If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pathology, VU University Medical Center, Amsterdam, The NetherlandsDepartment of Neuroimmunology Unit, Blizard Institute, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, United Kingdom
HERV-W Endogenous Retroviral protein MSRV-Env is detected in all MS Brains.
Intense staining of MSRV-Env protein coincides with areas of active demyelination.
It has lifelong expression from earliest demyelination to late progressive lesions.
It is pathogenic, inducing inflammation, autoimmunity and remyelination blockade.
MSRV-Env is a pathogenic target in MS for new and specific therapeutic strategies.
Attempts to identify a causative agent of Multiple Sclerosis (MS) among environmental viruses have consistently failed suggesting that development of MS is a result from gene-environment interactions. A new pathogenic player within human genes, a human endogenous retrovirus (HERV) was identified from MS cells, named MS-associated retrovirus element (MSRV) and unveiled homologous multicopy HERVs (HERV-W). As independent studies revealed biological features of HERV-W on immune-mediated inflammation and on remyelinating cells, the present study characterized the presence of HERV-W envelope protein (MSRV-Env) at the cellular level, in different MS lesion stages to extend and validate previous studies.
Immunohistological analysis of HERV-W envelope cellular expression in different lesion stages from a cohort of MS brains versus controls, using well-characterized and highly specific monoclonal antibodies.
HERV-W envelope protein was detected in all MS brains and quite essentially in lesions. Immunohistochemistry showed dominant expression in macrophages and microglia, coinciding with areas of active demyelination, spread over the active lesions, or limited to the rim of active microglia in chronic active lesions or in few surviving astrocytes of inactive plaques. Weak expression was seen in MS normal appearing white matter. In active plaques, few lymphoid cells and astrocytes were also stained. This HERV-W expression was not observed in control brains.
HERV-W was expressed in demyelinated lesions from MS brains, which were all positive for this endogenous pathogenic protein. Pronounced HERV-W immunoreactivity in active MS lesions was intimately associated with areas of active demyelination throughout the successive stages of lesion evolution in MS brains. Based on its pathogenic potential, this HERV-W (MSRV) endogenous toxin thus appears to be a novel therapeutic target in MS. It also has a unique positioning as an early and lifelong expressed pathogenic agonist, acting upstream the pathways in which dysregulated physiological effectors are usually targeted by present therapeutic strategies for MS.
Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system (CNS). Histopathological evaluation of MS brain tissue has provided insights into key pathological features of MS, including inflammation, demyelinated lesions in the white and the grey matter, oligodendrocyte loss, defects in remyelination by oligodendrocyte precursor cells (OPC), and axonal and neuronal degeneration (
). To date, a global understanding of etiological factors that could directly and/or indirectly be involved in MS, in its onset, in the lifelong mechanisms of inflammation-driven CNS damage and in the impairment of lesion remyelination, is likely to rely upon a gene-environment interplay (
It therefore appears relevant that the search for pathogenic players in MS led to peculiar elements with both genomic and viral characteristics: Human Endogenous RetroViruses (HERVs), which have repeatedly entered the genome of species through germ-line infections during evolution of species and are now known to represent about 8% of the human genome (
). So, when a new human retrovirus was identified in cultures from MS cells (MSRV, for Multiple sclerosis associated RetroVirus) but also unveiled a previously unknown family of homologous and endogenous retroviral sequences (HERV-W) (
), these findings were initially regarded as anecdotal. Nonetheless, as studies further explored and progressively unveiled unexpected genetic and biological features of these HERVs, it became apparent that HERVs were not merely fossils of past infection and inert relics of ancient genomic invasions, but could display activities with a substantial impact on cellular function, in both health and disease (
). Thus, cumulated results from numerous studies during past decades progressively led to understand how their unique genetic positioning may confer upon them the potential to drive pathogenic cascades leading to MS in response to environmental triggers. HERV-W sequences isolated from MS were shown to produce a bioactive envelope protein (MSRV-Env) engaging pathways leading to final pathognomonic features of Multiple Sclerosis (MS).
In particular, immunological effects of MSRV particles revealed to induce superantigen-like activation of T-lymphocytes (
). MSRV-Env was also shown to promote macrophage and dendritic cells differentiation, anti-myelin oligodendrocyte glycoprotein autoimmunity and experimental autoimmune encephalomyelitis (EAE) in C57/BL/6 mice, an animal model of MS, (
), but its molecular origin and mechanisms of action long remained unknown. However, it was recently discovered that transiently expressed TLR4 in early differentiating oligodendrocyte precursor cells (OPC) exposed them to an interaction with MSRV-Env causing a defect of myelin production through a differentiation blockade of OPC (
This appeared relevant when considering that other studies confirmed the enhanced d expression of HERV-W/MSRV elements in patients with MS and the ex-vivo and post-mortem detection of its envelope protein (
Importantly, the HERV-W envelope was detected in MS pathognomonic lesions, i.e. within brain demyelinated lesions, using monoclonal antibodies (mAbs) raised against the MSRV-Env protein encoded by virion RNA isolated from MS cell cultures (
). Successive studies have repeatedly and independently confirmed the presence of HERV-W envelope (here named MSRV-Env) within MS lesions with various specific mAbs, as detailed in Table 1. Such results obtained after examination of brain regions and lesions from a total of 55 MS cases from 6 different centers and countries have consolidated the evidence for the presence of HERV-W Env protein expression in MS.
Table 1Previous studies on HERV-W envelope Immunohistological detection in MS and control brains.
These observations reported its detection in astroglial or macrophage-like cells within MS lesions and in endothelial cells within Marburg's type hyperacute lesions. However, which immune or glial cells expressed this protein in the CNS and where, during the development of the lesions and according to their different stages of activity until the end of MS patient's life (when autopsy material is collected), still remained open questions. Further determining when such an expression arose and localized according to MS lesion development required a systematic analysis of MS brains, from first perivascular inflammatory infiltrates to late burnt out plaques.
To address these questions, the present study performed a systematic analysis of HERV-W envelope protein (MSRV-Env) distribution and cellular localization in different stages of MS lesion development, using a large cohort of well-characterized MS samples with one of the 3 highly specific monoclonal antibodies that were previously used to examine its expression in sections of MS lesions (
Here we show that MSRV-Env protein is abundantly present in MS brain lesions and is intimately associated with active areas of demyelination where macrophages and microglia represent the predominant cell type expressing MSRV Env. Only moderate expression was observed in reactive astrocytes throughout active and chronic active lesion areas, while rarely in inactive lesion areas. Expression in perivascular macrophages and in neighboring endothelial cells appeared quite essentially in early demyelinated areas.
2. Materials and methods
2.1 Autopsy material
Brain samples from 20 patients with clinically diagnosed and neuropathologically confirmed MS were obtained at rapid autopsy and immediately fixed in buffered formalin (in collaboration with The Netherlands Brain Bank, Amsterdam; Dr. I. Huitinga, coordinator). The Netherlands Brain Bank received permission to perform autopsies for the use of tissue and for access to medical records for research purposes from the ethics committee of the VU Medical Center (Amsterdam, The Netherlands). Six cases without neurological disease were selected as controls. Tissue samples from control cases were taken from the subcortical white matter or corpus callosum, regions where most MS lesions were encountered and therefore analyzed. MS tissue samples were selected on the basis of postmortem MRI and lesions were classified according to validated histopathological criteria as previously published (
). MRI-guided white matter lesions were thus obtained from different locations in every case. Hence, cases with different type of lesion stages was included in the present MS cohort. However, different types of lesions were also obtained from individual brains as lesional activity can still be detected at autopsy in most cases (
). Relevant clinical information was retrieved from the medical records and is summarized in Table 2. All patients and controls, or their next of kin, had given informed consent for autopsy and the use of their brain tissue for research purposes.
Table 2Detailed information on MS and control cases included in the immunohistochemical study.
All antibodies were mAb mouse IgGs obtained after immunization with either recombinant protein (Env) or immunization plasmid. Selected hybridomas were cultured in vitro for antibody production. mAb purification was made with protein-A sepharose columns and purity of the final product controlled by HPLC analysis. Their specificity was controlled by using MSRV-Env proteins produced in E. coli (non-glycosylated) or in transfected human cell cultures (glycosylated) and further purified as endotoxin-free recombinant full-length proteins for ELISA and WB analyses. Transfected cells were used for Immunofluorescence and immunocytochemistry analyses of mAb specificity (Cf. Supplementary material).
Formalin-fixed paraffin-embedded tissue from each selected region was serially sectioned at 5 µm within large and successive areas covering the entire lesion and stained for proteolipid protein (PLP; clone plpc1; Serotec) and major histocompatibility complex (MHC) class II (DAKO) to characterize lesion stage. HERV-W envelope protein (MSRV-Env) was detected using GN-mAb_03 at a concentration of 10 µg/ml. Sections were deparaffinized in xylene and rehydrated through graded alcohol into distilled water and endogenous peroxidase activity was quenched by incubating the slides in 0.3% hydrogen peroxide in methanol. Serial sections of each tissue sample were incubated overnight with appropriate antibodies and subsequently incubated with horseradish peroxidase-labeled anti-mouse/rabbit from the EnVision kit (Dako) for 30 min at room temperature and finally diaminobenzidine tetrachloride. Between incubation steps, the sections were thoroughly washed with PBS. After a short rinse in tap water the sections were incubated with hematoxylin for 1 min and extensively washed with tap water for 10 min. Finally, the sections were dehydrated with ethanol followed by xylol and mounted with Entellan (Merck, Darmstadt, Germany) for systematic examination of all lesion areas. All antibodies were diluted in PBS containing 0.1% bovine serum albumin (Boehringer Mannheim). For cellular localization sections were incubated overnight with antibodies directed against HERV-W Env protein followed by incubation with Alexa-594-labeled goat anti-mouse (1:400; Molecular Probes). Astrocytes were visualized using rabbit anti-glial fibrillary acidic protein (GFAP; 1:500; Dako) and microglia/macrophages using either rabbit anti-Iba1 (1:100; Wako, ref. 019-19741) or rabbit anti-p22phox (Santa Cruz Biotechnology, ref. sc-20781). Both antibodies were applied for 1 h and followed by incubation with Alexa-488-labeled goat anti-rabbit (1:400; Molecular Probes). After being washed, the slides were covered with Vectashield (Vector Laboratories) supplemented with 0.4% DAPI to stain nuclei. Microscopical analysis was performed with a Leica TCS SP2 AOBS confocal laser-scanning microscope (Leica Microsystems, Heidelberg, Germany).
Sections were prepared from tissue blocks of 20 MS patients and classification of lesion staging was based on immunohistochemical detection of inflammatory cells expressing MHC class II/HLA-DR antigens and on the detection pattern of PLP, which reveal areas of myelin loss or the presence of myelin in phagocytic cells as previously described (
). Based on these stainings we selected 6 active, 7 chronic active and 7 chronic inactive lesions in studied sections. “Active” lesion areas were classified as lesions with myelin loss and abundant parenchymal macrophages. “Chronic active” lesion areas were characterized by a hypocellular demyelinated gliotic center with astrogliosis and a hypercellular rim containing activated microglia and macrophages. “Chronic inactive” lesion areas, in contrast to “active” and “chronic active” lesion areas, were devoid of macrophages and only sporadically contained some activated microglia. White matter samples from all control samples did not show any sign of changes in myelin distribution and lacked activated microglia and infiltrated macrophages.
In control white matter samples, anti-Env antibody only weakly decorated isolated glial cells (Fig. 1). Glial cells with weak Env-positivity were seen in NAWM surrounding MS lesions (data not shown).
With marked contrast, when compared to control white matter samples and NAWM of MS brains, MSRV-Env was abundantly present in demyelinated areas with ongoing inflammation (Fig. 2 (A)–(C)). Particularly, infiltrating macrophages and activated microglia were markedly labeled with antibodies directed against MSRV-Env (Fig. 2(E) and (F)). Dense lymphoid infiltrates in the vicinity of this active plaque were observed and MSRV-Env was detected in a proportion of these lymphoid cells. Interestingly, it not only involved monocyte cells but also T-cells, as shown by co-staining of few CD3 positive cells (Fig. 2(G)). MSRV-Env-positive reactive astrocytes were also detected throughout active lesions (Fig. 2(C) and (F) with GFAP double staining).
In a very early perivascular demyelination process (Fig. 3(A)-(B)), perivascular macrophages infiltrating from the contiguous blood vessel were quite exclusively stained with anti-Env antibody as well as neighboring endothelial cells (Fig. 3(C)-(D)). Moreover, this MSRV-Env expression was restricted to the area of demyelination, as superposed with the small area with decreased or absence of PLP staining in the upper left part next to the blood vessel wall (Fig. 3(A)).Conversely, activated microglia (MHC-II positive in Fig. 3(B)) extended far beyond this initial demyelinated area, probably reflecting activation by secreted pro-inflammatory cytokines with large intraparenchymal diffusion. Thus, immunoinflammatory activation of microglia did not directly cause MSRV-Env expression, which was localized within the actively demyelinating area.
The cellular expression pattern of this HERV-W protein in chronic active lesion areas, while showing similarities with the distribution observed in active areas as MSRV-Env protein was markedly upregulated in activated microglia, was strikingly coinciding with the rim of a large demyelinating lesion as shown in Fig. 4. Here, the edge of an expanding lesion is decorated by an Env-positive line (4 C) coinciding with HLA DR-positive detection (4B). The fact that both activated microglia (actively demyelinating cells) and bipolar fusiform microglia (migrating cells) showed strong expression of MSRV-Env protein (4D), suggests that this “line of microglia” represents the “line of progression” of the demyelinating cells that expand the lesions in such chronic active plaques.
In chronic inactive lesion areas, the intensity of the staining was strikingly reduced and limited to some sporadic HERV-W Env-immunoreactive astrocytes and to faint dust-like staining of fibrotic extracellular structures where secreted MSRV-Env protein could remain present (Fig. 5).
Taking together all data from this extensive immunohistochemical survey of these different types of lesions from 20 different MS brains, it became obvious that HERV-W envelope protein expression is markedly upregulated within all inflammatory lesions or within actively demyelinating microglia at the rim of chronic lesions. It predominantly localizes to macrophages, activated microglia and reactive astrocytes, intimately coinciding with actively demyelinating localization as detected in all examined MS brains (n=20).
The HERV-W envelope protein (Env) is known to provoke immunoinflammatory and glial cythopathic effects on human cells in culture and in animal models. This was shown for MSRV-Env, as encoded by clones isolated from MS virus-like particles (
). The pathogenic effects of HERV-W envelope therefore coincide with the major pathways underlying hallmarks of MS pathology.
Detailed immunohistochemical analysis showed that the corresponding envelope protein was detected in the successive typical stages of lesion from MS in strict association with cells and histopathological characteristics constituting hallmarks of active demyelination.
HERV-W (MSRV-Env) expression appeared to be associated with active demyelination from the early stage, involving endothelial cells, perivascular macrophages and microglia neighboring macrophage infiltration within brain parenchyma, to the late lesion progression at the rim of reactive microglia figuring the limit between a large demyelinated plaque and the NAWM. Thus, a persisting weak detection in few surviving astrocytes within late inactive lesion also appeared consistent. At the cellular level, results of systematic observation from early to burnt-out demyelinated areas in different patients showed that (i) HERV-W Env protein production in perivascular macrophage and in microglia appear as a pivotal continuum of active demyelination, that (ii) reactive astrocytes were significantly stained in all lesion stages but to a lesser extent, and only in rare surviving astroglial cells of chronic inactive lesions, which may also correspond to Env protein capture at their surface, that (iii) occasional lymphoid infiltrates comprised positive T-cells, which may rather represent membrane bound protein than active production, and that (iv) along with perivascular macrophages, endothelial cells were stained in early demyelinated lesions. In the latter case, previous observations of similar endothelial staining in hyperacute lesions from Marburg's MS types (
) suggest an interplay between these cell types and MSRV-Env expression during the early and/or active phase of lesion formation at the blood-brain-barrier level. This is supported by negative detection in vascular elements in areas of older lesion as in NAWM. Concerning lymphocytes, as peripheral blood T-lymphocytes were reported not to express HERV-W RNA in MS, when natural killer (NK) cells did (
), further studies are needed to elucidate the phenotype of this T-cell subtype expressing HERV-W MSRV-Env protein in MS-brain. Nonetheless, we may simply observe what would be expected from the binding and/or the presentation by neighboring antigen-presenting cells of this endogenous retroviral protein to certain V-beta chains of the T-cell receptor, which is known to induce potent superantigen-like T-lymphocyte activation as previously shown (
), the present observations of macrophage and/or microglial activation with MSRV-Env expression in earlier stages of demyelinated areas are consistent with their role in a subsequent recruitment of T-cells, along with induced overexpression of integrins at the surface of neighboring endothelial cells (
). Interestingly, this could illustrate how the activation of a disproportionate number of T-cells may occur in MS CNS, thus releasing unexpectedly huge quantities of Interferon gamma and of other pro-inflammatory molecules.
When considering all sections studied in brains from the present cohort, 100% of MS cases were found positive for MSRV-Env protein. Since all MS cases tested in the previous studies also had shown positivity for this HERV-W envelope protein, the total number of MS cases with MSRV-Env positive brain lesions in dedicated studies is 75.
Taken altogether, our data indicate that HERV-W envelope may be a key factor in lesion progression but this endogenous toxin is still present in the end-stage of lesions with dust-like staining of extracellular fibrotic structures or of rare surviving astrocytes. A key point is that HERV-W, at least its MSRV-Env protein, is expressed in post-mortem lesion tissue implying that it is implicated in active demyelination from the early development to the chronic expansion of lesions, until the end of MS patients’ life. It is thus likely to be produced from the onset of the disease throughout its whole course. This does not correspond to “hit-and run” pathogenic triggers nor to transiently disease-activated factors. It neither correspond to a non-specific activation of mostly defective HERV copies that can be transcriptionally upregulated, impacting various RNA detection (
), this confirms the specificity of MSRV-Env in the observed pathogenicity. HERV-W envelope in MS (MSRV-Env) is thus more likely to correspond to a central element of the pathogenesis and to a potential therapeutic target in this disease. Its association with demyelinating cells in the lesions support the rationale of a therapeutic strategy using a neutralizing antibody targeting this endogenous protein. A humanized neutralizing antibody is now available as a recombinant human IgG4, GNbAC1 (