Radiolabeled Arginine-Glycine-Aspartic Acid Peptides to Image Angiogenesis in Swine Model of Hibernating Myocardium
Lynne L. Johnson, MD, FACC*,*,
Lorraine Schofield, BS ,
Tammy Donahay, BS,
Mark Bouchard, BS,
Athena Poppas, MD, FACC,
Roland Haubner, PhD ,
* Columbia University, New York, New York
Rhode Island Hospital, Providence, Rhode Island
Medizinische Universität Innsbruck, Innsbruck, Austria
Technische Universität München, Munich, Germany.
 |
Abstract
|
|---|
Objectives: Our aim was to image angiogenesis produced by endomyocardial injection of phVEGF165 in a swine model of hibernating myocardium using [123I]Gluco-arginine-glycine-aspartic acid (RGD) targeting the  vβ3 integrins.
Background: A noninvasive test to monitor the efficacy of therapy inducing angiogenesis is needed. The interaction between extracellular matrix and endothelial cells in sprouting capillaries is effected primarily by vβ3 integrins that bind through RGD motifs.
Methods: At 21 ± 4 days, after left circumflex coronary artery ameroid constrictor placement, 8 swine received endomyocardial injection of 1.2 mg phVEGF165 divided into 6 sites and 6 swine received saline (S) using nonfluoroscopic 3-dimensional endocardial mapping system (Noga)-guided delivery. After 20 ± 6 days, 13 animals were injected with 6.4 ± 1.7 mCi [123I]Gluco-RGD, 1 VEGF (vascular endothelial growth factor)-injected animal with I-123–labeled peptide control, and all animals with 2.5 ± 0.4 mCi of Tl-201 and underwent single-photon emission computed tomography imaging. Blood flow and echocardiographic measurements were made at both time points and tissue analyzed for fibrosis and capillary density by lectin staining.
Results: Hibernating myocardium in the ameroid constrictor territory at time of injections was documented by reduced wall thickening compared with remote. Ratio of myocardial blood flow in left circumflex coronary artery/left anterior descending coronary artery territories increased by 15 ± 11% in the VEGF animals and fell 13 ± 12% in S-injected (p < 0.01). There was a small increase in wall thickening in constrictor territory after VEGF (8 ± 17%) while in S-injected animals wall thickening fell by 23 ± 31% (p = 0.01 vs. VEGF). Lectin staining as percent positive tissue staining for ameroid territory was higher in VEGF-injected compared with S-injected animals (2.5 ± 1.5% vs. 0.87 ± 0.52%, p = 0.01). Focal uptake of [123I]Gluco-RGD corresponding to Tl-201 defects was seen in VEGF-injected but not in S-injected animals. [123I]Gluco-RGD uptake in the ameroid territory as percent injected dose correlated with lectin staining (R2 = 0.80, p = 0.002).
Conclusions: These data suggest that single-photon emission computed tomography imaging of radiolabeled RGD peptides may be a useful noninvasive method to monitor therapy that induces angiogenesis in the heart.
Key Words: angiogenesis • VEGF • integrins • cyclo-RGD peptides • myocardial hibernation
|
Abbreviations and Acronyms
| | HPLC = high pressure liquid chromatography | | ID = injected dose | | LV = left ventricle/ventricular | | LVEF = left ventricular ejection fraction | | PET = positron emission tomography | | RGD = arginine-glycine-aspartic acid | | SPECT = single-photon computed tomography | | VEGF = vascular endothelial growth factor |
|
Approaches to produce therapeutic angiogenesis in patients with chronic ischemic myocardium have included administration of recombinant human vascular endothelial growth factor (VEGF) protein into the coronary arteries (1), administration of naked plasmid deoxyribonucleic acid encoding for phVEGF2 injected into the myocardium either with a thoracotomy or percutaneously using a catheter-based myocardial injection and mapping system (Biosense-Webster, Irwindale, California) (2–5), and intracoronary injection of genes encoding for VEGF (6–8). The results of clinical trials have been mixed. Initial experience with cell-based therapy (CD34+ bone marrow-derived cells) has shown some efficacy in early trials. In the absence of strong evidence to support transformation of these progenitor cells into myocytes, a mechanism proposed for observed benefits is release by these cells of cytokines including VEGF (9,10). There is a need to develop a noninvasive test to monitor the efficacy of therapy that directly or indirectly induces angiogenesis.
The interaction between the extracellular matrix and endothelial cells is affected by a family of cell adhesion receptors (integrins) that bind to cells in sprouting capillaries (11,12). The vβ3 integrin receptor has been identified as a principle integrin expressed in neovascular growth (13–16). Binding of cells to the extracellular matrix occurs through arginine-glycine-aspartic acid (RGD) motifs on ligands. This observation forms the basis for development of radiolabeled ligands with RGD motifs (17). Haubner et al. (18–20) improved pharmacokinetics of vβ3 selective first-generation tracers by glycosylation of a modified derivative using sugar amino acid to achieve decreased lipophilicity and decreased hepatic uptake. It was the purpose of this study to test the hypothesis that angiogenesis produced by endomyocardial injection of phVEGF165 in a swine model of hibernating myocardium can be imaged using [123I]Gluco-RGD targeting the vβ3 receptors on smooth muscle cells that bind components of the extracellular matrix during capillary sprouting.
|
Methods
|
|---|
For all experiments, conditioned castrated male juvenile swine weighing 20 to 30 kg were used. All experiments were performed within the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and with the approval of the Rhode Island Hospital Animal Care Committee. Upon arrival at the Animal Care Facility, all study animals were fed a standard pig chow.
Surgery.
The left thoracotomy approach was used for placement of single left circumflex coronary artery ameroid. The left atrial appendage was lifted and a 1-cm segment of the proximal circumflex artery was dissected free from the surrounding myocardial tissue. The arterial segment was carefully lifted using silk suture at the proximal and distal ends and the carefully sized ameroid (Research Instruments, Escondido, California) placed snugly around the artery but not constricting it. The ameroid size for most of the experiments was 2.75 mm (range 2.50 to 3.00) with a gap size of 0.8 mm. The left atrial appendage was then released and the pericardium loosely closed. A chest tube was placed in the chest cavity, exiting through the skin. After re-expansion of the left lung and re-establishment of negative chest pressure, anesthetics were discontinued and the animal was allowed to awaken. Animals received daily oral antibiotics for 5 days.
Noga mapping and intramyocardial injection.
After 2 to 3 weeks, animals were anesthetized and a midline neck incision made exposing the right carotid artery and internal jugular vein. A 9-F sheath was placed in the artery and a catheter placed into the vein. Coronary angiography was performed through the carotid artery followed by transthoracic echocardiography. An Agilent 5500 Sonos platform (Philips, Andover, Massachusetts) with a phased array 2- to 4-MHz probe was used. Animals were positioned supine and imaged in 4 standard imaging planes: parasternal long-axis, parasternal short-axis, apical 4-chamber, and apical 2-chamber views.
Endocardial R-wave mapping was then performed using a nonfluoroscopic 3-dimensional endocardial mapping system (Noga) (Johnson & Johnson, Biosense-Webster, Diamond Bar, California). A reference was centered over the heart using the graphics workstation. The mapping catheter was passed through the carotid sheath and advanced into the left ventricular (LV) cavity. The position of the catheter tip was located in 3-dimensional space. The stability of the catheter-to-wall contact was evaluated at every site. With Noga-guided placement, 6 injections (1 ml) of VEGF were made in the 2 selected areas (3 per area) in 8 animals and saline in 6 animals (control group). A catheter was placed in the LV, and 1 injection of colored microspheres was given to document blood flow. The sheath was removed and the carotid artery repaired.
Final study.
Two weeks after intervention, the animal was returned to the Cardiovascular Research Laboratory for the final study. Coronary angiography was first performed followed by electromechanical mapping of the heart to obtain 60 to 90 points. Echocardiography was performed followed by injection of thallium-201 (2 to 3 mCi) and vβ3-selective [123I]Gluco-RGD in 13 animals and 123I-labeled negative control peptide in 1. One to 2 h later (based on blood pool clearance of [123I]Gluco-RGD), 2 sequential tomographic scans were performed without moving the animal on an ADAC Arc 3000 camera interfaced with a Nuclear Mac (Nuclear Cardiology Systems, Boulder, Colorado) over 180° orbit at 30 s per stop for 32 stops using the 70 keV photopeak and 15% window for Tl-201 and 160 keV photopeak with 15% window for I-123.
At completion of all the imaging procedures, anesthesia was deepened and the animal sacrificed with a bolus of potassium chloride and the heart removed. The heart was "breadloaf sliced" and imaged ex vivo on the detector for I-123. Tissue samples were removed from the heart for weighing, staining, fibrosis measurement, and blood flow analysis.
Tracer preparation, injection, and imaging.
The labeling precursor used was vβ3-selective glycopeptide cyclo(-Arg-Gly-Asp-D-Tyr-Lys[SAA]) (Gluco-RGD) where SAA stands for sugar amino acid, which is in this case is 3-acetamido-2,6-anhydro-3-deoxy-β-D-glycero-D-gulo-heptanoic acid (18). Gluco-RGD (200 to 300 µg; molecular weight 850.9 g/mol) was dissolved in phosphate buffered saline and transferred to an Eppendorf cap coated with Iodogen. Approximately 25 mCi of carrier added 123I-iodine (>2 x 108 GBq/mol) (MDS Nordion, Ottawa, Ontario, Canada) was added. After 30 min at room temperature, the solvent was removed from the solid oxidizing agent. The labeled compound was separated from precursor by high pressure liquid chromatography (HPLC). The HPLC conditions were 0% to 40% acetonitrile in water with 0.1% trifluoro acetic acid with flow of 1 ml/min, UV detection at 220 nm and 20 min gradient, using a Nucleosil C-18 5 µm, 125 x 4.6 mm column (Capital HPLC Ltd., West Lothian, Scotland, United Kingdom). The precursor has a retention time of 13 min and [123I]Gluco-RGD of 16 min. The 2 peaks were well separated. The specific activity was determined by the specific activity of the radioisotope. The solvent was removed before injection. The radiochemical yield was about 50%. Activities of 25 mCi resulted in approximately 13 mCi [I-123]Gluco-RGD. The negative control peptide, cyclo(-Arg-Ala-Asp-D-Tyr-Val-) was labelled using the same protocol as described for Gluco-RGD.
Nuclear data processing.
The raw tomographic nuclear data were processed using standard software. The thallium and [123I]Gluco-RGD scans were coregistered in polar map display format to display the relative distributions of the 2 tracers in each heart. The percent injected dose (ID) was calculated from the decay corrected injected activity, I-123 counts in the ameroid territory, tissue weight, and camera efficiency.
Echocardiographic data processing.
Analysis of LV global and segmental function was performed by experts blinded to the results of other imaging modalities and to pathology. Regional wall thickening was measured off-line using digital calipers for 4 equal segments (septal, anterior, lateral, and posterior) from the midshort-axis view. Wall thickening for each segment was calculated as percentage change in wall thickness from end diastole to end systole. Global left ventricular ejection fraction (LVEF) was calculated using a Simpson's rule algorithm.
R-wave voltage map processing.
Post-processing of the endocardial map was done using the Biosense work station. The final map for each animal was selected and displayed. The apex location was checked for accuracy, and the map was played back in cine (motion) mode. Poor points were deleted using established criteria.
Tissue processing and image analysis.
The heart was prepared for histology and microsphere measurements in the following manner. Each 1-cm breadloaf slice was divided into anterior, septal, posterior, and lateral regions. The slices were traced and the tracings annotated for localization of sample sources. Samples were taken from each region for microsphere counting. Weights from the lateral wall samples were used for the ID calculations. Two 4-µm sections from each of the 4 anatomical regions (anterior, septal, posterior, and lateral) from each 1-cm slice were taken for fibrosis measurements. Each 4-µm section was mounted and stained with trichrome. Each stained section was scanned into a computer and imported into Photoshop. Color channels for each image were split. Image math function in National Institutes of Health Image was used to quantify the percentage fibrosis and irreversibly damaged myocytes for each region (21).
To identify capillary sprouting, a 5-µm transmural tissue block was taken from the injection region for lectin staining. Dolichos biflorus lectin is a glycoprotein capable of recognizing and binding to carbohydrate moieties on endothelial cell membranes of sprouting capillaries. It was biotinylated to link with streptavidin-horseradish peroxide, which links to 3,3'-diamonobenzadine (brown chromagen). Sections were stained with 3,3'-diamonobenzadine and counterstained with hematoxylin. Ten fields were randomly captured at 20x from 3 different slides. Using Image Pro Plus (Media Cybernetics, Inc., Silver Springs, Maryland), the brown DAB color (lectin) was defined as was the entire field of tissue (blue). The combined areas for all 10 fields positive for lectin staining were expressed as percent lectin-positive staining over total tissue area. To identify vitronectin receptors in the injection site, vβ3 staining was performed on sections from 1 experiment injected with VEGF showing uptake of gluco-RGD and from 1 experiment injected with saline. For fluorescent staining, the vβ3-fluorescein antibody was applied (Vitronectin receptor, clone LM609, Chemicon, Tenecula, California). The tissue was fixed in acetone and blocked with 0.3% of H2O2 in methanol. The primary antibody was applied in a dilution of 1:100 (Vitronectin receptor, clone LM609, Chemicon), in an overnight incubation. Staining was validated with positive control in kidney tissue and negative control with FITC-labeled IgG in porcine myocardium.
Blood flow measurements.
Colored microspheres (15 ± 0.43 µm) used for blood flow determination were cross-linked polystyrene-divinylbenzene microspheres in 8 colors: red, blue, orange, green, yellow, coral red, violet, and black (E-Z Trac, Los Angeles, California). These colored microspheres are chemically stable and exhibit no dye leaching, even in tissue exposed to strong acid and base solutions. Regional myocardial blood flow values were measured by using the methods described by Hale et al. (22) and as reported previously for our laboratory (23).
Statistical analysis.
Numeric data for the 2 groups (VEGF and saline injected) were compared using a 2-sample t test with equal variance. All results are expressed as ± SD. The myocardial uptake of [123I]Gluco-RGD for each heart was plotted against the percent lectin staining using a simple linear regression.
|
Results
|
|---|
Ameroid constrictors.
Ameroid constrictors were placed on circumflex arteries in 14 swine. The mean time between ameroid placement and VEGF injection was 21 ± 6 days, and the mean time between VEGF injection and the final imaging and sacrifice was 20 ± 6 days. The initial weight was 25 ± 5 kg, and the final weight was 49 ± 9 kg. For the 8 VEGF-injected animals, 6 of 8 ameroids were patent at the time of injection and 5 of 8 were patent at the final study with collaterals present in 2 of 3 of the closed vessels. For the 6 saline-injected animals, 4 to 6 ameroids were patent at time of injection and 3 of 6 were patent at the final study with collaterals seen in one. Hibernation in the circumflex territory was supported by comparing R-wave amplitude, myocardial blood flow, and wall thickening in the circumflex territory to remote myocardium as shown in Figure 1. While wall thickening was reduced in the ameroid territory, R-wave amplitude was within "viable" range, and myocardial blood flow was mildly reduced compared with that in the remote myocardium.
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1 Documentation of Hibernating Myocardium
All swine underwent unipolar voltage (UPV) mapping (top panel), myocardial blood flow (MBF) measurements with microspheres (middle panel), and transthoracic echocardiography (bottom panel) before myocardial injection of either phVEGF125 or saline. The UPV scores from the Noga map in the ameroid territory were reduced compared with those in the remote myocardium but in the viable range. MBF in cc/g/min was mildly reduced in the ameroid territory compared with that in the remote myocardium. The percent wall thickening in the ameroid territory was reduced compared with that in the remote myocardium. These findings of abnormally functioning but viable myocardium fulfill the criteria for hibernation.
|
|
Blood flow and LV function.
Transmural myocardial blood flow in the ameroid territory increased by 0.63 ± 0.35 cc/gm/min between injection and final study in animals injected with VEGF, and fell by 0.14 ± 0.34 cc/gm/min in animals injected with saline (p = 0.004). The ratios of blood flow in the lateral wall (ameroid territory) over the remote myocardium at baseline were not different between VEGF- and saline-injected animals (0.96 ± 0.15 vs. 1.09 ± 0.02, p = 0.12). Ratios increased after 3 weeks in the VEGF-injected animals (15 ± 11%) and decreased in the saline-injected animals (–13 ± 12%) (p = 0.006) (Fig. 2).
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2 Change in Blood Flow Ratios in VEGF-Injected and Control Animals
Myocardial blood flow (MBF) is expressed as ratio of MBF in the ameroid territory to remote territory. The change (± SD) from immediately before intramyocardial injection of either phVEGF165 or saline to 3 weeks after injection for both treatment and control experiments are expressed as percent. The blood flow ratio increased in the vascular endothelial growth factor (VEGF)-injected animals (orange bar) and decreased in the saline-injected animals (yellow bar). LAD = left anterior descending coronary artery; LCx = left circumflex artery.
|
|
Baseline global and regional LV function was not different between the 2 groups: LVEF was 59 ± 11% for the VEGF group, and 51 ± 15% for the saline group (p = 0.28); percentage anterior wall thickening was 40 ± 30% for the VEGF group and 55 ± 31% for the saline group (p = 0.38), and percentage lateral wall thickening was 35 ± 18% for the VEGF group and 45 ± 25% for the saline group (p = 0.55). LV function improved slightly between the 2 studies in the VEGF-injected animals but continued to deteriorate in saline-injected animals. The percentage change in LVEF in the VEGF-injected animals was +4.3 ± 9.0% and for the saline-injected animals was –11.4 ± 12.5%. While this difference was statistically significant (p = 0.04), there was a wide range in values. In the VEGF-injected group, 5 experiments showed no change in LVEF and 3 showed an increase while in the saline-injected group; only 1 showed no change and 5 fell. There was a small increase in wall thickening in the ameroid territory after VEGF (8 ± 17%), while, in the saline-injected animals, wall thickening fell by 23 ± 31% (p = 0.01 vs. VEGF) (Fig. 3).
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3 Change in LV Function Over Time
Global and regional left ventricular (LV) function were measured using echocardiography immediately before and 3 weeks after intramyocardial injection with either phVEGF125
(yellow bars) or saline (orange bars). Mean value for left ventricular ejection fraction (LVEF) increased slightly but with a large standard deviation (error bar) while mean value for LVEF fell in the saline-treated animals but also with a large standard deviation. Wall thickening (WT) in the ameroid territory increased slightly in the vascular endothelial growth factor (VEGF)-treated animals and fell in the saline-treated animals.
|
|
Fibrosis and angiogenesis.
Fibrosis in the ameroid territory (as percent of the transmural sections) was higher in the saline-injected animals than in the VEGF-injected animals (6.2 ± 1.9% vs. 3.9 ± 1.6%) (p = 0.007). Histopathological evidence for neovascularization in the VEGF-injected animals compared with that in the saline injected was documented by lectin-positive staining (2.50 ± 1.5% vs. 0.87 ± 0.52%, p = 0.01) (Fig. 4). The VEGF-injected tissue stained for vitronectin receptors from several experiments showed positive fluorescence compared with saline-injected tissue (Fig. 4). Supporting the pathology were the Noga map findings that showed improvement in the R-wave voltage in the VEGF-injected territory compared with reduction in the saline-injected territory (Fig. 5).
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4 Immunohistology Documenting Angiogenesis in VEGF-Injected Animals
Values from quantitative immunohistology in the ameroid territory for the vascular endothelial growth factor (VEGF)-injected animals (yellow bars) and saline-injected animals (orange bars) are shown in the upper panel. The VEGF-injected experiments showed less fibrosis and greater capillary sprouting by lectin staining than the saline-injected control experiments. The bottom images show representative examples of tissue staining. Panel A shows a myocardial section stained for lectin from the injection territory from 1 VEGF experiment, and panel B shows 1 saline-injected animal. Brown staining of sprouting capillaries is seen in panel A and is absent in panel B. Panel C shows green fluorescence identifying integrin expression in capillary sprouts from the injection site of 1 VEGF-injected animal that showed in-vivo uptake of [123I]Gluco-RGD. FITC = fluorescein.
|
|
View larger version (64K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5 Noga Maps and Unipolar Voltage Score in VEGF-Injected and Control Animals
Images shown on the left are unipolar voltage maps at the time of injection with the injection sites marked for 1 saline-injected animal and 1 vascular endothelial growth factor (VEGF)-injected animal. The 2 Noga maps on the right were acquired during the final study on the same 2 animals. The color bar in the right upper corner of each image corresponds to unipolar voltage values from 6 mV (orange) to 15 mV (pink). The saline-injected region shows a fall in unipolar voltage score, while the VEGF-injected region shows an increase in unipolar voltage score.
|
|
Scan results.
Blood pool clearance curve of the radiolabeled peptide showed blood levels below 20% of peak by 2 h after injection. All animals showed mild resting thallium defects in the ameroid territory. The VEGF-injected animals showed focal uptake of [123I]Gluco-RGD in the same territory while none of the saline-injected animals or the radiolabled control peptide showed focal uptake of radiotracer in the heart (Fig. 6). There was no difference in thallium uptake as percentage peak counts in the ameroid territory between VEGF-injected and saline-injected myocardium (0.78 ± 0.07 vs. 0.70 ± 0.09, p = 0.12). The ratio of I-123 counts in the ameroid territory to remote myocardium was 1.71 ± 0.11 for the VEGF-injected animals versus 1.04 ± 0.13 for the saline-injected animals (p < 0.001). The counts per gram in the injection site were higher for the VEGF-injected animals than for the saline-injected animals (180 ± 153 vs. 51 ± 14) but the difference was not significant while the percent ID, also higher in VEGF-injected animals than in the saline-injected animals (0.26 ± 0.09 vs. 0.15 ± 0.07 x 103), was borderline significant (p = 0.04). The difference in counts per gram between the 2 groups was not significant, and the difference in percent ID was borderline significant (p = 0.04). When values for the percent ID (per gram tissue) for the 11 experiments in which all data for these calculations were available were plotted against quantitative lectin staining in the corresponding myocardial segments, there was a significant correlation (R2 = 0.80, p = 0.002) (Fig. 7). There was, however, overlap between the 2 groups. One of the saline-injected animals showed higher myocardial background uptake of [123I]Gluco-RGD and 1 of the VEGF-injected animals had a low percent ID corresponding to lower lectin staining.
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6 SPECT Scans for RGD-Based Targeting of Neoangiogenesis
On the left are shown reconstructed short-axis (SA), vertical (VLA), and horizontal long-axis (HLA) single-photon computed tomography (SPECT) slices from the thallium scans (top rows) and I-123 scans (bottom rows) from 1 representative animal treated with vascular endothelial growth factor (VEGF) and 1 animal treated with saline. The VEGF-injected animal shows a mild anterolateral thallium defect and focal uptake of [123I]Gluco-RGD into the anterolateral wall. The colocalization of the thallium defect and focal hotspot is better displayed on the polar maps. The saline-injected animal shows a mild-to-moderate anterolateral and apical defect without focal uptake of of [123I]Gluco-RGD in the heart. RGD = arginine-glycine-aspartic acid.
|
|
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7 [123I]Gluco-RGD Uptake Versus Angiogenesis
Graph shows percent injected dose (ID) per gram of tissue for [123I]Gluco-RGD plotted versus lectin staining as percent area for vascular endothelial growth factor-injected animals (yellow triangles) and saline injected (orange squares). Although the relationship is significant, there is not a perfect separation between the 2 groups. One saline-injected animal showed no focal hotspot in the myocardium but showed a higher level of background activity for I-123. One of the vascular endothelial growth factor-injected experiments had low uptake of [123I]Gluco-RGD, which corresponded to lower levels of angiogenesis by lectin staining. RGD = arginine-glycine-aspartic acid.
|
|
|
Discussion
|
|---|
The present study is the first to image angiogenesis using a radiolabeled cyclo-RGD peptide targeting vβ3 receptors in VEGF-injected hibernating myocardium. Ameroid placement in swine produced myocardial hibernation/stunning based on reduced regional wall thickening with preserved R-wave voltage and mildly reduced myocardial blood flow. Focal uptake of [123I]Gluco-RGD was seen on single-photon computed tomography (SPECT) imaging corresponding to regions of phVEGF165 injection at approximately 3 weeks after injection in treated animals. No tracer uptake was seen in the ameroid territory of saline injection. Tracer uptake corresponded to documentation of vβ3 receptors and lectin staining for capillary sprouting in tissue samples from VEGF-injected territory. Myocardial blood flow and LV function improved in VEGF injected animals. These data suggest that radiolabeled Gluco-RGD probes may be useful to monitor angiogenesis therapy in hibernating myocardium.
Approaches to stimulate myocardial angiogenesis in patients with intractable angina who are not candidates for percutaneous transluminal coronary angioplasty or coronary artery bypass grafting have included intracoronary infusion or intramyocardial injection of a growth factor gene or protein (1–8). None of these studies showed unequivocal efficacy, and initial enthusiasm for these approaches has been tempered. Subsequent trials of stem cell therapy in myocardial infarction do not support the role of these cells to differentiate into and function as myocytes but do seem to support a paracrine effect of the engrafted cells through secretion of VEGF in response to hypoxia (10,11). These findings renew interest in developing approaches to noninvasively monitor myocardial angiogenesis by targeting VEGF expression.
Approaches to image angiogenesis in the heart or blood vessel wall have been reported using different imaging platforms (24–28). Targeting molecular species expressed during angiogenesis with imaging probes has been investigated for cancer and cardiovascular applications. VEGFs are widely distributed in mammalian tissues. The member of the VEGF family that plays an important role in angiogenesis is VEGF-A. These cytokines are expressed in response to up-regulation of hypoxia-inducible factor-1. In response to hypoxia, m-RNA encodes VEGF molecules with 121, 145, 165 amino acid sequences (29). For capillary sprouting to occur, there needs to be an interaction between the endothelial cell growth and the surrounding supporting tissue or extracellular matrix. This interaction necessary for new growth is mediated by the integrins, a family of cell surface receptors that recognize extracellular matrix proteins such as vitronectin, fibrinogen, and fibronectin and are necessary for optimal activation of growth factor proteins (12–16).
Integrins are heterodimeric transmembrane glycoproteins consisting of an alpha and beta subunit (11,12). The integrins aggregate with growth factor receptors and thereby facilitate their activation. Certain integrins are preferentially associated with specific growth factor receptors. The integrin vβ3 is associated with platelet-derived growth factor and VEGF receptors (13). This important role of vβ3 in promoting the binding of growth factor receptor and ligand has made it an important target for development of agents to image angiogenesis. Extracellular matrix proteins like vitronectin, fibrinogen, and fibronectin interact with the integrins via the amino acid sequence RGD (16). Haubner et al. (18) were the first to report the synthesis and biological evaluation of the first glycoslylated RGD-containing peptides (18). In this study we used Gluco-RGD and labeled it with iodine-123 (19). Direct halogenation involves a multistep preparation and requires HPLC, and in vivo dehalogenation degrades image quality. Subsequent to the completion of this project, additional peptide compounds have been synthesized and tested in vivo in tumor models. These compounds include versatile chelating moieties that allow linkage with both SPECT and positron emission tomography (PET) tracers (17). They are based on cyclic pentapeptide conjugated via the  -amino function of a lysine with chelators or use an vβ3 binding disulphide-bridged undecapeptide coupled with 6-hydrazinopyridine-3-carboxylic acid and labeled with 99mTc.
The use of probes containing the RGD motif to image angiogenesis has been reported for nuclear (SPECT and PET) applications and for magnetic resonance imaging (24–28). Myocardial imaging of angiogenesis has been reported in a rat and dog model of chronic myocardial infarction using an In-111-labeled vβ3-targeted agent (111In-RP748) for vβ3 and thallium-201 for perfusion (26). Dual isotope SPECT imaging showed focal uptake of 111In-RP748 in regions of reduced perfusion corresponding to regions of angiogenesis by histological and immunohistological analysis (27). A PET approach to myocardial imaging using dual imaging for both gene expression and angiogenesis was reported (28). In this study, a nonreplicating adenovirus expressing mutant herpesviral thymidine kinase (HSV1-sr39tk) reporter gene and human VEGF121 gene was constructed and injected into normal pig myocardium. Two days later, animals were injected with [18F]fluoro-hydroxymethylbutylguanine probe for the reporter gene and [18F]Galacto-RGD for angiogenesis. This novel approach documented reporter gene expression, but there was no significant uptake of the [18F]Galacto-RGD probe. This failure to detect angiogenesis in the normal porcine myocardium was attributed to insufficient time to allow angiogenesis to occur, but lack of a hypoxic environment may have contributed. The ameroid model of chronic myocardial ischemia/hibernation may create a substrate for expression of related genes favorable to angiogenesis.
The swine model of hibernating myocardium produced by slow constriction of a coronary artery was developed by Fallavollita et al. (30). Swine ameroid models have been used in pre-clinical studies to document angiogenesis. A proof of concept paper by Vale et al. (31,32) showed that direct injection into the myocardium of phVEGF165 could be performed using nonfluoroscopic electromechanical LV mapping to localize hibernating myocardium and record injection sites (32). They documented VEGF expression by Western blots in the injection sites, and in a subsequent paper the same group showed improvement in collateral myocardial blood flow (33).
|
Conclusions
|
|---|
In these experiments, naturally occurring angiogenesis in response to hypoxia was not observed in the saline-injected animals based on either lectin staining or detectable uptake of [123I]Gluco-RGD. Failure of the hypoxic stimulus to produce beneficial levels of naturally occurring angiogenesis was also indicated by failure to show improvement in perfusion or LV function. A possible explanation for these findings is that the severity and/or duration of tissue hypoxia was insufficient stimuli to produce "therapeutic levels" of endogenous VEGF. The conditions of these experiments were not designed to address these questions.
Tracer retention in a region might be due to circulating blood levels of compound at the time of imaging combined with differences in intramyocardial blood volume, or interstitial activity associated with changes in vascular permeability, and not active angiogenesis. However, the correlation between lectin staining and percent ID supports actual binding of the [123I]Gluco-RGD to vβ3 indicating growth factor activity. The study would have been strengthened by evaluation of control hearts without ameroids and treatment, and control hearts with phVEGF treatment. That the correlation between [123I]Gluco-RGD uptake and lectin staining does not show a perfect separation between the VEGF-injected and saline-injected animals indicates that at least in one experiment high background activity of 123I probably relating to in vivo dehalogenation affected results. Peptides with more stable linkage to radio-probes including 111In and 99mTc via chelators have been developed for SPECT imaging. These new tracers eliminate this problem.
|
Footnotes
|
|---|
This study was supported by the National Institutes of Health (RO1HL65395).
* Reprint requests and correspondence: Dr. Lynne L. Johnson, Columbia University Medical Center, PH 10-405, New York, New York 10032. (Email: lj2129{at}columbia.edu).
Manuscript received January 17, 2008;
revised manuscript received April 30, 2008,
accepted May 1, 2008.
|
REFERENCES
|
|---|
- Henry TD, Annex BH, McKendall GR, et al. The VIVA trial: Vascular endothelial growth factor in Ischemia for Vascular Angiogenesis Circulation 2003;107:1359-1365.[Abstract/Free Full Text]
- Symes JF, Losordo DW, Vale PR, et al. Gene therapy with vascular endothelial growth factor for inoperable coronary artery disease Ann Thorac Surg 1999;68:830-837.[Abstract/Free Full Text]
- Vale PR, Losordo DW, Milliken CE, et al. Randomized, single-blind, placebo-controlled pilot study of catheter-based myocardial gene transfer for therapeutic angiogenesis using left ventricular electromechanical mapping in patients with chronic myocardial ischemia Circulation 2001;101:2138-2143.[Web of Science]
- Losordo DW, Vale PR, Hendel RC, et al. Phase placebo-controlled double-blind, dose-escalating trial of myocardial vascular endothelial growth factor 2 gene transfer by catheter delivery in patients with chronic myocardial ischemia Circulation 2002;105:2012-2018.[Abstract/Free Full Text]
- Kastrup J, Jorgensen E, Ruck A, et al. Direct intramyocardial plasmid vascular endothelial growth factor-A165 gene therapy in patients with stable severe angina pectoris J Am Coll Cardiol 2005;45:982-988.[Abstract/Free Full Text]
- Grines CL, Watkins MW, Helmer G, et al. Gene therapy (AGENT) trial in patients with stable angina pectoris Circulation 2002;105:1291-1297.[Abstract/Free Full Text]
- Grines CL, Watkins MW, Mahmarian JJ, et al. A randomized, double-blind, placebo-controlled trial of Ad5FGDF-4 gene therapy and its effect on myocardial perfusion in patients with stable angina J Am Coll Cardiol 2003;42:1339-1347.[Abstract/Free Full Text]
- Stewart DJ, Hilton JD, Arnold JMO, et al. Angiogenic gene therapy in patients with nonrevascularizable ischemic heart disease: a phase 2 randomized, controlled trial of AdVEGF (ADVEFG121) versus maximum medical treatment Gene Ther 2006;13:1503-1511.[CrossRef][Web of Science][Medline]
- Payne TR, Oshima H, Okada M, et al. A Relationship between vascular endothelial growth factor, angiogenesis, and cardiac repair after muscle stem cell transplantation into ischemic hearts J Am Coll Cardiol 2007;50:1677-1685.[Abstract/Free Full Text]
- Vandervelde S, van Luyn MJA, Harmsen MC. Signaling factors in stem cell-mediated repair of infarcted myocardium J Mol Cell Cardiol 2005;39:363-376.[CrossRef][Web of Science][Medline]
- Xiong JP, Stehle T, Diefenbach B, et al. Crystal structure of the extracellular segment of integrin aVβ3 Science 2001;294:339-346.[Abstract/Free Full Text]
- Giancotti FG, Ruoslahti E. Integrin signaling Science 1999;285:1028-1032.[Abstract/Free Full Text]
- Senger DR, Claffey KP, Benes JE, Perruzzi CA, Segiou AP, Detmar M. Angiogenesis promoted by vascular endothelial growth factor: regulation through a1β1 and a2β1 integrins Proc Natl Acad Sci U S A 1997;94:13612-13617.[Abstract/Free Full Text]
- Byzova TV, Goldman CK, Pampori N, et al. A mechanism for modulation of cellular responses to VEGF: activation of the integrins Mol Cell 2000;6:851-860.[Web of Science][Medline]
- Tsou R, Isik FF. Integrin activation is required for VEGF and FGF receptor protein presence on human microvascular endothelial cells Mol Cell Biochem 2001;224:81-89.[CrossRef][Web of Science][Medline]
- Bayless KJ, Salazar R, Davis GE. RGD-dependent vacuolation and lumen formation observed during endothelial cell morphogenesis in three-dimensional fibrin matrices involves the avβ3 and a5β1 integrins Am J Pathol 2000;156:1673-1683.[Abstract/Free Full Text]
- Haubner R. avβ3-Integrin imaging: a new approach to characterize angiogenesis? Eur J Nucl Med Mol Imaging 2006;33:S54-S63.[CrossRef][Web of Science]
- Haubner R, Wester HJ, Burkhart F, et al. Glycosylated RGD-containing peptides: tracer for tumor targeting and angiogenesis imaging with improved biokinetics J Nucl Med 2001;42:326-336.[Abstract/Free Full Text]
- Haubner R, Kuhnast B, Mang C, et al. [18F] Galacto-RGD: synthesis, radiolabeling, metabolic stability, and radiation dose estimates Bioconjug Chem 2004;15:61-69.[CrossRef][Web of Science][Medline]
- Haubner R, Wester HJ, Weber WA, et al. Noninvasive imaging of avβ3 integrin expression using 18F-labeled RGD-containing glycopeptides and positron emission tomography Cancer Res 2001;61:1781-1785.[Abstract/Free Full Text]
- Thambar ST, Schofield L, Poppas A, Bouchard M, Williams DO, Johnson LL. Validation of R wave voltage endomyocardial mapping to assess myocardial fibrosis: comparison with thallium and dobutamine echocardiography in a swine model J Interv Cardiol 2003;16:23-31.[CrossRef][Medline]
- Hale SL, Alker KF, Kloner RA. Evaluation of nonradioactive, colored microspheres for measurement of regional myocardial blood flow in dogs Circulation 1988;78:428-434.[Abstract/Free Full Text]
- Johnson LL, Schofield L, Mastrofrancesco P, Donahay T, Farb A, Khaw BA. Technetium-99m glucarate uptake in a swine model of limited flow plus increased demand J Nucl Cardiol 2000;7:590-598.[CrossRef][Web of Science][Medline]
- Winter PM, Morawski AM, Caruthers SD, et al. Molecular imaging of angiogenesis in early-stage atherosclerosis with avβ3-integrin-targeted nanoparticles Circulation 2003;108:2270-2274.[Abstract/Free Full Text]
- Sadeghi MM, Krassilnikova S, Zhang J, et al. Detection of injury-induced vascular remodeling by targeting activated avβ3 integrin in vivo Circulation 2004;110:84-90.[Abstract/Free Full Text]
- Hua J, Dobrucki LW, Sadeghi MM, et al. Noninvasive imaging of angiogenesis with a 99m Tc-labeled peptide targeted at avβ3 integrin after murine hindlimb ischemia Circulation 2005;111:3255-3260.[Abstract/Free Full Text]
- Meoli DF, Sadeghi MM, Krassilnikova S, et al. Noninvasive imaging of myocardial angiogenesis following experimental myocardial infarction J Clin Invest 2004;113:1684-1691.[CrossRef][Web of Science][Medline]
- Wagner B, Anton M, Nekolla SG, et al. Noninvasive characterization of myocardial molecular interventions by integrated positron emission tomography and computed tomography J Am Coll Cardiol 2006;48:2107-2115.[Abstract/Free Full Text]
- Ylä-Herttuala S, Rissanen TT, Vajanto I, Hartikainen J. Vascular endothelial growth factors J Am Coll Cardiol 2007;49:1015-1026.[Abstract/Free Full Text]
- Fallavollita JA, Logue M, Canty JM. Stability of hibernating myocardium in pigs with a chronic left anterior descending coronary artery stenosis: absence of progressive fibrosis in the setting of stable reductions in flow, function and coronary flow reserve J Am Coll Cardiol 2001;37:1989-1995.[Abstract/Free Full Text]
- Vale PR, Losordo DW, Tkebuchave T, Chen D, Milliken CE, Isner JM. Catheter-based myocardial gene transfer utilizing nonfluoroscopic electromechanical left ventricular mapping J Am Coll Cardiol 1999;34:246-254.[Abstract/Free Full Text]
- Vale PR, Losordo DW, Milliken CE, et al. Left ventricular electromechanical mapping to assess efficacy of ph VEGF 165 gene transfer for therapeutic angiogenesis in chronic myocardial ischemia Circulation 2000;102:965-974.[Abstract/Free Full Text]
- Tio RA, Tkebuchava T, Scheuermann TH, et al. Intramyocardial gene therapy with naked deoxyribonucleic acid encoding vascular endothelial growth factor improves collateral flow to ischemic myocardium Hum Gene Ther 1999;102953–660.
Related Article
-
Targeted Imaging Offers Advantages Over Physiological Imaging for Evaluation of Angiogenic Therapy
- Albert J. Sinusas
J. Am. Coll. Cardiol. Img. 2008 1: 511-514.
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
E. A. Osborn and F. A. Jaffer
The year in molecular imaging.
J. Am. Coll. Cardiol. Img.,
January 1, 2009;
2(1):
97 - 113.
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. J. Sinusas
Targeted imaging offers advantages over physiological imaging for evaluation of angiogenic therapy.
J. Am. Coll. Cardiol. Img.,
July 1, 2008;
1(4):
511 - 514.
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Methods