| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Basic Science Investigation |
1 Molecular Medicine Program, Mayo Clinic College of Medicine, Rochester, Minnesota; 2 Department of Radiology, Mayo Clinic College of Medicine, Rochester, Minnesota; and 3 Division of Cardiovascular Diseases, Mayo Clinic College of Medicine, Rochester, Minnesota
Correspondence: For correspondence or reprints contact: Noel M. Caplice MD, PhD, Division of Cardiovascular Sciences, Biosciences Institute, Room 4.07, University College Cork, Cork, Ireland. E-mail: n.caplice{at}ucc.ie
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Key Words: cell therapy • myocardial infarction • dynamic PET/CT cell tracking
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Initial human studies using 18F-FDG-labeled stem cells tracked with PET/CT indicate that retention of intracoronary injected stem cells within the heart at 1–2 h is low (ranging from 0.2% to 39%, varying with the cell fraction used) (6,7). However, the dynamics of cell washout from the myocardium within the first hour of injection are unknown. Insights into early cell trafficking may facilitate the development of more efficient delivery strategies, which may in turn impact therapeutic efficacy. We tested the feasibility of dynamic in vivo 3-dimensional (3D) PET/CT tracking of 18F-FDG-labeled CPC during intracoronary injection in a porcine model of acute myocardial infarction (MI).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Derivation of CPC
Yorkshire pigs (30-40 kg, Larson Products; 7 animals total) were used for harvest of porcine CPC and for the in vivo model of MI. One pig did not undergo induction of MI but served as a normal control to confirm feasibility. CPC were also obtained from normal human volunteers for ex vivo studies to determine labeling efficiency. From both sources, mononuclear cells were harvested from peripheral buffy coat preparations in Ficoll-Paque Plus (Stem Cell Technologies, Inc.). Cells were washed 3 times in MCDB 131 supplemented with hydrocortisone, antibiotics, and 10 ng/mL vascular endothelial growth factor (VEGF). Cells were then resuspended in XVivo-15 medium (Biowhittaker) supplemented with VEGF (1 ng/mL) and seeded on fibronectin-coated plates at a density of 4.9 x 103 cells per mm2 in a modification of previously described methodology (2).
Adherent cells were incubated at 37°C with DiI (1,1'-di-octadecyl-3,3,3,3'-tetramethylindocarbocyanine perchlorate)–labeled acetylated low-density lipoprotein (Ac-LDL) for 1 h and Alexa-Fluor 488–conjugated isolectin IB4 for 30 min (Molecular Probes). Fluorescence-activated cell sorting (FACS) was performed to identify cell-surface antigens on human and porcine CDC. Primary antibodies to CD31, CD105, von Willebrand Factor, eNOS, Flk-1, and VE-cadherin (R&D Systems) were used with secondary antibody detection with fluorescein isothiocyanate in each case. Isotype-matched IgG antibodies were used as controls, and the fluorescent intensity of stained cells was gated according to established methods.
Induction of MI
A closed-chest unconscious porcine model of acute MI was used (9). Each animal received prophylactic amiodarone (100 mg daily) and aspirin (325 mg daily) beginning 2 wk before induction of MI that was continued throughout the study. All animals also received lidocaine (60-mg intravenous bolus and infusion at 1.5–3.0 mg/min), unfractionated heparin (10,000 units intravenously), and benzathine penicillin (300,000 units intramuscularly) at procedure initiation. After anesthesia, mechanical ventilation with a Harvard apparatus was maintained on a fraction of inspired oxygen of 40%. Under sterile conditions, a carotid arterial cutdown was performed for placement of a 9-French arterial sheath. Fifty milliliters of blood were drawn into ethylenediaminetetraacetic acid (EDTA) tubes for progenitor cell isolation and culture as described. Through an 8-French coronary artery guide catheter, a slightly oversized angioplasty balloon catheter (1.1:1 balloon-to lumen ratio) was positioned in the middle portion of the left circumflex coronary artery and inflated to produce arterial occlusion for 90 min. Complete arterial occlusion was verified with angiographic contrast every 15–30 min. After the procedure, the arterial sheath was removed, followed by animal recovery and habitation. All animals survived to the end of PET acquisitions and were sacrificed thereafter.
18F-FDG Labeling of CPC
Forty-eight hours after induction of MI (pigs) or blood sampling (human volunteers) progenitor cells were harvested from culture. Porcine and human cells were detached using 0.5 mM EDTA. All cells were washed twice in phosphate-buffered saline (PBS). Cells were incubated under sterile conditions using previously described methodology (6) in a 10-mL test tube containing 100 MBq 18F-FDG in serum-free PBS (pH 7.2) and 10 U/mL heparin (Roche) and 0.1 U/mL recombinant human insulin (Novo Nordisk). To investigate the optimal temperature for labeling, porcine and human samples were divided into 3 groups: incubation for 30 min at 20°C, 28°C, or 37°C. On the basis of the results suggesting the highest labeling efficiency at 37°C, subsequent labeling studies were performed at 37°C. Cells were incubated in a lead-shielded incubator with gentle swirling for 10, 20, 30, 45, or 60 min. Labeled cell solutions were then centrifuged in a 3-step washing process in heparinized PBS (5 min at 150g each time). Activity in the final cell pellet and supernate was measured for determination of labeling efficiency. Stability of 18F-FDG-CPC resuspended in autologous serum at 37°C was assessed by measurement of cell-associated activity at baseline, 1 h, and 2 h after labeling.
For in vivo porcine experiments, 3 x 107 viable cells were resuspended in 12 mL of XVivo-10 medium (Biowhittaker) for balloon-occlusion intracoronary injection or in 4 mL of XVivo-10 medium for high-concentration, single-bolus intracoronary injection. The total activity of this cell solution was determined using a dose-calibrator immediately before injection. Separate aliquots were maintained in solution at the same cell concentration for measurements of 18F-FDG elution at various time points after injection. Cell viability was assessed using the trypan-blue exclusion method. Labeling efficiency was defined in this study as the percentage of cell-bound activity in the final cell solution (total activity of the solution minus activity in the supernatant, divided by total activity) after incubation and washing steps had been completed.
Baseline MRI Studies
Cardiac imaging with MRI was performed on all animals at 48 h after infarction but before progenitor cell therapy. A 1.5-T MRI system (Signa Horizon 9.0; GE Healthcare) with gradient strength of 22 mT/m was used, and a phased-array surface coil was used for each study. All images were electrocardiographically gated and acquired during suspended respiration (the animal was sedated and ventilated for the procedure). A detailed description of the MRI protocol is provided in a supplemental appendix (supplemental material is available online only at http://jnm.snmjournals.org).
In Vivo CPC Infusion and Simultaneous Systemic Blood Sampling
Animals were anesthetized, and a marginally oversized over-the-wire angioplasty balloon (1.1:1 balloon-to-lumen ratio) was placed at the same angiographic position as that used for induction of infarction. The balloon and catheter system was then secured in place. The animal was transported to the PET suite using a customized cradle that ensured stability of the catheter system during transfer.
Six pigs underwent induction of MI. These animals were divided into 2 groups, each containing 3 pigs. Group 1 underwent intracoronary injection of cell therapy using a balloon-occlusion delivery strategy. Group 2 underwent intracoronary injection of cell therapy using a high-concentration, single-bolus injection strategy (without balloon occlusion). For group 1, the intracoronary balloon was inflated at the same location within the coronary artery as that used for induction of MI, and a 4-mL aliquot (1 x 107 cells) of the 12-mL autologous CPC cell suspension was infused via the central lumen. The balloon remained inflated for a total of 4 min to prevent backflow of cells and to produce stop-flow beyond the site of inflation, and then deflated for a 4-min period of reperfusion. This cycle was repeated 2 more times in the same fashion, for a total infusion of 12 mL of cell suspension (containing 3 x 107 cells) per animal. For animals receiving high-concentration, single-bolus therapy, a 4-mL cell solution containing 3 x 107 labeled cells was injected through the central lumen of the balloon catheter but without inflation of the balloon. The cell bolus was injected over a 2-min period.
For both injection protocols, serial blood samples were drawn from the arterial sheath at 1-min intervals, the first sample to coincide with the beginning of the first injection cycle. Two samples of 5 mL were drawn at each time point. One sample was centrifuged immediately, and the supernatant was aspirated and passed through a 20-µm filter to remove all cells and debris. Both samples (whole blood and supernatant) were then passed through a 
-counter (1480 Wallac Wizard; Perkin Elmer), and the activity of each sample was measured in counts per minute. Percent cell-bound activity of the whole-blood sample was calculated by subtracting the activity of the supernatant from activity of the whole blood and dividing this figure by the activity of the whole-blood sample.
PET/CT Tracking of Intracoronary Injected 18F-FDG-CPC
Cardiac imaging with PET/CT was performed on a Discovery RX system (GE Healthcare); the PET component of this scanner has a high sensitivity (7.3 cps/kBq) (10). A CT scout acquisition was acquired from which the scan range of the helical and PET scans were prescribed, with the myocardium centered within the axial field of view. A helical CT scan was acquired with 140 kVp, 120 mA, 0.5-s rotation speed, and a pitch of 1.75. The slice thickness was 3.75 mm and the slice interval was 3.27 mm. Subsequently, a dynamic 3D PET acquisition was performed, with 30 frames at 1 min each. The PET acquisition was started on the injection of the labeled cells; the injected activity was 416.25 ± 160.7 kBq (11.25 ± 4.34 µCi). Immediately after the dynamic acquisition, a series of whole-body acquisitions were performed. Scout and helical CT images were acquired, followed by three 3D whole-body PET acquisitions consisting of 3 bed positions, 3 min per bed position, and a 7-slice overlap.
The PET images were reconstructed with Fourier rebinning ordered-subset expectation maximization using 21 subsets with 3 iterations and a 5-mm full width at half maximum gaussian postprocessing filter. The PET data were reconstructed into a 450-mm field of view with a pixel size of 3.52 mm and slice thickness of 3.27 mm. For all reconstructions, random events were estimated from the singles rate of each crystal, the CT images were used for the attenuation correction, and scatter was corrected for using a model-based technique for the 3D acquisition.
The PET/CT data were imported into PMOD (version 2.75; PMOD Technologies) for analysis. Regions of interest were drawn around the myocardium on the CT data and then copied onto the dynamic PET data. Time–activity curves were generated and were converted to absolute units of activity through the application of a cross-calibration factor. The cross-calibration factor was derived from acquisitions of spheres of known size and volume and from the measurement of an aliquot of activity in the -counter. In this manner, corrections are made for both the effects of partial volume and the different geometries of the scanner and -counter.
Data Analysis
All data are expressed as mean ± SE. A minimum of 3 independent in vitro experiments was performed for each set of labeling conditions. Comparison of results between multiple groups in cell labeling and in vivo experiments was by ANOVA. Statistical significance was set a priori at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Effects of Temperature and Incubation Time on 18F-FDG Cell Labeling
In initial experiments, CPC were incubated for 30 min at 20°C, 28°C, or 37°C. Labeling efficiency at 20°C was 41.2% ± 11.2% (human) and 30.3% ± 10.8% (pig), at 28°C was 53.6% ± 7.4% (human) and 54.3% ± 9.3% (pig). As shown in Figure 1A, labeling efficiency was significantly increased for both human and porcine CPC with incubation at 37°C (92.7% ± 5.9% and 90.4% ± 6.1%, respectively; P < 0.01). There was no significant difference between labeling at 20°C and 28°C. There was no significant difference between human and porcine cells with respect to labeling efficiency at any given temperature (P > 0.05 for each).
|
Assessment of stability of cell-bound 18F-FDG after labeling at 37°C for 30 min is presented in Figure 1C. Labeling efficiency of 18F-FDG-CPC resuspended for 1 h in autologous serum at 37°C decreased from 89.9% ± 4.8% to 74.3% ± 10.7% (human) and from 91.6% ± 6.4% to 77.7% ± 12.8% (porcine) (P > 0.05 for both when compared with baseline labeling). At 2 h, cell-bound activity had significantly declined when compared with baseline, to 62.1% ± 8.9% (human) and 68.6% ± 5.4% (porcine) (P < 0.01 for both human and porcine CPC).
Baseline MRI Characteristics and Cell Labeling for In Vivo Experiments
Six pigs underwent induction of MI. Baseline characteristics of experimental pigs in the 2 groups (group 1, balloon-occlusion delivery; group 2, high-concentration, single-bolus delivery) are presented in Table 1. There was no significant difference in any of the following baseline characteristics between group 1 and group 2 (P > 0.05 for each comparison). Body weight averaged 42.1 ± 0.2 kg in group 1 and 41.7 ± 0.5 kg in group 2. Baseline ejection fraction was 46.6% ± 4.2% (group 1) and 44.9% ± 5.1% (group 2). Left ventricular mass was 89.1 ± 6.1 g (group 1) and 86.5 ± 5.3 g (group 2). Infarct size was 14.5 ± 2.9 g or 16.3% ± 3.4% of LV mass (group 1) and 17.8 ± 1.9 g or 20.6% ± 2.7% of LV mass (group 2). The infarct zone was localized by delayed gadolinium enhancement to the inferolateral territory in all animals. End-diastolic volume was 64.2 ± 2.9 mL (group 1) and 67.3 ± 5.2 mL (group 2). End-systolic volume was 35.2 ± 5.8 mL (group 1) and 33.8 ± 4.9 mL (group 2). Cardiac output was 2.9 ± 0.1 L/min (group 1) and 2.7 ± 0.5 L/min (group 2). Stroke volume was 28.7 ± 2.3 mL (group 1) and 25.5 ± 2.2 mL (group 2).
|
|
Group 1 pigs were injected with 18F-FDG-CPC using a balloon-occlusion delivery strategy as described. 3D PET acquisition commenced with the first balloon inflation. Activity was localized to the appropriate left circumflex myocardial territory in each case (Fig. 2A), correlating with the MRI infarct zone. Quantitative assessment of cell trafficking in the myocardium during the cell delivery phase is illustrated in Figure 2B. This demonstrates 3 distinct, sharp increases in myocardial activity, each of which coincides exactly with the timing of balloon inflation and cell injection. Peak activity was reached within 2 min of balloon inflation, coinciding exactly with the completion of cell injection. Activity within the myocardium remained stable during the remaining 2 min of balloon inflation. Subsequent balloon deflation was associated with a marked drop in myocardial activity, with the loss of approximately 80% of peak activity per cycle with each balloon deflation (Fig. 2B). This abrupt drop in myocardial activity with each balloon deflation was associated on each occasion with the detection in the systemic arterial circulation of a transient spike of cell-bound (>90%) activity (Fig. 2B). This characteristic pattern of activity within the myocardium and systemic circulation was observed for all 3 animals.
|
PET acquisition was continued for 1 h after injection in each case. Peak activity during cell injection, 30-min activity, and 1-h activity within the myocardium were measured as a proportion of total initial injected activity (Fig. 3). For pigs treated with a balloon-occlusion delivery strategy, mean peak activity was estimated at 22.4% ± 3.1%, 30-min activity was 13.1% ± 2.1%, and activity at 1 h was 8.7% ± 1.5% of initial injected dose. For pigs treated with a high-concentration, single-bolus injection strategy, mean peak activity was 36.9% ± 6.8%, 30-min activity was 23.48% ± 10.2%, and activity at 1 h was 17.8% ± 7.9% of initial injected dose. Peak activity was significantly greater after high-concentration, single-bolus injection when compared with balloon-occlusion delivery (P < 0.05), but for 30-min and 1-h time points no significant difference was detected. For both groups 1 and 2, most of the injected activity at 1 h was localized to the lungs (>60%), with mild accumulation observed in the liver and spleen.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Previous human studies of stem cell tracking in vivo have been limited by an unavoidable delay between the timing of cell injection and subsequent PET acquisition (6,7). Intracoronary injections must be administered under direct fluoroscopic guidance in the cardiac catheterization laboratory. Patients are then transferred from the catheterization area to the PET facility. This involves a delay of at least 1 h between cell injection and the start of cell tracking (6,7). Experimental studies using 111In-labeled cells have not performed dynamic scanning and have not used intracoronary delivery (15–17). As a result, the trafficking of injected cells during intracoronary cell delivery is not understood.
We evaluated an established balloon-occlusion cell-delivery strategy using "real time" in vivo tracking and observed a pattern of activity that is consistent with significant loss of 18F-FDG-CPC from the myocardium after each cycle of balloon deflation. This was confirmed by first-pass detection in the systemic circulation of transient high levels of cell-bound activity that coincided with balloon deflation. These data identify an immediate washout of cells after reperfusion as a key target for strategies designed to augment myocardial engraftment. The effect of in vitro modification of progenitor cells (e.g., using cytokines or growth factors to upregulate adhesion molecules and integrins such as 
vß3) (18), or other strategies varying cell dose or balloon-occlusion time, could be evaluated using the preclinical model we describe.
One such strategy may involve use of higher concentrations of injected cell therapy. We tested an approach using a 3-fold higher concentration of cells delivered without balloon protection. Our findings confirm that differences in delivery strategy can be reliably characterized using this in vivo tracking technique. Not unexpectedly, peak myocardial activity after cell injection was greater when high-concentration, single-bolus injection was compared with 3-cycle balloon-occlusion delivery. Interestingly, a trend toward enhanced cardiac engraftment at 1 h was also observed. These data suggest that increased progenitor cell/endothelial interfacing associated with a single high bolus injection may obviate the need for multiple cycles of balloon occlusion, injection, and reperfusion. Apart from the simplicity of this approach, this strategy also eliminates the risk of coronary endothelial injury due to balloon inflation that is inherent in the conventional balloon-occlusion strategy.
The clinical relevance of this animal model is underscored by similarities we have demonstrated between human and porcine CPC labeling characteristics. Furthermore, we have previously demonstrated preservation of infarct territory mass associated with improved regional systolic function after therapy with autologous CPC in the pig (Noel M. Caplice, unpublished data, June 2007), effects that are consistent with those observed in human trials of progenitor cell therapy (19).
This dynamic PET/CT cell tracking technique may be applicable in a clinical setting in the future. However, such use would require greater integration of diagnostic and interventional imaging modalities than are currently available. Transport of patients with coronary equipment in situ and the performance of intracoronary injections without fluoroscopic guidance are the main concerns, as such maneuvers may pose a small, but unacceptable, hazard to patients (particularly for delivery strategies involving coronary balloon inflation). The ability to manipulate interventional catheters under fluoroscopic guidance is even more important for percutaneous intramyocardial injection of cell therapy, although assessment of engraftment after direct cell injection using an open surgical approach would still be possible in an experimental setting. Integrated fluoroscopy and PET/CT suites could overcome all of these limitations and greatly expand the scope of mechanistic insights that could be gained from human trials.
Limitations include the short half-life of 18F (110 min) and the relatively rapid elution of this radiolabel from the cells. The latter appears to be significant at 2 h, perhaps limiting specificity of the current technique to the first hour after injection. The development of radiolabels with longer half-lives and the evolution of labeling techniques that provide greater intracellular stability (without interfering with cell function) should further enhance the usefulness of this model.
Mean infarct size in this study was modest and did not result in significant impairment of left ventricular systolic function at 48 h after infarction in either of the 2 treatment groups. Use of the left anterior descending coronary artery for induction of infarction produces larger infarcts in pigs but is associated with prohibitively high mortality (up to 50%). Thus, the ability of this model to examine the impact of infarct size on cell engraftment may be somewhat limited.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONReferences |
|---|
Related articles in JNM:
This article has been cited by other articles:
|
|
 |
|
  D. L. Kraitchman and J. W.M. Bulte In Vivo Imaging of Stem Cells and Beta Cells Using Direct Cell Labeling and Reporter Gene Methods Arterioscler. Thromb. Vasc. Biol., July 1, 2009; 29(7): 1025 - 1030. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Willmann, R. Paulmurugan, M. Rodriguez-Porcel, W. Stein, T. J. Brinton, A. J. Connolly, C. H. Nielsen, A. M. Lutz, J. Lyons, F. Ikeno, et al. Imaging Gene Expression in Human Mesenchymal Stem Cells: From Small to Large Animals Radiology, July 1, 2009; 252(1): 117 - 127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Suuronen, P. Zhang, D. Kuraitis, X. Cao, A. Melhuish, D. McKee, F. Li, T. G. Mesana, J. P. Veinot, and M. Ruel An acellular matrix-bound ligand enhances the mobilization, recruitment and therapeutic effects of circulating progenitor cells in a hindlimb ischemia model FASEB J, May 1, 2009; 23(5): 1447 - 1458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 Y. Zhang, S. Thorn, J. N. DaSilva, M. Lamoureux, R. A. deKemp, R. S. Beanlands, M. Ruel, and E. J. Suuronen Collagen-Based Matrices Improve the Delivery of Transplanted Circulating Progenitor Cells: Development and Demonstration by Ex Vivo Radionuclide Cell Labeling and In Vivo Tracking With Positron-Emission Tomography Circ Cardiovasc Imaging, November 1, 2008; 1(3): 197 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Togel, Y. Yang, P. Zhang, Z. Hu, and C. Westenfelder Bioluminescence imaging to monitor the in vivo distribution of administered mesenchymal stem cells in acute kidney injury Am J Physiol Renal Physiol, July 1, 2008; 295(1): F315 - F321. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| JOURNAL OF NUCLEAR MEDICINE TECHNOLOGY | THE JOURNAL OF NUCLEAR MEDICINE |
