationBiotinylated, lipid-shelled microbubbles were as previously described. MBb were conjugated to either rat anti-mouse VCAM-1 monoclonal antibody (MBvM) or isotype control antibody (MBiM) via a magnetic streptavidin bridge. For a positive control microbubble, microbubbles were conjugated to a rat anti-mouse VCAM-1 monoclonal antibody via regular non-magnetic streptavidin bridge (MBv). For in vitro flow-chamber studies, microbubbles were fluorescently labeled by the addition of dioctadecyltetramethylindocarbocyanine (DiI) perchlorate to the aqueous suspension prior to sonication. Microbubble size, concentration and distribution were measured by electrozone sensing.3.2. Assessment of Microbubbles in a Magnetic Field (MF)The behavior of microbubbles within a MF was determined using an optical microscope. The microbubble suspension was gently shaken before one drop was applied to the microscope slide, and a cover slip applied. The images were recorded digitally with a CCD camera under a 10×objective lens and a 4×magnification tube. A magnet (5000 GS) was placed beside the visual field to assess microbubble behavior under MF.3.3. Parallel Plate Flow-Chamber Adhesion StudiesPBS droplets (200μl) containing 1000 ng of recombinant mouse VCAM-1 Fc chimera were placed and fixed in a 1 cm diameter circular area on culture dishes. The flow chamber was placed on a microscope in a custom-designed stage and the culture dish side was inverted, and the MF placed underneath. A suspension of MBvM, MBv or MBiM (5×106 ml-1) was drawn through the flow chamber with an adjustable withdrawal pump with or without MF-guidance at an initial shear stress of 1 dyn/cm2. Because the MF was used to manipulate microbubble behavior at different shear stress conditions (1-24 dyn/cm2), MF-guidance was implemented for the first 5 min of microbubble infusion, after which it was removed and followed by a 5 min "flush". The number of microbubbles attached to the plate was determined over 20 optical fields (total area,0.5 mm2) at the 5 min and 10 min time points using Image Pro-Plus (IPP, Media Cybernetics) software to auto track and count microbubbles.4. Ultrasound molecular imaging of early stage artherosclerosis with magnetic microbubbles targeted to VCAM-14.1. Animals Preparation and DietWe used 4 different mouse models:apolipoprotein E-deficient (APOE-/-) mice on a hypercholesterolemic diet, APOE-/-mice on a regular diet, wild-type mice (C57BL/6) on a hypercholesterolemic diet, and wild-type mice on a regular diet which served as normal controls. Animals were anaesthetized and a jugular vein was cannulated for intravenous administration of microbubbles.4.2. Contrast Enhanced Ultrasound (CEU) Molecular ImagingUltrasound imaging was performed with a high-frequency linear-array probe (17L5) held in place. A MF (5000 GS) was placed under the abdomen of anesthetized mice. CEU was performed with Contrast Pulse Sequencing. Real-time imaging at 0.18 MI was performed after intravenous bolus injection of 1×106 MBvM, MBv or MBiM performed in random order. After 5 min of imaging the MF was manually removed, and continuous imaging continued for 10 minutes. The mechanical index was transiently increased to 1.0 for 3 s, to destroy adhered microbubbles, and subsequent post-destruction images were acquired at 0.18 MI, to obtain background images. To determine signal from retained microbubbles alone,3 post-destruction contrast frames representing any freely circulating microbubbles were averaged and digitally subtracted from 3 averaged pre-destruction frames use the Yabko MCE2.7 software (University of Virginia, USA) and then color-coded. Background-subtracted signal intensity was measured from a region of interest placed over the abdominal aorta.Results1. Binding capability of microbubbles targeted to VCAM-1 under pulsatile high-shear flow conditions1.1. Both MBv and MBi were prepared successfully, the concentration of MBv was about 2.9 x 108/ml, the mean sizes was 2.55±0.75μm respectively.1.2. The marked binding of MBv were seen in pulsatile and continuous flow conditions at low-shear flow conditions of 0.5-2 dyn/cm2, but the binding rate in the pulsatile flow group was higher (P< 0.05) than that in the continuous flow conditions. Furthermore, the marked binding of MBv was still noted at the highest shear rates (4-8 dyn/cm2) under pulsatile flow conditions, while it was not observed under continuous flow conditions. Microbubble detachment was assessed by increasing the flow every 30 s and observing the number of microbubbles remaining bound. Better retention of microbubbles was found on the 1000 ng/ml VCAM-1 Fc surfaces (F= 12.011, P< 0.001). The half detachment rate of MBv was high up to 22.1±2.6 dyn/cm2.2. Ultrasound molecular imaging of early stage artherosclerosis with microbubbles targeted to VCAM-12.1 There were no significant differences in aortic peak velocity, aortic diameter and heart rates between the different animal groups.2.2. Contrast Enhanced Ultrasound (CEU) Molecular Imaging The obviously signal enhancement was only observed for MBv in APOE-HCD mice (10.21±1.60), and was greater than any microbubbles used in any other animal groups (F=62.203, P<0.001).2.3. ImmunohistochemistryOn immunostaining, VCAM-1 expression was detected on the regions of luminal endothelial surface of the aorta in wild-type mice on hypercholesterolemic diet. In APOE-/-mice, there were visually intimal thickening and large atherosclerotic plaques protruding into the lumen, particularly in animals on hypercholesterolemic diet. Immunohistochemistry in APOE-/-mice demonstrated VCAM-1 expression on the endothelium, which was higher in animals on hypercholesterolemic diet.3. Preparation of magnetic microbubbles targeted to VCAM-1 and in-vitro assessment.3.1. Microbubble size, concentration and distribution were not significantly different between groups. After addition of magnetic streptavidin and biotinylated antibody to the biotinylated microbubble, the size distribution profile remained unchanged, suggesting a lack of aggregation.3.2. While the behavior of microbubbles bearing magnetic streptavidin, magnetic microbubble and inactive magnetic microbubble, were influenced by the presence of a magnetic field, non-magnetic microbubble, remained stationary.3.3. Both MBvM and MBv demonstrated similar attachment to plates coated with VCAM-1 Fc within the flow chamber in the absence of MF-guidance (F=0.131, P=0.877), while MBiM had minimal bubble attachment at the initial shear stress of 1 dyn/cm2. For all microbubbles, attachment decreased exponentially as shear stress increased, becoming minimal at shear rates>8 dyn/cm2. Upon MF-guidance, attachment of MBvM was significantly higher than MBv (F= 70.236, P< 0.0001). When MF-guidance was applied for the initial 5 min of infusion, both magnetic microbubble and inactive magnetic microbubble demonstrated binding at higher shear flow (8-16 dyn/cm2), while the attachment of MBiM remained unchanged. After termination of MF-guidance and a 5 min "flush", minimal MBiM attachment remained at low shear flow; however MBvM remained firmly attached even at high shear flows of 12-16 dyn/cm2 and was significantly greater than MBv and MBiM at each level of shear stress (F= 60.398, P< 0.0001).4. Ultrasound molecular imaging of early stage artherosclerosis with magnetic microbubbles targeted to VCAM-1Background-subtracted color-coded CEU images showed the obviously signal enhancement was observed for MBvM in APOE-HCD mice. After termination of MF-guidance and 5 min "flush", in wild-type mice on regular diet, signal for MBvM was low and similar to MBiM and MBv. In the other three groups, background-subtracted signal intensity for MBvM was greater than MBv and MBiM.The interaction between targeted microbubble agent and animal group was highly significant (F= 77.545, P< 0.001), suggesting that the difference in signal for MBvM depended on the animal group (disease severity).Conclusions1. The targeted microbubbles binding to VCAM-1 specific and effective at high-shear stress under pulsatile flow conditions. The molecular ultrasound imaging could be potentially use in the high-shear conditions artery system.2. MBv can use to ultrasound molecular imaging of early stage inflammation in artherosclerosis, but still need improvement.3. MBvM can be manipulated by a magnetic field and have the better binding than MBv and MBiM, using this method could provide better ultrasound molecular imaging of artherosclerosis.4. Magnetic field-guided molecular CEU imaging using magnetic microbubbles targeted to endothelial VCAM-1 improves the detection of the early stages of atherosclerosis in mice.