Citrullinated human and murine MOG35-55 display distinct biophysical and biochemical behavior

Abstract

The peptide spanning residues 35 to 55 of the protein myelin oligodendrocyte glycoprotein (MOG) has been studied extensively in its role as a key autoantigen in the neuroinflammatory autoimmune disease multiple sclerosis. Rodents and nonhuman primate species immunized with this peptide develop a neuroinflammatory condition called experimental autoimmune encephalomyelitis, often used as a model for multiple sclerosis. Over the last decade, the role of citrullination of this antigen in the disease onset and progression has come under increased scrutiny. We recently reported on the ability of these citrullinated MOG35-55 peptides to aggregate in an amyloid-like fashion, suggesting a new potential pathogenic mechanism underlying this disease. The immunodominant region of MOG is highly conserved between species, with the only difference between the murine and human protein, a polymorphism on position 42, which is serine in mice and proline for humans. Here, we show that the biophysical and biochemical behavior we previously observed for citrullinated murine MOG35-55 is fundamentally different for human and mouse MOG35-55. The citrullinated human peptides do not show amyloid-like behavior under the conditions where the murine peptides do. Moreover, we tested the ability of these peptides to stimulate lymphocytes derived from MOG immunized marmoset monkeys. While the citrullinated murine peptides did not produce a proliferative response, one of the citrullinated human peptides did. We postulate that this unexpected difference is caused by disparate antigen processing. Taken together, our results suggest that further study on the role of citrullination in MOG-induced experimental autoimmune encephalomyelitis is necessary.

Keywords: amyloid; animal model; antigen processing; citrullination; multiple sclerosis; peptide chemical synthesis.

Figure 1

Comparison of the MOG_35-55sequence of different species and the effect of the amino acid polymorphism on the citrullination-induced amyloid-like aggregation.A, amino acid sequence of the extracellular part of myelin oligodendrocyte glycoprotein (MOG) for human (hMOG), common marmoset (Callithrix jacchus) (cjMOG), mouse (mMOG), and rat (rMOG). Residues that are mutated compared with the human protein are marked red, whereas minor mutations are marked in blue. The immunodominant portion, MOG_35–55, is underlined. B–G, thioflavin T (ThT) aggregation assays showing the aggregation properties of hMOG_35–55 (P1–3) and mMOG_35–55 (P4–6) and their citrullinated variants at pH 5.0 (20 mM NaOAc buffer) and 37 °C.

Figure 2

Thioflavin T (ThT) aggregation assays showing the aggregation properties of MOG_35–55-Cit_41,46and MOG_35–55-Cit_46,52. When Ser₄₂ is replaced with either alanine (A and B), threonine (C and D), l-2-aminobutyric acid (Abu) (E and F), or l-4-hydroxyproline (Hyp) (G and H). MOG, myelin oligodendrocyte glycoprotein.

Figure 3

Aggregation of peptides in EBV-infected B cells. EBV-infected B cells cultured at standard conditions were incubated with 50 μm of the indicated peptides for 48 h. Cells were stained with DRAQ5 (red) to visualize the nucleus and with thioflavin-S (ThS, green) to visualize amyloid-like aggregates. Representative images are shown. Scale bars represent 25 μm. (A) control (no peptide), (B) hMOG_35–55 (P1), (C) hMOG_35–55-Cit_41,46 (P2), (D) hMOG_35–55-Cit_46,52 (P3), (E) mMOG_35–55 (P4), (F) mMOG_35–55-Cit_41,46 (P5), and (G) mMOG_35–55-Cit_46,52 (P6). EBV, Epstein–Barr virus; MOG, myelin oligodendrocyte glycoprotein.

Figure 4

Murine and human citrullinated MOG_35-55peptides show distinct T-cell activation in a marmoset B-cell cross-presentation assay. B cells isolated from rhMOG-immunized marmosets were incubated with different peptides at the indicated concentrations for 1 h. Marmoset axillary lymph node (ALN) cells were added, and proliferation was measured using a [³H]thymidine incorporation assay. A stimulation index (ratio of proliferation for stimulated T cells versus unstimulated control, SI) was plotted for each condition, and SI >2 was taken as the cutoff for T-cell activation. The experiment was carried out using B cells and ALN cells from three different animals, with individual responses shown as three distinct colors. rhMOG, recombinant human myelin oligodendrocyte glycoprotein .

Figure 5

Evaluation of the amyloid-like aggregation of truncated citrullinated mMOG_35-55peptides by ThT assay.A, ThT aggregation curves of the three citrullinated mMOG_40–48 epitopes at 50 μM concentration. B, ThT aggregation curves of the stepwise extension of the mMOG_40–48-Cit₄₆ epitope. This graph shows representative curves at 50 μM. mMOG, myelin oligodendrocyte glycoprotein from mouse; ThT, thioflavin T.

Figure 6

Effect of citrullination on the cathepsin G sensitivity of human and murine MOG_35-55.A, degradation of peptides with human cathepsin G (hCatG) at 37 °C and pH 5.0. Peptides (10 μM) were incubated with hCatG (250 ng/ml), and samples were taken out at the stated time points, and enzymatic activity was stopped by dilution of a 45:45:10 mixture of H₂O/MeCN/TFA. The samples were analyzed using LC–mass spectrometry, and amount of degradation was determined by comparison of peak area against an internal standard. The graph shows the average of two independent experiments for each peptide. B, mass spectrometry analysis was used to determine the cleavage sites of hCatG by analyzing the mass spectra of the fragments formed after degradation. Arrows indicated detected cleavage sites. The immunodominant epitope MOG_40–48 is underlined, and the citrullinated arginine residues are highlighted in bold. MOG, myelin oligodendrocyte glycoprotein.