Multiple sclerosis (MS) is a chronic inflammatory and degenerative disease of the central nervous system (CNS). Although its etiology is unknown, it is considered an autoimmune disease facilitated by genetic and environmental predispositions (Milo and Kahana, 2010). Already in the 1990’s, the presence of retrovirus-like particles was associated with MS pathology (Perron et al., 1997). The RNA of these particles was characterized as the MS-associated endogenous retrovirus (MSRV), which shares good homology with the human endogenous retrovirus family W (HERV-W) (Blond et al., 1999). Both the presence and expression of RNA and proteins of MSRV/HERV-W is elevated in MS patients and levels coincide with progression of the disease (Garcia-Montojo et al., 2013; Mameli et al. al., 2009; Perron et al., 2012). Although MS mostly affects the brain, increased expression of MSRV/HERV-W has also been observed in the peripherical blood mononuclear cells (PBMC). HERV-W is a family composed of more than 600 similar elements integrated throughout the human genome (Pavlicek et al., 2002). While similar, detailed sequence analysis allows for discrimination between elements (Lander et al., 2001). My studies have focussed on elucidating the exact relationship between MSRV and HERV-W on the one hand, and understanding how HERV-W overexpression contributes to MS on the other hand.
I hypothesized that MSRV could be a HERV-W element that is not fixed in the current population, and therefore not described in the genome database. My first objective was to identify copies of HERV-W not represented in the database. A PCR-based method described previously to localize transposable elements (Uren et al., 2009), was adapted to HERV-W type sequences. Using the variant method, several HERV-W elements present in PBMCs from 4 MS patients have been successfully amplified, cloned and localized in the human genome. However, all of the HERV-W sequences identified in this way corresponded to copies represented in the database. NGS has allowed for identification of HERV-W copies not represented in the database. Their implication in MS is presently unknown.
Thanks to a collaboration with the department of Neurology of Miguel Servet University Hospital (Zaragoza, Spain), I have obtained blood samples from MS patients and healthy controls from which I have extracted PBMCs. The number of copies of HERV-W in the samples was analysed, and I have detected that it is not enhanced in MS patients. HERV-W expression was analysed in PBMCs, and expression was slightly elevated in MS patients. Overexpression among patients could be linked to clinically active patients. I have analysed which HERV-W elements might be overexpressed or overrepresented in the transcriptome of MS patients. Relative expression levels of HERV-W that are most abundantly transcribed in PBMCs do not differ between MS patients and controls. By contrast, several specific HERV-W elements are exclusively transcribed in MS patients, although their relative expression levels are not high (between 1-5%).
Taking other HERVs elements as an example, regulation of HERV-W may be controlled, among other mechanisms, by epigenetic mechanisms such as DNA methylation. As the methylation status of HERV-W has not been analysed in the context of MS, the possibility that the methylation of HERV-W loci would control their expression was investigated. Methylation assays for selected HERV-W elements were designed, validated and applied to genomic DNA obtained from MS patients. Results show that the analysed loci were highly methylated in PBMCs of both MS patients and controls. No relation between HERV-W expression and methylation levels could be established. Similarly, methylation levels were not altered in MS and did not correlate with the pathology.
Toll-like receptors (TLRs) detect viral proteins or nucleic acids and mediate an antiviral response that includes interferon beta (INFb) induction. While dsRNA transcribed from HERV loci is capable of triggering a TLR3-mediated immune response in tumour cells (Chiappinelli et al., 2015), a similar mechanism in MS has not been studied. The possibility that overexpression of HERV-W may induce an inflammatory response within the CNS, specifically in oligodendrocytes and neuronal precursor cells (NPCs), was investigated. Lentivirus-mediated HERV-W overexpression induced expression of INFb in NPCs, possibly through TLR.
The combined results of my studies suggest that the increase in HERV-W expression observed in MS patients does not originate from a single deregulated HERV-W element that may be associated with the pathology. On the contrary, they suggest several copies of HERV-W that are expressed at low levels, do so exclusively in patients. My results suggest that the deregulation of HERV-W could activate the innate system within the CNS and thus contribute to the neuroinflammation present in MS.

PhD name: 
Start date: 
End date: 
Research centre : 
Institution :