Regulatory T cells sequentially migrate from inflamed tissues to draining lymph nodes to suppress the alloimmune response

Abstract

To determine the site and mechanism of suppression by regulatory T (Treg) cells, we investigated their migration and function in an islet allograft model. Treg cells first migrated from blood to the inflamed allograft where they were essential for the suppression of alloimmunity. This process was dependent on the chemokine receptors CCR2, CCR4, and CCR5 and P- and E-selectin ligands. In the allograft, Treg cells were activated and subsequently migrated to the draining lymph nodes (dLNs) in a CCR2, CCR5, and CCR7 fashion; this movement was essential for optimal suppression. Treg cells inhibited dendritic cell migration in a TGF-beta and IL-10 dependent fashion and suppressed antigen-specific T effector cell migration, accumulation, and proliferation in dLNs and allografts. These results showed that sequential migration from blood to the target tissue and to dLNs is required for Treg cells to differentiate and execute fully their suppressive function.

Figure 1. Treg migration to islet allografts and lymphoid tissues

(A) 1×10⁶ nTreg labeled with PKH26 and intravenously transferred to Foxp3^GFP islet allograft recipients at the time of transplantation. Grafts, dLN, peripheral axillary LN, and spleens harvested and sectioned 4 days later. Adoptively transferred Treg (red) and endogenous Treg (green) shown by fluorescent microscopy. Pictures representative of 10 sections/sample from 3 mice. Magnification x100. (B) 1×10⁶ nTreg, TGFβTreg or TGFβTEaTreg from Foxp3^GFP mice or CSFE labeled naïve CD4⁺CD25^- T cells intravenously transferred to recipients on the day of transplantation. Four days after transplantation, grafts, dLN, ndLN and spleen were isolated, and T cell migration with pooled samples from 2 mice/group analyzed with flow cytometry. Results are gated on CD3⁺ cells and representative of 3 independent experiments. (C) Comparison of CD62E ligand, CD62P ligand and CD62L expression on T cell subsets (gated on CD4⁺Foxp3GFP⁺ cells). Results representative of 3 independent experiments. Gating on CD4⁺GFP⁺or CD4⁺GFP^- population. (D) qRT-PCR analysis of chemokine receptor and S1P₁ expression in T cell subsets from Foxp3-GFP mice. Freshly isolated nTreg or naïve CD4⁺ T cells, or cells stimulated in culture and Treg and non-Treg CD4⁺ subsets separated by GFP expression. Results representative of 3 independent experiments. (*p<0.05 compared to counterpart Treg group).

Figure 2. Treg migration to islet allografts is important for their effector function

1×10⁶ wild type, CD62L^-/-, FucT IV-VII^-/-, CCR2^-/-, CCR4^-/-, CCR5^-/- or CCR7^-/- nTreg intravenously transferred at the time of transplantation. (A) Treg labeled with PKH26, and grafts and dLN harvested at day 4 and sectioned for fluorescence microscopy. Pictures representative of 12 sections/sample from 3 mice/group. Magnification x100. (B) PKH26 stained Treg were counted in the whole graft area or dLN area in 12 consecutive sections/sample from 3 mice/group (* p<0.05 compared to wild type control). (C) In vitro suppressive activity of nTreg from the various mouse strains (p>0.05 CCR2^-/-, CCR4^-/-, CCR5^-/-, CCR7^-/-, CD62L^-/-, and FucT IV-VII^-/- vs. wild type Treg). (D) Graft survival after intravenous transfer of 1×10⁶ nTreg from the indicted mouse strains (p>0.05 for FucT IV-VII^-/-, CCR2^-/-, CCR4^-/-, and CCR5^-/- groups vs. no Treg group, p<0.05 for wild type, CD62L^-/- and CCR7^-/- groups vs. no Treg group). (E) 1×10⁶ wild type, CCR4^-/-, CD62L^-/- or FucT IV-VII^-/- nTreg transferred intravenously, or mixed with islets and then transplanted under the kidney capsule, and graft survival determined (p<0.05 for wild type Treg local vs. iv, FucT IV-VII^-/- Treg local vs. iv, and CCR4^-/- local vs. iv groups). D and E compared the same no, WT, and Fuct IV-VII^-/- Treg iv groups. Results are representative of 3 separate preparations of islets and Treg.

Figure 3. TGFβTEaTreg express CD62E ligand and are more potent suppressors for islet allograft survival

(A) CD4⁺ T cells from TEaFoxp3^GFP double transgenic mice cultured with or without TGFβ for 5 days, then stained for CD62E ligand, CD62P ligand and CD62L. Results are representative of 4 independent experiments. Gating on CD4⁺ cells. (B) 1×10⁶ TGFβTreg from Foxp3^GFP or TEaFoxp3^GFP transgenic mice transferred intravenously into recipients, or mixed with islets and then transplanted under the kidney capsule, and graft survival determined (p<0.05 for TGFβTEaTreg iv vs. TGFβTreg iv group, and TGFβTEaTreg local vs. TGFβTreg local group). Results are representative of 3 separate preparations of islets and Treg.

Figure 4. Treg migration from islet allografts to dLN is required for optimal Treg function

1×10⁶ wild type, CCR2^-/-, CCR4^-/-, CCR5^-/- or CCR7^-/- nTreg transferred with grafts under the kidney capsule. (A) Treg labeled with PKH26, and grafts and dLN harvested at day 4 and sectioned for fluorescence microscopy. Pictures representative of 12 sections/sample from 3 mice/group. Magnification x100. (B) PKH26 stained Treg counted in whole graft area or dLN area in 12 consecutive sections/sample from 3 mice/group (* p<0.01, # p<0.05 vs wild type). (C) nTreg from wild type, CCR2^-/- and CCR7^-/- Treg were transferred either intravenously or locally to grafts; and migration to grafts and dLN determined 1, 4, 7 and 10 days after transfer. Treg were counted in the whole graft area or dLN area in 12 consecutive sections/sample from 3 mice/group (* p<0.05 compared to wild type control). (D) 1×10⁶ nTreg transferred intravenously or locally with grafts, and then sorted 4 days after transplantation from 3 pooled grafts or dLNs per experiment. Chemokine receptor and S1P₁ expression analyzed by qRT-PCR. Representative of 3 experiments. (E) Graft survival determined after local or combined local plus intravenous transfer of 1×10⁶ nTreg from the indicated mouse strains (p>0.05 for CCR2^-/-, CCR5^-/-, or combined groups vs. CCR7^-/- group, and p<0.05 vs. wild type group). Results representative of 3 separate preparations of islets and Treg. (F and G) Wild type, CCR2^-/-, or CCR7^-/- nTreg were labeled with PKH26, transferred as indicated into islet allograft recipients, isolated 4 days after transplantation, and analyzed by (F) fluorescent flow cytometry and (G) qRT-PCR (pooled samples from 4 mice/group and gated on CD4⁺PKH26⁺ population, * p<0.05 vs nTreg before transfer). Results representative of 2 independent experiments in (E and F).

Figure 5. TGFβTEaβTreg preserve islet structure and inhibit effector T cell infiltration of islet allografts

(A and B) 1×10⁶ TGFβTEaTreg transferred intravenously or locally to grafts. (A) Grafts harvested 7 days after transplantation, processed for H&E staining. Pictures representative of 10 sections/graft from 3 mice/group. Magnification x100. (B) Allografts and dLN harvested 7 days after transplantation dispersed into single cell suspensions and stained with anti-CD4 mAb for flow cytometry. (C and D) 1×10⁶ CSFE labeled naive TEa CD4⁺ T cells intravenously injected, and 1×10⁶ PKH26 labeled nTreg transferred intravenously or locally into recipients at the time of transplantation. (C) Grafts, dLN, ndLN, and spleens harvested 4 days after transplantation and analyzed with flow cytometry. Cell percentages and numbers shown. (D) Proliferation of CSFE labeled TEa CD4⁺ T cells analyzed by CSFE dilution; percent undivided cells shown. Samples from 2 mice/group per experiment pooled for analysis. Results representative of 3 independent experiments in (B-D).

Figure 6. Treg inhibit the migration of islet DC by secreting TGFβ and IL-10

(A) PKH26 labeled TGFβTreg were intravenously transferred along with CX3CR1^GFP BALB/c islets to C57BL/6 recipient mice. Transferred TGFβaTreg and DC in were monitored 1 and 4 days later by microscopy and cell counting. Treg, red; DC, green. Pictures representative of 12 sections/graft from 2 mice. Magnification x100, left panel; x450, middle and right panels. GFP+ DC in islet grafts or dLN were counted in the whole graft or dLN areas in 12 consecutive sections/sample from 3 mice/group (* p<0.05 compared to CD4+ control group). (B) Islets from CX3CR1^GFP mice were cultured with or without CCL21 (0.5 μg/ml) for 24 hours and GFP⁺ DC were visualized by fluorescent (upper panel) and light microscopy (lower panel). Pictures representative of observations in 9 independent experiments. Magnification x50. (C) CX3CR1^GFP islets incubated with the indicated chemokine and T cells (* p<0.05 vs. CCL21 group). (D) CX3CR1^GFP islets incubated with the indicated chemokine and cytokines (* p<0.05 vs. CCL21 group). (E) CX3CR1^GFP islets incubated with the indicated chemokine, Treg, and mAbs (* p<0.05 vs. CCL21 group). Calculations based on 100 islets/group per experiment, and results representative of 3 independent experiments in (C-E). (F) Graft survival after local transfer of 1 × 10⁶ wild type or IL-10^-/- nTreg with islets. Results are representative of 3 separate preparations of islets and Treg. (G) C57BL/6 mice received BALB/c CX3CR1^GFP islets, local transfer of wild type or IL-10^-/- nTreg, and intravenous transfer of PKH26 labeled naïve TEa CD4⁺ T cells. Islet DC and TEa CD4⁺ T cell migration were analyzed by flow cytometry after 4 days. Samples from 2 mice/group per experiment pooled for flow cytometry analysis. Results representative of 3 independent experiments. (H) SWAP-70^-/- 129/Sv EMS or 129/Sv EMS islets transplanted to BALB/c recipients. Results are representative of 2 separate preparations of islets and Treg.