Intravenous mesenchymal stem cells prevented rejection of allogeneic corneal transplants by aborting the early inflammatory response

Abstract

Mesenchymal stem/progenitor cells (MSCs) were reported to enhance the survival of cellular and organ transplants. However, their mode of action was not established. We here used a mouse model of corneal allotransplantation and demonstrated that peri-transplant intravenous (i.v.) infusion of human MSCs (hMSCs) decreased the early surgically induced inflammation and reduced the activation of antigen-presenting cells (APCs) in the cornea and draining lymph nodes (DLNs). Subsequently, immune rejection was decreased, and allograft survival was prolonged. Quantitative assays for human GAPDH revealed that <10 hMSCs out of 1 × 10(6) injected cells were recovered in the cornea 10 hours to 28 days after i.v. infusion. Most of hMSCs were trapped in lungs where they were activated to increase expression of the gene for a multifunctional anti-inflammatory protein tumor necrosis factor-α stimulated gene/protein 6 (TSG-6). i.v. hMSCs with a knockdown of TSG-6 did not suppress the early inflammation and failed to prolong the allograft survival. Also, i.v. infusion of recombinant TSG-6 reproduced the effects of hMSCs. Results suggest that hMSCs improve the survival of corneal allografts without engraftment and primarily by secreting TSG-6 that acts by aborting early inflammatory responses. The same mechanism may explain previous reports that MSCs decrease rejection of other organ transplants.

Figure 1

Intravenous (i.v.) human mesenchymal stem cells (hMSCs) prolonged the survival and prevented the immune rejection of B6 corneal grafts in BALB/c mice. (a) Representative photographs of the cornea 14 days after transplantation and the Kaplan–Meier survival curve of corneal grafts. The graft survival was significantly prolonged by i.v. hMSCs. Seven out of 12 B6 corneal grafts in BALB/c mice (allografts) were rejected within 28 days (median survival time 23.1 days). In contrast, 11 out of 12 allografts survived in mice that received i.v. hMSCs both one day before surgery (day 1) and immediately after surgery (day 0), and 9 of 12 allografts survived in mice that received a single injection of hMSCs at day 0 (MSC at day 1 and 0 vs. Hank's balanced salt solution (HBSS), P = 0.006; MSC at day 0 vs. HBSS, P = 0.047; generalized Wilcoxon test). n = 12 in each group.(b) Hematoxylin–eosin staining of corneal grafts at day 28 showed heavy infiltration of inflammatory cells in the rejected allografts of control animals and much less inflammatory cell infiltration in the allografts that received hMSCs. (c) Immunohistochemical staining showed that many CD3⁺ T cells infiltrated the allografts of control animals, whereas there were rare T cells in the grafts that received hMSCs.

Figure 2

Time course of gene expression levels in the cornea after transplantation surgery. Real-time reverse transcription (RT)-PCR showed that the levels of proinflammatory cytokines (interleukin (IL)-6, IL-1β, and IL-12a) were upregulated at similar levels in (a) autografts and (b) allografts at days 3 and 7 after transplantation, which defines the early phase of surgery-induced inflammation. The levels of T cell-derived cytokines (IFN-γ) were increased up to day 28 in allografts, but not in autografts, which indicates the late phase of the allogeneic immune rejection. In allografts that received intravenous (i.v.) human mesenchymal stem cells (hMSCs) (c), levels of IL-6, IL-1β, and IL-12a were significantly lower at days 3 and 7, and levels of IFN-γ were markedly decreased at day 28 compared to autografts or allografts that did not receive hMSCs. n = 5 at each time-point in all experimental groups.

Figure 3

Intravenous (i.v.) human mesenchymal stem cells (hMSCs) decreased the early inflammatory response in corneal allografts. (a) The amount of myeloperoxidase as a measure of neutrophil infiltration was significantly decreased by hMSCs at day 7 after transplantation. (b–f) Also, the levels of proinflammatory cytokines, IL-6, IL-1β, and IL-12 were significantly decreased by i.v. hMSCs at day 7. Results indicate that inflammatory responses in the early postoperative period after transplantation surgery were suppressed by i.v. hMSCs. Auto: autografts, Allo: allografts, Allo MSCx1: allografts that received hMSCs once immediately after transplantation, Allo MSCx2:allografts that received hMSCs on the day before transplantation and immediately after transplantation. n = 5 in each group. *P < 0.05; **P < 0.01.

Figure 4

Intravenous (i.v.) human mesenchymal stem cells (hMSCs) suppressed the late T cell-mediated immune response in corneal allografts. (a,b) The transcript levels of activated CD4 T-cell cytokines (IL-2 and IFN-γ) and protein level of (c) IFN-γ were significantly decreased in the allografts that received i.v. hMSCs at day 28 after surgery, compared to the grafts without i.v. hMSCs. (d–f) Also, the transcript levels of CD8 T-cell effector molecules (granzyme A, granzyme B, and perforin) were significantly decreased by i.v. hMSCs. Auto: autografts, Allo: allografts, Allo MSCx1: allografts that received hMSCs once immediately after transplantation, Allo MSCx2:allografts that received hMSCs on the day before transplantation and immediately after transplantation. n = 5 in each group. *P < 0.05; **P < 0.01.

Figure 5

Intravenous (i.v.) human mesenchymal stem cells (hMSCs) decreased the number of MHC class II⁺cells in the cornea 1 week following transplantation. Immunostaining for (a) murine Ia^d+ and enumeration of MHC class II⁺cells in the whole-mounted epithelial sheets of the cornea (b) showed that the number of MHC class II⁺cells indicating Langerhans cells was significantly increased both in autografts and allografts in response to transplantation surgery. However, the number of MHC class II⁺cells was significantly decreased in the allografts by i.v. hMSCs. n = 3 in each group. Auto: autografts, Allo: allografts, Allo MSCx1: allografts that received hMSCs once immediately after transplantation, Allo MSCx2:allografts that received hMSCs on the day before transplantation and immediately after transplantation.

Figure 6

Intravenous (i.v.) human mesenchymal stem cells (hMSCs) decreased the number of activated antigen-presenting cells in ipsilateral cervical lymph nodes 1 week following transplantation. (a) Flow cytometry showed that the proportions of dendritic cells (DCs), both MHC II⁺CD11b⁺CD11c⁺ cells (b) and MHC II⁺CD11b⁻CD11c⁺ cells (c),were significantly decreased in mice treated with hMSCs. In addition, the proportion of MHC II⁺CD11b⁺CD11c⁻cells representing macrophages was significantly decreased by i.v. (d) hMSCs. n = 4 in each group. Auto: autografts, Allo: allografts, Allo MSCx1: allografts that received hMSCs once immediately after transplantation, Allo MSCx2:allografts that received hMSCs on the day before transplantation and immediately after transplantation.

Figure 7

intravenous (i.v.) human mesenchymal stem cells (hMSCs) with TSG-6 siRNA knockdown did not suppress the early surgery-induced inflammation and did not prolong the survival of corneal allografts. (a) The Kaplan–Meier survival curve of B6 corneal grafts in BALB/c mice. Three out of six allografts were rejected in mice that received a two-time injection of hMSCs with TSG-6 knockdown (TSG-6 siRNA MSC), whereas all allografts survived in mice that received a two-time injection of hMSCs with scrambled siRNA (SCR MSC). (b) The myeloperoxidase (MPO) amount as a measure of neutrophil infiltration was not significantly decreased by hMSCs with TSG-6 knockdown at day 7 after grafting. n = 3 in each group. The levels of activated T cell-derived cytokines (c–e) and effector enzymes (f–h) were not decreased in corneal grafts of mice treated with TSG-6 knockdown hMSCs, whereas hMSCs with scrambled siRNA significantly decreased the levels of T-cell cytokines and enzymes in corneal grafts. n = 3 in each group. Auto: autografts, Allo: allografts, Allo MSCx1: allografts that received hMSCs once immediately after transplantation, Allo MSCx2:allografts that received hMSCs on the day before transplantation and immediately after transplantation. n = 6 in each group. *P < 0.05; **P < 0.01.

Figure 8

Intravenous (i.v.) injection of recombinant TSG-6 suppressed the early surgery-induced inflammation and the late immune rejection of the corneal allografts. (a) The Kaplan–Meier survival curve of B6 corneal grafts in BALB/c mice. Six out of nine allografts survived in mice that received a single injection of rhTSG-6, whereas two out of nine allografts survived in mice that received phosphate-buffered saline (PBS). (b–f) The myeloperoxidase (MPO) amount and the levels of transcripts for proinflammatory cytokines were significantly decreased in the corneal allografts at day 7 by i.v. recombinant TSG-6 (35 µg) injected immediately after surgery. n = 3 in each group. (g–j) The levels of transcripts for T cell-related cytokines were also decreased by i.v. recombinant TSG-6 at day 28. n = 6 in each group.