Center for Medical Education

Many autoimmune diseases preferentially affect women; however, the underlying mechanisms for the sex differences are poorly understood. We examined sex-dependent differences in the immunologic response to myelin proteins in 22 multiple sclerosis (MS) patients and 22 healthy controls. Using ELISA spot assay (ELISPOT) methodology, interferon (IFN) gamma and IL-5 secretions were examined at the single cell level in response to overlapping proteolipid protein (PLP) peptides. As previously reported, we observed an overall disease effect in the IFNgamma response, such that MS patients were significantly higher than controls. With respect to PLP-induced IFNgamma secretion, both MS and control females responded higher than their corresponding males. Female MS patients demonstrated the highest responses compared to MS males or healthy controls of either sex. Although MS females had high IFNgamma responses to PLP, they had no IL-5 responses at all, suggesting strong Th1 skewing. In contrast, MS males had more IL-5 than control males, who lacked IL-5 responses. These IL-5 responses suggested that disease and gender are not independent, but rather interact to influence the cytokine response to myelin. The data suggest a gender bias towards Th1 responses in MS, which may contribute to the female predominance in this disease.

Sex hormones play a central role as modulators of immune responses and autoimmune diseases. We hypothesized that suppression of MS disease during pregnancy may be mediated by sex steroid hormones via regulation of costimulatory molecules such as CD40L or CD80/CD86 (B7-1/B7-2). We tested two sex hormones that are implicated in immune suppression during pregnancy: estriol and progesterone. We also examined whether this regulation is gender-specific or disease-related. PBMC from untreated relapsing remitting multiple sclerosis (RR MS) patients and controls were examined for expression of T cell and monocyte costimulatory molecules following mitogen stimulation in the presence or absence of sex hormones. In the absence of hormones, we confirmed that mitogen stimulation induced significantly more CD40L on the surface of CD4(+)T cells in MS patients compared to controls, and we extend these findings by showing there were no gender differences in induction of CD40L. Although supra-physiologic doses of hormones mildly suppressed CD40L expression on activated T cells, in vitro exposure to typical pregnancy-related physiologic doses of estriol or progesterone showed very little or no suppression of CD40L. On monocytes, neither estriol nor progesterone significantly altered the expression of CD80/CD86. These results suggest that physiologic doses of estriol or progesterone cannot alter CD40L on T cells or CD80/CD86 on monocytes sufficiently to explain the improvement observed in MS during pregnancy. Thus, although amelioration of MS and other autoimmune diseases during pregnancy is thought to be due to increased sex hormones, the present results do not support a role for suppression of costimulation via estriol or progesterone.

The relationship between autoreactivity to myelin antigens and disease progression in multiple sclerosis (MS) is not fully understood. We addressed this relationship by cross-sectionally comparing an objective measure of MS disability with immune cytokine responses to myelin proteins. The ELISPOT assay was used to determine the ex vivo interferon gamma (IFNgamma) and interleukin-10 (IL-10) production by peripheral blood mononuclear cells (PBMCs) in response to peptides spanning the entire proteolipid protein (PLP) and myelin basic protein (MBP) molecules in 20 patients with relapsing-remitting (RR) MS and 27 age- and sex-matched healthy controls. MS patients showed significantly higher MBP-induced IFNgamma responses and PLP-induced IL-10 responses compared with healthy controls. Using the Multiple Sclerosis Functional Composite (MSFC), a new multifactorial measure of disability, MS patients showed a significant correlation between the IFNgamma response to PLP peptides and MBP peptides, and disability. In contrast, in MS patients, there was no correlation between the MSFC and the response to unrelated control antigens or mitogens. These data show that myelin-specific T lymphocytes secreting the inflammatory cytokine IFNgamma correlate with functional impairment in MS, supporting an antigen-specific link between the immune response to myelin and disability in MS.

Mutation of the hypoxanthine-guanine phosphoribosyl transferase (HPRT) gene in a T-cell is believed to be an indication that the T-cell has been activated and has proliferated in vivo. HPRT mutant T-cell lines were generated from peripheral blood mononuclear cells from patients with MS and control subjects. More lines were isolated from the MS patients than from the control subjects. Using stringent criteria for recognition, none of the lines from MS-affected or control subjects recognized intact myelin basic protein (MBP) or myelin proteolipid protein (PLP) molecules. Using stringent criteria, two of the 10 MS patients harbored mutant lines each recognizing distinct PLP peptides (PLP peptide 40-60 recognized by 3 lines from one patient and PLP peptide 178-191 recognized by 2 lines from the other patient). A single line recognizing PLP peptide 89-106 was derived from 1 of 7 normal controls. HPRT mutant lines recognizing multiple epitopes of PLP which spanned much of the molecule could be isolated from MS patients, and to a lesser extent, normal subjects.

Myelin proteolipid protein (PLP) is a prime candidate autoantigen for multiple sclerosis. In order to define potential immunodominant epitopes, T cell lines (TCL) from the peripheral blood of HLA-DR 15(2) MS patients were established which responded to the intact molecule of PLP. These TCL were then tested in individual proliferation assays with a variety of PLP peptides spanning most of the PLP molecule. Multiple peptides were recognized by TCL from the MS population, with more than one peptide often recognized by lines from the same individual. Three immunodominant peptides were identified which were recognized by the majority of MS patients. Estimated frequency analyses were then performed on the peripheral blood of HLA-DR15(2)-positive MS and control subjects using TCL initiated by the three immunodominant peptides, 40-60, 95-117, and 185-206. TCL from HLA-DR15 MS subjects recognized peptide 95-117 significantly more often than TCL from control subjects.

Multiple sclerosis affects more women than men. The reasons for this are unknown. Previously, we have shown significant differences in women versus men in inflammatory cytokine responses to the major protein component of myelin, proteolipid protein (PLP), which is thought to be a target in MS patients. Here, using the ELISPOT assay, we examined sex differences in single-cell secretion of Th1 and Th2 cytokines from freshly isolated PBMC between relapsing remitting (RR) MS patients and healthy individuals. Cells were stimulated with MS-associated antigens including proteolipid protein (PLP), myelin basic protein (MBP), myelin oligodendrocyte glycoprotein (MOG), and non-disease related antigens. Our data show a sex bias in the cytokine responses to multiple MS-relevant myelin antigens: Women with MS show IFNgamma-skewed responses and men with MS show IL-5-skewed responses. These data extend our previous findings [Pelfrey, C.M., Cotleur, A.C., Lee, J.C., Rudick, R.A. 2002. Sex differences in cytokine responses to myelin peptides in multiple sclerosis. J. Neuroimmunol. 130, 211-223.]: (1) by demonstrating gender skewing in cytokine responses to an expanded myelin antigen repertoire, which includes MBP, MOG and PLP; (2) by showing TNFalpha and IL-10 do not display comparable gender skewing compared to IFNgamma and IL5; (3) by defining the patient population as early, untreated RRMS patients to avoid confounding factors, such as different disease stages/disability and immunomodulatory therapy; and (4) by showing HLA type does not appear to underlie the gender differences. These findings may explain increased susceptibility to MS in women and could contribute to the differences in disease severity between men and women.

We have designed and expressed in bacteria a recombinant fetal form of human myelin basic protein (21.5 kDa isoform; rhMBP21.5), a candidate autoantigen in multiple sclerosis. An exon 2 insertion, carboxy-terminal histidine tag and preferred bacterial codons differentiate the MBP21.5 gene from that encoding the adult, brain-derived form of human MBP (18.5 kDa isoform; hMBP18.5). MBPs were expressed at high levels in E. coli and extracted from whole cells by simultaneous acid solubilization and mechanical disruption. A nearly two-fold increase in recombinant protein was detected in strains harboring MBP genes with bacterial preferred codons compared to genes containing human codons. The recombinant molecules were purified in two steps, first by reversed-phase chromatographic separation and then by metal affinity chromatography. Dimeric forms of recombinant MBP21.5 were detected under physiological conditions, however, substitution of a serine for the single cysteine at amino acid residue 81 resulted in only monomer formation. All forms of recombinant MBPs induced proliferative responses of human T lymphocytes specific for epitopes in MBP18.5 kDa. In contrast, human T cell lines that recognize an exon 2-encoded epitope of MBP responded to the 21.5 kDa isoform of MBP, but not the 18.5 kDa isoform.

The relationship between multiple sclerosis (MS) disease activity and myelin protein-induced cytokine responses over time is not elucidated. We addressed this relationship by examining longitudinal cytokine responses to myelin proteins every three months for one year, in the context of gadolinium (gad)-enhancing brain lesions and of clinical relapses. The ELISPOT assay was used to determine the ex vivo cytokine production in response to nine amino acid long peptides spanning the entire proteolipid protein (PLP) and myelin basic protein (MBP) molecules in relapsing-remitting (RR) MS patients and matched healthy controls. We identified three longitudinal levels of myelin-induced cytokine secretion by adding up the positive responses for all PLP or MBP peptides obtained for five timepoints, at three-month intervals: low reactivity (< 200 cumulative cytokine-secreting cells), isolated peptide reactivity (201-450 cumulative cytokine-secreting cells) and recurrent protein-wide bursts of cytokine reactivity (> 451 cumulative cytokine-secreting cells). The majority of MS patients showed recurrent bursts to PLP and MBP. In contrast, controls showed a more even distribution between all levels of cytokine reactivity. The majority of patients with gad-enhancing lesions showed PLP/IFN gamma and MBP/IFN gamma recurrent burst responses. This is the first longitudinal study on MS patients in which nine amino acid long myelin peptides are used to reveal the broad range of PLP- and MBP-peptide cytokine reactivity across the whole molecule of these two major myelin proteins. This study also reveals the extremely dynamic nature of the immune reactivity to numerous regions of myelin, which can fluctuate dramatically over time. Such fluctuation could hamper the efficacy of antigen-based therapies for MS.

Adhesion molecules are likely to play a critical role in the immunopathogenesis of multiple sclerosis (MS). The interaction of vascular cell adhesion molecule-1 (VCAM-1) with its lymphocyte ligand very late antigen-4 (VLA-4) may mediate migration of lymphocytes into the CNS. We have previously demonstrated that MS patients treated with interferon beta (IFN-beta) have a significant increase in soluble VCAM-1 (sVCAM-1) soon after the initiation of treatment, and this effect correlated with the resolution of contrast-enhancing MRI lesions. We studied the cell surface expression of VLA-4 by flow cytometry in 10 MS patients before and during IFN-beta treatment. We found a significant decrease in mean VLA-4 fluorescence of MS patients' lymphocytes on treatment and no change in untreated controls. In vitro treatment of lymphocytes with IFN-beta did not reproduce this effect, but the addition of sVCAM-1 did result in a decrease in VLA-4 expression. These data indicate that the previously identified increase in sVCAM-1 may lead to a decrease in VLA-4 expression and that this effect may partially explain the mechanism of action of IFN-beta.