ABSTRACT
Recent work from our laboratory has shown that the spread of damage after a traumatic injury to the central nervous system (CNS) can be slowed down by a controlled adaptive immune response. The immune response is mediated by T-cells directed against a CNS-associated self-antigen, such as myelin basic protein (MBP),1-3 myelin oligodendrocyte protein (MOG), or proteolipid protein (PLP), or against peptides (with or without encephalitogenic activity) derived from these proteins.4 T-cells directed against encephalitogenic epitopes were as effective as those directed against cryptic epitopes in displaying neuroprotection,2,4 indicating that the observed neuroprotection was not related to the virulence of the autoimmune response. The response could be achieved either by active immunization with the relevant proteins or peptides or by passive transfer of T-cells activated by them.5 On the basis of these findings, we suggested that autoimmune Tcells can protect CNS neurons from the post-injury spread of damage. We further showed that the neuroprotective autoimmunity is not only the result of an experimental manipulation, but is an endogenous response that is stimulated by the damaged neurons, though apparently not strongly enough to be effective (Yoles et al, unpublished results). It thus appears that protective autoimmunity is a physiological mechanism whereby the body attempts to cope with trauma-related nerve damage to the nervous system, but-presumably because of an evolutionary trade-off-the recruited autoimmune response, in its natural state, is neither timely nor effective.6-8
The beneficial autoimmunity can, in principle, gain access to the damaged tissue at any time, as even the healthy CNS is permissive to surveillance by T-cells, which (unlike immunoglobulins or
macrophages) are not restricted by the blood-brain barrier. We found that T-cells which patrol the CNS accumulate preferentially at sites of injury.2,9
In individuals suffering from a neurodegenerative disease, it is conceivable that at any given time some neurons have already degenerated and died, some are actively undergoing degeneration, and some are still healthy or only marginally damaged but, in the absence of therapeutic intervention, will inevitably succumb to secondary degeneration.10-13 This progressive spread of damage occurs not only in chronic degenerative diseases, but also after acute traumatic injuries to the CNS, where the functional outcome is often more severe than might be expected from the severity of the injury. Intensive research has therefore been devoted to the problem of progressive degeneration, in an attempt to understand the underlying mechanisms and develop therapies to arrest or retard the spread of damage. Many of these studies employ animal models of acute injury to the optic nerve or spinal cord. Since the spread of damage in neurodegenerative diseases may be viewed as the outcome of a continuous series of acute mini-injuries, findings in the acute injury model are expected to be applicable to chronic syndromes.14,15
One of the compounds responsible for neuronal losses after CNS injury is glutamate, an amino acid which normally serves as a ubiquitous neurotransmitter in brain functions such as learning and
however, it can become cytotoxic. Extracellular glutamate is normally buffered via uptake by astrocytes, and is recycled after being converted to glutamine.21-23 Under abnormal conditions, caused for example by acute or chronic CNS insults, the local buffering capacity in the CNS is apparently unable to control the inevitable increase in glutamate. Studies in our laboratory have shown that glutamate, when injected intravitreally at different concentrations into rats, exerts a dose-dependent effect on the recruitment of the immune system to counteract the cytotoxicity. This was demonstrated by comparing retinal ganglion cell (RGC) death in normal mice and mice devoid of T-cells (nude mice) from a strain capable of producing a protective T-cell-mediated response (Kipnis et al, unpublished results; Schori et al, unpublished results; Yoles et al, unpublished results). Interestingly, absence of T-cells was correlated with greater neuronal loss.
