Hla-DR2-restricted responses to proteolipid protein 95-116 peptide cause autoimmune encephalitis in transgenic mice

Abstract

In multiple sclerosis (MS) patients who carry the Class II major histocompatibility (MHC) type HLA-DR2, T cells specific for amino acids 95-116 in the proteolipid protein (PLP) are activated and clonally expanded. However, it remains unclear whether these autoreactive T cells play a pathogenic role or, rather, protect against the central nervous system (CNS) damage. We have addressed this issue, using mice transgenic for the human MHC class II region carrying the HLA-DR2 (DRB1* 1502) haplotype. After stimulating cultured lymph node cells repeatedly with PLP95-116, we generated 2 HLA-DR2-restricted, PLP95-116-specific T-cell lines (TCLs) from the transgenic mice immunized with this portion of PLP. The TCLs were CD4+ and produced T-helper 1 (Th1) cytokines in response to the peptide. These TCLs were adoptively transferred into RAG-2/2 mice expressing HLA-DR2 (DRG1* 1502) molecules. Mice receiving 1 of the TCLs developed a neurological disorder manifested ataxic movement without apparent paresis on day 3, 4, or 5 after cell transfer. Histological examination revealed inflammatory foci primarily restricted to the cerebrum and cerebellum, in association with scattered demyelinating lesions in the deep cerebral cortex. These results support a pathogenic role for PLP95-116-specific T cells in HLA-DR2+ MS patients, and shed light on the possible correlation between autoimmune target epitope and disease phenotype in human CNS autoimmune diseases.

Figure 1

Expression of DRα/I-Eβ^b and HLA-DR2 molecules in HLA-DR2–B6 mice. (a) Splenocytes from B6, DRA-B6, DRB1*1502–B6, and HLA-DR2–B6 mice were stained with PE-labeled anti–HLA-DRα (L243) or FITC-labeled anti–HLA-DRβ (TÜ36) mAb. The bars marked M1 indicate the population that stained significantly with the mAb, whereas M2 shows the positive population selected on more strict criteria. (b) Splenocytes from HLA-DR2–B6 mice were doubly stained with PE-conjugated anti–HLA-DRα (L243) and FITC-conjugated anti–HLA-DRβ (TÜ36) mAbs.

Figure 2

Proliferative response of TCLs B3-4 and B3-23 generated from HLA-DR2–B6 mice immunized with PLP95-116. Peptide-specific proliferation of TCLs B3-4 (a) and B3-23 (b) were evaluated as described in Methods. By using splenocytes from HLA-DR2–B6 (HLA-DR2⁺, DRα/I-Eβ^b+, I-A^b+) or DRA-B6 (DRα/I-Eβ^b+, I-A^b+) mice as APCs, T-cell proliferative response to control peptide MBP143-168 or PLP95-116 (25 μg/mL) was measured. Blocking effects of anti–HLA-DR mAb (G46-6) or isotype-matched control IgG2a (G155-178) were also examined as indicated. Data represent mean ± SD of the cpm obtained by triplicate assays in 4 independent experiments.

Figure 3

Cytokine production profile of TCLs B3-4 and B3-23. TCLs B3-4 and B3-23 were cultured for 48 hours with or without PLP95-116 in the presence of syngeneic splenocytes, and the culture supernatants were collected. Shown are the levels of IFN-γ, IL-2, TNF-α, and IL-4 detected in the supernatants of PLP95-116–stimulated cultures (no cytokine could be detected in the supernatants of the cultures without PLP95-116). Data represent mean ± SD of the values obtained by duplicate assays in 3 independent experiments.

Figure 4

IFN-γ production by TCLs B3-4 and B3-23 in the presence of human PBMCs. TCLs B3-4 (a) and B3-23 (b) were cultured for 48 hours with or without PLP95-116 in the presence of human PBMCs from subject MN (HLA-DR2 [DRB1*1502]/4) or YM ([HLA-DR1]/8), and the supernatants were collected for measurement of IFN-γ. Blocking effects of anti–HLA-DR mAb (G46-6) or isotype-matched control IgG2a (G155-178) were also examined as indicated. Data represent mean ± SD of the values obtained by duplicate assays in 4 independent experiments.

Figure 5

Flow cytometric detection of intracellular IFN-γ after stimulation with PLP95-116. Intracellular IFN-γ synthesis of TCLs responding to PLP95-116 was examined using different sources of APCs: syngeneic splenocytes from HLA-DR2–B6 mice, or human PBMCs (from subject MN expressing HLA-DR2 [DRB1*1502]/4). The line cells B3-4 (a) and B3-23 (b) were incubated for 48 hours in the presence of the APCs with (+) or without (–) PLP95-116 (25 μg/mL) and rhIL-2 (25 μg/mL). GolgiPlug^® (brefeldin A) was added for the last 10 hours of incubation, and the cells were stained for intracellular IFN-γ as described in Methods. Only the CD4⁺ T-cell population was gated for analysis, and the frequency (%) of the IFN-γ⁺ population is shown (above bars). Note that exogenous rhIL-2 markedly enhanced IFN-γ production in the presence of the human APCs.

Figure 6

Histology of CNS sections from moribund HLA-DR2/RAG-2^–/– mice receiving TCL B3-23. Shown are representative inflammatory foci in the cerebrum and cerebellum from mice receiving TCL B3-23. (a–d and h) H&E staining. (e, f, and g) Luxol fast blue staining. Bars = 100 μm. Higher magnifications of the boxed areas B, H, F, and G are illustrated in b, h, f, and g, respectively. The inflammatory lesions include perivascular infiltrates in the cerebral cortex (g) and periventricular white matter (c) as well as inflammatory infiltrates within the ventricle (c) and cerebellar meninges (d). These infiltrates were primarily composed of mononuclear cells, but neutrophils were also identified (h). Some of the inflammatory foci located in the deep cerebral cortex (a and e) were accompanied by demyelinating lesions (b and f).