Co-transplantation of autologous MSCs delays islet allograft rejection and generates a local immunoprivileged site

Abstract

Aims: Mesenchymal stem cells (MSCs) are multipotent cells with immunomodulatory properties. We tested the ability of MSCs to delay islet allograft rejection.

Methods: Mesenchymal stem cells were generated in vitro from C57BL/6 and BALB/c mice bone marrow, and their immunomodulatory properties were tested in vitro. We then tested the effect of a local or systemic administration of heterologous and autologous MSCs on graft survival in a fully allogeneic model of islet transplantation (BALB/c islets into C57BL/6 mice).

Results: In vitro, autologous, but not heterologous, MSCs abrogated immune cell proliferation in response to alloantigens and skewed the immune response toward a Th2 profile. A single dose of autologous MSCs co-transplanted under the kidney capsule with allogeneic islets delayed islet rejection, reduced graft infiltration, and induced long-term graft function in 30 % of recipients. Based on ex vivo analysis of recipient splenocytes, the use of autologous MSCs did not appear to have any systemic effect on the immune response toward graft alloantigens. The systemic injection of autologous MSCs or the local injection of heterologous MSCs failed to delay islet graft rejection.

Conclusion: Autologous, but not heterologous, MSCs showed multiple immunoregulatory properties in vitro and delayed allograft rejection in vivo when co-transplanted with islets; however, they failed to prevent rejection when injected systemically. Autologous MSCs thus appear to produce a local immunoprivileged site, which promotes graft survival.

Keywords: Immunoprivileged site; Immunoregulation; Islet transplantation; Mesenchymal stem cells.

Fig. 1

Characterization of MSCs. C57BL/6 MSCs were characterized through flow cytometry, ELISPOTs, RT-PCR, and MSCs delineation. Flow cytometry was performed to characterize the phenotype of C57BL/6 MSCs. Results show that C57BL/6 highly express MHC Class I and CD13, but express low amounts of CD28, Thy-1, and CD34 (a). b Shows the delineation of MSCs into chondrocytes, osteoblasts, and adipocytes (i.e., color = positive effect) when MSCs were cultured accordingly to protocol (b). c and d, MSCs secretome was assessed and C57BL/6 MSCs were plated alone, MSCs produced high amount of IL-4 cytokine at a concentration of 100,000 cells and some percentage of IFN-γ (IFN-γ **p <0.01, IL-4 ***p <0.001) (c, d). Lastly, RT-PCR results show that C57BL/6 MSCs are positive for SDF-1α and TGF-β (e, f). Data are shown as mean ± SEM and are representative of n = 3 mice and of at least two independent experiments. **p <0.01; ***p <0.001

Fig. 2

Immunomodulatory properties of MSCs. 1 × 10⁶ C57BL/6 responder splenocytes were stimulated with irradiated stimulator BALB/c splenocytes in the presence or absence of 1 × 10⁵ autologous-C57BL/6 or heterologous BALB/c MSCs, and in an ELISPOT assay the production of INF-γ (a) and IL-4 was tested (c). While heterologous MSCs had only marginal effect on cytokines production (b, d), the use of autologous MSCs greatly modified cytokine production profile compared to cells cultured in the absence of MSCs (***p <0.001; b, d). An MLR experiment shows a dose-dependent reduction of C57BL/6 splenocytes challenged with irradiated BALB/c splenocytes and autologous MSCs (e). Data are shown as mean ± SEM and are representative of at least two independent experiments. ***p <0.001

Fig. 3

Survival of allogeneic islet transplant with (systemically or locally) or without MSCs. The systemic injection of autologous C57BL/6 MSCs is not able to promote allogeneic islet graft survival (a), on the contrary the local infusion (into kidney capsule together with islet graft) of the same number of MSCs promote graft survival (b) (**p <0.01 vs. untreated). Conversely, the local infusion of heterologous BALB/c MSCs has no effect on islet graft survival (c). P values were calculated with Student’s t test. Data are representative for at least five mice per group. **p <0.01

Fig. 4

Pathology of the graft at day 14 after transplantation. The islet graft of untreated mice showed no preserved structures or insulin staining (a, b, k), while in locally MSC-treated mice, the islet graft showed preserved structures and positive insulin staining (f, g, k). CD4⁺ and CD8⁺ T cell infiltrate is evident (l) but mostly confined outside the islets (h–i) in locally MSC-treated mice, as compared to untreated mice, which showed overall more infiltrate (c–d, l). CD31 staining, a marker of endothelium and vascularization, appeared to be more evident in islet graft of mice locally MSC-treated mice (j) as compared to samples obtained from untreated mice (e). All the generated data are representative of at least n = 3 mice per group. *p <0.05; **p <0.01

Fig. 5

Immunophenotype of treated mice. Proliferation toward donor antigens was reduced in the locally autologous MSC-treated mice compared to systemically MSC-treated mice (a). We found significant difference in terms of cytokine production toward graft antigens between untreated, locally and systemically MSC-treated mice, with a reduction in the number of IFN-γ-producing, and an increase in IL-4-producing, cells in mice treated with MSCs locally (b–c). The percentage of CD4⁺ effector T cells in the spleen (d) and in the blood (i) is unaffected by both treatments with MSCs, and MSC co-transplantation reduces the percentage of IL-17⁺ (e, J) and IFN-γ⁺ (f, k) cells. While the percentage of FoxP3⁺ regulatory CD4⁺ T cells is unaltered (h, m), a tenfold increase in the percentage of IL-10⁺ CD4⁺ T cells was evident in the spleen (g), but not in the blood (l), of mice locally treated with MSCs. Scale bars show mean ± SEM for at least n = 3 mice, and data are representative of two independent experiments. *p <0.05; **p <0.01; ***p <0.001

Fig. 6

MSCs tracking. In order to track administered MSCs in mice, GFP⁺MSCs were injected in islet-transplanted mice and GFP/CD105 double staining was performed and quantified (a–l). Flow cytometry analysis revealed that MSCs administered locally and systemically traffick to the spleen (d–f) but poorly to the liver (j–l). In untreated islet transplanted mice, WT mice, and Isotype control, GFP/CD105 double-positive cells are poorly represented (a–c, g–i). Data are shown as mean ± SEM and are representative of n = 3 mice and of at least two independent experiments. **p <0.01