Replying again to myself, I have found some previous evidence about the possibility of EBV-infected B-cells. The first one I have found is from 2007 "Dysregulated Epstein-Barr virus infection in the multiple sclerosis brain" (
http://jem.rupress.org/content/204/12/2899?papetoc=). They say:
"Here, we show that intracerebral accumulation of EBV-infected B cells and plasma cells is a regular feature of MS and identify ectopic B cell follicles as main sites of viral persistence. We also provide evidence for a relationship between acute inflammation and EBV reactivation in the MS brain"
The ectopic B-cell follicles are accumulations of B-cells in the meninges. They seem to be related to the damage to the gray matter and the progressive phase.
Here I copypaste some other results, summarising the situation of the research in this field.
Maybe the most clear indicator of the B-cell problems is the efectivity of Rituxan. In fact B-cell depletion should occur in a natural way in MS, being the T-cells responsible for killing the sick B-cells (CD8+ T-cells). In MS patients this does not happen ("Defective T-cell control of Epstein–Barr virus infection in multiple sclerosis",
http://www.nature.com/cti/journal/v6/n1 ... 1687a.html)
Moreover, in a small study in one secondary progressive MS patient, the injection of cytotoxic T cells designed to kill the host EBV-infected B cells produced a remission of the disease (
http://www.cell.com/trends/molecular-me ... 0149-6.pdf)
The mouse EAE models do not work as expected. Instead, it has been found that a kind of monkeys, the marmosets, have a similar behaviour having EAE and B-cells infection. Instead of using EBV they have an equivalent virus (CalHV3,
http://journals.sagepub.com/doi/abs/10. ... 7317690184). In this cases, the CD8+ Tcells targeted myelin instead of the infected B-cells. Several reports point out that there is a problematic protein named EBNA1 (
http://onlinelibrary.wiley.com/doi/10.1 ... 21886/full,
http://www.neurology.org/content/62/12/2277.short,
http://link.springer.com/article/10.100 ... 010-9201-3)
It seems that the EBV-infected marmoset is the most accurate model that we have for MS. Instead of human EBV they use a similar family of viruses named lymphocryptovirus (LCVs) of which the previous CalHV3 is one of them. In these experiments the mutual influence between Tcells and Bcells is similar to the corresponding situation in human MS (
http://www.nature.com/cti/journal/v6/n2 ... 0171a.html,
http://www.jimmunol.org/content/197/4/1074.short)
Finally there is a good review about the subject at
http://www.cell.com/trends/molecular-me ... 16)30149-6