Virus-induced demyelination: from animal models to human diseases.

Demyelinating disorders of the human central nervous system (CNS), principally multiple sclerosis (MS) but also postinfectious encephalomyelitis, commonly result in profound neurologic dysfunction. These diseases seem to involve primarily cell-mediated immune mechanisms and probably have an initiating environmental factor—that is, a virus. This relationship clearly exists for postinfectious encephalomyelitis and, on the basis of epidemiologic investigations, is also likely for MS. Evidence from studies of twins indicates that environmental and genetic susceptibility factors are important in MS.

Murine Models of CNS Demyelination.

Two murine models of CNS demyelination—experimental allergic encephalomyelitis and Theiler's murine encephalomyelitis—provide opportunities for investigating autoimmune and virus-associated disease, respectively. Experimental allergic encephalomyelitis is of considerable interest both as an autoimmune disease of the CNS and as a model of immunotherapy for CNS demyelinating diseases. A myelin basic protein-specific, CD4+ T-cell population is required for initiation of experimental allergic encephalomyelitis. MS and experimental allergic encephalomyelitis have the following common features: CNS demyelination, perivascular CD4+ T cells, association with major histocompatibility complex (MHC) class II antigens, and restricted T-cell receptor variable-gene segment utilization. The murine adoptive transfer model in the SJL/J mouse has another important feature of MS—the chronic relapsing clinical course. This recurrent course is ideal for manipulating the immune response not only during induction and onset of disease but also during relapse. The pathologic findings are characterized by inflammation and prominent demyelination. This autoimmune disease, however, is not associated with an environmental factor.

Theiler's murine encephalomyelitis virus (TMEV) is an important model of immune-mediated demyelination because it has parallels with postinfectious encephalomyelitis and MS. In this model, a mild or even subclinical viral encephalitis is followed by a 4-week period of quiescence and then the onset of demyelination. The demyelination in the TMEV model is mediated by viral-specific T lymphocytes. During the demyelinating phase of the disease, however, the virus is persistent; this finding implicates either a low-level expression of viral polypeptides or an immunologic cross-reactivity between virus and myelin antigen (or antigens). As in MS and experimental allergic encephalomyelitis, T cells seem to initiate immune-mediated demyelination in TMEV.

In this issue of the Mayo Clinic Proceedings (pages 829 to 838), Lindsley and associates describe the association of viral and MHC class I antigen expression in areas of demyelination induced by TMEV infection. The observation that these molecules are coexpressed in the white matter of the spinal cord in susceptible mice but not in resistant mice is a major contribution to the understanding of this fascinating animal model of CNS demyelination. Interestingly, only a few glial cells in the spinal cords were infected, and many more cells expressed MHC class I antigen only. This result may be attributed to the secretion of cytokines within the CNS by lymphocytes or endothelial cells or even glial-induced “upregulation” of MHC class I molecules.

These findings prompt three important questions:

What are the roles of MHC class I (CD8+ T cell-mediated) and MHC class II (CD4+-mediated) immune responses in the pathogenesis of demyelination after TMEV infection?

How does TMEV-induced demyelination compare with other models of autoimmune or virus-induced demyelination?

Is a viral cause reasonable to pursue in the human demyelinating diseases—postinfectious encephalomyelitis and MS?

Role of MHC Immune Response.

Both CD8+ MHC class I-restricted and CD4+ MHC class II-restricted T lymphocytes are important in the initial immune response that limits infection and in subsequent demyelination after TMEV infection. Pioneering studies that demonstrated that immunosuppression could decrease demyelination were followed by investigations of depletion of specific lymphocytic phenotypes by administration of monoclonal antibodies. Monoclonal antibodies specific for CD8+, CD4+, and I-A+ cells all result in decreased demyelination. Class I MHC antigen (H-2D) has been implicated as a genetic susceptibility factor. The presence of CD8+ cells but not CD4+ cells in the CNS parenchyma during demyelination in conjunction with expression of viral antigen and MHC class I antigen on potential target cells suggests that the CD8+ T lymphocytes are effector cells involved in a cytotoxic process.

The overall role of the CD8+ subset seems complex in that the CD8+ cells are implicated as a characteristic of disease resistance in the resistant strains of mice. These lymphocytes are important in the initial immune response to the virus and for clearance of the virus from the CNS. The results of recent investigative studies of TMEV infection in mice deficient in β2-microglobulin and therefore deficient in MHC class I antigen demonstrate striking demyelination and implicate the class I-restricted immune response. Therefore, the CD8+ T lymphocytes are important for limiting infection and both potentiating and inhibiting demyelination. A dual role for CD8+ T lymphocytes is now established in the murine model of relapsing experimental allergic encephalomyelitis, in which the CD8+ cells function both as effectors and as regulators. CD8+ T cells may also have an important role in the pathogenesis of MS.

CD4+ T lymphocytes have been shown to be critical in the immune response to myelin basic protein, which produces murine relapsing experimental allergic encephalomyelitis. In fact, CD4+ encephalitogenic cells that are class II restricted have cytotoxic MHC class II-restricted activity in vitro. The TMEV-induced demyelination studies with anti-CD4 monoclonal antibodies have demonstrated that if the CD4+ subset is depleted before infection, the result is mortality due to the initial infection. Alternatively, if anti-CD4+ monoclonal antibody is administered after the initial infection, the clinical manifestations of demyelination are decreased. Current evidence suggests that, in TMEV, the T-cell response to viral polypeptides is readily observed and that reactivity to myelin antigens is rare or nonexistent. An intriguing study has substantiated that a virus (VP2)-specific CD4+ T-cell line can adoptively augment the TMEV-induced demyelination. These findings may be consistent with a recent model of CD4 T-cell subtypes.

A model of functionally different subpopulations of CD4+ T cells—T helper (TH) cells with delayed-type hypersensitivity—has been evaluated in mice. The distinction is based on patterns of cytokine production in response to T-cell receptor binding to antigen or immobilized anti-CD3. TH1 cells are characterized by secretion of interleukin 2, interferon-γ, and lymphotoxin (tumor necrosis factor-β), whereas TH2 cells produce interleukins 4, 5, 6, and 10. THO cells, which represent a third population, secrete only interleukin 2 and may be precursors of TH1 and TH2 cells. Expansion of TH1 cells is favored by the presence of interleukin 2 and interferon-γ. Proliferation of TH2 cells is favored by depletion of CD8+ cells or the presence of only interleukin 2. Recent investigations demonstrated that TH1 and TH2 interactions are complex and may be extremely important in immunoregulation. TH1 cells provide some B-cell assistance, especially for production of IgG2a, and major support for delayed-type hypersensitivity, cytotoxic T-lymphocyte, and antiviral immunity. TH2 cells are the major source of B-cell assistance, especially for production of IgGl, IgM, IgA, and IgE, and are directly involved in hypersensitivity reactions. Cross-regulation between TH1 and TH2 cells may be an integral part of immune system control.

Thus, the initial response to infection by CD4+ cells could be promotion of antibody formation by TH2 cells, and the secondary response may be TH1 cells contributing to the demyelination. The further characterization of the CD4+ and CD8+ contributions to the cellular immune response in Theiler's murine encephalomyelitis will clarify the pathogenetic interactions and the interaction between these subsets and may suggest which immunotherapeutic intervention will be most effective. A new spectrum of immunotherapies has been successfully applied to models of experimental allergic encephalomyelitis, including the following examples: (1) cytokine (T-cell growth factor-β1), (2) lymphokine-toxin treatment, (3) anti-T-cell receptor variable β-gene-specific monoclonal antibody treatment, (4) T-cell vaccination, and (5) blocking peptides. Application of similar therapies to prevent TMEV-induced demyelination will be of considerable interest. Two factors must be considered: (1) Will new therapies activate persistent virus? (2) Will these selective therapies be effective for a long-term period in the setting of continuous low-level expression of viral antigen?

Model Systems for Human Demyelinating Disease.

Other model systems for human demyelinating disease with a viral pathogenesis, including canine distemper virus, JHM virus (a coronavirus), and visna virus, have been reviewed previously. Of particular interest relative to the study by Lindsley and colleagues is a recent observation that measles virus infection of the CNS in Lewis rats results in neuronal infection, with white matter inflammation of antigen-presenting cells expressing class II antigens in the absence of virus. A model system with similarities to MS in humans has been produced by intracerebral injection of a coronavirus into primates. This system includes demyelination in the absence of viral products, as measured by in situ hybridization, polymerase chain reaction, and immunohistochemical studies.

These models of virus-induced demyelination are important for understanding the human disease postinfectious encephalomyelitis. Different mechanisms of demyelination depend on the virus involved. Persistent virus in glial cells can provide target cells for a cytotoxic process in the white matter. Alternatively, the virus may provoke an immune response to myelin antigens either by direct CNS injury that releases antigens or by molecular mimicry in which viral epitopes with sequence homology to myelin proteins produce a cross-reactive immune response.

A major question is whether such models are relevant to MS. As discussed previously, MS is characterized by three major elements: an environmental factor, genetic susceptibility, and an immunologic response. Many recent studies have investigated detailed aspects of the immune response, especially T-cell responses to myelin antigens. Among laboratories, no consensus exists about the restriction or heterogeneity of T cells responding to myelin basic protein. A potential linkage or association of MS with genes of the immune system is the focus of recent investigation. The evidence for an environmental cause is compelling. Most likely, this factor would be exposure to either an infectious agent such as a virus or a pathogen that produces a superantigen capable of activating T cells. Of concern is the possibility that a heterogeneous group of environmental factors could elicit apparently similar symptoms and pathologic changes.

Many viruses have been isolated from the CNS of patients with MS. To date, however, no consistent isolate has been found. The recent attribution and repudiation of human T-cell lymphotropic virus type I as the agent in MS demonstrate the need for use of meticulous techniques for detecting viruses in MS.18, 19 Human T-cell lymphotropic virus type I remains important because of its features in producing tropical spastic paraparesis. The similarities to MS include inflammation, demyelination, probable cell-mediated pathogens, and evidence of disease of the cerebral white matter on magnetic resonance imaging (Table 1 ). The possibility of the environmental factor (or factors) in MS being a retrovirus or any type of virus necessitates continued consideration. In a recent study, coronavirus RNA and antigen were demonstrated in the brain of 11 of 21 patients with MS and 2 of 21 patients without MS. This investigation once again suggests the possibility of a viral cause for MS; however, careful confirmation and assessment of this association will be necessary.

Table 1

Characteristics of Human Demyelinating Diseases and Animal Models*

	Human diseases		Animal models
Characteristic	MS	CPM	EAE	TMEV	Coronavirus
Inflammation	Yes	Yes	Yes	Yes	Yes
Demyelination	Yes	Yes	Yes	Yes	Yes
Environmental factor	Yes	Yes	No	Yes	Yes
Genetic factor	Yes	No	Yes	Yes	?
HLA	Yes	?	Yes	Yes	?
T cell mediated	Prob	Prob	Yes	Yes	Prob

CPM = chronic progressive myelopathy associated with human T-cell lymphotropic virus type I; EAE = experimental allergic encephalomyelitis; HLA = human leukocyte antigen; MS = multiple sclerosis; Prob = probable; TMEV = Theiler's murine encephalomyelitis virus; ? = unknown.

Conclusion.

The model systems available—and the TMEV model in particular—provide an exceptional opportunity to study the interaction of the immune response and virus in the pathogenesis of CNS demyelination. The investigation of virus-associated CNS demyelination in the model systems will yield invaluable information directly applicable to human demyelinating disorders.