Intravitreal Injections of Neurotrophic Factors Secreting Mesenchymal Stem Cells Are Neuroprotective in Rat Eyes following Optic Nerve Transection

Abstract

Purpose.: To evaluate the neuroprotective effect of intravitreal injections of neurotrophic factors secreting mesenchymal stem cells (NTF-SCs) on the survival of retinal ganglion cells (RGCs) in rat eyes after optic nerve transection (ONT).

Methods.: Rat and human bone marrow–derived mesenchymal stem cells (MSCs) were induced to secrete high levels of NTF. The neuroprotective effect from intravitreally injected untreated MSCs or NTF-SCs was compared with that from PBS injections using an ONT model in 146 rats. RGCs were labeled by applying rhodamine dextran to the orbital optic nerve or by injecting Fluorogold into the superior colliculus. Cell- and saline-treated eyes were compared 8 days after ONT. For tracking, MSCs and NTF-SCs were labeled with PKH26 and analyzed at 2 hours and at 10, 17, and 24 days using immunohistochemistry and RT-PCR.

Results.: Mean RGC survival at 8 days after transection increased significantly after intravitreal injections of human NTF-SCs (69% ± 3%) or of untreated human MSCs (66% ± 5%) versus PBS (46% ± 3%; P = 0.0005 and P = 0.03, respectively). In an additional set of experiments, human NTF-SCs versus PBS were significantly neuroprotective, but bone marrow–derived rat NTF-SCs were not (P = 0.001 and P = 0.1, respectively). Immunohistochemistry demonstrated that human-derived MSCs, human NTF-SCs, and rat-derived NTF-SCs survived at least 24 days after intravitreal injection.

Conclusions.: Bone marrow–derived MSCs can deliver NTFs by intravitreal injection and can be neuroprotective after ONT. This approach might be further studied to deliver NTFs by autotransplantation in glaucomatous eyes.

Optic neuropathies, including glaucoma, are characterized by progressive degeneration of retinal ganglion cells (RGCs). This characteristic process makes this group of diseases suitable candidates for cellular therapy. ¹ Stem cell-based treatment had become a promising frontier for treating neurodegenerative diseases, thus making it a potential therapeutic approach for optic neuropathies as well. ^{1 –8} There are several candidate cells for use in stem cell therapy to the retina, including embryonal stem cells, oligodendrocyte precursor cells, human Müller stem cells, and bone marrow–derived stem cells. ^{9 –13} Bone marrow–derived mesenchymal stem cells (MSCs) are self-renewing precursors that reside in the bone marrow. They are best characterized by their capacity to differentiate into connective tissue cell types but have also been reported to exhibit the ability to differentiate in vitro and in vivo into nonmesodermal cell types, such as neural-like cells. ^{14 –17} Numerous studies have evaluated the efficacy of MSCs to promote repair and regeneration of nervous tissue within the central and peripheral nervous systems, although the exact mechanisms responsible for a beneficial outcome remain indeterminate. ^{18 –24} Possible effects include cell replacement, trophic factor delivery, and immunomodulation. ²⁵ Previous reports have suggested that the adult retina does not provide a permissive environment in which transplanted stem cells can easily migrate and integrate. ^3,26,27 In addition, replacement of lost RGCs will require the construction of complex axonal connections within the retina and to the brain. As such, the trophic factor delivery characteristic of MSCs may present a more practical approach and may be beneficial as supportive therapy for dying RGCs. According to this approach, MSC effectiveness may be credited to the production of neurotrophic factors that support neuronal cell survival, induce endogenous cell proliferation, and promote nerve fiber regeneration at sites of injury. Indeed, mounting evidence suggests that supplemental therapy with neurotrophic factors either by intravitreal injections or by gene therapy is neuroprotective in models of optic nerve injuries, including optic nerve transection (ONT) and glaucoma. ^{28 –34} To take this approach one step forward, we used a novel approach that we designed for inducing the differentiation of human and rat MSCs toward NTF-SCs to deliver neurotrophic factors by transplantation of stem cells. 

NTF-SCs produce and secrete high levels of brain-derived neurotrophic factor (BDNF), glial-cell-derived neurotrophic factor (GDNF), and vascular endothelial growth factor (VEGF) compared with MSC. ELISA showed that untreated MSCs secreted low levels of BDNF and GDNF but, after 6 days of induction, the NTF-SCs secreted more than five times more BDNF and more than twice the amount of GDNF. ³⁵ NTF-SC was found to be neuroprotective in several models of brain injury, including the 6-hydroxydopamine (6-OHDA)–induced lesion, a rat model for Parkinson's disease, and were also shown to migrate toward a quinolinic acid–induced striatal lesion model for Huntington's disease. ^{20,35 –37}  

In this study, we evaluated the survival, function, and neuroprotective effect of intravitreal injections of MSCs and NTF-SCs in a rat model of ONT. 

Materials and Methods

Animals

Adult Wistar rats (weight range, 375–425 g) were used in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research in protocols approved and monitored by the Animal Care Committee of the Tel Aviv University School of Medicine. The animals were housed with a 14-hour light/10-hour dark cycle with standard chow and water ad libitum. 

Study Design

The present study includes several sets of experiments, each consisting of an intravitreal injection of stem cells or of phosphate-buffered saline (PBS) before the performance of ONT. The experiment for the neuroprotection study included unilateral intravitreal injections of NTF-SCs, MSCs, or PBS (controls) followed by ONT 3 days later. Eyes were enucleated 8 or 9 days after transection (11 days after injection), and the number of surviving RGCs was counted. For cell tracking purposes, human MSCs, human NTF-SCs, and rat NTF-SCs were prelabeled with PKH26 fluorescent dye (Sigma-Aldrich, St. Louis, MO) and were transplanted as previously described. ONT was performed, and surviving stem cells were counted. In addition, the survival and function of human NTF-SCs were analyzed from 2 hours to 24 days after transection (n = 4 rats for each of the four time points) using immunohistochemistry for neurotrophic factors and polymerase chain reaction (PCR). The contralateral eyes served as controls. All our experiments were performed under masked conditions. 

MSC Culture and Treatment

The production and culture of rat-derived ³⁶ or human-derived ³⁵ MSCs were described previously. Cells were grown in polystyrene plastic flasks (Corning, Corning, NY) in a growth medium consisting of Dulbecco's modified Eagle's medium (DMEM) supplemented with 15% fetal calf serum, 2 mM l-glutamine, 100 μg/mL streptomycin, 100 U/mL penicillin, 12.5 U/mL nystatin (SPN) (all from Biological Industries, Kibbutz Beit-Haemek, Israel), and 0.001% 2- mercaptoethanol (Sigma) for rat-derived cells. Human-derived cells were grown in a medium consisting of DMEM supplemented with SPN, glutamine, and heparin and 5% platelet-rich plasma. MSC cultures were passaged approximately every 2 weeks. The medium was changed twice a week, and the cells were maintained at 37°C in a humidified 5% CO₂ incubator. Experiments on the cells were performed after two to seven passages. 

Induction of MSCs into NTF-SCs

Human- or rat-derived MSCs were first placed in DMEM supplemented with SPN, 2 mM l-glutamine, 20 ng/mL human epidermal growth factor (hEGF), 20 ng/mL human basic fibroblast growth factor (hbFGF; R&D Systems, Minneapolis, MN), and N2 supplement (Invitrogen, Carlsbad, CA). After 72 hours, the medium was replaced with DMEM supplemented with 1 mM dibutyryl cyclic AMP (dbcAMP), 0.5 mM isobutylmethylxanthine (Sigma-Aldrich), 5 ng/mL human platelet-derived growth factor (PDGF), 50 ng/mL human neuregulin 1-β1/ HRG1-β1 EGF domain, and 20 ng/mL hbFGF (all from R&D Systems) for 3 more days. On the last day of this induction protocol, the cells were trypsinized, washed in PBS, and counted. They were then suspended for transplantation in PBS in a concentration of 400,000 cells/5 μL. 

ELISA-Based Measurements of NTF Secretion

At the end of the NTF-SC induction process, the cell culture supernatant was measured for human GDNF and BDNF concentrations by a sandwich ELISA procedure according to the manufacturer's instructions (DuoSet; R&D Systems). Absorbance at 450 nm and 570 nm was recorded on a microplate reader (Multiskan MS; Labsystems, Thermo Scientific, Waltham, MA). The results were calculated for one million cells. Each result represents three different cultures, each tested in duplicate. 

Intravitreal Injections

Intravitreal injections of 400,000/5 μL NTF-SCs, untreated MSCs, or PBS were performed randomly and unilaterally. Animals were anesthetized with intraperitoneal (IP) ketamine (50 mg/kg) and xylazine (5 mg/kg). Guided by an operating microscope, the cells or PBS were injected in the superotemporal area posterior to the lens using a metal needle connected to a glass syringe (Hamilton Co., Reno, NV), taking care to avoid penetration of the lens or damage to the vortex veins. The injection site was 3 mm posterior to the superotemporal limbus. Injections were given slowly (i.e., over 1 minute). The injection site was visualized with a standard indirect ophthalmoscope and a 90-D condensing lens (Volk Optical Inc., Mentor, OH) after each injection. 

Immunosuppression Treatment

Subcutaneous injections of 10 mg/kg cyclosporine (Sandimmune; Sandoz, Princeton, NJ) were given daily in one set of experiments to induce immunosuppression. Animals were observed for any side effects and weight changes. 

Optic Nerve Transection

ONT was performed 3 days after the intravitreal injections and 13 days after labeling with Fluorogold. The surgical procedure was performed unilaterally under anesthesia with intraperitoneal ketamine (50 mg/kg) and xylazine (5 mg/kg) and topical proparacaine 1% eye drops. The optic nerve was transected by a diamond knife 3 to 4 mm behind the globe, with care taken not to interfere with the blood supply. The retinas were examined with an ophthalmoscope to ensure blood vessel patency. 

RT-PCR

For RNA production, enucleated eyes were immediately frozen in liquid hydrogen and stored at −80°C until they were used. Total RNA was isolated from the enucleated eyes using a commercial reagent (TriReagent; Sigma-Aldrich). The amount and quality of RNA were determined spectrophotometrically using a spectrophotometer (ND-1000; NanoDrop, Wilmington, DE). First-strand cDNA synthesis was carried out from 1 μg of the total RNA using random primer (Invitrogen) and reverse transcriptase (SuperScript II RT; Invitrogen). The reverse transcription reaction was performed at 25°C for 10 minutes, 42°C for 120 minutes, 70°C for 15 minutes, and 95°C for 5 minutes 

PCR amplifications were performed in a 25 μL final volume containing 2 μL reverse-transcribed RNA (cDNA), 0.5 μM sense and antisense primers, and 1× master mix (GoTaq Green; Promega, Madison, WI). Primers were chosen to specifically identify human-derived BDNFs according to known databases (Basic Local Alignment Search Tool, National Center for Biotechnology Information, http://blast.ncbi.nlm.nih.gov/Blast.cgi): upstream ACTTCGGGAAGGTGTCTGGACATT and downstream GTGTGAAATGGGCTGAATGGGCTT (size, 271 bp). The housekeeping rat gene GAPDH was tested with the primers TGCACCACCAACTGCTTA and GGATGCAGGGATGATGTTC. PCR conditions were 94°C for 5 minutes, 94°C for 30 seconds, 62°C for 30 seconds, and 72°C for 45 seconds. The annealing temperature was reduced by 2°C every six cycles for a total of 18 cycles, following 16 cycles at 56°C. The PCR reaction was resolved on a 1% agarose gel. The bands were observed under ultraviolet light and photographed. 

Immunohistochemistry

Eyes were enucleated and cryopreserved in sucrose/OCT (Sakura Finetek Inc., Torrance, CA). They were then cryosectioned into 10-μm thick sections. Blocking was performed in PBS containing 5% normal goat serum (Biological Industries), 1% bovine serum albumin (Sigma-Aldrich), and 0.05% Triton. Primary antibody was dissolved in blocking solution diluted 1:1 with PBS and incubated over the tissue section overnight at 4°C. The primary antibodies used were rabbit anti-BDNF (1:500; Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-GDNF (1:500; Santa Cruz), and mouse anti-human nuclei (1:200; Chemicon, Temecula, CA). For sections stained for human nuclei, an antigen unmasking process was performed by microwave boiling in citric buffer. After primary antibody incubations, the sections were incubated for 1 hour with a secondary conjugated anti-mouse antibody (1:500; Molecular Probes). For the rest of the stains, the sections were treated with biotin-conjugated antibody for 1 hour (ready to use for mouse [Zymed] and 1:500 for rabbit [Invitrogen]) after 1 hour of incubation with streptavidin (1:500; Invitrogen). The sections were then stained with 4′,6-diamidino-2-phenylindole (DAPI, 1:500; Sigma-Aldrich) and viewed with a fluorocentric microscope (Olympus Optical Co., Tokyo, Japan). To validate double staining of a specific protein within the PKH-26–positive cells, several sections from day 24 after transplantation were also analyzed by confocal microscopy (FluoView [Olympus], 60× magnification; green laser: excitation 488, emission 510–530; red laser: excitation 543, emission 660 and up). 

Negative controls included nonimmune serum of the same species as the primary antibody at the same protein concentration and incubation buffer alone. Images of representative slides were captured digitally by a standardized microscope and camera settings (Olympus Optical Co.). 

Stem Cell Survival

The number of human MSCs, human NTF-SCs, or rat NTF-SCs (n = 4 for each cell type) were counted 24 days after injection. The cells were prelabeled with PKH-26, and then cryopreserved sections were stained with 4′,6-diamidino-2-phenylindole (DAPI, 1:1000; Sigma-Aldrich) and viewed with a fluorocentric microscope (Olympus). All cell-containing fields were photographed. Cells were counted in a masked fashion using ImageJ software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html) software (n = 10 slides/eye equal to 10% of all slides). 

Labeling and Counting of RGCs

Labeling of RGCs with Fluorogold.

Ten days before intravitreal injections and 13 days before ON transection, RGCs were labeled by applying fluorescent dye (Fluorogold; Fluorochrome, Denver, CO) to the superior colliculus by bilateral stereotactic injections. Eight days after transection, the animals were anesthetized and killed, their eyes were enucleated, and retinal whole mounts from both eyes were placed on slides. 

RGC Labeling with Rhodamine Dextran.

RGCs were labeled by backfilling with rhodamine dextran (Molecular Probes, Eugene, OR) applied to the orbital optic nerve 1 day before kill. The rats were anesthetized as described, and ONT was performed unilaterally in the experimental eye 1 to 2 mm behind the globe (i.e., closer to the globe than the initial ONT). Twenty-four hours later, the eyes were enucleated, and the retinas were prepared as whole mounts. The rats were killed by intraperitoneal injection of pentobarbital (60 mg/mL). The surgeon performing all the procedures was masked to treatment assignment groups. 

RGC Counting Procedure.

Retinal whole mounts were viewed with a fluorescence microscope (BX51; Olympus) and an appropriate filter to identify rhodamine dextran- or Fluorogold-labeled cells. Labeled RGCs were counted with a 40× super wide-field objective along two radii in four directions (i.e., superior, temporal, inferior, and nasal areas) centered on the position of the optic nerve head. Four fields were counted along each radius, yielding a total of 32 fields per retina. An experienced observer masked to the procedure and to the designated treatment counted the RGCs. RGCs were identified by their Fluorogold labeling and morphology. 

For each rat, the percentage of surviving RGCs was calculated as a ratio of RGCs in the experimental eye compared with the contralateral control eye. 

Statistical Analysis

The area of each field in our microscope is 0.134 mm², yielding a total counted area of 4.3 mm², which is a 8.6% sample of the average 50.1 mm² Wistar rat retina. Data are presented as mean ± SEM. The neuroprotective effect of NTF-SCs was evaluated by comparing the number of surviving RGCs in the NFC-SC–treated group to the number of surviving RGCs in the MSC- and PBS-treated groups using analysis of variance (ANOVA), with a significance level set at P < 0.05. Post hoc analysis included Tukey HSD, Bonferroni, and Scheffé tests. Statistical calculations were performed using SPSS software (version 13; SPSS, Chicago, IL; http://www.spss.com). 

Results

This study included a total of 146 rats; 108 were killed for evaluation of RGC survival within different treatment groups, and 38 were killed for RT-PCR and immunohistochemistry evaluation of stem cell survival and function in the eye. 

Human-Derived NTF-SCs Secrete More BDNF and GDNF than Rat-Derived Cells

In ELISA-based quantitative measurements, we found that rat-derived NTF-SCs secreted very low levels of GDNF (Fig. 1A) that did not change after induction (45.99 ± 8.47 vs. 35.15 ± 34.75 pg/million cells for untreated MSCs). Human-derived MSCs secreted GDNF at similar levels (40.92 ± 40.92 pg/million cells); however, NTF induction increased it more than eightfold, to 348.61 ± 46.04 pg/million cells for untreated MSCs. The NTF induction protocol increased BDNF levels in both human- and rat-derived cells (101.93 ± 16.23 vs. 6573.02 ± 1312.84 for rat-derived cells and 1610.94 ± 264.97 vs. 10,126.24 ± 664.87 for human-derived cells; Fig. 1B). Interestingly, in a post hoc analysis for the ANOVA trial, we found that human-derived NTFs are different in a statistically significant manner from all other groups. 

Figure 1.

 

Human-derived NTF-SCs secreted more BDNF and GDNF than did rat-derived cells. (A) In a quantitative ELISA measurement, we found an eightfold increase in GDNF secretion levels in the supernatant of human-derived NTF-SCs, whereas rat-derived cells secreted very low levels of GDNF. (B) Although rat-derived NTF-SCs secreted large amounts of BDNF, those amounts were statistically significantly lower than in human cells (*P < 0.05 and **P < 0.01 compared with untreated MSCs of each strain).

Assessment of Stem Cell Survival and Function in the Eye

The survival of human MSCs, rat NTF-SCs, and human NTF-SCs after ONT was evaluated after injection and ONT. The cells were usually found in clusters in the vitreous cavity between the retina and the lens (Fig. 2). Histologic studies confirmed the presence of transplanted cells in the eye. The mean total cell count/10 slides 24 days after injection was 140 ± 30 for human MSCs, 210 ± 48 for rat NTF-SCs, and 418 ± 100 for human NTF-SCs (ANOVA, P = 0.08). 

Figure 2.

 

Human NTF-SCs in the vitreous cavity 10 days after intravitreal injection. (A–C) NTF-SCs were prelabeled with the red fluorescent dye PKH-26 and were located between the retina and the lens. The retina was stained with Hoechst staining. Magnification, 10×.

Human NTF-SCs prelabeled with PKH-26 showed positive staining to BDNF, GDNF, and human nuclei antigens within the surviving cells (Fig. 3) at all time points, as evidenced by double staining and colocalization of PKH-26, DAPI and BDNF, GDNF, or human nuclei. The presence of human BDNF mRNA was evident by RT-PCR. The specific BDNF transcript was found in the NTF-SC–transplanted eyes at all time points, whereas there was no expression in the PBS-injected rats (Fig. 4). The housekeeping gene GAPDH was positive for all injected eyes and controls. The presence of human mRNA 24 days after intravitreal injection indicated the survival of the transplanted human NTF-SCs. 

Figure 3.

 

Human NTF-SCs survived in the tissue and maintained their phenotype for at least 24 days after transplantation, as shown by immunohistochemistry. (A–D) Colocalization of human nuclei antigen with PKH-26 fluorescent marker. (E–H) Colocalization of BDNF with PKH-26 fluorescent marker. (I–L) Colocalization of GDNF with PKH-26 fluorescent marker. (M–P) Colocalization of CD68 (ED1) with PKH-26 fluorescent marker. (Q) Colocalization of BDNF and PKH-26 as shown by confocal microscopy. (R) Colocalization of GDNF and PKH-26 as shown by confocal microscopy. Scale bar, 100 μm.

Figure 4.

 

NTF-SCs survive in the tissue and maintain their phenotype for at least 24 days after transplantation, as shown by RT-PCR. Specific primers were used to detect RNA expression for BDNF at different times up to 24 days. The test was performed in duplicate at all four time points.

Neuroprotective Effect of Transplantation of Bone Marrow–Derived Human NTF-SCs

Eight days after transection of the optic nerve, treatment with intravitreal injections of bone marrow–derived human NTF-SCs or bone marrow–derived human MSCs was significantly neuroprotective compared with intravitreal injections of PBS (P = 0.0005 and P = 0.03, respectively). The mean RGC survival with intravitreal injections of 400,000/5 μL bone marrow–derived human NTF-SCs was 69% ± 3% compared with 66% ± 5% for bone marrow–derived human MSCs and 46% ± 3% in the PBS-injected group (Fig. 5). These results suggest that both types of MSCs show a neuroprotective effect. The two MSCs groups were not statistically different. 

Figure 5.

 

The neuroprotective effect of bone marrow–derived human MSCs and NTF-SCs 8 days after ONT. Mean RGC survival rates with intravitreal injection of 400,000/5 μL bone marrow–derived human NTF-SCs were 69% ± 3% compared with 66% ± 5% with bone marrow–derived human MSCs and 46% ± 3% for the saline-injected group (P = 0.0005 and P = 0.03, respectively). *P < 0.5.

Neuroprotective Effect of Bone Marrow–Derived Rat NTF-SCs

To overcome a potential immunologic response against human-originated cells, we investigated the neuroprotective effect of bone-marrow–derived rat NTF-SCs and compared it with human NTF-SCs in an additional set of experiments. As we had found before, intravitreal injections of 400,000/5 μL bone marrow–derived human NTF-SCs were significantly neuroprotective compared with intravitreal injections of PBS 9 days after ONT (P = 0.001). The effect of injections of bone marrow–derived rat NTF-SCs was not significantly neuroprotective (P = 0.1). The mean RGC survival with intravitreal injections of 400,000/5 μL bone marrow–derived human NTF-SCs was 48% ± 4% compared with 35% ± 6% with rat NTF-SCs and 22% ± 2% with PBS (Fig. 6). 

Figure 6.

 

The neuroprotective effect of bone marrow–derived rat NTF-SCs and human NTF-SCs at 9 days after ONT. Intravitreal injections of bone marrow–derived human NTF-SCs were significantly neuroprotective compared with intravitreal injections of saline (P = 0.001) and more neuroprotective than injections of bone marrow–derived rat NTF-SCs (P = 0.1).

The higher rate of RGC death in this set of experiments compared with the first one could be attributed to the fact that the eyes were enucleated 9 days after transection compared with 8 days in the former experiment. In addition, it should be considered that differences between in vivo experiments are known to be not uncommon. 

Effect of Immunosuppression on the Neuroprotective Effect of NTF-SCs

Systemic cyclosporine treatment was used in an additional set of experiments to eliminate any potential immunologic and inflammatory reactions. Subcutaneous injections of 10 mg/kg cyclosporine (Sandimmune; Novartis, Basel, Switzerland) were given daily to induce immunosuppression. In this additional experiment, we found that the mean RGC survival with intravitreal injections of 400,000/5 μL bone marrow–derived human NTF-SCs was 45% ± 3% compared with 38% ± 3% for PBS (P = 0.017). The mean RGC survival with intravitreal injections of bone marrow–derived human MSCs was 38% ± 4% (P = 0.97 compared with PBS; Fig. 7). Unfortunately, the rats experienced severe side effects, including weight loss, apathy, and behavioral changes that could have affected the results of this experiment and interfered with the neuroprotective effect. 

Figure 7.

 

With systemic immunosuppression, bone marrow–derived human NTF-SCs were more neuroprotective than bone marrow–derived human MSCs and saline injection. The neuroprotective effect was impaired, however, because of severe adverse effects associated with immunosuppression.

Discussion

The results of this study suggest that intravitreal injection of human MSCs and NTF-SCs effectively rescued RGCs in a model of ONT in rats, indicating that the increased secretion of neurotrophic factors by these cells has significant importance in preventing RGC death and in protecting the injured optic nerve. Human NTF-SCs secrete higher levels of neurotrophic factors than do human MSCs. The fact that both types of cells were neuroprotective suggests that even small amounts of different types of neurotrophic factors can be neuroprotective. However, the lack of significant difference between the effects of the two types of cells suggests that in a severe injury such as ONT, delivering higher amounts of different neurotrophic factors is not sufficient to prevent death. 

NTF-SCs survived up to 24 days after intravitreal injection without any additional immunosuppression treatment. Previous studies using MSCs in a model of glaucoma found viable cells 4 weeks after transplantation. ³⁸ Another report, ²⁷ however, found that MSCs from umbilical cord blood can survive in the vitreous up to 1 week and up to 3 weeks with immunosuppression. We observed no indication that the induction protocol used in the present study interfered with cell survival or function, implying that NTF-SCs are suitable and as good as MSCs for intravitreal injections. 

Interestingly, human NTF-SCs achieved a better neuroprotective effect than rat NTF-SCs. Indeed, human-derived cells secreted more neurotrophic factors than did rat-derived cells. However, we cannot rule out the possibility that some inflammatory reaction occurred in response to human cell transplantation that could have contributed to the protective effect. On the other hand, the immunomodulating and immunosuppressive properties MSCs are believed to have may play a critical role in their therapeutic potential. 

The therapeutic potential of different types of stem cells for treating retinal and optic nerve diseases has already been investigated. ^10,11,38,39 Most studies confirmed that the adult retina has limited levels of graft-host integration, whereas only few studies demonstrated successful integration of stem cells to the retina. ⁴⁰ MacLaren et al. ⁴¹ showed that adult degenerating mammalian retinas can effectively incorporate rod photoreceptor precursor cells into the outer nuclear layer and that these cells form functional synaptic connections and even contribute to visual function. Those authors found that the success of this treatment depends mainly on the optimal ontogenetic stage of the donor cells. Qiu and Selier ⁴² were able to enhance photoreceptor cell differentiation and integration of retinal progenitor cells after subretinal transplantation into retinal degenerate rats by optimization of isolation, expansion, and transplantation procedures. Arnold et al. ⁴³ found that mesenchymal stem cells can prolong photoreceptor survival in the rhodopsin knockout mouse, also providing evidence of a therapeutic benefit in retinitis pigmentosa. The situation was not better for inner retinal diseases. Previous reports found that the inner retina, especially the inner limiting membrane, obstructs the integration of intravitreally injected cells into the retina. Martin et al. ² observed minimal retinal integration (<1% of cells) when MIO-M1 stem cells or oligodendrocytes precursor cells (OPCs) were transplanted. Bull et al. ¹¹ found that the Müller stem cell line MIO-M1 in glaucomatous eyes produced cells that expressed neuronal and glial cell markers but that the retina was relatively resistant to transplant integration and long-term xenograft survival was limited. Interestingly, local modulation of the retinal environment enhanced the integration of MIO-M1 cells into the glaucomatous retina. ¹¹ On the other hand, when OPCswere used, a significant neuroprotective effect was achieved. This effect did not appear to be contact mediated or to be conferred by the myelination of naked axons; rather, it was most likely due to the release of diffusible neurotrophic factors by activated OPCs. ¹⁰  

All these data suggest that it may be more advantageous to use stem cells as vectors that secrete and deliver neurotrophic factors. At the same time, efforts should be expended to find ways to improve the integration of stem cells in the RGC layer, to facilitate their differentiation into RGCs, and to induce axonal sprouting to form neuronal connections within the retina and to the brain. 

One of the main advantages of MSC and NTF-SC therapy is the option of autologous transplantation using the patient's own bone marrow–derived stem cells. This approach entirely circumvents the problem of immune rejection, does not cause the formation of teratomas, and is free of ethical or political concerns. This novel approach is a safe and efficacious method of generating NTFs in the eye and may be neuroprotective in glaucomatous eyes. 

Footnotes

 Supported in part by the Norma and Alan Aufzien Chair for Research of Parkinson's Disease, Tel Aviv University.

Footnotes

 Disclosure: H. Levkovitch-Verbin, None; O. Sadan, None; S. Vander, None; M. Rosner, None; Y. Barhum, None; E. Melamed, P; D. Offen, P; S. Melamed, None