Autoantibodies are present in many patients with demyelinating diseases and have been shown to have pathogenic potential in the case of people with neuromyelitis optica (NMO), however it is still debated whether antibodies that are present in people with multiple sclerosis (MS) are pathogenic. There are multiple different mechanisms by which autoantibodies could exert their effects in MS, and they could potentially target any cell/structure within the central nervous system (CNS), although the focus of antibody studies in MS has been almost solely on myelin antigens. Recently, however, some evidence has suggested that astrocytes could also be targeted in MS. Therefore, there is a need for a simple but robust in vitro model to allow the testing of antibodies against oligodendrocytes and astrocytes. The hypothesis tested in this thesis is that a cell culture model containing a mixture of mature glial cells could be a useful model by which to test the pathogenicity of the antibodies from people with MS, and the antigens that they are targeting. In order to test this hypothesis, this thesis set out to develop a mature mixed glial cell culture model, to test whether one cell type could be killed without impacting on the health of the other cells, and to then use this model to test antibodies from people with MS.
Chapter 2 describes the development of a rat neonatal-derived mature mixed glial culture. A method was developed that allowed cultures with a mixture of protoplasmic (type 1) and fibrous (type 2) astrocytes to grow alongside mature oligodendrocytes. This method did not rely on addition of extra growth factors, presumably due to the production of all factors required for growth of each cell type by the other cells in the mixed culture. Because of this co-dependence on the other cells in the mixed culture for growth and development, the next step was to test whether specific targeting (killing) of one cell would result in the death of the other cells in culture, as that could potentially interfere with the interpretation of results in later chapters investigating the effects of antibodies in patient sera. In Chapter 3, it was found that oligodendrocytes could be selectively targeted (using a commercially-available antibody against the galactocerebroside, GalC) and the astrocytes remained unaffected. It proved to be more difficult to find a cell surface-expressed molecule that could target only astrocytes and not have adverse effects on oligodendrocytes. In the literature, a molecule that is generally considered to be only expressed on astrocytes is GLAST, a glutamate transporter. Initial studies using an antibody against the intracellular N-terminus of GLAST confirmed that only astrocytes in culture were stained by this antibody. An antibody was therefore generated that was specific for an extracellular epitope of GLAST (GLAST152-172) expressed on the surface of the astrocyte. However, when this antibody was tested for its ability to kill astrocytes in culture, both oligodendrocytes and astrocytes were killed. Further investigation revealed that the GLAST152-172 antibody labelled both oligodendrocytes and astrocytes in vitro and in vivo.
As there are various splice variants of GLAST known, their expression in oligodendrocytes in vivo and in vitro was tested in Chapter 4. Two of the splice variants, GLAST 1a and 1c, were found to be expressed in both oligodendrocytes and astrocytes, whereas 1b was specific for astrocytes. Because GLAST was also expressed on oligodendrocytes, it was not a useful target to try to specifically kill astrocytes. Therefore, in Chapter 5, the use of a chemical (DL-amino adipic acid or DL-AAA), which has been widely reported in the literature to specifically kill astrocytes, was investigated. Unexpectedly, however, DL-AAA killed the oligodendrocytes in the mixed cultures earlier than astrocytes. DL-AAA works through a cysteine-glutamate antiporter (C-GAP), which had only ever been reported to be expressed on astrocytes. However, immunocytochemistry using an antibody specific for C-GAP showed that this molecule is expressed on oligodendrocytes in culture.
Because it appeared that it might prove much more difficult than anticipated to specifically kill astrocytes in vitro, it was decided to return to testing MS patient sera. Patient sera was tested for several potential antibody-mediated mechanisms. Firstly, sera were screened for their ability to label cells in culture and, in the presence of complement, to kill oligodendrocytes and/or astrocytes. Some sera were found to have this potential. In addition, given the earlier results showing expression of glutamate transports on oligodendrocytes and astrocytes, the MS sera were also tested in ELISA to determine their ability to bind to the glutamate transports; some sera showed reactivity to glutamate transporter proteins. It was shown that almost all MS sera could opsonise human myelin, thereby increasing its uptake by macrophages
This thesis therefore showed that a mature mixed glial cell culture could be generated and used to study some functional properties of antibodies from MS patient sera. In addition, novel findings showing expression of a range of molecules involved in glutamate uptake and exchange on oligodendrocytes, which could potentially also be targets of autoimmune attack in MS.