The Mesenchymal Stem Cell-like Phenotype And Multilineage Potential And The Inhibitory Effect On Dendritic Cells Maturation Of Murine Corneal Stroma Cells

Corneal stroma cells (CSCs), a unique population of neural crest-derived cells embedded in the corneal stroma, play a major role in maintaining corneal transparency. Since the cell number in vivo is scarce, CSCs must be expanded in vitro. Previous studies have indicated that when cultured in the complete medium (containing fetal bovine serum, FBS), CSCs readily lost their biological characteristic and transformed into some other cells. Unfortunately, CSCs cultured in the serum-free medium do not proliferate. Therefore, expanding CSCs while maintaining their normal biological characteristic in vitro is very desirable.Recent studies have shown that between birth and eyelid opening, the number of proliferating CSCs decreases dramatically, and at the time of eyelid opening, CSCs have withdrawn from the cell cycle, remaining in G0 rather than undergoing complete terminal differentiation. Moreover, CSCs express stem cell markers and have transdifferentiation potency, which are similar to mesenchymal stem cells. However, the study on the mesenchymal stem cell characteristic of murine CSCs is rare.Dendritic cells (DCs) are the most efficient antigen-presenting cells that initiate or control adaptive immune responses to invading pathogens, and are found in two distinct functional states. Immature DCs can uptake antigens and induce immunity tolerance; mature DCs are uniquely able to stimulate naive T cell responses efficiently. Since corneal DCs play a critical role in corneal transplantation and corneal disorders, and some approaches targeted DCs have been applied, further studies on regulation of DCs maturation should be required.Recent studies have demonstrated that DCs are uniformly immature in the central cornea, but mainly mature in the peripheral region. And local microenvironment has been widely recognized as an important regulator for DCs maturation. Therefore, we deem that CSCs should have regulative effect on DCs maturation. However, to date, no systematic study has been performed.Consequently, this study was carried out to explore how to expand murine CSCs, and to investigate the mesenchymal stem cell characteristic and the inhibitory effect on DCs maturation of murine CSCs. As follows: Part 1 Isolation, identification, cultivation, and expansion of murine corneal stroma cellsObjective: To investigate whether murine CSCs expanded in the KSFM medium still hold the original biological characteristic.Methods: After incubated in EDTA solution (20mmol/L) for 45 minutes, the corneal epithelium and endothelium were carefully peeled away from the corneal stroma with fine forceps. And then, central corneal stromas were digested with collagenase I (300U/mL) for 4 hours. Following centrifugation, isolated single cells were harvested, and seeded on plastic in the DMEM basic medium (serum-free) or the DMEM complete medium (containing 10% FBS), or in the KSFM medium. The cells were cultured at 37℃in a 5% CO2 atmosphere, and subcultured with EDTA solution (containing 1U/mL dispase). Meanwhile, the cells were observed and further the cell growth curve was drawn; the gene expression of keratocan, aldehyde dehydrogenase (ALDH), cytokeratin 12 (CK12), and neuron-specific enolase (NSE) was examined by reverse transcription polymerase chain reaction (RT-PCR); the protein expression of keratocan was analyzed by immunofluorescence and Western Blot.Results: After collagenase digestion, cell suspension obtained from two murine corneal stromas yielded about 1×10~4 single cells. The data of RT-PCR indicated that the primary cell exhibited positive expression of keratocan and ALDH, which are considered as hallmarks for keratocytes, and negative expression of CK12 and NSE, which are expressed in corneal epithelium and endothelium respectively; the data of immunofluorescence and Western Blot further showed that these cells expressed keratocan protein. And thus, the primary cells in this study were of stromal origin. In the DMEM basic medium, primary CSCs could not proliferate; in the DMEM complete medium, CSCs proliferated, but passage 3 cells lost the gene expression of ALDH and keratocan and the protein expression of keratocan; in the KSFM medium, CSCs also proliferated, and passage 3 cells still maintained the gene expression of ALDH and keratocan and the protein expression of keratocan, with no significant difference compared with primary CSCs (P>0.05).Conclusion: KSFM medium can maintain the biological characteristic of murine CSCs while promoting cell proliferation.Part 2 The mesenchymal stem cell-like phenotype and multilineage potential of murine corneal stroma cellsObjective: To investigate whether murine CSCs, cultured and expanded in the KSFM medium, share the same phenotype and multilineage potential with mesenchymal stem cell.Methods: The central region of murine cornea was treated with collagenase digestion after the epithelium and endothelium were removed. Then the single cells were cultured and expanded in the KSFM medium. Passage 2 CSCs were harvested and incubated with hematopoietic marker antibodies (CD34-FITC, CD45-PE) and mesenchymal marker antibodies (CD105-PE, CD90-FITC, CD71-FITC, CD29-APC) for 30 minutes at 4℃in dark. Then the stained cells were analyzed on a flow cytometer. CSCs were maintained in the KSFM medium. At day 2 post-confluence, the medium was changed with the osteogenic differentiation medium (DMEM supplemented with 10% FBS, 100nmol/L dexamethasone, 10mmol/Lβ-glycerophosphate, and 50mg/L ascorbic acid), the adipogenic differentiation medium (DMEM supplemented with 10% FBS, 0.5μmol/L dexamethasone, 0.5mmol/L isobutylmethylxanthine, and 10mg/L insulin), or the control medium (DMEM supplemented with 10% FBS). The medium was changed every other day. After 21 days, the cells, cultured in the osteogenic differentiation medium and the control medium, were stained with 2% alizarin red S solution, and the gene expression of alkaline phosphatase and osteocalcin was examined by RT-PCR; and the cells, cultured in the adipogenic differentiation medium and the control medium, were stained with 0.3% oil red O solution, and the gene expression of lipoprotein lipase and peroxisome proliferator activated receptorγwas examined by RT-PCR.Results: The phenotypic characterization of passage 2 CSCs was analyzed by flow cytometry. Data showed that cells were negative for CD34 (3.68%±1.44%) and CD45 (9.56%±1.83%); but positive for CD29 (96.85%±1.91%), CD90 (93.62%±1.65%), CD105 (50.91%±2.56%), and CD71 (45.27%±3.56%). Within 3 days after osteoblastic induction, cells continued to exhibit fibroblast-like morphology similar to cells maintained in the control medium. After 7 days, cells had transformed to a many-horned shape, with black particles appeared in cytoplasm. One week later, center of the colony increased gradually and finally formed mineralization nodules, which were stained by alizarin red S. Cells cultured in the control medium did not show any morphological sign of osteoblastic differentiation and were not stained by alizarin red S. Moreover, the expression of osteoblast-specific markers was analyzed by RT-PCR. Data showed that, cells under osteoblastic conditions exhibited positive expression of alkaline phosphatase and osteocalcin, whereas cells cultured in the control medium showed lower expression of alkaline phosphatase and negative expression of osteocalcin. After 7 days under adipogenic conditions, cells changed their shape from spindle to round, coincided with the accumulation of intracellular droplets. Two weeks after initial induction, the cytoplasm was completely filled with lipid rich vacuoles, which were stained positively by oil red O. Differentiation was demonstrated further by RT-PCR analysis. The lipoprotein lipase and peroxisome proliferator activated receptorγmRNA were both detected. These changes were not found in cells cultured in the control medium.Conclusion: Similar to mesenchymal stem cells, the murine central cornea-derived CSCs, cultured and expanded in the KSFM medium, have special phenotypic marker expression profile and can differentiate into adipocytes and osteoblasts.Part 3 The inhibitory effect of murine corneal stroma cells culture supernatant on dendritic cells maturationObjective: To investigate whether murine CSCs culture supernatant can inhibit lipopolysaccharide-induced DCs maturation.Methods: Splenic T cells from BALB/c mice were collected by nylon wool columns, and the purity of cultured T cells was determined by flow analysis of surface CD3 staining. Three days after primary murine CSCs (105/mL) were cultured in the serum-free RPMI 1640 medium, the medium was semi-changed. Three days later, the culture supernatant was harvested and tested. Bone marrow mononuclear cells were prepared from C57BL/6 mouse femur bone marrow suspension by depletion of red cells and then cultured in the RPMI 1640 medium supplemented with 10% FBS and 10ng/mL recombinant murine granulocyte macrophage colony stimulating factor. The medium was wholly changed on day 3 and semi-changed on day 5. On day 7, nonadherent and loosely adherent cells were harvested as immature DCs, the purity of which was tested by flow cytometry with anti-CD11c antibody staining. To induce DCs maturation, lipopolysaccharide (1μg/mL) was added for another 48 hours of culture. To explore the effect of CSCs culture supernatant on DCs maturation, various concentrations of culture supernatant (25%, 50%) was added into the culture medium during the DCs maturation stage. And then, to evaluate the phenotypic maturation of DCs, the cellular surface markers for maturation, including CD80, CD86 and major histocompatibility complex classⅡ(MHC-Ⅱ) were analyzed by flow cytometry. Furthermore, to evaluate the functional maturation of DCs, the capability of stimulating the proliferation of T lymphocytes was measured by allogeneic mixed lymphocyte reactions and the function of endocytosis was assessed by fluorescein isothiocyanate-dextran uptake.Results: For T cells, following lysis of red cells, splenic cells were passed through nylon wool columns and nonadherent small cells were harvested. Phenotypic analysis by flow cytometry indicated that these cells were positive for CD3 (93.97%±3.06%), which is considered as a hallmark for T cells. For DCs, on day 7 of culture, bone marrow mononuclear cells generated many distinctive cell clusters with nonattachment and loose attachment to plate bottoms. These cells displayed different protruding veils, and expressed high level of CD11c (78.61%±4.27%), but low levels of the maturation markers CD80, CD86, and MHC-Ⅱ. An additional 48 hours of stimulation with lipopolysaccharide later, the levels of the maturation markers were increased. In the next step, the effect of CSCs culture supernatant on DCs maturation was explored. After adding the culture supernatant (25%, 50%) during the mature stage of DCs, compared with the control group, the expression of CD80, CD86, and MHC-Ⅱwas down-regulated (P<0.01), and the expression of CD11c was not altered (P>0.05); the capability of stimulating the proliferation of T lymphocytes was decreased (P<0.05); and the function of endocytosis was increased (P<0.01). Furthermore, the inhibitory effect seemed in a dose-dependent manner (25% vs. 50%, P<0.05). Conclusion: Murine CSCs culture supernatant can inhibit lipopolysaccharide-induced phenotypic and functional maturation of DCs dose-dependently. And thus, we speculate that CSCs could inhibit DCs maturation via secretion of soluble immunomodulatory cytokines.Part 4 Murine corneal stroma cells inhibit dendritic cells maturation partially through transforming growth factorβ2 and prostaglandin E2-mediated mechanism in vitroObjective: To explore which kind of immunomodulatory cytokines secreted by murine CSCs inhibits DCs maturation.Methods: The gene expression of transforming growth factorβ2 (TGF-β2), macrophage colony stimulating factor (M-CSF), interleukin 10 (IL-10), and prostaglandin endoperoxide synthase 2 (PTGS2) in primary murine CSCs was examined by RT-PCR. After that, the levels of TGF-β2 and prostaglandin E2 (PGE2) in CSCs culture supernatant and the fresh RPMI 1640 medium were analyzed by enzyme linked immunosorbent assay (ELISA). Then, to further identify whether TGF-β2 and PGE2 are involved in the inhibitory effect on DCs maturation mediated by CSCs, the neutralizing TGF-β2 antibody and the EP2 receptor antagonist AH6809 were applied. During the DCs maturation stage, different treatments were executed: 1, LPS; 2, LPS + 50% CSCs culture supernatant; 3, LPS + 50% CSCs culture supernatant + AH6809; 4, LPS + 50% CSCs culture supernatant + antibody; 5, LPS + 50% CSCs culture supernatant + AH6809 + antibody. Subsequently, the cellular surface markers for DCs, including CD11c, CD80, CD86, and MHC-Ⅱ, were analyzed by flow cytometry; the capability of stimulating the proliferation of T lymphocytes was evaluated by allogeneic mixed lymphocyte reactions and the function of endocytosis was assessed by fluorescein isothiocyanate-dextran uptake.Results: The data of RT-PCR indicated that primary murine CSCs exhibited high positive expression of TGF-β2 and PTGS2, low positive expression of M-CSF, and negative expression of IL-10; and the data of ELISA showed a higher concentration of TGF-β2 (1.46±0.38 ng/mL) and PGE2 (21.27±0.94 ng/mL) in murine CSCs culture supernatant than in the fresh RPMI 1640 medium. After adding neutralizing TGF-β2 antibody into CSCs culture supernatant, the phenotypic and functional modifications mediated by the supernatant were partially reversed (P<0.05 or P<0.01). After pretreating immature DCs with AH6809, the functional modification (P<0.05) and inhibition of CD86 and MHC-Ⅱexpression (P<0.05 or P<0.01) mediated by CSCs culture supernatant were also partially reversed, but the expression of CD80 was not altered (P>0.05). After applying AH6809 and neutralizing TGF-β2 antibody simultaneously, the expression of MHC-Ⅱand the capability of stimulating the proliferation of T lymphocytes were up-regulated, with statistical difference in interaction (P<0.05); the expression of CD86 and CD80 was elevated and the function of endocytosis were down-regulated, with no statistical difference in interaction (P>0.05). Furthermore, simultaneous application of AH6809 and neutralizing TGF-β2 antibody could not reversed the modification mediated by CSCs culture supernatant completely (P<0.05 or P<0.01).Conclusion: TGF-β2 and PGE2 contribute to the inhibitory effect on DCs maturation mediated by murine CSCs in vitro, and further have additive effect on the immunosuppression of DCs.