Immunopeptidome mining reveals a novel ERS-induced target in T1D

Abstract

Autoreactive CD8⁺ T cells play a key role in type 1 diabetes (T1D), but the antigen spectrum that activates autoreactive CD8⁺ T cells remains unclear. Endoplasmic reticulum stress (ERS) has been implicated in β-cell autoantigen generation. Here, we analyzed the major histocompatibility complex class I (MHC-I)-associated immunopeptidome (MIP) of islet β-cells under steady and ERS conditions and found that ERS reshaped the MIP of β-cells and promoted the MHC-I presentation of a panel of conventional self-peptides. Among them, OTUB2_58-66 showed immunodominance, and the corresponding autoreactive CD8⁺ T cells were diabetogenic in nonobese diabetic (NOD) mice. High glucose intake upregulated pancreatic OTUB2 expression and amplified the OTUB2_58-66-specific CD8⁺ T-cell response in NOD mice. Repeated OTUB2_58-66 administration significantly reduced the incidence of T1D in NOD mice. Interestingly, peripheral blood mononuclear cells (PBMCs) from patients with T1D, but not from healthy controls, showed a positive IFN-γ response to human OTUB2 peptides. This study provides not only a new explanation for the role of ERS in promoting β-cell-targeted autoimmunity but also a potential target for the prevention and treatment of T1D. The data are available via ProteomeXchange with the identifier PXD041227.

Keywords: CD8+ T cells; ERS; OTUB2; T1D; immunopeptidome.

Fig. 1

High glucose (HG) and thapsigargin (TG) treatment reshapes the NIT-1 cell-derived MHC-I-associated immunopeptidome (MIP). A The workflow for identifying NIT-1 cell-derived MIPs. B The total numbers of H2-Kd-restricted peptides and their potential source proteins in the MIP derived from nonstressed NIT-1 cells (NIT-1-NC) and ER-stressed NIT-1 cells treated with HG (NIT-1-HG) or TG (NIT-1-TG). C Length distributions of MIPs derived from NIT-1-NC, NIT-1-HG and NIT-1-TG cells (8 ~11 amino acids). D The predicted affinities of peptides by NetMHC-4.0, SB: strong binder (%Rank<0.5); WB: weak binder (%Rank:0.5 ~ 2). E The binding motifs of the 9-mer peptides identified in the MIP. The x-axis represents the residue position within the 9-mer peptide sequence. The y-axis represents the information content, with the size of each amino acid symbol proportional to its frequency. F Overlap of H-2Kd-restricted MIPs between NIT-1-NC and NIT-1-HG cells and between NIT-1-NC and NIT-1-TG cells at the source protein level. G Overlap of the source proteins for H-2Kd-restricted peptides exclusively present in the MIPs of NIT-1-HG and NIT-1-TG cells

Fig. 2

OTUB2_58-66 is an immunodominant self-peptide in nonobese diabetic (NOD) mice. A The workflow for the recall IFN-γ ELISPOT assay. B Representative ELISPOT images showing IFN-γ production by the indicated peptide-primed NOD mouse CD4^- splenocytes restimulated with DCs alone in the presence of an isotype Ab (DC + no peptide + isotype Ab), the indicated peptide-pulsed DCs in the presence of an isotype Ab (DC + indicated peptide + isotype Ab) or the indicated peptide-pulsed DCs in the presence of an anti-H-2Kd antibody (DC + indicated peptide + anti-H-2Kd Ab). C Summary data for the assay in (B) showing the mean IFN-γ spot number per 2 × 10⁵ CD4^- splenocytes ± SD in duplicate cultures. The data are one representative result of eight independent experiments. *p < 0.05, **p < 0.01, determined by Student’s t test. D The reactivity of CD4^- splenocytes from NOD mice harvested at different ages to the indicated peptides, as determined by the recall IFN-γ ELISPOT assay. Readouts are expressed as the mean spot numbers/2 × 10⁵ CD4^- splenocytes in duplicate cultures. The average readout from CD4^- splenocytes restimulated with DCs alone was set as the basal response. A positive response against the indicated peptide was defined as readouts above basal + 3 SD (underlined); ND indicates no data

Fig. 3

Characterization of OTUB2_58-66-specific CD8⁺ T cells in nonobese diabetic (NOD) mice. A Representative flow cytometry plots of CD8⁺dextramer⁺ T cells in the spleens, pancreases and pancreatic draining lymph nodes (pLNs) of NOD mice. B Representative flow cytometry histograms showing carboxyfluorescein succinimidyl ester (CFSE) dilution in NOD mouse splenic CD8⁺ T cells stimulated with or without OTUB2_58-66 (50 μg/mL). C Summary data for the assay in (B) are presented as the mean proliferation rate ±SD in triplicate for one representative result of three independent experiments. D Representative flow cytometry plots of peptide-loaded or unloaded NOD mouse splenocytes labeled with CFSE and incubated for 4 h in the presence (target + effector) or absence (target control) of NOD mouse CD8⁺ T cells (effector/target ratio, 100:1). Population 1 contained fluorescent microspheres that could be discriminated on the basis of their scattering pattern and fluorescence. Population 2 contained CFSE-negative cells (effector T cells). Population 3 contained CFSE-positive cells (target cells). E Summary data for the assay in (D) are presented as the mean killing rate ± SD in triplicate for one representative result of three independent experiments. **P < 0.01, ***p < 0.001, determined by Student’s t test

Fig. 4

OTUB2_58-66 peptide-specific CD8⁺ T cells are diabetogenic in nonobese diabetic/severe combined immunodeficiency (NOD/scid) mice. A The workflow for the adoptive transfer experiment. B Representative images of HE staining of the pancreas of NOD/scid mice receiving OTUB2_58-66-activated or control splenocytes (magnification, 200×). C Insulitis scores of NOD/scid recipient mice receiving OTUB2_58-66-activated or control splenocytes. D T1D incidence curve of NOD/scid mice adoptively transferred with OTUB2_58-66-activated or control splenocytes. Significance was determined by the log-rank (Mantel-Cox) test, *P < 0.05. E Representative laser confocal fluorescence microscopy images of pancreatic sections from NOD/scid mice stained for insulin (red) and CD8 (green). The cells were counterstained with the nuclear dye DAPI (blue). F Summary data for the assay in (E) are representative of two independent experiments and are presented as the mean ± SD, n = 5, **p < 0.01, as determined by Student’s t test

Fig. 5

Sustained high glucose intake increases the peptide-specific CD8⁺ T-cell response against OTUB2_58-66 in nonobese diabetic (NOD) mice. A Summary data showing the absolute cell number of pancreas-infiltrating lymphocytes in 6- and 12-week-old NOD mice fed high-glucose water (HG) or normal water (NC) (n = 4); *p < 0.05, determined by Student’s t test. B Representative flow cytometry plots of CD8⁺ dextramer⁺ T cells in the pancreas of 6-week-old NOD mice fed high-glucose water (HG) or normal water (NC). C Summary data for the assay in B are representative of two independent experiments and are presented as the mean ± SD (n = 4). D Representative flow cytometry plots showing the production of IFN-γ and TNF-α by NOD mouse-derived splenic CD8⁺ T cells stimulated with phorbol-12-myristate-13-acetate (PMA, 50 ng/mL) plus ionomycin (1 μg/mL). E Summary data for the assay in D are representative of two independent experiments and are presented as the mean ± SD (n = 4). F Summary data showing the proportions of IFN-γ- and TNF-α-producing cells in 6- and 12-week-old NOD mouse splenocytes stimulated with or without the indicated peptide (50 μg/mL). The data are expressed as the mean ± SD, n = 4

Fig. 6

High glucose treatment increases OTUB2 expression in both the pancreases of nonobese diabetic (NOD) mice and NIT-1 cells. A Representative image of immunohistochemical analysis of OTUB2 expression in pancreatic sections from NOD mice. B, C Western blot analysis of OTUB2 expression in the pancreas of NOD mice given normal water (NC-NOD) or high-glucose water (HG-NOD). D, E Western blot analysis of the protein level of OTUB2 in untreated NIT-1 cells (NIT-1-NC) and NIT-1 cells treated with HG (NIT-1-HG) or TG (NIT-1-TG). F Quantitative PCR analysis of relative OTUB2 mRNA expression in untreated NIT-1 cells (control), NIT-1 cells treated with HG or TG, and NIT-1 cells treated with HG or TG in the presence of tauroursodeoxycholic acid (TUDCA). The data are representative of three independent experiments and are expressed as the mean ± SD. *P < 0.05, **P < 0.01, ***p < 0.001, ****p < 0.0001, as determined by Student’s t test

Fig. 7

Repeated vaccination with the OTUB2_58-66 peptide prevents type 1 diabetes (T1D) onset in nonobese diabetic (NOD) mice. A The workflow for the peptide vaccination experiment. B Incidence of diabetes in female NOD mice injected with the OTUB2_58-66 peptide plus poly IC (red line, n = 14) or with poly IC alone (black line, n = 14). *p < 0.05 compared by the log-rank (Mantel-Cox) test. C Summary data showing the absolute number of pancreas-infiltrating lymphocytes in NOD mice after 2 or 3 injections of the OTUB2_58-66 peptide mixed with poly IC (OTUB2_58-66) or poly IC alone (control). The data are representative of two independent experiments and are presented as the mean ± SD (n = 4 per group). D Representative flow cytometry plots of CD8⁺dextramer⁺ T cells in the pancreas of NOD mice after 2 or 3 injections of the OTUB2_58-66 peptide plus poly IC (OTUB2_58-66) or poly IC alone (control). E Summary data for the assay in (D) showing the frequency of CD8⁺ dextramer⁺ T cells in the pancreas of NOD mice after 2 or 3 injections. The data are representative of two independent experiments and are presented as the mean ± SD (n = 4 per group). F Representative flow cytometry plots depicting PMA- and OTUB2_58-66-stimulated IFN-γ and TNF-α production by splenic CD8⁺ T cells from NOD mice after 2 or 3 injections of the OTUB2_58-66 peptide plus poly IC (OTUB2_58-66) or poly IC alone (control). G Statistical summary of the flow cytometry data shown in F. The data are representative of two independent experiments and are presented as the mean ± SD (n = 4 per group). *p < 0.05, **p < 0.01, ***p < 0.001, as determined by Student’s t test

Fig. 8

Autoreactive T-cell responses to human OTUB2 peptides in patients with type 1 diabetes (T1D). A Representative ELISPOT images showing IFN-γ production by the peripheral blood mononuclear cells (PBMCs) of HLA-A2⁺ patients with T1D and healthy controls stimulated with HLA-A*0201-restricted OTUB2 peptide pool 1 and pool 2 (50 μg/mL of each peptide), IGRP_215-233 (50 μg/mL), phytohemagglutinin (PHA, 2.5 μg/mL) or medium alone. The number represents the number of IFN-γ spots in each well. B Summary data for the IFN-γ ELISPOT assay in A. Each dot shows the stimulation index (Sl) of triplicate wells for each subject. The horizontal dashed line represents the cutoff of positivity (SI ≥ 3). The data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, determined by the Mann‒Whitney U test