Overexpression of Galectin-9 in Islets Prolongs Grafts Survival via Downregulation of Th1 Responses

Abstract

The differential activation of T helper (Th) cells and production of cytokines contribute to graft rejection or tolerance. In general, the Th1-type cytokines and cytotoxic T-cells are detected consistently in a host who is undergoing rejection, whereas Th2 responses are linked to a tolerance condition. Galectin-9 modulates Th1 cell immunity by binding to the T-cell immunoglobulin mucin-3 (Tim-3) molecule expressed on the Th1 cells. We investigate whether overexpression of galectin-9 in islets prolongs grafts survival in diabetic recipients. Islets were transduced with lentiviruses carrying galectin-9 and were then transplanted to streptozotocin-induced diabetic NOD/SCID recipients. The normoglycemic recipients then received splenocytes from diabetic NOD mice. Blood glucose concentration was monitored daily after adoptive transfer. The histology of the islet grafts and flow cytometric analyses were assessed at the end of the study. Overexpression of galectin-9 in islets prolonged graft survival in NOD/SCID mice after challenge with diabetogenic splenocytes (mean graft survival, 38.5 vs. 26.0 days, n = 10, respectively; p = 0.0096). The galectin-9-overexpressed grafts showed decreased infiltration of IFN-γ-producing CD4⁺ and CD8⁺ T-cells, but not of IL-17-producing CD4⁺ T-cells. Strikingly, this islet-specific genetic manipulation did not affect the systemic lymphocyte composition, indicating that galectin-9 may regulate T-cell-mediated inflammation in situ. We demonstrate that galectin-9 protects grafts from Th1 and Tc1 cellmediated rejections, suggesting that galectin-9 has preventive and/or therapeutic benefit in transplant therapy for autoimmune diabetes and may be applied further to the transplantation of other organs or tissues.

Keywords

Islet transplantation Galectin-9 T-cell immunoglobulin mucin-3 (Tim-3)T helper 1 (Th1) cells Autoimmune diabetes

Introduction

Galectin-9 is a β-galactoside binding protein that was originally found to induce thymocytes apoptosis in mouse (29). Recently, galectin-9 was shown to have a variety of functions in the immune system, including the induction of apoptosis of activated T-cell (11) and modulation of inflammatory response (1). T-cell immunoglobulin mucin-3 (Tim-3) is expressed mainly on interferon (IFN)-γ-secreting CD4⁺ T helper 1 (Th1) and CD8⁺ T cytotoxic 1 (Tc1) cells and acts as a cellular receptor of galectin-9 (15, 36). Previous reports have demonstrated that the galectin-9–Tim-3 pathway controls the T-cell-mediated immune responses in many disease models. For example, overexpression of galectin-9 ameliorates the Th1 cell-mediated autoimmunity in encephalomyelitis, arthritis, and diabetes models (3,4,36). Conversely, disruption of Tim-3 pathway by anti-Tim-3 antibody skews the T-cell response toward Th1 immunity in allergic disease (12). Moreover, administration of Tim-3-Ig fusion protein leads to resistance to acquisition of the tolerance to transplantation induced by donor-specific transfusion plus costimulation blockade (22). These findings indicate that the galectin-9–Tim-3 pathway plays a pivotal role in the control of Th1 immunity.

Islet transplantation is considered an appropriate treatment to achieve insulin independence and provides improved metabolic control in type 1 diabetic patients. However, autoimmune-mediated destruction is one of the major obstacles to the long-term survival of islet grafts in this potential therapy. Previous reports have demonstrated that Th1 cells and Tc1 cells are highly closed to the rejection of islet grafts and also participated in other organ rejection responses (10,25,26,30). Therefore, to improve the outcomes of islet transplantation and control Th1 immunity, we tested the hypothesis that over-expression of galectin-9 in islet grafts attenuates the Th1-mediated graft destruction. In the present study, we demonstrated that the genetically engineered galectin-9-expressing islets down regulate the detrimental Th1- and Tc1-mediated rejection responses within the graft microenvironment and prolong graft survival.

Materials and Methods

Animals

Nonobese diabetic (NOD)/Sytwu (K^d, D^b, L^d, I-A^g7) mice were purchased originally from the Jackson Laboratory (Bar Harbor, ME, USA) and NOD/severe combined immunodeficient (SCID) mice were purchased from Tzu Chi University (Hualian, Taiwan). All mice were bred and maintained under specific pathogen-free conditions at the Animal Center of the National Defense Medical Center (Taipei, Taiwan), which was accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). For induction of diabetes in NOD/SCID mice, streptozotocin (STZ) (Sigma, St. Louis, MO, USA) was prepared in sodium citrate buffer (5 μM, pH 4.5; Sigma) and was given to mice at 50 mg/ kg per day for 5 days consecutively.

Construction of the Plasmid and Generation of the Lentiviral Vector

Mouse galectin-9 cDNA was cloned from splenocytes as previously described (4) and inserted into a pWPI-internal ribosome entry site-enhanced green fluorescent protein (pWPI-IRES-eGFP)-transducing self-inactivating (SIN) vector, which drives the target gene with a elongation factor 1α promoter and coexpresses eGFP by the IRES sequence. Vesicular stomatitis virus glycoprotein (VSV-G)-pseudotyped recombinant HIV-based virus was produced by three-plasmid cotransfection of TE671 cells (ATCC, Manassas, VA, USA; number CCL-136) with the packaging helper construct, the envelope expression construct, and the transducing self-inactivating vector. The viruses were concentrated by ultracentrifugation and titered by transduction of confluent TE671 cells.

Islet Isolation and Viral Transduction

Islets were purified from 6- to 8-week-old male NOD mice using the collagenase-digesting method as described previously (5). Purified islets were suspended in 0.5 ml of culture medium containing 8 mg/ml polybrene (Sigma) and infected with lentivirus at a multiplicity of infection (MOI) of 10. An islet is a three-dimensional structure that is composed of a, b, d, e and pancreatic polypeptide producing (PP) cells and contains on average 1,000 cells. The MOI was calculated according to the assumption that islets contain on average 1,000 cells. Islets were incubated at 37°C for 3 h with lentivirus in about 0.5 ml of medium and then cultured in F12K (Invitrogen, Carlsbad, CA, USA) medium supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin (10,000 units/ml), and 1% l-glutamine (2 mmol/l) (Invitrogen) at 37°C in 5% CO₂ before use for transplantation. When using the replication-defective lentiviral vectors to infect islets, only part of cells surrounded of the islet could be efficiently infected and the transduction efficiency is around 5% of total islet cells.

Immunoblot Analysis of Galectin-9 Expression

NIT-1 cells (ATCC number CRL-2055) were transduced with lentiviruses carrying eGFP and galectin-9 at an MOI of 10. Total protein from cellular lysates was prepared, size fractionated using 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (Invitrogen), and electroblotted onto a nitrocellulose membrane (Millipore Corporation, Billerica, MA, USA). Galectin-9 was detected using a goat anti-mouse galectin-9 antibody (R&D Systems, Minneapolis, MN, USA). Anti-β-actin (Sigma) and anti-tubulin (Abcam plc, Cambridge, UK) antibodies were used for the detection of the expression of internal control proteins.

Isolation of splenocytes

Splenocytes were isolated following a previously published protocol (13). Briefly, spleen was sliced into small pieces and pressed with plunger of a syringe (BD Pharmingen, San Jose, CA, USA) in a petri dish (Techno Plastic Products AG, Zollstrasse, Trasadingen, Switzerland). Then the cell suspension was expelled into a centrifuge tube (BD Pharmingen) through a 200-μm nylon cell strainer (BD Pharmingen). Cells were further treated with Trisbuffered ammonium chloride (Sigma) for erythrocyte depletion.

Islet Transplantation and Adoptive Transfer

Twenty-four hours after virus transduction, marginal islets were collected and washed, and ~400 islets were implanted into the left renal capsule of streptozotocininduced diabetic NOD/SCID mice with a blood glucose concentration of 300-500 mg/dl. Monitoring of the blood glucose of mice was performed by applying one drop of blood from the tail vein to a blood glucose test strip and glucose was detected by a FreeStyle Optium Blood Glucose Monitoring System (Abbott, Abbott Park, IL, USA). At 4-6 days after transplantation, the normoglycemic recipients received 2 × 10⁷ splenocytes isolated from diabetic NOD mice, via intraperitoneal injection. Blood glucose concentration was monitored daily after islet transplantation. Loss of graft function was defined as a blood glucose concentration of >300 mg/dl on two consecutive days.

Flow Cytometry

Control and experimental paired recipients were sacrificed at the time at which control islet grafts were destroyed (26 days on average). Mice were perfused with normal saline to remove blood cells, and lymphocytes from the spleen, pancreatic lymph node (PLN), and grafts were harvested for flow cytometric analysis. For intracellular cytokine staining, cells were stimulated with phorbol 12-myristate 13-acetate (PMA) and ionomycin in the presence of monensin (all Sigma) for 4 h, which was followed by intracellular cytokine staining. Cells were stained with anti-cluster of differentiation 4 (CD4), anti-CD8, anti-CD69, anti-CD25, anti-interferon (IFN)-γ, anti-interleukin (IL)-17, and anti-forkhead box P3 (FOXP3) antibodies and mouse IgG1 isotype (eBioscience, San Diego, CA, USA). Cells were analyzed using a FACS-Calibur^™ flow cytometer and the CELLQUEST software (BD Pharmingen, San Jose, CA, USA).

Histology

For histology, graft-bearing kidneys were removed, embedded in OCT Compound (Sakura Finetek, Torrance, CA, USA), and cryosectioned. The sections were stained with hematoxylin and eosin (Sigma). For immunohistochemical analysis, tissue sections were stained with either anti-insulin (Abcam plc) or anti-galectin-9 (R&D Systems, Minneapolis, MN, USA) primary antibodies, followed by a horseradish peroxidase (HRP)-conjugated secondary antibody (R&D Systems). Chromogen substrate 3-amino-9-ethylcarbazole (Sigma) was added for enzymatic stain development, which resulted in a rose red to brownish red precipitate at the antigen sites. Mayer's hematoxylin and eosin (Sigma) was applied for H&E stain or counterstain (Mayer's hematoxylin).

Methods for the supplementary data (ERK phosphorylation, glucose-stimulated insulin secretion test, cell viability and ELISAs for IFN, IL-17 and galectin-9, suppression of T-cell proliferation in a coculture system) can be found in the online supplementary material (https://skydrive.live.com/?cid=355A165A5117E0AE&id=355A165A5117E0AE!105 and back up site: https://docs.google.com/file/d/0ByGzRMkw5qqISFI2VUg2TzJqdzQ/edit?usp=sharing).

Statistical Analysis

Prism v4.00 software (GraphPad, San Diego, CA, USA) was used for graph generation and statistical analysis. Graft survival curve was analyzed by Kaplan-Meier test. The comparison of T-cell populations between two groups was made by unpaired t test. A value of p < 0.05 was defined as significant.

Results

Galectin-9 Expression in Lentiviral Vector-Transduced Cells

We have demonstrated previously that systemic overexpression of galectin-9 downregulates the Th1 cell population and reduces the diabetic incidence in a nonobese diabetic mouse model (4). To investigate further the therapeutic application of galectin-9 in transplantation for autoimmune diabetes, we used a galectin-9-carried lentiviral vector to transduce islets and investigated whether overexpression of galectin-9 prolongs the survival of islet grafts after challenge with diabetogenic splenocytes. We used a dual expression lentiviral system that expresses galectin-9 and eGFP (Lt-Gal-9-IRES-eGFP) simultaneously. Therefore, expression of the transgene was monitored directly by the detection of the expression of eGFP. First, we evaluated the transduction efficiency and determined the galectin-9 protein expression in NIT-1 insulinoma cells. At a MOI of 10, lentiviruses transduced NIT-1 cells efficiently, as monitored using fluorescent microscopic imaging of eGFP protein expression (Fig. 1A). Expression of the galectin-9 protein in Lt-Gal-9-IRES-eGFP-transduced NIT-1 cells was confirmed by immunoblot analysis (Fig. 1B). We tested further the transduction efficiency of this lentivirus in islets. Islets were isolated from 6- to 8-week-old male NOD mice and morphologically intact islets were selected for the transplantation therapy experiment. The freshly isolated islets were transduced successfully by the lentiviruses carrying eGFP (Fig. 1C) and galectin-9 (Fig. 1D). As galectin-9 is expressed widely in different tissues and will be induced by IFN-γ or IL-1β (8), our results reveal that galectin-9 is barely detected in islets transduced with lentiviruses carrying eGFP only, indicating that islets are not a tissue that expresses galectin-9 endogenously without stimulation by proinflammatory cytokines. To determine the possible effects of galectin-9 on β-cells, in terms of cell growth, extracellular signalregulated kinase 1/2 (ERK1/2) signaling (28), and insulin secretion, we determined the islet cell viability and analyzed the phosphorylation level of ERK1/2 after lentiviral transduction. Besides, we also evaluated the islet function by a glucose-stimulated insulin secretion test after the lentiviral transduction. Lt-Gal-9-IRES-eGFP-transduced islets displayed similar cell viability, ERK1/2 phosphorylation level and insulin secretion levels, compared to the Lt-eGFP-transduced islets (Supplementary Fig. 1; https://skydrive.live.com/?cid=355A165A5117E0AE&id=355A165A5117E0AE!105 and back up site: https://docs.google.com/file/d/0ByGzRMkw5qqISFI2VUg2TzJqdzQ/edit?usp=sharing). Taken together, these results demonstrate that the lentiviral vector delivers the target genes into islets effectively and represents a potential tool for the genetic manipulation of these almost nondividing islets.

Figure 1.

Transduction of NIT-1 and islet cells using a lentiviral vector. (A) Fluorescence microscopy image displaying the expression of the eGFP protein in lentivirus-galectin-9-internal ribosome entry site-enhanced green fluorescent protein vector (Lt-Gal-9-IRES-eGFP)-transduced NIT-1 cells. Original magnification: 200×. (B) Immunoblot analysis of galectin-9 expression in Lt-Gal-9-IRES-eGFP-transduced NIT-1 cells. (C) The expression of the eGFP protein in Lt-Gal-9-IRES-eGFP-transduced islet cells was monitored by fluorescence microscopy. Original magnification: 100×. (D) Immunoblot analysis of galectin-9 expression in Lt-Gal-9-IRES-eGFP-transduced islets. Multiplicity of infection (MOI) = 10.

Survival of Galectin-9-Transduced Islet Grafts After Challenge with Diabetogenic Splenocytes

Systemic administration of galectin-9 protein not only regulates T-cell immune responses in several autoimmune diseases (23,36) but also prolongs the survival of skin and heart allografts (6,32,33). To investigate whether galectin-9 has a therapeutic potential to protect islet grafts from autoimmune attack in a tissue-specific manner, we used an autoimmune diabetes model and evaluated the regulatory effects of galectin-9 in islet transplantation. Lt-Gal-9-IRES-eGFP-transduced islets and Lt-eGFP-transduced islets were transplanted into streptozotocin-induced diabetic NOD/SCID mice, respectively, which were then challenged with splenocytes from newly diabetic NOD mice. Islet grafts overexpressing galectin-9 showed increased resistance to autoimmune attack compared with control islets (mean graft survival, 38.5 vs. 26.0 days, n = 10, respectively; p = 0.0096) (Fig. 2). To evaluate the function of and transgene expression in islet grafts in vivo, islet-bearing kidneys were removed on day 26 (i.e., the mean time of graft survival in the control group) for histological analysis. Lt-Gal-9-IRES-eGFP-transduced islet grafts showed less leukocytic infiltration than the control islet grafts (Fig. 3A, B). Tissue sections were stained immunohistochemically with anti-insulin (Fig. 3C, D) or anti-galectin-9 (Fig. 3E, F) antibodies. Galectin-9-expressing islet grafts maintained insulin-secreting function and expressed galectin-9 stably, whereas the control group expressed negligible levels of insulin and no, or undetectable, levels of endogenous galectin-9. In summary, islet-specific overexpression of galectin-9 reduced the infiltration of lymphocytes in grafted area, which led to the subsequent maintenance of the function of islet grafts.

Figure 2.

Islet graft survival in NOD/SCID recipients after challenge with splenocytes from diabetic mice. Lt-Gal-9-IRES-eGFP-transduced islets and Lt-eGFP-transduced islets were transplanted into streptozotocin (STZ)-induced diabetic nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice, respectively. The normoglycemic recipients then received 2 × 10⁷ splenocytes from diabetic mice and the blood glucose concentration was monitored daily after islet implantation. Loss of graft function was defined as a blood glucose concentration of >300 mg/dl on two consecutive days. Kaplan–Meier method, Lt-Gal-9-IRES-eGFP versus Lt-eGFP, p = 0.0096.

Figure 3.

Histology of islet grafts (IGs) under the capsule (CA) of the kidney (K). Lt-eGFP- and Lt-Gal-9-IRES-eGFP-transduced islets were transplanted into STZ-induced diabetic NOD/SCID mice, respectively, followed by transfer of splenocytes from newly diabetic NOD mice. IG-implanted kidneys were removed on the day at which the Lt-eGFP-transduced IG failed but the Lt-Gal-9-IRES-eGFP-transduced IG was still functional and maintained euglycemia. Hematoxylin and eosin (H&E) stain of IGs from mice that received Lt-eGFP-transduced islets (A) and Lt-Gal-9-IRES-eGFP-transduced islets. (B) Sections of Lt-eGFP-transduced or Lt-Gal-9-IRES-eGFP-transduced IGs were stained immunohistochemically with anti-insulin (C, D) and anti-galectin-9 antibodies (E, F). Scale bars: 50 μm (A, B), 200 μm (C, D), and 20 μm (E, F).

Flow Cytometric Analyses of Differential T-Cell Subsets in NOD/SCID Recipients Challenged with Diabetogenic Splenocytes

In the present study, we take advantage of the lentiviral vector to effectively and specifically deliver galectin-9 to islet grafts, which attenuates leukocytic infiltration at graft sites. To investigate further the activation status and differential populations of T-cells in NOD/SCID recipients challenged with diabetogenic splenocytes, we analyzed the lymphocyte composition, homing, and production of effector cytokines of peripheral lymphoid organs and grafted tissues at days 20–26 after adoptive transfer. The CD4⁺ and CD8⁺ T-cells from Lt-eGFP and Lt-Gal-9-IRES-eGFP groups displayed a similar activation phenotype (Fig. 4A) and similar IFN-γ expression levels (Fig. 4B, C) in spleen and PLN (the primary homing lymph node of diabetogenic lymphocytes). To evaluate whether the galectin-9-mediated graft protection is through the downregulation of Th1 or Tc1 immune response at the inflammatory lesion, we isolated graft-infiltrated lymphocytes (GIL) and analyzed them for the presence of IFN-γ-producing T-cells. Strikingly, the Lt-Gal-9-IRES-eGFP group exhibited a significant decrease in IFN-γ-producing CD4⁺ T-cells and CD8⁺ T-cells (Fig. 4B, C, right). Recently, Th17 cells have been demonstrated to mediate several allogeneic organ and tissue rejection responses (7). To evaluate the effects of galectin-9 overexpression on Th17 cells in the islet transplantation model, we analyzed the Th17 cell population in the spleen, PLN, and islet grafts from both groups of recipients challenged with diabetogenic splenocytes. There was no significant difference in the Th17 cell population in the spleen, PLN, and islet graft of the two groups. Indeed, there was a much lower infiltration of Th17 cells in the islet graft compared with Th1 cells (Fig. 4D). These results suggest that islet graft destruction is mediated mainly by Th1 cells and that Th17 cells may participate but play a minor role in islet graft rejection in this model. As regulatory T-cells play crucial roles in transplant tolerance, we checked further the CD4⁺CD25⁺FOXP3⁺ regulatory T-cell population in these two groups of recipients. There were no significant changes in the regulatory T-cell population between the Lt-GPF and Lt-Gal-9-IRES-eGFP groups in the spleen, PLN, and GIL (Fig. 4E). To address the issue regarding the decreased Th1 and Tc1 cells in the galectin-9-expressing islet grafts, we counted the absolute cell numbers of spleen, PLN, and GIL in reconstituted recipients in Table 1 and summarized the cell numbers of each T-cell subsets of GIL in Table 2. Our data revealed that a significant decrease of total cell numbers was observed only in the galectin-9-overexpressed islet grafts, compared to Lt-eGFP-transduced islet grafts, whereas the total cell numbers in spleen or PLN showed no significant difference between two groups. Besides, we collected cells from spleen, PLN, and graft of reconstituted recipients for ELISA assay. The T-cell compositions displayed no significant difference between Lt-eGFP and Lt-Gal-9-IRES-eGFP groups in all three sources and the cells were stimulated with 0.1 mg plate-bound anti-CD3 antibody for 48 or 72 h. Cells from spleen or PLN secreted similar levels of IFN-γ and IL-17 between Lt-eGFP and Lt-Gal-9-IRES-eGFP groups; however, the graft-infiltrated cells from Lt-Gal-9-IRES-eGFP group produced much lower amount of IFN-γ and moderately lower IL-17 than the Lt-eGFP group (Supplementary Fig. 2; https://skydrive.live.com/?cid=355A165A5117E0AE&id=355A165A5117E0AE!105 and back up site: https://docs.google.com/file/d/0ByGzRMkw5qqISFI2VUg2TzJqdzQ/edit?usp=sharing). Taken together, our results demonstrate that local expression of galectin-9 protected islets against the T-cell attack in situ via the downregulation of Th1 and Tc1 cell responses specifically at the graft microenvironment.

Figure 4.

Flow cytometric analysis of the T-cell population in NOD/SCID mice after challenge with splenocytes from diabetic mice. Lymphocytes from the spleen, pancreatic lymph node (PLN), and graft-infiltrated lymphocytes (GIL) of paired NOD/SCID recipients were isolated and analyzed using flow cytometry. The activation status of the T-cells in the spleen and PLN was evaluated via staining using the early activation marker cluster of differentiation 69 (CD69) (A). The CD4⁺interferon-γ (IFN-γ)⁺ (B), CD8⁺IFN-γ⁺ (C), and CD4⁺interleukin 17 (IL-17)⁺ (D) populations in the spleen, PLN, and GIL were analyzed via intracellular cytokine staining. (E) CD4⁺CD25⁺forkhead box P3 (FOXP3)⁺ T-cells in the spleen, PLN, and GIL. Graft tissues were dissected after removal of peripheral blood cells using perfusion, and graft-infiltrated lymphocytes were analyzed via intracellular cytokine staining. Eight to 11 pairs of mice were used for these analysis and the results of statistical analyses are shown in the bottom right corner of each panel. Unpaired t test; significance was set at p < 0.05.

Table 1.

Absolute Cell Numbers of Spleen, PLN, and GIL in Reconstituted Recipients

Total Cell Number Spleen (×10⁶)	Lt-eGFP 21.12±6.04	Lt-Gal-9-IRES-eGFP 17.48±4.01
PLN (×10⁵)	5.42±1.66	9.55±3.79
GIL (×10⁵)	22.49±2.55	13.91±2.39*

n = 6~8 mice in each group and expressed as means±SEM. *p < 0.05, compared to Lt-eGFP (lentivirus-enhanced green fluorescent protein vector) group. PLN, pancreatic lymph node; GIL, graft-infiltrated lymphocytes; Lt-Gal-9-IRES-eGFP, lentivirus-galectin-9-internal ribosome entry site-enhanced green fluorescent protein vector.

Table 2.

Absolute Cell Numbers of Each T-Cell Population in Graft-Infiltrated Lymphocytes (×10³)

T-Cell Subsets	Lt-eGFP	Lt-Gal-9-IRES-eGFP
CD4⁺	540.66±61.25	374.74±64.25^*
CD8⁺	507.37±57.48	368.06±63.11^*
CD4⁺IFN-g⁺	283.25±17.06	140.94±17.29^*
CD4⁺IL-17⁺	33.15±3.76	26.97±4.62
CD8⁺IFN-g⁺	391.54±44.36	233.39±40.02^*
CD8⁺IL-17⁺	5.12±0.58	3.76±0.64
CD4⁺CD25⁺Foxp3⁺	36.08±4.09	29.51±5.06

Data are expressed as means±SEM.

p < 0.05, compared to Lt-eGFP group. CD4, cluster of differentiation 4; IFN-γ, interferon-γ; IL-17, interleukin-17; Foxp3, forkhead box P3.

Discussion

Islet transplantation therapy is an attractive approach to cure type 1 diabetes, as it provides better metabolic control than insulin injection and prevents nephropathy, retinopathy, and vascular and heart disease. However, inflammation, allorejection, and recurrent autoimmunity contribute to early graft loss and are major obstacles to the success of islet transplantation. In this study, we demonstrate that overexpression of galectin-9 in islet grafts leads to profound prolongation of graft survival after challenge with diabetogenic splenocytes in a NOD/SCID mouse model. Moreover, this galectin-9-mediated immune suppression is restricted to the graft, in contrast with the global inhibition of systemic T-cell immunity or the induction of regulatory T-cells.

In the past decades, Th1 cells and related cytokines have been considered as being involved in graft rejection (25); however, the role of different subsets of Th cells in graft rejection has been reexamined after the identification and thorough characterization of Th17 cells. Several lines of evidence have demonstrated that Th17 cells have the capacity to cause rejection of cardiac allografts (2,35). It is not clear whether Th1 and Th17 cells work synergistically or sequentially to cause graft rejection. The suppressive effects of galectin-9 on T-cells have been reported in several models of allotransplantation (6,32,33) and autoimmune diseases (23,36). The differential effects of galectin-9 on the downregulation of Th1 and/or Th17 populations may be due to the different disease models. In transplantation models, the contribution of Th1 or Th17 in the graft rejection is not well characterized, and the results were highly dependent on the types of grafts (7). Moreover, Th17 cells expressed 100-fold less Tim-3 compared with Th1 cells (3), suggesting that Th17 may be less sensitive to galectin-9-mediated suppression than Th1 cells when limited galectin-9 is available. Furthermore, in our model, we observed that IFN-g-producing T-cells (52% of CD4⁺ cells; 77% of CD8⁺ cells) constituted the major population in the graft, whereas IL-17-producing T-cells (6% of CD4⁺ cells; 1% of CD8⁺ cells) are dramatically less than IFN-g-producing cells. Based on these results, we draw the conclusion that overexpression of galectin-9 in islet preferentially suppresses Th1 response and mediates minimal effect on Th17 cells.

Until now, there is no conclusive evidence to support that galectin-9 induces the differentiation of regulatory T-cells or directly modulates the function of regulatory T-cells (21 –23,34). To investigate whether galectin-9 has modulatory effects on regulatory T-cell in NOD mouse system, we have analyzed regulatory T-cell population in NOD mice treated with recombinant galectin-9 protein (100 mg/mouse/week for 4 IP injections). Our data revealed that galectin-9 treatment did not increase the regulatory T-cell population (data not shown), similar to our previous results observed in NOD mice injected by a plasmid encoding galectin-9 (4). Besides, regulatory T-cells isolated from galectin-9 plasmid-treated and control mice preserved the similar ability to suppress responder T-cell proliferation in a coculture system, indicating that overexpression of galectin-9 in vivo did not affect regulatory T-cell function in our model (Supplementary Fig. 3; https://skydrive.live.com/?cid=355A165A5117E0AE&id=355A165A5117E0AE!105 and back up site: https://docs.google.com/file/d/0ByGzRMkw5qqISFI2VUg2TzJqdzQ/edit?usp=sharing). Based on these data, we concluded that galectin-9-mediated islet graft protection is mainly attributed to the downregulation of IFN-γ-producing T-cells, although the potential effect of galectin-9 on infiltrated T regulatory cells could not be completely excluded. Nevertheless, our results are also consistent with another report using galectin-9 protein to prolong the survival of fully mismatched cardiac allograft in mice (6).

In addition to the suppressive effects of galectin-9 on Tim-3⁺ T-cells, it enhances Tim-3-expressing dendritic-cell-mediated inflammation (1). Therefore, it is reasonable to speculate that systemic administration of galectin-9 may promote tissue inflammation via the activation of dendritic cells, although the net effect of galectin-9 in vivo in those models is immunosuppressive. Moreover, prolonged use of galectin-9 protein in vivo induces anaphylactic-like reactions in animals (36), which limits the therapeutic benefits of this protein. In the present study, galectin-9 was expressed exclusively in islet grafts, which provides graft-specific protection and prevents unexpected side effects.

Many approaches have been proven to protect islets from immune attack in vivo and in vitro, including transgenic mouse-based studies and transduction of islets using viral vectors. In NOD mouse models, transgenic overexpression of coinhibitory molecules (24,31), anti-inflammatory cytokines (16,17), antiapoptotic molecules and antioxidative molucules (9,10), or other immunomodulatory molecules (14,27) in islets protects islets from autoimmune damage. However, the application of these protective genes to transplant therapy is not an overt success (18,31). In general, therapeutic targets that have paracrine action would exert more marked biological effects than membrane-bound or intracellular molecules. In present study, we have analyzed the galectin-9 production in culture supernatant of Lt-Gal-9-IRES-eGFP-transduced islets by an ELISA assay and demonstrated that galectin-9 can be released (Supplementary Fig. 4; https://skydrive.live.com/?cid=355A165A5117E0AE&id=355A165A5117E0AE!105 and back up site: https://docs.google.com/file/d/0ByGzRMkw5qqISFI2VUg2TzJqdzQ/edit?usp=sharing). The level of galectin-9 released from Lt-Gal-9-IRES-eGFP-transduced islets to the culture medium is increased after prolonged culture. However, the level of galectin-9 released from transduced islets is much lower than the therapeutic dosages used in transplantation and autoimmune disease models (6,9,32). Because the amount of gelectin-9 quantified by ELISA assay may just reflect a portion of total galectin-9 production from Lt-Gal-9-IRES-eGFP-transduced islets and the potential membrane-form galectin-9, which may also contribute to the protective effect, is not determined. Besides, galectin-9 in transduced graft microenvironment may accumulate and reach higher concentration to exhibit its biological function. Therefore, the galectin-9 protein that was produced by grafts may build up a shield that attenuates T-cell infiltration. Although only the surrounding cells of the islets were infected by the lentiviral vector, the expression of galectin-9 in the outer layer of cells seems to be a barrier that protects the entire islet and delays T-cell-mediated destruction and infiltration. However, the expression of galectin-9 in islets did not modulate the immune response adequately toward a long-term graft survival. This is probably due to an insufficient amount of galectin-9 in the graft microenvironment. Better results may have been obtained if a higher MOI were used, which may generate more galectin-9 protein.

In conclusion, galectin-9 has a narrow group of cellular targets, which could promote the suppression of Th1-, Th17-, and Tc1-mediated immunity (11,23,32,36) without disrupting the functions of other immune cells, such as regulatory T-cells or Th2 cells. Because of its powerful immunomodulatory functions, it is believed that galectin-9 will be an attractive therapeutic target to develop novel therapeutic strategies against detrimental immune responses. Our results provide a type of protection for transplanted islets, without systemic immune suppression.

Footnotes

Acknowledgments

This work was supported by the National Science Council, Taiwan, ROC (NSC102-2321-B-016-006, NSC100-3112-B-016-001, NSC99-2320-B-016-001-MY3 to H.-K. Sytwu), Tri-service General Hospital Foundation (TSGH-C101-009-0S01, TSGH-C102-007-009-S01), National Health Research Institutes (NHRI-100A1-PDCO-0809111), Ministry of Economic Affairs (100-EC-17-A-20-S1-028), and C.Y. Foundation for Advancement of Education, Sciences, and Medicine. The authors declare no conflict of interest.

References

Anderson

A. C.

Anderson

D. E.

Bregoli

Hastings

W. D.

Kassam

Lei

Chandwaskar

Karman

E. W.

Hirashima

Bruce

J. N.

Kane

L. P.

Kuchroo

V. K.

Hafler

D. A.

Promotion of tissue inflammation by the immune receptor Tim-3 expressed on innate immune cells. Science 318(5853): 1141–1143; 2007.

Chen

Ahmed

Wang

Ochando

Chong

A. S.

Alegre

M. L.

TLR signals promote IL-6/IL-17-dependent transplant rejection. J. Immunol. 182(10): 6217–6225; 2009.

Chen

Langrish

C. L.

McKenzie

Joyce-Shaikh

Stumhofer

J. S.

McClanahan

Blumenschein

Churakovsa

Low

Presta

Hunter

C. A.

Kastelein

R. A.

Cua

D. J.

Anti-IL-23 therapy inhibits multiple inflammatory pathways and ameliorates autoimmune encephalomyelitis. J. Clin. Invest. 116(5): 1317–1326; 2006.

Chou

F. C.

Shieh

S. J.

Sytwu

H. K.

Attenuation of Th1 response through galectin-9 and T-cell Ig mucin 3 interaction inhibits autoimmune diabetes in NOD mice. Eur. J. Immunol. 39(9): 2403–2411; 2009.

Chou

F. C.

Sytwu

H. K.

Overexpression of thioredoxin in islets transduced by a lentiviral vector prolongs graft survival in autoimmune diabetic NOD mice. J. Biomed. Sci. 16: 71; 2009.

Fang

Wang

Zhou

Gao

Zhang

W. N.

Niki

Hirashima

Yuan

Chen

Z. K.

Galectin-9 significantly prolongs the survival of fully mismatched cardiac allografts in mice. Transplantation 88(6): 782–790; 2009.

Heidt

Segundo

D. S.

Chadha

Wood

K. J.

The impact of Th17 cells on transplant rejection and the induction of tolerance. Curr. Opin. Organ Transplant. 15(4): 456–461; 2010.

Hirashima

Kashio

Nishi

Yamauchi

Imaizumi

T. A.

Kageshita

Saita

Nakamura

Galectin-9 in physiological and pathological conditions. Glycoconj. J. 19(7-9): 593–600; 2004.

Hotta

Tashiro

Ikegami

Niwa

Ogihara

Yodoi

Miyazaki

Pancreatic β-cell-specific expression of thioredoxin, an antioxidative and antiapoptotic protein, prevents autoimmune and streptozotocin-induced diabetes. J. Exp. Med. 188(8): 1445–1451; 1998.

10.

Huang

S. H.

Chu

C. H.

J. C.

Chuang

W. C.

Lin

G. J.

Chen

P. L.

Chou

F. C.

Chau

L. Y.

Sytwu

H. K.

Transgenic expression of haem oxygenase-1 in pancreatic β-cells protects non-obese mice used as a model of diabetes from autoimmune destruction and prolongs graft survival following islet transplantation. Diabetologia 53(11): 2389–2400; 2010.

11.

Kashio

Nakamura

Abedin

M. J.

Seki

Nishi

Yoshida

Nakamura

Hirashima

Galectin-9 induces apoptosis through the calcium-calpain-caspase-1 pathway. J. Immunol. 170(7): 3631–3636; 2003.

12.

Kearley

McMillan

S. J.

Lloyd

C. M.

Th2-driven, allergen-induced airway inflammation is reduced after treatment with anti-Tim-3 antibody in vivo. J. Exp. Med. 204(6): 1289–1294; 2007.

13.

Kruisbeek

A. M.

In vitro assays for mouse lymphocyte function. Curr. Prot. Immunol. 75: 3.0.1–3.0.5; 2006.

14.

Lin

G. J.

Huang

S. H.

Chen

Y. W.

Hueng

D. Y.

Chia

W. T.

Chien

M. W.

Yen

B. L.

Sytwu

H. K.

Transgenic expression of murine chemokine decoy receptor D6 by islets reveals the role of inflammatory CC chemokines in the development of autoimmune diabetes in NOD mice. Diabetologia 54(7): 1777–1787; 2011.

15.

Monney

Sabatos

C. A.

Gaglia

J. L.

Ryu

Waldner

Chernova

Manning

Greenfield

E. A.

Coyle

A. J.

Sobel

R. A.

Freeman

G. J.

Kuchroo

V. K.

Th1-specific cell surface protein Tim-3 regulates macrophage activation and severity of an autoimmune disease. Nature 415(6871): 536–541; 2002.

16.

Moritani

Yoshimoto

Wong

S. F.

Tanaka

Yamaoka

Sano

Komagata

Miyazaki

Kikutani

Itakura

Abrogation of autoimmune diabetes in non-obese diabetic mice and protection against effector lymphocytes by transgenic paracrine TGF-β1. J. Clin. Invest. 102(3): 499–506; 1998.

17.

Mueller

Krahl

Sarvetnick

Pancreatic expression of interleukin-4 abrogates insulitis and autoimmune diabetes in nonobese diabetic (NOD) mice. J. Exp. Med. 184(3): 1093–1099; 1996.

18.

Mysore

T. B.

Shinkel

T. A.

Collins

Salvaris

E. J.

Fisicaro

Murray-Segal

L. J.

Johnson

L. E.

Lepore

D. A.

Walters

S. N.

Stokes

Chandra

A. P.

O'Connell

P. J.

d'Apice

A. J.

Cowan

P. J.

Overexpression of glutathione peroxidase with two isoforms of superoxide dismutase protects mouse islets from oxidative injury and improves islet graft function. Diabetes 54(7): 2109–2116; 2005.

19.

Niwa

Satoh

Matsushima

Hosoya

Saeki

Niki

Hirashima

Yokozeki

Stable form of galectin-9, a Tim-3 ligand, inhibits contact hypersensitivity and psoriatic reactions: A potent therapeutic tool for Th1- and/or Th17-mediated skin inflammation. Clin. Immunol. 132(2): 184–194; 2009.

20.

O'Connell

P. J.

Pacheco-Silva

Nickerson

P. W.

Muggia

R. A.

Bastos

Kelley

V. R.

Strom

T. B.

Unmodified pancreatic islet allograft rejection results in the preferential expression of certain T-cell activation transcripts. J. Immunol. 150(3): 1093–1104; 1993.

21.

Oomizu

Arikawa

Niki

Kadowaki

Ueno

Nishi

Yamauchi

Hirashima

Galectin-9 suppresses Th17 cell development in an IL-2-dependent but Tim-3-independent manner. Clin. Immunol. 143(1): 51–58; 2012.

22.

Sanchez-Fueyo

Tian

Picarella

Domenig

Zheng

X. X.

Sabatos

C. A.

Manlongat

Bender

Kamradt

Kuchroo

V. K.

Gutierrez-Ramos

J. C.

Coyle

A. J.

Strom

T. B.

Tim-3 inhibits T helper type 1-mediated auto- and alloimmune responses and promotes immunological tolerance. Nat. Immunol. 4(11): 1093–1101; 2003.

23.

Seki

Oomizu

Sakata

K. M.

Sakata

Arikawa

Watanabe

Ito

Takeshita

Niki

Saita

Nishi

Yamauchi

Katoh

Matsukawa

Kuchroo

Hirashima

Galectin-9 suppresses the generation of Th17, promotes the induction of regulatory T-cells, and regulates experimental autoimmune arthritis. Clin. Immunol. 127(1): 78–88; 2008.

24.

Shieh

S. J.

Chou

F. C.

P. N.

Lin

W. C.

Chang

D. M.

Roffler

S. R.

Sytwu

H. K.

Transgenic expression of single-chain anti-CTLA-4 Fv on β-cells protects nonobese diabetic mice from autoimmune diabetes. J. Immunol. 183(4): 2277–2285; 2009.

25.

Strom

T. B.

Roy-Chaudhury

Manfro

Zheng

X. X.

Nickerson

P. W.

Wood

Bushell

The Th1/Th2 paradigm and the allograft response. Curr. Opin. Immunol. 8(5): 688–693; 1996.

26.

Suarez-Pinzon

Rajotte

R. V.

Mosmann

T. R.

Rabinovitch

Both CD4⁺ and CD8⁺ T-cells in syngeneic islet grafts in NOD mice produce interferon-γ during β-cell destruction. Diabetes 45(10): 1350–1357; 1996.

27.

Sung

H. H.

Juang

J. H.

Lin

Y. C.

Kuo

C. H.

Hung

J. T.

Chen

Chang

D. M.

Chang

S. Y.

Hsieh

S. L.

Sytwu

H. K.

Transgenic expression of decoy receptor 3 protects islets from spontaneous and chemical-induced autoimmune destruction in nonobese diabetic mice. J. Exp. Med. 199(8): 1143–1151; 2004.

28.

Tanikawa

Okada

Nakano

Hirashima

Yamauchi

Hosokawa

Tanaka

Interaction of galectin-9 with lipid rafts induces osteoblast proliferation through the c-Src/ERK signaling pathway. J. Bone Miner. Res. 23(2): 278–286; 2008.

29.

Wada

Ota

Kumar

Wallner

E. I.

Kanwar

Y. S.

Developmental regulation, expression, and apoptotic potential of galectin-9, a β-galactoside binding lectin. J. Clin. Invest. 99(10): 2452–2461; 1997.

30.

Walsh

P. T.

Strom

T. B.

Turka

L. A.

Routes to transplant tolerance versus rejection; the role of cytokines. Immunity 20(2): 121–131; 2004.

31.

Wang

C. J.

Chou

F. C.

Chu

C. H.

J. C.

Lin

S. H.

Chang

D. M.

Sytwu

H. K.

Protective role of programmed death 1 ligand 1 (PD-L1) in nonobese diabetic mice: The paradox in transgenic models. Diabetes 57(7): 1861–1869; 2008.

32.

Wang

Yuan

Zhou

Zhang

Chen

Z. K.

Activation of Tim-3-Galectin-9 pathway improves survival of fully allogeneic skin grafts. Transpl. Immunol. 19(1): 12–19; 2008.

33.

Wang

Zhou

Yuan

Chen

Z. K.

The Tim-3 ligand galectin-9 negatively regulates CD8⁺ alloreactive T-cell and prolongs survival of skin graft. Cell. Immunol. 250(1-2): 68–74; 2007.

34.

Wang

Wan

Zhang

Zheng

Chen

Z. K.

Tim-3-Galectin-9 pathway involves the suppression induced by CD4⁺CD25⁺ regulatory T-cells. Immunobiology 214(5): 342–349; 2009.

35.

Yuan

Paez-Cortez

Schmitt-Knosalla

D'Addio

Mfarrej

Donnarumma

Habicht

Clarkson

M. R.

Iacomini

Glimcher

L. H.

Sayegh

M. H.

Ansari

M. J.

A novel role of CD4 Th17 cells in mediating cardiac allograft rejection and vasculopathy. J. Exp. Med. 205(13): 3133–3144; 2008.

36.

Zhu

Anderson

A. C.

Schubart

Xiong

Imitola

Khoury

S. J.

Zheng

X. X.

Strom

T. B.

Kuchroo

V. K.

The Tim-3 ligand galectin-9 negatively regulates T helper type 1 immunity. Nat. Immunol. 6(12): 1245–1252; 2005.