Amniotic fluid promotes the appearance of neural retinal progenitors and neurons in human RPE cell cultures

Isolation and cultivation of RPE cells

Enucleated eye globes were obtained from newborn human cadavers, up to one year of age, less than 24 h after death at the Central Eye Bank of Iran. To isolate the RPE cells, the initial surrounding tissues, including the fat tissue and optic nerves, were removed. Then, a circular incision was made at the anterior region of the globe next to the iris. The anterior segment and its contents, including the iris, cornea, lens, and vitreous humor, were discarded. The remaining part of the globe was opened, and the neural retina was peeled off. The rest of the tissue, including the RPE layer and choroid, were washed several times with PBS. (1X; was prepared by dissolving 8 g NaCl, 0.2 g KCl, 2.68 g Na₂HPO₄-7H₂O and 0.24 g KH₂PO₄ in 800 ml ddH₂O and then pH was adjusted to 7.4 and final volume was adjusted to 1 l).The RPE tissue was then gently separated from the sclera, chopped into small pieces, and incubated in 1 U/ml of dispase I solution (Invitrogen, Brussels, Belgium) at 37 °C for 50 min. Subsequently, the cells were centrifuged at 300 g and 4 °C for 5 min. The pellet was resuspended in DMEM/F-12 (Sigma, Munich, Germany). The suspension was incubated in DMEM/F12 supplemented with 20% FBS; Gibco, Canyon, Australia), 120 µg/ml penicillin, 100 µg/ml streptomycin, and 250 µg/ml amphotericin B (the latter was used only for freshly isolated cells) in a 25 cm² flask, which was precoated for at least 2 h at 37 °C with FBS in a humidified incubator containing O₂ 95% and CO₂ 5% at 37 °C. The medium was changed every 3–4 days. When the RPE cultures reached confluence, the cells were subcultured with a solution of Trypsin 0.25% (w/v; Sigma)–1 mM EDTA (Merck, Darmstadt, Germany) in PBS, and the detached cells were seeded at a density of 5 ×10³ cells/cm². Confluent RPE cultures between passages 3 and 7 were used for all examinations.

Human amniotic fluid preparation

Human amniotic fluid (HAF) samples were obtained from pregnant women between the 14^th and 16^th weeks of pregnancy. The women had been advised to undergo an amniocentesis procedure for a genetic evaluation, irrespective of the current study. Researchers received the extra quantity of HAF to treat RPE cultures. After centrifuging the HAF at 300 ×g at 4 °C for 5 min, the HAF cells were removed for karyotype analysis; the remaining supernatants that were devoid of chromosome abnormalities were pooled, passed through a 0.2 μm filter for sterilization, and preserved at −70 °C. The ethics committee of the National Institute of Genetic Engineering and Biotechnology and Ophthalmic Research Center approved the collection process for these samples. Amniotic fluid samples were obtained from thirty pregnant women who underwent amniocentesis for the assessment of genetic deficiencies in the first trimester of gestation. Amniotic fluid cells were removed for karyotype analysis. The remaining supernatants, in cases with no evidence of chromosomal abnormalities, were pooled and used in our downstream experimental procedures. The collection of these samples was approved by the ethics committees of the NIGEB and the Ophthalmic Research Center. The AF samples were centrifuged at 300 ×g for 5 min at 4 °C, and the resulting supernatants were then sterilized using a 0.2 μm membrane filter (OrangeScientific, Brussels, Belgium) and stored at −70°C until the time of analysis.

Treating RPE cultures with HAF

To examine the effects of HAF on the morphology of the RPE cultures, 1×10⁵ RPE cells/well were cultured in 1 ml of DMEM/F12 supplemented with HAF 10% (v/v) or FBS 10% (v/v) in a 24-well microplate (Nunc, Roskilde, Denmark). As a control, the same number of cells were cultured in DMEM/F12. Cultures were examined daily by phase contrast microscopy (Axiovert; ZEISS, Germany) for more than 30 days, and morphological changes were regularly surveyed and documented with photographs.

Cell proliferation and death enzyme-linked immunosorbent assays

Cell proliferation and death enzyme-linked immunosorbent assay (ELISA) were performed to investigate whether HAF could induce proliferation or even unfavorable apoptosis in RPE cultures. To evaluate the effect of HAF on either cell proliferation or apoptosis of the cultured RPE cells, the cells were plated at a density of 1×10⁴ cells/well in 200 μl of DMEM/F12 supplemented with HAF 10, 20, or 30% (v/v) or FBS 10% (v/v) in a 96-well microplate. The control cultures received DMEM/F12 instead of the HAF- or FBS-supplemented media. Twenty-four hours after incubation, the medium was replaced with fresh medium containing BrdU (for cell proliferation assay; concentration of BrdU stock was 1000X. BrdU stock was diluted 1 to 1000 in each reaction mixture.) and incubated for another 24 h. Then, the treated and control cultures were assayed for cell proliferation (based on the measurement of BrdU incorporation during DNA synthesis) and cell death (qualitative and quantitative in vitro determination of cytoplasmic histone-associated-DNA fragments) according to the protocols recommended by the manufacturer (cell proliferation ELISA, BrdU colorimetric, and cell death detection ELISA kits, Roche, Grenzach-Wyhlen, Germany). The absorbance values for the proliferation (at 450 nm and 690 nm) and cell death (at 405 nm and 492 nm) assays were recorded using a scanning multiwell spectrophotometer (Titertek multiscan ELISA reader, Labsystems Multiscan, Roden, Netherlands).

Immunocytochemistry

The harvested RPE cells were seeded onto FBS- or HAF-coated glass coverslips at a density of 1×10⁵ cells/well in 1 ml of DMEM/F12 supplemented with HAF 30% (v/v), FBS 10% (v/v), or DMEM/F12 (as a control) in 24-well microplates. Seven days later, a standard immunocytochemistry (ICC) procedure was performed, as described by Ghaderi et al. [21]. Briefly, the medium was removed, and the cells were washed with PBS and fixed with cold methanol (−10 °C) for 5 min at room temperature (RT). The air-dried cells were washed with PBS three times and then blocked with bovine serum albumin (BSA) 1% (w/v; Sigma) in PBS for 20 min at RT. The cells were subsequently rinsed with PBS and incubated for 1 h at RT with primary antibodies that had been diluted 1:50 in bovine serum albumin (BSA 1% [w/v])/PBST 1% (v/v). To prepare 1% BSA in PBST, PBST was prepared by dissolving 1 ml Triton X-100 in 99 ml PBS buffer (1X). Then 1 g of BSA was dissolved in each 100 ml of PBST. The primary antibodies used were the rabbit polyclonal anti-human NESTIN antibody (Santa Cruz, Carlsbad, CA), goat polyclonal anti-CRABPI (Santa Cruz, Carlsbad, CA), rabbit polyclonal anti–PKCα (Santa Cruz), rabbit anti-human RPE65 polyclonal antibody (Santa Cruz), and mouse anti-human cytokeratin 8/18 (Santa Cruz). After three washes with PBS for 5 min each, the cells were incubated with secondary antibodies that had been diluted 1:100 in BSA 1.5% (w/v) / PBST 1% (v/v) in the dark for 45 min at RT. The secondary antibodies were FITC-conjugated goat anti-rabbit IgG (1:100, Santa Cruz), FITC-conjugated goat anti-mouse IgG (1:100, Santa Cruz), and FITC-conjugated donkey anti-goat IgG (1:100, Santa Cruz). Finally, the cell nuclei were stained with 1 mg/ml of 4´,6-diamidino-2-phenylindole dihydrochloride (DAPI, Santa Cruz) for 10 min in the dark at RT. The coverslips were mounted onto slides using fluorescence mounting media (glycerol 90%, PBS 10%, and phenyl diamine 10% [w/v]). The negative controls were processed as described above but without any primary antibody. An Axiophot Zeiss fluorescence microscope equipped with a 460 nm filter for DAPI and a 520 nm filter for the FITC was used to observe and image the cells.

RNA extraction and cDNA synthesis

Total RNA was extracted (using the RNeasy kit; Qiagen, Hilden, Germany) from RPE cultures that were treated for 24 h with 10, 20, or 30% (v/v) HAF and/or 10% (v/v) FBS treated cultures as a control, to normalize the obtained data. Subsequently, the first strand of cDNA was synthesized using the cDNA synthesis kit (Qiagen) according to the manufacturer’s instructions.

Real-time quantitative PCR analysis

For real-time PCR assays using SYBR Green I, the primers for all the genes of interest were designed using the Beacon Designer software 7.00. A BLAST search was performed to check the specificity of the primer sequences (for primer details see Table 1). To quantitatively measure the expression of the NESTIN, PKCα, CRABPI, CRABPII, pigment epithelium-derived factor (PEDF), and VEGF genes, real-time PCR was performed using Light Cycler apparatus version 3.5 (Roche). The housekeeping gene glyceraldehyde phosphate dehydrogenase (GAPDH) was used as an internal control. The real-time PCR program included the following three main steps: an initial denaturation step at 95 °C for 10 min; 40 cycles of amplification that consisted of a denaturation step (10 s at 95 °C), an annealing step (40 s at 57 °C for NESTIN, PKCα, CRABPII, and GAPDH, 30 s at 60 °C for PEDF, 30 s at 52 °C for VEGF, and 50 s at 57 °C for the CRABPI gene); and finally, an extension step (15 s at 72 °C). To draw the melting curve and confirm the precision of the reactions, the following steps were performed: step 1, 95.0 °C for 0 s; step 2, 65.0 °C for 15 s; step 3, 95.0 °C for 0 s; step 4, 40.0 °C for 30 s. The size of the PCR product for each gene was confirmed using gel electrophoresis. To determine the PCR efficiency of the amplifications, a standard curve was prepared by making serial dilutions of the cDNA samples (a twofold dilution series for NESTIN, CRABPI, CRABPII, PKCα, VEGF, and GAPDH and a fivefold dilution series for PEDF). Relative expressions of all the genes of interest under the treatment of three different doses of HAF measured according to the 2^-∆∆Ct method, which yields a normalized, relative gene expression value with respect to FBS-treated cultures, using the Bio-Rad software. Experiments were replicated two or three times.

Statistical analysis

Results of the ELISA assays for cell proliferation and cell death were based on at least two different experiments with two replicates in each studied group. The averages of the absorbance measurements for all the repetitions were calculated and statistically with the t test. To calculate the percentage of cells that were immunopositive for PKCα and CRABPI, the number of cells in three different fields of vision was counted under the fluorescent microscope, and the average of the counted immunoreactive cells was reported. To evaluate the expression of the genes of interest, real-time PCR was performed for two series of treated and control cDNA samples obtained from distinct experiments. Minimally, each sample was assessed in duplicate.

The establishment of RPE cells as a monolayer culture

Freshly isolated human RPE cells were seeded in DMEM/F12 supplemented with 20% FBS. During the first week of cultivation, the cells were small and had various shapes, including triangular, star-like, and branching (Figure 1A,B). Afterward, they became elongated and had a fibroblast-like morphology (Figure 1C). Regardless of the morphological features, the RPE cells were heavily pigmented in the developing cultures, although their pigment content became more diluted with consecutive passages. When the cultures formed a monolayer and reached full confluence, the cells became hexagonal and organized into “cobblestone” structures (Figure 1D). More interestingly, we observed the development of colonies consisting of several hundred cells in the established cultures (Figure 1E,F). When the cultures reached the fifth passage, the RPE cells gradually flattened and their ability to proliferate decreased (Figure 1G,H). These alterations implied that the cultures were undergoing progressive senescence [30]. An immunocytochemical examination of the newly isolated cells showed that almost all cells were labeled by the anti-RPE65 and anti-cytokeratin 8/18 antibodies, confirming the identity and the purity of the culture as epithelial cells, and particularly RPE cells (Figure 2A,B). The presence of RPE65 and cytokeratin 8/18 were tracked in the cultures over a 6-week period. We found that the number of RPE65-immunoreactive cells decreased significantly, while a large number of cells strongly expressed cytokeratin 8/18. As negative controls, secondary antibody–only control was also included; goat anti-rabbit and goat anti-mouse secondary antibodies were used for RPE65 and cytokeratin 8/18, respectively (Figure 2 E,F).

The RPE–derived cells formed neurosphere-like colonies when treated with HAF

To determine the effect of HAF on the morphology of the RPE cells, the cultures from the middle passages (passage 3 was chosen as the middle point of the exponential phase for RPE cultures, data not shown) were treated with three doses of HAF (10%, 20%, and 30%) for 30 days. Twenty-four hours after the treatment, the cultures formed a well-adherent monolayer of cells that had a uniform morphology and that were arranged in parallel to each other (Figure 3A). After 7 days, the cells began to aggregate and formed clusters of a few cells in the peripheral area of the vessels. Upon subsequent observation, the clusters of cells grew in number and size and formed sphere-like colonies containing both pigmented and nonpigmented cells (Figure 3B,C). The RPE-derived spheres were consequently detached from the base of the plate and floated into the medium (Figure 3D). The sphere formation rate was associated with HAF concentration, that is, higher doses caused more sphere-like colony formation in a shorter period of time (data not shown). In contrast, sphere-like colony formation was delayed considerably in the control cultures.

The RPE–derived colonies consisting of neural progenitor cells

The examination for the development of progenitor cells was based on tracing the NESTIN-positive cells. Neural progenitor cells were found in the neurospheres, but not in monolayer-associated cells. In cultures treated with 30% HAF, nearly all cells forming neurosphere-like colonies expressed NESTIN, as did neural progenitors (Figure 4). This result suggests that HAF stimulates the RPE cultures to dedifferentiate into cells that phenocopy neural progenitor cells.

HAF–promoted proliferation of RPE cells

The proliferation of RPE cells in the presence of HAF, FBS, and DMEM/F12 was evaluated. We found that the proliferation of the cultures was promoted when the HAF concentration in the DMEM/F12 medium was increased (Figure 5). The rate of cell death for the RPE cells in the HAF-containing media remained almost the same as for the cells in the FBS-containing media (Figure 6). This result is consistent with those obtained from the cell proliferation assay.

The RPE–derived cells treated with HAF proceeded toward retinal neurons

Seven days after HAF treatment, we used immunocytochemistry to characterize the cellular components of the cultures. The number of cells that were immunoreactive for the RPE markers cytokeratin 8/18 and RPE65 decreased significantly in the treated cultures, while a few cells expressed interneuron-specific markers, including PKCα (Figure 7A) and CRABPI (Figure 7B). Approximately 19% and 25% of cells in 30% HAF-treated cultures were positive for PKCα (bipolar cell marker) and CRABPI (amacrine cell marker), respectively. As control, FBS-treated cultures showed 5% and 12% of cells positive for PKCα (Figure 7E) and CRABPI (Figure 7F), respectively. These results suggest that the HAF-treated RPE cultures transdifferentiate into retinal neurons similar to the interneurons located in the inner nuclear layer.

An evaluation of the transcript levels of NESTIN, CRABPI, PKCα, and GAPDH, as well as CRABPII, PEDF, and VEGF verified the data obtained from the immunocytochemistry and demonstrated that the expression of NESTIN, PKCα, and CRABPI in HAF-treated cultures were enhanced 2-, 20-, and 22-fold, respectively. Nevertheless, the expression levels of these genes in cells treated with 10 and 20% HAF were the same or less than those of the cells treated with FBS.

Although the transcript levels of PEDF, VEGF, and CRABPII indicated increased expression of these genes (7.5-, 5.5-, and 2.7-fold, respectively) in cells treated with 10% HAF, further increasing in concentration of HAF significantly decreased expression levels for genes of interest, especially for PEDF. The amounts of mRNA for CRABPI, CRABPII, and VEGF genes were also decreased in the samples treated with 20% HAF (Figure 8).