Anti-S-nitrosocysteine antibodies are a predictive marker for demyelination in experimental autoimmune encephalomyelitis: implications for multiple sclerosis

Abstract

Multiple sclerosis (MS) is characterized by inflammation within the CNS. This inflammatory response is associated with production of nitric oxide (NO) and NO-related species that nitrosylate thiols. We postulated that MS patients would exhibit an antibody (Ab) response directed against proteins containing S-nitrosocysteine (SNO-cysteine) and showed that anti-NO-cysteine Abs of the IgM isotype are in fact present in the sera of some MS patients (Boullerne et al., 1995). We report here the presence of a seemingly identical Ab response directed against SNO-cysteine in an acute model of MS, experimental autoimmune encephalomyelitis (EAE) induced in Lewis rats with the 68-84 peptide of guinea pig myelin basic protein (MBP(68-84)). Serum levels of anti-SNO-cysteine Abs peaked 1 week before the onset of clinical signs and well before the appearance of anti-MBP(68-84) Abs. The anti-SNO-cysteine Ab peak titer correlated with the extent of subsequent CNS demyelination, suggesting a link between Ab level and CNS lesion formation. In relapsing-remitting MS patients, we found elevated anti-SNO-cysteine Ab at times of relapse and normal values in most patients judged to be in remission. Two-thirds of patients with secondary progressive MS had elevated anti-SNO-cysteine Ab levels, including those receiving interferon beta-1b. The data show that a rise in circulating anti-SNO-cysteine Ab levels precedes onset of EAE. Anti-SNO-cysteine Abs are also elevated at times of MS attacks and in progressive disease, suggesting a possible role for these Abs, measurable in blood, as a biological marker for clinical activity.

Fig. 1.

First series of experiments on young rats aged 6–7 weeks (n = 7). A, Biphasic pattern of clinical course of EAE in 6- to 7-week-old rats. Clinical scores were graded daily for the scores 0–3 over a 35 d follow-up. Daily average clinical scores are plotted (mean ± SD).B, Serum anti-SNO-cysteine-BSA IgM titers peak 1 week after immunization in rats induced for EAE. Values are expressed as mean OD ± SEM of Ab titers. Anti-SNO-cysteine-BSA IgM titers in sera from EAE rats (open bars) were significantly elevated at 7 dpi when compared with other time points (***p < 0.001). IgG titers (closed bars) rose less than IgM titers but were nonetheless elevated at 7 dpi compared with other time points (*p < 0.05). No rise of anti-SNO-cysteine-BSA IgM was observed in the control group immunized with CFA alone (shaded bars).C, IgM binding to immobilized SNO-cysteine-BSA is inhibited by SNO-cysteine-BSA in liquid phase (competitor) in a dose-dependent manner, indicating specific binding of IgM to SNO-cysteine-BSA. Each plot is the mean OD ± SEM of two independent experiments each performed in duplicate.D, Serum Ab response against MBP_68–84peptide in rats with EAE. Values are expressed as mean OD ± SEM of Ab titers obtained from two independent analyses each in duplicate. Anti-MBP_68–84 IgM titers (open bars) were significantly elevated at 14 and 21 dpi when compared with other time points (*p < 0.05). Anti-MBP_68–84 IgG titers (closed bars) were significantly elevated at 21 and 30 dpi (*p < 0.05).

Fig. 2.

Second series of experiments on older rats aged 9–10 weeks (n = 9). A, Monophasic pattern of clinical course of EAE in 9- to 10-week-old rats showing a milder disease. Clinical scores were graded daily for the scores 0–3 over a 35 d follow-up. Daily average clinical scores are plotted (mean ± SD). B, Anti-SNO-cysteine-BSA IgM titers in sera from EAE rats (open bars) were significantly elevated at 6 dpi when compared with other time points (***p < 0.001). No rise of anti-SNO-cysteine-BSA IgM was observed in the control group immunized with CFA alone at 6 dpi (shaded bars). Values are expressed as mean OD ± SEM of Ab titers obtained from two independent analyses each in duplicate. C, Anti-MBP_68–84 IgG titers (open bars) were significantly elevated at 30 dpi when compared with other time points (***p < 0.001). No rise of anti-MBP_68–84 IgG titers was observed in the control group immunized with CFA alone at any time (shaded bars). Values are expressed as mean OD ± SEM of Ab titers obtained from two independent analyses each in duplicate.

Fig. 3.

Demyelination in cervical spinal cord (C1–C6) assessed by Spielmeyer staining at 35 dpi varies from one rat to another. The ring of white matter is stained in black by Spielmeyer, whereas the central gray matter appears graybecause of the combined hematoxylin coloration. A control rat immunized with CFA alone shows no myelin loss (A), whereas rats with EAE show mild demyelination (B), moderate demyelination (C), or extensive demyelination (D).

Fig. 4.

Anti-SNO-cysteine-BSA IgM titers at 6 dpi correlate with subsequent demyelination and inflammation in spinal cord and cerebellum in rats with EAE of the second experiment (n = 9). Values of Ab titers are expressed as mean OD of two independent experiments in duplicate. A, Demyelination was assessed by Spielmeyer staining of C1–C6 spinal cord sections at 35 dpi, and demyelination values are expressed as percentage of myelin loss. A direct relationship was found between demyelination and anti-SNO-cysteine-BSA IgM titers for sera of EAE rats at 6 dpi (closed squares) with Spearman coefficientr = 0.93 (p = 0.0007).B, Inflammation was assessed by hematoxylin and eosin staining of C1–C6 spinal cord sections at 35 dpi, and values of inflammation are expressed as an inflammation index. A direct relationship was found between inflammation and anti-SNO-cysteine-BSA IgM titers for sera of rats with EAE at 6 dpi (closed squares) with Spearman coefficient r = 0.85 (p = 0.006). C, Anti-SNO-cysteine-BSA IgM titers at 6 dpi correlate with subsequent demyelination of cerebellum at 35 dpi in rats with EAE assessed by MBP content. Values of MBP content are expressed as mean OD (arbitrary units). A direct inverse relationship was found between MBP content and anti-SNO-cysteine-BSA IgM titers for sera of rats with EAE at 6 dpi (closed squares) with Spearman coefficientr = −0.82 (p = 0.011).D, Demyelination assessed by Spielmeyer staining in C1–C6 spinal cord sections correlates with demyelination of cerebellum assessed by MBP content at 35 dpi for rats with EAE (closed squares). A direct inverse relationship was found, with Spearman coefficient r = −0.87 (p = 0.0045).

Fig. 5.

Anti-SNO-cysteine-BSA IgM Abs are elevated during clinical activity in MS patients when compared with healthy controls (group 1; closed squares; n = 19). The ordinate gives relative optical density (see Materials and Methods). Secondary progressive MS patients were divided into an untreated group (group 2; open triangles;n = 14) and a group treated with IFNβ-1b (group 3; closed triangles; n = 9). Relapsing–remitting MS patients were classified into patients in acute relapse and untreated (group 4; open diamonds;n = 8) and in remission and receiving IFNβ-1b treatment (group 5; closed circles;n = 7) or in remission without treatment (group 6;open circles; n = 11). Relative ODs were significantly elevated in some MS groups: group 1 versus groups 2 and 4 (p < 0.001) and versus group 3 (p < 0.05); group 2 versus groups 5 (p < 0.001) and 6 (p < 0.01); and group 4 versus groups 5 (p < 0.01) and 6 (p < 0.05).

Fig. 6.

Anti-SNO-cysteine-BSA IgM Abs are enriched in the IgM of MS patients. Immunoglobulin fractions were tested at a protein concentration of 25 μg/ml. Anti-SNO-cysteine-BSA IgM is elevated in total IgM purified from the serum of an MS patient during relapse on a gel filtration column (closed squares) compared with a healthy donor (open squares). The mean of two independent experiments in duplicate is plotted. MS and control sera representative of four MS and four healthy donor sera tested are shown.

Fig. 7.

MS sera are specific for the SNO-cysteine epitope. IgM binding to immobilized SNO-cysteine-BSA is inhibited by a competitor in the liquid phase in a dose-dependent manner. Binding inhibition is calculated by dividing OD of each competitor concentration by OD without competitor (ODo). A, Specific binding of IgM to immobilized SNO-cysteine-BSA (closed squares) with IC₅₀ at 4 × 10⁻⁷m. Absence of nitrosylation, i.e., with cysteine-BSA (open squares) or SNO-cysteine-BSA treated with HgCl₂ (closed circles), prevented IgM binding inhibition. Each plot is the average OD ± SD of three sera tested in two independent experiments, each performed in duplicate. B, Specific binding of IgM to immobilized SNO-cysteine-BSA (closed squares) with IC₅₀ at 2 × 10⁻⁸m. Conformational modifications of the BSA carrier induced by acidic nitrosylation (NO-BSA-g; open circles) or cysteine-BSA treated with HCl treatment (open squares) or of cysteine-BSA nitrosylated by SNAP (stars) did not abolish binding. Eachplot is the average OD ± SD of three other sera tested in two independent experiments, each performed in duplicate.C, Other oxidative modifications of cysteine-BSA induced by the peroxynitrite donor SIN-1 (closed diamonds) or by hydrogen peroxide (closed circles) did not inhibit IgM binding to immobilized SNO-cysteine-BSA (closed squares). Each plot is the average OD of one serum tested in two independent experiments, each performed in duplicate.

Fig. 8.

Western blot of one MS serum on various nitrosylated antigens showed specificity for SNO-cysteine-BSA. Proteins were loaded at 6 μg/lane. SNO-cysteine-BSA (lane 1) was well stained; NO-BSA-g (lane 2) and DTT-treated NO-BSA-g (lane 4) were not recognized; DTT-treated SNO-cysteine-BSA (lane 3) was stained; and HgCl₂-treated SNO-cysteine-BSA (lane 5) was stained faintly.

Fig. 9.

Time course study of nitrite and nitrate release from SNO-cysteine-BSA on exposure to light and room temperature. Nitrate was enzymatically reduced into nitrite before evaluation by Griess assay. Nitrite levels (open bars) and combined nitrite and nitrate levels (NO ${}_{\text{x}}^{-}$; shaded bars) released from an SNO-cysteine-BSA solution rose slowly from day 1 to day 13 (13 d). Incubation of SNO-cysteine-BSA with 10 mm DTT for 2 hr on day 22 (22 d) markedly increased nitrite and nitrate levels. After 3 d exposure to 10 mm DTT (25d), complete denitrosylation was achieved. Nitrite or nitrate levels were not detected at any time point in the cysteine-BSA control.