PubMed1998
This monograph reviews the literature and presents experimental data on the intracerebral presentation of antigen(s) to the immune system as a consequence of neuronal cell death. "Which cells are the antigen presenting cells (APC) of the brain?" is the main question of this book. The immune surveillance of the CNS occurs through specialized resident cells, which present (auto)antigen(s) to the immune system and thus initiate an (auto)immune response. There are four established prerequisites necessary to identify resident APC of the brain. First, the APC must be capable to phagocytose dead neurons. Second, in order to be recognized by T lymphocytes, these neuronophages must express Major Histocompatibility Complex (MHC) cells II glycoproteins on their surface. Third, in order to present (auto)antigen, the MHC class II-positive neuronophages must also be able to contact T lymphocytes. Fourth, in order to exert a stimulatory effect on T lymphocytes, the APC should be able to produce the cytokine interleukin-1 beta (IL-128 Mb). Three main tools were used to identify and characterize the APC of the brain. First, a lesion model was employed that yields a slowly progressing neuronal cell loss without disruption of the blood-brain barrier. This model consisted of resection of 10 mm of the facial nerve, which caused a slowly occurring neuronal death so that one year after resection the amount of facial neurons was about 44% of the control value. Second, neuronophages were labeled in vivo in situ via phagocytosis of the permanent fluorescent marker Fluoro-Gold (FG) from decaying pre-loaded facial motoneurons. Third, the FG-labeled neuronophages were immunocytochemically characterized with the new method "immunoquenching of fluorescence". Sections of the brainstem containing FG-labeled, i.e. fluorescent, neuronophages were incubated with a variety of primary antibodies, followed by avidin-HRP and DAB-nickel as a dark brown reaction product for bright-field microscopy. In the fluorescent mode this DAB reaction product selectively quenches the fluorescence of all immunopositive cells, i.e. only those neuronophages that do not bind to the primary antibody remain fluorescent. Combining FG-labeling of neuronophages with immunoquenching, a population of small round fluorescent cells was discovered, localized in the immediate vicinity of the motoneurons long after the neuronofugal migration of microglia. As the fluorescence of these cells was not quenched after a triple immunostaining with anti-neuronal-specific enolase, anti-GFAP and OX-42 (quenching all fluorescence from neurons, astroglia, and microglia), they seem to represent a new, immunologically unidentified neuronophage. Following this triple immunostaining, a broad panel of antibodies was tested to stain, quench fluorescence, and thus immunotype these enigmatic phagocytes. Only the monoclonal antibody ED2, the classical marker for perivascular cells, specifically stained the small round neuronophages. Although the perivascular cells are in the vicinity of the basal lamina of the cerebral vasculature, they must not be confused with the pericytes, which are not able to perform phagocytosis. In contrast, the perivascular cells are macrophages-ED2 recognizes an established macrophage membrane antigen. In addition, after neuronal injury a subset of the perivascular cells starts to synthesize MHC class II glycoproteins and IL-1 beta. Hence this population of cells seems to possess the complete machinery required for antigen presentation: They are macrophages, upregulate MHC class II molecules and IL-1 beta, and due to their anatomical location, have access to circulating T lymphocytes. What was still lacking, however, was a direct proof of neuronophagia. Our experiments provided this proof. (ABSTRACT TRUNCATED)