Engagement of TLR2 reverses the suppressor function of conjunctiva CD4+CD25+ regulatory T cells and promotes herpes simplex virus epitope-specific CD4+CD25- effector T cell responses

Abstract

PURPOSE. The authors recently reported that Foxp3(+)CD4(+) CD25(+(Bright)) "natural" regulatory T cells (nT(reg) cells) are abundant in rabbit conjunctiva and suppress herpes simplex virus (HSV)-1-specific CD4(+) and CD8(+) effector T cells (T(eff) cells). However, little is known about the overall regulatory mechanisms of these nT(reg) cells. The authors investigate the regulation of conjunctiva-resident nT(reg) cells through Toll-like receptors (TLRs) and their effect on ocular mucosal T(eff) cell immunity. METHODS. CD4(+)CD25(+) nT(reg) cells were purified from naive rabbit conjunctivas, and their TLR expression profile was determined. The effects of TLR engagement on nT(reg) cell-mediated suppression of CD4(+) T(eff) cells were determined in vitro and in vivo. RESULTS. The authors found that conjunctiva-resident nT(reg) cells express high levels of TLR2 and TLR9; exposure to the TLR2 ligand lipoteichoic acid (LTA) led to the increased activation and proliferation of nT(reg) cells, and the addition of autologous APCs further increased nT(reg) cell expansion; in contrast, the TLR9 ligand CpG(2007) inhibited the proliferation of nT(reg) cells, and the addition of autologous APCs had no effect on such inhibition; nT(reg) cells treated with LTA, but not with CpG(2007), expressed IFN-γ and IL-10 mRNA, but not TGF-β; consistent with in vitro data, rabbits immunized by topical ocular drops of HSV-gD peptides + TLR2 ligand (LTA) displayed enhanced CD4(+)CD25(-) T(eff) cell immune responses when compared with HSV-gD peptides + TLR9 ligand (CpG(2007)). CONCLUSIONS. Although conjunctiva-resident CD4(+)CD25(+) nT(reg) cells express high level of TLR2 and TLR9, their suppressive function is more significantly reversed after the administration of TLR2 ligand (LTA; P < 0.005) than of TLR9 ligand (CpG(200); P > 0.005). These findings will likely help optimize the topical ocular administration of immunotherapies.

Figure 1.

Intracellular staining of Toll-like receptors in CD4⁺CD25⁺ regulatory T cells purified from rabbit conjunctiva. (A) Purified CD4⁺CD25⁺ cells (0.66 × 10⁵ per assay) were first surface stained with 1 μL α rabbit CD25-biotin and then intra cellularly stained with 1 μL α human TLR2-PE, TLR3 PE, TLR4-PE, TLR8-PE, TLR9-PE, as shown on each histogram. Dotted line: CD4⁺CD25⁺ cells only; solid line: cells stained with anti-human TLR-PE antibody. (B) Isotype control (dotted line) where CD4⁺CD25⁺ cells (0.66 × 10⁵) were first stained with 1 μL anti-rabbit CD25-biotin followed by staining with 1 μL mouse IgG1 PE. (C, D) Level of TLRs in CD4⁺CD25⁺ and CD4⁺CD25⁻ cells as a function of MFI. *P < 0.05 when compared with the MFI of TLR2 and TLR9 expression in CD4⁺CD25⁺ and CD4⁺CD25⁻ populations.

Figure 2.

(A) Intracellular staining of Toll-like receptors in CD11b⁺ APCs isolated from rabbit conjunctiva. CD11b⁺ cells (0.24 × 10⁶ per assay) were first surface stained with 1 μL anti-rabbit CD11b-FITC antibody and then intracellularly stained with 3 μL anti-human TLR2-PE, TLR3-PE, TLR4-PE, TLR8-PE, TLR9-PE in each set. Dotted line: CD11b⁺ cells only; solid line: cells stained with anti-human TLR-PE antibody. (B) TLR level in CD11b⁺ cells as a function of MFI. *P < 0.05 compared with the MFI of TLR2 and TLR9 expression in the CD11b⁺ populations.

Figure 3.

In vitro proliferation of CD4⁺CD25⁺ T_reg cells purified from rabbit conjunctiva. CFSE-labeled CD4⁺CD25⁺ T_reg cells were stimulated with 1 μg/mL soluble α-CD3 (A) or α-CD3 ⁺ APCs (B) for 6 days at 37°C in a 5% CO₂ incubator, in the presence and absence of TLR ligand 2 (LTA) and TLR ligand 9 (CpG₂₀₀₇) at a concentration of 10 μg/mL and 1 μg/mL, respectively. Cells were harvested and stained with anti-rabbit CD4-PE or anti-rabbit CD25-PE and analyzed by flow cytometry. Values in each bar indicate the average ± SD of two wells. Each well has 5 × 10⁴ CFSE-labeled CD4⁺CD25⁺ T_reg cells and with and without 1 × 10⁵ APCs (mitomycin C treated), as designated in the figure.

Figure 4.

Quantification of INF-γ (A, D), IL-10 (B, E), and TGF-β (C, F) expression in rabbit conjunctiva-purified CD4⁺CD25⁺ cells using real-time PCR. One million nT_reg cells were plated in a six-well tissue culture plate and were pretreated with LTA (10 μg/mL; A–C) or CpG (1 μg/mL; D–F) for 12 hours, followed by stimulation for another 12 hours in the presence of anti-CD3 (1 μg/mL) and IL-2 (20 U/mL) at 37°C. RNA was isolated from harvested cells and was quantified. An equal amount of RNA was reversed transcribed. cDNA was amplified using real time PCR and specific primers for rabbit INF-γ, IL-10, and TGF-β. GAPDH was used as a housekeeping gene. RNA quantification was calculated using the comparative Ct method, also known as the 2-ΔΔCt method, where ΔΔCt = ΔCt sample − ΔCt reference (none). Here, ΔCT sample is the Ct value for any sample normalized to the endogenous housekeeping gene, and ΔCt reference (none) is the Ct value for the calibrator also normalized to the endogenous housekeeping gene.

Figure 5.

In vitro suppression of CD⁺CD25⁻ effector cells by CD4⁺CD25⁺ nT_reg. Conjunctiva-purified CD⁺CD25⁻ effector cells (5 × 10⁴ cells) were labeled with CFSE and stimulated with soluble α human CD3 (1 μg/mL) and mitomycin C (50 μg/mL)–treated autologous APCs (1 × 10⁵ cells) for 5 days at 37°C. For the suppression assay, CFSE-labeled CD4⁺CD25⁻ effector cells (5 × 10⁴ cells) were mixed with nonlabeled CD4⁺CD25⁺ cells (10 × 10⁴ cells) in culture medium containing soluble α human CD3 (1 μg/mL) and autologous APCs (1 × 10⁵ cells) and were incubated in the presence and absence of LTA (10 μg/mL) and CpG₂₀₀₇ (1 μg/mL) for 5 days at 37°C. Cells were harvested and stained with α rabbit CD4-PE and analyzed by flow cytometry. (A, dotted line) Absolute number of CD4⁺CD25⁻ effector cells in the absence of anti-human CD3 and autologous APCs. (B) Percentage of T_eff cells calculated from (A). Values in each bar indicate the average ± SD of two wells. (C) Histogram that represents nonstimulated CFSE-labeled CD4⁺CD25⁻ T_eff cells (dashed lines) overlaid with histograms (bold lines) that represent CFSE-labeled CD4⁺CD25⁻ T_eff cells stimulated with (a) an anti-CD3⁺ mAb alone or with (b) an anti-CD3⁺ mAb in the presence of untreated CD4⁺CD25⁺ T_reg cells, (c) an anti-CD3⁺ mAb in the presence of LTA-treated CD4⁺CD25⁺ T_reg cells, or (d) an anti-CD3⁺ mAb in the presence of CpG-treated CD4⁺CD25⁺ T_reg cells.

Figure 6.

Six rabbits (two per group) were immunized ocularly three times at an interval of 14 days with a mixture of four HSV-gD peptide (gD_144–179, gD_287–317, gD_49–82, and gD_332–358) mixed with either CpG₂₀₀₇ (group 1) or LTA (group 2) as a topical ocular mucosal immunoadjuvants. The control group received saline alone, LTA alone, or CpG₂₀₀₇ alone (mock, group 3). Ten days after the third immunization, rabbits were euthanatized, upper and lower conjunctivas were harvested from each rabbit, and lymphocytes were isolated. Lymphocytes were labeled with CFSE and stimulated with HSV-gD peptide (gD_144–179, gD_287–317, gD_49–82, or gD_332–358)–pulsed autologous APCs for 5 days at 37°C in a CO₂ incubator. Cells were harvested and stained with anti-rabbit CD4-PE, and their proliferation was analyzed by flow cytometry. (A) Absolute number of CFSE-labeled proliferating CD4⁺ cells under the marked conditions. (B) Dot plot representation with the percentage of CFSE-labeled proliferating CD4⁺ cells.

Figure 7.

Foxp3 expression on rabbit conjunctival Treg by TLR2 and TLR9 ligands. Rabbit conjunctiva–purified CD4⁺CD25⁺ Treg cells (0.5 × 10⁶ cells) were either left untreated or stimulated overnight with anti-human CD3 (1 μg/mL) ± treatment with LTA (10 μg/mL), CpG (1 μg/mL), or Pam3CSK4 (50 ng/mL). TLR ligand treatments and anti-CD3 stimulations were performed together in a 90-well, flat-bottomed tissue culture plate at 37°C/5% CO₂ incubator for overnight. (A, B), Cells were harvested and surface stained with anti-rabbit CD4-FITC (10 μL/assay) followed by intracellular staining with anti-human Foxp3-PE (20 μL/assay) and were analyzed by FACS. (C) Cells from each well (identical to those described) were lysed with 100-μL lysis buffer (RIPA with PI cocktail). Equal volumes (20 μL) of denatured cell lysates containing ∼1 × 10⁵ cells were loaded in each lane. Details of the Western blot procedure have been previously described. (A, dot plot) MFI of CD4⁺/Foxp3-positive cells for rabbit conjunctiva-purified Treg cells (top) and human PBMC-purified Treg cells (bottom). Human PBMC-purified Treg cells were used as a positive control in this assay. (B, overlapped histograms) Expression of Foxp3 in Treg by FACS. (C) Foxp3 protein in rabbit conjunctiva-purified CD4⁺CD25⁺ Treg cells detected by Western blot analysis.