Myelin-specific T cells in animals with Japanese macaque encephalomyelitis

Abstract

Objective: To determine whether animals with Japanese macaque encephalomyelitis (JME), a spontaneous demyelinating disease similar to multiple sclerosis (MS), harbor myelin-specific T cells in their central nervous system (CNS) and periphery.

Methods: Mononuclear cells (MNCs) from CNS lesions, cervical lymph nodes (LNs) and peripheral blood of Japanese macaques (JMs) with JME, and cervical LN and blood MNCs from healthy controls or animals with non-JME conditions were analyzed for the presence of myelin-specific T cells and changes in interleukin 17 (IL-17) and interferon gamma (IFNγ) expression.

Results: Demyelinating JME lesions contained CD4⁺ T cells and CD8⁺ T cells specific to myelin oligodendrocyte glycoprotein (MOG), myelin basic protein (MBP), and/or proteolipid protein (PLP). CD8⁺ T-cell responses were absent in JME peripheral blood, and in age- and sex-matched controls. However, CD4⁺ Th1 and Th17 responses were detected in JME peripheral blood versus controls. Cervical LN MNCs from eight of nine JME animals had CD3⁺ T cells specific for MOG, MBP, and PLP that were not detected in controls. Mapping myelin epitopes revealed a heterogeneity in responses among JME animals. Comparison of myelin antigen sequences with those of JM rhadinovirus (JMRV), which is found in JME lesions, identified six viral open reading frames (ORFs) with similarities to myelin antigen sequences. Overlapping peptides to these JMRV ORFs did not induce IFNγ responses.

Interpretations: JME possesses an immune-mediated component that involves both CD4⁺ and CD8⁺ T cells specific for myelin antigens. JME may shed new light on inflammatory demyelinating disease pathogenesis linked to gamma-herpesvirus infection.

Figure 1

Analysis of CNS mononuclear cell (MNC) infiltrates and periphery blood mononuclear cells (PBMCs) of JME animals. (A) Flow cytometry was used to analyze T‐cell populations infiltrating CNS lesions isolated from five JME animals (JMs 34027, 29756, 31869, 34478, and 34421) that were stimulated with pools of sequential 15 amino acid (aa) peptides, overlapping by 11 aa, representing macaque myelin oligodendrocyte glycoprotein (MOG), macaque myelin basic protein (MBP), and macaque proteolipid protein (PLP) and assayed by intracellular cytokine staining (ICS). A representative gating strategy that was utilized for all animals is shown for JM 31869. Antigen‐stimulated and antibody‐stained CNS‐MNCs positive for IFNγ or IL‐17 were determined by first removing CD45‐negative cell populations and then selecting for CD3⁺ cells. This CD45⁺ CD3⁺ cell population was further gated for lymphocytes, followed by gating for singlets, and then live cell populations. Within this live cell population, gates were drawn for CD4⁺ or CD8⁺ single positive cells expressing CD69 and then analyzed for cells expressing IFNγ or IL‐17. (B) Graphical representation of the ICS analysis of CNS‐MNCs from the five JME animals based upon Th1 (percent of IFNγ ⁺CD4⁺CD69⁺), Th17 (percent of IL‐17A⁺CD4⁺CD69⁺), CTL (percent of IFNγ ⁺CD8⁺CD69⁺), and Tc17 (percent of IL‐17A⁺CD8⁺CD69⁺). (C) PBMCs from the nine JME animals and 11 age‐ and sex‐matched healthy controls (HC) were stimulated with the overlapping peptide pools to MOG, MBP, or PLP, assayed by ICS, and analyzed by FCM. P values of less than 0.05 were considered significant. NS indicates the differences between groups were greater than 0.05 and non‐significant.

Figure 2

IFNγ ELISpot measurements of immune responses against MOG, MBP, and PLP. (A) Cervical lymph node (CV LN) MNCs from the nine JME animals and four control JM that were euthanized for non‐neurological conditions were screened for responses against pools of peptides representing MOG, MBP, and PLP. Representative images of IFNγ‐ELISpot assays from JME animals 34027, 34361, 34421, and 35604 stimulated with Staphylococcal enterotoxin B (SEB) as positive control, DMSO and media alone as negative controls for background determination, or pools of specific MOG, MBP, and PLP peptides. ELISpot assays were performed in duplicate with frozen CV LN MNCs that were thawed and revived in RPMI media at 37°C for 1 h prior to stimulation with the peptides. Wells were read by the AID ELISpot reader for spot‐forming cells (SFC) and further analyzed as described in Material with Subjects and Methods. Some images are weak and barely visible, and were not enhanced for reproduction. (B) Graphical representation of the IFNγ‐ELISpot analysis from the JME and C animals.

Figure 3

IFNγ ELISpot measurements of individual animal immune responses against reactive MOG‐, MBP‐, and PLP‐specific peptides. CV LN MNCs from JME animals were screened against individual peptides from MOG, MBP, and/or PLP pools that yielded SPC‐OB from IFNγ‐ELISpot assays described in Figure 2. As controls, CV LN MNCs were incubated with SEB, media, or DMSO alone. Peptides that were insoluble were not included. (A) Shown are the representative images from animals 34421 and 34361 against specific peptides performed in duplicate. (B) Graphical representation of the IFNγ‐ELISpot analysis from animals 34421, 34361, and 34027 that were found to be reactive to individual myelin peptides. Shown are the average of the duplicate wells with background SFC from media alone or DMSO not subtracted.

Figure 4

Sequences within macaque MOG, MBP, and PLP that are targets of JM T cells. (A) Shown are the primary amino acid (aa) sequences for macaque MOG, MBP, and PLP. Amino acid sequences that were identified as targets by JM CVLN T cells in our ELISpot assays are highlighted in red and those that are shared with MS patient T cells are underlined. Amino acid sequences reported to be targeted by T cells from MS patients are in bold. (B) Peptide sequences from JMRV ORFs with aa sequence homologies were identified by in silico analysis. Peptides, 15‐mers overlapping by 11 aa, were synthesized for each region with homology to myelin peptide (Table 3). Each peptide was evaluated by ELISpot analysis to determine if CVLN MNC from JME animals 34027, 34361, and 34421 exhibited SFC over background in duplicate. CV LN MNCs from three JMRV seropositive age‐ and gender‐matched control animals 33720, 34037, and 38114 are included. Shown are the average of the duplicate wells with background SFC from media alone or DMSO not subtracted.