Multiple studies highlighted the overtly self-reactive T cell repertoire in the diabetes-prone NOD mouse. This autoreactivity has primarily been linked to defects in apoptosis induction during central tolerance. Previous studies suggested that thymus-specific serine protease (TSSP), a putative serine protease expressed by cortical thymic epithelial cells and thymic dendritic cells, may edit the repertoire of self-peptides presented by MHC class II molecules and shapes the self-reactive CD4 T cell repertoire. To gain further insight into the role of TSSP in the selection of self-reactive CD4 T cells by endogenous self-Ags, we examined the development of thymocytes expressing distinct diabetogenic TCRs sharing common specificity in a thymic environment lacking TSSP. Using mixed bone marrow chimeras, we evaluated the effect of TSSP deficiency confined to different thymic stromal cells on the differentiation of thymocytes expressing the chromogranin A–reactive BDC-2.5 and BDC-10.1 TCRs or the islet amyloid polypeptide–reactive TCR BDC-6.9 and BDC-5.2.9. We found that TSSP deficiency resulted in deficient positive selection and induced deletion of the BDC-6.9 and BDC-10.1 TCRs, but it did not affect the differentiation of the BDC-2.5 and BDC-5.2.9 TCRs. Hence, TSSP has a subtle role in the generation of self-peptide ligands directing diabetogenic CD4 T cell development. These results provide additional evidence for TSSP activity as a novel mechanism promoting autoreactive CD4 T cell development/accumulation in the NOD mouse.

Type 1 diabetes (T1D) is an autoimmune disease resulting from the selective destruction of insulin-secreting pancreatic islet β cells. Both genetic and environmental factors contribute to the progressive mononuclear cell infiltration of pancreatic islets that leads to disease (1, 2). T lymphocytes are critical for T1D development (36), and both CD8 and CD4 T cells contribute to pathogenesis (7, 8). The strong association between T1D susceptibility and particular MHC class II alleles appears to involve specific features of the MHC class II β-chain that influence MHC class II:peptide complex stability, as well as the repertoire of bound peptides that drives CD4 thymocyte selection (9). Therefore, enzymatic activities impacting the nature of the available peptide pool are likely to further modulate the selection of autoreactive CD4 T cells.

Thymus-specific serine protease (TSSP) is a putative serine protease highly expressed by cortical thymic epithelial cells (cTECs) (10, 11) and expressed at lower levels by thymic dendritic cells (DCs) (12). Although the precise function of TSSP is unknown, indirect evidence suggests that this protease may have a subtle role in the MHC class II presentation pathway in the thymus (13). TSSP-deficient mice show normal development of CD4 T cells (12, 14). However, although globally immunocompetent, we found that TSSP-deficient NOD mice (Tssp° NOD) show a reduced capacity to mount a CD4 T cell response to the foreign Ag hen egg lysozyme (15). Moreover, we found that Tssp° NOD mice are completely protected from spontaneous diabetes and severe insulitis, with no evidence for enhanced regulatory T cell activity (12). Diabetes resistance reflected the limited diabetogenic potential of the mature CD4 T cell compartment that was imposed during intrathymic development of the polyclonal CD4 T cell repertoire by both cTECs and bone marrow (BM)-derived APCs. These observations suggested that the lack of TSSP expression by the different thymic stromal compartments may impair the development of CD4 T cells essential for diabetes initiation/development. Although selection of CD4 thymocytes expressing the highly diabetogenic NY4.1 TCR or the IA2β-specific phogrin18 TCR was impaired in Tssp° NOD mice, the development of CD4 T cells specific for the InsB9-23 epitope of insulin, for glutamic acid decarboxylase (GAD)65, or islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP) was not affected by TSSP deficiency, thus questioning the relative impact of TSSP deficiency on the intrathymic development of the diabetogenic CD4 T cell repertoire. Understanding how TSSP may edit the self-peptide repertoire presented by class II molecules in the thymus and the impact of such editing on the development of diabetogenic CD4 T cells is critical given the major role of TSSP in diabetes development in the NOD mouse and the genetic linkage of Prss16, the gene coding for TSSP, to a diabetes-susceptibility locus of the extended HLA region (1618).

BDC clones are the first set of islet cell Ag-reactive, diabetogenic CD4 T cell clones isolated from newly diabetic NOD mice (1921) and remain the largest panel of highly diabetogenic T cell clones characterized so far (22). The overtly diabetogenic character of CD4 T cells bearing the BDC-2.5, BDC-10.1, or BDC-6.9 TCR is shown, for example, in engineered NOD mice expressing retrovirus-encoded TCRs (TCR retrogenics [Rgs]). Unlike TCRs reacting with insulin, IA2β, or GAD65 epitopes, which induced limited or no disease, both BDC-2.5 and BDC-10.1 TCRs induced insulitis and 100% diabetes incidence, with a lower ID50 value in the case of the BDC-10.1 TCR (23). In comparison, the diabetogenic potential of the BDC-6.9 TCR was less marked in the Rg setting, yet ∼60% of BDC-6.9 TCR–Rg mice became diabetic (23). In addition, diabetes is accelerated in some BDC-6.9 TCR–transgenic (Tg) mice (24).

The elusive Ags targeted by these diabetogenic CD4 T cells were identified recently. The BDC-2.5 and BDC-10.1 CD4 T cell clones both weakly react with the naturally occurring 358-371 cleavage product (WE14) of chromogranin A (ChgA), a bioactive peptide precursor found in the secretory vesicles of endocrine cells and neurons (25). The BDC-2.5 CD4 T cells also were reported to recognize a distinct hormonal polypeptide derived from ChgA, the vasostatin-1–derived peptide ChA29–42 (26). In contrast, the BDC-5.2.9 T cell clone recognizes the β cell–specific peptide hormone islet amyloid polypeptide (IAPP) (27). The diabetogenic BDC-6.9 T cell clone (28) is also very likely to recognize an IAPP epitope, because a genetic mapping analysis revealed that its Ag reactivity is linked to the iapp gene on chromosome 6 (29), and IAPP-deficient islets fail to induce BDC-6.9 CD4 T cell activation (27). However, an in vitro response of BDC-6.9 to IAPP or IAPP-derived peptides could not be directly documented (27). Interestingly, CD4 T cells reactive to IAPP were detected in the pancreas of prediabetic NOD mice using tetramer technology (30), suggesting that IAPP targeting occurs early during diabetes pathogenesis in mice.

The shared specificity of the BDC-2.5 and BDC-10.1 TCRs and of the BDC-5.2.9 and BDC-6.9 TCRs, as well as the highly diabetogenic potential of these TCRs, offers a unique experimental system to decipher the role of TSSP in the presentation of self-Ags by thymic stromal cells for selection of the islet-reactive CD4 T cell repertoire. We show in this study that TSSP deficiency differentially impacts the development of these TCRs. Collectively, the results point to a subtle role for TSSP in the class II presentation pathway and highlight a novel mechanism contributing to selection of the overtly self-reactive CD4 T cell repertoire of the NOD mouse.

NOD (NOD/ShiLtJ), NODscid (NOD.CB17-Prkdcscid/J), Rag-1−/− NOD [NOD.129S7(B6)-Rag-1tm1Mom/J], and TCR Cα−/− NOD [NOD.129.P2(C)-Tcratm1Mjo/DoiJ](NOD-Cα°) mice were purchased from Charles River Laboratories (l’Arbresles, France). BDC-2.5 NOD mice (31) were a gift from Dr. B. Salomon (Hopital La Pitié Salpêtrière, INSERM U959, Paris, France). TSSP-deficient NODscid (Tssp° NODscid), TSSP-deficient NOD-Cα°, and TSSP-sufficient controls were described previously (12). Mice were maintained under pathogen-free conditions. All mouse experiments were performed in accordance with national and European regulations and INSERM institutional guidelines, and experimental protocols were approved by the local ethics committee of Midi-Pyrénées, Toulouse, France.

Cells were stained with a combination of FITC-, PE-, PE-Cy7–, allophycocyanin-, allophycocyanin-eFluor 780–, Brilliant Violet 421–, Alexa Fluor 405–, Brilliant Violet 605–, eFluor 506–, or PerCP-Cy5.5–conjugated mAbs. The anti-CD4 (clone RM4-5), CD8α (53-6.7), CD24 (clone M1/69), CD5 (clone 53-7.3), CD69 (H1.2F3), anti-TCR Cβ (H57-597), and Vβ4 (KT4) Abs were from eBioscience or BD Biosciences. Events were collected within a lymphoid gate based on forward scatter and side scatter profiles, and dead cells were excluded using TO-PRO-3 or Fixable Viability Dye staining. Data were acquired on a BD FACSCalibur (Figs. 1, 3, 6) or a BD Fortessa flow cytometer (Figs. 4, 5, 7) and analyzed with FlowJo software (TreeStar).

FIGURE 1.

Normal developmental profile of CD4 T cells expressing the ChgA WE14-reactive diabetogenic BDC-2.5 TCR in a thymic microenvironment lacking TSSP expression. (A) A 3:1 mix of T cell–depleted BM cells from BDC-2.5 TCR–Tg NOD mice and NODscid BM cells (Tg + WT) or Tssp° NODscid BM cells (Tg + KO) were injected i.v. into lethally irradiated NOD-Cα° (WT) or Tssp° NOD-Cα° (KO) hosts, as indicated (BM → host). The CD4 versus CD8 profile of thymocytes and TCR Vβ4 expression by immature CD4+CD8+ DP cells and mature CD4+CD8 (CD4SP) cells is shown for a representative chimera of each group. The percentages of DP and CD4SP cells are indicated. For line graphs, the values represent the geometric mean of fluorescence intensity. The shaded graphs represent unstained thymocytes. (B) Absolute thymocyte number found in the different chimeras. Percentage (C) and number (D) of DP and CD4SP cells in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group. For clarity, the TSSP status of TEC and BM cells is indicated on the x-axis.

FIGURE 1.

Normal developmental profile of CD4 T cells expressing the ChgA WE14-reactive diabetogenic BDC-2.5 TCR in a thymic microenvironment lacking TSSP expression. (A) A 3:1 mix of T cell–depleted BM cells from BDC-2.5 TCR–Tg NOD mice and NODscid BM cells (Tg + WT) or Tssp° NODscid BM cells (Tg + KO) were injected i.v. into lethally irradiated NOD-Cα° (WT) or Tssp° NOD-Cα° (KO) hosts, as indicated (BM → host). The CD4 versus CD8 profile of thymocytes and TCR Vβ4 expression by immature CD4+CD8+ DP cells and mature CD4+CD8 (CD4SP) cells is shown for a representative chimera of each group. The percentages of DP and CD4SP cells are indicated. For line graphs, the values represent the geometric mean of fluorescence intensity. The shaded graphs represent unstained thymocytes. (B) Absolute thymocyte number found in the different chimeras. Percentage (C) and number (D) of DP and CD4SP cells in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group. For clarity, the TSSP status of TEC and BM cells is indicated on the x-axis.

Close modal
FIGURE 3.

Impaired development of WE14-reactive BDC-10.1 TCR–Rg thymocytes in the absence of TSSP. (A) Chimeric mice combinations were performed, as described in Fig. 1, with the exception that BDC-10.1 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. Transduced thymocytes were electronically selected based on GFP expression (left panels). The CD4/CD8 distribution of GFP+ thymocytes is shown for a representative chimera of each group. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from four independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

Impaired development of WE14-reactive BDC-10.1 TCR–Rg thymocytes in the absence of TSSP. (A) Chimeric mice combinations were performed, as described in Fig. 1, with the exception that BDC-10.1 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. Transduced thymocytes were electronically selected based on GFP expression (left panels). The CD4/CD8 distribution of GFP+ thymocytes is shown for a representative chimera of each group. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from four independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal
FIGURE 6.

Impaired development of IAPP-reactive BDC-6.9 TCR–Rg thymocytes in mice lacking TSSP. (A) Chimeric mice combinations were performed as described in Fig. 3, with the exception that BDC-6.9 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. The CD4/CD8 distribution of GFP+ thymocytes is shown. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05, **p < 0.01, ****p < 0.0001.

FIGURE 6.

Impaired development of IAPP-reactive BDC-6.9 TCR–Rg thymocytes in mice lacking TSSP. (A) Chimeric mice combinations were performed as described in Fig. 3, with the exception that BDC-6.9 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. The CD4/CD8 distribution of GFP+ thymocytes is shown. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05, **p < 0.01, ****p < 0.0001.

Close modal
FIGURE 4.

TSSP-deficient hematopoietic cells induce the deletion of ChgA WE14-reactive BDC-10.1 TCR–Rg thymocytes. Chimeric mice combinations were performed as described in Fig. 3. (A) CD4/CD8 distribution of GFP+ thymocytes, CD5/CD69 distribution of DP thymocytes (gate I), and CD24/TCR Cβ profile of DP and CD4SP thymocytes (gate II) for one representative chimera of each group. Percentage of CD69+CD5+ DP thymocytes and the CD24 or TCR Cβ Geo Mean on immature CD24highTCRlow and mature CD24lowTCRhigh thymocytes in the different chimeras (B) Representative CD5 fluorescence intensity for immature and mature thymocytes. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05.

FIGURE 4.

TSSP-deficient hematopoietic cells induce the deletion of ChgA WE14-reactive BDC-10.1 TCR–Rg thymocytes. Chimeric mice combinations were performed as described in Fig. 3. (A) CD4/CD8 distribution of GFP+ thymocytes, CD5/CD69 distribution of DP thymocytes (gate I), and CD24/TCR Cβ profile of DP and CD4SP thymocytes (gate II) for one representative chimera of each group. Percentage of CD69+CD5+ DP thymocytes and the CD24 or TCR Cβ Geo Mean on immature CD24highTCRlow and mature CD24lowTCRhigh thymocytes in the different chimeras (B) Representative CD5 fluorescence intensity for immature and mature thymocytes. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05.

Close modal
FIGURE 5.

Normal development of IAPP-reactive BDC-5.2.9 TCR–Rg thymocytes in a thymic environment lacking TSSP. (A) Chimeric mice combinations were performed as described in Fig. 3, with the exception that BDC-5.2.9 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. The CD4/CD8 distribution of GFP+ thymocytes is shown. The representative line graphs represent the surface expression of the Rg TCR by DP and CD4SP cells. The percentages of GFP+, DP, and CD4SP cells, as well as TCR-Cβ+ cells, among CD4SP cells are shown. For line graphs, the geometric mean of fluorescence intensity (gMFI) of DP and CD4SP thymocytes is indicated. The shaded graphs represent unstained thymocytes. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05.

FIGURE 5.

Normal development of IAPP-reactive BDC-5.2.9 TCR–Rg thymocytes in a thymic environment lacking TSSP. (A) Chimeric mice combinations were performed as described in Fig. 3, with the exception that BDC-5.2.9 TCR retrovirally transduced LinRag1−/− NOD BM cells were used. The CD4/CD8 distribution of GFP+ thymocytes is shown. The representative line graphs represent the surface expression of the Rg TCR by DP and CD4SP cells. The percentages of GFP+, DP, and CD4SP cells, as well as TCR-Cβ+ cells, among CD4SP cells are shown. For line graphs, the geometric mean of fluorescence intensity (gMFI) of DP and CD4SP thymocytes is indicated. The shaded graphs represent unstained thymocytes. (B) Absolute thymocyte number in the different chimeras. Percentage (C) and number (D) of GFP+ DP and CD4SP thymocytes in the different chimeras. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05.

Close modal
FIGURE 7.

TSSP deficiency in TECs impacts positive selection of IAPP-reactive BDC-6.9 TCR–Rg thymocytes. Chimeric mice combinations were performed as described in Fig. 6. (A) CD4/CD8 distribution of GFP+ thymocytes, CD5/CD69 distribution of DP thymocytes (gate I), and CD24/TCR Cβ profile of DP and CD4SP thymocytes (gate II) are shown for one representative chimera of each group. Percentage of CD69+CD5+ DP thymocytes and the CD24 or TCR Cβ Geo Mean on immature CD24highTCRlow and mature CD24lowTCRhigh thymocytes in the different chimeras. (B) Representative CD5 and CD69 fluorescence intensity for immature and mature thymocytes. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05, **p < 0.01.

FIGURE 7.

TSSP deficiency in TECs impacts positive selection of IAPP-reactive BDC-6.9 TCR–Rg thymocytes. Chimeric mice combinations were performed as described in Fig. 6. (A) CD4/CD8 distribution of GFP+ thymocytes, CD5/CD69 distribution of DP thymocytes (gate I), and CD24/TCR Cβ profile of DP and CD4SP thymocytes (gate II) are shown for one representative chimera of each group. Percentage of CD69+CD5+ DP thymocytes and the CD24 or TCR Cβ Geo Mean on immature CD24highTCRlow and mature CD24lowTCRhigh thymocytes in the different chimeras. (B) Representative CD5 and CD69 fluorescence intensity for immature and mature thymocytes. Each symbol corresponds to an individual chimera; n = number of mice/group from three independent experiments. *p < 0.05, **p < 0.01.

Close modal

The mouse stem cell virus–based retroviral constructs encoding the TCRα- and β-chain cDNAs of BDC-10.1 and BDC-6.9, along with an internal ribosome entry site–GFP cassette, were described previously (23) and were a generous gift from Dr. D.A. Vignali (Immunology Department, St. Jude Children’s Research Hospital, Memphis, TN). The retroviral construct encoding the TCRα- and β-chain cDNAs of BDC-5.2.9 CD4 T cell clone was generated by RT-PCR cloning, as previously described (32). In brief, total RNAs from BDC-5.2.9 T cells were isolated and reverse transcribed using standard procedures. cDNAs were PCR amplified using standard TCRCα- and TCRCβ-specific primers along with a panel of Vα- and Vβ-specific primers to identify V-domain usage. TRAV2*01- and TRBV19*01-specific sense primers and TCRCα- and TCRCβ-specific anti-sense primers were engineered to introduce flanking restriction sites and a P2A sequence between the full-length α- and β-chain cDNAs in a two-step PCR using the Phusion DNA polymerase. The TCRα-P2A-TCRβ sequence was cloned into the mouse stem cell virus–driven retroviral vector cited above and confirmed by sequencing. The functionality and specificity of the retrovirally encoded TCRs were validated in vitro using retrovirally transduced BWαβ thymoma expressing CD4 (data not shown).

Rg mice were generated as previously described (33). In brief, the Phoenix Eco packaging line was transfected with the retroviral construct of interest using Lipofectamine 2000 and Opti-MEM I reagents (Invitrogen). BM cells were recovered from 6–14-wk-old Rag-1−/− NOD mice and enriched in lineage Ag-negative (Lin) cells using the mouse Lineage Cell Depletion Kit (Miltenyi Biotec), according to the manufacturer’s instructions. Lin BM cells were cultured in StemSpan medium (STEMCELL Technologies) supplemented with 10% FCS and recombinant murine cytokines: IL-3 (10 ng/ml), IL-6 (20 ng/ml), and stem cell factor (40 ng/ml) (PeproTech). The next day, the media were replaced with retrovirus-containing supernatant containing 20 μg/ml Polybrene (hexadimethrine bromide; Sigma) and cytokines prior to centrifugation at 1800 × g for 2 h. Another spin-transduction was performed the next day. Transduced BM cells were used on day 3 postinitial culture.

Mixed BM chimeras were generated by reconstitution of lethally irradiated (9.5 Gy) 8–12-wk-old Tssp° NOD-Cα° or NOD-Cα° mice with 1–2 × 106 retrovirally transduced Lin BM cells (Rg) or 5 × 106 T cell–depleted BM cells from sex-matched BDC-2.5 TCR–Tg mice, together with either NODscid or Tssp° NODscid BM cells at a Rg/Tg to scid BM ratio of 3:1. Mice were housed with antibiotics in their drinking water for 2 wk and analyzed 4–5 wk after reconstitution because of rapid disease occurrence (23). Rg mice with >10% GFP+ thymocytes at the time of analysis were considered reconstituted.

The CD4 T cell clones BDC-2.5, BDC-10.1, and BDC-5.2.9 (2 × 104) were stimulated with 2.5 × 104 peritoneal exudate cells pulsed with graded doses of peptides. Twenty-four hours later, supernatants were harvested, and IFN-γ concentrations were determined by specific ELISA. The β cell membrane is a granule-enriched membrane fraction obtained from β cell tumors as a source of Ag (34). The following synthetic peptides were used for the stimulation assay (GeneCust, >95% purity): pS3 (SRLGLWVRME), mBDC (RVLPVWVRME), WE14 (ChgA358-371: WSRMDQLAKELTAE), DD14 (ChgA29-42: DTKVMKCVLEVISD), SE14 (ChgA410-423: SREDSVEARSDFEE), and KS20 (KCNTATCATQRLANFLVRSS).

The unpaired two-tailed Mann–Whitney test was used for statistical analyses. Significant p values are indicated in the figures.

To examine the development of immature thymocytes expressing the BDC-2.5 TCR in the context of a thymic microenvironment lacking TSSP, we generated mixed BM chimeras in which either the thymic epithelial cell (TEC) or the APC compartment lacked TSSP expression. Hence, T cell–depleted BM cells from BDC-2.5 TCR–Tg NOD donors were coinjected with either NODscid or Tssp° NODscid BM cells as a source of TSSP-deficient hematopoietic cells, such as DCs, into lethally irradiated NOD-Cα° or Tssp° NOD-Cα° mice. A ratio of 3:1 NOD BDC-2.5 TCR BM cells and wild-type (WT) or Tssp° NODscid BM cells was used, given that <10% of Ag-loaded APCs is sufficient to induce complete deletion of Ag-specific thymocytes (35). The thymi of chimeric mice were analyzed by flow cytometry–coupled immunofluorescence 4 wk after reconstitution (Fig. 1A). The CD4/CD8 distribution profile and the TCR expression level on immature and mature BDC-2.5 TCR–Tg thymocytes were unchanged in the thymus in which both the thymic epithelium and BM-derived cells were lacking TSSP expression relative to control (Fig. 1A). Likewise, there was no significant difference between TSSP-deficient and control thymic microenvironments in terms of thymic cellularity (Fig. 1B) or the percentage and numbers of CD4+CD8+ immature (double-positive [DP]) and CD4+CD8 mature (CD4 single-positive [CD4SP]) BDC-2.5 TCR–Tg thymocytes (Fig. 1C, 1D). In agreement, reconstituted mice with TSSP deficiency restricted to either the radioresistant or radiosensitive thymic stromal cell compartment reflected the fact that neither TSSP-deficient epithelial cells nor TSSP-deficient BM-derived APCs significantly influenced the development of ChgA-reactive BDC-2.5 TCR–Tg thymocytes (Fig. 1). Likewise, although very low in WT chimeric mice, the development of regulatory T cells was not affected by TSSP deficiency in either thymic stromal cell compartment (data not shown).

Collectively, the results show that TSSP deficiency has no significant impact on the development of BDC-2.5 TCR–Tg thymocytes.

Both BDC-2.5 and BDC-10.1 TCRs react to the WE14 motif of ChgA (25). However, the fine specificity of these two TCRs appears to be distinct (Fig. 2) (25). Thus, both BDC-2.5 and BDC-10.1 reacted similarly to β cell membrane extracts, weakly to the WE14 peptide but not to the vasostatin-1–derived peptide ChgA29–42 or to the ChgA410–423, the only ChgA peptide eluted from I-Ag7 molecules expressed on Nit-1 insulinoma (Fig. 2A–D) (36). Both clones also responded strongly to the pS3 mimotope SRLGLWVRME, as previously reported (Fig. 2E) (25, 26). However, we found that the BDC-2.5 TCR, but not the BDC-10.1 TCR, responded to the BDC-2.5 mimotope (mBDC) RVLPVWVRME (Fig. 2F). In addition, the diabetogenic potential of CD4 T cells expressing the BDC-10.1–Rg TCR was higher than that of CD4 T cells expressing the BDC-2.5–Rg TCR, suggesting that the BDC-10.1 TCR may have a higher avidity for its natural ligand than does the BDC-2.5 TCR (23). Given these differences, we examined whether the selection of immature BDC-10.1 thymocytes is impacted by TSSP deficiency. To address this issue, we used the published TCR retrogenesis approach (32, 33, 37). Despite the fact that, for a given TCR specificity, the development of TCR-Rg thymocytes is not necessarily identical to that seen in TCR-Tg mice in terms of cellularity and frequency, TCR retrogenesis allows for the differentiation of fully functional T cells in the proper coreceptor lineage (12, 15, 23, 32, 33, 38, 39). Thus, Lin Rag-1−/− NOD BM cells were infected with viral particles encoding the α- and β-chains of the BDC-10.1 TCR, along with GFP. The transduced BM cells were used to generate mixed BM chimeras with either the TECs or APCs lacking TSSP, as described above. The analysis of 4-wk-old reconstituted mice revealed that TSSP deficiency in the thymic stromal compartment significantly altered the development of BDC-10.1 TCR–Rg thymocytes (Fig. 3). The total number of Rg BDC-10.1 thymocytes was reduced in the TSSP-deficient thymus (Rg + knockout [KO] → KO) compared with WT thymus (Rg + WT → WT); such a reduction affected both the DP and CD4SP subsets, indicating that, in the absence of TSSP, both immature and mature BDC-10.1 TCR–Rg thymocytes were subjected to deletion (Fig. 3B–D).

FIGURE 2.

The ChgA-reactive BDC-2.5 and BDC-10.1 TCRs have different fine specificities. The CD4 T cell clones BDC-2.5, BDC-10.1, and BDC-5.2.9 were stimulated with the indicated amount of β cell membrane extracts (A), ChgA-derived peptides WE14 (B), ChgA29–42 (C), ChgA410–423 (D), the pS3 peptide (E), the mBDC peptide (F), or the IAPP-derived KS20 peptide (G). IFN-γ production measured by ELISA is expressed as ng/ml. One representative experiment out of two performed is shown.

FIGURE 2.

The ChgA-reactive BDC-2.5 and BDC-10.1 TCRs have different fine specificities. The CD4 T cell clones BDC-2.5, BDC-10.1, and BDC-5.2.9 were stimulated with the indicated amount of β cell membrane extracts (A), ChgA-derived peptides WE14 (B), ChgA29–42 (C), ChgA410–423 (D), the pS3 peptide (E), the mBDC peptide (F), or the IAPP-derived KS20 peptide (G). IFN-γ production measured by ELISA is expressed as ng/ml. One representative experiment out of two performed is shown.

Close modal

Reconstituted mice with TSSP deficiency restricted to either the radioresistant or radiosensitive thymic stromal cell compartment revealed the relative contribution of these two tissues to the impaired development of BDC-10.1 thymocytes (Fig. 3). Mice with a TSSP-deficient BM cell compartment (Rg + KO → WT) displayed a thymic phenotype quite similar to that observed in mice lacking TSSP in all thymic stromal cells: both DP and CD4SP populations were reduced in number, suggesting that self-peptide:MHC class II complex presentation by TSSP-deficient thymic DCs induced the deletion of BDC-10.1 TCR–Rg thymocytes (Fig. 3D).

Reconstituted mice with TSSP deficiency restricted to the thymic epithelium (Rg + WT → KO) revealed no signs of negative selection of BDC-10.1 thymocytes (Fig. 3). However, relative to Rg + WT → WT control mice, the thymic cellularity was elevated as the result of more DP thymocytes without significant change in the CD4SP subset cellularity (Fig. 3B, 3D). These features could reflect a modulation of DP thymocyte homeostasis by TSSP-deficient TECs or reveal a mild negative selection induced by TSSP-sufficient TECs.

To confirm and gain insight into these observations, we examined the expression of CD24, CD5, and CD69 by developing thymocytes in the different chimeras. Downregulation of CD24 and upregulation of CD69 and CD5 are hallmarks of DP thymocytes undergoing positive selection. The expression level of the negative regulator of TCR signaling, CD5, is indeed determined by TCR signal intensity during thymocyte development (40). Consistent with normal positive selection, the frequency of CD69+CD5+ BDC-10.1 DP thymocytes was comparable in all chimeras, regardless of whether the mice expressed TSSP in a given stromal compartment (Fig. 4). Although statistical significance was not reached (p < 0.06), we found that immature (CD24highTCRlow) and mature (CD24lowTCRhigh) thymocytes developing in chimeras lacking TSSP expression in cTECs expressed overall higher levels of CD5 than did thymocytes developing in chimeras expressing TSSP, suggesting that TSSP-deficient cTECs may express increased cognate ligand density/affinity that may slightly favor the development or survival of BDC-10.1 thymocytes (Fig. 4B). Such a modified ligand presentation by epithelial cells could account for the augmented number of DP thymocytes seen in Rg + WT → KO chimeras (Fig. 3D). In chimeras lacking TSSP expression in DCs, only mature thymocytes showed an increase in CD5 expression, which, together with the observation that the number of CD4SP is reduced in these chimeras, suggests that TSSP-deficient thymic DCs induce the deletion of BDC-10.1 thymocytes (Fig. 4B). The fact that immature BDC-10.1 TCR–Rg thymocytes developing in Rg + KO → WT chimeras had an augmented level of CD24 confirms that TCR ligands expressed by TSSP-deficient BM cells impact the fate of this thymocyte subset, as suggested by the reduced number of DP cells seen in these chimeras (Figs. 3, 4A).

Altogether, data from BDC-10.1 TCR–Rg mice show that TSSP deficiency affects the development of BDC-10.1 TCR–Rg thymocytes, with primarily TSSP-deficient thymic DCs inducing a marked deletion of BDC-10.1 TCR–Rg thymocytes at all developmental stages.

Recent studies showed that the BDC-5.2.9 TCR reacts with the IAPP-derived peptide KCNTATCATQRLANFLVRSS (KS20) but not with any ChgA-derived peptides (Fig. 2, data not shown) (27). We cloned and sequenced the TCRα- and TCRβ-chain cDNAs of the BDC-5.2.9 T cell clone (Table I) and generated the corresponding retroviral construct that was functionally validated in vitro (data not shown).

Table I.
Amino acid sequence of the junctional region for the α- and β-chains of the BDC-2.5, BDC-10.1, BDC-6.9, and BDC-5.2.9 TCRs
α-Chain
TCRCDR3CDR3 aa (Net Charge)
BDC-2.5 CAA SLAGSWQL IFGSG TRAV7D-6*01 TRAJ22*01 8 (0) 
BDC-10.1 CAL EGHYGGSGNKL IFGTG TRAV17*01 TRAJ32*01 11 (+1) 
BDC-6.9 CAA SAVGYKL TFGTG TRAV5D-4*01 TRAJ9*01 7 (+1) 
BDC-5.2.9 CIV TASSGSWQL IFGSG TRAV2*01 TRAJ22*01 9 (0) 
β-Chain
TCRCDR3CDR3 aa (Net Charge)
BDC-2.5 CAS SQGGTTNSDY TFGSG TRBV2*01 TRBJ1-2*01 10 (−1) 
BDC-10.1 CGA RGSGDTQ YFGPG TRBV20*01 TRBJ2-5*01 7 (0) 
BDC-6.9 CAS SQGLGWAETL YFGSG TRBV2*01 TRBJ2-3*01 10 (−1) 
BDC-5.2.9 CAS SPPDSYAEQ FFGPG TRBV19*01 TRBJ2-1*01 9 (−2) 
α-Chain
TCRCDR3CDR3 aa (Net Charge)
BDC-2.5 CAA SLAGSWQL IFGSG TRAV7D-6*01 TRAJ22*01 8 (0) 
BDC-10.1 CAL EGHYGGSGNKL IFGTG TRAV17*01 TRAJ32*01 11 (+1) 
BDC-6.9 CAA SAVGYKL TFGTG TRAV5D-4*01 TRAJ9*01 7 (+1) 
BDC-5.2.9 CIV TASSGSWQL IFGSG TRAV2*01 TRAJ22*01 9 (0) 
β-Chain
TCRCDR3CDR3 aa (Net Charge)
BDC-2.5 CAS SQGGTTNSDY TFGSG TRBV2*01 TRBJ1-2*01 10 (−1) 
BDC-10.1 CGA RGSGDTQ YFGPG TRBV20*01 TRBJ2-5*01 7 (0) 
BDC-6.9 CAS SQGLGWAETL YFGSG TRBV2*01 TRBJ2-3*01 10 (−1) 
BDC-5.2.9 CAS SPPDSYAEQ FFGPG TRBV19*01 TRBJ2-1*01 9 (−2) 

The domain denomination used is from ImMunoGeneTics nomenclature. Note the distinctions between couple of TCRs reacting to both ChgA (BDC-2.5 versus BDC-10.1) and IAPP (BDC-6.9 versus BDC-5.2.9) in terms of CDR3 length and net charge. For instance, the CDR3α of the BDC-2.5 TCR is shorter and neutral relative to that of the BDC-10.1 TCR (8 versus 11 aa; 0 versus +1), and the CDR3β region is longer and is more acidic (10 versus 7 aa; −1 versus 0).

We next examined the requirement for TSSP expression by different thymic stromal cells in the intrathymic development of BDC-5.2.9 thymocytes using TCR retrogenesis, as described above. The CD4/CD8 profiles of thymocytes were comparable for the different chimeric mice analyzed, regardless of whether the thymic epithelium or BM-derived APCs expressed TSSP (Fig. 5A). Likewise, although there were some variations among the different mice analyzed, the total cellularity and total number of DP or CD4SP thymocytes expressing the BDC-5.2.9 TCR were comparable among chimeric mice (Fig. 5B–D). In addition, despite some variability among the different mice analyzed, the level of TCR expression by DP or CD4SP thymocytes was not significantly affected by the absence of TSSP expression by the different thymic stromal compartments (data not shown). Hence, lack of TSSP expression by either TECs or thymic DCs does not impact the development of BDC-5.2.9 TCR–expressing thymocytes.

Like BDC-5.2.9, the BDC-6.9 T cell clone responds to NOD-derived islet cells, but not to IAPP-deficient islet cells, suggesting that both clones recognize IAPP (27). In addition, genetic-mapping studies showed a linkage of the iapp gene with the Ag locus for the BDC-6.9 T cell clone (29). However, in contrast to the BDC-5.2.9 T cell clone, the BDC-6.9 T cell clone did not respond to IAPP or IAPP-derived peptides in vitro, suggesting differences in the fine specificity of these two clones (27).

Therefore, we examined whether TSSP deficiency had any impact on the development of thymocytes expressing the BDC-6.9 TCR through the TCR retrogenesis approach described above (Fig. 6). The analysis of Rg thymi revealed that the overall development of BDC-6.9 TCR–Rg thymocytes on the NOD control background (Rg + WT → WT) was less efficient than that of BDC-5.2.9 or BDC-10.1 thymocytes, as evidenced by the reduced thymic cellularity (compare Fig. 3B or 5B and Fig. 6B; p = 0.003 and 0.03, respectively).

In contrast to the IAPP-specific BDC-5.2.9 TCR, the development of BDC-6.9 thymocytes was altered in the TSSP-deficient thymic microenvironment (Rg + KO → KO). Thus, relative to control Rg + WT → WT TCR Rg thymi, the cellularity of Rg + KO → KO TCR Rg thymi was significantly increased, and this phenomenon correlated with more DP thymocytes (Fig. 6B, 6D). In addition, the percentage of CD4SP cells was significantly reduced in these chimeras compared with WT chimeras, indicating that the transition from DP to CD4SP was impaired in the TSSP-deficient thymic environment and further suggesting that, in the absence of TSSP, the positive selection of immature BDC-6.9 TCR–Rg thymocytes was impaired (Fig. 6B–D). In agreement with the notion of impaired positive selection, we found an increased cellularity of total and DP thymocytes and reduced CD4SP percentage in reconstituted host mice lacking TSSP expression only in TECs (Rg + WT → KO, Fig. 6B–D). Further compatible with impaired positive selection, the frequency of CD69+CD5+ DP thymocytes was reduced in these chimeras compared with control Rg + WT → WT (Fig. 7A), as was the case for the level of CD5 and CD69 expression by immature thymocytes (Fig. 7B).

In contrast, mice with TSSP deficiency restricted to the BM-derived cell compartment (Rg + KO → WT) showed no significant change in thymic cellularity or the number of DP or CD4SP subsets (Fig. 6). Furthermore, the frequency of CD69+CD5+ DP thymocytes and the expression level of CD69 and CD5 by immature and mature thymocytes were comparable in Rg + KO → WT and Rg + WT → WT TCR–Rg mice, further indicating that lack of TSSP expression by thymic DCs had no impact on the development of BDC-6.9–Rg thymocytes (Fig. 7).

Hence, expression of TSSP by the thymic epithelium appears mandatory for positive selection of IAPP-reactive immature BDC-6.9 TCR–Rg thymocytes.

We showed previously that TSSP is a critical regulator of diabetes pathogenesis through enhanced selection of the autoreactive CD4 T cell repertoire by naturally expressed endogenous Ags. In this article, we provide direct evidence that TSSP favors the development of highly diabetogenic CD4 T cells. Indeed, we found that the development of thymocytes expressing the MHC class II–restricted BDC-10.1 and BDC-6.9 TCRs that react to ChgA and IAPP autoantigens, respectively, was impaired when the thymic stromal cells lacked TSSP. Combined with our previous demonstration that TSSP expression was necessary for the development of thymocytes expressing the NY4.1 diabetogenic TCR or for thymocytes specific for the IA2β or S100β Ags (12, 41), these novel findings clearly show that TSSP has a substantial impact on the generation of the islet-reactive CD4 T cell repertoire.

Although the precise function of TSSP remains to be formally demonstrated, our data suggest that this protease has a subtle role in the MHC class II presentation pathway. With regard to TECs, expression of TSSP appears necessary for the positive selection of some, but not all, MHC class II–restricted TCRs. Hence, the positive selection of BDC-6.9, but not that of BDC-5.2.9, BDC-2.5, or BDC-10.1, immature thymocytes is impaired when TECs lack TSSP expression. Therefore, in TECs, TSSP appears to be mandatory for the generation of some positively selecting peptide(s), independently of invariant chain processing (14). In contrast, in thymic DCs, TSSP likely impairs the presentation of some self-ligands, thus preventing central tolerance. Hence, TSSP-deficient BM-derived cells, most certainly thymic DCs, caused a marked deletion in immature BDC-10.1 thymocytes (this study), as well as NY4.1-Tg thymocytes and IA2β- and S100β-specific thymocytes (12, 41). These observations suggest that, although expressed at a lower level in thymic DCs than in cTECs, TSSP substantially affects the assembly/presentation of MHC class II:peptide complexes by thymic DCs. Possibly, the Ag(s) involved in such a presentation may originate from neighboring epithelial cells (42, 43). Collectively, these results suggest that TSSP somehow edits the self-peptide repertoire presented by MHC class II molecules in the thymus.

The comparative analyses of the effect of TSSP deficiency on the development of two ChgA-reactive TCRs further highlight the subtle effect of TSSP on the MHC class II presentation pathway. Because both BDC-2.5 and BDC-10.1 TCRs react to the WE14 portion of ChgA (25, 30), it is remarkable that the development of BDC-10.1 TCR–Rg immature thymocytes is markedly sensitive to TSSP deficiency in thymic APCs, whereas that of BDC-2.5 TCR Tg immature thymocytes is not. One plausible explanation is that TSSP deficiency in BM-derived cells could lead to the neoexpression of self-peptide:MHC class II complex(es) at a level sufficient to induce deletion of BDC-10.1 thymocytes which is not seen by thymocytes expressing the BDC-2.5 TCR. Such a self-Ag is possibly ChgA itself because Chga transcripts are detected in the thymus (44), although we cannot formally exclude the involvement of another self-epitope. The BDC-2.5 and BDC-10.1 TCRs differ by a number of important molecular features. In particular, both the variable (V) and junction (J) domains differ between the two TCRs, and this is true for both the α- and β-chains. In addition, the CDR3 of these chains are quite distinct (Table I). Such distinctions could contribute to the differential reactivity of the two TCRs to several ChgA mimotopes [this study and (25)] and to the differential sensitivity to TSSP deficiency. Furthermore, although enzymatic modification of ChgA WE14 by transglutaminase significantly increased the response of both T cell clones, the BDC-2.5 clone appeared significantly more sensitive than the BDC-10.1 clone (30). Finally, in a Rg setting, the BDC-10.1 TCR showed higher reactivity to islet extracts and a stronger diabetogenic potential than did the BDC-2.5 TCR, suggesting that the BDC-10.1 TCR had a higher avidity for its natural peripheral ligand than did the BDC-2.5 TCR (23). Therefore, higher avidity of the BDC-10.1 TCR could favor deletion, especially if the density of the ligand presented by TSSP-deficient DCs is limited.

Along the same lines, we found that the two IAPP-specific TCRs, BDC-5.2.9 and BDC-6.9, are differentially impacted by TSSP deficiency. Hence, the positive selection of BDC-6.9, but not that of BDC-5.2.9, was impaired when TECs lacked TSSP expression. Unlike BDC-10.1 TCR–Rg mice, in which DP thymocyte numbers were reduced upon confrontation with TSSP-deficient BM cells, BDC-6.9 TCR–Rg mice showed more DP thymocytes in the TSSP-deficient thymic microenvironment due to inefficient positive selection on the TSSP mutant epithelium. In this case too, the BDC-5.2.9 and BDC-6.9 TCRs differ in terms of Vα-Jα and Vβ-Jβ gene segment usage and CDR3α/CDR3β-specific features (Table I). Furthermore, although both TCRs appear to recognize IAPP, only the BDC-5.2.9 clone responded to IAPP protein or the IAPP-derived peptide KS20 in vitro (27). Therefore, the specific features of both TCRs and fine specificity of positively selecting ligand(s) (45, 46) could explain the differential effect of TSSP deficiency on positive selection of these two TCRs.

The strong resistance of TSSP-deficient NOD mice to insulitis and diabetes development (12) suggests that islet-specific, MHC class II–restricted T cell specificities that are absent in TSSP-deficient NOD mice might play an important role in the early steps of T1D pathogenesis, perhaps by contributing to a substantial level of infiltration. Therefore, it was surprising to find that TSSP-deficient NOD mice responded normally to the two major insulin epitopes (InsB9-23 and InsB49-66), the GAD65-derived peptides GAD206–220 and GAD65524–543, and the three known epitopes of IGRP (IGRP4–22, IGRP123–145, and IGRP195–214) (12). In addition, we found that the development of thymocytes expressing Rg TCRs specific for InsB9-23 (Ins12-4-1) or GAD65206–220 (PA19-5E11) was unaffected by TSSP deficiency (C. Viret and S. Guerder, unpublished observations). However, we showed that thymocytes expressing the highly diabetogenic NY4.1 TCR and the IA2β-specific phogrin18 TCR were deleted in a thymic microenvironment lacking TSSP (12). Thus, the results obtained for BDC-10.1 and BDC-6.9 TCR–Rg thymocytes provide additional evidence that TSSP deficiency significantly affects development of the islet-reactive CD4 T cell repertoire. This substantial reduction in the frequency of autoreactive CD4 T cells may permit an efficient control of the activation/agressivity of the remaining pool of islet-reactive CD4 T cells by regulatory T cells. Although we cannot formally exclude that TSSP may be necessary for the development of CD4 T cells specific for an unknown essential islet Ag, it is possible that the protection from T1D in TSSP-deficient mice primarily results from this substantial reduction in the autoreactive CD4 T cell repertoire. This notion fits with a “cumulative Ag-targeting” model of diabetes pathogenesis and is compatible with the idea that T1D initiation/development involves multiple autoantigens (47). Further studies are clearly needed to understand the mechanisms by which TSSP deficiency prevents T1D.

Several studies supported the notion that the failure to censor autoreactive T cells, a characteristic of the NOD background, is due to multiple defects in the apoptotic pathway (4853). More recently, it was shown that genetic components of the NOD background impact the homeostasis of the DP thymocyte pool and that negative selection proceeds basically unperturbed in NOD mice (54). In addition to the issue of genetic susceptibility to negative selection, we provide evidence that the accumulation of self-reactive CD4 T cells in the NOD mouse is linked to altered presentation of self-Ag by thymic stromal cells due to the unique properties of TSSP.

We thank Martine Guiraud for technical assistance, Dr. D.A. Vignali for BDC-6.9 and BDC-10.1 TCR retroviral constructs, Dr. Benoit Salomon for BDC-2.5 TCR–Tg mice, and Drs. P. Romagnoli and J. van Meerwijk (Centre de Physiopathologie de Toulouse Purpan) for critical reading of the manuscript.

This work was supported in part by institutional grants from Institut National de la Santé et de la Recherche Médicale and Centre National de la Recherche Scientifique and by a grant from Agence Nationale de la Recherche (ANR-10-BLAN-1332).

Abbreviations used in this article:

BM

bone marrow

CD4SP

CD4 single positive

ChgA

chromogranin A

cTEC

cortical thymic epithelial cell

DC

dendritic cell

DP

double positive

GAD

glutamic acid decarboxylase

IAPP

islet amyloid polypeptide

IGRP

islet-specific glucose-6-phosphatase catalytic subunit-related protein

KO

knockout

Rg

retrogenic

T1D

type 1 diabetes

TEC

thymic epithelial cell

Tg

transgenic

TSSP

thymus-specific serine protease

WT

wild-type.

1
Tisch
R.
,
McDevitt
H.
.
1996
.
Insulin-dependent diabetes mellitus.
Cell
85
:
291
297
.
2
Wicker
L. S.
,
Todd
J. A.
,
Peterson
L. B.
.
1995
.
Genetic control of autoimmune diabetes in the NOD mouse.
Annu. Rev. Immunol.
13
:
179
200
.
3
Christianson
S. W.
,
Shultz
L. D.
,
Leiter
E. H.
.
1993
.
Adoptive transfer of diabetes into immunodeficient NOD-scid/scid mice. Relative contributions of CD4+ and CD8+ T-cells from diabetic versus prediabetic NOD.NON-Thy-1a donors.
Diabetes
42
:
44
55
.
4
Ikehara
S.
,
Tanaka
H.
,
Nakamura
T.
,
Furukawa
F.
,
Inoue
S.
,
Sekita
K.
,
Shimizu
J.
,
Hamashima
Y.
,
Good
R. A.
.
1985
.
The influence of thymic abnormalities on the development of autoimmune diseases.
Thymus
7
:
25
36
.
5
Like
A. A.
,
Butler
L.
,
Williams
R. M.
,
Appel
M. C.
,
Weringer
E. J.
,
Rossini
A. A.
.
1982
.
Spontaneous autoimmune diabetes mellitus in the BB rat.
Diabetes
31
(
Suppl. 1 Pt. 2
):
7
13
.
6
Whalen
B. J.
,
Greiner
D. L.
,
Mordes
J. P.
,
Rossini
A. A.
.
1994
.
Adoptive transfer of autoimmune diabetes mellitus to athymic rats: synergy of CD4+ and CD8+ T cells and prevention by RT6+ T cells.
J. Autoimmun.
7
:
819
831
.
7
DiLorenzo
T. P.
,
Graser
R. T.
,
Ono
T.
,
Christianson
G. J.
,
Chapman
H. D.
,
Roopenian
D. C.
,
Nathenson
S. G.
,
Serreze
D. V.
.
1998
.
Major histocompatibility complex class I-restricted T cells are required for all but the end stages of diabetes development in nonobese diabetic mice and use a prevalent T cell receptor alpha chain gene rearrangement.
Proc. Natl. Acad. Sci. USA
95
:
12538
12543
.
8
Groen
H.
,
Klatter
F.
,
Pater
J.
,
Nieuwenhuis
P.
,
Rozing
J.
.
2003
.
Temporary, but essential requirement of CD8+ T cells early in the pathogenesis of diabetes in BB rats as revealed by thymectomy and CD8 depletion.
Clin. Dev. Immunol.
10
:
141
151
.
9
Marrack
P.
,
Kappler
J. W.
.
2012
.
Do MHCII-presented neoantigens drive type 1 diabetes and other autoimmune diseases?
Cold Spring Harb. Perspect. Med.
2
:
a007765
.
10
Bowlus
C. L.
,
Ahn
J.
,
Chu
T.
,
Gruen
J. R.
.
1999
.
Cloning of a novel MHC-encoded serine peptidase highly expressed by cortical epithelial cells of the thymus.
Cell. Immunol.
196
:
80
86
.
11
Carrier
A.
,
Nguyen
C.
,
Victorero
G.
,
Granjeaud
S.
,
Rocha
D.
,
Bernard
K.
,
Miazek
A.
,
Ferrier
P.
,
Malissen
M.
,
Naquet
P.
, et al
.
1999
.
Differential gene expression in CD3epsilon- and RAG1-deficient thymuses: definition of a set of genes potentially involved in thymocyte maturation.
Immunogenetics
50
:
255
270
.
12
Viret
C.
,
Leung-Theung-Long
S.
,
Serre
L.
,
Lamare
C.
,
Vignali
D. A.
,
Malissen
B.
,
Carrier
A.
,
Guerder
S.
.
2011
.
Thymus-specific serine protease controls autoreactive CD4 T cell development and autoimmune diabetes in mice.
J. Clin. Invest.
121
:
1810
1821
.
13
Guerder
S.
,
Viret
C.
,
Luche
H.
,
Ardouin
L.
,
Malissen
B.
.
2012
.
Differential processing of self-antigens by subsets of thymic stromal cells.
Curr. Opin. Immunol.
24
:
99
104
.
14
Gommeaux
J.
,
Grégoire
C.
,
Nguessan
P.
,
Richelme
M.
,
Malissen
M.
,
Guerder
S.
,
Malissen
B.
,
Carrier
A.
.
2009
.
Thymus-specific serine protease regulates positive selection of a subset of CD4+ thymocytes.
Eur. J. Immunol.
39
:
956
964
.
15
Viret
C.
,
Lamare
C.
,
Guiraud
M.
,
Fazilleau
N.
,
Bour
A.
,
Malissen
B.
,
Carrier
A.
,
Guerder
S.
.
2011
.
Thymus-specific serine protease contributes to the diversification of the functional endogenous CD4 T cell receptor repertoire.
J. Exp. Med.
208
:
3
11
.
16
Lie
B. A.
,
Todd
J. A.
,
Pociot
F.
,
Nerup
J.
,
Akselsen
H. E.
,
Joner
G.
,
Dahl-Jørgensen
K.
,
Rønningen
K. S.
,
Thorsby
E.
,
Undlien
D. E.
.
1999
.
The predisposition to type 1 diabetes linked to the human leukocyte antigen complex includes at least one non-class II gene.
Am. J. Hum. Genet.
64
:
793
800
.
17
Lie
B. A.
,
Viken
M. K.
,
Akselsen
H. E.
,
Flåm
S. T.
,
Pociot
F.
,
Nerup
J.
,
Kockum
I.
,
Cambon-Thomsen
A.
,
Thorsby
E.
,
Undlien
D. E.
.
2007
.
Association analysis in type 1 diabetes of the PRSS16 gene encoding a thymus-specific serine protease.
Hum. Immunol.
68
:
592
598
.
18
Viken
M. K.
,
Blomhoff
A.
,
Olsson
M.
,
Akselsen
H. E.
,
Pociot
F.
,
Nerup
J.
,
Kockum
I.
,
Cambon-Thomsen
A.
,
Thorsby
E.
,
Undlien
D. E.
,
Lie
B. A.
.
2009
.
Reproducible association with type 1 diabetes in the extended class I region of the major histocompatibility complex.
Genes Immun.
10
:
323
333
.
19
Haskins
K.
,
Portas
M.
,
Bradley
B.
,
Wegmann
D.
,
Lafferty
K.
.
1988
.
T-lymphocyte clone specific for pancreatic islet antigen.
Diabetes
37
:
1444
1448
.
20
Haskins
K.
,
Portas
M.
,
Bergman
B.
,
Lafferty
K.
,
Bradley
B.
.
1989
.
Pancreatic islet-specific T-cell clones from nonobese diabetic mice.
Proc. Natl. Acad. Sci. USA
86
:
8000
8004
.
21
Haskins
K.
,
McDuffie
M.
.
1990
.
Acceleration of diabetes in young NOD mice with a CD4+ islet-specific T cell clone.
Science
249
:
1433
1436
.
22
Haskins
K.
2005
.
Pathogenic T-cell clones in autoimmune diabetes: more lessons from the NOD mouse.
Adv. Immunol.
87
:
123
162
.
23
Burton
A. R.
,
Vincent
E.
,
Arnold
P. Y.
,
Lennon
G. P.
,
Smeltzer
M.
,
Li
C. S.
,
Haskins
K.
,
Hutton
J.
,
Tisch
R. M.
,
Sercarz
E. E.
, et al
.
2008
.
On the pathogenicity of autoantigen-specific T-cell receptors.
Diabetes
57
:
1321
1330
.
24
Pauza
M. E.
,
Dobbs
C. M.
,
He
J.
,
Patterson
T.
,
Wagner
S.
,
Anobile
B. S.
,
Bradley
B. J.
,
Lo
D.
,
Haskins
K.
.
2004
.
T-cell receptor transgenic response to an endogenous polymorphic autoantigen determines susceptibility to diabetes.
Diabetes
53
:
978
988
.
25
Stadinski
B. D.
,
Delong
T.
,
Reisdorph
N.
,
Reisdorph
R.
,
Powell
R. L.
,
Armstrong
M.
,
Piganelli
J. D.
,
Barbour
G.
,
Bradley
B.
,
Crawford
F.
, et al
.
2010
.
Chromogranin A is an autoantigen in type 1 diabetes.
Nat. Immunol.
11
:
225
231
.
26
Nikoopour
E.
,
Sandrock
C.
,
Huszarik
K.
,
Krougly
O.
,
Lee-Chan
E.
,
Masteller
E. L.
,
Bluestone
J. A.
,
Singh
B.
.
2011
.
Cutting edge: vasostatin-1-derived peptide ChgA29-42 is an antigenic epitope of diabetogenic BDC2.5 T cells in nonobese diabetic mice.
J. Immunol.
186
:
3831
3835
.
27
Delong
T.
,
Baker
R. L.
,
Reisdorph
N.
,
Reisdorph
R.
,
Powell
R. L.
,
Armstrong
M.
,
Barbour
G.
,
Bradley
B.
,
Haskins
K.
.
2011
.
Islet amyloid polypeptide is a target antigen for diabetogenic CD4+ T cells.
Diabetes
60
:
2325
2330
.
28
Peterson
J. D.
,
Haskins
K.
.
1996
.
Transfer of diabetes in the NOD-scid mouse by CD4 T-cell clones. Differential requirement for CD8 T-cells.
Diabetes
45
:
328
336
.
29
Dallas-Pedretti
A.
,
McDuffie
M.
,
Haskins
K.
.
1995
.
A diabetes-associated T-cell autoantigen maps to a telomeric locus on mouse chromosome 6.
Proc. Natl. Acad. Sci. USA
92
:
1386
1390
.
30
Delong
T.
,
Baker
R. L.
,
He
J.
,
Barbour
G.
,
Bradley
B.
,
Haskins
K.
.
2012
.
Diabetogenic T-cell clones recognize an altered peptide of chromogranin A.
Diabetes
61
:
3239
3246
.
31
Katz
J. D.
,
Wang
B.
,
Haskins
K.
,
Benoist
C.
,
Mathis
D.
.
1993
.
Following a diabetogenic T cell from genesis through pathogenesis.
Cell
74
:
1089
1100
.
32
Holst
J.
,
Szymczak-Workman
A. L.
,
Vignali
K. M.
,
Burton
A. R.
,
Workman
C. J.
,
Vignali
D. A.
.
2006
.
Generation of T-cell receptor retrogenic mice.
Nat. Protoc.
1
:
406
417
.
33
Viret
C.
,
Cresson
C.
,
Broccardo
C.
,
Guerder
S.
.
2012
.
Retroviral transduction of lineage antigen-negative (Lin−) cells: a valuable alternative for the generation of T cell receptor (TCR) retrogenic mice.
J. Immunol. Methods
384
:
103
110
.
34
Bergman
B.
,
McManaman
J. L.
,
Haskins
K.
.
2000
.
Biochemical characterization of a beta cell membrane fraction antigenic for autoreactive T cell clones.
J. Autoimmun.
14
:
343
351
.
35
Merkenschlager
M.
,
Power
M. O.
,
Pircher
H.
,
Fisher
A. G.
.
1999
.
Intrathymic deletion of MHC class I-restricted cytotoxic T cell precursors by constitutive cross-presentation of exogenous antigen.
Eur. J. Immunol.
29
:
1477
1486
.
36
Suri
A.
,
Walters
J. J.
,
Rohrs
H. W.
,
Gross
M. L.
,
Unanue
E. R.
.
2008
.
First signature of islet beta-cell-derived naturally processed peptides selected by diabetogenic class II MHC molecules.
J. Immunol.
180
:
3849
3856
.
37
Holst
J.
,
Vignali
K. M.
,
Burton
A. R.
,
Vignali
D. A.
.
2006
.
Rapid analysis of T-cell selection in vivo using T cell-receptor retrogenic mice.
Nat. Methods
3
:
191
197
.
38
Arnold
P. Y.
,
Burton
A. R.
,
Vignali
D. A.
.
2004
.
Diabetes incidence is unaltered in glutamate decarboxylase 65-specific TCR retrogenic nonobese diabetic mice: generation by retroviral-mediated stem cell gene transfer.
J. Immunol.
173
:
3103
3111
.
39
Zhang
L.
,
Jasinski
J. M.
,
Kobayashi
M.
,
Davenport
B.
,
Johnson
K.
,
Davidson
H.
,
Nakayama
M.
,
Haskins
K.
,
Eisenbarth
G. S.
.
2009
.
Analysis of T cell receptor beta chains that combine with dominant conserved TRAV5D-4*04 anti-insulin B:9-23 alpha chains.
J. Autoimmun.
33
:
42
49
.
40
Azzam
H. S.
,
Grinberg
A.
,
Lui
K.
,
Shen
H.
,
Shores
E. W.
,
Love
P. E.
.
1998
.
CD5 expression is developmentally regulated by T cell receptor (TCR) signals and TCR avidity.
J. Exp. Med.
188
:
2301
2311
.
41
Serre
L.
,
Fazilleau
N.
,
Guerder
S.
.
2015
.
Central tolerance spares the private high-avidity CD4+ T-cell repertoire specific for an islet antigen in NOD mice.
Eur. J. Immunol.
45
:
1946
1956
.
42
Derbinski
J.
,
Schulte
A.
,
Kyewski
B.
,
Klein
L.
.
2001
.
Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self.
Nat. Immunol.
2
:
1032
1039
.
43
Koble
C.
,
Kyewski
B.
.
2009
.
The thymic medulla: a unique microenvironment for intercellular self-antigen transfer.
J. Exp. Med.
206
:
1505
1513
.
44
Derbinski
J.
,
Gäbler
J.
,
Brors
B.
,
Tierling
S.
,
Jonnakuty
S.
,
Hergenhahn
M.
,
Peltonen
L.
,
Walter
J.
,
Kyewski
B.
.
2005
.
Promiscuous gene expression in thymic epithelial cells is regulated at multiple levels.
J. Exp. Med.
202
:
33
45
.
45
Ebert
P. J.
,
Jiang
S.
,
Xie
J.
,
Li
Q. J.
,
Davis
M. M.
.
2009
.
An endogenous positively selecting peptide enhances mature T cell responses and becomes an autoantigen in the absence of microRNA miR-181a.
Nat. Immunol.
10
:
1162
1169
.
46
Lo
W. L.
,
Felix
N. J.
,
Walters
J. J.
,
Rohrs
H.
,
Gross
M. L.
,
Allen
P. M.
.
2009
.
An endogenous peptide positively selects and augments the activation and survival of peripheral CD4+ T cells.
Nat. Immunol.
10
:
1155
1161
.
47
DiLorenzo
T. P.
2011
.
Multiple antigens versus single major antigen in type 1 diabetes: arguing for multiple antigens.
Diabetes Metab. Res. Rev.
27
:
778
783
.
48
Lundholm
M.
,
Motta
V.
,
Löfgren-Burström
A.
,
Duarte
N.
,
Bergman
M. L.
,
Mayans
S.
,
Holmberg
D.
.
2006
.
Defective induction of CTLA-4 in the NOD mouse is controlled by the NOD allele of Idd3/IL-2 and a novel locus (Ctex) telomeric on chromosome 1.
Diabetes
55
:
538
544
.
49
Motta
V.
,
Lejon
K.
,
Holmberg
D.
.
2007
.
The NOD allele of the Idd5 locus on chromosome 1 mediates a non-cell-autonomous defect in negative selection of T cells.
J. Autoimmun.
28
:
216
223
.
50
Zucchelli
S.
,
Holler
P.
,
Yamagata
T.
,
Roy
M.
,
Benoist
C.
,
Mathis
D.
.
2005
.
Defective central tolerance induction in NOD mice: genomics and genetics.
Immunity
22
:
385
396
.
51
Kishimoto
H.
,
Sprent
J.
.
2001
.
A defect in central tolerance in NOD mice.
Nat. Immunol.
2
:
1025
1031
.
52
Lesage
S.
,
Hartley
S. B.
,
Akkaraju
S.
,
Wilson
J.
,
Townsend
M.
,
Goodnow
C. C.
.
2002
.
Failure to censor forbidden clones of CD4 T cells in autoimmune diabetes.
J. Exp. Med.
196
:
1175
1188
.
53
Liston
A.
,
Lesage
S.
,
Gray
D. H.
,
O’Reilly
L. A.
,
Strasser
A.
,
Fahrer
A. M.
,
Boyd
R. L.
,
Wilson
J.
,
Baxter
A. G.
,
Gallo
E. M.
, et al
.
2004
.
Generalized resistance to thymic deletion in the NOD mouse; a polygenic trait characterized by defective induction of Bim.
Immunity
21
:
817
830
.
54
Mingueneau
M.
,
Jiang
W.
,
Feuerer
M.
,
Mathis
D.
,
Benoist
C.
.
2012
.
Thymic negative selection is functional in NOD mice.
J. Exp. Med.
209
:
623
637
.

The authors have no financial conflicts of interest.