Rbpj expression in regulatory T cells is critical for restraining TH2 responses

Delacher, Michael; Schmidl, Christian; Herzig, Yonatan; Breloer, Minka; Hartmann, Wiebke; Brunk, Fabian; Kägebein, Danny; Träger, Ulrike; Hofer, Ann-Cathrin; Bittner, Sebastian; Weichenhan, Dieter; Imbusch, Charles D.; Hotz-Wagenblatt, Agnes; Hielscher, Thomas; Breiling, Achim; Federico, Giuseppina; Gröne, Hermann-Josef; Schmid, Roland M.; Rehli, Michael; Abramson, Jakub; Feuerer, Markus

doi:10.1038/s41467-019-09276-w

Download PDF

Article
Open access
Published: 08 April 2019

Rbpj expression in regulatory T cells is critical for restraining T_H2 responses

Nature Communications volume 10, Article number: 1621 (2019) Cite this article

5731 Accesses
29 Citations
30 Altmetric
Metrics details

Subjects

Abstract

The transcriptional regulator Rbpj is involved in T-helper (T_H) subset polarization, but its function in T_reg cells remains unclear. Here we show that T_reg-specific Rbpj deletion leads to splenomegaly and lymphadenopathy despite increased numbers of T_reg cells with a polyclonal TCR repertoire. A specific defect of Rbpj-deficient T_reg cells in controlling T_H2 polarization and B cell responses is observed, leading to the spontaneous formation of germinal centers and a T_H2-associated immunoglobulin class switch. The observed phenotype is environment-dependent and can be induced by infection with parasitic nematodes. Rbpj-deficient T_reg cells adopt open chromatin landscapes and gene expression profiles reminiscent of tissue-derived T_H2-polarized T_reg cells, with a prevailing signature of the transcription factor Gata-3. Taken together, our study suggests that T_reg cells require Rbpj to specifically restrain T_H2 responses, including their own excessive T_H2-like differentiation potential.

The transcription factor BCL-6 controls early development of innate-like T cells

Article 27 July 2020

Marianthi Gioulbasani, Alexandros Galaras, … Mihalis Verykokakis

NFAT primes the human RORC locus for RORγt expression in CD4+ T cells

Article Open access 16 October 2019

Hanane Yahia-Cherbal, Magda Rybczynska, … Elisabetta Bianchi

The role of transcription factors in shaping regulatory T cell identity

Article 19 June 2023

Jorge L. Trujillo-Ochoa, Majid Kazemian & Behdad Afzali

Introduction

Regulatory T cells (T_reg) are important mediators of peripheral tolerance, and their absence leads to catastrophic autoimmunity in men (IPEX¹) and mice (Scurfy²). T_reg cells are characterized by both expression of the hallmark transcription regulator Foxp3^3,4,5 and a unique epigenetic profile^6,7,8,9. T_reg cells specialize to fulfill their diverse regulatory functions¹⁰. They can engage defined molecular pathways to specifically suppress either T_H1-polarized, T_H2-polarized, or T_H17-polarized immune effector cells¹¹. For instance, under T_H1 conditions, T_reg cells up-regulate expression of the T_H1-specific transcription factor T-box 21 (T-bet) and accumulate at inflammatory sites¹². Correspondingly, under T_H2 conditions, T_reg cells express Gata-binding protein 3 (Gata-3) and interferon regulatory factor 4 (Irf4), and T_reg-specific IRF4-deletion leads to IL-4 cytokine production of effector T cells and lymphoproliferative disease¹³. Up-regulation of signal transducer of activated T cells 3 (Stat3) is critical for the capacity of T_reg cells to control T_H17-mediated inflammation, while its T_reg-specific deletion results in enhanced IL-17 production by effector cells and intestinal inflammation¹⁴. Therefore, T_reg cells integrate unique parts of T_H subtype-specific transcriptional programs to specifically control the respective T_H-polarized immune response.

Recombination signal-binding protein for immunoglobulin kappa J region (Rbpj) is a transcription factor commonly known for its function as a co-factor during Notch signaling, translating extracellular signals into gene expression changes¹⁵. In the context of T cell differentiation and function, Rbpj has been associated with T_H1/T_H2 cell fate decisions^16,17. Indeed, in CD4⁺Foxp3⁻ conventional T (T_conv) cells, Rbpj in a complex with the Notch intracellular domain (NICD) was shown to be critical for regulation of Gata-3, an important molecular switch for optimal T_H2 responses^18,19. In contrast to this, forced expression of the NICD in T_reg cells rendered them incapable of suppressing T effector cells and caused autoimmunity²⁰. This indicates that, based on the cellular context, Rbpj and Notch have a different impact on cellular responses. While the importance of Rbpj is well documented in T_H2 subset polarization, its function in T_reg cells remains unclear.

Here we unveil a previously unappreciated role of Rbpj in regulating the capacity of T_reg cells to restrain T_H2 responses. Loss of Rbpj renders T_reg cells more sensitive to T_H2-inducing conditions and fosters the extensive generation of Gata-3-positive tissue-type T_reg cells.

Results

Deletion of Rbpj causes defined organ pathology

We specifically deleted Rbpj in T_reg cells by crossing Foxp3^Cre,YFP mice with mice harboring floxed Rbpj alleles (called Δ/Δ). We compared these to littermate control Foxp3^Cre,YFP mice with wildtype Rbpj alleles (termed WT). We closely monitored our mice for 20 weeks, and about 40% of mice spontaneously developed splenomegaly and lymphadenopathy within this time interval, while about 60% of animals remained healthy (Fig. 1a, b). We confirmed the T_reg-specific deletion of Rbpj on DNA, RNA, and protein level (Supplementary Fig. 1a–d). First, we analyzed Δ/Δ and WT mice for the presence of CD4⁺CD25⁺Foxp3⁺ T_reg cells in spleen and other tissues (Fig. 1c). We observed a strong increase in the fraction of T_reg cells among CD4⁺ T cells from about 12% in WT spleens to 28% in spleens from affected Δ/Δ animals. In absolute numbers, lymph nodes and spleen from Δ/Δ animals harbored about 10–20 times more T_reg cells than their WT counterparts (Fig. 1c, right panel). This increase was not seen in the thymus, indicating normal thymic T_reg cell output, or in mesenteric lymph nodes. Analysis of CD44 and L-selectin (CD62L) indicated activation of the T_reg compartment in affected Δ/Δ mice (Fig. 1d). Furthermore, affected Δ/Δ animals showed a higher density of Foxp3-positive T_reg cells by immunohistology in spleen (Fig. 1e) and lymph nodes (Supplementary Fig. 1e). Affected animals developed noticeable skin pathology at snout, abdominal and tail regions, which served as useful biomarker to identify sick animals. Affected skin areas showed thickening of the epidermis and mononuclear cell infiltrates in the corium (Fig. 1f). Global defects in T_reg cell function normally lead to a severe autoimmune manifestation, with destructive immune cell infiltration in a diverse set of organs². Detailed histological examination of different organs including small intestine, large intestine, stomach, kidneys, salivary gland, eye, liver, and lung showed no signs of obvious immune cell infiltration and tissue destruction (Supplementary Figure 2a, b), indicating that Rbpj deficiency in T_reg cells did not lead to a global loss of T_reg-mediated immune control. This was supported by data from a standard in vitro suppression assay with TCR-stimulated T_responder cells, were we did not detect significant changes in the in vitro suppressive potential (Supplementary Fig. 3). In summary, these data indicate that RBPJ deficiency affected a more specific segment of T_reg function.

Germinal center formation and B-cell polarization

Given that the T_reg-specific deletion of Rbpj affected secondary lymphoid organs and skin, we performed gene expression analysis of total LN RNA from WT and affected Δ/Δ animals (Fig. 2a). The most strongly up-regulated genes among the 4388 differentially expressed probes were involved in immune globulin (Ig) chain rearrangement and antibody production (Fig. 2a, highlighted in blue). Furthermore, B cell-specific markers such as Cd22 and Cd19 were over-expressed in LN from Δ/Δ mice. In addition, Il4 was increased and pointed towards T_H2 subtype polarization (Fig. 2a and Supplementary Fig. 4a). B cells also down-modulated CD62L, and total numbers of B cells expanded about 10–20-fold in affected animals (Fig. 2b). To determine whether this increase in absolute B cell numbers and their maturation lead to changes in antibody Ig subtype distribution, we analyzed blood serum Ig levels via ELISA (Fig. 2c). Interestingly, we detected a significant up-regulation of IgG1 and IgE, while IgG3 was repressed in the serum (Fig. 2c), indicative of a classical IL-4 (=T_H2)-induced B cell antibody class switch²¹. As another hallmark of ongoing B cell differentiation and antibody production, we detected the formation of numerous germinal centers in LNs from affected Δ/Δ animals, but not in healthy WT animals (Fig. 2d and Supplementary Fig. 4b). Since IgE levels showed an almost 100-fold increase in affected Δ/Δ animals (Fig. 2c), and skin-resident mast cells express a high-affinity IgE receptor, we performed Giemsa staining of affected skin tissue. Indeed, we found an accumulation of mast cells rich in dark-stained granulae, presumably contributing to skin pathology in Δ/Δ animals (Fig. 2e and Supplementary Fig. 4c).

Because of the spontaneous germinal center formation, we analyzed whether the produced antibodies could bind self-antigens. To this end, protein was extracted from different organs isolated from Rag2-deficient animals, separated by SDS–PAGE, and incubated with serum from WT or affected Δ/Δ animals. We observed antibody binding to self-proteins extracted from organs such as lung, stomach, small intestine, pancreas, and eye with blood serum from Δ/Δ animals, although antibody-binding patterns were different between individual animals (Supplementary Fig. 5). In conclusion, our B-cell analysis revealed spontaneous germinal center formation with T_H2-specific Ig class-switch.

T_H2 polarization of T_conv cells

Il4 expression was increased in LN from Δ/Δ mice (Fig. 2a and Supplementary Fig. 4a), therefore we studied T_conv polarization. In spleen and LN, T_conv cells became activated and differentiated into effector/memory T cells by down-regulation of CD62L and up-regulation of CD44. Furthermore, the absolute number increased by about 5–10-fold (Fig. 3a). Treatment of T cells from WT and affected Δ/Δ animals with phorbol-12-myristate-13-acetate (PMA) and Ionomycin to measure intracellular cytokine expression revealed higher frequencies of T_conv cells from affected Δ/Δ animals producing IL-2, IL-4, and IL-13, while T_reg cells remained unchanged (Fig. 3b). This indicated a T_H2 polarization. Since Gata-3 is the master transcription factor of T_H2 cells, we analyzed Gata-3 expression in the T_conv compartment and detected increased frequencies of Gata-3-positive T_conv cells from about 3% in WT to 13% in Δ/Δ animals by flow cytometry (Fig. 3c, d), which was confirmed on RNA level (Fig. 3d, right panel). Co-staining of stimulated T_conv cells with IL-4, IL-13, and Gata-3 revealed that IL-4 and IL-13 was mainly produced by Gata-3-positive T_H2 cells (Fig. 3c). To examine closely the link between the observed T_H2 bias and the profound B cell phenotype, we analyzed T-follicular helper (Tfh) cells in LNs by Cxcr-5 and PD-1 co-staining. We observed an increase in the Cxcr-5⁺PD-1⁺ Tfh cell fraction and total numbers (Fig. 3e). Furthermore, we identified significantly more Gata-3-polarized Tfh cells in Δ/Δ animals (Fig. 3f).

Environmental conditions influence disease development

Approximately 40% of Foxp3^Cre,YFPRbpj^Δ/Δ mice developed a T_H2-polarized disease, while 60% of animals did not get sick during a 20-week observation period (Fig. 1a). Since most mice that developed a pathology were older than 10 weeks, environment-related causes might influence the phenotype in Δ/Δ animals. Therefore, we transferred embryos into a new breeding facility with individually ventilated cages and a defined altered Schaedler flora (details in Methods section). In this new environment, the incidence rate dropped and about 90% of animals remained healthy until 20 weeks of age (Fig. 4a). These findings indicate that environmental factors influence T_H2-disease onset in Δ/Δ animals. To further investigate this, we infected healthy young Δ/Δ and control animals from the new breeding facility with the parasitic nematode Strongyloides ratti (S. ratti). Infective larvae penetrate the skin and migrate within 3 days to the small intestine. Normally, immune competent mice mount a canonical T_H2 response and resolve the infection within 2–4 weeks²². Final expulsion of parasites from the intestine is predominantly mediated by IL-9 and activated mucosal mast cells²³. Especially, Foxp3⁺ T_reg cells have been shown to expand following S. ratti infection²⁴. Therefore, we infected two cohorts of Δ/Δ and WT animals with S. ratti infectious larvae into the footpad and examined animals after 6 (cohort 1) and 14 days (cohort 2, Fig. 4b). Already 6 days post infection (p.i.), T_reg frequency of CD4⁺ T cells in the spleen increased in Δ/Δ animals to about 28%, while WT animals had normal T_reg frequencies at that time point (about 9%) (Fig. 4c). The three-fold increase in T_reg frequency was persistent and also measured at 14 days p.i. (Fig. 4c). At the 14-day time point, T_conv cells expressed more Gata-3 in Δ/Δ animals (Fig. 4d), as well as more IL-4, IL-13, and IL-9, but not IFN-γ (Fig. 4e–g). The fraction of Gata-3-high expressing T_reg cells increased in the Δ/Δ animals from day 6 to day 14 p.i. to 56%, while at the same 14-day time point, the WT animals had significant lower numbers (27%, Fig. 4d). Furthermore, we detected increased levels of IgE, but not IgM, in blood serum of Δ/Δ animals (Fig. 4h), mirroring the phenotype observed during our initial breeding (Fig. 2c). Interestingly, on day 6, Δ/Δ animals with stronger T_H2-polarization and increased IL-9 production displayed reciprocally reduced output of S. ratti DNA (Fig. 4i) and a strong trend towards reduced numbers of parasitic females in the intestine (Fig. 4j). In summary, our data indicate that disease development in Δ/Δ animals was influenced by environmental factors, such as the breeding environment, and that the T_H2-inducing parasite S. ratti could trigger the onset of this phenotype in Δ/Δ animals.

Characterization of Rbpj-deficient T_reg cells

Mice with Rbpj-deficiency in T_reg cells spontaneously developed a T_H2-type disease. Since the Rbpj-deficiency was specific to T_reg cells, we aimed at dissecting the molecular properties of those. Therefore, we measured the expression of classical T_reg associated proteins, such as Foxp3, cytotoxic T-lymphocyte-associated protein 4 (Ctla-4), or Ikaros family zinc finger 2 (Helios), and did not detect obvious differences between T_reg cells from WT or affected Δ/Δ animals (Fig. 5a). Analogously, the methylation status of the T_reg-specific demethylated region (TSDR), a well-described methylation-sensitive cis-regulatory region required for durable expression of the Foxp3 gene²⁵, was unchanged (Fig. 5b). The investigation of the T cell receptor (TCR) repertoire revealed no abnormal TCR beta J-chain usage (Fig. 5c), nor any dominant clones (Fig. 5d), and only a small increase in overall clonality and decreased entropy (Fig. 5e). As T_reg cells were abundantly present in lymphatic organs of affected Δ/Δ animals, we analyzed peripheral tissues. T_reg frequency and total numbers were either equal or elevated (40-fold in affected skin tissue and about three-fold in the lung, Fig. 5f–g and Supplementary Fig. 6) in affected Δ/Δ animals. Since T_reg cells co-opt parts of T_H subset programs to specifically suppress those T_H subset responses, we analyzed Gata-3 protein expression. Indeed, we could detect a strong increase of Gata-3 high-expressing T_reg cells isolated from affected Δ/Δ compared to WT animals (about 45% vs. 12%, respectively), which could be confirmed on RNA level (Fig. 5h). Furthermore, we investigated T follicular regulatory cells based on the expression of Cxcr5⁺ and PD1⁺ (Tfr; CD4⁺CD25⁺Foxp3⁺Cxcr5⁺PD1⁺) in WT and Δ/Δ animals. The total number of PD1⁺ Tfr cells was increased in LNs of affected Δ/Δ animals (Fig. 5i), excluding the possibility that a loss of Tfr cells was responsible for the observed phenotype. In addition, we co-stained PD-1 and Gata-3 and noticed a strong increase in Gata-3⁺PD-1⁺ T_reg cells in Δ/Δ mice (Fig. 5j).

Altered gene expression in Rbpj-deficient T_reg cells

To dissect molecular characteristics, we performed array-based gene expression analysis of T_reg cells isolated from affected Δ/Δ and WT animals (Fig. 6a). Six hundred and ninety probes were found to be differentially expressed. For example, we observed a strong up-regulation of IL-7 receptor (Il7r) and Killer cell lectin-like receptor subfamily G member 1 (Klrg1) genes, while Bcl-2-like protein 11 (Bcl2l11) and Deltex-1 (Dtx1) were under-represented in Δ/Δ T_reg cells (Fig. 6a, b). Bcl2l11 serves as translocator of apoptosis-inducing factors and regulator of mitochondrial depolarization^26,27. Indeed, Rbpj-deficient T_reg cells showed less caspase-3 activity when compared to WT T_reg cells (Fig. 6c and Supplementary Fig. 7). Besides differences in apoptosis, Rbpj-deficient T_reg cells also up-regulated the Il7r gene encoding for the IL-7 receptor (IL-7R, CD127). In thymus and spleen, about 75% of T_conv cells expressed the IL-7R, and this was not different in WT and Δ/Δ animals. In contrast to this, T_reg cells from Δ/Δ mice showed a strong increase in IL-7R expression in the spleen (Fig. 6d). Since the IL-7R is involved in survival and proliferation²⁸, we correlated IL-7R expression with the overall frequency of T_reg cells in unaffected, as well as phenotypically affected Δ/Δ animals of varying age. A good correlation between both parameters could be observed (r = 0.81), suggesting that IL-7R expression supports T_reg accumulation (Fig. 6e). To further validate this, we co-stained IL-7R expression with KI-67, a widely accepted cell proliferation marker²⁹ (Fig. 6f). The majority of T_reg cells in Δ/Δ animals expressed both IL-7R and KI-67, which was not seen in WT animals (Fig. 6f), indicating that IL-7R high-expressing T_reg cells were the proliferating fraction in Δ/Δ animals. Next, we measured the phosphorylation of Stat-5, a downstream component of the IL-7R signaling cascade³⁰. With escalating doses of IL-7, Rbpj-deficient T_reg cells phosphorylated significantly more Stat-5, indicating elevated cytokine sensitivity (Fig. 6g). Since Klrg1 was up-regulated in the gene expression profile (Fig. 6a), we co-stained Klrg1 and IL-7R expression (Fig. 6h). Indeed, while a Klrg1⁺IL7R⁺ double-positive population was almost absent in T_conv cells or WT T_reg cells, the spleens of affected Δ/Δ animals harbored about 60% Klrg1⁺IL7R⁺ T_reg cells (Fig. 6h). Therefore, both IL-7R and Klrg1 were valuable parameters to identify the intensively proliferating subpopulation of T_reg cells in affected Δ/Δ mice.

IL-7R⁺Klrg1⁺ T_reg cells are reminiscent of tisT_regST2

In a recent publication, we described a tissue-resident T_reg population characterized by the expression of, amongst others, IL-7R, Klrg-1, IL-33 receptor alpha (ST2), and Gata-3⁷. This T_reg subset, mainly present within tissues and T_H2-polarized, was called tisT_regST2⁷. Since we detected a stong enrichment of Gata-3, Klrg1, and IL-7R-expressing T_reg cells specifically in affected Δ/Δ animals, we performed a co-staining for ST2 and Klrg1 (Fig. 7a). About 2% of T_reg cells from spleens of WT animals co-express ST2 and Klrg1 compared to 50–60% of spleen T_reg cells from affected Δ/Δ animals (Fig. 7a). A co-staining with KI-67 and Gata-3 revealed that Klrg1⁺ST2⁺ T_reg cells in both WT and Δ/Δ animals were T_H2 polarized and proliferating (Fig. 7a, middle panel). Klrg1⁺ST2⁺ were also significantly increased in lymph nodes (WT: 2%, Δ/Δ: 27%) and skin tissue (WT: 48%, Δ/Δ: 65%, Fig. 7b). To compare the tissue-like gene expression program of Klrg1⁺ tisT_regST2-like cells in Δ/Δ animals on a broader scale, we sorted Δ/Δ Klrg1⁺, Δ/Δ Klrg1⁻, WT Klrg1⁺ and WT Klrg1⁻ T_reg cells from spleen and performed RNA sequencing (RNA-seq) analysis. We extracted RNA-seq data from fat, skin and LN-derived bulk T_reg cells from a previous study⁷ and normalized all datasets. We then plotted 106 reported tisT_regST2 genes in a heatmap (Fig. 7c). Interestingly, there was a strong gene expression overlap between fat and skin T_reg-differentially regulated genes with spleen Δ/Δ-derived and WT-derived Klrg1⁺ T_reg cells, indicating that the majority of Δ/Δ-T_reg cells from affected mice indeed displayed a tisT_regST2-like signature. Still, when listing key genes identifying tissue T_reg cells from fat (Pparg) or skin (Gpr55), as well as tissue T_reg effector molecules such as Il10 and amphiregulin (Areg), the Δ/Δ Klrg1⁺ tisT_regST2-like population in the spleen of Δ/Δ animals did not express comparable levels of these markers (Fig. 7d). This indicated that they were generated in the lymphoid tissue rather than extravasated from non-lymphoid tissues. To analyze differences in more detail, we prepared MA plots for comparisons between all four groups (Fig. 7e). The comparison between WT Klrg1⁺ vs. Δ/Δ Klrg1⁺ T_reg cells revealed 2036 differential expressed genes, which include the molecular changes associated with loss of Rbpj (Fig. 7e; left panel). This group contains well-known Rbpj target, such as Dtx1^31,32, as well as suppression-related proteins such as Id-3³³. The comparison between WT Klrg1⁻ vs. Δ/Δ Klrg1⁺ T_reg cells revealed 3330 differential expressed genes (Fig. 7e; right panel), which include tissue-T_reg-related genes, such as Klrg1 and Il1rl1, as well as suppression-related proteins such as Bach2³⁴. In summary, we showed that affected Δ/Δ animals harbor a strongly increased tisT_regST2-like population in their lymphoid tissues.

Genome-wide chromatin accessibility of WT and Δ/Δ T_reg cells

In affected Δ/Δ animals, Klrg1⁺ tissue-like T_reg cells constitute the majority of all spleen T_reg cells, while in WT animals, Klrg1⁻ non-tissue T_reg cells dominate the T_reg pool at large. To obtain insights into their gene-regulatory landscapes, we isolated both populations and performed the Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq)³⁵. In total, across cell types and replicate experiments, we detected 68,214 ATAC-seq peaks throughout the genome (Fig. 8a). Reference genome annotation revealed that 17.5% of the peaks located to promoters, 41.5% to introns, 2.0% to exons, and 39% to intergenic genomic regions (Fig. 8b). In Δ/Δ T_reg cells, about 3400 regions were more accessible compared to WT T_reg cells, while in WT T_reg cells, 10,816 regions were more accessible in comparison to Δ/Δ T_reg cells (Fig. 8b, c). We next asked which transcription factor motifs were enriched in differential accessible chromatin regions to identify potential drivers of WT-specific and Δ/Δ T_reg-specific gene-regulatory programs. De novo motif analysis revealed a strong Gata transcription factor signature in Δ/Δ T_reg-specific regions (24.38%) vs. background sequences (13.22%) with a high p-value (10⁻⁶⁸) and score (0.95) (Fig. 8d, Supplementary Fig. 8a). Of the relevant Gata family members, only Gata-3 was significantly induced in Δ/Δ T_reg cells (Fig. 8d, lower panel). In addition, we identified enrichment of Ets, Klf, and AP-1-binding sites in Δ/Δ T_reg-specific regions (Supplementary Fig. 8a). WT T_reg-specific regions were dominated by Ets, Tcf, Stat, as well as Nur77 motifs, and we also identified significant enrichment of a motif highly similar to the recently described Rbpj consensus-binding site (Supplementary Fig. 8b). Examples of ATAC-seq signals along with occurrences of Rbpj motifs (both de novo identified and previously identified consensus motifs) at key genes are shown in Fig. 8e–h and Supplementary Fig. 8c. Earlier, we showed that Foxp3 expression was unchanged between Δ/Δ and WT T_reg cells (Fig. 5a) and Foxp3 TSDR demethylation was unaffected by loss of Rbpj (Fig. 5b). Accordingly, the ATAC-seq profile at the Foxp3 locus remained unchanged between both groups (Fig. 8e). Peak calling identified the TSDR region as highly accessible region in both WT and Δ/Δ T_reg cells, but no Rbpj-binding motif has been detected in this or any other part of the Foxp3 gene. The Rbpj locus was accessible in both WT and Δ/Δ T_reg cells, indicating that Rbpj was not required to open its own locus (Supplementary Fig. 8c). In contrast to this, tissue-T_reg-related genes, such as Klrg1 or Il1rl1 (ST2) showed significant ATAC-seq signals around the promoter and potential enhancer sites in Δ/Δ T_reg cells, and Rbpj-binding motifs were also found in these regions (Fig. 8f). These changes in ATAC-seq signals translated into enhanced expression of Il1rl1 and Klrg1 (Fig. 8i). In addition to tissue T_reg-related genes, T_reg suppressive function-associated genes, such as Il2ra, Dtx1, and Bach2 also displayed a distinct ATAC-seq profile: in intragenic and/or enhancer sites, WT T_reg cells had enriched signals compared to Rbpj-deficient T_reg cells (Fig. 8g). Again, Rbpj-binding motifs were detected in differential peaks and ATAC-seq profiles correlated well with changes in gene expression: Il2ra gene expression was significantly down-modulated in Δ/Δ T_reg cells, and expression of Dtx1 and Bach2 was almost completely lost (Fig. 8i). Interestingly, Bach2 has recently been described as a key transcription factor involved in regulating T_H2 polarization by inhibiting Gata-3 expression³⁴. Other T_H2-polarized-related regions were also differentially accessible in Δ/Δ T_reg cells, e.g., the T_H2 locus control region (Rad50), Il10 and Areg, but the corresponding genes were not expressed (Fig. 8h, Fig. 7d and Supplementary Fig. 8c). Taken together, our data suggest that both Rbpj and Gata3 influence the expression of key tisT_regST2-related genes, and that the genomic deletion of Rbpj in concert with a Gata-3-inducing T_H2-type inflammatory environment lead to the massive differentiation and expansion of tisT_regST2-like cells in Δ/Δ animals.

T_H2-polarized T_reg fail to suppress T_H2 responses in vitro

But why are these T_H2-polarized T_reg cells not controlling the T_H2-response anymore? Recently, it was shown that Lilrb4a (encoding the protein ILT3) expressing T_reg cells were unable to regulate T_H2-responses due to their inability to control the maturation of a specific T_H2-promoting DC subset³⁶. This subset of DCs is characterized by the expression of PD-L2 and IRF-4^36,37,38. Our ATAC-seq data identified a highly accessible region at the Lilrb4a promoter in Klrg1⁺ST2⁺ T_reg cells isolated from affected Δ/Δ animals, and a Rbpj-binding site was also predicted in this region (Fig. 9a). Indeed, enhanced activity at the Lilrb4a promoter resulted in increased Lilrb4a expression in Klrg1⁺ST2⁺ T_reg cells from affected Δ/Δ animals (Fig. 9b). In addition, Klrg1⁺ST2⁺ T_reg cells from WT animals expressed more ILT-3 than Klrg1⁻ST2⁻ non-tissue type T_reg cells, suggesting a general mechanism of ILT3-expression during differentiation of the tisT_regST2-like gene expression program. Our de novo motif analysis indicated that Gata-3 was responsible for large parts of the tisT_regST2-like signature in Klrg1⁺ST2⁺ T_reg cells from affected Δ/Δ animals (Fig. 8d). To study the link between ILT3⁺Gata-3 overexpression and control of T_H2 responses, we performed in vitro polarization studies with T_reg cells. IL-4 is the prototype cytokine to induce Gata-3 expression and T_H2 differentiation, and IL-33 is linked to the generation of tisT_regST2 cells⁷. Therefore, we FACS-sorted highly pure T_reg cells from Foxp3^GFP animals and expanded them with anti-CD3/CD28 microbeads, IL-2, IL-4, and IL-33 or without the latter two cytokines as control, for 6 days. Both groups of expanded T_reg cells stayed highly Foxp3 positive (Fig. 9c). Interestingly, we were able to co-induce ILT3 and Gata-3 expression specifically in the IL-4 and IL-33-treated T_reg cells (Fig. 9c). Using this model, we studied the ability of ILT3-expressing T_H2-polarized T_reg cells to influence DC maturation and DC-mediated T_H2 polarization of FACS-sorted CD4⁺Foxp3^-CD62L⁺ naive T cells in vitro. Our data revealed that ILT3-expressing T_H2-polarized T_reg cells profoundly promoted the differentiation of PD-L2⁺IRF4⁺ DCs, a subset described to support T_H2 polarization in vivo^36,37,38 (Fig. 9d). In addition, ILT3-expressing T_reg cells were unable to suppress the T_H2 differentiation of IL-4 and anti-CD3-stimulated naive T cells into Gata-3-polarized effector T cells in the presence of DCs (Fig. 9e). These data strongly indicate that overexpression of Gata-3 and ILT3 renders T_reg cells less able to suppress T_H2 responses in vitro. Finally, we investigated the sensitivity of Rbpj-deficient T_reg cells to T_H2-inducing conditions. To this end, we FACS-sorted and expanded Rbpj-deficient T_reg cells from healthy, young animals, with no pre-existing T_H2 polarization, and compared them to WT T_reg cells. Both groups were treated with escalating doses of IL-4 in vitro (Fig. 9f). Indeed, Rbpj-deficient T_reg cells were more sensitive to the T_H2-inducing IL-4 treatment, translating into enhanced Gata-3 protein and mRNA induction in Δ/Δ T_reg cells (Fig. 9f). This elevated sensitivity towards Gata-3 induction could explain the profound expansion of Gata-3⁺Klrg1⁺ST2⁺ T_H2-polarized T_reg cells in affected Δ/Δ animals, with ameliorated T_H2-suppressive potential.

Discussion

In this study, we identify a previously unrecognized role for Rbpj in T_reg cell-mediated immune homeostasis. Upon T_reg-specific Rbpj deletion in Foxp3^CreRbpj^Δ/Δ animals, mice developed a lymphoproliferative disease with type-2 effector polarized B-cell and T-cell responses. Disease development was environment-related and could be induced by infection with the parasitic nematode S. ratti. The finding that disease development was environment-related could explain the discrepancy to a published study using mice with Rbpj^Δ/Δ T_reg cells, where the authors did not report the lymphoproliferative characteristic²⁰.

But what happened once the proper environmental trigger has been received? Based on our data, we would argue that deleting Rbpj confined the functional capacity of T_reg cells in several ways. First, augmented proliferation potential: the down-modulation of Bcl2l11 could lead to enhanced resistance to apoptosis^26,27, while the up-regulation of the Interleukin-7 receptor promoted proliferation and supported a strong increase in total T_reg numbers in lymphoid tissues. Potential Rbpj-binding sites at the Il7r promoter region have already been reported³⁹. Second, Rbpj is important to restrict the T_H2 differentiation potential of T_reg cells: Rbpj-deficient T_reg cells were more sensitive to IL-4 polarization and concomitantly overexpressed Gata-3 and other T_H2-associated proteins compared to WT T_reg cells. This was also observed upon in vivo infection with parasitic nematodes. Our data showed that about two-times more T_reg cells expressed high levels of Gata-3 in Δ/Δ mice as compared to infected WT mice. As a consequence of Gata-3 expression, Δ/Δ T_reg cells differentiated into T_H2-polarized Klrg1⁺ST2⁺ tisT_regST2-like cells. This differentiation integrated a third critical restriction of T_reg function and suppressive capacity: the loss of T_H2-suppressive capacity via the down-modulation of Bach2, Dtx1, and Il2ra, and the induction of ILT3. Il2ra, encoding for the IL-2 receptor alpha chain (CD25), is required for T_reg suppressive capacity⁴⁰. Bach2 has been described as an important transcription factor required to inhibit Gata-3 expression and T_H2 polarization^34,41, including the expression of ST2⁴², and the Bach2–Batf interaction was shown to control T_H2-type immune responses⁴³. In addition, Bach2 supports the suppressive capacity of T_reg cells, as a loss-of-function study demonstrated that Bach2-deficient T_reg cells failed to prevent disease in a colitis model³⁴. Dtx1, previously reported to interact with Rbpj³², was also shown to be important for T_reg cell suppressive function in a transfer model of colitis³¹. Finally, the induction of ILT3 can lead to a severe defect in controlling T_H2-polarized immune responses via the induction of IRF4⁺PD-L2⁺ DCs³⁶.

This cumulative effect on T_reg suppressive capacity was finally leading to a loss of T_H2 suppressive potential, a state where effector T_H2 cells produced more IL-4, leading to even more Gata-3 expression in Δ/Δ T_reg cells. Gata-3 expression is required to maintain high Foxp3 expression levels and it is important to prevent T_reg differentiation into an effector phenotype^44,45,46. But Gata-3 does not function in a binary on–off mode. It has been reported that Gata-3 over-expression in T cell progenitors changes the identity of developing double-negative thymocytes and drives them into the mast cell lineage^47,48,49. These studies indicate that a well-defined Gata-3 dosage is required for proper T cell development and function. A recent report showed that the IL-4 signaling strength is important for T_reg cell function. By using mice carrying an IL-4Rα chain mutation leading to enhanced IL-4 signaling, the authors demonstrated that these T_reg cells, which were polarized towards a T_H2 cell-like phenotype with high Gata-3 expression levels, had an impaired functionality⁵⁰.

Our data indicate that Rbpj could act as a Gata-3 dosage modifier, adjusting the balance of Gata-3 expression by restricting IL-4 sensitivity as a powerful molecular switch. In addition, published findings report that the Gata3 gene itself is a direct target of Rbpj¹⁸. Therefore, the regulation of Gata-3 expression could be both on the transcriptional, as well as the IL-4 cytokine sensitivity level. Our findings should be considered in the current discussion that T_H2-polarized Gata-3⁺ T_reg cells are better suppressors of the corresponding T_H2-polarized effector T cells, a model proposed based on the complete deficiency of IRF4¹³. We could show that IL-4 and IL-33-induced Gata3^high T_reg cells express significantly more ILT3 upon in vitro expansion and differentiation, a surface receptor shown to interfere with efficient control of T_H2 effector cells³⁶. Gata-3^high T_reg cells were unable to inhibit T_H2 differentiation of IL-4 exposed naive CD4 cells in vitro. In contrast to this, they supported the maturation of a T_H2-promoting IRF4⁺PD-L2⁺ DC subpopulation in vitro.

Our motif analysis of the ATAC-sequencing data revealed a strong Gata signature in Δ/Δ T_reg-specific differentially accessible regions. Many of the affected genes were shared with tisT_regST2 cells, a T_H2-biased T_reg subset normally present within non-lymphoid tissues⁷. Rbpj may regulate the Gata-3-dependent T_regST2 differentiation pathway and, thereby, might limit the access to the T_regST2 compartment in lymphoid organs to allow the maintenance of a diverse T_reg subset pool.

Methods

Mice

Wildtype C57BL/6, congenic B6.SJL-Ptprc^aPepc^b/BoyCrl (CD45.1⁺), and congenic B6.PL-Thy1^a/CyJ (CD90.1⁺) mice were obtained from Charles River Breeding Laboratories (Wilmington, MA, USA) or the Jackson Laboratory (Bar Harbor, ME, USA). B6N.129(Cg)-Foxp3^tm3Ayr mice (Foxp3.IRES-DTR/GFP)⁵¹ were bred to CD45.1⁺ or CD90.1⁺ mice in the animal facility of the German Cancer Research Center (DKFZ). B6.129(Cg)-Foxp3^{tm4(YFP/cre)Ayr}/J, Jackson (FOXP3.IRES-YFP/Cre)⁵² were crossed to Rbpj^fl/fl mice⁵³ to specifically delete Rbpj in T_reg cells. Age-matched littermate controls (Foxp3^Cre,YFP-positive and wildtype for the Rbpj alleles) were used throughout the study. Details about hygiene status and barrier breeding conditions are explained in the following paragraph. Rag2-deficient (B6-Rag2tm1Fwa) lines were used to isolate protein for autoantibody detection. All animals were housed under specific pathogen-free conditions at the respective animal care facilities, and the governmental committee for animal experimentation (Regierungspräsidium Karlsruhe, Regierung von Unterfranken and Behörde für Gesundheit und Verbraucherschutz Hamburg) approved all experiments involving animals. Relevant ethical regulations for animal testing and research were complied with.

Breeding conditions and disease-free survival in barriers

Data from Fig. 1a are derived from animals housed under specified pathogen-free conditions in a specific mouse facility (called barrier 3) of the German Cancer Research Center, fulfilling the criteria given in the FELASA recommendations (animal number, health monitoring, age, agents, methods). All animals were housed in open cages allowing transmission of agents. Research personnel had access to the unit. Routine testing included testing for ectoparasites, endoparasites, bacteria and viruses. In barrier 3, murine norovirus (MNV), Pneumocystis sp. and Staphylococcus aureus have been detected, along with occasional detection of additional opportunistic agents. In our breeding, we observed 79 animals with T_reg lineage-specific bi-allelic Rbpj deletion (Foxp3^CreRbpj^Δ/Δ), of which 21 were sacrificed due to sickness and used for experimentation. Twenty-nine animals <20 weeks old were otherwise healthy but used for experimentation and censored for survival analysis. Thirty-two animals grew older than 20 weeks and were marked healthy during the observation period, although some turned sick later and were used for experimentation. Out of the 67 animals with mono-allelic Rbpj deletion (Foxp3^CreRbpj^wt/Δ) and 79 wildtype animals (Foxp3^CreRbpj^wt/wt), 0 animals showed signs of disease. All Foxp3^CreRbpj^wt/Δ and Foxp3^CreRbpj^wt/wt were littermates of the Foxp3^CreRbpj^Δ/Δ mice, housed in the same cages as their siblings.

To identify the influence of breeding conditions (and therefore the environmental influence) on disease development of Foxp3^CreRbpj^Δ/Δ animals, we transferred the colony to a newly established mouse facility (called barrier A) via embryo transfer (results in Fig. 4a). Mice in this barrier were colonized with a defined Altered Schaedler Flora and are housed in individually ventilated cages. Access is limited to animal caretaker personnel. Until study end, the barrier was completely free of infectious agents listed in the FELASA recommendations. In this barrier, we observed 55 animals with T_reg lineage-specific bi-allelic Rbpj deletion (Foxp3^CreRbpj^Δ/Δ), of which three were sacrificed due to sickness and used for experimentation. Seventeen animals <20 weeks old were otherwise healthy but used for experimentation and censored for survival analysis. Thirteen animals grew older than 20 weeks and were marked healthy during the observation period, although some turned sick later and were used for experimentation. Thirty-three animals were between 10 and 20 weeks old when the observation was stopped and the experiment concluded. Out of the 45 animals with mono-allelic Rbpj deletion (Foxp3^CreRbpj^wt/Δ) and 40 wildtype animals (Foxp3^CreRbpj^wt/wt), 0 animals showed signs of disease.

Rbpj genotyping

Animal tails were digested in digest buffer (50 mM KCl, 20 mM Tris–HCl pH 8.8, 0.00045% Tween20 and Igepal CA-630) with proteinase K overnight at 56 °C. Following inactivation for 10 min at 96 °C, a PCR reaction with Taq polymerase, dNTPs, 2.5 mM MgCl₂, and 0.4 µM Rbpj primers (GTGGAACTTGCTATGTGCTTTG, CTGCCATATTGCTGAATGAAAA, CACATTCCCATTATGATACTGAGTG) was started (95 °C, 5 min; 95°-30 s-58°-30 s-72°-30 s × 35; 72 °C, 5 min). Samples were separated on an agarose gel and gene status analyzed. Similarly, Foxp3-driven presence of Cre recombinase was detected via PCR (AGGATGTGAGGGACTACCTCCTGTA, TCCTTCACTCTGATTCTGGCAATTT; 94 °C, 3 min; 94°-40s-60°-40s-72°-60 s × 30; 72 °C, 3 min).

Flow cytometry and fluorescence-activated cell sorting

Target organs were isolated and single-cell suspensions were established. If applicable, tissues were treated with collagenases and pre-purified according to protocol⁷. Red blood cells were lysed in hypotonic buffers, and cells were either pre-enriched with magnetic bead technology (Miltenyi Biotec) or directly stained with fluorochrome-labeled antibodies. Surface stainings were performed for 30 min at 4 °C, with all antibodies used at 1:100 dilution if not stated otherwise. If applicable, cells were fixed and afterwards permeabilized with the Foxp3 Fix/Perm Buffer set for 60 min at RT. Upon permeabilization, cells were stained intracellularly for 60 min at RT. Flow cytometry samples were acquired on a LSR II, Fortessa II or Canto II flow cytometer (BD Biosciences). For cell counting, AccuCheck counting beads were used (ThermoFisher). Samples for RNA isolation, DNA isolation, or subsequent cultivation were acquired and sorted on ARIA II or ARIA Fusion cell sorting systems (BD Biosciences) with four-way purity settings and an 85 µM nozzle. For RNA collection, cells were sorted in 500 µL of RNA lysis buffer, followed by RNA isolation based on manufacturer’s instructions (RNEasy Mini Kit or RNEasy Micro Plus Kit, Quiagen). For DNA collection, samples were sorted into 500 µL of DNA lysis buffer and DNA was purified according to manufacturer’s protocol (DNEasy Blood and Tissue Kit, Quiagen). To harvest protein, cells were sorted into FCS-containing buffer and afterwards pelleted. Cells were lysed in RIPA buffer.

Real-time PCR

For RNA isolation, antibody-labeled cells were sorted directly into RLT+lysis buffer and purified according to manufacturer’s protocol (RNEasy mini Kit, Quiagen). RNA from whole tissues was isolated with mechanical tissue dissemination using ceramic beads followed by column-based RNA isolation (Innuprep RNA Kit, Analytik Jena). RNA was reversely transcribed into cDNA according to manufacturer’s protocol (Reverse Transcriptase II, life technologies). cDNA was used with Taqman probes and Taqman master mix or with Sybr primers and Sybr master mix in a Viia7 real-time PCR system (all ThermoFisher). Gene expression was normalized to housekeeping gene expression (Hprt) with the formula: relative gene expression = 2^{−(Ct (gene X)—Ct (Hprt))}. Primer sequences and designations are listed in Supplementary Table 1.

Detection of Rbpj protein via Western Blot

T_reg (CD3⁺CD45⁺CD4⁺CD8⁻CD25⁺Foxp3-GFP⁺) and T_conv cells (CD3⁺CD45⁺CD4⁺CD8⁻CD25⁻Foxp3-GFP⁻) were isolated via FACS. Cells were lysed in RIPA buffer and supplemented with Laemmli buffer containing beta-mercaptoethanol. Samples were then heated to 95 °C for 10 min and afterwards separated by SDS–PAGE with pre-cast gels (Biorad). Gels were blotted onto PVDF membranes according to standard protocol. Membranes were blocked with 5% Milk-PBST for one hour at RT followed by incubation overnight with an anti-mouse Rbpj primary monoclonal antibody (Cell signaling clone D10A4) at 1:3000 in 5% Milk-PBST. Membranes were washed and RBPJ antibody was labeled with an HRP-conjugated secondary antibody at 1:10,000 dilution for one hour at RT. Membranes were washed and developed with a chromogenic detection reagent (Thermo Fisher).

Histology and microscopy

Immediately after the animals were sacrificed, organs were carefully excised and stored overnight in freshly prepared 4% formaldehyde solution. Afterwards, samples were embedded, thin-cut (thickness between 3 and 5 µM) and stained. Haematoxylin and eosin (H&E) stainings, periodic acid-Schiff reaction (PAS) stainings, and Giemsa stainings were prepared according to literature⁵⁴. Representative images were acquired on a Zeiss Axioplan microscope equipped with a AxioCam ICc 3 color camera with ZEN 2011 lite (BLUE EDITION) software. Intensity and contrast settings were adjusted for each organ but kept consistent between control and test sample. Foxp3-staining on embedded tissues was performed as follows: first, samples were thin-cut (3–5 µm); second, paraffin was melted (72 °C, 30 min) followed by Xylol (2 × 5 min), and alcohol treatment; third, samples were incubated at 120 °C in Tris–EDTA buffer for 5 min, followed by blocking with peroxidase-block (Dako); then, incubation with primary antibody (Foxp3 FJK-16 s, 1:50, 30 min at RT) and secondary antibody (anti-rat HRP, 30 min at RT) with intermittent washing steps for 5 min (Washing buffer from Dako); last, chromogenic detection solution was added for 10 min at RT (DAB+ substrate, Dako) followed by 1 min incubation with Hematoxylin (Merck) and washing steps.

Immunohistochemistry and slide scanning/image analysis

Foxp3 stainings of embedded tissues were prepared with a fluorescence-labeled anti-Foxp3 antibody as described in literature⁵⁵. To visualize germinal centre formation, lymph node samples were recovered from animals, immediately frozen in TissueTec buffer (Sakura Fineteck Europe) on cold carbon dioxide pellets and stored at −80 °C. Individual samples were cut from the tissue block and fixed with acetone for 10 min. Following blocking with 10% FCS, slides were incubated with AF488-labeled anti-mouse GL7 antibody (1:20), AF594-labeled anti-mouse IgD antibody (1:20), and AF 647-labeled anti-mouse CD4 antibody (1:20) overnight at 4 °C. After washing, samples were mounted with a fluorescent mounting medium (Dako). Unstained or single-stained samples were prepared and imaged as background staining controls. Samples were imaged on a motorized Zeiss inverted Cell Observer.Z1 with a mercury arc burner HXP 120/Colibri LED module, as well as a gray scale CCD camera AxioCam and a color CCD camera AxioCam MRc. Images were sequentially scanned and assembled with ZEN 2011 lite (BLUE EDITION) software. In order to obtain full-size images of lymph nodes, we first scanned the green channel (AF488, GL-7) followed by red channel (AF594, IgD), and blue channel (AF647, CD4). Contrast and fluorescence intensity were adjusted for all samples in parallel. To reduce robotic scanning errors, images were stitched to smooth transition areas. For Foxp3 stainings, staining intensity was normalized for each image.

Gene expression analysis with bead chips and statistics

For whole lymph-node gene expression analysis, inguinal lymph nodes from WT and affected Δ/Δ animals underwent mechanical tissue dissemination using ceramic beads followed by column-based RNA isolation, as described earlier. For gene expression analysis of T cells, we FACS-isolated spleen-derived T_reg (CD3⁺CD4⁺CD8⁻CD45⁺CD25⁺Foxp3-YFP⁺) and T_conv cells (CD3⁺CD4⁺CD8^-CD45⁺CD25⁻Foxp3-YFP⁻) from WT and Δ/Δ animals. RNA was isolated (RNEasy mini kit) and the DKFZ Genomics and Proteomics Core Facility amplified and hybridized material to the Illumina MouseWG-6 v2.0 Expression BeadChip. Microarray scanning was done using an iScan array scanner. Data extraction was done for all beads individually, and outliers were removed when the absolute difference to the median was >2.5 times MAD (2.5 Hampelís method). Expression values were quantile normalized and log2-transformed. Differentially expressed probes between groups were identified using the empirical Bayes approach based on moderated t-statistics as implemented in the Bioconductor package limma. All p-values in limma were adjusted for multiple testing using Benjamini–Hochberg correction in order to control the FDR. Probes with an FDR <5% were considered statistically significant.

Statistical analysis of data

Data were analyzed with Prism software. We used a log-rank Mantel–Cox test in Kaplan–Meier survival curves (Figs. 1a and 4a), Mann–Whitney testing (Figs. 1b, c, d; 2b, c; 3a, b), unpaired t testing (Figs. 3d, e, f; 4c, d, f, g, i, j; 5e, g–j; 6c, d, f, h; 7a, b; 8d, i; 9b, c, Supplementary Fig. 1a, c, d; Supplementary Fig. 2a), paired t testing (Figs. 4h and 9d–f), one-way ANOVA with Bonferroni post-testing (Figs. 5b and 6b; Supplementary Fig. 1b, Supplementary Fig. 2b) or Newmann–Keuls post-testing (Fig. 6b), or two-way ANOVA with Bonferroni post-testing (Figs. 6g and 9f, Supplementary Fig. 3). RNA-sequencing data in Figs. 2a, 6a and 7d, e were statistically evaluated as described in the respective paragraph of the Methods section. The respective number of animals (n) as well as the statistical test is also listed in the figure legend. All statistical test results and data used to calculate statistics are incorporated into the source data file.

Blood serum Ig subtype analysis

Blood was extracted via cardiac puncture from sacrificed animals and allowed to clot for at least 15 min. Afterwards, samples were centrifuged at 13,000 × g for 15 min and serum was collected. ELISA plates (Costar #9018) were pre-coated with goat-anti-mouse IgG + IgM (Dianova #115-005-0068) or goat-anti-mouse IgE (Biozol #1110-01) overnight at 4 °C, followed by washing and blocking (PBS 0.2% gelatine 0.1% NaN₃). Wells were incubated with serial dilutions of serum or control antibody for one hour at RT, followed by four washing steps. Peroxidase-conjugated secondary antibodies were added at 1:1000 dilution in PBS and incubated for one hour at RT, again followed by four washing steps. Plates were developed with 1 mg/mL OPD in 0.1 M KH₂PO₄ (pH 6.0) solution with 1 µL/mL of 30% H₂O₂ solution. Once colorimetric reaction was complete, incubation was stopped with 25 µL 1 M H₂SO₄ and read on a ELISA photometer at 490 nm wave length. All antibodies are listed in Supplementary Table 1.

Isolation of blood plasma and blood serum

Blood was collected from sacrificed mice via cardiac puncture. Blood was mixed with Heparin–PBS to a final concentration of 20 U/mL Heparin. Samples were centrifuged at 3000 × g for 15 min at 4 °C. Blood parameters were measured by photometric analysis on the ADVIA 2400 system (Siemens Healthcare Diagnostics) in the Zentrallabor (Medical Clinic-1, Analysezentrum) of the Heidelberg University Clinic. For blood serum collection, blood was extracted via cardiac puncture and allowed to clot for at least 15 min. Afterwards, samples were centrifuged at 13,000 × g for 15 min and blood serum was collected.

Autoantibody detection via Western Blot

Following organs and tissues from RAG2^-/- animals were isolated: brain, eye, spleen, lymph nodes, pancreas, salivary gland, stomach, liver, kidneys, heart, lung, testis, small, and large intestine. Tissues were weighted and adjusted to 100 mg, followed by addition of 500 µL ProteoJET mammalian cell lysis reagent (Fermentas) supplemented with 2X protease inhibitor (Roche Diagnostics). Tissues were mechanically dissected using scissors and incubated for 10 min at RT on an orbital shaker. Samples were then centrifuged for 15 min at 16,000 × g and supernatant containing protein lysate was harvested. Protein concentration was determined by BCA assay (Pierce). Twenty micrograms of protein were loaded per well on a pre-cast SDS gel (Biorad), followed by gel electrophoretic separation. Each gel contained eight wells with the same protein load (e.g. pancreas). After transfer on a PVDF membrane, this membrane was cut into eight strips containing the protein of interest. PVDF membranes were blocked with 5% Milk–PBST solution for one hour at RT following incubation with peripheral blood serum of individual mice (4 WT, 4 Δ/Δ) overnight at 4 °C (concentration 1:500 in 5% Milk–PBST). Individual strips were washed three times with PBST, followed by incubation with a HRP-conjugated donkey-anti-mouse IgG antibody (concentration 1:3000). After washing, strips were re-assembled and HRP activity was detected with a chromogenic substrate (Thermo Fisher).

Infection of WT and Δ/Δ animals with S. ratti

Animal experimentation was conducted at the animal facility of the Bernhard Nocht Institute for Tropical Medicine in agreement with the German animal protection law under the supervision of a veterinarian. The experimental protocols have been reviewed and approved by the responsible federal health authorities of the state of Hamburg, Germany, the Behörde für Gesundheit und Verbraucherschutz. Mice were sacrificed by cervical dislocation under deep CO₂ narcosis. Two cohorts with five WT and five healthy Δ/Δ animals each were infected on day 0 by injection of 1000 S. ratti larvae subcutaneously into the footpad. Cohort 1 was sacrificed 6 days after infection and intestinal parasite count, stool PCR, and flow cytometry (T_reg frequency, cytokine restimulation) were performed. Cohort 2 was sacrificed 14 days after infection; stool collection on day 6, 8, 10, 14 followed by stool-PCR; flow cytometry (T_reg frequency, cytokine restimulation) on day 14. To count the number of adult parasitic females in the gut, the small intestine was flushed slowly with tap water to remove feces, sliced open longitudinally and incubated at 37 °C for 3 h in a Petri dish with tap water. The released adult females were collected by centrifugation for 5 min at 1200 rpm and counted. To quantify the release of S. ratti larvae by infected mice, the feces of individual mice was collected over 24 h periods and DNA from representative 200 mg samples was extracted as described⁵⁶. Two hundred nanograms of DNA was used as a template for qPCR specific for S. ratti 28 S ribosomal RNA gene as described⁵⁷. For analysis of serum antibodies, blood was collected from infected mice at the indicated time points and allowed to coagulate for 1 h at RT. Serum was collected after centrifugation at 10,000×g for 10 min at RT and stored at −20 °C for further analysis. Strongyloides-specific IgM in the serum was quantified by ELISA, as described⁵⁷. Serum concentration of IgE was quantified using the IgE ELISA kit according to the manufacturers recommendations. In flow cytometry experiments, spleens were collected, red blood cells lysed and samples stained as described previously. For cytokine restimulation experiments, single-cell suspensions were incubated with cell stimulation cocktail plus transport inhibitor for 6 h at 37 °C followed by intracellular cytokine staining. To measure secreted IL-9 levels, mesenteric LN-derived cells were mashed and incubated with anti-CD3 antibody (1 µg/mL) for 72 h at 37 °C with 1 × 10⁶ cells per well. IL-9 levels were quantified using the IL-9 ELISA kit according to the manufacturers recommendations.

Methylation of the TSDR

Genomic DNA of sorted cell populations was purified according to manufacturer’s guidelines using the DNEasy Blood and Tissue kit (Quiagen). DNA purity and concentration were measured with a NanoDrop^® photometer. Bisulfite-conversion was performed using the EpiTect Bisulfite Conversion Kit (Quiagen). Barcode-labeled primers for the Foxp3 CNS2 (TSDR) were used to generate PCR amplicons from bisulfite-converted DNA (ForP: TGGGTTTTTTTGGTATTTAAGAAAG; RevP: AAAAAACAAATAATCTACCCCACAA). PCR amplicons were separated from primer dimers on a 2% agarose gel and purified using a Quick Gel Extraction Kit (Life Technologies). PCR amplicons were processed on a GS Junior Sequencer (Roche). Sequence reads were aligned to the BS-converted mouse genome and visualized.

In vitro T_reg suppression assay

First, we isolated MHCII-positive antigen-presenting cells (CD8⁻CD25⁻Foxp3-GFP⁻MHCII⁺CD90.1⁻CD90.2⁻) as well as CD4-positive T-responder cells (CD4⁺CD8⁻CD25⁻Foxp3-GFP⁻MHCII⁻CD90.1⁺CD90.2⁻) from Foxp3^{GFP, CD90.1} mice. T-responder cells were then labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE) at 1 µM concentration in 10 mL cell culture medium for 15 min at RT, followed by several washing steps. Next, T_reg (CD4⁺CD8⁻CD25⁺Foxp3-YFP⁺MHCII⁻CD90.1⁻CD90.2⁺) and T_conv (CD4⁺CD8⁻CD25⁻Foxp3-YFP⁻MHCII⁻CD90.1⁻CD90.2⁺) cells from WT and affected Δ/Δ mice were isolated by FACS and serially diluted. In each well, 50,000 T-responder cells, 100,000 MHCII-positive APCs, and serially diluted T_reg or T_conv cells were added. To stimulate APC-driven T-responder cell proliferation, 2 µg/mL anti-CD3 antibody was added (Biolegend clone OKT3). Cells were incubated for 5 days at 37 °C, followed by re-staining for flow cytometric analysis of CFSE-dye dilution in T-responder cells.

TCR sequencing

Single cell suspensions from spleen and lymph nodes (combined axial, cervical, brachial) from individual WT vs. Δ/Δ mice were established, and red blood cells lysed. T_reg cells (CD3⁺CD4⁺CD8⁻CD45⁺CD25⁺Foxp3-YFP⁺) were pre-enriched with CD25-magnetic bead-based purification and sorted via FACS. Genomic DNA was isolated with the DNEasy blood and tissue kit according to manufacturer’s instructions and measured on nanodrop photometer. Five hundred nanograms of gDNA per individual mouse were shipped to Adaptive Biotechnologies (Seattle, WA) for TCR sequencing. Data were analyzed with online tools provided by Adaptive Biotechnologies.

Active caspase-3 assay and intracellular cytokine secretion

For measurement of active caspase-3, single-cell suspensions were resuspended in FCS-containing cell culture medium (Gibco) and 1 µL of Red-DEVD-FMK (abcam) was added to each well in a 96-well tissue culture plate. Samples were incubated for 60 min at 37 °C, followed by washing with supplied wash buffer and surface antibody staining. As a positive control, splenocytes were incubated for 5 min at 42 °C. To measure intracellular cytokines, splenocytes, or lymph node-derived cell suspension were resuspended in FCS-containing cell culture medium. For stimulation, cells received 1X PMA-Ionomycin cocktail plus transport inhibitor, whereas controls received 1X transport inhibitor only cocktails (eBiosciences). Samples were incubated for 6–8 h at 37 °C, followed by surface and intracellular staining with Foxp3 Fix/perm staining mix.

Detection of phosphorylated Stat5 or Stat6

Spleen-derived single-cell suspensions were resuspended in FCS-containing cell culture medium (Gibco) and incubated at 37 °C in a 96-well tissue culture plate. Then, mouse recombinant IL-2, IL-4, or IL-7 were added in a 1+9 titration curve with a pre-warmed 2X mixture to cell suspension. Cells were incubated with the respective cytokine for 10 min at 37 °C. Afterwards, cells were centrifuged at 1000 × g and 4 °C for 2 min, followed by two washing steps with cold FACS buffer. Cells were fixed with 1X Fixation buffer (BD Fixation Buffer) for 30 min at 4 °C. Afterwards, cells were centrifuged and resuspended in −20 °C cooled Perm Buffer (BD PermBuffer III) and incubated for 30 min at 4 °C. Afterwards, cells were washed twice and stained with surface and intracellular antibodies for 30 min at 4 °C, followed by analysis via flow cytometry.

RNA sequencing

cDNA was generated and amplified using 4.8 ng of total RNA (RNEeasy Mini Kit) and SMARTer Ultra Low Input RNA for Illumina Sequencing—HV (Clontech Laboratories, Inc.) according to the manufacturer’s protocol. Then, sequencing libraries were prepared using the NEXT ChIP-Seq Library Prep Master Mix Set for Illumina (New England Biolabs) according to the manufacturer’s instructions with the following modifications: The adapter-ligated double-stranded cDNA (10 µL) was amplified using NEBNext Multiplex Oligos for Illumina (New England Biolabs, 25 µM primers), NEBNext High-Fidelity 2x PCR Master Mix (New England Biolabs) and 15 cycles of PCR. Final libraries were validated using Agilent 2100 Bioanalyzer (Agilent Technologies) and Qubit flourometer (Invitrogen), normalized and pooled in equimolar ratios. 50 bp single-read sequencing was performed on the Illumina HiSeq 2000 v4 according to the manufacturer’s protocol.

Mapping of RNA seq data, statistical evaluation, and plotting

For all samples, low-quality bases were removed with Fastq_quality_filter from the FASTX Toolkit 0.0.13 (http://hannonlab.cshl.edu/fastx_toolkit/index.html) with 90% of the read needing a quality phred score >20. Homertools 4.7⁵⁸ were used for PolyA-tail trimming, and reads with a length <17 were removed. PicardTools 1.78 (https://broadinstitute.github.io/picard/) were used to compute the quality metrics with CollectRNASeqMetrics. With STAR 2.3⁵⁹, the filtered reads were mapped against mouse genome 38 using default parameters. Count data and RPKM tables were generated by mapping filtered reads against union transcripts (derived from Mouse Ensembl 90) using a custom pipeline. Mapping was carried out with bowtie2 version 2.2.4⁶⁰ against union mouse genes: every gene is represented by a union of all its transcripts (exons). The count values (RPKM and raw counts) were calculated by running CoverageBed from Bedtools v2.26.0⁶¹ of the mapped reads together with the mouse annotation file (Ensembl 90) in gtf format and parsing the output with custom perl scripts. The input tables containing the replicates for groups to compare were created by a custom perl script. For DESeq2⁶², DESeqDataSetFromMatrix was applied, followed by estimateSizeFactors, estimateDispersions, and nbinomWald testing. The result tables were annotated with gene information (gene symbol, gene type) derived from the gencode.vM16.annotation.gtf file. The results were then filtered for protein-coding genes according to the gencode.vM16.annotation.gtf file. To account for the gender differences between male and female mice, 4571 differently expressed genes with an adj. p-value < 0.05 in the comparison of wildtype spleen-derived TCRβ⁺CD4⁺CD8⁻CD25⁺Foxp3(GFP)⁺Klrg1⁻ST2⁻ (female) with wildtype spleen-derived TCRβ⁺CD4⁺CD8⁻CD25⁺Foxp3(GFP)⁺Klrg1⁻ST2⁻ (male) cells were excluded from the input tables into DESeq2 and the analyses were rerun with the remaining genes. MA plots were generated as described in ref. ⁶³. PCA plots and tables were generated using the plotPCA function of DESeq2 for the most variable 500 genes after applying the DESeq2 varianceStabilizingTransformation to the data.

ATAC-seq

The assay for transposase-accessible chromatin using sequencing (ATAC-seq) was performed according to Buenrostro et al. ³⁵ with some modifications. In brief, about 50,000 FACS-isolated cells were pelleted on with 10,000 × g for 3 min and supernatant removed. Cells were tagmented at 55 °C for 8 min in 50 µL 1x tagmentation buffer including 2.5 µL transposome from the Nextera DNA library prep kit and 0.01% digitonin. The transposome was inactivated by addition of 20 µL 5 M guanidinium thiocyanate, and the DNA was purified with two volumes, i.e., 140 µL, of DNA-binding beads (HighPrep beads). The DNA was PCR amplified with a LightCycler 480 (Roche) in a 50 µL reaction with the NEBNext High Fidelity mix including 0.5 µL 1x SYBRGreen, forward primer Tn5McP1n (AATGATACGGCGACCACCGAGATCTACACTCGTCGGCAGCGTC) and barcoded reverse primers Tn5mCBar (CAAGCAGAAGACGGCATACGAGAT (8–9N barcode) GTCTCGTGGGCTCGG); 72 °C, 5 min; 98 C, 30 s (gap repair and initial melting); then cycling with 98 °C, 10 s; 63 °C, 30 s; 72 °C, 30 s for 15 cycles when all amplifications turned into saturation. PCR products were purified with 1.4 volumes (70 µL) magnetic beads. Ten microliters of eluates were sequenced on a HiSeq2000, 125 bp paired-end.

ATAC-seq data analysis

ATAC-seq read data were processed as described previously⁶⁴. Reads were trimmed using Skewer⁶⁵ and aligned to the mm10 assembly of the murine genome using Bowtie2⁶⁰ with the ‘-very-sensitive’ parameter. Duplicate reads were removed using sambamba markdup⁶⁶, and only properly paired reads with mapping quality >30 were kept. Reads were shifted as previously described to account for the transposition event³⁵. Bigwig files were created using bedtools⁶¹. Preprocessed ATAC-seq data was analyzed with the HOMER package⁵⁸. In brief, mapped sequencing reads were transformed to tag directories using the option ‘-sensitivity 3’. Peaks were called on all reads in all samples using the ‘findPeaks’ command with the options ‘-style factor -size 350 -minDist 300 -L 2’. Peaks from the mitochondrial genome and ENCODE blacklisted regions⁶⁷ were removed. Based on this consensus peak set, differential peaks were called using the DEseq2 implementation of HOMER. Specifically, reads form all individual experiments were counted in all individual peaks of consensus peak set using the ‘annotatePeaks.pl’ command with the option ‘-raw’. Based on the read count at each peak, significantly differential peaks were called using the ‘getDiffExpression.pl’ command using the ‘-norm2total’ option. Motifs were then called on differential peaks using the ‘findMotifsGenome.pl’ command using the ‘hypergeometric’ option for p-value calculations. As a background, the non-differential peak set was used. The final motif sets were filtered for motifs with a significance <1e−10 and with an occurrence of <0.5% in the background regions set in case the non-differential peaks were used as a background file for motif finding using the ‘compareMotifs.pl’ command. To create heatmaps of differential peaks, sequencing reads were annotated to the differential peaks in 25 bp bins in a region of −750 bp to +750 bp from the peak center. The data matrix was clustered using Cluster 3.0⁶⁸ with the k-means option, and the heatmap was visualized with the TreeView 3⁶⁹ software.

In-vitro T_reg differentiation with IL-4 and IL-33

Spleen and lymph-nodes were isolated and single-cell suspensions established. T_reg cells were pre-purified using CD25-based magnetic bead staining and column-based isolation, followed by FACS-based sorting of CD4⁺CD25⁺Foxp3-GFP⁺ T_reg cells. 20,000 cells each were supplemented with either recombinant mouse IL-4 (500 ng/mL) plus recombinant mouse IL-33 (500 ng/mL) (IL4 + IL33 T_reg group) or without additional cytokines (Crtl T_reg group). In addition, both T_reg groups were cultured with 5000 U/mL IL-2 and anti-CD3/28 microbeads at 4:1 bead to cell ratio were added to each well. Cells were incubated with the respective cytokine mix for 6 days at 37 °C, medium was re-supplemented on day 4. On day 6, cells were counted using a flow cytometer and expression of ILT3 and Gata-3 was measured. Cells were washed three times to remove remaining cytokines and used for the in vitro DC differentiation assay and for the in vitro T_eff polarization assay.

In vitro DC differentiation assay

Spleen and mesenteric LNs from donor animals were isolated and single-cell suspensions established. DCs were pre-purified using CD11c microbeads and column-based isolation, followed by FACS-based sorting of a MCHII⁺CD11c⁺ population. DCs were stimulated with LPS (100 ng/mL), and anti-CD3 (4 µg/mL) was added. In vitro expanded T_reg cells were added at indicated ratios (APC:T_reg 1:3, 1:1, 3:1, 1:0) and incubated for 16 h at 37 °C. Cells were fixed, and DCs were analyzed after antibody surface and intracellular staining with the BD Fix/Perm Buffer kit.

In vitro T_eff polarization assay

Spleen and mesenteric LNs from donor animals were isolated and single-cell suspensions established. T_eff cells were pre-purified using CD4 microbeads and column-based isolation, followed by FACS-based sorting of CD4⁺CD25⁻Foxp3-GFP⁻CD62L⁺ naive T_eff cells. 5 × 10⁴-purified T_eff cells were co-cultured with expanded T_reg cells at various ratios (T_eff:T_reg 1:0.5, 1:0.25, 1:0.125, 1:0.06, 1:0) and 1 × 10⁴ FACS-sorted DCs (MCHII⁺CD11c⁺). Anti-CD3 (4 µg/mL) and recombinant mouse IL-4 (20 ng/mL), as well as mouse IFNγ-blocking mAb (20 µg/mL) were added and cells were incubated for 96 h at 37 °C. Cells were fixed, followed by surface and intracellular staining with the Foxp3 Fix/Perm Buffer kit.

Measurement of Gata-3 induction upon cytokine challenge

Spleen and lymph-node-derived T_reg cells (CD4⁺CD8⁻CD25⁺Foxp3-YFP⁺) were isolated from WT and healthy Δ/Δ animals via FACS. 75,000 cells each were supplemented with IL-4 at different concentrations, along with 5000 U/mL IL-2, 20 µg/mL IL-12 blocking mAb, 20 µg/mL IFNγ-blocking antibody, and CD3/CD28 microbeads at 4:1 bead to cell ratio. Cells were incubated with the respective cytokine for 40 h at 37 °C, followed by surface and intracellular staining with the Foxp3 Fix/Perm Buffer kit.

Reporting summary

Further information on experimental design is available in the Nature Research Reporting Summary linked to this article.

Data availability

RNA microarray data, RNA sequencing data and ATAC-sequencing data that support the findings of the study have been deposited in the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi) under the accession number GSE119169. The source data underlying Figs. 1a–d, 2a–c, 3a, b, 3d–f, 4a, c, d, f–j, 5b, d–e, g–j, 6a–h, 7a–d, 8a–d, i, 9b–f, and Supplementary Figs. 1–4 and 8,a, b are provided as a Source Data file. All other data are available from the author upon reasonable request.

References

Wildin, R. S. et al. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nat. Genet. 27, 18–20 (2001).
Article CAS Google Scholar
Brunkow, M. E. et al. Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat. Genet. 27, 68–73 (2001).
Article CAS Google Scholar
Khattri, R., Cox, T., Yasayko, S. A. & Ramsdell, F. An essential role for Scurfin in CD4 + CD25+ T regulatory cells. Nat. Immunol. 4, 337–342 (2003).
Article CAS Google Scholar
Hori, S., Nomura, T. & Sakaguchi, S. Control of regulatory T cell development by the transcription factor Foxp3. Science 299, 1057–1061 (2003).
Article ADS CAS Google Scholar
Fontenot, J. D., Gavin, M. A. & Rudensky, A. Y. Foxp3 programs the development and function of CD4 + CD25 + regulatory T cells. Nat. Immunol. 4, 330–336 (2003).
Article CAS Google Scholar
Ohkura, N. et al. T cell receptor stimulation-induced epigenetic changes and Foxp3 expression are independent and complementary events required for Treg cell development. Immunity 37, 785–799 (2012).
Article CAS Google Scholar
Delacher, M. et al. Genome-wide DNA-methylation landscape defines specialization of regulatory T cells in tissues. Nat. Immunol. 18, 1160–1172 (2017).
Article CAS Google Scholar
Kitagawa, Y. et al. Guidance of regulatory T cell development by Satb1-dependent super-enhancer establishment. Nat. Immunol. 18, 173–183 (2017).
Article CAS Google Scholar
Schmidl, C. et al. Lineage-specific DNA methylation in T cells correlates with histone methylation and enhancer activity. Genome Res. 19, 1165–1174 (2009).
Article CAS Google Scholar
Feuerer, M. et al. Lean, but not obese, fat is enriched for a unique population of regulatory T cells that affect metabolic parameters. Nat. Med. 15, 930–939 (2009).
Article CAS Google Scholar
Campbell, D. J. & Koch, M. A. Phenotypical and functional specialization of FOXP3 + regulatory T cells. Nat. Rev. Immunol. 11, 119–130 (2011).
Article CAS Google Scholar
Koch, M. A. et al. The transcription factor T-bet controls regulatory T cell homeostasis and function during type 1 inflammation. Nat. Immunol. 10, 595–602 (2009).
Article CAS Google Scholar
Zheng, Y. et al. Regulatory T-cell suppressor program co-opts transcription factor IRF4 to control T(H)2 responses. Nature 458, 351–356 (2009).
Article ADS CAS Google Scholar
Chaudhry, A. et al. CD4 + regulatory T cells control TH17 responses in a Stat3-dependent manner. Science 326, 986–991 (2009).
Article ADS CAS Google Scholar
Radtke, F., Fasnacht, N. & Macdonald, H. R. Notch signaling in the immune system. Immunity 32, 14–27 (2010).
Article CAS Google Scholar
Tu, L. et al. Notch signaling is an important regulator of type 2 immunity. J. Exp. Med. 202, 1037–1042 (2005).
Article CAS Google Scholar
Amsen, D., Antov, A. & Flavell, R. A. The different faces of Notch in T-helper-cell differentiation. Nat. Rev. Immunol. 9, 116–124 (2009).
Article CAS Google Scholar
Amsen, D. et al. Direct regulation of Gata3 expression determines the T helper differentiation potential of Notch. Immunity 27, 89–99 (2007).
Article CAS Google Scholar
Fang, T. C. et al. Notch directly regulates Gata3 expression during T helper 2 cell differentiation. Immunity 27, 100–110 (2007).
Article CAS Google Scholar
Charbonnier, L. M., Wang, S., Georgiev, P., Sefik, E. & Chatila, T. A. Control of peripheral tolerance by regulatory T cell-intrinsic Notch signaling. Nat. Immunol. 16, 1162–1173 (2015).
Article CAS Google Scholar
Stavnezer, J. Immunoglobulin class switching. Curr. Opin. Immunol. 8, 199–205 (1996).
Article CAS Google Scholar
Breloer, M. & Abraham, D. Strongyloides infection in rodents: immune response and immune regulation. Parasitology 144, 295–315 (2017).
Article Google Scholar
Reitz, M. et al. Mucosal mast cells are indispensable for timely termination of Strongyloides ratti infection. Mucosal Immunology 10, 481–492 (2017).
Article Google Scholar
Blankenhaus, B. et al. Foxp3(+) regulatory T cells delay expulsion of intestinal nematodes by suppression of IL-9-driven mast cell activation in BALB/c but not in C57BL/6 mice. PLoS. Pathog. 10, e1003913 (2014).
Article Google Scholar
Polansky, J. K. et al. DNA methylation controls Foxp3 gene expression. Eur. J. Immunol. 38, 1654–1663 (2008).
Article CAS Google Scholar
Concannon, C. G. et al. AMP kinase-mediated activation of the BH3-only protein Bim couples energy depletion to stress-induced apoptosis. J. Cell. Biol. 189, 83–94 (2010).
Article CAS Google Scholar
Kilbride, S. M. et al. AMP-activated protein kinase mediates apoptosis in response to bioenergetic stress through activation of the pro-apoptotic Bcl-2 homology domain-3-only protein BMF. J. Biol. Chem. 285, 36199–36206 (2010).
Article CAS Google Scholar
Surh, C. D. & Sprent, J. Homeostasis of naive and memory T cells. Immunity 29, 848–862 (2008).
Article CAS Google Scholar
Soares, A. et al. Novel application of Ki67 to quantify antigen-specific in vitro lymphoproliferation. J. Immunol. Methods 362, 43–50 (2010).
Article CAS Google Scholar
Foxwell, B. M., Beadling, C., Guschin, D., Kerr, I. & Cantrell, D. Interleukin-7 can induce the activation of Jak 1, Jak 3 and STAT 5 proteins in murine T cells. Eur. J. Immunol. 25, 3041–3046 (1995).
Article CAS Google Scholar
Hsiao, H. W. et al. Deltex1 antagonizes HIF-1alpha and sustains the stability of regulatory T cells in vivo. Nat. Commun. 6, 6353 (2015).
Article CAS Google Scholar
Yamamoto, N. et al. Role of Deltex-1 as a transcriptional regulator downstream of the Notch receptor. J. Biol. Chem. 276, 45031–45040 (2001).
Article CAS Google Scholar
Miyazaki, M. et al. Id2 and Id3 maintain the regulatory T cell pool to suppress inflammatory disease. Nat. Immunol. 15, 767–776 (2014).
Article CAS Google Scholar
Roychoudhuri, R. et al. BACH2 represses effector programs to stabilize T(reg)-mediated immune homeostasis. Nature 498, 506–510 (2013).
Article ADS CAS Google Scholar
Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. & Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213–1218 (2013).
Article CAS Google Scholar
Ulges, A. et al. Protein kinase CK2 enables regulatory T cells to suppress excessive TH2 responses in vivo. Nat. Immunol. 16, 267–275 (2015).
Article CAS Google Scholar
Williams, J. W. et al. Transcription factor IRF4 drives dendritic cells to promote Th2 differentiation. Nat. Commun. 4, 2990 (2013).
Article Google Scholar
Gao, Y. et al. Control of T helper 2 responses by transcription factor IRF4-dependent dendritic cells. Immunity 39, 722–732 (2013).
Article CAS Google Scholar
Wang, H. et al. NOTCH1-RBPJ complexes drive target gene expression through dynamic interactions with superenhancers. Proc. Natl Acad. Sci. USA 111, 705–710 (2014).
Article ADS CAS Google Scholar
Furtado, G. C., Curotto de Lafaille, M. A., Kutchukhidze, N. & Lafaille, J. J. Interleukin 2 signaling is required for CD4(+) regulatory T cell function. J. Exp. Med. 196, 851–857 (2002).
Article CAS Google Scholar
Kim, E. H. et al. Bach2 regulates homeostasis of Foxp3 + regulatory T cells and protects against fatal lung disease in mice. J. Immunol. 192, 985–995 (2014).
Article CAS Google Scholar
Tsukumo, S. et al. Bach2 maintains T cells in a naive state by suppressing effector memory-related genes. Proc. Natl Acad. Sci. USA 110, 10735–10740 (2013).
Article ADS CAS Google Scholar
Kuwahara, M. et al. Bach2-Batf interactions control Th2-type immune response by regulating the IL-4 amplification loop. Nat. Commun. 7, 12596 (2016).
Article ADS CAS Google Scholar
Yu, F., Sharma, S., Edwards, J., Feigenbaum, L. & Zhu, J. Dynamic expression of transcription factors T-bet and GATA-3 by regulatory T cells maintains immunotolerance. Nat. Immunol. 16, 197–206 (2015).
Article CAS Google Scholar
Wang, Y., Su, M. A. & Wan, Y. Y. An essential role of the transcription factor GATA-3 for the function of regulatory T cells. Immunity 35, 337–348 (2011).
Article CAS Google Scholar
Wohlfert, E. A. et al. GATA3 controls Foxp3(+) regulatory T cell fate during inflammation in mice. J. Clin. Invest. 121, 4503–4515 (2011).
Article CAS Google Scholar
Taghon, T. et al. Enforced expression of GATA-3 severely reduces human thymic cellularity. J. Immunol. 167, 4468–4475 (2001).
Article CAS Google Scholar
Scripture-Adams, D. D. et al. GATA-3 dose-dependent checkpoints in early T cell commitment. J. Immunol. 193, 3470–3491 (2014).
Article CAS Google Scholar
Taghon, T., Yui, M. A. & Rothenberg, E. V. Mast cell lineage diversion of T lineage precursors by the essential T cell transcription factor GATA-3. Nat. Immunol. 8, 845–855 (2007).
Article CAS Google Scholar
Noval Rivas, M. et al. Regulatory T cell reprogramming toward a Th2-cell-like lineage impairs oral tolerance and promotes food allergy. Immunity 42, 512–523 (2015).
Article CAS Google Scholar
Kim, J. M., Rasmussen, J. P. & Rudensky, A. Y. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat. Immunol. 8, 191–197 (2007).
Article CAS Google Scholar
Rubtsov, Y. P. et al. Regulatory T cell-derived interleukin-10 limits inflammation at environmental interfaces. Immunity 28, 546–558 (2008).
Article CAS Google Scholar
Nakhai, H. et al. Conditional ablation of Notch signaling in pancreatic development. Development 135, 2757–2765 (2008).
Article CAS Google Scholar
Romeis. Mikroskopische Technik, Vol. 19 (Springer Spektrum, Berlin, Heidelberg, 2015).
Federico, G. et al. Tubular Dickkopf-3 promotes the development of renal atrophy and fibrosis. JCI Insight 1, e84916 (2016).
Article Google Scholar
Nouir, N. B. et al. Vaccination with Strongyloides ratti heat shock protein 60 increases susceptibility to challenge infection by induction of Th1 response. Vaccine 30, 862–871 (2012).
Article Google Scholar
Eschbach, M. L. et al. Strongyloides ratti infection induces transient nematode-specific Th2 response and reciprocal suppression of IFN-gamma production in mice. Parasite Immunol. 32, 370–383 (2010).
Article CAS Google Scholar
Heinz, S. et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589 (2010).
Article CAS Google Scholar
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Article CAS Google Scholar
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Article CAS Google Scholar
Quinlan, A. R. & Hall, I. M. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842 (2010).
Article CAS Google Scholar
Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol. 11, R106 (2010).
Article CAS Google Scholar
Anders, S. et al. Count-based differential expression analysis of RNA sequencing data using R and Bioconductor. Nat. Protoc. 8, 1765–1786 (2013).
Article Google Scholar
Rendeiro, A. F. et al. Chromatin accessibility maps of chronic lymphocytic leukaemia identify subtype-specific epigenome signatures and transcription regulatory networks. Nat. Commun. 7, 11938 (2016).
Article ADS CAS Google Scholar
Jiang, H., Lei, R., Ding, S. W. & Zhu, S. Skewer: a fast and accurate adapter trimmer for next-generation sequencing paired-end reads. BMC Bioinforma. 15, 182 (2014).
Article Google Scholar
Tarasov, A., Vilella, A. J., Cuppen, E., Nijman, I. J. & Prins, P. Sambamba: fast processing of NGS alignment formats. Bioinformatics 31, 2032–2034 (2015).
Article CAS Google Scholar
Consortium, E. P. An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74 (2012).
Article ADS Google Scholar
de Hoon, M. J., Imoto, S., Nolan, J. & Miyano, S. Open source clustering software. Bioinformatics 20, 1453–1454 (2004).
Article Google Scholar
Saldanha, A. J. Java Treeview—extensible visualization of microarray data. Bioinformatics 20, 3246–3248 (2004).
Article CAS Google Scholar

Download references

Acknowledgements

We thank Alexander Rudensky (Memorial Sloan-Kettering Cancer Center) for providing mice, Frank Lyko for help with amplicon sequencing, Claudia Schmitt, Ulrike Rothermel, Sabine Schmitt, Anna von Landenberg, Kristin Lobbes (all DKFZ), Marina Wuttke, Brigitte Ruhland, Veronika Hofmann, Kathrin Schambeck, Luise Eder, Rudolf Jung (all University Regensburg) and Marie-Luise Brunn (BNITM Hamburg) for excellent technical support. Furthermore, we thank the DKFZ core facilities for preclinical research, microscopy, flow cytometry and genomics & proteomics for outstanding support. This work was supported by grants from the Helmholtz Association of German Research Centers (HZ-NG-505) and the European Research Council (ERC-CoG, #648145 REGiREG) and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation- Projektnummer 324392634-TRR 221) to M.F., M.D. was supported by the German–Israeli Helmholtz Research School in Cancer Biology.

Author information

Authors and Affiliations

Chair for Immunology, University Regensburg and University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053, Regensburg, Germany
Michael Delacher, Sebastian Bittner & Markus Feuerer
Regensburg Center for Interventional Immunology (RCI), University Regensburg and University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053, Regensburg, Germany
Michael Delacher, Christian Schmidl, Sebastian Bittner, Michael Rehli & Markus Feuerer
Immune Tolerance Group, Tumor Immunology Program, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Michael Delacher, Danny Kägebein, Ulrike Träger, Ann-Cathrin Hofer & Markus Feuerer
Department of Immunology, Weizmann Institute of Science, 234 Herzl Street, 76100, Rehovot, Israel
Yonatan Herzig & Jakub Abramson
Bernhard Nocht Institute for Tropical Medicine, Bernhard-Nocht-Straße 74, 20359, Hamburg, Germany
Minka Breloer & Wiebke Hartmann
Division of Developmental Immunology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Fabian Brunk
Division of Epigenomics and Cancer Risk Factors, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Dieter Weichenhan
Division of Applied Bioinformatics, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Charles D. Imbusch
Genomics and Proteomics Core Facility, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Agnes Hotz-Wagenblatt
Division of Biostatistics, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Thomas Hielscher
Division of Epigenetics, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Achim Breiling
Division of Cellular and Molecular Pathology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 280, 69120, Heidelberg, Germany
Giuseppina Federico & Hermann-Josef Gröne
Department of Internal Medicine, Technical University of Munich, Ismaninger Straße 22, 81675, Munich, Germany
Roland M. Schmid
Department of Internal Medicine III, Hematology and Oncology, University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053, Regensburg, Germany
Michael Rehli

Authors

Michael Delacher
View author publications
You can also search for this author in PubMed Google Scholar
Christian Schmidl
View author publications
You can also search for this author in PubMed Google Scholar
Yonatan Herzig
View author publications
You can also search for this author in PubMed Google Scholar
Minka Breloer
View author publications
You can also search for this author in PubMed Google Scholar
Wiebke Hartmann
View author publications
You can also search for this author in PubMed Google Scholar
Fabian Brunk
View author publications
You can also search for this author in PubMed Google Scholar
Danny Kägebein
View author publications
You can also search for this author in PubMed Google Scholar
Ulrike Träger
View author publications
You can also search for this author in PubMed Google Scholar
Ann-Cathrin Hofer
View author publications
You can also search for this author in PubMed Google Scholar
Sebastian Bittner
View author publications
You can also search for this author in PubMed Google Scholar
Dieter Weichenhan
View author publications
You can also search for this author in PubMed Google Scholar
Charles D. Imbusch
View author publications
You can also search for this author in PubMed Google Scholar
Agnes Hotz-Wagenblatt
View author publications
You can also search for this author in PubMed Google Scholar
Thomas Hielscher
View author publications
You can also search for this author in PubMed Google Scholar
Achim Breiling
View author publications
You can also search for this author in PubMed Google Scholar
Giuseppina Federico
View author publications
You can also search for this author in PubMed Google Scholar
Hermann-Josef Gröne
View author publications
You can also search for this author in PubMed Google Scholar
Roland M. Schmid
View author publications
You can also search for this author in PubMed Google Scholar
Michael Rehli
View author publications
You can also search for this author in PubMed Google Scholar
Jakub Abramson
View author publications
You can also search for this author in PubMed Google Scholar
Markus Feuerer
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.D., M.B., J.A. and M.F. designed experiments; M.D., Y.H., W.H., F.B., D.K., U.T., A.-C.H., S.B., D.W., A.B. and G.F. performed the experiments; R.M.S. provided material; M.D., C.S., M.B., W.H., C.I., A. H.-W., T.H., H.J.G., M.R., J.A. and M.F. analyzed data; M.D. and M.F. wrote the manuscript.

Corresponding author

Correspondence to Markus Feuerer.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Journal peer review information: Nature Communications thanks Derk Amsen and the other anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Peer Review File

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Delacher, M., Schmidl, C., Herzig, Y. et al. Rbpj expression in regulatory T cells is critical for restraining T_H2 responses. Nat Commun 10, 1621 (2019). https://doi.org/10.1038/s41467-019-09276-w

Download citation

Received: 13 April 2018
Accepted: 26 February 2019
Published: 08 April 2019
DOI: https://doi.org/10.1038/s41467-019-09276-w

This article is cited by

Breaking tolerance: the autoimmune aspect of atherosclerosis
- Amir Khan
- Payel Roy
- Klaus Ley
Nature Reviews Immunology (2024)
Tissue-specific enhancer–gene maps from multimodal single-cell data identify causal disease alleles
- Saori Sakaue
- Kathryn Weinand
- Soumya Raychaudhuri
Nature Genetics (2024)
Recirculating IL-1R2+ Tregs fine-tune intrathymic Treg development under inflammatory conditions
- Eirini Nikolouli
- Yassin Elfaki
- Jochen Huehn
Cellular & Molecular Immunology (2021)
Elucidating different pattern of immunoregulation in BALB/c and C57BL/6 mice and their F1 progeny
- Wiebke Hartmann
- Birte Blankenhaus
- Minka Breloer
Scientific Reports (2021)
Preceding Viral Infections Do Not Imprint Long-Term Changes in Regulatory T Cell Function
- Felix Rost
- Katharina Lambert
- Nicole Joller
Scientific Reports (2020)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.