Phenotype of hRPE cells and mouse ESCs
The primary adherent pigmented hRPE cells at first plating (P0) reached 90% confluence after seven days of cultivation (Fig. 1A). They grew as cobblestone cultures and contained a great quantity of pigmentation. During culture, the pigment was diluted upon cell division and the cells gradually acquired a fusiform, largely depigmented morphology. The results of western blotting indicated that the differentiation marker proteins CRALBP, S-100 and RPE65 were expressed in the hRPE cells at P4 (Fig. 1B). Immunofluorescence staining also showed CRALBP expression in the hRPE cells (Fig. 1C).
The mouse ESCs exhibited a clonal or islet appearance. Under a light microscope, the clone was bright and round with a clear, sharp boundary (Fig. 1D). Immunofluorescence assays showed that the stem cell markers OCT4 and KLF4 were expressed in the ESCs (Fig. 1E).
Effects of coculture with ESCs on morphological changes in hRPE cells
The role of ESCs in regulating the morphology of hRPE cells was investigated. The hRPE cells at P5 in the control group presented a fusiform pattern (Fig. 2A). On the other hand, the P5 hRPE cells in the hRPE + ESC group showed an epithelioid shape with a homogeneous morphology that is more similar to the primary cultured cells originating from the eye cups, which maintained a normal cell property of contact inhibition. Immunofluorescent staining indicated reduced OCT4 expression in ESCs cocultured with hRPE cells (Fig. 2B). After culturing for 72 h, hRPE cells from all groups were collected for experiments below.
Coculture with ESCs enhances the proliferative capacity of hRPE cells
We next investigated the potential effects of ESCs on the proliferation of hRPE cells. The CCK-8 test (a common assay for detecting cell proliferation) was used to obtain the growth curve of hRPE cells from each group. During the slow-growing latent phase in days 1 and 2, no marked differences of optical density (OD) values were detected among the three groups (Fig. 3A). However, on the third day, the hRPE + ESC group showed significantly higher OD values than the control group. All three groups entered the logarithmic growth phase on the fourth day but the slope of the growth curve for the hRPE + ESC group was higher than the other two groups. These observations indicate that the hRPE cells treated with ESCs possessed a relatively stronger growth capacity. In contrast, there were no significant differences in the growth curves between the hRPE group and the hRPE + CEC group.
To study the effects of ESCs on cell apoptosis, we employed Annexin V-APC/7-AAD staining and flow cytometry analysis. We found that the percentages of apoptotic hRPE cells from the Ctrl, hRPE + CEC, and hRPE + ESC groups were 13.73 ± 0.9912%, 13.7 ± 1.512%, and 9.473 ± 1.835%, respectively (Fig. 3B). There were fewer apoptotic cells in the hRPE + ESC group than in the hRPE group (P = 0.0065), while the percentages of apoptotic cells were almost the same in the hRPE and hRPE + CEC groups. These results suggest that ESCs inhibit apoptosis in hRPE cells.
Cell cycle progression was further evaluated by flow cytometry. As shown in Fig. 4A., the percentage of hRPE cells entering the cell cycle was significantly higher in the hRPE + ESC group than the other groups (P < 0.05). In the hRPE + ESC group, 33.94%±2.191% of cells were entering S phase, whereas in the hRPE group, 14.71%±2.468% of cells were in S phase. Consistent with the flow cytometry results, hRPE cells cocultured with ESCs showed higher expression levels of the cell cycle promoters cyclin A2, cyclin B1, and cyclin D1, and lower expression levels of the cell cycle negative regulators p21 and p27, both transcriptionally and translationally (Fig. 4B-D). In contrast, neither the cell cycle distribution nor the cell cycle-related protein expression levels of hRPE cells was significantly changed by coculture with CECs.
Coculture with ESCs enhances the stem cell phenotype of hRPE cells
As shown in Fig. 2A, the hRPE cells cocultured with ESCs had smaller cell sizes with bigger nucleus to cytoplasm (N:C) ratios than cells in the control group (Fig. 5A). No obvious differences in morphology and N:C ratio was observed between the control and hRPE + CEC groups. To evaluate the clonal growth capacity, the hRPE cells from the three groups were seeded without a feeder layer to assess colony-forming efficiency (CFE). At day 7, the CFE of the hRPEs in the ESC-treated group reached 10.65%±0.6856%, whereas the CFE of the control and hRPE + CEC groups were 3%±0.4082% and 2.925%±0.3775%, respectively (Fig. 5B).
RT-qPCR analysis revealed that the mRNA expression of the RPE differentiation markers CRALBP and PEDF by hRPE cells in the hRPE + ESC group was significantly decreased by 0.264- and 0.315-fold, respectively, in comparison to the control group (Fig. 5C). However, the expression of KLF4, a marker associated with early stem cells and reprogramming was markedly increased by 167-fold in hRPE cells after coculturing with ESCs (Fig. 5C). Consistent with the mRNA expression profiles, western blot analysis revealed that hRPE cells from the control group strongly expressed CRALBP, while hRPE cells in the hRPE + ESC group showed weaker CRALBP expression (Fig. 5D). KLF4 was barely detectable in the hRPE cells in the control group but markedly upregulated in ESC-treated hRPE cells (Fig. 5D). In contrast, the expression of RPE-specific and stem cell-associated markers in hRPEs from the hRPE + CEC group was not significantly different from that of the control group. These observations indicate that ESCs can enhance the stem cell phenotype of hRPE cells.
ESCs enhance the proliferative capacity of hRPE cells by activating the PI3K pathway
Our previous finding 27 demonstrated that the ESC microenvironment promoted the proliferation of corneal epithelial cells via activation of the PI3K/Akt signaling pathway. Therefore, we further examined whether coculturing with ESCs enhances the proliferative capacity of hRPE cells by upregulating the PI3K pathway. We performed RT-PCR validation of key PI3K pathway genes, including PAR2, FAK, PI3K, PDK1/2 and AKT, and found that they were significantly upregulated in hRPE cells after exposure to ESCs (Fig. 6A-C). In contrast, the expressions of these genes were not enhanced in hRPE cells from the CEC group compared with the hRPE group.
To determine whether PI3K pathway activity is necessary for the pro-proliferative effect of the ESC microenvironment, we added the PI3K antagonist, LY294002 to disrupt PI3K signaling (Fig. 6D). LY294002 abolished the pro-proliferative effect of the ESCs on hRPE cells in the ESC co-culture system; the growth of hRPE cells was reduced and ESC-inhibited apoptosis was blocked (Fig. 3,4). Meanwhile, expression of the cell cycle promoters, cyclin A2, cyclin B1, and cyclin D1 decreased, while expression of p21 and p27 increased significantly (Fig. 4B-D). These findings indicate that the PI3K pathway is indeed activated by ESCs in hRPE cells, which in turn boosts proliferation. Interestingly, although LY294002 decreased the N:C ratio and clonal growth capacity of hRPEs (Fig. 5A,B), it did not affect their expression of RPE-specific and stem cell-associated markers (Fig. 5C,D).