Mesenchymal stem/stromal cells protect the ocular surface by suppressing inflammation in an experimental dry eye

Abstract

Dry eye syndrome (DES) is one of the most common ocular diseases affecting nearly 10% of the US population. Most of the currently available treatments are palliative, and few therapeutic agents target biological pathway of DES. Although DES is a multifactorial disease, it is well-known that inflammation in the ocular surface plays an important role in the pathogenesis of DES. Mesenchymal stem/stromal cells (MSCs) have been shown to repair tissues by modulating excessive immune responses in various diseases. Therefore, we here investigated the therapeutic potential of MSCs in a murine model of an inflammation-mediated dry eye that was induced by an intraorbital injection of concanavalin A. We found that a periorbital administration of MSCs reduced the infiltration of CD4(+) T cells and the levels of inflammatory cytokines in the intraorbital gland and ocular surface. Also, MSCs significantly increased aqueous tear production and the number of conjunctival goblet cells. Subsequently, corneal epithelial integrity was well-preserved by MSCs. Together, the results demonstrate that MSCs protect the ocular surface by suppressing inflammation in DES, and suggest that MSCs may offer a therapy for a number of ocular surface diseases where inflammation plays a key role.

Figures

Figure 1

Establishment of inflammation-induced dry eye in mice. (a) Concanavalin A (ConA) was injected into an intraorbital space in mice. Phosphate-buffered solution (PBS) was injected as vehicle control. One week later, the tissues were subjected to assays. (b) Immunostaining showed extensive infiltration of CD3+ T cells in the intraorbital gland. (c) Aqueous tear production as measured by phenol red thread test was significantly reduced by ConA injection. (d,e) Real-time RT-PCR assay showed that either 10 or 20 μl injection of ConA (10 mg/ml) significantly increased the levels of IL-2 and IFN-γ transcripts which are the cytokines derived from activated T cells. (f) In the ocular surface including the cornea and conjunctiva, the levels of IL-2 and IFN-γ transcripts were increased by ConA injection (10 mg/ml, 20 μl), while the levels of TNF-α, IL-1β, and IL-6 were not affected. (g) The corneal dye staining indicated the development of corneal epithelial erosion in ConA-injected eyes. n = 10 to 20 in each group. The data are presented as the mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 2

Effects of mesenchymal stem/stromal cells in inflammation-induced dry eye. (a) After concanavalin A (ConA) injection (10 mg/ml, 20 μl), human or mouse bone marrow-derived mesenchymal stem/stromal cells (hMSC, mMSC; 1 × 103 or 1 × 105 cells/20 μl balanced salt solution; BSS), human dermal fibroblasts (hFb; 1 × 105 cells/20 μl BSS), or BSS were administered periorbitally. One week later, the tissues were subjected to assays. (b) The amount of aqueous tear production as assayed by phenol red thread test was significantly higher in the hMSC- or mMSC-treated eyes, compared to the BSS- or hFb-treated eyes. (c,d) The levels of IL-2 and IFN-γ transcripts were significantly lower in the intraorbital gland and ocular surface of the hMSC- or mMSC-treated group. However, the levels of IL-2 and IFN-γ were markedly increased in the hFb-treated eyes. (e) The corneal epithelium had severe punctate erosions in the BSS-treated control, and the corneal staining score was lower in the hMSC- or mMSC-treated eyes than in the BSS- or hFb-treated eyes. n = 10–20 in each group. The data are presented as the mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 3

Histologic and flow cytometric analyses of the intraorbital gland. (a) Representative picture of hematoxylin-eosin staining showed disruption of the intraorbital gland structure with extensive acinar atrophy and periductal infiltration of inflammatory cells 7 days after concanavalin A (ConA) injection (10 mg/ml, 20 μl). Most of the infiltrating inflammatory cells were CD3+ T cells. In contrast, the structure of the intraorbital gland was well preserved with few inflammatory infiltrates by treatment with human or mouse bone marrow-derived mesenchymal stem/stromal cells (hMSC, mMSC). (b) Flow cytometric analysis of cells revealed that the percentages of CD4+ cells or IFN-γ+CD4+ cells in the intraorbital gland were increased by ConA injection, and significantly reduced by treatment with hMSC. The data are presented as the mean ± SEM. *P < 0.05; **P < 0.01.

Figure 4

PAS staining of the conjunctiva. (a) Representative photograph of Periodic Acid Schiff (PAS) staining of the conjunctiva showed that the number of goblet cells was markedly reduced by concanavalin A (ConA) injection (10 mg/ml, 20 μl), and improved by treatment with human or mouse bone marrow-derived mesenchymal stem/stromal cells (hMSC, mMSC). (b) The goblet cell counts in the conjunctiva were significantly higher in the hMSC- or mMSC-treated eyes, compared to the BSS-treated controls. The data are presented as the mean ± SEM. **P < 0.01; ***P < 0.001.

Figure 5

Effects of mesenchymal stem/stromal cells on T-cell proliferation and differentiation. (a) Fluorescence carboxyfluorescein succinimidyl ester (CFSE)-labeled murine CD4+ cells were stimulated with ConA (5 µg/ml), and co-cultured with human bone marrow-derived mesenchymal stem/stromal cells (hMSC) in transwell for 5 days. Flow cytometry to detect CFSE dilution showed that CD4+ cell proliferation was significantly suppressed by co-cultures with hMSC. (b) CFSE-labeled murine CD4+ cells were stimulated with anti-CD3/CD28 antibodies (2.5 µg/ml), and co-cultured with either hMSC or human dermal fibroblasts (hFb) for 5 days. CD4+ cell proliferation was significantly suppressed by co-cultures with hMSC, but not by hFb. (c) CD4+ cells were co-cultured with hMSC in transwell for 5 days in the presence of IL-1β and IL-23, and evaluated for IFN-γ expression by flow cytometry. Flow cytometry demonstrated that hMSC markedly reduced the percentage of IFN-γ–secreting cells indicating the suppression of Th1 differentiation by hMSC. n = 3 in each group. Data are presented in mean ± SD and represent three independent experiments. *P< 0.05; ***P < 0.001.