SEARCH

SEARCH BY CITATION

Keywords:

  • Epigenetic modification;
  • Lineage differentiation;
  • Regulatory T cell

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

Compelling evidence suggests that Foxp3-expressing CD25+CD4+ regulatory T cells (Treg) are generated within the thymus as a separate lineage. However, Foxp3+CD4+ Treg can also be generated de novo in a TGF-β-dependent process from naive T cells by TCR triggering. Recently, we have shown that naturally occurring, but not in vitro TGF-β-induced Foxp3+ Treg display stable Foxp3 expression that was associated with selective demethylation of an evolutionarily conserved element within the Foxp3 locus named TSDR (Treg-specific demethylated region). Here, we report that inhibition of DNA methylation by azacytidine, even in absence of exogenous TGF-β, not only promoted de novo induction of Foxp3 expression during priming, but also conferred stability of Foxp3 expression upon restimulation. Most notably, such stable Foxp3 expression was found only for cells displaying enhanced TSDR demethylation. In contrast, in vitro TSDR methylation diminished its transcriptional activity. Foxp3+ Treg generated in vivo by DEC-205-mediated targeting of agonist ligands to dendritic cells showed long-term survival in the absence of the inducing antigen and exhibited efficient TSDR demethylation. Together, our data suggest that TSDR is an important methylation-sensitive element regulating Foxp3 expression and demonstrate that epigenetic imprinting in this region is critical for establishment of a stable Treg lineage.

Supporting Information for this article is available at www.wiley-vch.de/contents/jc_2040/2008/38105_s.pdf

Abbreviations:
Aza:

azacytidine-derivatives

iTreg:

induced Treg

nTreg:

naturally occurring Treg

RA:

retinoic acid

TSDR:

Treg-specific demethylated region

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

Regulatory T cells (Treg) play an essential role in balancing the immune system and preventing severe autoimmune-reactions mediated by self-reactive effector T cells that escape negative selection in the thymus. CD25+CD4+ naturally occurring Treg (nTreg) are the best understood among all T cell populations that have been described in recent years to possess suppressive capacities 1. The nTreg express the forkhead-box transcription factor Foxp3, which is essential for Treg lineage identity 2, 3 and their suppressive function 4. Absence of functional Foxp3 protein due to mutations in the Foxp3 gene results in the development of severe autoimmune disorders as can be observed in the ‘scurfy’ mouse mutant 5 and patients suffering from immune dysregulation, polyendocrinopathy, enteropathy, X–linked syndrome (IPEX) 6. Recently, others and we have shown that these autoimmune symptoms can indeed be attributed to the absence of Foxp3+ Treg 7, 8. Vice versa, ectopic expression of Foxp3 confers suppressive capacity to initially non-regulatory CD4+ T cells 9, 10. In mice, expression of Foxp3 is largely confined to the nTreg compartment 2, 7, 11, while in humans, the significance of transient FOXP3 up-regulation in activated non-regulatory T cells is still debated 1217.

The thymus plays an important role in the generation of nTreg with Foxp3 expression becoming prominent mainly in CD4 single-positive thymocytes 18. However, conversion of Foxp3CD4+ T cells into Foxp3+CD4+ Treg in the periphery has been demonstrated in various in vivo models when antigen was delivered under tolerogenic conditions 1921. Yet, to what extent Treg induced outside the thymus contribute to the peripheral Treg pool and whether they show a stable phenotype has not been thoroughly analyzed so far. Although a recent study failed to demonstrate in vivo conversion of BDC2.5 T cells expressing a transgenic TCR, which targets an unknown pancreatic self-antigen in NOD mice 22, de novo generated Treg are likely to play an important role in scenarios such as tolerance to food-born antigens or to the commensal gut flora 2325. Foxp3 expression can also be induced in CD25CD4+ T cells by TCR triggering in the presence of transforming growth factor β (TGF-β) in vitro2628. These in vitro induced Foxp3+ Treg (iTreg) were suppressive in vitro and in vivo28, 29. However, the acquired Treg phenotype was unstable in vitro, since the vast majority of cells lost Foxp3 expression as well as suppressive capacity upon restimulation in the absence of exogenous TGF-β, which was not observed in ex vivo isolated nTreg 30.

The difference in stability of nTreg and in vitro iTreg suggests that during intrathymic generation, Treg-specific gene expression patterns get heritably fixed in developing Foxp3+ thymocytes, which does not occur during TGF-β-mediated conversion in vitro resulting in a loss of Treg characteristics as soon as the instructive signal is withdrawn. Increasing evidence points to an important role of epigenetic gene regulation in the fixation of cell differentiation events. Several molecular mechanisms contribute to epigenetic imprinting of essential lymphocyte lineage markers (e.g. cytokine genes), including selective demethylation of CpG motifs and various permissive modifications of histones 3133.

Recently, we reported that epigenetic modifications could be detected in the Foxp3 locus. We identified an evolutionary conserved CpG-rich element within the locus that was selectively demethylated in nTreg, but neither in Foxp3CD25CD4+ conventional T cells nor in in vitro generated Foxp3+ iTreg 30. This Treg-specific demethylated region (TSDR) was also associated with modified histones in nTreg but not in Foxp3 T cells, further strengthening the idea that epigenetics contributes to the control of Foxp3 gene expression 30. Yet, experimental evidence that a demethylated state of TSDR and open chromatin conformation in the Foxp3 locus is not merely associated with transcriptional activity, but in addition involved in conferring the stable, “imprinted” phenotype of permanently differentiated Foxp3+ Treg was lacking so far.

Here, we demonstrate that manipulation of the TSDR methylation status in vitro affects activity as well as stability of Foxp3 expression. Moreover, tolerogenic vaccination by DEC-205-mediated targeting of peptide-agonist ligands to DC in vivo converts naive T cells into Treg displaying both stable Foxp3 expression and demethylated TSDR.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

TGF-β-induced Foxp3 expression is unstable in vitro

Recently, we have reported that murine Foxp3+ iTreg induced by short-term in vitro stimulation in the presence of TGF-β rapidly lost Foxp3 expression in subsequent restimulation cultures lacking exogenous TGF-β 30. This unstable phenotype was accompanied by maintenance of a largely methylated state of TSDR. As permanent differentiation of functional CD4+ T cell subsets usually requires repeated stimulation under polarizing conditions 3436, we here asked whether prolonged exposure to TGF-β would stabilize Foxp3 expression and would induce stronger TSDR demethylation. Naive Foxp3CD25CD4+ T cells from Foxp3-GFP reporter mice 2 were stimulated in vitro in the presence of exogenous TGF-β. On day 5 of Foxp3 induction cultures, Foxp3(GFP)+ cells were sorted to high purity and restimulated either in the presence or in absence of TGF-β (Fig. 1). As described earlier 30, the vast majority of Foxp3+ cells restimulated in the absence of exogenous TGF-β lost Foxp3 expression with 7.2% remaining Foxp3+ cells on day 10 (i.e. day 5 of restimulation). In contrast, 95% of cells restimulated in the presence of exogenous TGF-β maintained Foxp3 expression. However, this Foxp3 expression was still unstable when such cells were purified and restimulated in the absence of TGF-β with 9.2% remaining Foxp3+ cells on day 15 of the experiment (Fig. 1). Even continuous stimulation of cells for five additional days in the presence of TGF-β failed to stabilize Foxp3 expression (data not shown).

thumbnail image

Figure 1. Prolonged exposure to TGF-β fails to mediate stable Foxp3 expression and TSDR demethylation. Highly pure naive Foxp3CD25CD4+ T cells from Foxp3-GFP reporter mice were stimulated with plate-bound anti-CD3/anti-CD28 antibodies and IL-2 (10 ng/mL) in the absence or presence of TGF-β (5 ng/mL). On day 5 of culture, Foxp3 expression was assessed based on GFP expression. Foxp3+ T cells were sorted from TGF-β-containing cultures to high purity and restimulated under the same conditions as described above in the absence or presence of exogenous TGF-β. Foxp3 expression in such restimulation cultures was then assessed on day 10. Foxp3+ cells were purified from TGF-β-containing cultures and restimulated for additional 5 days before the experiment was ended on day 15 with the analysis of Foxp3 expression. The TSDR methylation patterns corresponding to Foxp3+ cells on days 5 and 10 are shown on the right. Amplicons are subdivided by horizontal lines, each representing an individual CpG motif. The methylation status of individual CpG motifs is color-coded according to the degree of methylation at that site. The color code ranges from yellow (0% methylation) to blue (100% methylation) according to the color scale (bottom right). Numbers in FACS-plots indicate percentage of CD4+Foxp3+ cells.

Download figure to PowerPoint

Analysis of highly pure Foxp3+ iTreg populations of TGF-β-containing cultures on days 5 and 10 revealed that instability of Foxp3 expression correlated with strong TSDR methylation (Fig. 1 and Supporting Information Table 1). These data indicate that even prolonged exposure to TGF-β fails to mediate the epigenetic modulations required for stable Foxp3 expression.

Azacytidine-derivatives (Aza) stabilize Foxp3 expression in iTreg

To test whether efficient DNA demethylation is indeed critical for the induction of stable Foxp3 expression in vitro, we used DNA-hypomethylating Aza (5-azacytidine or 5-aza-deoxycytidine) to interfere with de novo DNA methylation. During replication these nucleoside analogs are integrated into DNA (and RNA for 5-azacytidine) and subsequently interfere with the function of DNMT1 (DNA methyltransferase 1) leading to rapid passive DNA demethylation 37. To test the impact of DNA methylation on the stability of Foxp3 expression, we first induced Treg in vitro by activation of Foxp3CD25CD4+ T cells in the presence of TGF-β as described above. After sorting of Foxp3+ iTreg to high purity (97%), cells were restimulated either alone or in the presence of TGF-β or Aza. As depicted in Fig. 2, Aza significantly increased the proportion of cells expressing Foxp3 compared to restimulation cultures without exogenous TGF-β and Aza, which rapidly lost Foxp3 expression over time (Fig. 2B). On day 4 of restimulation, the percentage of Foxp3+ cells in Aza-treated cultures was more than 4-fold increased (Fig. 2A, 47% and 9.3% cells maintained Foxp3 expression in Aza-treated and non-treated cultures, respectively). TGF-β-containing control cultures maintained high Foxp3 expression with 91% Foxp3+ cells on day 4 (Fig. 2A). These results strengthen the idea of an essential role for DNA (de)methylation in the regulation of Foxp3 stability.

thumbnail image

Figure 2. Aza stabilizes TGF-β-induced Foxp3 expression. Naive Foxp3CD25CD4+ T cells were purified and stimulated in vitro as described in Fig. 1. On day 5, Foxp3+ cells were sorted from TGF-β-containing cultures and restimulated with anti-CD3/anti-CD28 antibodies and IL-2 alone or in the presence of TGF-β or Aza (5-Azacytidine, 5 µM). On day 4 of restimulation, the percentage of Foxp3-expressing cells in Aza-containing cultures without exogenous TGF-β was markedly increased compared to non-treated cultures (A). Increased percentages of Foxp3+ cells in the presence of Aza compared to non-treated cultures were observed during the duration of restimulation cultures (B). TGF-β containing restimulation cultures maintained Foxp3 expression. Numbers in FACS-plots indicate percentage of Foxp3+ cells.

Download figure to PowerPoint

TSDR methylation determines transcriptional activity

Our previous observation that the TSDR encoding sequence mediates transcriptional activity only when cloned in front of minimal promoter 30 suggested that TSDR might serve as an enhancer in the regulation of Foxp3 expression. In these initial studies a non-methylated construct was used. Therefore, we now aimed to determine whether the TSDR methylation status is crucial for its transcriptional activity. For this, luciferase vectors either carrying a methylated or demethylated TSDR insert were used in transient luciferase assays employing the murine CD4+ T cell line RLM-11–1. As observed before 30, the TSDR transcriptional activity was dependent on TCR stimulation, here mimicked by treatment with 4-beta-phorbol 12-myristate 13-acetate (PMA; Fig. 3). The increase in luciferase activity in response to PMA was drastically reduced when the methylated construct was assessed (75% reduction compared to the demethylated construct), demonstrating that the transcriptional activity is strictly dependent on a demethylated TSDR status.

thumbnail image

Figure 3. Methylation-dependent transcriptional activity of TSDR. To determine whether the methylation status is crucial for TSDR transcriptional activity, methylated or demethylated TSDR was inserted into the pGL3Promoter luciferase vector (pGL3-Pro). RLM-11–1 cells were cotransfected with either empty pGL3-Pro, demethylated or methylated pGL3-Pro-TSDR, and a Renilla internal control vector. Four hours post-transfection cells were stimulated with PMA or left untreated. Luciferase activity was measured 24 h post-stimulation and normalized to Renilla activity (relative light units; RLU).

Download figure to PowerPoint

Aza promotes stable Foxp3 expression in vitro

The finding that Aza stabilized TGF-β-dependent Foxp3 expression in iTreg in vitro prompted us to test whether this demethylating drug also promotes the initial induction of Foxp3 expression. In fact, Aza in combination with TGF-β increased the proportion of Foxp3+ cells compared to cultures containing TGF-β alone (Fig. 4A, 87 to 60%, respectively). Interestingly, Aza alone was sufficient to induce Foxp3 expression in a significant proportion of CD25CD4+ T cells as early as 3 days after stimulation (Fig. 4A, 32% Foxp3+ cells). Titration experiments indicated that increasing concentrations of Aza correlated with increased proportions of Foxp3+ cells (Fig. 4B). These results demonstrate that Foxp3 can be induced by interference with DNA methylation independent of exogenous TGF-β.

thumbnail image

Figure 4. Aza induced Foxp3 expression in vitro. Naive Foxp3CD25CD4+ T cells expressing a transgenic TCR recognizing the influenza hemagglutinin peptide 107–119 (HA107–119) were FACS-purified from RAG2-deficient mice, labeled with CFSE and stimulated in vitro using anti-CD3/anti-CD28 coated beads and IL-2 (100 U/mL), either alone or in the presence of TGF-β (5 ng/mL) and/or Aza (5-aza-deoxycytidine, 1 μM). Foxp3 expression was assessed by intracellular staining on day 3 of culture (A). Increasing amounts of Aza in the absence or presence of TGF-β increased the proportion of Foxp3+ cells (B). Numbers in FACS-plots indicate the percentage of Foxp3+ or Foxp3 cells.

Download figure to PowerPoint

Since Aza treatment leads to random demethylation of the genome, we wondered whether in those cells that acquired Foxp3 expression in Aza-containing cultures, a demethylated TSDR could be observed. To clarify this, we induced Foxp3 expression in vitro in the presence of TGF-β and/or Aza. On day 5 of induction cultures, we sorted Foxp3-expressing or non-expressing cells to assess the TSDR methylation status. In contrast to Foxp3+ cells from Aza-free cultures, Foxp3+ cells from Aza-containing cultures displayed efficient TSDR demethylation (Fig. 5A and Supporting Information Table 1). Most notably, however, was the finding that TSDR in Foxp3 populations from the same Aza-containing cultures was completely methylated (Fig. 5A and Supporting Information Table 1), suggesting that Aza-mediated demethylation events are stochastic and only those cells acquire Foxp3 expression that have encountered demethylation in the Foxp3 locus. As a further control, a CpG-rich control region outside the Foxp3 locus was analyzed. No differences in the degree of methylation among Foxp3+ and Foxp3 cells from the Aza-treated samples were observed, ruling out an overall increased demethylated state of Foxp3+ cells (data not shown). In addition, we confirmed that Foxp3 mRNA expression was observed only in the fraction of Aza-treated cells that also expressed the Foxp3 protein (data not shown), confirming transcriptional regulation of Foxp3 expression rather than an indirect effect of Aza post-transcriptionally stabilizing the Foxp3 protein.

thumbnail image

Figure 5. Efficient TSDR-demethylation and increased Foxp3 stability of Aza-induced Foxp3+ cells. Naive Foxp3CD25CD4+ TCR transgenic T cells were purified from TCR-HA107–119 mice, which were crossed to Foxp3-GFP reporter mice. Foxp3 expression was induced in vitro using Aza and/or TGF-β as described in Fig. 4A. On day 5, cells were sorted according to Foxp3 expression to high purity (A and B, top panels). The various populations were then subjected to TSDR methylation analysis (A, bottom panels). Part of the cells were restimulated with anti-CD3/anti-CD28 coated beads in the presences of IL-2 (100 U/mL), without Aza or TGF-β. On day 5 of such restimulation cultures, Foxp3 expression was assessed based on GFP fluorescence (B, lower panels). The methylation status of individual CpG motifs is color-coded according to the degree of methylation at that site. The color code ranges from yellow (0% methylation) to blue (100% methylation) according to the color scale on the right. Numbers in FACS-plots indicate percentage of Foxp3+ cells.

Download figure to PowerPoint

To test whether this difference in TSDR methylation of Aza- and TGF-β-induced Foxp3+ cells correlates with stability of Foxp3 expression, we sorted populations of Aza- and/or TGF-β-induced Foxp3+ cells to high purity (Fig. 5B) and restimulated them in the absence of both instructive signals. Cells induced by TGF-β alone showed little stability of Foxp3 expression with 17% Foxp3+ cells remaining at day 5 (Fig. 5B), confirming previously published data 30, 38. In contrast, more stable Foxp3 expression could be observed upon restimulation of Foxp3+ cells that had been induced in the presence of Aza (87% Foxp3+ cells in Aza-only-induced cells and 69% Foxp3+ cells in Aza/TGF-β-induced cells).

Viewed as a whole, these data suggest a functional link between demethylation of TSDR and stable Foxp3 expression, although the contribution of additional yet to be identified methylation-sensitive loci cannot be formally excluded.

In vivo induced Treg exhibit stable Foxp3 expression and complete TSDR demethylation

To analyze whether also de novo induction of stable Treg in vivo is linked to TSDR demethylation, we took advantage of a recently described approach to extrathymically generate antigen-specific Treg by targeting peptide-agonist ligands to steady-state DC. As shown previously, a substantial proportion of adoptively transferred initially naive CD25CD4+ T cells expressing a transgenic TCR, which recognizes the influenza hemagglutinin peptide 107–119 (HA107–119), converted into CD25+Foxp3+ Treg after injection of recipient mice with a single dose of 40 ng recombinant anti-DEC-205-HA107–119 fusion antibodies 20, 39. While efficient conversion required subimmunogenic conditions, such induced Foxp3+ Treg can survive for at least 9 weeks in the absence of the inducing antigen (Fig. 6A) and maintain a stable Foxp3+ suppressor phenotype even under immunogenic conditions in vivo20. To more directly assess the stability of induced Foxp3 expression in this system, Foxp3+ and Foxp3 populations were sorted to high purity based on CD25 expression levels 3 weeks after adoptive transfer of initially naive T cells and anti-DEC-205-HA107–119 injection (Fig. 6A). Upon in vitro stimulation with anti-CD3/anti-CD28 coated beads in the presence of IL-2, both populations extensively proliferated with an approximately 20-fold expansion rate at day 8 of culture (data not shown). While the CD25 population remained Foxp3, the vast majority of CD25+ Treg maintained Foxp3 expression (Fig. 6B), confirming the stability of in vivo induced Foxp3+ Treg.

thumbnail image

Figure 6. Foxp3 stability and TSDR demethylation status of in vivo induced Treg. FACS-purified naïve Foxp3CD25CD4+ T cells from Thy1.2+RAG2–/– TCR-HA107–119 mice were adoptively transferred into Thy1-mismatched BALB/c mice. The following day, recipient mice were intraperitoneally injected with 40 ng recombinant anti-DEC-205-HA107–119 fusion antibody. Induced Foxp3 expression of transferred populations was assessed by intracellular staining at the indicated time-points (A). Three weeks after adoptive transfer and DEC-205 fusion antibody injection, Thy1.2+CD4+ T cells were sorted according to CD25 expression (B, upper panel). Sorted populations were subsequently expanded in vitro with anti-CD3/anti-CD28-coated beads in the presence of IL-2 (100 U/mL). On day 5 of expansion culture, Foxp3 expression was assessed by intracellular staining (B, lower panel) and the TSDR-methylation status was analyzed (C). The methylation status of individual CpG motifs is color-coded according to the degree of methylation at that site. The color code ranges from yellow (0% methylation) to blue (100% methylation) according to the color scale. Numbers in FACS-plots indicate percentage of cells in the indicated quadrant.

Download figure to PowerPoint

Importantly, analysis of the methylation status in these in vivo induced and in vitro expanded cells demonstrated that TSDR was efficiently demethylated in Foxp3+ cells, in contrast to Foxp3 cells derived from the same mice (Fig. 6C and Supporting Information Table 1). In fact, the extent of demethylation was comparable to thymus-derived Treg with the same antigen specificity (data not shown).

These data demonstrate that conditions leading to conversion of naive T cells into Treg with stable Foxp3 expression also induce epigenetic modification of regulatory regions in the Foxp3 locus.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

While the essential role of Foxp3 in Treg lineage specification has been firmly established, the molecular mechanisms involved in the regulation of Foxp3 expression remain poorly defined. A recent characterization of the human FOXP3 promoter indicated that its transcriptional activity is dependent on TCR signaling, which is consistent with the presence of several AP1- and NFAT-binding sites within the promoter 40. In addition, binding of signal transducer and activator of transcription (STAT) 5 and STAT3 to conserved binding sites within the Foxp3 locus have been implicated in the regulation of FOXP3 expression in human Treg 41. We have recently identified the conserved CpG-rich TSDR in the 5′ untranslated region of the Foxp3 gene that showed transcriptional activity 30, 42. The findings provided first evidence that epigenetic control mechanisms might be involved in the regulation of Foxp3 expression, as the TSDR was selectively demethylated in nTreg and associated with modified histones 30. By use of a TSDR reporter assay, we here provide experimental evidence that the transcriptional activity of TSDR indeed is controlled by its methylation status. This is in agreement with data by Kim and Leonard 43, who analyzed the transcriptional activity of a CpG-motif containing element from the Foxp3 locus, which is located within TSDR and which is methylation-sensitive. Effects of DNA methylation on Foxp3 regulation have also been reported for natural killer cells, in which inhibition of DNA methylation in the presence of IL-2 induced Foxp3 expression 41.

The major advance of the present study is the finding that DNA methylation in TSDR not only regulates Foxp3 gene transcription, but also is critically involved in maintaining stable Foxp3 expression. Thus, epigenetic imprinting provides the key to establishing a permanent Treg lineage. TGF-β-mediated conversion of CD25CD4+ T cells has been used in various studies to generate Foxp3+ iTreg with suppressive capacity 2628, 30, 38, 44. However, the TGF-β-induced Foxp3+ phenotype was found to be unstable upon restimulation in the absence of exogenous TGF-β 30, 38. In the present study, we show that even multiple rounds of TGF-β treatment for extended periods are not sufficient to stabilize Foxp3 expression. Interestingly, even after prolonged TGF-β signaling, acquisition of Foxp3 expression was not associated with a major demethylation in TSDR, indicating that TGF-β alone is not sufficient to induce epigenetic imprinting of Foxp3 expression. Thus, TGF-β-induced Foxp3 expression in vitro is unstable and seems to be independent of the TSDR methylation status. As we did not observe major effects of TGF-β in the TSDR reporter assay (data not shown), we assume that TGF-β acts independently of TSDR. To confirm the latter hypothesis, a conditional TSDR-deficient mouse line is currently being generated.

The addition of DNA methylation-inhibiting Aza to the cultures was sufficient to induce stable Foxp3 expression, and Aza stabilized TGF-β-induced Foxp3+ Treg in restimulation cultures. These effects could only be observed in a narrow concentration window of Aza, since higher concentrations induced extensive cell death and lower concentrations showed minimal effects (data not shown). A serious challenge of conclusions drawn from global inhibition of DNA methylation is the potential role of indirect effects. Other factors, the expression of which is also controlled by epigenetic mechanisms, could be induced (or silenced) and provide an instructive signal for Foxp3 transcription. While this could appear conceivable for the induction of Foxp3 expression, a role for induction of stability would be much more difficult to explain. Most strikingly, cells from Aza-treated cultures displayed significant demethylation of TSDR only among Foxp3+ cells, but not in Foxp3 cells, whereas methylation of a CpG-rich control region outside the Foxp3 locus exhibited no differences between Foxp3+ and Foxp3 cells. This indicates that Aza-treatment, at the conditions used, only leads to restricted, stochastic demethylating events during mitosis and that demethylation of the TSDR co-distributes with induced and stable Foxp3 expression, strongly suggesting a direct causal relationship. The stochastic basis of Aza-mediated Foxp3 induction may also explain, why Foxp3+ cells from Aza + TGF-β containing cultures show slightly less TSDR demethylation than Aza-only-induced cells: the former population also includes Foxp3+ cells that had not been demethylated at TSDR, but rather induced by TGF-β. These circumstances also let them appear to be less stable after restimulation. Together, these data point to a critical role of an accessible chromatin structure at TSDR for a prolonged transcription of Foxp3.

Since TGF-β-mediated conversion proved insufficient to induce stable Foxp3 expression, we were interested whether Foxp3 instability was a general phenomenon of extra-thymic Foxp3 induction and therefore extended our analysis to an in vivo conversion model. Foxp3+ T cells can be induced in vivo from initially Foxp3 naive T cells by their cognate antigen fused to the DEC-205 antibody, which allows antigen targeting to steady state DC. Indeed, such de novo induced Foxp3+ cells showed stable Foxp3 expression and efficient TSDR demethylation.

These results indicate that epigenetic modification, which results in imprinting of Foxp3 expression and stable Treg populations, is not restricted to nTreg differentiating within the thymus, but can still be initiated in peripheral Foxp3 T cells. The biological signals leading to the induction of TSDR demethylation remain elusive and are obviously lacking in TGF-β-induction cultures, but were provided during the DEC-205-HA-driven conversion in vivo. Several recent studies have implicated all-trans retinoic acid (RA)-producing CD103+ DC in the induction of Foxp3 expression 2325. In vitro, RA increased the proportion of Foxp3+ cells in TGF-β-containing cultures 4547. Moreover, RA induced histone acetylation at the Foxp3 promoter 47 and influenced stability of Foxp3 expression 45. It remains to be tested whether RA is suitable to stabilize Foxp3 expression during TGF-β-mediated induction by increased TSDR demethylation.

The molecular mechanisms involved in the regulation of Foxp3 expression are still far from being understood and need further clarification in order to exploit induced Treg populations in clinical settings in the future. From this study, we conclude that stable conversion of CD25CD4+ T cells into Foxp3+ Treg can only occur under conditions that also induce epigenetic fixation of the Treg phenotype by DNA demethylation. We further propose that TSDR is the critical element for this imprinting of Foxp3 expression. However, the instructive signals and the exact molecular pathways remain to be defined.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

Mice

C57BL/6 Foxp3-GFP reporter mice, kindly provided by Alexander Rudensky, were bred at the Forschungseinrichtung fuer Experimentelle Medizin (FEM, Berlin), and the Dana-Farber Cancer Institute (Boston). RAG2-deficient Thy-1.2 BALB/c mice expressing a transgenic TCR recognizing the influenza hemagglutinin peptide 107–119 (TCR-HA107–119) in the context of H2-IEd, C57BL/6 Foxp3-GFP reporter mice crossed to TCR-HA107–119 transgenic mice for ⩾ nine generations, and congenic Thy-1.1 BALB/c mice were bred at the Dana-Farber Cancer Institute. All animals were kept under specific pathogen-free conditions. Animal care and all procedures were performed in accordance with institutional, state and federal guidelines.

Flow cytometry and cell sorting

Cytometric analysis was performed using BD FACSCaliburTM or BD FACSAriaTM instruments and CellQuest (BD) or FlowJo (Tree Star) software. All sorted populations were ⩾96–99% pure. The mAb to CD4 (L3T4), CD25 (PC6.1) and Thy1.2 (53–2.1) were purchased from Becton Dickinson. The mAb to the TCR-HA (6.5) was purified and conjugated with FITC following standard protocols. Foxp3 expression was analyzed using the intracellular staining set and the anti-Foxp3-PE mAb FJK-16s from eBioscience according to manufacturers’ instructions. For CFSE labeling, cells were incubated in 10 µM CFSE (Molecular Probes) plus 0.1% BSA for 10 min at 37°C.

In vitro and in vivo Foxp3 induction

CD4+ T cells were enriched from single-cell suspensions of pooled spleen and lymph nodes using anti-CD4 microbeads and the AutoMACS magnetic separation system (Miltenyi Biotec). Cells were subsequently stained for CD25 expression and Foxp3(GFP)CD25CD4+ cells were sorted by FACS. To purify naive Foxp3CD25CD4+ TCR transgenic T cells from TCR-HA107–119 mice, cells were additionally stained with the clonotypic 6.5 antibody. Cell culture was done in RPMI 1640 (Gibco) supplemented with 10% FBS (Sigma). Sorted cells were stimulated with plate-bound anti-CD3 (145.2C11) and anti-CD28 (37.51) antibodies or magnetic bead-coated anti-CD3/anti-CD28 antibodies (Dynal), in the presence of indicated amounts of IL-2 (R&D). TGF-β (5 ng/mL; R&D) and 5-Azacytidine or 5-aza-deoxycytidine (both Sigma) were added to cultures as indicated. DEC-205 mediated de novo generation of antigen-specific Treg was performed as described previously 20, 39.

DNA methylation analysis

DNA methylation analysis was performed by bisulfite sequencing as described previously 30.

Luciferase vectors

Cloning of the pGL3-Pro-TSDR vector was described previously 30. The TSDR-insert was excised by restriction-digestion using KpnI and XmaI and purified by gel extraction using the QIAquick®Gel Extraction Kit 50 (Qiagen). For methylation, 1 µg of purified insert was incubated with 4 U of M. SssI methylase (NEB) or was left untreated as control. Successful methylation was verified by digestion with HpaI and MspI, a methylation-sensitive and a methylation-insensitive enzyme, respectively, which recognize the same restriction sequence. Methylated or unmethylated inserts were then ligated into KpnI- and XmaI-cut pGL3Promoter (gGL3-Pro, Promega) using T4 DNA Ligase (NEB) and used in luciferase assays.

Luciferase assay

RLM-11–1 cells 48, which were kindly provided by Marc Ehlers (DRFZ, Berlin, Germany), were transfected using 1 µg of pGL3-Pro or methylated or unmethylated pGL3-Pro-TSDR. Synthetic Renilla luciferase reporter vector (pRL-TK, 0.2 µg; Promega) was used as an internal control for transfection efficiency. Four hours after transfection via nucleofection (Amaxa), RLM-11–1 cells were stimulated with PMA (10 ng/mL; Sigma) and cultured in IMEM (Gibco) supplemented with 10% FBS (Sigma) for 18–24 h. Cells were harvested and luciferase activity was measured using the dual luciferase assay system (Promega). Experiments were performed in triplicates. Data were normalized to Renilla luciferase activity.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

We thank A. Rudensky, Seattle, for generously providing us with the Foxp3-GFP reporter mouse strain, Katharina Raba and Toralf Kaiser for FACS sorting and Kerstin Schlawe and Alexander Hellwag for expert technical assistance. RLM-11–1 cells were kindly provided by Marc Ehlers (DRFZ, Berlin, Germany). This work was supported by the Wilhelm Sander Foundation, by the BMBF (NGFN II, SIPAGE, FKZ 01GS0413) and by the Collaborative Research Center programs SFB 633 and SFB 650 of the German Research Foundation (DFG).

  • 1

    WILEY-VCH

  • 2

    WILEY-VCH

  • 3

    WILEY-VCH

  • 4

    WILEY-VCH

  • 5

    WILEY-VCH

  • 6

    WILEY-VCH

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

The authors declared no financial or commercial conflict of interest.

  • 1
    Shevach, E. M., From vanilla to 28 flavors: multiple varieties of T regulatory cells. Immunity 2006 25: 195201.
  • 2
    Fontenot, J. D., Rasmussen, J. P., Williams, L. M., Dooley, J. L., Farr, A. G. and Rudensky, A. Y., Regulatory T cell lineage specification by the forkhead transcription factor foxp3. Immunity 2005 22: 329341.
  • 3
    Gavin, M. A., Rasmussen, J. P., Fontenot, J. D., Vasta, V., Manganiello, V. C., Beavo, J. A. and Rudensky, A. Y., Foxp3-dependent programme of regulatory T-cell differentiation. Nature 2007 445: 771775.
  • 4
    Williams, L. M. and Rudensky, A. Y., Maintenance of the Foxp3-dependent developmental program in mature regulatory T cells requires continued expression of Foxp3. Nat. Immunol. 2007 8: 277284.
  • 5
    Brunkow, M. E., Jeffery, E. W., Hjerrild, K. A., Paeper, B., Clark, L. B., Yasayko, S. A., Wilkinson, J. E. et al., Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat. Genet. 2001 27: 6873.
  • 6
    Bennett, C. L., Christie, J., Ramsdell, F., Brunkow, M. E., Ferguson, P. J., Whitesell, L., Kelly, T. E. et al., The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat. Genet. 2001 27: 2021.
  • 7
    Lahl, K., Loddenkemper, C., Drouin, C., Freyer, J., Arnason, J., Eberl, G., Hamann, A. et al., Selective depletion of Foxp3+ regulatory T cells induces a scurfy-like disease. J. Exp. Med. 2007 204: 5763.
  • 8
    Kim, J. M., Rasmussen, J. P. and Rudensky, A. Y., Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat. Immunol. 2007 8: 191197.
  • 9
    Fontenot, J. D., Gavin, M. A. and Rudensky, A. Y., Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. 2003 4: 330336.
  • 10
    Hori, S., Nomura, T. and Sakaguchi, S., Control of regulatory T cell development by the transcription factor Foxp3. Science 2003 299: 10571061.
  • 11
    Wan, Y. Y. and Flavell, R. A., Identifying Foxp3-expressing suppressor T cells with a bicistronic reporter. Proc. Natl. Acad. Sci. USA 2005 102: 51265131.
  • 12
    Walker, M. R., Kasprowicz, D. J., Gersuk, V. H., Benard, A., Van Landeghen, M., Buckner, J. H. and Ziegler, S. F., Induction of FoxP3 and acquisition of T regulatory activity by stimulated human CD4+CD25 T cells. J. Clin. Invest. 2003 112: 14371443.
  • 13
    Morgan, M. E., van Bilsen, J. H., Bakker, A. M., Heemskerk, B., Schilham, M. W., Hartgers, F. C., Elferink, B. G. et al., Expression of FOXP3 mRNA is not confined to CD4+CD25+ T regulatory cells in humans. Hum. Immunol. 2005 66: 1320.
  • 14
    Wang, J., Ioan-Facsinay, A., van der Voort, E. I. H., Huizinga, T. W. J. and Toes, R. E. M., Transient expression of FOXP3 in human activated nonregulatory CD4+ T cells. Eur. J. Immunol. 2007 37: 129138.
  • 15
    Pillai, V., Ortega, S. B., Wang, C. K. and Karandikar, N. J., Transient regulatory T-cells: A state attained by all activated human T-cells. Clin. Immunol. 2007 123: 1829.
  • 16
    Allan, S. E., Crome, S. Q., Crellin, N. K., Passerini, L., Steiner, T. S., Bacchetta, R., Roncarolo, M. G. and Levings, M. K., Activation-induced FOXP3 in human T effector cells does not suppress proliferation or cytokine production. Int. Immunol. 2007 19: 345354.
  • 17
    Tran, D. Q., Ramsey, H. and Shevach, E. M., Induction of FOXP3 expression in naive human CD4+FOXP3 T cells by T cell receptor stimulation is TGF-beta-dependent but does not confer a regulatory phenotype. Blood 2007 110: 29832990.
  • 18
    Fontenot, J. D., Dooley, J. L., Farr, A. G. and Rudensky, A. Y., Developmental regulation of Foxp3 expression during ontogeny. J. Exp. Med. 2005 202: 901906.
  • 19
    Apostolou, I. and von Boehmer, H., In vivo instruction of suppressor commitment in naive T cells. J. Exp. Med. 2004 199: 14011408.
  • 20
    Kretschmer, K., Apostolou, I., Hawiger, D., Khazaie, K., Nussenzweig, M. C. and von Boehmer, H., Inducing and expanding regulatory T cell populations by foreign antigen. Nat. Immunol. 2005 12: 12191227.
  • 21
    Siewert, C., Lauer, U., Cording, S., Bopp, T., Schmitt, E., Hamann, A. and Huehn, J., Experience-driven development: effector/memory-like αE+Foxp3+ regulatory T cells originate from both naïve T cells and naturally occurring naïve-like regulatory T cells. J. Immunol. 2008 180: 146155.
  • 22
    Wong, J., Mathis, D. and Benoist, C., TCR-based lineage tracing: no evidence for conversion of conventional into regulatory T cells in response to a natural self-antigen in pancreatic islets. J. Exp. Med. 2007 204: 20392045.
  • 23
    Sun, C. M., Hall, J. A., Blank, R. B., Bouladoux, N., Oukka, M., Mora, J. R. and Belkaid, Y., Small intestine lamina propria dendritic cells promote de novo generation of Foxp3 T reg cells via retinoic acid. J. Exp. Med. 2007 204: 17751785.
  • 24
    Coombes, J. L., Siddiqui, K. R., Arancibia-Carcamo, C. V., Hall, J., Sun, C. M., Belkaid, Y. and Powrie, F., A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J. Exp. Med. 2007 204: 17571764.
  • 25
    von Boehmer, H., Oral tolerance: is it all retinoic acid? J. Exp. Med. 2007 204: 17371739.
  • 26
    Chen, W., Jin, W., Hardegen, N., Lei, K. J., Li, L., Marinos, N., McGrady, G. and Wahl, S. M., Conversion of peripheral CD4+CD25 naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J. Exp. Med. 2003 198: 18751886.
  • 27
    Fantini, M. C., Becker, C., Monteleone, G., Pallone, F., Galle, P. R. and Neurath, M. F., Cutting edge: TGF-beta induces a regulatory phenotype in CD4+CD25 T cells through Foxp3 induction and down-regulation of Smad7. J. Immunol. 2004 172: 51495153.
  • 28
    Fantini, M. C., Becker, C., Tubbe, I., Nikolaev, A., Lehr, H. A., Galle, P. R. and Neurath, M. F., TGF-beta induced Foxp3+ regulatory T cells suppress Th1-mediated experimental colitis. Gut 2006 55: 671680.
  • 29
    DiPaolo, R. J., Brinster, C., Davidson, T. S., Andersson, J., Glass, D. and Shevach, E. M., Autoantigen-specific TGFbeta-induced Foxp3+ regulatory T cells prevent autoimmunity by inhibiting dendritic cells from activating autoreactive T cells. J. Immunol. 2007 179: 46854693.
  • 30
    Floess, S., Freyer, J., Siewert, C., Baron, U., Olek, S., Polansky, J., Schlawe, K. et al., Epigenetic control of the foxp3 locus in regulatory T cells. PLOS Biol. 2007 5: e38.
  • 31
    Lee, G. R., Kim, S. T., Spilianakis, C. G., Fields, P. E. and Flavell, R. A., T helper cell differentiation: regulation by cis elements and epigenetics. Immunity 2006 24: 369379.
  • 32
    Kioussis, D. and Georgopoulos, K., Epigenetic flexibility underlying lineage choices in the adaptive immune system. Science 2007 317: 620622.
  • 33
    Ansel, K. M., Djuretic, I., Tanasa, B. and Rao, A., Regulation of Th2 differentiation and Il4 locus accessibility. Annu. Rev. Immunol. 2006 24: 607656.
  • 34
    Lee, D. U., Agarwal, S. and Rao, A., Th2 lineage commitment and efficient IL-4 production involves extended demethylation of the IL-4 gene. Immunity 2002 16: 649660.
  • 35
    Lohning, M., Richter, A., Stamm, T., Hu-Li, J., Assenmacher, M., Paul, W. E. and Radbruch, A., Establishment of memory for IL-10 expression in developing T helper 2 cells requires repetitive IL-4 costimulation and does not impair proliferation. Proc. Natl. Acad. Sci. USA 2003 100: 1230712312.
  • 36
    Chang, H. D., Helbig, C., Tykocinski, L., Kreher, S., Koeck, J., Niesner, U. and Radbruch, A., Expression of IL-10 in Th memory lymphocytes is conditional on IL-12 or IL-4, unless the IL-10 gene is imprinted by GATA-3. Eur. J. Immunol. 2007 37: 807817.
  • 37
    Creusot, F., Acs, G. and Christman, J. K., Inhibition of DNA methyltransferase and induction of Friend erythroleukemia cell differentiation by 5-azacytidine and 5-aza-2′-deoxycytidine. J. Biol. Chem. 1982 257: 20412048.
  • 38
    Selvaraj, R. K. and Geiger, T. L., A kinetic and dynamic analysis of Foxp3 induced in T cells by TGF-beta. J. Immunol. 2007 179: 13901401.
  • 39
    Kretschmer, K., Heng, T. S. and von Boehmer, H., De novo production of antigen-specific suppressor cells in vivo. Nat. Protoc. 2006 1: 653661.
  • 40
    Mantel, P. Y., Ouaked, N., Ruckert, B., Karagiannidis, C., Welz, R., Blaser, K. and Schmidt-Weber, C. B., Molecular mechanisms underlying FOXP3 induction in human T cells. J. Immunol. 2006 176: 35933602.
  • 41
    Zorn, E., Nelson, E. A., Mohseni, M., Porcheray, F., Kim, H., Litsa, D., Bellucci, R. et al., IL-2 regulates FOXP3 expression in human CD4+CD25+ regulatory T cells through a STAT dependent mechanism and induces the expansion of these cells in vivo. Blood 2006 108: 15711579.
  • 42
    Baron, U., Floess, S., Wieczorek, G., Baumann, K., Grützkau, A., Dong, J., Thiel, A. et al., DNA demethylation in the human FOXP3 locus discriminates regulatory T cells from activated FOXP3+ conventional T cells. Eur. J. Immunol. 2007 37: 23782389.
  • 43
    Kim, H. P. and Leonard, W. J., CREB/ATF-dependent T cell receptor-induced FoxP3 gene expression: a role for DNA methylation. J. Exp. Med. 2007 204: 15431551.
  • 44
    Park, H. B., Paik, D. J., Jang, E., Hong, S. and Youn, J., Acquisition of anergic and suppressive activities in transforming growth factor-beta-costimulated CD4+CD25 T cells. Int. Immunol. 2004 16: 12031213.
  • 45
    Benson, M. J., Pino-Lagos, K., Rosemblatt, M. and Noelle, R. J., All-trans retinoic acid mediates enhanced T reg cell growth, differentiation, and gut homing in the face of high levels of co-stimulation. J. Exp. Med. 2007 204: 17651774.
  • 46
    Mucida, D., Park, Y., Kim, G., Turovskaya, O., Scott, I., Kronenberg, M. and Cheroutre, H., Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 2007 317: 256260.
  • 47
    Kang, S. G., Lim, H. W., Andrisani, O. M., Broxmeyer, H. E. and Kim, C. H., Vitamin A metabolites induce gut-homing FoxP3+ regulatory T cells. J. Immunol. 2007 179: 37243733.
  • 48
    Grueter, B., Petter, M., Egawa, T., Laule-Kilian, K., Aldrian, C. J., Wuerch, A., Ludwig, Y. et al., Runx3 regulates integrin alpha E/CD103 and CD4 expression during development of CD4/CD8+ T cells. J. Immunol. 2005 175: 16941705.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. Appendix
  9. Supporting Information

Supporting information for this article is available on the WWW under http://www.wiley-vch.de/contents/jc_2040/2008/38105_s.pdf or from the author.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.