Identification of Angiogenesis Rich-Viable Myocardium using RGD Dimer based SPECT after Myocardial Infarction

Cardiac healing after myocardial ischemia is a complex biological process. Advances in understanding of wound healing response have paved the way for clinical testing of novel molecular imaging to improve clinical outcomes. A key factor for assessing myocardial viability after ischemic injury is the evaluation of angiogenesis accompanying increased expression of integrin αvβ3. Here, we describe the capability of an αvβ3 integrin-targeting SPECT agent, 99mTc-IDA-D-[c(RGDfK)]2, for identification of ischemic but viable myocardium, i.e., hibernating myocardium which is crucial to predict functional recovery after revascularization, the standard care of cardiovascular medicine. In vivo SPECT imaging of rat models with transient coronary occlusion showed significantly high uptake of 99mTc-IDA-D-[c(RGDfK)]2 in the ischemic region. Comparative measurements with 201Tl SPECT and 18F-FDG PET, then, proved that such prominent uptake of 99mTc-IDA-D-[c(RGDfK)]2 exactly matched the hallmark of hibernation, i.e., the perfusion-metabolism mismatch pattern. The uptake of 99mTc-IDA-D-[c(RGDfK)]2 was non-inferior to that of 18F-FDG, confirmed by time-course variation analysis. Immunohistochemical characterization revealed that an intense signal of 99mTc-IDA-D-[c(RGDfK)]2 corresponded to the vibrant angiogenic events with elevated expression of αvβ3 integrin. Together, these results establish that 99mTc-IDA-D-[c(RGDfK)]2 SPECT can serve as a sensitive clinical measure for myocardial salvage to identify the patients who might benefit most from revascularization.

Scientific RepoRts | 6:27520 | DOI: 10.1038/srep27520 patients since 1970's, limitations are becoming apparent since single photon emission computed tomography (SPECT) imaging with 201 Tl or 99m Tc sestamibi cannot provide any molecular and pathophysiological insight on the defect region and utilize relatively high ionizing radiation. 18 F-fluoro-2-deoxy-D-glucose ( 18 F-FDG) positron emission tomography (PET) is another gold standard to measure myocardial viability more sensitively. The uptake and retention of 18 F-FDG reflects the activity of the various glucose transporters and hexokinase-mediated phosphorylation. In the setting of ischemic heart failure, viable myocardium often exhibits a shift in substrate utilization from aerobic (free fatty acids) to anaerobic (glucose) metabolism, thus, 18 F-FDG imaging provides an in vivo assessment tool for glycolytic activity of the ischemic myocardium and can be used to evaluate myocardial viability. A combination of perfusion and glucose metabolism imaging enables classification of myocardium, i.e., fibrous scar, when there is a decrease both in perfusion and metabolism; viable myocardial hibernation, when a perfusion/metabolic mismatch occurs; and normal tissue, when myocardial perfusion and metabolism were preserved 4 . Despite such well-established value for clinical assessment, this provides only little insight into the underlying biological processes after initiation of MI which makes it difficult to predict future cardiovascular events and assess individual efficacy of reperfusion therapy. To distinguish ischemically compromised but viable "hibernating myocardium" before manifestation of adverse remodeling, specifically targeted imaging technique of evaluating critical molecular processes is needed.
Myocardial ischemia results in hypoperfusion and tissue hypoxia, leading to the stimulation of angiogenesis, i.e., formation of new capillaries from existing microvessels. Accordingly, angiogenesis is considered as an important component of infarct healing which can be a key biomarker to delineate viable myocardium early after MI. In addition, it has been a target of molecular therapies to direct myocardial repair with several clinical trials [5][6][7] . Integrin α v β 3 , i.e., a cell membrane glycoprotein receptor that is highly expressed on endothelial cells during angiogenesis, has been identified as a favorable target for imaging angiogenesis and thus has attracted great interest in the field of MI staging 8,9 . Cumulative studies have revealed that its expression is up-regulated within the first few weeks after ischemic myocardial injury in the infarcted and border zone regions as part of the early infarct healing process [10][11][12] . Most notably, a recent study showed that a strong early integrin imaging signal from the ischemic region in rats was associated with less ventricular remodeling in subsequent weeks, suggesting that integrin expression is a potential biomarker of cardiac repair 13 . Imaging studies mainly focused on cyclic arginyl-glycyl-aspartic acid (RGD) peptide, a potent inhibitor of α v β 3 integrin, which was radiolabeled for SPECT or PET imaging of experimental MI models 10,14-20 . However, as yet none have shown comparison with combined use of perfusion and metabolic imaging, i.e., the most established clinical protocols. In addition, most trials have not carefully considered pathological advance and therapeutic process of MI by using simple permanent coronary artery ligation model, revealing limitations to return to the bedside for clinical testing.
We recently developed a new radiotracer, 99m Tc-labeled RGD peptide ( 99m Tc-IDA-D-[c(RGDfK)] 2 ) and successfully applied for diagnostic SPECT imaging of glioblastoma 21 and atherosclerosis 22 . Considering high integrin-binding affinity, specific in vivo targeting, and desirable pharmacokinetic properties as shown in the previous studies 21,22 , the developed RGD dimer agent is expected to be suitable to pinpoint hibernating myocardium which is clinically characterized by perfusion defect and enhanced FDG uptake. To prove such utility of 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT imaging approach, we particularly focused on in vivo imaging of transient coronary occlusion model to mimic reversible myocardial infarction and reperfusion in the clinical setting. Comparative measurements with 201 Tl SPECT and 18 F-FDG PET, then, showed that focal uptake of 99m Tc-IDA-D-[c(RGDfK)] 2 matched the hallmark of hibernation, i.e., the perfusion-metabolism mismatch pattern. Together, we demonstrate a molecular imaging strategy that uses α v β 3 integrin-targeted probe 99m Tc-IDA-D-[c(RGDfK)] 2 with SPECT to assess myocardial viability after ischemic injury to identify the patients who might benefit most from revascularization.

Generation of MI/reperfusion Model and Work Flow of Molecular Imaging.
To investigate the ability of α v β 3 integrin-targeted probe to detect hibernating myocardium which is crucially important in clinical decision-making, we established clinically relevant models of myocardial infarction (MI)/reperfusion by transient coronary occlusion of rats (n = 4). Figure 1 shows schematic work flow of model generation and molecular imaging. The anterior descending branch of the left coronary artery was ligated to create MI in rats, followed by reperfusion after 20 minutes to induce dysfunctional but viable (hibernating) myocardium (Fig. 1A). After 7, 14, and 28 days, animals were subjected to single photon emission computed tomography (SPECT) and positron emission tomography (PET) imaging, and sacrificed for histological analyses (Fig. 1A). We performed three different tracers' imaging, i.e, 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT, 201 Tl SPECT, and 18 F-FDG PET at different time points (7, 14, and 28 days post-injury) on each single animal repetitively. This made it possible to minimize the number of animals used and improve reliability of comparative imaging data by testing each tracer on the exactly same state of myocardium. Specifically, SPECT scans were performed for 30 minutes after serial injection of 99m Tc-IDA-D-[c(RGDfK)] 2 and 201 Tl intravenously (IV), immediately followed by 20-min computed tomography (CT) scans (Fig. 1B). 18 F-FDG PET/CT was conducted with the same animals at the day of SPECT/CT imaging performed. The PET images were obtained for 90 minutes with IV injection of 18 F-FDG after CT scans for 20 minutes (Fig. 1C).

SPECT Imaging of Viable Myocardium with 201 Tl and 99m Tc-IDA-D-[c(RGDfK)] 2 . Using a radiola-
beling protocol similar to that is described in our previous reports 21, 22 , we produced 99m Tc-IDA-D-[c(RGDfK)] 2 , a diagnostic imaging agent for angiogenesis, with chemical and radiochemical purities greater than 99% and specific activity greater than 55 GBq/μ mol. This agent, an integrin-binding RGD dimer peptide was designed to have increased hydrophilicity for optimized in vivo imaging. Its superior pharmacokinetic properties and high metabolic stability have been verified in the previous studies 21,22 . Herein we demonstrated the feasibility of SPECT imaging using 99m Tc-IDA-D-[c(RGDfK)] 2 to noninvasively detect hibernating (i.e., ischemic but viable) myocardium in surgically generated rat models of transient coronary occlusion. uptake ratio of ischemic to remote myocardium using standardized uptake value (SUV) showed that the value was peak at day 7 (1.63 ± 0.51) and gradually decreased in process of time, but high enough for clear differentiation between ischemic and normal regions even at day 14 (1.4 ± 0.21) and day 28 (1.23 ± 0.14) (Fig. 2D). Nonspecific uptake of 99m Tc-IDA-D-[c(RGDfK)] 2 in the liver was identified by strong SPECT signal at the bottom in Fig. 2A-C, but it does not interfere myocardial signals.

F-FDG PET Imaging of Viable Myocardium in MI/reperfusion Model. 'Perfusion defect' identi-
fied by loss of 201 Tl signal means the presence of scar (i.e., irreversible damage) or hibernating (i.e., reversible damage) myocardium 2,4 . To clarify whether intense uptake of 99m Tc-IDA-D-[c(RGDfK)] 2 would come from fibrous scar (non-viable myocardium) or viable myocardium, we next performed glucose metabolic imaging with 18 F-FDG PET by using the same rats on the day of SPECT imaging. As shown in Fig

Discussion
Despite considerable diagnostic and therapeutic advances for ischemic cardiovascular disease (CVD) over the past 40 years [26][27][28][29] , there remains a significant population of patients who are not managed well by current treatment approaches 1 . One of the main reason for this failure is clinical challenge to assess individual risk factor. The identification of a so-called 'hibernating myocardium' is the most important in predicting which patients will experience functional recovery after revascularization and which patients will not. Although perfusion and metabolic imaging have been widely used for diagnosing CVDs, such strategies do not reflect the activity of key biomarkers identifying myocardial viability such as angiogenesis. Another reason resulting in high mortality of CVD patients is the absence of novel revascularization therapy capable of sufficient cardiac repair. Angiogenic therapy is an attractive approach for regeneration of ischemic myocardium, however, as yet none have shown sufficient efficacy to be approved in clinical trials 7,30,31 . As more and more specific therapies emerge with a variety of angiogenic agents 5,6 , the need for equally specific diagnostic tests is growing to lead us into clear success story. Accordingly, there is an unmet clinical need to develop a specific imaging tool for evaluation of myocardial salvage, suitable candidate selection, and efficacy monitoring of novel therapy by targeting angiogenesis and angiogenesis rich-viable myocardium.
In this study, we showed the feasibility of SPECT imaging with α v β 3 integrin-targeted 99m Tc-IDA-D- Current revascularization procedures aimed at reopening the obstructed artery, such as thrombolysis, angioplasty, and bypass surgery improved the post-MI survival rate. To date, perfusion and metabolism imaging have become the gold standard for diagnosis, prognosis, and follow-up of the patients treated by revascularization. In contrast, specific molecular imaging has not yet been used extensively to characterize injured myocardium although the quest for new, potentially more specific targeting strategies has continued. Angiogenesis is a key factor in the process of cardiac healing after myocardial ischemia. As a result, much attention is paid to targeting angiogenesis for both diagnosis and therapy. However, molecular imaging strategy to track angiogenesis never been rigorously evaluated by comparison with current clinical standard, i.e., the combined measurement of myocardial perfusion and metabolism. The main advance of this work comes from this point. We clarified that angiogenesis targeting with the developed SPECT agent 99m Tc-IDA-D-[c(RGDfK)] 2 can exactly pinpoint injured but viable myocardium coincident with current diagnostic criteria (i.e., perfusion defect and metabolic activation). This was possible by the use of a carefully designed and validated transient coronary occlusion model. The demonstration of comparable imaging performance to assess myocardial viability in comparison with gold standard approach will facilitate clinical translation of 99m Tc-IDA-D-[c(RGDfK)] 2 based angiogenesis imaging technique to manage CVDs.
Much work is still ahead to clinically apply 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT for the use in ischemic heart disease to enable reliable detection of hibernation, offering information on the likelihood of response to a given therapy. Of note, an urgent need exists to investigate the capability of the α v β 3 integrin targeted 99m Tc-IDA-D-[c(RGDfK)] 2 to capture multiple components to boost cardiac recovery. Wound healing after MI entails a cascade of events including angiogenesis, inflammation and a complex interplay between several different cell types such as macrophages and myofibroblasts. Interestingly, α v β 3 integrin is highly expressed not only in endothelial cells during angiogenesis but also in macrophages 32,33 and myofibroblasts 12,34 . As such, it is worth to explore the specific range of in vivo targeting by the developed radiotracer. The strong correlation between 99m Tc-IDA-D-[c(RGDfK)] 2 and 18 F-FDG (Fig. 4) indicates that 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT can record not only angiogenesis but also macrophage activity because 18 F-FDG is usually accumulated not in cardiac muscle cells but in activated macrophages under fasting condition. In addition, further work needs to be accomplished to clarify the complex relationship between the timing of imaging, signal strength, and functional outcome. Comparison with other suggested imaging strategy such as 201 Tl SPECT with glucose-insulin-potassium (GIK) infusion will be also valuable to more accurately verify hibernating myocardium targeting ability of 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT. The potential of 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT imaging to distinguish delicate differences amongst different ischemic myocardium must be evaluated especially in order to clearly pinpoint hibernating myocardium from irreversible scar using different degree of occlusion models. To confirm whether 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT signal high region after myocardial injury, indicative of hibernating myocardium, could really show successful improvement in contractile function after revascularization will be of another great interest. Such assessment can be performed by physiological measurement on the regional left ventricular function using echocardiography or electrocardiography. Generalized, large, and prospective clinical trials are mandatory to reach beyond the tools available in laboratory research. Most importantly, however, this current study had laid the foundation for possible application of 99m Tc-IDA-D-[c(RGDfK)] 2 from fundamental research of cardiac healing process to patient management and new revascularization drug development.
Scientific RepoRts | 6:27520 | DOI: 10.1038/srep27520 In summary, the presented study suggests that noninvasive identification of viable myocardium under ischemic heart condition is feasible using α v β 3 integrin-targeted 99m Tc-IDA-D-[c(RGDfK)] 2 which can be readily incorporated into clinical practice to identify the patients who might benefit most from revascularization. As a sensitive angiogenesis detection probe to monitor revascularization therapy effect, this imaging platform can also pave the way to the successful development of a drug capable of revascularization of cardiac tissues which would be a major milestone in the history of cardiovascular medicine.

Materials and Methods
Animal Model of Myocardial Infarction and Reperfusion. All animal experiments were carried out in accordance with the approved guidelines. All animal experimental protocols were approved by the Institutional Animal Care and Use Committee of Preclinical Research Institute in the Seoul National University Bundang Hospital. A total of eight Sprague-Dawley rats (280-400 g, male, Charles River Laboratories, Wilmington, MA, USA) were used for this study. Generation of transient coronary occlusion (MI) followed by reperfusion (n = 4) has been fully described by previous investigators 10, 16,35 . Briefly, rats were anaesthetized with intraperitoneal administration of zoletil (30 mg/kg), xylazine (5 mg/kg) and then intubated and mechanically ventilated with a mixture of O 2 and 1.5% isoflurane using a rodent ventilator (Harvard model 683, Harvard Apparatus Inc., Boston, MA, USA). Animals were placed in a supine position and the body temperature monitored and maintained at 35-37 °C. A left thoracotomy was then performed through the third intercostal space by a horizontal incision of pectoralis muscles to expose the heart. The anterior descending branch of the left coronary artery (LCA) was ligated using a 5-0 polypropylene suture with a small curved needle. The coronary occlusion was confirmed by the regional appearance of pale color on the anterior surface of the left ventricle (LV). Twenty minutes after the ligation, the suture was cut and removed to achieve reperfusion. The model of transient 20 min coronary occlusion was selected to induce MI in approximately half of the region at risk and generate ischemic but viable "hibernating myocardium" as previously reported 10 . The chest was closed, and then the animal was gradually weaned from the respirator as soon as they are able to breathe spontaneously. Additionally, control sham-operated rats underwent the same surgical procedures except the occlusion of the LCA (n = 4).
SPECT imaging was followed by CT scans with the animal exactly in the same position. The animal CT scanner system consisting of a low-energy X-ray tube and a precision motion translation stage was used. The X-ray source and detectors are mounted on a circular gantry allowing it to rotate 360° around the rat positioned on a stationary bed. The images at 180 projections were acquired with the X-ray source set at 45 kVp and 177 μ A. Two-dimensional slices of the bed position were reconstructed using an Exact Cone Beam Filter Back Projection algorithm with a Shepp-Logan filter. Finally, the CT image was used to correct attenuation error of the gamma-ray signal emitted from the 201 Tl or 99m Tc-IDA-D-[c(RGDfK)] 2 .
In Vivo PET/CT Imaging. In addition to the SPECT/CT, a PET/CT scan with 18 F-FDG was done at the same day. All animals were fasted for 24 hours before imaging to follow the protocol of current clinical practice 36 , thus minimize 18 F-FDG uptake in normal myocardium. Rats were anesthetized with 1.5-2% isoflurane in 100% oxygen (2 L/min flow rate). Following a 20-min CT scan for attenuation correction, the PET data acquisition was started at the time of intravenous injection with 18 F-FDG (1.20 ± 0.09 mCi). The 3D static images were collected for 90 minutes with an energy window of 400-600 keV.
All PET scans were performed using a small animal PET scanner, NanoPET/CT (Bioscan Inc, USA), which provides a minimum axial coverage of 9.48 cm, a 12.3 mm transaxial field of view, 0.3 mm sampling distance, 0.58 mm in-plane reconstructed resolution and 7.7% of absolute sensitivity at the center of field of view for an energy window at 250-750 keV. Image Analysis. The acquired images were processed with the comprehensive image analysis software, PMOD (version 3.13, PMOD Technologies, Inc.). All emission images were co-registered with regard to the respective CT scan. The images ( 99m Tc-IDA-D-[c(RGDfK)] 2 SPECT, 201 Tl SPECT, and 18 F-FDG PET) were then Scientific RepoRts | 6:27520 | DOI: 10.1038/srep27520 analyzed to calculate the standardized uptake values (SUVs) for a given subject at different time points. SUV at time point t is defined as follows: where c is the measured tissue radioactivity concentration (mCi/mL) and injected activity is the amount of radiation (mCi/mL) injected extrapolated to time point t. Three-dimensional volumes of interest (VOI) were drawn based on the radioactivity normalized SPECT or PET images. Subsequently, the mean SUV in each VOI of hibernating and remote myocardium was calculated. The uptake ratio of hibernating to remote zone for 99m Tc-IDA-D-[c(RGDfK)] 2 and 18 F-FDG was independently calculated by division of the mean SUV of hibernating VOI with that of remote VOI.
Statistical Analysis. All quantitative data are expressed as mean ± SD. The correlation between quantitative parameters was evaluated by Spearman's rank correlation. Statistical significance was tested using ANOVA. Differences with a p value less than 0.01 were considered statistically significant. All analyses were conducted with SPSS 15.0 statistical package (SPSS, Chicago, IL, USA) or MedCalc (MedCalc version 6.15.000).
Histology and Immunohistochemistry. Following euthanasia of the rats underwent in vivo SPECT/CT and PET/CT imaging, myocardial tissue was isolated to confirm angiogenesis and integrin expression in the hibernating region by histopathological analysis. The excised tissues were fixed with 10% formalin, embedded in paraffin, cut into 5-μ m sections and deparaffinized. The sections were then stained with hematoxylin and eosin or anti integrin α ν β 3 monoclonal antibody (1:50, Abcam, ab7166) to identify characteristics of the recorded maximum and minimum radioactivity corresponding to the hibernating and remote myocardium, respectively. Bright field color micrographs were obtained on a BX51 microscope equipped with DP71 camera (Olympus Optical Co., Ltd., Tokyo, Japan).