Identification of a genomic enhancer that enforces proper apoptosis induction in thymic negative selection

Hojo, Miki Arai; Masuda, Kyoko; Hojo, Hiroaki; Nagahata, Yosuke; Yasuda, Keiko; Ohara, Daiya; Takeuchi, Yusuke; Hirota, Keiji; Suzuki, Yutaka; Kawamoto, Hiroshi; Kawaoka, Shinpei

doi:10.1038/s41467-019-10525-1

Download PDF

Article
Open access
Published: 13 June 2019

Identification of a genomic enhancer that enforces proper apoptosis induction in thymic negative selection

Miki Arai Hojo^1,2,
Kyoko Masuda³,
Hiroaki Hojo^2,3,4,
Yosuke Nagahata³,
Keiko Yasuda³,
Daiya Ohara³,
Yusuke Takeuchi³,
Keiji Hirota³,
Yutaka Suzuki¹,
Hiroshi Kawamoto³ &
…
Shinpei Kawaoka ORCID: orcid.org/0000-0003-4064-6781^2,3,4

Nature Communications volume 10, Article number: 2603 (2019) Cite this article

5338 Accesses
9 Citations
3 Altmetric
Metrics details

Subjects

Abstract

During thymic negative selection, autoreactive thymocytes carrying T cell receptor (TCR) with overtly strong affinity to self-MHC/self-peptide are removed by Bim-dependent apoptosis, but how Bim is specifically regulated to link TCR activation and apoptosis induction is unclear. Here we identify a murine T cell-specific genomic enhancer E^{BAB (Bub1-Acoxl-Bim)}, whose deletion leads to accumulation of thymocytes expressing high affinity TCRs. Consistently, E^BAB knockout mice have defective negative selection and fail to delete autoreactive thymocytes in various settings, with this defect accompanied by reduced Bim expression and apoptosis induction. By contrast, E^BAB is dispensable for maintaining peripheral T cell homeostasis via Bim-dependent pathways. Our data thus implicate E^BAB as an important, developmental stage-specific regulator of Bim expression and apoptosis induction to enforce thymic negative selection and suppress autoimmunity. Our study unravels a part of genomic enhancer codes that underlie complex and context-dependent gene regulation in TCR signaling.

Thymic-specific regulation of TCR signaling by Tespa1

Article 17 July 2019

A double-negative thymocyte-specific enhancer augments Notch1 signaling to direct early T cell progenitor expansion, lineage restriction and β-selection

Article 31 October 2022

How autoreactive thymocytes differentiate into regulatory versus effector CD4+ T cells after avoiding clonal deletion

Article Open access 23 March 2023

Introduction

T cell population in the thymus is highly heterogeneous, harboring a diverse T cell receptor (TCR) repertoire^1,2,3,4. The massive diversity in TCR sequences is on one hand useful, as it puts the immune system on stand-by for numerous foreign antigens such as pathogens, while on the other hand it is risky because it could generate T cells harboring TCRs that strongly recognize self, potentially causing autoimmunity. To suppress autoimmunity, organisms have evolved a sophisticated mechanism called negative selection, establishing central T cell tolerance. In negative selection, interaction between TCR and self-peptide presented on major histocompatibility complexes (self-pMHCs) is converted into apoptotic output: high-affinity TCR clones are considered as autoreactive and die by apoptosis^1,2,3,4.

A part of high-affinity TCR clones, instead of being deleted, are diverted into regulatory T (T_reg) cells that are potent to suppress autoreactive T cells in periphery^1,5. Suppression by T_reg cells is one of the peripheral tolerance mechanisms for organisms to deal with autoreactive T cells that have evaded negative selection. Other mechanisms are induction of T cell anergy and peripheral deletion by apoptosis¹.

Pro-apoptotic Bim promotes the mitochondrial apoptosis cascade, contributing to numerous biological pathways^{6,7,8,9,10,11,12,13,14,15,16,17,18}. In central T cell tolerance, Bim is considered as a downstream target of TCR signal: TCR signal activates Bim expression, and Bim knockout (KO) mice show defective negative selection⁶. However, little is known about how TCR signal strength is linked to Bim expression^2,17.

Bim is genetically required not only for establishing central T cell tolerance^6,7,8, but also for depleting activated T cells in periphery^11,12, B cell homeostasis, embryonic development, and so on¹⁸. Therefore, Bim should be able to distinguish multiple biological pathways in different cell types, depending on signals that cells receive. The molecular mechanism underlying how Bim is regulated to work at an appropriate place and time remains elusive.

Enhancers are genomic elements that regulate gene expression in a signal and cell type dependent manner^19,20. Although epigenome analyses have enabled systematic identification and characterization of enhancers, it is still difficult to directly study their physiological roles in vivo for the following reasons. First, enhancers are located often several hundreds of kilobases to even megabases away from their target genes, making it difficult to confidently predict a target(s) of an enhancer. Second, some genes may have multiple functionally redundant enhancers. Third, making enhancer KOs through genetic ablation has been labor-intensive and time-consuming, especially in mice. Recent progress in CRISPR–Cas9 technology²¹ has reduced the cost and time needed for generating enhancer KO mice, and most importantly, has enabled us to produce large genomic deletions without leaving unwanted footprints of exogenous DNAs. CRISPR–Cas9 technology is indeed beginning to uncover physiological functions of novel enhancers in vivo^22,23,24.

Here, we utilize enhancer genetics to understand how Bim is specifically regulated to induce apoptosis during thymic negative selection, and find a cis-regulatory enhancer specifically contributing to this process. With the aid of epigenome analyses, we identify a cis-regulatory enhancer E^{BAB (Bub1-Acoxl-Bim)} that is specific to thymocytes and splenic T cells. We generate E^BAB KO mice by CRISPR–Cas9 technology and find that a high-affinity TCR repertoire accumulates in the E^BAB KO thymus. E^BAB KO thymocytes are defective in apoptosis due to incomplete activation of Bim. By contrast, Bim-mediated homeostasis of T_reg cells and peripheral T cells is not affected by E^BAB KO, thereby implicating a specific function of E^BAB in thymic negative selection. This study is an example of utilizing enhancer KO approach to dissect regulation of enhancer activity and subsequent gene function in vivo to address biological questions.

Results

Identification of a murine T cell-specific enhancer E ^BAB

Our analyses on various publicly available ChIP-seq (chromatin-immunoprecipitation and following sequencing) data on multiple mouse tissues identified a T cell-specific enhancer-like region (H3K27ac high and H3K4me3 low)¹⁹ in the mouse Bub1-Acoxl-Bim locus (Fig. 1a and Supplementary Fig. 1). This region was located at approximately 200-kb upstream of Bim, within the ninth intron of Acoxl gene (unexpressed in T cells), and at approximately 90-kb upstream of Bub1 (a mitotic checkpoint factor), and thus was named E^BAB. E^BAB was approximately 8-kb in length and contained two prominent H3K27ac peaks E1 and E2 (Fig. 1a). Both E1 and E2 were highly specific to the thymus (Fig. 1a) and well conserved between human and mice (Fig. 1a and Supplementary Fig. 1b). H3K27ac peaks corresponding to E^BAB were identified also in the spleen to a lesser extent (Fig. 1a). The signals in the spleen were likely derived from splenic peripheral T cells because naïve peripheral T cells, but not CD19⁺ B cells, retained DNase hypersensitivity sites in the locus (Fig. 1b), the observation further supported by other publicly available ChIP-seq data sets (Supplementary Fig. 1c).

To investigate a physiological role of E^BAB in vivo, we generated E^BAB KO mice by using the CRISPR–Cas9 system (Fig. 1c and Supplementary Data 1). Three founder lines (line#44, #47 and #50) were successfully obtained, each harboring a distinct pattern of deletion (Fig. 1d and Supplementary Fig. 2). Because offspring from these different founders showed no phenotypic differences (Supplementary Fig. 3), we refer to these three KO alleles simply as ΔE^BAB in this paper.

To ask if E^BAB KO affects expression of Bim and Bub1, we performed quantitative polymerase chain reaction (qPCR) experiments on several organs and cell types including thymocytes and splenocytes (Fig. 1e, f). Bim and Bub1 expression were slightly decreased in thymocytes and splenocytes in ΔE^BAB mice (7–17 weeks old), while unaltered in the lung, liver, kidney, and pancreas (Fig. 1e, f). Thus, E^BAB is an enhancer specific to thymocytes and splenic T cells (Fig. 1a, b and Supplementary Fig. 1), deletion of which only slightly affected its proximal genes Bim and Bub1 in total thymocytes and splenocytes (Fig. 1e, f).

E ^BAB KO accumulates high-affinity TCR clones in the thymus

Next, we asked whether ΔE^BAB impairs T cell homeostasis in the thymus. For this purpose, thymocytes were analyzed using flow cytometry with anti-CD4 and anti-CD8 antibodies (Fig. 2a–d). While the extent of reduction in Bim and Bub1 expression was very modest in ΔE^BAB thymocytes (Fig. 1e, f), severe abnormalities in thymocyte population were observed both in young (7–17 weeks old) and aged (30–36 weeks old) mice (Fig. 2a, b): ΔE^BAB increased proportion of double negative (DN), CD4 single positive (CD4 SP), and CD8 SP thymocytes, whereas proportion of double positive (DP) had decreased (Fig. 2a, b). These were attributed to the increased number of DN, CD4 SP, and CD8 SP thymocytes (Fig. 2c, d). No obvious phenotypic differences were observed between WT and heterozygotes, between ΔE^BAB males and ΔE^BAB females, and among the three founder lines (Supplementary Fig. 3).

To understand the nature of altered T cell homeostasis in the ΔE^BAB thymus, we performed whole transcriptome analysis of ΔE^BAB thymocytes (two littermate pairs) utilizing RNA-seq (Fig. 2e). As shown in Fig. 2e, only a small number of genes showed statistically significant (P < 0.05) more than twofold changes (see methods and Supplementary Data 2 for the detail). Yet, differentially expressed genes (DEGs) affected by ΔE^BAB indicated that the ΔE^BAB thymus accumulated high affinity TCR clones (Fig. 2e): the ΔE^BAB thymus exhibited a higher level of Nr4a1 expression when compared to the WT thymus. Nr4a1 is a faithful responder for TCR signal, and the expression level of Nr4a1 positively correlates with TCR signal strength^15,25,26. Elevated expression of Nr4a1 is thus one of the hallmarks of high affinity TCR clones in the thymus. Arhgap20 and Tnfrsf9, the top two most elevated genes in RNA-seq data in the ΔE^BAB thymus (Fig. 2e), are also known as being upregulated in TCR-activated T cells²⁶, supporting the notion that high-affinity TCR clones accumulate in ΔE^BAB mice.

To obtain more evidence that the ΔE^BAB thymus piles up high-affinity TCR clones, we stained thymocytes with anti-TCRβ and anti-CD69 antibodies (Fig. 2f, g). We found that proportion of post-selection (TCRβ^highCD69^high) thymocytes was increased in the ΔE^BAB thymus both in young and aged mice (Fig. 2f, g). Moreover, as determined by cell-sorting followed by qPCR, Nr4a1 expression in post-selection thymocytes was much higher in ΔE^BAB than in WT (Fig. 2h). Thus, it was likely that high affinity TCR clones were accumulated in the ΔE^BAB thymus.

We also analyzed TCR⁺ population in the DN subset: thymic precursors of TCRαβ⁺CD8αα⁺ intestinal intraepithelial lymphocytes (IELs)^27,28 and TCR Vα14⁺ invariant natural killer T cells (iNKT cells)^29,30. Our data demonstrated that ΔE^BAB increased the number of IEL precursors (Supplementary Fig. 4a, b), while iNKT population was much less affected by ΔE^BAB (Supplementary Fig. 4c–e). These changes may contribute to the increased number of DN thymocytes in ΔE^BAB mice (Fig. 2c, d).

E ^BAB is required for eliminating high-affinity TCR clones

Accumulation of high-affinity TCR clones in the ΔE^BAB thymus led us to hypothesize that E^BAB is required for negative selection and/or more general apoptosis processes. To test this hypothesis, we treated primary thymocytes with four apoptotic stimuli: dexamethasone (DEX)^7,8, phorbol 12-myristate 13-acetate (PMA)^7,8, ionomycin^7,8 (Supplementary Fig. 5), and anti-CD3 and anti-CD28 antibodies^6,25 (Fig. 3a). ΔE^BAB did not affect survival of DEX- or PMA-treated thymocytes (Supplementary Fig. 5a, b) but partially rescued thymocytes from cell death caused by ionomycin or anti-CD3 and anti-CD28 antibodies (Fig. 3a and Supplementary Fig. 5c–e). The rescue in ionomycin experiments was observed in a dose-dependent manner (Supplementary Fig. 5c–e). PMA, ionomycin, and anti-CD3 and anti-CD28 antibodies are often used for artificially activating TCR stimulation and thus for mimicking negative selection ex vivo^6,7,8,25. Thus, these results suggest that ΔE^BAB thymocytes were defective in apoptosis following TCR signal activation, leading to a hypothesis that E^BAB plays a role in depleting high-affinity TCR clones.

To assess the possibility that E^BAB is required for deleting high affinity TCR clones in vivo, we took advantage of three TCR transgenic mouse models: HY transgenic (tg)⁶ (Fig. 3b–f), OT-II tg^6,31,32 (Fig. 4), and OT-I tg^33,34,35 (Supplementary Fig. 6).

HY tg mice express αβTCRs that recognize a male-specific endogenous antigen called HY, which is presented by H-2D^b class I MHC molecules⁶ (Fig. 3b-f). ΔE^BAB; HY tg mice were analyzed at 6–8 weeks old as described previously⁶ by flow cytometry. As shown by the lower abundance of DP and CD8 SP in E^BAB+/−; HY tg male mice compared to female mice (Fig. 3b–f), thymocytes expressing HY-TCR are negatively selected in a male-specific manner. Notably, ΔE^BAB significantly rescued DP (both in proportion and number) and CD8 SP (in proportion) thymocytes in HY tg male mice (Fig. 3b–f).

OT-II tg thymocytes express Vα2/Vβ5 TCRs that primarily recognize chicken ovalbumin-derived peptide (ISQAVHAAHAEINEAGR, OVA_323–339 peptide) presented by I^-A^b class II MHC molecules^6,31,32. ΔE^BAB; OT-II tg mice were generated by crossing and injected with OVA_323–339 peptide or the control peptide intraperitoneally. WT; OT-II tg or E^BAB+/−; OT-II tg mice were also used as a control. In response to injection with OVA_323–339 peptide, the CD4 SP proportion of control thymocytes was massively reduced (Fig. 4a, b). Strikingly, E^BAB KO rescued this reduction (Fig. 4a, b). Staining thymocytes with anti-TCRβ and anti-CD69 antibodies revealed that ΔE^BAB almost completely prevented the deletion of post-selection thymocytes caused by OVA_323–339 peptide injection (Fig. 4c, d).

Fetal thymic organ culture (FTOC) with the OT-I tg system allows us to investigate effects of antigens of interests on intrathymic T cell development^33,34,35,36. OT-I tg thymocytes express Vα2/Vβ5 TCRs that bind chicken ovalbumin peptide residues 257–264 (SIINFEKL, OVA_257–264 peptide) in the context of H-2K^b class I MHC molecules³⁴. Of note, a series of OVA_257–264 variants with different TCR affinity can be used in the OT-I tg FTOC system^34,35. We cultured fetal thymus (FT) of E^BAB+/−; OT-I tg and ΔE^BAB; OT-I tg in the presence of OVA_257–264, Q4R7 (SIIQFERL), or gp33 (KAVYNFATC)³⁴ (Supplementary Fig. 6). The previous publications establish OVA_257–264 and Q4R7 as negative selectors (affinity to OT-I TCR: OVA_257–264 > Q4R7)^34,35. gp33, which does not bind to OT-I TCRs, was used as a control. ΔE^BAB compared to E^BAB+/− significantly rescued DP thymocytes from Q4R7-dependent selection but not those from OVA_257–264-dependent selection (Supplementary Fig. 6). Thus, ΔE^BAB rescued OT-I TCR⁺ thymocytes from negative selection in a TCR affinity-dependent manner.

Taken the results from the three transgenic models together, we concluded that E^BAB plays an important role in apoptosis of high-affinity TCR clones in thymic negative selection ex vivo and in vivo.

E ^BAB is essential for TCR-dependent activation of Bim

The above-described phenotypes observed in the ΔE^BAB thymus, four ex vivo culture systems, and three transgenic models are strikingly similar to those observed in Bim KO mice as described in multiple publications^6,7,8,9,10 and as validated by ourselves (Figs. 2–4 and Supplementary Figs. 4–7). For example, the extent of increase for post-selection thymocytes caused by ΔE^BAB was similar to that by Bim KO (Fig. 2g. versus Supplementary Fig. 7e). Such similarity was also observed for the ex vivo TCR stimulation (Fig. 3a), OT-II (Fig. 4), and OT-I (Supplementary Fig. 6) experiments, whereas Bim KO seemingly more efficiently rescued HY-TCR⁺ thymocytes than E^BAB KO did in the HY experiments (Fig. 3: see discussion regarding interpretation on this data). Moreover, ΔE^BAB did not affect DEX-induced cell death, which was in fact Bim-independent (Supplementary Fig. 5a). Some of the thymocytes abnormalities are seen also in mice ectopically expressing Bim-antagonist Bcl2 (Bcl2 tg) in a T cell-specific manner, which leads to defective apoptosis in the thymus^25,37,38. Although ΔE^BAB had a minor impact on Bim expression in total thymocytes (Fig. 1e), we next examined whether the deficient negative selection of ΔE^BAB thymocytes is due to abnormal regulation of Bim. To this end, we again artificially activated TCR by treating thymocytes with anti-CD3 and anti-CD28 antibodies ex vivo, and then measured expression of Bim, Nr4a1, Bub1, and Bcl2 by qPCR (Fig. 5a–d). Bim expression in ΔE^BAB thymocytes treated with anti-CD3 and anti-CD28 antibodies was clearly lower than that in TCR-stimulated WT thymocytes (Fig. 5a). Expression of Nr4a1, a faithful responder for TCR signal, was comparable between WT and ΔE^BAB (Fig. 5b), suggesting that TCR signal was properly activated in ΔE^BAB thymocytes. Consistent with the report that a dominant negative form of Bub1 is insufficient to cause T cell abnormalities³⁹, Bub1 expression was not influenced by TCR stimulation and E^BAB KO (Fig. 5c). Bcl2, a TCR-downstream anti-apoptotic gene, exhibited an expected response to TCR stimulation⁴⁰, and this response was comparable between WT and ΔE^BAB (Fig. 5d). Thus, it was likely that E^BAB primarily targets Bim following TCR signal.

To assess the role of E^BAB in TCR-dependent Bim activation in vivo, we sorted pre- and post-selection thymocytes and measured expression of Bim and Bub1 (Fig. 5e–h). If E^BAB regulates Bim in a TCR signal dependent manner, Bim expression should be affected by ΔE^BAB at the post-selection stage while less affected at the pre-selection stage. As expected, expression of Bim in the post-selection (TCRβ^highCD69^high) thymocytes was markedly lower in ΔE^BAB than in WT (ΔE^BAB/WT = 0.31) (Fig. 5e, f). Although Bim expression was moderately affected by E^BAB KO in the pre-selection (TCRβ^lowCD69^low) stage (ΔE^BAB/WT = 0.64) (Fig. 5e, f), expression ratio of ΔE^BAB to WT in the post-selection stage was significantly lower than that in the pre-selection stage (Fig. 5f). These results were validated at the protein level, demonstrating that protein expression of BimEL and BimL, two of major isoforms of Bim⁴¹, was compromised by ΔE^BAB (Supplementary Fig. 9). Together, ΔE^BAB affected Bim expression in a post-selection-biased manner, indicating an important role of E^BAB in activating Bim upon TCR activation in vivo. Bub1 expression was not significantly altered between ΔE^BAB and WT mice both in pre- and post-selection thymocytes (Fig. 5g, h), again excluding Bub1 in explaining the T cell phenotypes we observed (Figs. 2–4). Collectively, our data demonstrated that E^BAB KO disrupts transcriptional activation for Bim upon TCR stimulation, resulting in the rescue of high affinity TCR clones in the thymus (Figs. 2–5). We reasoned that Bim expression in total thymocytes looked only mildly affected by E^BAB KO (Fig. 1e) as only 10–30% of thymocytes is at the post-selection stage where transcriptional regulation of Bim is strongly affected by ΔE^BAB.

Furthermore, we investigated whether E^BAB regulates Bim in cis or trans. To test this, we generated E^BAB+/−; Bim^−/+ mice (Supplementary Fig. 10a). If E^BAB controls Bim in cis, a phenotype of E^BAB+/−; Bim^−/+ thymocytes should be similar to ΔE^BAB (Supplementary Fig. 10a). If E^BAB can regulate Bim in trans, E^BAB+/−; Bim^−/+ mice should show no phenotype in the thymus (Supplementary Fig. 10a). Staining with anti-CD4, anti-CD8, anti-TCRβ, and anti-CD69 antibodies revealed that E^BAB+/⁻; Bim⁻^/+ thymocytes exhibited ΔE^BAB-like phenotypes (Supplementary Fig. 10b–f). Given that Bim^+/− exhibits almost no phenotype in DN, DP, CD4 SP, CD8 SP, and TCRβ^highCD69^high thymocytes (Supplementary Fig. 10g), these results validated that E^BAB is a cis-regulatory element regulating Bim.

E ^BAB is dispensable for T_reg cells and peripheral T cells

Expression levels of Nr4a transcription factors including Nr4a1 are markers not only for high affinity TCR clones, but also for thymic T_reg cells that suppress autoreactive T cells in periphery^1,5. Nr4a transcription factors cooperatively promote T_reg differentiation through directly activating expression of Foxp3, the master regulator for T_reg identity^42,43,44. Thus, it is thought that high-affinity TCR clones are a precursor of T_reg cells⁵. Bim KO and Bcl2 tg mice accumulate both of high affinity TCR clones and T_reg cells^{6,7,8,14,15,25,37,38} (Supplementary Figs. 7d, e, 8c). To examine whether accumulation of high-affinity TCR clones resulted in the increased number of T_reg cells in ΔE^BAB mice, we analyzed thymic and splenic T_reg cells by staining with anti-Foxp3 antibody (Fig. 6a, b). We found that the number of thymic and splenic T_reg cells were comparable between WT and ΔE^BAB (Fig. 6a, b). Thus, although high-affinity TCR clones are accumulated in the ΔE^BAB thymus (Fig. 2), E^BAB does not affect homeostasis of T_reg cells. These results indicate that Bim-dependent homeostasis of T_reg cells in the thymus and spleen is independent on E^BAB.

Bim KO and Bcl2 tg mice also accumulate peripheral T cells^{6,7,8,12,13,25,37,38} (Supplementary Fig. 8a, b). However, the number of splenic CD4⁺ and CD8⁺ T cell were not affected by ΔE^BAB both in young and aged mice (Fig. 6c, d). In addition, splenic B cells, whose homeostasis depends on Bim⁸, were unaffected by E^BAB KO (Supplementary Fig. 11). These results indicated that homeostasis of peripheral T cells is properly maintained in ΔE^BAB mice in contrast to Bim KO mice^6,7,8,12,13 (Supplementary Fig. 8a, b).

To gain an additional insight into a role of E^BAB in peripheral T cell homeostasis, we performed Interleukin-2 (IL-2) deprivation experiments and activation-induced cell death (AICD) experiments in cultured activated CD4⁺ and CD8⁺ splenocytes (Fig. 7).

Bim-deficient peripheral T cells are resistant to the absence of IL-2 ex vivo^7,45 while WT and ΔE^BAB splenic T cells needed IL-2 for their continuous survival in culture (Fig. 7a, b). In accordance with this, E^BAB was dispensable for upregulation of Bim following IL-2 withdrawal (Fig. 7c, d). Expression of Bub1 was merely affected by IL-2 (Fig. 7e, f). Hence, IL-2 deprivation-dependent T cell death in periphery requires Bim but not E^BAB.

AICD is known as a peripheral cell death cascade where repeatedly activated T cells undergo apoptosis⁴⁶. In fact, reactivating peripheral T cells by anti-CD3 antibody treatment strongly induced cell death (Fig. 7g, h). Using this system, we found that neither Bim nor E^BAB was required for AICD ex vivo (Fig. 7g, h).

From these experiments, we concluded that at least two peripheral apoptotic pathways are intact in ΔE^BAB mice, which can be one explanation for why the increase in thymic high-affinity TCR clones did not lead to accumulation of peripheral T cells.

We additionally analyzed splenic activated T cell proportion by staining splenocytes with anti-CD44 and anti-CD62L antibodies to investigate autoimmune pathology (Fig. 8a–c). In line with the fact that two peripheral apoptosis pathways are intact in ΔE^BAB mice (Fig. 7), activated T cell (CD44^highCD62L^low) proportion was not increased in the ΔE^BAB spleen even in aged mice (Fig. 8a–c and Supplementary Fig. 12a–e). These data were in line with RNA-seq data from WT and ΔE^BAB spleen (two littermate pairs), where only two candidate DEGs were identified (Supplementary Fig. 12f and Supplementary Data 3). In contrast, Bim-deficient mice accumulated activated CD8⁺ T cells in the spleen already at younger age (Supplementary Fig. 8d–f).

Furthermore, we assessed a role of E^BAB in an in vivo autoimmune disease model, experimental autoimmune encephalomyelitis (EAE)⁴⁷. As shown in Fig. 8d, clinical scores of ΔE^BAB were comparable to those of WT: ΔE^BAB mice neither exhibited severer nor ameliorated EAE phenotypes, suggesting that E^BAB is dispensable for EAE. In contrast, we confirmed previous publication that Bim plays a role in EAE (Supplementary Fig. 13)¹⁶. Consistently, we did not find any histological sign for autoimmunity such as massive infiltration of leukocytes (i.e., inflammation) in several nonlymphoid organs in ΔE^BAB mice (Fig. 8e).

Thus, ΔE^BAB mice did not show any autoimmune phenotype, which was consistent with that E^BAB was not essential for peripheral T cell homeostasis (Figs. 6–8, Supplementary Fig. 8, and Supplementary Figs. 12 and 13). These results highlighted an extraordinary specialized role of E^BAB in thymic negative selection (Fig. 9).

Discussion

It is thought that TCR signal activates Bim to promote apoptosis in thymic negative selection process^1,2,3. However, it has been unclear how TCR signal strength is linked to the Bim expression level. Even whether activation of Bim expression is critical for negative selection has been unanswered due to lack of a mutant specifically deficient for TCR-dependent Bim expression¹⁷. Here, by means of CRISPR–Cas9-mediated KO approach in mice, we generated a mutant that lacks a cis-regulatory enhancer for Bim, E^BAB (Figs. 1–5 and Supplementary Fig. 10). E^BAB is T cell-specific and evolutionarily conserved (Fig. 1a, b and Supplementary Fig. 1), and ΔE^BAB mice were unable to efficiently eliminate high affinity TCR clones (Figs. 2–4 and Supplementary Fig. 6). This defect appeared to be attributed to insufficient activation of Bim during TCR activation (Fig. 5 and Supplementary Fig. 9). Thus, we propose that E^BAB links TCR signal and Bim expression, contributing to depletion of high affinity TCR clones in the thymus and thus to central T cell tolerance (Fig. 9).

Bim KO affects homeostasis of nearly all of T cell types^{6,7,8,9,10,11,12,13,14,15,16,17,18} (Supplementary Fig. 4 and Supplementary Figs. 7 and 8). Comparing T cell phenotypes between Bim KO and ΔE^BAB mice provides insights into the extent of contribution of E^BAB-dependent Bim regulation on phenotypes of interests. Among the phenotypes observed in the ΔE^BAB thymus, accumulation of post-selection thymocytes in the ΔE^BAB thymus was as severe as that in the Bim KO thymus (Fig. 2f, g and Supplementary Fig. 7d, e). Moreover, both of ΔE^BAB and Bim KO rescued thymocytes from negative selection in the ex vivo TCR stimulation model and three transgenic systems (Figs. 3 and 4 and Supplementary Fig. 6). The degrees of rescues by E^BAB KO and Bim KO were similar in the ex vivo TCR stimulation, OT-I, and OT-II experiments (Figs. 3 and 4 and Supplementary Fig. 6). In contrast, in the HY experiments, it appeared that Bim KO better rescued HY-TCR⁺ thymocytes than E^BAB KO. When interpreting these results, it should be noted that readouts used (e.g., proportion of cell types of interests) do not always specifically reflect negative selection. In particular, Bim KO is able to rescue apoptosis induced by various stimuli¹⁸, even death of control-treated cells (Supplementary Fig. 5). Thus, increased survival seen in Bim KO potentially reflects both negative selection-dependent and -independent events such as post-selection lifetime. Importantly, our analyses showed that function of E^BAB appears to be specific for TCR signal-dependent Bim expression in the thymus. Hence, we assume that the more efficient rescue observed in Bim KO in the HY experiments may be attributed to pleiotropic effects of Bim KO on negative selection-independent phenomena in the thymus. Alternatively, the observations in the HY experiments imply that there may be another enhancer controlling TCR-dependent expression of Bim. Collectively, we suggest that a role for Bim in eliminating high affinity TCR clones in the thymus requires E^BAB. These results establish E^BAB-mediated transcriptional regulation for Bim as a contributor for Bim-dependent thymic negative selection.

Bim KO and Bcl2 tg mice accumulate T_reg cells, a basis for the current model that T_reg cells are differentiated from high-affinity TCR clones in the thymus^1,5,15,25 (Supplementary 8c). Intriguingly, in the ΔE^BAB thymus, the number of T_reg cells was comparable with that in the littermate controls even though high affinity TCR clones were increased (Fig. 2f, g and Fig. 6a, b). This indicates that the residual amount of Bim in the ΔE^BAB thymus is sufficient for maintaining T_reg homeostasis. It is also plausible that a non-cell autonomous mechanism may be dominant for T_reg homeostasis in the thymus. Supporting this, a previous study shows that Foxp3 overexpression induces Bim phosphorylation and enhances apoptosis in newly arising thymic T_reg cells⁴⁸. In addition, T_reg homeostasis is controlled by availability of pro-survival cytokines such as IL-2 derived from the surrounding environment^48,49. These altogether indicate that T_reg apoptosis is precisely tuned by multiple mechanisms independent from E^BAB, and demonstrate that the role of E^BAB is specific for thymic negative selection (Fig. 9). Alternatively, the increased proportion of IEL precursors both in Bim^−/− and ΔE^BAB mice (Supplementary Fig. 4a, b) suggests that E^BAB controls fate determination of survived high-affinity TCR clones: ΔE^BAB may bias fate of thymocytes escaped from negative selection toward IEL rather than T_reg cells.

Bim KO and Bcl2 tg mice show abnormalities not only in thymic negative selection, but also in peripheral T cell homeostasis^{6,7,8,11,12,13,25,37,38} (Supplementary Fig. 8a, b and Supplementary Fig. 12). In contrast, peripheral T cells did not accumulate in the ΔE^BAB spleen (Fig. 6c, d). Consistent with this, ΔE^BAB affects neither peripheral T cell activation nor experimental autoimmunity (Fig. 8 and Supplementary Fig. 12). This suggests that high-affinity TCR clones were peripherally depleted in ΔE^BAB mice. One possible explanation for this observation is that peripherally activated T cells properly undergo IL-2 withdrawal-dependent apoptosis, a process that requires Bim but not E^BAB (Fig. 7a, b). This idea is further supported by that Bim was properly upregulated upon IL-2 withdrawal in activated T cells even in the absence of E^BAB (Fig. 7c, d). Additional peripheral T cell-specific enhancer(s) for Bim, or completely different mechanisms such as post-translational modifications, may also ensure Bim-mediated maintenance for peripheral T cell homeostasis. A different apoptosis pathway such as the Fas pathway could also be a major contributor in eliminating peripheral autoreactive T cells^1,46,50,51. The Fas pathway is known to be required for AICD, which we found requires neither Bim nor E^BAB (Fig. 7g, h). These results are consistent with that deletion of E^BAB had a very minor impact on genome-wide gene expression in peripheral T cells (Supplementary Fig. 12f). Collectively, our findings emphasize that the role of E^BAB is dedicated to thymic negative selection (Fig. 9).

A physiological contribution of a gene in a certain biological phenomenon depends on quantity of functional gene products that exist in that context. Regarding classical protein-coding genes, concentration of functional gene products-(modified) proteins is affected by transcriptional regulation, mRNA stability, translational efficacy, post-translational modifications, and protein stability, which are interconnected with each other. Each of these mechanisms profoundly contributes to determine the steady-state concentration of gene products, depending on genes and biological contexts. However, it has been technically difficult to dissect the relative importance of each mechanism for a gene to exert its physiological role(s) in a certain biological situation. Our study is unique in that we disrupted the single gene expression program by targeting the specific cis-element. This enabled us to reveal the degrees of contribution of Bim and E^BAB in thymic negative selection measured by different readouts, which represent cutting-edge molecular dissection of the cis-regulatory control for thymic negative selection in vivo. Although we do not exclude a possibility that downstream mechanisms such as Bim phosphorylation might be also important, we provide the evidence that the enhancer-driven gene expression regulation of Bim is an important upstream mechanism in thymic negative selection. In addition, the present work provides an example for using enhancer KO approach, instead of conventional genetic-manipulation of protein-coding genes, to address important biological questions.

Methods

Mice

All animal protocols were approved by the Animal Care and Use committee of Advanced Telecommunications Research Institute International (permission numbers: AN20140002, AN20150002, AN20160002, AN20170002, and AN20180002) and Kyoto university (permission numbers: K-16-12-11 and L-18-1). For generating ΔE^BAB mice, four gRNAs were individually cloned into the pX330 vector (#42230, Addgene, MA, USA) that harbors a Cas9 transgene. Briefly, two pairs of oligos (Supplementary Data 1) were annealed at 95 °C followed by natural cooling to 25 °C, and ligated with BbsI (New England Biolabs, MA, USA)-digested pX330 by using Ligation high (TOYOBO, Osaka, Japan). The obtained plasmids were sequence-validated (FASMAC, Kanagawa, Japan). The four gRNAs were together injected into fertilized eggs of C57BL/6N mice at the concentration of 2.5 μg/μl each (Transgenic, Fukuoka, Japan), generating the three different founders #44, #47, and #50. These founders were crossed with WT to obtain F1. Deleted regions were validated by sequencing analyses (FASMAC) against DNA from F0 and F1. The obtained F1 offspring were then crossed to generate ΔE^BAB mice. Bim^−/− (B6.129S1-Bcl2l11^tm1.1Ast/J) and OT-II tg (B6.Cg-Tg(TcrαTcrβ)425Cbn/J) mice were purchased from The Jackson Laboratory (ME, USA). OT-II tg, OT-I tg and HY tg mice were crossed with ΔE^BAB and Bim^−/⁻ mice for negative selection assays.

In all mice experiments, KO and the littermate control (WT or heterozygote), thus in total two to three mice were sacrificed and analyzed side-by-side on the same day to be considered as a littermate pair or trio. All experiments were performed without randomization. Blinding was performed in the injection process of OT-II negative selection assay. In other experiments, blinding was not done.

DNA isolation and genomic PCR

Genomic DNAs were prepared by using MightyAMP DNA polymerase kit (TaKaRa, Shiga, Japan) according to the manufacturer’s instruction. Genomic PCR experiments were performed using MightyAMP DNA polymerase or KOD FX-neo (TOYOBO) and the primers listed in Supplementary Data 1.

Epigeome analysis

Epigenome analyses were performed with the UCSC genome browser (https://genome.ucsc.edu/). For analyzing epigenome from peripheral naïve T cells, ChIP-seq data retrieved (accession numbers: GSE67443 and GSE60005) were mapped to the mm9 of mouse genome by using Bowtie2⁵². The obtained sam format files were converted into bam format files by using samtools⁵³ that were then subjected to peak calling with MACS⁵⁴. Wig format files were then uploaded into the UCSC genome browser to be visualized.

RNA isolation, cDNA synthesis, and qPCR

For nonlymphoid organs, total RNAs were prepared by using Trizol reagent (Thermo Fisher Scientific, MA, USA). The obtained supernatants containing total RNAs were further purified with RNeasy mini kit (Qiagen, Venlo, Netherlands) according to the manufacturer’s instruction. Total RNAs from lymphocyte samples were prepared by RNeasy mini kit with the equipped lysis buffer. 0.1–2 μg of total RNAs were then reverse transcribed with SuperScript III first-strand synthesis system (Thermo Fisher Scientific) according to the manufacturer’s instruction. The obtained cDNAs were 5- or tenfold-diluted and subjected into qPCR experiments by using LightCycler480 Instrument II system and SYBR Green Master Mix (Roche, Basel, Switzerland). The obtained data were analyzed using the delta-Ct method.

Antibodies

Anti-CD4 antibody (Clone: GK1.5 (FITC), SONY, Tokyo, Japan), anti-CD8a antibody (Clone: 53-6.7 (APC), SONY), anti-CD8b antibody (eBioH35-17.2 (PE), Invitrogen, CA, USA), anti-TCRβ antibody (Clone: H57-597 (FITC), SONY), anti-TCRβ antibody (Clone: H57-597 (APC/Cy7), BioLegend, CA, USA), anti-CD69 antibody (Clone: H1.2F3 (APC), BioLegend), anti-CD5 antibody (Clone: 53-7.3 (APC), BioLegend), anti-CD44 antibody (Clone: IM7 (Alexa Fluor 700), BioLegend), anti-CD62L antibody (Clone: MEL-14 (PE), BioLegend), anti-TCR Vβ5.1, 5.2 antibody (Clone: MR9-4 (PE), BioLegend), anti-TCR HY antibody (Clone: T3.70 (PE), eBioscience, CA, USA), anti-B220 antibody (Clone: RA3-6B2 (FITC), SONY), LEAF-purified anti-CD3ε antibody (Clone: 145-2C11, BioLegend), LEAF-purified anti-CD28 antibody (Clone: 37.51, BioLegend), Biotin anti-TCRβ antibody (Clone: H57-597, BioLegend) and Biotin anti-CD69 antibody (Clone: H1.2F3, BioLegend) were used in this study.

Flow cytometry

The thymus and spleen were harvested to obtain single lymphocytes suspension. For splenocytes, red blood cells (RBCs) were lysed with RBC lysis buffer (0.015 M NH₄Cl, 0.1 mM KHCO₃, 0.01 mM Na₂EDTA). Live cell numbers were counted by using Countess system (Thermo Fisher Scientific). The obtained samples (1 × 10⁶ cells) were then stained with anti-CD4 antibody (1:200), anti-CD8a antibody (1:100), anti-CD8b antibody (1:200), anti-TCRβ antibody (1:200 (FITC), 1:100 (APC/Cy7)), anti-CD69 antibody (1:100), anti-CD5 antibody (1:100), anti-CD44 antibody (1:200), anti-CD62L antibody (1:100), anti-TCR Vβ5.1, 5.2 antibody (1:100), anti-TCR HY antibody (1:200) or anti-B220 antibody (1:100) in 100 μl of FCM buffer (10% fetal bovine serum (FBS) in phosphate-buffered saline (PBS)) for >30 min at 4 °C in the dark. For IEL precursors and iNKT measurement, 1 × 10⁶ thymocytes were stained with α-GalCer-loaded CD1d tetramer (TS-MCD-1 (PE), MBL, Aichi, Japan) in 50 μl of FCM buffer for >30 min at 4 °C in the dark, and then stained with additional antibodies (see Supplementary Fig. 4 for the details). To analyze T_reg cells, the obtained single cell suspension (3–5 × 10⁶ cells) was stained using Foxp3 staining kit (Clone: FJK-16s, eBioscience, CA, USA), essentially according to the manufacturer’s instruction. Cell fixation was performed for 10 min. Samples were analyzed on EC800 (SONY) and FACS CantoII (BD Bioscience, NJ, USA). Gating strategies are shown in Supplementary Fig. 14.

RNA-seq and bioinformatic analysis

Total RNAs were extracted from the thymus and spleen as described in the RNA isolation section. RNA-seq libraries were generated using the SureSelect Strand-Specific RNA Library Prep for Illumina (Agilent) according to the manufacturer’s instructions. Sequencing experiments were performed with Hiseq2500 (Illumina; Single End 36 bp). The obtained reads were mapped to the mouse genome mm9 using Illumina Eland with the default parameter setting. Uniquely aligned reads were retrieved allowing up to 2 bp mismatches, and the number of exon-mapped overlapping reads were counted. The obtained gene list with reads per million per a kilobase (RPKM) scores were shown in Supplementary Datas 2 and 3. To identify DEGs, we first focused on the well-annotated protein-coding genes. RPKM scores from two replicates were averaged, and the ratio ΔE^BAB/WT were calculated. In this calculation, 1 was added to all averaged RPKM scores to ignore scores below “1”, and to make analyses more stringent. The obtained ratios were used to sort genes to find candidate DEGs, followed by statistical analyses (unpaired two-tailed Student’s t-test) and qPCR validation.

Cell sorting

To purify pre- and post-selection thymocytes, single cell suspension (1–2 × 10⁷ cells) was stained with anti-TCRβ antibody (1:100 or 1:200) and anti-CD69 antibody (1:100 or 1:200) in 100 μl of FCM buffer for >30 min at 4 °C in the dark. The stained cells were then tenfold diluted and sorted using SH800 cell sorter system (SONY).

Ex vivo thymocytes cell death assay

Totally, 1 × 10⁵ total thymocytes were cultured with DEX (10 nM), PMA (2 ng/ml) or ionomycin (1 μg/ml or 0.1 μg/ml) in 200 μl of RPMI1640 media (10% FBS, 1% penicillin streptomycin, 50 μM 2-mercaptoethanol, 1 × nonessential amino acids (nacalai tesque, Kyoto, Japan) and 1 × sodium pyruvate (nacalai tesque)) in a 96-well flat bottom plate. DMSO- or EtOH-treated cells served as controls. Viability was measured by Annexin V FLUOS staining kit (Roche) at days 0–2.

Ex vivo TCR stimulation

Ex vivo TCR stimulation experiments were performed essentially as described previously^7,25. Briefly, 12-well plates were coated with LEAF-purified anti-CD3ε and anti-CD28 antibodies (0 or 10 μg/ml in 500 μl of PBS, BioLegend) for 2 h at 37 °C. After 3 times-wash with 0.5 ml of PBS, 1 × 10⁶ cells of total thymocytes were cultured in 2 ml of DMEM media (10% FBS, 1% penicillin streptomycin, 2 mM l-glutamine, 50 μM 2-mercaptoethanol, 10 mM HEPES (pH 7.4), 1 × nonessential amino acids (nacalai tesque) and 1 × sodium pyruvate (nacalai tesque)) for 3 h (for qPCR experiments) or 9 h (for Annexin V and PI staining using Annexin V FLUOS staining kit (Roche)).

Peptides

OVA_323–339 peptide (ISQAVHAAHAEINEAGR, BEX, Tokyo, Japan), OVA_257–264 peptide (SIINFEKL, BEX), gp33 (KAVYNFATC, BEX), and Q4R7 (SIIQFERL, BEX) were certified as >98% pure by HPLC. Peptides were dissolved in dimethyl sulfoxide (DMSO) at concentration of 100 mg/ml and stored at −80 °C. The peptide/DMSO solutions were diluted in PBS just before the experiment.

OT-II negative selection assay

OT-II negative selection assay was performed as described previously⁶. Briefly, WT; OT-II⁺, E^BAB+/−; OT-II⁺ and ΔE^BAB; OT-II⁺ mice were injected intraperitoneally with 1 mg OVA_323–339 peptide or OVA_257–264 peptide as a control resuspended in 500 μl of PBS. Thymocytes were harvested after 72 h for analysis.

OT-I tg fetal thymic organ culture

FTOC was performed as described previously^33,34,35. Briefly, fetal thymic lobes were excised at embryonic day 15 (E15) and cultured on Whatman Nuclepore Track-Etched Membrane (WHA110409, GE Healthcare, Little Chalfont, England) floated on RPMI1640 media (10% FBS, 1% penicillin streptomycin, 50 μM 2-mercaptoethanol, 1 × nonessential amino acids (nacalai tesque, Kyoto, Japan) and 1 × sodium pyruvate (nacalai tesque)) in the presence of 2 μM OVA_257–264, 2 μM Q4R7, or 20 μM gp33. On day 4, thymocytes were analyzed by flow cytometry.

Immunoblotting assay

CD69⁺ and CD69⁻ T cells were enriched from thymocytes using biotinylated anti-CD69 antibody (BioLegend) and MojoSort Streptavidin Nanobeads (BioLegend). CD69^- T cells were further purified by depleting TCRβ⁺ cells using biotinylated anti-TCRβ antibody (BioLegend) and MojoSort Streptavidin Nanobeads (BioLegend). Expression of Bim and ACTB proteins in each subset was detected with anti-Bim antibody (1:1000, #2819, Cell Signaling, MA, USA) and anti-ACTB antibody (1:10,000, NB600-532, Novus Biologicals, CO, USA). As the secondary antibody, anti-rabbit IgG (1:15,000, 711-035-152, Jackson ImmumoResearch, PA, USA) was used. Signals were visualized with ECL plus Western Blotting Detection Reagents (GE Healthcare) and analyzed by the CCD digital imaging system LAS-4000 Luminescent Image Analyzer (GE Healthcare). Whole proteins were stained with SYPRO Ruby Protein Gel Stain (S12000, Thermo Fisher Scientific) and detected by the CCD digital imaging system LAS-4000 Luminescent Image Analyzer (GE Healthcare). Uncropped scans are available in the Source data file.

Cytokine withdrawal assay

CD4⁺ or CD8⁺ T cells were isolated from splenocytes using MojoSort Mouse CD4 or CD8 T Cell Isolation Kit (BioLegend), respectively according to the manufacturer’s instruction. T cells were activated by culturing cells in RPMI1640 media (10% FBS, 1% penicillin streptomycin, 50 μM 2-mercaptoethanol, 1 × nonessential amino acids (nacalai tesque) and 1 × sodium pyruvate (nacalai tesque)) plus IL-2 (100 U/ml, Wako, Osaka, Japan) in 96 well flat bottom plate coated with LEAF-purified anti-CD3ε and anti-CD28 antibodies (5 μg/ml in 50 μl of PBS, BioLegend) for 48 hr. Activated T cells were then cultured for 1 day with RPMI1640 plus IL-2 (100 U/ml, Wako) in normal 96-well plate. After that, IL-2 was removed from the media and viability of CD4⁺ and CD8⁺ T cells was measured using Annexin V FLUOS staining kit (Roche) on days 1–3 after IL-2 withdrawal.

Activation induced cell death assay

Isolation and activation of CD4⁺ or CD8⁺ T cells were performed as described in the “cytokine withdrawal assay” section. After activation and 1-day culture in the presence of IL-2, T cells were further incubated in RPMI1640 media (10% FBS, 1% penicillin streptomycin, 50 μM 2-mercaptoethanol, 1 × nonessential amino acids (nacalai tesque) and 1 × sodium pyruvate (nacalai tesque)) without IL-2 in a 96-well flat bottom plate coated with LEAF-purified anti-CD3ε (5 μg/ml in 50 μl of PBS, BioLegend) for 6 h. Cell viability was determined using Annexin V FLUOS staining kit (Roche).

Histochemistry

Mouse tissues were fixed in 10% neutral buffered formalin. Samples were then processed into 6 micron sections, mounted, and Hematoxylin and eosin (H&E)-stained (Genostaff, Tokyo, Japan). The prepared samples were observed with Nikon Ni-E microscope (Nikon, Tokyo, Japan).

Experimental autoimmune encephalomyelitis

Experimental autoimmune encephalomyelitis (EAE) experiments were performed essentially as described previously⁴⁷. Briefly, mice (WT, ΔE^BAB, and Bim^−/⁻) were subjected to subcutaneous injection at two different sites with an emulsion of complete Freund’s adjuvant (CFA; 250 μg heat-killed Mycobacterium tuberculosis H37Ra (Difco) dissolved in 50 μl incomplete Freund’s adjuvant (Sigma)) and 250 μg myelin oligodendrocyte glycoprotein (MOG) peptides (amino acids 35–55) in 50 μl PBS. Mice were then injected with 200 ng of Pertussis Toxin (Calbiochem) intraperitoneally at day 0 and day 2 after EAE induction. The clinical scores were assessed as follows: (1) flaccid tail, (2) impaired righting reflex and/or gait, (3) partial hind limb paralysis, (4) total hind limb paralysis, and (5) total hind limb paralysis with partial forelimb paralysis.

Statistics

The sample size was chosen as follows. First, the number of animals was minimized as much as possible in light of animal ethics. Second, against effect size estimated in each experiment, ≥80–90% power was favored. Third, in most cases, n = 5 was set as a threshold according to the previous report⁵⁵. These three criterion functioned to determine the sample size as n = 5–8 in most experiments. Moreover, dot-plot/box plot representation of data provided insights into how the samples were distributed, and thus into the extent of difference of two groups. This, in the specific cases, led us to conclude that relatively small number of animals (i.e., less than five animals) was enough to support our conclusion.

According to the observed differences, our sample size appeared to be appropriate to meet≥80–90% power. In addition, dot-plot/box plot representation of data was useful to estimate sample distribution. Clearly abnormal distribution was not observed in our experiments, further supporting that our statistical tests were appropriate.

Significant differences between two groups were examined using one or two-tailed, unpaired t test, or Mann–Whitney U test. One-tailed test was chosen when we had hypothesis and/or previous knowledge regarding direction of changes (e.g., increased or decreased) in experiments.

Reporting summary

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

Data availability

The RNA-seq data have been deposited in DNA Data Bank of Japan (DDBJ) under the accession code DRA004726 and DRA008123. Source data underlying graphs and blots are supplied as a Source data file. All other data are included in the supplemental information or available from the authors upon reasonable requests.

References

Xing, Y. & Hogquist, K. A. T-cell tolerance: central and peripheral. Cold Spring Harb. Perspect. Biol. 4, a006957 (2012).
Article Google Scholar
Klein, L., Kyewski, B., Allen, P. M. & Hogquist, K. A. Positive and negative selection of the T cell repertoire: what thymocytes see (and don’t see). Nat. Rev. Immunol. 14, 377–391 (2014).
Article CAS Google Scholar
Hogquist, K. A. & Jameson, S. C. The self-obsession of T cells: how TCR signaling thresholds affect fate ‘decisions’ and effector function. Nat. Immunol. 15, 815–823 (2014).
Article CAS Google Scholar
Richards, D. M., Kyewski, B. & Feuerer, M. Re-examining the nature and function of self-reactive T cells. Trends Immunol. 37, 114–125 (2016).
Article CAS Google Scholar
Wirnsberger, G., Hinterberger, M. & Klein, L. Regulatory T-cell differentiation versus clonal deletion of autoreactive thymocytes. Immunol. Cell Biol. 89, 45–53 (2011).
Article Google Scholar
Bouillet, P. et al. BH3-only Bcl-2 family member Bim is required for apoptosis of autoreactive thymocytes. Nature 415, 922–926 (2002).
Article CAS ADS Google Scholar
Bouillet, P. et al. Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science 286, 1735–1738 (1999).
Article CAS Google Scholar
Labi, V. et al. Deregulated cell death and lymphocyte homeostasis cause premature lethality in mice lacking the BH3-only proteins Bim and Bmf. Blood 123, 2652–2662 (2014).
Article CAS Google Scholar
Herold, M. J. et al. Impact of conditional deletion of the pro-apoptotic BCL-2 family member BIM in mice. Cell Death Dis. 5, e1446 (2014).
Article CAS Google Scholar
Li, K. P. et al. Temporal expression of Bim limits the development of agonist-selected thymocytes and skews their TCRbeta repertoire. J. Immunol. 198, 257–269 (2017).
Article CAS Google Scholar
Hildeman, D. A. et al. Activated T cell death in vivo mediated by proapoptotic bcl-2 family member bim. Immunity 16, 759–767 (2002).
Article CAS Google Scholar
Hale, J. S., Nelson, L. T., Simmons, K. B. & Fink, P. J. Bcl-2-interacting mediator of cell death influences autoantigen-driven deletion and TCR revision. J. Immunol. 186, 799–806 (2011).
Article CAS Google Scholar
Li, W. Q., Guszczynski, T., Hixon, J. A. & Durum, S. K. Interleukin-7 regulates Bim proapoptotic activity in peripheral T-cell survival. Mol. Cell Biol. 30, 590–600 (2010).
Article Google Scholar
Chougnet, C. A. et al. A major role for Bim in regulatory T cell homeostasis. J. Immunol. 186, 156–163 (2011).
Article CAS Google Scholar
Stritesky, G. L. et al. Murine thymic selection quantified using a unique method to capture deleted T cells. Proc. Natl Acad. Sci. USA 110, 4679–4684 (2013).
Article CAS ADS Google Scholar
Ludwinski, M. W. et al. Critical roles of Bim in T cell activation and T cell-mediated autoimmune inflammation in mice. J. Clin. Invest. 119, 1706–1713 (2009).
Article CAS Google Scholar
Strasser, A., Puthalakath, H., O’Reilly, L. A. & Bouillet, P. What do we know about the mechanisms of elimination of autoreactive T and B cells and what challenges remain. Immunol. Cell Biol. 86, 57–66 (2008).
Article CAS Google Scholar
Sionov, R. V., Vlahopoulos, S. A. & Granot, Z. Regulation of Bim in health and disease. Oncotarget 6, 23058–23134 (2015).
Article Google Scholar
Calo, E. & Wysocka, J. Modification of enhancer chromatin: what, how, and why? Mol. Cell 49, 825–837 (2013).
Article CAS Google Scholar
Vernimmen, D. & Bickmore, W. A. The hierarchy of transcriptional activation: from enhancer to promoter. Trends Genet. 31, 696–708 (2015).
Article CAS Google Scholar
Sander, J. D. & Joung, J. K. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat. Biotechnol. 32, 347–355 (2014).
Article CAS Google Scholar
Canver, M. C. et al. BCL11A enhancer dissection by Cas9-mediated in situ saturating mutagenesis. Nature 527, 192–197 (2015).
Article CAS ADS Google Scholar
LaFlam, T. N. et al. Identification of a novel cis-regulatory element essential for immune tolerance. J. Exp. Med. 212, 1993–2002 (2015).
Article CAS Google Scholar
Thomas, G. D. et al. Deleting an Nr4a1 super-enhancer subdomain ablates ly6c(low) monocytes while preserving macrophage gene function. Immunity 45, 975–987 (2016).
Article CAS Google Scholar
Burger M. L., Leung K. K., Bennett M. J., Winoto A. T cell-specific inhibition of multiple apoptotic pathways blocks negative selection and causes autoimmunity. Elife 3, 03468 (2014). https://doi.org/10.7554/eLife.03468
Fassett, M. S., Jiang, W., D’Alise, A. M., Mathis, D. & Benoist, C. Nuclear receptor Nr4a1 modulates both regulatory T-cell (Treg) differentiation and clonal deletion. Proc. Natl Acad. Sci. USA 109, 3891–3896 (2012).
Article CAS ADS Google Scholar
Konkel, J. E. et al. Control of the development of CD8alphaalpha+ intestinal intraepithelial lymphocytes by TGF-beta. Nat. Immunol. 12, 312–319 (2011).
Article CAS Google Scholar
McDonald, B. D., Bunker, J. J., Ishizuka, I. E., Jabri, B. & Bendelac, A. Elevated T cell receptor signaling identifies a thymic precursor to the TCRalphabeta(+)CD4(−)CD8beta(−) intraepithelial lymphocyte lineage. Immunity 41, 219–229 (2014).
Article CAS Google Scholar
Matsuda, J. L. et al. Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192, 741–754 (2000).
Article CAS Google Scholar
Pellicci, D. G. et al. A natural killer T (NKT) cell developmental pathway Involving a thymus-dependent NK1.1(−)CD4(+) CD1d-dependent precursor stage. J. Exp. Med. 195, 835–844 (2002).
Article CAS Google Scholar
Kisielow, P., Bluthmann, H., Staerz, U. D., Steinmetz, M. & von Boehmer, H. Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+ 8+ thymocytes. Nature 333, 742–746 (1988).
Article CAS ADS Google Scholar
Barnden, M. J., Allison, J., Heath, W. R. & Carbone, F. R. Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements. Immunol. Cell Biol. 76, 34–40 (1998).
Article CAS Google Scholar
Hogquist, K. A. et al. T cell receptor antagonist peptides induce positive selection. Cell 76, 17–27 (1994).
Article CAS Google Scholar
Daniels, M. A. et al. Thymic selection threshold defined by compartmentalization of Ras/MAPK signalling. Nature 444, 724–729 (2006).
Article CAS ADS Google Scholar
Takada, K. et al. TCR affinity for thymoproteasome-dependent positively selecting peptides conditions antigen responsiveness in CD8(+) T cells. Nat. Immunol. 16, 1069–1076 (2015).
Article CAS Google Scholar
Nitta, T., Ohigashi, I. & Takahama, Y. The development of T lymphocytes in fetal thymus organ culture. Methods Mol. Biol. 946, 85–102 (2013).
Article CAS Google Scholar
Strasser, A., Harris, A. W., von Boehmer, H. & Cory, S. Positive and negative selection of T cells in T-cell receptor transgenic mice expressing a bcl-2 transgene. Proc. Natl Acad. Sci. USA 91, 1376–1380 (1994).
Article CAS ADS Google Scholar
Strasser, A., Harris, A. W. & Cory, S. bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell 67, 889–899 (1991).
Article CAS Google Scholar
Cowley, D. O., Muse, G. W. & Van Dyke, T. A dominant interfering Bub1 mutant is insufficient to induce or alter thymic tumorigenesis in vivo, even in a sensitized genetic background. Mol. Cell Biol. 25, 7796–7802 (2005).
Article CAS Google Scholar
Schmitz, I., Clayton, L. K. & Reinherz, E. L. Gene expression analysis of thymocyte selection in vivo. Int Immunol. 15, 1237–1248 (2003).
Article CAS Google Scholar
O’Connor, L. et al. Bim: a novel member of the Bcl-2 family that promotes apoptosis. EMBO J. 17, 384–395 (1998).
Article Google Scholar
Sekiya, T. et al. The nuclear orphan receptor Nr4a2 induces Foxp3 and regulates differentiation of CD4+ T cells. Nat. Commun. 2, 269 (2011).
Article Google Scholar
Sekiya, T. et al. Nr4a receptors are essential for thymic regulatory T cell development and immune homeostasis. Nat. Immunol. 14, 230–237 (2013).
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 CAS ADS Google Scholar
Kelly, P. N. et al. Individual and overlapping roles of BH3-only proteins Bim and Bad in apoptosis of lymphocytes and platelets and in suppression of thymic lymphoma development. Cell Death Differ. 17, 1655–1664 (2010).
Article CAS ADS Google Scholar
Krammer, P. H., Arnold, R. & Lavrik, I. N. Life and death in peripheral T cells. Nat. Rev. Immunol. 7, 532–542 (2007).
Article CAS Google Scholar
Yasuda, K. et al. Satb1 regulates the effector program of encephalitogenic tissue Th17 cells in chronic inflammation. Nat. Commun. 10, 549 (2019).
Article ADS Google Scholar
Tai, X. et al. Foxp3 transcription factor is proapoptotic and lethal to developing regulatory T cells unless counterbalanced by cytokine survival signals. Immunity 38, 1116–1128 (2013).
Article CAS Google Scholar
Lio, C. W. & Hsieh, C. S. A two-step process for thymic regulatory T cell development. Immunity 28, 100–111 (2008).
Article CAS Google Scholar
Fortner, K. A., Bouillet, P., Strasser, A. & Budd, R. C. Apoptosis regulators Fas and Bim synergistically control T-lymphocyte homeostatic proliferation. Eur. J. Immunol. 40, 3043–3053 (2010).
Article CAS Google Scholar
Ju, S. T. et al. Fas(CD95)/FasL interactions required for programmed cell death after T-cell activation. Nature 373, 444–448 (1995).
Article CAS ADS Google Scholar
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Article CAS Google Scholar
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Article Google Scholar
Zhang, Y. et al. Model-based analysis of ChIP-Seq (MACS). Genome Biol. 9, R137 (2008).
Article Google Scholar
Krzywinski, M. & Altman, N. Points of significance: comparing samples—part I. Nat. Methods 11, 215–216 (2014).
Article CAS Google Scholar

Download references

Acknowledgements

We thank Dr. Thomas. N. Sato (T.N.S), the director of The TNS BioMEC-X Laboratories, ATR, and JST ERATO Sato Live Bio-forecasting project, for supporting all aspects of the study. We thank Tomoko Kuroda, Tomoko Ninomiya, Hitomi Anabuki, Satsuki Endo, Fumihiko Sagawa, Satoshi Kozawa, Terumi Horiuchi, and Kiyomi Imamura for technical assistance. We thank Ryoko Takahashi, Erika Kojima, and Toshiya Morie for administrative assistance. We are thankful to Dr. Pieter Bas Kwak and Dr. Bryce Nelson for critically reading the manuscript. We also thank Dr. Kosuke Yusa and Dr. Takeshi Watanabe for providing insightful comments on the manuscript. Dr. Thomas. N. Sato and the members of the TNS BioMEC-X Laboratories provided insightful comments on the manuscript. This work was supported by JST ERATO (to S.K. via T.N.S; JPMJER1303), JSPS KAKENHI (26890030, 15H01478, 18K15409, and 18H04810: S.K.), The Uehara Memorial Foundation (S.K.), The Shimizu Foundation for Immunology and Neuroscience Grant for 2018 (S.K) and Japan Foundation for Applied Enzymology (S.K.).

Author information

Authors and Affiliations

Graduate School of Frontier Science, The University of Tokyo, Kashiwa-shi, Chiba, 277-8562, Japan
Miki Arai Hojo & Yutaka Suzuki
The Thomas N. Sato BioMEC-X Laboratories, Advanced Telecommunications Research Institute International (ATR), Soraku-gun, Kyoto, 619-0237, Japan
Miki Arai Hojo, Hiroaki Hojo & Shinpei Kawaoka
Institute for Frontier Life and Medical Sciences, Kyoto University, Kyoto-shi, Kyoto, 606-8507, Japan
Kyoko Masuda, Hiroaki Hojo, Yosuke Nagahata, Keiko Yasuda, Daiya Ohara, Yusuke Takeuchi, Keiji Hirota, Hiroshi Kawamoto & Shinpei Kawaoka
ERATO Sato Live Bio-forecasting Project, Japan Science and Technology Agency (JST), Soraku-gun, Kyoto, 619-0237, Japan
Hiroaki Hojo & Shinpei Kawaoka

Authors

Miki Arai Hojo
View author publications
You can also search for this author in PubMed Google Scholar
Kyoko Masuda
View author publications
You can also search for this author in PubMed Google Scholar
Hiroaki Hojo
View author publications
You can also search for this author in PubMed Google Scholar
Yosuke Nagahata
View author publications
You can also search for this author in PubMed Google Scholar
Keiko Yasuda
View author publications
You can also search for this author in PubMed Google Scholar
Daiya Ohara
View author publications
You can also search for this author in PubMed Google Scholar
Yusuke Takeuchi
View author publications
You can also search for this author in PubMed Google Scholar
Keiji Hirota
View author publications
You can also search for this author in PubMed Google Scholar
Yutaka Suzuki
View author publications
You can also search for this author in PubMed Google Scholar
Hiroshi Kawamoto
View author publications
You can also search for this author in PubMed Google Scholar
Shinpei Kawaoka
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.A.H. performed and analyzed the most of the experiments except EAE and wrote the paper. K.M., Y.N. and H.K. performed the TCR transgenic mice experiments. H.H., K.Y., D.O., Y.T. and K.H. performed the EAE experiments. Y.S. and S.K. performed the RNA-seq and bioinformatics analyses. S.K. conceived and supervised the project, and wrote the paper. All authors provided intellectual input and reviewed the paper.

Corresponding author

Correspondence to Shinpei Kawaoka.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Journal peer review information: Nature Communications thanks Stephen Jameson and 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

Peer Review File

Reporting Summary

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

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

Hojo, M.A., Masuda, K., Hojo, H. et al. Identification of a genomic enhancer that enforces proper apoptosis induction in thymic negative selection. Nat Commun 10, 2603 (2019). https://doi.org/10.1038/s41467-019-10525-1

Download citation

Received: 24 April 2018
Accepted: 19 May 2019
Published: 13 June 2019
DOI: https://doi.org/10.1038/s41467-019-10525-1

This article is cited by

Remote solid cancers rewire hepatic nitrogen metabolism via host nicotinamide-N-methyltransferase
- Rin Mizuno
- Hiroaki Hojo
- Shinpei Kawaoka
Nature Communications (2022)

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.