Ancient viral protein enrages astrocytes in multiple sclerosis
Mark P Mattson
& Dennis D Taub
Mark Mattson is in the Laboratory of Neurosciences and Dennis Taub is in the Laboratory of Immunology, National Institute on Aging Intramural Research Program, 5600 Nathan Shock Drive, Baltimore, Maryland 21224, USA. mattsonm@grc.nia.nih.gov
Syncytin is a viral envelope protein encoded in the human genome. New work in this issue indicates that it is activated in multiple sclerosis astrocytes and microglia, contributing to the inflammation-induced myelin destruction that causes disease symptoms.
Multiple sclerosis (MS) is a devastating disease that can cause repeated unpredictable bouts of motor disturbances, partial paralysis, sensory abnormalities and visual impairment. These symptoms result from an inflammatory process that selectively attacks and destroys oligodendrocytes, the cells that form the myelin sheaths around axons in the brain and spinal cord1. Several lines of evidence suggest a role for an autoimmune mechanism of disease pathogenesis. It has been suggested that exposure to a viral, bacterial or other pathogen may trigger the disease process, perhaps through a molecular mimicry mechanism where a protein in the pathogen is similar to protein(s) in myelin, eliciting an autoimmune response2. Current hypotheses mostly blame activated T lymphocytes and microglia/macrophages, which can produce cytotoxic cytokines and reactive oxygen molecules, for the destruction of oligodendrocytes in MS2.
However, the findings of Antony et al. reported in this issue3 suggest that normally docile astrocytes are among the executioners. In individuals with MS, the authors report, astrocytes express a protein called syncytin (Fig. 1), leading to their activation and the synthesis of reactive oxygen species. Syncytin is a viral envelope glycoprotein, but it does not come from any viral pathogen infecting MS patients. Instead, syncytin is produced from a human endogenous retrovirus (HERV-W, at the ERVWE1 locus), a remnant of a virus that invaded humans during primate evolution, probably more than a million years ago. Approximately 8% of the human genome originated from retroviral genomes and, although most of the viral sequences are not expressed, some are, and the functions of these viral proteins in physiology and disease are of considerable interest.
Figure 1. Role of syncytin in the pathogenesis of MS.
Both autoimmune processes and infectious agents have been implicated in the pathogenesis of multiple sclerosis. Activated T cells, which recognize self-antigens expressed in the CNS, and macrophages enter the brain from the peripheral circulation and initiate the inflammatory lesion. Through the production of cytotoxic factors (CF), inflammatory cytokines and reactive oxygen species (ROS), activated T lymphocytes and microglia/macrophages are believed to be responsible for the demyelination and destruction of oligodendrocytes, the oligodendrocytes that form the myelin sheaths around axons in the brain and spinal cord. Astrocytes, which are usually docile, may have a major role in this cytolytic process. Astrocytes and microglia derived from patients with MS show increased expression of the HERV-W envelope protein, syncytin. Syncytin is a cytoplasmic protein induced upon mitogenic activation of astrocytes, microglia and macrophages. Syncytin expression results in increased production of IL-1, ROS and other CF that are toxic to oligodendrocytes. This toxicity can be prevented by the antioxidant ferulic acid, anti-inflammatory drugs and inhibitors of nitric oxide production. Thus, the HERV-W gene encoding syncytin is activated in astrocytes in MS, where it may induce the production of cytokines and ROS that may damage oligodendrocytes resulting in demyelination and associated symptoms.
Syncytin has apparently acquired important functions in humans because, unlike those of other HERV-W elements, which are defective, the open reading frame for syncytin is intact and has been preserved for thousands of years. Indeed, syncytin is highly expressed in the developing placenta, where it is important in trophoblast cell fusion and syncytium formation4. Syncytin is a 518-amino-acid membrane glycoprotein that may exert biological actions by binding to a receptor called ASCT2 (alanine, serine, cysteine transporter 2), which is both an amino acid transporter and a retrovirus receptor5. Viral envelope glycoproteins are known to affect immune responses, and syncytin has amino acid sequences that would be predicted to affect the activation of lymphocytes and macrophages4.
Antony et al. found that levels of syncytin mRNA and protein were significantly higher in frontal cortex white matter tissue samples taken from MS patients as compared to samples taken from patients with Alzheimer disease or HIV encephalitis or from subjects without neurological disease. Syncytin expression was increased specifically in astrocytes and microglia associated with damaged oligodendrocytes, but not in the oligodendrocytes or neurons. When the authors exposed cultured human astrocytes or microglia to a phorbol ester to simulate immune activation, syncytin expression was increased. Overexpression of syncytin in astrocytes and macrophages was sufficient to cause the cells to produce high amounts of the proinflammatory cytokine interleukin-1 (IL-1) and reactive oxygen radicals. The culture medium from astrocytes overexpressing syncytin was toxic to oligodendrocytes, and this toxicity was prevented by the antioxidant ferulic acid, by an anti-inflammatory drug and by inhibitors of nitric oxide production. Whether reactive oxygen intermediates are responsible for the observed toxicity or whether they are intermediates in the generation of additional cytolytic mediators remains to be determined.
To determine whether syncytin can cause demyelination in vivo, the authors injected a syncytin-producing viral vector into the corpus callosum of mice3. Astrocytes were infected and produced large amounts of syncytin, causing damage to oligodendrocytes and impaired sensorimotor function. When mice were administered ferulic acid, the astrocytes still produced syncytin, but oligodendrocyte damage did not occur and sensorimotor function was preserved. Thus, the HERV-W gene encoding syncytin is activated in astrocytes in MS, where it may induce the production of oxygen radicals that then damage adjacent oligodendrocytes, resulting in demyelination and associated symptoms.
Why is the expression of syncytin increased in astrocytes in MS? In placental cells, the production of syncytin is decreased in response to hypoxia, and this is associated with decreased cell fusion and placental abnormalities in pre-eclampsia, a common clinical problem in pregnant women5. The syncytin gene promoter contains binding sites for the transcription factors CBF (CCAAT binding factor), Oct-1 (octamer protein-1), AP-1 and Sp1. CBP and Oct-1 sites are critical for transcriptional regulation of the gene in trophoblast cells6. Indirect evidence suggests that one or more of these transcription factors may be involved in the induction of syncytin in MS. For example, CBP induces the expression of granulocyte/macrophage colony-stimulating factor (GM-CSF)7; GM-CSF expression is increased in association with demyelination in MS; and mice lacking GM-CSF are resistant to demyelination in the experimental autoimmune encephalomyelitis (EAE) model of MS8. In addition to IL-1 , the pro-inflammatory cytokine osteopontin is expressed at high levels by cells associated with MS lesions, including astrocytes and macrophages9. Mice lacking osteopontin are resistant to demyelination in the EAE model10. Sp1 and Oct-1 may be responsible for the upregulation of osteopontin in MS11, consistent with the possibility that these two transcription factors also induce syncytin expression in astrocytes in MS. The pathways upstream of these transcription factors in astrocytes are unknown, but might include cytokines induced in lymphocytes and macrophages early in the MS disease process.
How does syncytin cause the production of proinflammatory cytokines and oxygen radicals in astrocytes? There is ample precedent for the induction of oxidative stress by viral envelope proteins. For example, the HIV-1 coat protein gp120 can be cytotoxic to neurons and glial cells by a mechanism involving membrane-associated oxidative stress and upregulation of inflammatory cytokines IL-1 and other genes involved in oxidative stress responses12, suggesting further similarities in the pathogenic actions of syncytin and gp120. In the case of syncytin, its interaction with ASCT2 could induce the production of oxygen radicals. Indeed, related amino acid transporters are known to influence cellular systems involved in oxygen radical production and removal13. The production of IL-1 in response to syncytin may also contribute to the production of oxygen radicals by enhancing nitric oxide production in astrocytes and possibly the activation and cytokine production by MS lesion associated T cells and macrophages. Why the oxygen radicals produced by astrocytes damage oligodendrocytes and not neurons in MS is unclear. However, oligodendrocytes may have very low levels of antioxidants, rendering them highly susceptible to oxidative stress.
Further studies will be required to determine the relative importance of syncytin in MS. Antony et al.3 show that syncytin is present in glial cells associated with MS lesions and, when overexpressed, can induce astrocytes to produce substances that are toxic to oligodendrocytes. However, their findings do not establish a key role for endogenous syncytin in MS pathogenesis. It will be important to determine whether selective blockade of syncytin expression or action ameliorates the disease process in animal models of MS. The relationship between syncytin expression and relapsing and remitting phases that occur in MS patients should be established. It would also be of considerable interest to determine whether variations in the syncytin gene affect the risk of MS. Moreover, given that syncytin appears to be selectively expressed within the activated glial cells of MS patients, it will also be important to determine whether T cells derived from MS patients are capable of mounting an immune response to this retroviral protein. Sequestered antigens such as syncytin are not readily available for recognition by the immune system during lymphoid development, and thus the release of such antigens, often as a result of tissue injury or trauma, results in a normal immunologic response. Such responses may result in generation of autoantibodies and autoreactive T-lymphocyte populations. Whether such syncytin-specific responses are occurring in MS subjects remains to be determined.
Current treatments that provide temporary symptomatic relief for MS patients, such as interferon- and glatiramer, target immune cell signaling molecules14. Two other drugs that have recently been shown to be effective in reducing symptom severity in MS are mitoxantrone, which may act by suppressing the activation of macrophages and T and B cells, and natalizumab, an antibody against 41 integrin, which acts by inhibiting migration of leukocytes into the CNS. How might the new knowledge about syncytin and astrocyte-derived oxygen radicals contribute to the development of new treatments for MS? New drugs might block the production of syncytin or interfere with its interaction with ASCT2, for example. Certain antioxidants are effective in suppressing the disease process in the EAE model and some of these are moving towards clinical trials15. It is also possible that current therapeutic regimens affect syncytin expression by MS astrocytes and microglia; this would provide further support for an effector role for this protein in the development of MS. Finally, it is important to know whether syncytin has physiological functions in tissues other than the placenta, as such functions could be compromised by treatments aimed at syncytin.
, ROS and other CF that are toxic to oligodendrocytes. This toxicity can be prevented by the antioxidant ferulic acid, anti-inflammatory drugs and inhibitors of nitric oxide production. Thus, the HERV-W gene encoding syncytin is activated in astrocytes in MS, where it may induce the production of cytokines and ROS that may damage oligodendrocytes resulting in demyelination and associated symptoms.
4
