Real-Time Intraoperative MRI Intracerebral Delivery of Induced Pluripotent Stem Cell-Derived Neurons

iPSC Generation and Neural Differentiation

An iPSC line derived from the skin biopsy of an adult rhesus monkey was used for the study. Fibroblasts were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS) ¹² . iPSCs were generated using CytoTune^®-iPS 2.0 Sendai Reprogramming Kit (Thermo Fisher Scientific, Waltham, MA, USA). Pluripotency was confirmed by teratoma assay ¹³ . Differentiation toward dopaminergic (DA) progenitors was performed as described previously ¹⁴ . Briefly, iPSCs at about 10% confluence (1 day after passaging) on a mouse embryonic fibroblast (MEF) feeder layer were cultured in neural induction medium consisting of DMEM/F12, N2 supplement, nonessential amino acids, and heparin (Invitrogen, Carlsbad, CA, USA) in the presence of bone morphogenetic protein (BMP)/SMAD pathway inhibitors SB431542 (10 μM) and LDN193189 (200 nM; both from Stemgent, Cambridge, MA, USA) to minimize differentiation toward mesodermal fate. Sonic hedgehog (SHH; C25II; 500 ng/ml; R&D Systems, Minneapolis, MN, USA) and the GSK3 inhibitor CHIR99021 (Stemgent) were added to the cultures from days 1 to 16. Then CHIR99021 was removed, SHH was reduced to 20 ng/ml, and fibroblast growth factor 8b (FGF8b; 100 ng/ml) was added to the culture to expand the progenitors in suspension for 12 days (D16-28). The neuroprogenitor spheres were then dissociated using Accutase and plated back down to Matrigel-coated coverslips for immunostaining and fed with neural differentiation media consisting of neural basal media (21103-049; Gibco, Waltham, MA, USA), minimum essential medium-nonessential amino acids (MEM-NEAA; 11140-050; Gibco), N2 Supplement (17502-048; Gibco), B27 supplement (17504-044; Gibco), brain-derived neurotrophic factor (BDNF; 10 ng/ml; 450-02; PeproTech, Rocky Hill, NJ, USA), glial cell line-derived neurotrophic factor (GDNF; 10 ng/ml; 450-10; PeproTech), TGF-β3 (1 ng/ml; 100-36E; PeproTech), ascorbic acid (200 μM; A0278; Sigma-Aldrich, St. Louis, MO, USA), and cyclic adenosine monophosphate (cAMP; 1 μM; A9501; Sigma-Aldrich). The cultures were fed every other day, and the neuroprogenitor clusters were triturated into smaller cell aggregates with a polished glass Pasteur pipette every week. DA progenitors were labeled with Hoechst 33342 (Thermo Fisher Scientific) 1 day before transplantation. To label the cells, 1 μg/ml of Hoechst was applied to the culture media for 30 min at 37°C. Cells were triple washed with artificial cerebrospinal fluid (aCSF) before transplantation.

MRI-Compatible Infusion System

A fused silica single-end port cannula was used in this study (Engineering Resources Group, Inc., Pembroke Pines, FL, USA). The 100.0-mm-long shaft has an inner diameter of 0.490 mm and an outer diameter of 0.650 mm. A clear infusion line (FEP fluoropolymer infusion line; IDEX Health & Science, Lake Forest, IL, USA) with an inner diameter of 0.500 mm was used to connect the cannula to a pressure transducer fixed to a 100-μl Hamilton syringe (Hamilton, Reno, NV, USA). An MRI-compatible syringe pump (PHD 2000; Harvard Apparatus, Inc., Holliston, MA, USA) was used to drive the syringe.

Uploading of aCSF and cell suspension into the system was done through the cannula using the MRI-compatible syringe pump attached to the control mechanism of the standard Harvard apparatus PHD 2000. The cannula attached to the loading lines was secured in place by an MRI-compatible uploader device (Engineering Resources Group, Inc.) (Fig. 1A).

OPEN IN VIEWER

Figure 1.

(A) Cell uploading system with a PCR tube [magnified in (a)] in the uploading platform and the cannula secured in place by the screw top system. (B) For cell inoculation into the dish, the cannula was secured in place by the uploader, while a dish was placed on a platform below. Arrows point to spheres within the cannula. (C-F) Microscopic images of neurospheres obtained 24 h after collection. The spheres were uploaded into the system and kept at room temperature for 0 (C), 30 (D), 60 (E), and 120 (F) min before being inoculated in a dish. The main finding was that after 120 min, the spheres came out of the cannula as small cylindrical aggregates (f), suggesting some adherence between neurospheres, which could be easily dispersed. (G) Cell survival after 24 h was not different across groups or condition. (H, I) Control and 120-min uploaded spheres maintained identity given by βIII-tubulin (red) and TH (green) expression after 2 weeks in culture. Scale bars: 200 μm (C-F), 500 μm (f), 50 μm (H, I).

During procedures, the pump was placed at approximately the same height as the cannula. A pressure-monitoring and infusion pump controller (Engineering Resources Group, Inc.) was used to regulate the infusion and monitor the pressure in the infusion line, as previously described^9,11. Pressure graphs were generated in real time, providing immediate feedback on the functionality of the system, including efficiency of cell suspension uploading and expulsion (Fig. 2).

OPEN IN VIEWER

Figure 2.

In-line pressure monitoring provides real-time feedback on cell suspension uploading and infusion. Pressure graphs for the 30-min time-lapse trials without (A) or with (B) submersion of the cannula tip in aCSF during the waiting period. Note the higher increase in in-line pressure (A) at the beginning of the infusion when the tip is not kept in aCSF.

Agarose Gel Infusion Boxes

A 0.6% agarose gel was used as a brain surrogate for testing the cell delivery system before the in vivo procedure, as its ratio of distribution volume to infusion, elicited infusion pressure, and resistance to insertion closely resemble those of the brain matter ¹⁵ . To prepare the gel, a 1× Tris/borate/EDTA (TBE) buffer solution was first made by mixing one part 10× TBE to nine parts water; then 0.6 g of agarose was added for every 100 ml of 1× TBE solution. The mixture was heated to a boil, allowed to cool down to 60°C, and then poured into clear plastic boxes, which serve as molds for the gel. A few milliliters of water was added on top of the solidified gel to prevent it from drying out. The boxes were covered and placed in the refrigerator for next-day use. During the gel infusions, a box top with an attached pivot-point base was used to mimic a trajectory guide (Fig. 3A).

OPEN IN VIEWER

Figure 3.

(A) Cell inoculations in gel were performed using the MRI-compatible delivery system, including the remote introducer for cannula insertion. Images of gel 1 (B) and gel 2 (C) after completing three and two cell deposits, respectively, following the cannula tract. (D) Pressure graph obtained during cell inoculation in gels 1 and 2. Scale bar: 1 mm (B, C).

In Vivo RT-IMRI Targeting and Infusion Monitoring

The in vivo validation of the system was done in one adult male rhesus monkey (Macaca mulatta; 15 kg). The animal was housed individually on a 12-h light/dark cycle and received food and water ad libitum. The experiment was performed in strict accordance with the recommendations of the National Research Council Guide for the Care and Use of Laboratory Animals (2011) and in an AAALAC-accredited facility (Wisconsin National Primate Research Center, University of Wisconsin-Madison). Experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Wisconsin-Madison. All efforts were made to minimize the number of animals used and to ameliorate any distress.

All imaging was performed in a 3-T GE SIGNA (GE Healthcare, Waukesha, WI, USA) MRI scanner. A custom 3-inch-diameter, receive-only surface coil (MR Instruments, Inc., Minneapolis, MN, USA) was used for scanning. For the imaging procedures, the animal was food deprived overnight. After sedation with ketamine [10-15 mg/kg intramuscular (IM) injection; MWI Veterinary Supply, Boise, ID, USA], the animal was intubated and an intravenous (IV) catheter was placed for fluid administration. Anesthesia was continued under isoflurane (1%-3%; Phoenix Pharmaceuticals, St. Joseph, MO, USA) for the duration of the scan. Atropine (0.02-0.05 mg/kg, IM; Phoenix Pharmaceuticals), buprenorphine (0.01-0.03 mg/kg IM; MWI Veterinary Supply), meloxicam (0.2 mg/kg SQ; Norbrook Laboratories, Newry, UK; Boehringer Ingelheim Vetmedica Inc., St. Joseph, MO, USA), and cefazolin (20-30 mg/kg IV; Apotex Corporation, Weston, FL, USA) were also administered. Vital signs (heart rate, blood oxygen, respiration, and temperature) were monitored during the procedure using MRI-compatible instruments. The animals were wrapped to retain body heat and placed in an MRI-compatible stereotaxic frame.

RT-IMRI guidance of the cannula utilized a pivot point-based MRI-compatible external trajectory guide ⁹ (Medtronic Inc.) (Fig. 4A). Before surgery, three-dimensional (3D) T1-weighted (T1W) MRIs were used to visualize the target structure in three planes (sagittal, axial, and coronal) to identify the entry point of the cannula, which defines the placement of the trajectory guide base, and to develop a roadmap of the trajectory that the cannula should follow. Placement of the system base was performed under sterile conditions using a custom-built MRI-guided real-time prospective stereotaxic surgery platform. A laser calibration rod attached to the stereotaxic micromanipulator and inserted in the trajectory guide stem was used to place the base, which was mounted to the skull over the craniotomy with three small self-tapping screws (Fig. 4C).

OPEN IN VIEWER

Figure 4.

(A) Base with remote introducer and guiding insert in place (modified from Emborg et al. ⁹ ). (B) Cone of projection of the base using 3D T1W MRI reconstruction. The points labeled T1 and T2 correspond to the two selected targets in commissural and postcommissural ventral putamen nucleus. (C) Surgical placement on primate skull using laser calibration rod inserted in the stem. The base was secured in place with three self-tapping screws.

After the placement of the base, the monkey was transported to the MRI suite and positioned in the MRI bed while remaining in the MRI-compatible stereotaxic frame. During RT-IMRI targeting and before cannula insertion, an alignment stem (dummy cannula) filled with 0.2 mM gadobenate dimeglumine (MultiHance; Bracco Imaging, Monroe Township, NJ, USA) dissolved in sterile saline was inserted in the trajectory guide stem in order to visualize positioning of the trajectory guide ⁹ . Before each injection, a 3D T1W scan with the IR-fGRE protocol was run to select the target, trajectory angle, and depth to the target. When the trajectory angle (anteroposterior, mediolateral direction) of the contrast fluid-filled dummy cannula was confirmed to be on target, the stem was locked into position. The fluid-filled cannula was removed, the remote introducer was fastened to the stem, and the adapter insert was placed in the stem.

Immediately before each of the two intracerebral injections, the system was set up and loaded with the neurosphere cell suspension. The loaded cannula was threaded through the remote introducer and guiding insert and fastened to the remote introducer. After the pressure was stabilized, the relative pressures were zeroed. The cannula was lowered into position, and a new targeting 3D T1W MRI was performed to confirm that the cannula had reached its target prior to infusion.

The first infusion target was the putamen nucleus at the coronal level of the anterior commissure. Following the cannula tract, three deposits of 15 μl each separated by 2 mm in the ventral-to-dorsal plane were performed (infusion rate: 2 μl/min). Five minutes after the completion of the third deposit, the cannula was slowly retracted and taken out of the brain with the remote introducer. The second infusion target was the putamen nucleus at the coronal level 2 mm posterior to the anterior commissure. Following the cannula tract, two deposits of 15 and 20 μl, respectively, each separated by 2 mm in the ventral-to-dorsal plane, were performed (infusion rate: 3 μl/min). After the last inoculation was completed and a 5-min postinfusion period lapsed, the cannula was retracted, and the animal was transported to the surgical room, where the base was removed and the incision was closed in layers ¹⁰ .

Two weeks after the surgery, the monkey was anesthetized with sodium pentobarbital (25 mg/kg IV injection) and transcardially perfused with heparinized saline, followed by 4% paraformaldehyde (PFA) ¹⁶ . The brain was postfixated in 4% PFA for 24 h and cryoprotected by immersion in a graded (10%-30%) sucrose/0.1 M phosphate-buffered saline (PBS; pH 7.2) solution. The tissue was cut frozen (40-μm sections) on a sliding knife microtome; the sections were kept in a cryoprotected solution until processing.

Immunocytochemistry and Cell Evaluation

Cell viability analysis was performed with standard trypan blue dye exclusion staining and quantification by hemocytometer ¹⁵ .

Cell identity was evaluated by immunofluorescence staining on coverslip cultures as described previously ¹⁴ . Primary antibodies utilized were βIII-tubulin (neuron-specific class III β-tubulin; AB9354; 1:10,000; Fisher Scientific, Waltham, MA, USA) and tyrosine hydroxylase (TH; a DA neuronal marker; P40101; 1:1,000; Pel-Freez Biologicals, Rogers, AR, USA). The immunostainings were developed with appropriate fluorescein-tagged secondary antibodies (Alexa Fluor 488- or Alexa Fluor 594- conjugated secondary antibodies; 1:1,000; Invitrogen, Molecular Probes, Eugene, OR, USA). Confocal images were obtained using a Nikon A1 confocal microscope (Tokyo, Japan).

Brain Processing and Evaluation

Immunohistochemical staining of brain sections was performed to evaluate host response, as previously described ⁵ . Primary antibodies used were CD68 (microglia/macrophage marker; M0814; 1:3,000; Dako Cytomation, Glostrup, Denmark), CD45 (monocyte marker; M0701; 1:1,000; DakoCytomation), and glial fibrillary acidic protein (GFAP; astroglia marker; Z0334; 1:2,000; Dako Cytomation); CD68 and CD45 staining were enhanced by the addition of ammonium nickel sulfate and counter-stained with Nissl. For each marker, immunoreactivity (-ir) in grafted areas was evaluated in two coronal slices using a Zeiss (Oberkochen, Germany) light microscope under low magnification (10×). CD68⁺ microglia was defined as resting when cells presented small bodies and thin elongated processes and as activated when the cells had enlarged cell bodies and thickened or intensely immunostained processes. Activated astrocytes were identified by GFAP-ir with intense hyper-ramified processes ¹⁷ .

Immunofluorescence staining was performed to evaluate neuronal identity of grafted cells, as previously described ¹⁶ . Primary antibody used was nestin (AB5922; 1:1,000; EMD Millipore, Billerica, MA, USA), followed by an appropriately tagged secondary Alexa Fluor 594 (1:1,000; Invitrogen). Confocal images were obtained using a Nikon A1 confocal microscope.

The total number of grafted Hoechst⁺ cells was determined using unbiased stereological cell-counting methods as described previously ¹⁸ . Briefly, the optical dissector system consisted of an AxioImager Zeiss microscope coupled to an imaging video camera system, and neuronal counts were performed using the optical fractionator probe in StereoInvestigator (MBF Bioscience, Williston, VT, USA). Sixteen sections were used for analysis. The graft was outlined under low magnification (2.5×), dissector counting frames 150 × 90 μm, tissue thickness manually adjusted, and cells counted using a 63× oil immersion objective.

Statistical Analysis

Prism version 5.0 (Graphpad Software Inc., La Jolla, CA, USA) was used for statistical analysis. Cell quantification data were evaluated using multiple comparisons analysis of variance (ANOVA). A value of p < 0.05 was considered statistically significant.

Optimization of Uploading

After the loading lines were primed with PBS, 10 or 20 μl of aCSF was uploaded as a buffer zone before introducing the cell suspension (50,000 cells/μl of aCSF). During this first set of experiments to optimize uploading, we identified that a greater volume of aCSF (e.g., 20 vs. 10 μl) provided a better separation zone for preventing cells from mixing with PBS. Uploading of the neurospheres presented some challenges. The cells were not easily withdrawn from a 500-μl conical tube, as they could freely move in the aCSF while the silica cannula was secured in place by the uploader screwtop (Fig. 1A). This issue was minimized by transferring the cell suspension into 200-μl PCR tubes and by gently raising the tube from the base holder (using forceps) to ensure that the cells were drawn through the cannula without losing contact with the suspension and potentially trapping air in the system. After the uploading was optimized, two trials were performed. The neurospheres were uploaded at a rate of 10 and 20 μl/min (total of 100,000 cells per trial). The neurospheres were easily visualized in the clear silica cannula and loading line tubing (Fig. 1B). After the pressure in the lines was stabilized, the cells were immediately expelled into a dish at 20 μl/min. The collected cells were observed under the microscope at 10× magnification; presence of aggregation or disruption of the neurospheres was recorded. As no changes in neurosphere integrity were noted, we next performed assessments of cell survival and differentiation over time.

In Vitro Evaluation of Cell Survival and Differentiation over Time

For each of the trials, the loading lines were first primed with injectable saline, and then 40 μl of aCSF was uploaded at a rate of 20 μl/min, followed by 30 μl of the neurosphere suspension containing 500,000 cells at a rate of 10 μl/min. The cells were uploaded in three 1-min periods; between each upload, approximately 1 min passed to allow for any necessary readjustments of the uploader and mixing of the sphere suspension. After uploading, the cell-loaded system was maintained at room temperature for 0, 30, 60, or 120 min. As controls, equivalent samples of cells were kept at room temperature for the same amount of time. Each timed trial was done in triplicate. During the waiting period, the tip of the cannula was submerged in aCSF with the exception of one 0-and one 30-min trial. Expulsion of a few microliters after inoculation of the prescribed loading volume revealed the presence of a few cells, suggesting that some cells were lagging in the line. Therefore, an additional 40 μl was expelled after sample collection to ensure that fresh aCSF and cells were evaluated for each trial. The 0-, 30-, and 60-min infusion trials showed no aggregation or occluding of the cannula tip by the spheres; however, during the 120-min infusion trials, it was observed that the spheres were expelled as small cylindrical aggregates. Observation under the microscope revealed that the neurospheres seemed to be stuck together, although the spheres looked healthy and were easily dissociated (Fig. 1C-F). Monitoring of the cell suspension in the loading lines using high-definition pictures did not reveal aggregation in the lines over time, although some cell displacement was observed if the lines were not maintained at the same height than the cannula and pump. To further validate the viability and identity of the cells after the 120-min trials, the cell suspension was expelled through the cannula at a rate of 2 μl/min into a dish with cell culture media consisting of DMEM/F12, N2 supplement, NEAA, heparin, SHH (20 ng/ml), and FGF8b (100 ng/ml). Evaluation of cell survival at every time point by cell counting using a hemocytometer and trypan blue dye did not show differences between samples compared to controls (one-way ANOVA with post hoc Bonferroni's multiple comparison test, p = 0.8412) (Fig. 1G), demonstrating that the cells could remain at room temperature in the system up to 2 h before injecting without significant cell death.

To assess differentiation capability of the neurospheres formed by DA progenitors, two of the cell collections were exposed to our differentiation paradigm for 2 weeks and then fixed with 4% PFA. Immunohistochemical staining confirmed that the cells had maintained their DA neural identity given by coexpression of βIII-tubulin and TH (Fig. 1H and I). Most of the cells became neurons (above 95%) with about 30% TH⁺ neurons in two trials of the control and 120-min uploaded groups.

The infusion trials at 0 min revealed small in-line pressure variations associated with neurosphere inoculation. In comparison, during the first 30-min trial, a significant increase followed by a drop in the line pressure was noted. This matched the time of expulsion of the first cells, indicating possible occlusion of the cannula (Fig. 2A). As the occlusion seemed to be caused by dried-out cells, for the second 30-min trial, the cannula tip was placed in aCSF after cell loading. Submersion of the cannula tip in aCSF prevented this problem (Fig. 2B) and therefore was performed for all subsequent trials.

Cell Inoculation in Gel as Brain Surrogate

To assess the cell infusions in a brain-like media, two cannula tract injections were performed in different aga-rose gel-filled boxes (Fig. 3A-C). For each of the two trials, the loading lines were first primed with PBS, then 40 μl of aCSF was uploaded at 20 μl/min, followed by a suspension containing approximately 2,000,000 cells loaded in four 1-min periods at a rate of 10 μl/min, for a total of 40 μl. The loaded cannula and line were threaded through the remote introducer and a cannula adapter insert of the gel box top and fastened to the remote introducer. Using the remote introducer, the cannula was slowly inserted into the gel and lowered 30 mm.

For the first trial (Fig. 3B), three deposits of 15 μl each were infused through the cannula at a rate of 2 μl/min. After the first deposit was completed, the cannula was retracted 2 mm, where a second deposit was infused, and then retracted 2 mm more for the placement of the third deposit. For the second trial (Fig. 3C), two deposits of 15 and 20 μl, respectively, were infused through the cannula at a rate of 3 μl/min. After the first deposit was completed, the cannula was retracted 2 mm, where a second deposit was placed. In both trials, 5 min after completing the last deposit, the cannula was slowly retracted out of the gel with the remote introducer. A spike in pressure was observed at the beginning of each initial infusion for both gel trials, probably due to coring of the gel (Fig. 3D). Afterward, similar to the inoculations in a dish, minimal changes in line pressure were observed with each subsequent deposit. Cell distribution matched the pattern of deposits across the cannula tract and was confined to the immediate region (maximal distribution of 1.5 and 2.0 mm at the widest point for gels 1 and 2, respectively). After the postinfusion 5-min wait period, the cannula was retracted, and a few cells were dragged upward into the cannula tract.

Nonhuman Primate Intracerebral Infusion Validation

To validate the system in vivo, two intracerebral infusions were performed in a rhesus monkey. RT-IMRI guidance of the cannula was performed using a pivot point-based MRI-compatible external trajectory guide system shown here attached to a gel box ⁹ (Fig. 4A). Pre-operative planning for base placement was completed for the animal after a 3D T1W MRI to identify the entry point of the cannula. This process defined the placement of the trajectory guide base and developed a roadmap of the trajectory that the cannula should follow. Efforts were made to maximize perpendicular entry to the cortical surface to prevent deflection of the cannula, as well as to avoid deep sulci that contain larger blood vessels. A cone of projection depicting angular limits was used to assist placement of the base on the skull, shown here retroactively placed on the skull of the transplanted monkey's 3D T1W MRI (Fig. 4B). Placement of the system base was performed under sterile conditions using a custom-built MRI-guided real-time prospective stereotaxic surgery platform. A laser calibration rod attached to the stereotaxic micromanipulator and inserted in the trajectory guide stem was used to place the base, which was mounted to the skull over the craniotomy with three small self-tapping screws (Fig. 4C).

Before each of the infusions, the neurospheres were uploaded into the system. Similar to the gel tests, the loading lines were first primed with sterile injectable saline, and then 40 μl of aCSF was uploaded at a rate of 20 μl/min, followed by a neurosphere suspension containing approximately 2,000,000 cells loaded in four 1-min periods at a rate of 10 μl/min, for a total of 40 μl. Approximately 60 min elapsed between cell uploading and inoculation, due to different issues with targeting and surgery. Changes in line pressure were observed during the course of the infusion that resembled the pres sure pattern recorded during in vitro inoculations (Fig. 5). A spike in pressure occurred at the beginning of the first infusion for each needle tract, most likely due to coring of brain tissue. Although 40 μl of cell suspension was uploaded, an additional 5 μl was injected (total of 45 μl) to ensure that any cells lagging in the cannula were expelled.

OPEN IN VIEWER

Figure 5.

Pressure graphs of in vivo infusions during the first deposit at sites 1 (A) and 2 (B). Note the spike in pressure at the beginning of the infusion.

Postmortem analysis of Nissl-stained brain coronal sections (2 weeks after grafting) corroborated needle tract positioning and targeting the commissural and post-commisural putamen nucleus, respectively (Fig. 6), and hypercellularity in grafted areas with preserved host neuroanatomy, consistent with previous reports (Fig. 7A) ¹⁶ . Similar to the gel trials, the distribution seemed to be confined to the immediate region of injection. Fluorescence microscopy confirmed the presence of Hoechst-labeled grafted cells, the majority of which were nestin⁺ (Fig. 7B). Stereological quantification of Hoechst⁺ cells was difficult as the graft had a high number of labeled nuclei that overlapped as regions of solid blue; approximately 1,600,000 individual nuclei were identified, although this is probably a very low estimate of grafted cells.

OPEN IN VIEWER

Figure 6.

Sequential coronal brain sections of tissue (blue-colored images) and T1W MRI obtained after completion of the second IMRI cell infusion (black and white images). Arrows in Nissl-stained sections point to grafts, which are observed as areas of hyper-cellularity (dense Nissl areas). In the MRI images, the darker gray areas within the putamen correspond to the infusion cloud. In the cerebral cortex and part of the white matter, the infusion cannula can be seen as a dark line. Note the similar distribution of the infusate in the in vivo and postmortem images. Scale bar: 15 mm.

OPEN IN VIEWER

Figure 7.

(A) Nissl staining of the graft site shows hypercellularity. (B) Immunofluorescence of graft 2 weeks after surgery. The injected cells were neuroprogenitor positive for nestin (red) and labeled with Hoechst nuclear marker (blue). Scale bar: 100 μm.

To evaluate the host immune response within the graft site, CD68, CD45, and GFAP were analyzed. There was significant infiltration of CD68⁺ cells within the graft site (Fig. 8A). There were no observable CD45-labeled cells within the graft site (Fig. 8B), although a few could be seen confined to the vasculature. GFAP⁺ astrocytes were present almost exclusively outside or along the border of the graft (Fig. 8C), some of them extending processes into the graft. (Fig. 8C, inset).

OPEN IN VIEWER

Figure 8.

Host immune response evaluated by the expression of CD68 (A), CD45 (B), and GFAP (C). Numerous CD68-immunoreactive (-ir) cells infiltrated the graft after 2 weeks, while no CD45-ir cells were visible. GFAP-ir astrocytes were observed in the host tissue with some process extension into the host/graft interface. Scale bar: 100 μm (A-C), 25 μm (a-c).