Animals
Rhesus monkeys (Macaca mulatta) from the Wisconsin National Primate Research Center (WNPRC) at the University of Wisconsin-Madison, an AAALAC accredited facility, were used in this experiment. All procedures were performed in strict accordance with the recommendations in the National Research Council Guide for the Care and Use of Laboratory animals (2011) and were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Wisconsin-Madison (experimental protocol G00673). All efforts were made to minimize the number of animals used and to ameliorate any distress caused by the experimental procedures outlined in this report.
Six adult male rhesus macaques (9-13 years old; 8-19 kg) were rendered hemiparkinsonian by the administration of a unilateral (right) intracarotid artery injection of the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) under sterile surgical conditions and isoflurane anesthesia, as previously described (28) (see Table 1 for animal information). Three to 12 months later, the monkeys received injections of allogeneic iPSC-mDA into the putamen ipsilateral to the MPTP dosing (right) under sterile surgical conditions and isoflurane anesthesia using validated methods (29). Briefly, iPSC-mDA were generated from rhesus fibroblasts (30), genomic edited to express mCherry and intracerebrally delivered using Real Time-Intraoperative Magnetic Resonance Imaging (RT-IMRI) for targeting (29,31,32). Each animal received approximately 15-20 million iPSC-mDA, distributed across three needle tracts with 2-3 deposits of 20-30 µL per deposit. Twenty four months following brain surgery, the monkeys underwent [18F]FEPPA PET scans (details below) under isoflurane anesthesia, as previously described (33). Approximately 72 hours after [18F]FEPPA PET, animals were anesthetized with sodium pentobarbital (25 mg/kg iv) and transcardially perfused with heparinized saline, followed by 4% paraformaldehyde (PFA) as previously described (28,34) to minimize variation in the response to the allograft.
[18F]FEPPA Radiochemistry
The synthesis of [18F]FEPPA was modified from previous methods (35,36) to reduce precursor mass and improve high performance liquid chromatography (HPLC) separation of product from impurities. Radiolabeling was performed on an automated chemistry process control unit (CPCU). Aqueous [18F]-fluoride was produced with a 16 MeV GE PETtrace cyclotron in a silver target via the 18O(p,n)18F reaction using enriched [18O]H2O. The solution was passed through an Accel Plus QMA Light Sep-Pak and eluted into a reaction vessel with 700 µL of a 20% water in acetonitrile solution of potassium carbonate and kryptofix222, and rinsed with 700 µL acetonitrile. Anhydrous acetonitrile was added to the solution and dried by azeotropic distillation. FEPPA tosylate precursor (2.0 mg) was dissolved in 0.4 mL anhydrous acetonitrile and added to the dry [18F]-fluoride vial. Nucleophilic substitution was performed by heating at 90ºC for 10 minutes to produce [18F]FEPPA. The reaction mixture was taken up in 1.0 mL ethanol and passed through an Alumina N Plus Light Sep-Pak to remove free [18F]-fluoride and kryptofix222. The solution containing [18F]FEPPA was diluted with 0.5 mL DI H2O and injected onto semipreparative HPLC for purification (Column: Phenomenex Luna C18, 10 µm, 250x10 mm; mobile phase: 20/80 acetonitrile/H2O + 0.5% formic acid; flow rate: 10 mL/min; UV wavelength: 254 nm). The collected fraction was diluted with 50 mL sterile water for injection, passed through a C18 Light Plus Sep-Pak, and eluted with 1.0 mL ethanol and 9.0 mL bacteriostatic saline through a 0.22 µm filter into a sterile, pyrogen-free vial. 30 µL of the final product was injected onto analytic HPLC (Column: Intersil ODS-4, 5 µm, 150x4.6 mm; mobile phase: 40/60 acetonitrile/0.1 N ammonium formate; flow rate: 2.5 mL/min; UV wavelength: 254 nm) to assess the impurity profile of the final product.
PET imaging
Twenty four months following brain surgery, all animals underwent [18F]FEPPA scans in a microPET Focus 220 scanner under isoflurane anesthesia, (1–3% in 100% O2, 1 L/min); vital signs (respiration, temperature, heart rate) were monitored throughout as described elsewhere (33). The monkeys were placed in the prone position with their head secured in a stereotaxic frame. After a 15-min transmission scan, radioligand was injected as an i.v. bolus (~185 MBq) over 30 s. Dynamic PET data were obtained for 2 h with conventionally increasing frame durations (6 x 30 s, 3 x 60 s, 2 x 120 s, 22 x 300 s). PET frames were reconstructed by 2D filtered back-projection using a ramp filter. PET images were processed and analyzed using Statistical Parametric Mapping 12 software (SPM12; Wellcome Department of Cognitive Neurology, London, UK). Time-activity curves were generated from the reconstructed PET time series data. Image frames from the 90-120 min duration were averaged, smoothed with a 4 mm Gaussian kernel, and registered to the MRI using a resting-state frame from the RT-IMRI following the final cell deposition. The 90-120 minute frame data was chosen at a sufficiently late time to avoid the bias of regional differences in radiotracer delivery (i.e. blood flow dependence), although the sensitivity to using earlier time windows was not examined. Region of interest (ROI) masks were generated for each iPSC-mDA graft site through segmentation of static frames of the RT-IMRI data collected during iPSC-mDA delivery. Precise ROIs of the grafts were drawn utilizing the signal voids in the RT-IMRI immediately following injection of the cell delivery vehicle using MRIcron (University of South Carolina, South Carolina, USA) and were utilized for visual confirmation of [18F]FEPPA binding at the graft location. Masks for the ipsi/contralateral cerebellum and ipsi/contralateral putamen were generated from a rhesus atlas in standardized space (37,38). The RT-IMRI data was spatially normalized to the rhesus template, and the resulting deformation fields were used to transform the ROI masks of the cerebellum and putamen into the native RT-IMRI space. PET voxel activity concentrations were normalized to the injected dose and NHP weight to generate standard uptake value (SUV) images utilizing the SPM12 software. The ROI masks were applied to the PET images to extract mean SUVs for each region. An asymmetry index (Equation 1) was calculated for the ipsilateral (SUVR) and the contralateral putamen (SUVL) to assess asymmetry in [18F]FEPPA between hemispheres. Using the cerebellum as a control region, an asymmetry index was calculated for the ipsi- and contralateral cerebellum to assess whether any asymmetry was a global artifact of the image.
See formula 1 in the supplementary files.
Postmortem brain tissue processing and analysis
After the brains were retrieved, they were post-fixated for 24-48 hours in 4% PFA, cryoprotected in graded sucrose solutions, and then cut frozen (40 μm sections) on a standard sliding knife microtome (American Optical Corp. Model 860, Buffalo, NY, USA) as previously described (28,34). Serial coronal brain sections spanning the putamen from all monkeys were immunostained with antibodies against mCherry (1:2000, Thermo Fisher Scientific, Waltham, MA, USA; 1:200, secondary Goat Anti-Rat IgG Antibody, Vector Lab, Burligham, CA, USA) or CD68 (1:3000, DakoCytomation, Glostrup, Denmark; 1:200 secondary Horse Anti-Mouse IgG Antibody, Vector Lab, Burlingame, CA, USA) and counterstained with Nissl to visualize brain anatomy. Immunostaining of tissue sections from all animals was performed in parallel and included negative and positive controls for each antibody (28,34).
Coronal brain sections immunostained against mCherry and CD68 were evaluated in a Zeiss Axioplan 2 imaging photomicroscope coupled to a MAC5000 high precision computer-controlled x-y-z motorized stage, and a MicroFire CX9000 camera. To minimize potential bias, an independent investigator evaluated mCherry immunostained tissue and selected the commissural and post-commissural putamen for evaluation of microglia/macrophage cell response. A different investigator, blind to the treatment, assessed the ipsi- and contralateral putamen of all subjects in three representative coronal brain sections (starting at the level of the anterior commissure and 960 μm apart) per subject that corresponded to the core of the graft. Tissue sections were first evaluated at low magnification (10X) to identify areas of CD68 immunoreactivity (CD68-ir). High magnification (40X) was then used to assess microglia/macrophage morphology.
Activated CD68-ir microglia/macrophages were defined as cells with dense rounded bodies (amoeboid conformation) and no or small thickened processes; resting microglia/macrophages were identified as small CD68-ir cells with thin ramifications (39,40) (Fig 1). Based on these criteria, the ipsi- and contralateral putamen of each brain tissue section was rated for CD68-ir microglia/macrophages activation from 0 to 4 using a semi-quantitative rating scale. The scoring system was: 0 = None, no more than 4 individual, small amoeboid CD68 labeled microglia/macrophages throughout the entire putamen; 1 = Weak, small amoeboid microglia/macrophages forming no more than 2 small, sparsely populated clusters, with no other microglia/macrophages scattered in the putamen; 2 = Minimal, small amoeboid microglia forming no more than 5 small, more dense clusters, with some amoeboid microglia/macrophages scattered individually throughout a localized area of the putamen; 3 = Moderate, slightly larger amoeboid microglia/macrophages, more numerous, may form small clusters, mostly spread throughout a localized area of the putamen; 4 = Strong, large, dark, rounded microglia/macrophages forming dense clusters that accumulate into distinct, localized areas of dark CD68 immunoreactivity with some spread of smaller microglia outside of the clusters. The final score for the ipsi- and contralateral putamen of each subject was obtained by averaging the results of three tissue sections. An AI for CD68-ir was then calculated as the difference between the contra- and ipsilateral putamen scores, to provide a qualitative index for comparison with the [18F]FEPPA PET analysis.
Statistics
Data collection and analysis were performed by investigators blind to the treatment. Statistical analyses were performed with R (v3.5.3). Results are presented as mean±SEM. The p-values for data sets were calculated using a two-tailed, paired samples Student’s t-test. The relationship between [18F]FEPPA AI and CD68-ir AI was analyzed using Spearman’s correlations. A p < 0.05 was considered significant.