Conjunctival reconstruction via enrichment of human conjunctival epithelial stem cells by p75 through the NGF-p75-SALL2 signaling axis

Abstract

Severe conjunctival diseases can cause significant conjunctival scarring, which seriously limits eye movement and affects patients' vision. Conjunctival reconstruction remains challenging due to the lack of efficient methods for stem cells enrichment. This study indicated that p75 positive conjunctival epithelial cells (CjECs) were mainly located in the basal layer of human conjunctival epithelium and showed an immature differentiation state in vivo. The p75 strongly positive (p75++) CjECs enriched by immuno-magnetic beads exhibited high expression of stem cell markers and low expression of differentiated keratins. During continuous cell passage cultivation, p75++ CjECs showed the strongest proliferation potential and were able to reconstruct the conjunctiva in vivo with the most complete structure and function. Exogenous addition of NGF promoted the differentiation of CjECs by increasing nuclear localization of SALL2 in p75++ CjECs while proNGF played an opposite role. Altogether, p75++ CjECs present stem cell characteristics and exhibit the strongest proliferation potential so can be used as seed cells for conjunctival reconstruction, and NGF-p75-SALL2 signaling pathway was involved in regulating the differentiation of CjECs.

Keywords: NGF; SALL2; conjunctival epithelial stem cells; conjunctival reconstruction; p75; proNGF.

FIGURE 1

In vivo expression pattern of p75 in human conjunctival epithelium. A. Representative immunofluorescence images of p75 (green), K4 (red), and DAPI (4',6‐diamidino‐2‐phenylindole, blue) in human conjunctival epithelium. The p75 strongly positive conjunctival epithelial cells (CJECs) are shown by the double arrows and the p75 weakly positive CjECs are denoted by the single arrow. Scale bar = 50 μm. B, The percentages of p75 positive and p75 negative CjECs (based on total 20 sections from five different donors). C, Statistics of the K4+ and K4− CjECs among p75 positive CjECs. D, The percentages of K4+ CjECs, respectively, among p75 strongly positive (p75++) and p75 weakly positive (p75+) CjECs. Data are presented as the mean ± SD and statistically tested by a two‐tailed t test, ***P < .001; n = 5

FIGURE 2

Enrichment and identification of p75++, p75+, and p75− conjunctival epithelial cells (CjECs). A, Schematic diagram of cell sorting by immunomagnetic beads and the statistics of cell numbers in each subgroup. During the process, cells absorbed through the magnetic field after the first strong elution were recorded as p75++ CjECs. In the second sorting of the remaining CjECs, cells absorbed after a weak elution were recorded as p75+ CjECs, while those that were still unable to be adsorbed were recorded as p75− CjECs. The percentages of the three cell subsets were 16.6%, 27.8%, and 55.6%, respectively, which were consistent in multiple rounds of sorting. B, mRNA expression level of p75 in each cell subgroup analyzed by reverse transcription‐polymerase chain reaction (RT‐PCR). GAPDH (glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as an endogenous control. C, Representative Western blot analysis of p75 in each subgroup. β‐actin was used as an endogenous control. D, Quantification of relative p75 protein expression. The gray values of the p75 bands in each group were compared to those of β‐actin. Statistical results were standardized to the total. E, Representative immunofluorescence images of p75 (green) and DAPI (4',6‐diamidino‐2‐phenylindole, blue) in each cell subset when attached to the dish 6 hours after isolation. Scale bar = 50 μm. F, Quantification of the percentage of p75 positive cells. Data are presented as the mean ± SD and statistically tested by a two‐tailed t test, ***P < .001; n = 3

FIGURE 3

Expression levels of stem cell markers and differentiation‐associated keratins in each subgroup. A, Representative images of cell morphology in each subgroup when attached to the dish 6 hours after isolation. Scale bar = 50 μm. B, mRNA expression levels of ABCG2 and p63 in each subgroup analyzed by reverse transcription‐polymerase chain reaction (RT‐PCR). C, Representative immunofluorescence images of ABCG2 (green) and DAPI (4',6‐diamidino‐2‐phenylindole, blue) of each cell subgroup when attached to the dish 6 hours after isolation. Scale bar = 50 μm. D, Quantification of the percentage of ABCG2 positive cells. E, Representative immunofluorescence images of p63 (red) and DAPI (blue) of each cell subgroup when attached to the dish 6 hours after isolation. Scale bar = 50 μm. F, Quantification of the percentage of p63 positive cells. G, mRNA expression levels of K4 and K13 in each subgroup analyzed by RT‐PCR. H, Representative Western blot analysis of K4 and K13 in each subgroup. I, Quantification of relative K4 and K13 protein expressions. The gray values of K4 and K13 bands in each group were compared to the gray values of β‐actin. Statistical results were standardized with the total. J, Representative images of colonies formed by each subgroup 7 days after isolation. K, Quantification of CFE. CFE was calculated as the number of clones formed compared to the cells initially seeded. Data are presented as the mean ± SD and statistically tested by a two‐tailed t test, **P < .01, ***P < .001; n = 3. CFE, colony‐forming efficiency

FIGURE 4

Proliferative ability of each subgroup conjunctival epithelial cells (CjECs) during short and continuous cell passage cultivation in vitro. A, The short‐term proliferation rate of each cell subgroup analyzed by CCK8 assay and quantification of relative absorbance at 450 nm. B, Proliferation potential of each cell subset. A total of 106 cells were seeded in a 10 cm culture dish and cultured continuously in vitro. Cells were passaged in a ratio of 1:3 when they became confluent. Cell counts were performed every 5 days. C, Representative immunofluorescence images of ki67 (red) and DAPI (4',6‐diamidino‐2‐phenylindole, blue) of each cell subset on the 10th day during continuous cell passage cultivation. Scale bar = 50 μm. D, Quantification of the percentage of ki67 positive cells. Data are presented as the mean ± SD and statistically tested by a two‐tailed t test, ***P < .001; n = 3

FIGURE 5

Conjunctival reconstruction ability of each subgroup conjunctival epithelial cells (CjECs). A, Representative H&E staining images of laminated conjunctival epithelium constructed by each subgroup of cells on an amniotic basement membrane cultured for 10 days. B, Quantification of tissue layers. C, Representative immunohistochemical staining images of the laminated conjunctival epithelium constructed by each subgroup of cells. The expression levels of differentiated keratin K13, functional protein MUC5AC, and ZO‐1 were analyzed. Scale bar = 50 μm. D, Repair of conjunctival defect in rabbits by each subgroup of cells. The initial size of the conjunctival defect is 7 mm in diameter. The area of conjunctival defect was stained with Lissamine Green B. E, Quantification of the percentage of damaged area. F, Histological analysis of conjunctival defect reconstruction. The goblet cells are shown by the black arrow. Scale bar = 50 μm. Data are presented as the mean ± SD and statistically tested by a two‐tailed t test, ***P < .001; n = 3

FIGURE 6

The role of the NGF‐p75‐SALL2 signaling pathway in the regulation of p75++ conjunctival epithelial cells (CjECs). A, Representative Western blot analysis of different cell subsets treated with nerve growth factor (NGF) at 100 ng/mL and pro‐NGF at 1000 ng/mL for 72 hours. NGF promoted differentiation of CjECs while proNGF inhibited the differentiation, which was more obvious in p75++ CjECs but not in p75− CjECs. B, Quantification of relative K13 protein expression in each subgroup. The gray values of the K13 bands in each group were compared to the gray values of β‐actin. Statistical results were standardized with the blank control group. C, Representative Western blot analysis of the distribution of SALL2 in the nucleus and cytoplasm before and after treated with NGF at 100 ng/mL for 2 hours. glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as a control for cytoplasm and SP1 for the nucleus. The addition of NGF could increase nuclear localization of SALL2. D, Quantification of relative SALL2 protein expression. The gray values of the SALL2 bands were compared with the gray values of GAPDH and SP1. E, Percentage of SALL2 protein expression in the nucleus accounted for the total before and after NGF treatment. F, Representative immunofluorescence images of SALL2 (red), p75 (green), and DAPI (4',6‐diamidino‐2‐phenylindole, blue) in CjECs, before and after treatment with NGF at 100 ng/mL for 2 hours. The p75 positive CjECs are shown by double arrows, and the p75 negative CjECs are shown by a single arrow. The nuclear localization of SALL2 was increased by NGF in p75++ CjECs but not in p75− CjECs. Scale bar = 20 μm. G, Percentage of SALL2 FI in the nucleus accounting for total cells before and after NGF treatment. H, Representative co‐immunoprecipitation (CO‐IP) analysis of the interaction of SALL2 with p75 before and after treated with NGF at 100 ng/mL for 2 hours. p75 was coimmunoprecipitated from cell lysates and Western blot was probed with antibodies to p75 and SALL2. β‐actin was used as an endogenous control. The addition of NGF reduced the interaction of SALL2 with p75. I, mRNA expression level of downstream genes of SALL2 (p16, c‐MYC, SOX2, and OCT4) before and after treated with NGF analyzed by reverse transcription‐polymerase chain reaction (RT‐PCR). GAPDH was used as an endogenous control. NGF treatment significantly increased the expression level of p16 and decreased the expression level of c‐MYC, SOX2, and OCT4. Data are presented as the mean ± SD and statistically tested by t test, *P < .05, ***P < .001; n = 3. FI, fluorescence intensity