Retinal stem cells (RSCs) are promising in cell replacement strategies for retinal diseases. RSCs can migrate, differentiate, and integrate into retina. However, RSCs transplantation needs an adequate support; chitosan membrane (ChM) could be one, which can carry RSCs with high feasibility to support their integration into retina. RSCs were isolated, evaluated for phenotype, and subsequently grown on sterilized ChM and polystyrene surface for 8 hours, 1, 4, and 11 days for analysing cell adhesion, proliferation, viability, and phenotype. Isolated RSCs expressed GFAP, PKC, isolectin, recoverin, RPE65, PAX-6, cytokeratin 8/18, and nestin proteins. They adhered (28 ± 16%, 8 hours) and proliferated (40 ± 20 cells/field, day 1 and 244 ± 100 cells/field, day 4) significantly low
Retina is exposed over life to degenerative conditions. This leads into retinal dystrophies, followed by retinal diseases, and ultimately produces visual impairment [
From past few years, identification and characterization of stem cells of different origin have opened new avenues in cell replacement therapy [
Cell transplantation, cell integration in tissue, and its proper function are still open issues of research. Different types of stem cells such as RSCs, neural stem cells (NSCs), bone marrow derived stem cells (BMSCs), and embryonic stem cells (ESCs) have achieved partial success in retinal transplantation studies [
Chitosan, poly[
RSCs differentiation potential and chitosan characteristics encouraged us to investigate the feasibility of chitosan membrane (ChM) application in delivering retinal stem cells into retina.
RSCs were isolated from porcine eyes following already published papers [
ChMs were prepared and provided by INA-University of Zaragoza, Spain, for evaluating RSCs growth, viability, and characteristics. In brief, chitosan (Aldrich, high molecular weight) was dissolved in a 2 wt% acetic acid (Alfa Aesar, glacial) aqueous solution by stirring for 24 hours at 80°C. After filtration, 2.3 mL of chitosan 1 wt% solution was cast on PS Petri dishes and was evaporated at room temperature (RT) for 2 days. The 10
Each experiment was performed in 8-well chamber slides (8 mm × 8 mm), of which four were covered with ChM pieces of size 6 mm × 6 mm. The remaining four wells were used as a control (slide’s polystyrene surface). Chitosan has the property to swell in wet conditions. Therefore, few drops of medium were poured into each well before inserting a ChM piece in a well. When a ChM piece swelled, it covered almost whole area of a well (8 mm × 8 mm). Following insertion of ChM, each chamber slide was sterilized by overnight UV exposure and subsequently 2-hour incubation with medium containing antibiotics. RSCs (20,000 cells/well) were seeded in complete RSCs culture medium and incubated for 8 hours, 1 day, 4, and 11 days in standard culture conditions. At each time point, cells were analyzed for cell adhesion, proliferation, and viability, and for expression of different proteins using a viability/cytotoxicity assay kit (Biotium Inc., USA) and immunostaining technique followed by observations in a phase-contrast and fluorescence microscope Leica AF6000 (Leica Microsystems, Mannheim, Germany).
ChM and polystyrene surfaces were washed with PBS at each experimental time point to remove nonadhered cells. Numbers of adhered cells were determined by manual counting using a phase-contrast and fluorescence microscope. Cells were nuclear-stained with DAPI for 2 minutes at RT, mounted in a fluorescent mounting medium (Invitrogen, Paisley, UK), and visualized using a fluorescence microscope. Twenty fields (×10) were photographed at random per substrate. The cells, and their nuclei, contained in each field were counted using Adobe Photoshop Elements software. The mean number of nuclei per field of view (×10) was calculated for each time interval for each treatment and presented as a histogram showing the average nuclear count per field of view ±1 standard deviation (SD) versus time.
To determine the cell attachment to both surfaces, cells were counted after 8 hours. Average number of cells on the polystyrene substrate (positive control) was set to 100% and the average number of cells on the ChM surface was calculated as a percentage of the cells growing on the polystyrene surface for quantifying the percentage of cell adhesion and presenting as a histogram. The following formula was used to analyse the percentage of cell adhesion on ChM:
RSCs viability was evaluated using a cell viability/cytotoxicity assay kit for live and dead cells in accordance with manufacturer protocol at each time point of the experiment. The kit includes two-color fluorescent stains: green fluorescence for live cells and red fluorescence for dead cells using two probes; calcein acetoxymethyl ester (calcein AM) stains live cells green and Ethidium homodimer III (EthD-III) stains dead and damaged cells red. After staining, RSCs were visualized using a fluorescence microscope and were photographed at random per well.
Percentages of cell viability were determined using the following formulas:
RSCs were immunostained with antibodies against GFAP, PKC, isolectin, recoverin, and RPE65 for evaluating the expression of markers of different cells types. RSCs characteristics stability on ChM and polystyrene surfaces was evaluated by immunostaining for detecting the markers of epithelial (panCytokeratin), retinal stem (PAX6), and neural stem (nestin) cells as well as a marker of transdifferentiation towards fibroblast-like cells (alpha-SMA). In brief, cells were washed with PBS (3 × 5 min), fixed with methanol for 10 minutes at −20°C. At this step, the slides can be stored at −20°C in refrigerator. Cells were blocked for 1 hour in antibody blocking buffer (10% normal goat serum in PBS) at RT. Cells were then incubated overnight with different concentration (Table
List of antibodies used in the study.
Molecular marker | Antibody | Source | Working dilution |
---|---|---|---|
Cytokeratin 8/18 | Mouse monoclonal | Abcam, Cambridge, UK | 1 : 100 |
Nestin | Mouse monoclonal | Abcam, Cambridge, UK | 1 : 100 |
PAX6 | Rabbit polyclonal | Covance, Emeryville, CA, USA | 1 : 100 |
Alpha-smooth muscle actin (alpha-SMA) | Mouse monoclonal | Abcam, Cambridge, UK | 1 : 200 |
Glial fibrillary acidic protein (GFAP) | Rabbit polyclonal | DakoCytomation Inc., USA | 1 : 200 |
Isolectin | Mouse monoclonal | Sigma-Aldrich | 1 : 100 |
Protein kinase C, |
Rabbit polyclonal | Santa Cruz Biotechnology, Inc., USA | 1 : 50 |
Recoverin | Rabbit polyclonal | Millipore, CA, USA | 1 : 100 |
RPE65 | Mouse monoclonal | Novus biological, UK | 1 : 100 |
All experiments were repeated three times to check the reproducibility of the trends observed. The data were obtained from repetition of the experiments subjected to the statistical analysis through Microsoft Excel software. Average, standard deviation (SD), and
Ciliary margin isolated RSCs began to form floating cell spheres after one week in standard culture medium (Figure
Morphology and pigmentation of RSC spheres and RSCs in 10x microscope field observation. (a) Free floating RSC sphere at 1 week, (b) RSC spheres of different sizes at day 18. (c) Pigmented and nonpigmented RSCs at day 24. (d) Confluent RSCs layer with epithelial-like to fibroblast-like cell morphology. (a), (b) and (c) are taken in 10x microscopic field but trimmed down to increase the size of image to show clearly cells in neurosphere (a), scale bar (b) and pigments in cells (c).
Immunostaining results showed that RSCs expressed the proteins related to different retinal cell types. Almost all RSCs expressed GFAP (astrocytes, Figure
RSCs cultivated in standard culture medium supplemented with 10% FBS expressed the proteins (green) of different retinal cell types. Expression of protein; (a) GFAP (astrocyte), (b) isolectin (microglial cells), (c) PKC (rod bipolar cells), (d) recoverin (photoreceptor), (e) RPE65 (RPE cells). Rhodamine-phalloidin and DAPI staining showed actin (red) and nucleus (blue). The arrow showed the cells which are not expressing the proteins studied.
Very few RSCs adhered faintly on ChM surface during 4-5 hours. At 8 hours, RSCs adhered significantly less (28%) on ChM than that observed on polystyrene (Figure
Percentage of RSCs adhered on surfaces, ChM and polystyrene, at 8 hours. The data presents the mean number of cells (phase-contrast microscopy as well as nuclear counts of cells, assuming 1 nucleus per cell) per field (×10) attached to ChM surface as a percentage of the control (polystyrene) ±1 SD, at 8 hours. The histogram confirmed that ChM surface is less favourable for RSCs adhesion than polystyrene surface. Single asterisk (*) represents the significant
Average number of RSCs grown on surfaces, ChM and polystyrene, at days 1 and 4 determined using cell viability/cytotoxicity assay kit. The data presents the average number of cells per field (×10) attached to both surfaces ±1 SD, at days 1 and 4. The histogram confirmed that ChM surface is less favourable for RSCs growth than polystyrene surface. Single asterisk (*) represents the significant
Viability and morphology of RSCs on ChM and polystyrene surfaces detected using cell viability/cytotoxicity assay kit. The green fluorescence represents live cells and red fluorescence represents dead cells. (a) RSCs on ChM surface at day 1. (b) RSCs on polystyrene surface at day 1. (c) RSCs on ChM surface at day 4. (d) RSCs on polystyrene surface at day 4. The figures are representative figures of various photos of RSCs grown on both surfaces and subsequently stained for live and dead cells analysis. White arrows show fibroblast-like morphology.
Viability/cytotoxicity assay showed that very few dead cells (1–7) were present in each photo of 10x microscopic field taken for both surfaces at days 1 and 4 (Figure
Viability of RSCs on ChM and polystyrene surfaces at days 1 and 4. Cells were quantified using cell viability/cytotoxicity assay kit on both surfaces. The data are presented as percentage of viable and dead cells per field (×10) ±1 SD following the formulas written in Section
Phase contrast microscopy (data not shown) as well as cell viability/cytotoxicity assay kit showed that, at day 1, RSCs were rounded on ChM surface while RSCs on polystyrene surface began to take fibroblast-like shape (Figures
Immunostaining results showed that RSCs grown on ChM and polystyrene surfaces expressed cytokeratin 8/18, PAX6 as well as nestin proteins at day 11 (Figures
Immunostaining of proteins expressed by RSCs grown on ChM and polystyrene surfaces for 11 days. Expression of protein: cytokeratin 8/18 on ChM (a) and polystyrene (d), nestin on ChM (b) and polystyrene (e), PAX6 on ChM (c) and polystyrene (f), alpha-SMA on polystyrene (h) and ChM (i). Alpha-SMA protein expression of fibroblasts (g). Polystyrene surface and fibroblast were used as controls; (g) is taken in 20x microscopic field and the rest are in 10x microscopic field.
Although adult mammalian retina retains RSCs in quiescent form in
Porcine eye resembles human eye in many properties such as similar size, anatomy, and histology. Furthermore, retinal development in pig eye shows substantial similarity to human retinal development. These characteristics make pig eyes and their retinal cells an ideal model for performing preclinical tests [
The data obtained in this study indicates that RSCs can be isolated and cultured
The controversial reports [
RSCs were faintly attached to ChM surface than to polystyrene surface in the first few hours. However, with time, RSCs adherence on surfaces increased significantly and, at 8 hours, 28% of RSCs were attached on ChM surface. This showed that RSCs adapted the unknown internal changes to favour their attachment on ChM surface along with time. Cell viability/cytotoxicity analysis at days 1 and 4 showed that RSCs grew well on ChM surface with time but significantly less than that grew on polystyrene surface. The difference of the cell numbers on surfaces (ChM versus polystyrene) increased significantly along with time. RSCs maintained same viability (<90%) on ChM surface as observed on polystyrene surface. The reason behind low RSCs adherence and proliferation on ChM surface is unknown but it confirms the role of structure and property of chitosan molecules acting as biophysical or biochemical cues for cultivation of RSCs on ChM surfaces.
Image analysis showed that RSCs were rounded at 8 hours (figure not shown) and they started to adopt fibroblast-like shape at day 1. But the cells with fibroblast-like shape were significantly less on ChM than polystyrene surface. On polystyrene surface, RSCs started to adopt fibroblast-like shape in early hours (day 1), and, at day 4, numerous cells with fibroblast-like shape could be observed. This confirmed that polystyrene surface is more favourable for RSCs for adopting fibroblast-like shape. Fibroblast-like shape indicates the cells under transition to another type of cells such as epithelial cells or fibroblast or specific retinal cells; therefore, the results confirmed that ChM promotes the RSCs transition.
Immunofluorescence study of RSCs confluent layer at day 11 showed that RSCs expressed cytokeratin8/18, PAX6, and nestin proteins on both surfaces. As previously mentioned, cytokeratin 8/18 protein is an epithelial cell protein, PAX6 is a retinal stem cell protein, and nestin is a neural stem cell protein. These proteins were selected for immunostaining in this study because the results can provide preliminary data about the RSCs differentiation direction to retinal, nonretinal, or epithelial cells on the different surfaces. RSCs gown on both surfaces showed the expression of these proteins, confirming that RSCs maintained the characteristics on ChM as observed on polystyrene. Anti-alpha-SMA antibody is used to detect transdifferentiation of cells into fibroblast. Alpha-SMA protein expression was detected in RSCs grown neither on ChM surface nor on polystyrene surface; however, it was detected in fibroblast used as a control. This further confirmed that RSCs maintained similar characteristics on ChM and polystyrene surfaces. The level of expression of these proteins, detecting other proteins such as proteins for type of retinal neurons differentiated in such conditions and cell quantification, is still an issue of investigation. Thus, RSCs growing on ChM are under cellular transition, forming fibroblast-like shapes, but not driven towards differentiation to fibroblast. It could be stimulated to differentiate into another type of cells which is still under investigation.
Chitosan is a FDA approved biomaterial used in clinics for various purposes due to its safe, tolerable, and biocompatible nature. Although polystyrene supports sufficiently the cellular behavior as well as growth, it cannot be alternative of chitosan because it does not have similar nature as chitosan contains. It is well known that commercially available polystyrene dishes are treated to release or activate chemical groups responsible for cell adhesion and proliferation; therefore, such treatments, if feasible, with ChM, would improve RSCs behavior on ChM surface. However, as previously mentioned, the fate of RSCs to differentiate into different retinal cell type could depend on the stimulus obtained in retinal environment. Therefore, ChM could be applicable in RSCs transplantation as a cell carrier because it would not affect the cell behaviour. Additionally after RSCs transplantation, poor cell growth and adhesion support the proliferating RSCs to migrate and integrate into retina and differentiate into appropriate cell types in accordance with the stimulus obtained. However, a further
RSCs express proteins associated with epithelial, stem, and different retinal cells confirming its potentiality to move towards undifferentiated or differentiated cells depending on stimulus received. RSCs adhere and grow poorly on ChM surface but they maintain similar viability and characteristics on ChM and polystyrene surfaces. Thus, ChM application as a RSCs carrier in cell transplantation increases the feasibility that proliferating RSCs can migrate, differentiate, and integrate in retina. However, further
The authors declare that there is no conflict of interests regarding the publication of this paper.
Girish K. Srivastava and David Rodriguez-Crespo equally contributed to this work.
Authors thank the staff of Justino Gutierrez S. L. slaughterhouse (Valladolid, Spain) for providing the porcine eye globes used in this work and Dr. D. Jimeno, Instituto de Neurociencias de Castilla y León (INCYL), University of Salamanca, for demonstrating RSCs isolation from ciliary marzine of porcine eyes. Authors are thankful for funding by (1) National Plan of I+D+I 2008–2011 and ISCIII-Subdireccion General de Evaluación y Fomento de la Investigación (PS09/00938) (MICINN) cofinanced by FEDER, (2) Castilla and Leon Regenerative Medicine and Cell Therapy Network Center, (3) JCYL BIO/39/VA26/10, Junta de Castilla y León, Spain, (4) Aragon Government, and (5) AECI, Spanish Ministry of Foreign Affairs and Cooperation. A part of this work was presented in Annual Meeting of European Association for Vision and Eye Research (EVER, 2011), XX Biennial Meeting of the International Society for Eye Research (ISER, 2012), and European College of Veterinary Ophthalmologists Conference (ECVO, 2013).