Soluble OX40L and JAG1 Induce Selective Proliferation of Functional Regulatory T-Cells Independent of canonical TCR signaling

Abstract

Regulatory T-cells (Tregs) play a pivotal role in maintaining peripheral tolerance. Increasing Treg numbers/functions has been shown to ameliorate autoimmune diseases. However, common Treg expansion approaches use T-Cell Receptor (TCR)-mediated stimulation which also causes proliferation of effector T-cells (Teff). To overcome this limitation, purified patient-specific Tregs are expanded ex vivo and transfused. Although promising, this approach is not suitable for routine clinical use. Therefore, an alternative approach to selectively expand functional Tregs in vivo is highly desired. We report a novel TCR-independent strategy for the selective proliferation of Foxp3+Tregs (without Teff proliferation), by co-culturing CD4+ T-cells with OX40 L+Jagged(JAG)-1+ bone marrow-derived DCs differentiated with GM-CSF or treating them with soluble OX40 L and JAG1 in the presence of exogenous IL-2. Tregs expanded using soluble OX40 L and JAG1 were of suppressive phenotype and delayed the onset of diabetes in NOD mice. Ligation of OX40 L and JAG1 with their cognate-receptors OX40 and Notch3, preferentially expressed on Tregs but not on Teff cells, was required for selective Treg proliferation. Soluble OX40L-JAG1-induced NF-κB activation as well as IL-2-induced STAT5 activation were essential for the proliferation of Tregs with sustained Foxp3 expression. Altogether, these findings demonstrate the utility of soluble OX40 L and JAG1 to induce TCR-independent Treg proliferation.

Figure 1. G-BMDC-induced Treg proliferation in NOD mice is mediated through OX40L-JAG1 co-signaling.

(A) CD4+ T-cells were co-cultured with either splenic DCs or G-BMDCs in 1:1 ratio for 5 days. Extent of Treg proliferation was analyzed by flow cytometry based on cell trace violet dilution. Numbers in upper right and left quadrants indicate percentages of resting and proliferating Treg cells. (B) Bar graph showing percentages of resting (Black) and proliferating (Grey) Tregs. Values are expressed as Mean ± SEM (n = 3; ***p < 0.001 Vs splenic DCs). (C) G-BMDCs were pretreated with indicated blocking antibodies (High = 10 μgml, Low = 5 μg/ml) for 2 hrs, co-cultured with CD4+ T-cells and extent of Treg cell proliferation was measured by flow cytometry as mentioned for Fig. 1A. (D) Bar graph showing effect of blocking antibodies on G-BMDC induced Treg proliferation. Values are expressed as Mean ± SEM (n = 3; **p < 0.01 Vs None).(E) CD4+ T-cells were treated with combinations of soluble OX40 L (5 μg/ml), JAG1 (5 μg/ml) and IL-2 (10IU/ml) for 3 days as indicated. Treg proliferation was assayed by flow cytometry based on cell trace violet dilution. (F) Bar graph showing percentages of resting and proliferating Tregs from three independent experiments. Values are expressed as Mean ± SEM (n = 3; *p < 0.05, **p < 0.01 Vs IL-2).

Figure 2. Soluble OX40L-JAG1 can cause selective Treg proliferation independent of TCR stimulation.

(A) CD4+ T-cells were treated with IL-2 (Control), OX40L-JAG1-IL-2 and anti-CD3/CD28-IL-2 for 3 days. Extent of CD4+Foxp3- (Teff) and CD4+Foxp3+ (Treg) cell proliferation was analyzed by flow cytometry. (B) From the above experiments, percentages of CD25, CD44 and CD69 expressing Teff (Grey) and Treg (Black) cells were gated and indicated as numerical. (C,D) Bar graph showing percentages of Teff cells and Treg cells expressing CD25, CD44 and CD69. Values are expressed as Mean ± SEM (n = 3; *p < 0.05, **p < 0.01, ***p < 0.001 Vs control).

Figure 3. Soluble OX40L-JAG1 can cause Treg proliferation in vivo.

(A) Ten week old NOD mice were treated with either PBS or 100 μg of OX40 L and JAG1 once a week for three weeks. Two weeks later, mice were sacrificed and Treg cell numbers in spleen, pancreatic LNs and peripheral LNs were analyzed. Upper and lower panels show percentages of Tregs in PBS and OX40L-JAG1 treated mice respectively. Numerical in upper right quadrant indicates percentages of Foxp3 Tregs. (B) Bar graph showing percentages of Treg in spleen, pancreatic and peripheral lymph nodes of PBS (Grey) and OX40L-JAG1 (Black) treated mice. Values are expressed as Mean ± SEM (n = 3; **p < 0.01, ***p < 0.001 Vs PBS). (C) Splenocytes from PBS and OX40L-JAG1 treated mice were stained for CD4, Foxp3 and Ki67 (proliferation marker). Ki67+ cells among CD4+Foxp3- (Grey-Teff) and CD4+Foxp3+ (Black-Treg) cells were gated and shown as histograms. (D) Bar graph summarizing results shown in Fig. 3C. Values are expressed as Mean ± SEM (n = 3; **p < 0.01 Vs PBS). (E) MHC Class-II^−/− mice were treated with either PBS or 100 μg of OX40 L and JAG1 as mentioned above and Treg numbers in spleens were analyzed. Numbers in the quadrants indicate percentages of CD4-Foxp3+, CD4+ Foxp3+ and CD4+Foxp3-T-cells. (F) Bar graph showing percentages of above mentioned cell types in spleens of PBS (Grey) and OX40L-JAG1 (Black) treated mice. Values are expressed as Mean ± SEM (n = 3; *p < 0.05, **p < 0.01 Vs PBS).

Figure 4. Phenotypic characterization of OX40L-JAG1-IL-2 expanded Treg cells and in vitro suppression assay.

(A) Control (grey) and OX40L-JAG1 (black) expanded Treg cells were analyzed for the expression of Treg suppressive markers such as CTLA4, CD39, Helios and TIGIT and CD25 by FACS analysis. Numerical indicate respective MFI values of CTLA4, CD39, Helios and TIGIT expression in control Vs OX40L-JAG1 expanded Treg cells (n = 3). (B) Bar graph summarizing results showin Fig. 7A (Values represent Mean ± SEM, n = 3, **p < 0.01, ***p < 0.001 Vs Control) (C) Control and OX40L-JAG1 expanded CD4+CD25+ Treg cells from NOD mice were co-cultured with cell trace violet labeled fresh CD4+CD25- Teff cells at indicated ratios and stimulated with anti-CD3/CD28 for 3 days. Extent of Teff proliferation was measured by flow cytometry. (D) Percentage of suppression was calculated as ratio between proliferating Teff cells from Treg:Teff co-cultures to no Treg control. Graph summarizing % of suppression calculated from 4 C (Values represent Mean ± SEM, n = 4).

Figure 5. Treatment of NOD mice with OX40L-JAG1 delays onset of diabetes.

(A) NOD mice were administered with OX40 L and JAG1 once a week for 10–12 weeks. After OX40L-JAG1 treatment blood glucose level was monitored weekly. Kaplan-Meier survival graph shows significantly delayed onset of diabetes in NOD mice upon OX40L-JAG1 treatment (*p < 0.05 Vs PBS-treated). (B) Spleens of 28 week old PBS and OX40-JAG1 treated NOD were analyzed for the percentage of CD4+ Foxp3+ Treg cells by flow cytometry. (B) Bar graph summarizing % of Tregs in spleens of PBS and OX40L-Jag1 treated mice after 28 weeks. Values are expressed as Mean ± SEM (n = 10, *p < 0.05). (C) H & E staining analysis of pancreatic sections from PBS and OX40L-JAG1 treated NOD mice (n = 10).(D) Insulitis scoring was done as described in materials & methods with the following scoring scheme: 0-no insulitis, 1-peri-islet insulitis, 2-intermediate insulitis, 3-intraislet insulitis, 4-complete islet insulitis. (E) Pancreatic sections were stained for insulin by immunohistochemistry (n = 10). (F) Splenocytes from PBS and OX40L-JAG1 treated mice were stimulated with PMA/Ionomycin and mRNA expression of indicated cytokines was analyzed by RT-qPCR. Expression values are expressed as fold induction over stimulated control cells after normalization with GAPDH. (Values represent Mean ± SEM, n = 7, *p < 0.05, **p < 0.01).

Figure 6. Differential gene expression analysis of resting and proliferating Treg cells.

(A) Resting and proliferating Treg cells were sorted out based on Cell Trace violet dilution and subjected to microarray analysis to probe the differential gene expression pattern between these cells. Respective heat maps show the significantly altered expression of surface molecules, Treg functional markers, functional partners of Foxp3 and signaling molecules. (B) Validation of micro array results in a candidate gene approach by RT-qPCR. Fold induction values over resting Treg cells were expressed as Mean ± SEM (n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Vs Resting). (C) Resting and proliferating Treg cells stained for OX40, CTLA4 (CD152), CD25, CD39 and Foxp3. Respective MFI values of resting (Grey) and proliferating (black) Treg cells are indicated.

Figure 7. Characterization of OX40L-JAG1 induced Treg proliferation in OX40^−/− and Notch3−/− mice.

(A) Extent of Treg proliferating induced by OX40L-JAG1-IL-2 was compared between C57BL6 wild type (black) Vs OX40−/− mice (grey), and B6129SF1 wild type (black) Vs Notch3−/− (grey) mice. Numerical represent percentages of proliferating Treg cells. (B) Bar graph summarizing results shown in (A). Values are expressed as Mean ± SEM (n = 3; *p < 0.05, ***p < 0.001 Vs respective wild type controls). (C) C57BL6 wild type, OX40^−/−, B6129SF1 wild type and Notch3^−/− mice were treated with soluble OX40 L and JAG1 as mentioned in Fig. 3A. Spleens were analyzed for Treg cell numbers. Upper and lower panels show percentages of Tregs in PBS and OX40L-JAG1 treated mice. Numbers in upper right quadrant indicate percentages of Foxp3 Tregs (n = 3). (D,E) Bar graphs (D,E) show percentages of Tregs in C57BL6-WT Vs OX40^−/− and B6129SF1-WT Vs Notch3^−/− mice treated with either PBS control or OX40L-JAG1 (*p < 0.05, ***p < 0.001 Vs WT-control; ^#p < 0.05, ^###p < 0.001 Vs OX40^−/− or Notch3^−/− OX40L-JAG1).

Figure 8. Role of NF-κB and STAT5 signaling pathways in OX40L-JAG1-IL-2 induced Treg proliferation and Foxp3 expression.

(A) CD4+ T-cells from NOD mice were pre-treated with pharmacological inhibitors of indicated cell signaling pathways and treated with soluble OX40L-JAG1-IL-2. Effect of these pathway inhibitors on Treg cell proliferation was measured by flow cytometry analysis. (B) RT-qPCR analysis showing effect of inhibitors of NF-κB and STAT5 signaling pathways on Foxp3 mRNA expression (Values represent Mean ± SEM, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Vs control, ^#p < 0.05, ^##p < 0.01 Vs None-OX40L-JAG1-IL-2). (C) Western blot analysis showing effect of inhibitors of NF-κB and STAT5 signaling pathways on Foxp3 protein expression. Western blot analysis of the time dependent effect of soluble OX40L-JAG1-IL-2 on (D) Foxp3 expression, (E) NF-κB p65 phosphorylation and (F) STAT5 phosphorylation in CD4+ T-cells. Western blot analysis of the effects of permutation combinations of soluble OX40 L, JAG1 and IL-2 on (G) Foxp3 expression, (H) NF-κB p65 phosphorylation, and (I)STAT5 phosphorylation in CD4+ T-cells (Values represent Mean ± SEM, n = 3, *p < 0.05, **p < 0.01, ***p < 0.001 Vs Control).