Axonal Injury Is The Major Determinant Of Permanent Neurologic Disability In Multiple Sclerosis

Even the earliest literature on MS mentioned axonal degeneration. Studies using contemporary technology such as magnetic resonance imaging and confocal microscopy more definitively demonstrated that axonal transection begins at disease onset and that cumulative axonal loss provides the pathologic substrate for the progressive disability that most long-term MS patients experience. Moreover, postmortem studies have shown that several histopathologic abnormalities including axonal loss, can be detected in the normal appearing white matter (NAWM)4 and cortical gray matter12 of patients with MS, suggesting a more diffuse pathology than previously thought.

Figure 1. Axons end in large terminal ovoids (arrow) indicating axonal transection during demyelination (From Trapp et al,4 with permission).

2.1 Axonal Injury Begins at the Early Stage of the Disease

Axonal amyloid precursor protein (APP) was demonstrated on acute MS lesions . Accumulation of this protein is considered a marker for axonal dysfunction or injury since it is detected immunohistochemically only in axons with impaired axonal trans-port14. Many APP-immunoreactive structures resembled axonal ovoids, characteristic of newly transected axons. Hence, these results suggested axonal dysfunction within inflammatory MS lesions and indicated that many of these axons were transected. These observations were confirmed and extended by morphological investigation of lesions from MS brains with various degrees of inflammation and disease duration4. Axonal ovoids were identified through confocal microscopy as terminal ends of transected axons immunostained for non-phosphorylated neurofilaments (Figure 1). Over 11,000 transected axons were found per mm3 in active lesions and over 3,000 per mm3 at the edge of chronic active lesions. The core of chronic active lesions contained on average 875 transected axons per mm3. In contrast, less than one axonal ovoid per mm3 was detected in control white matter. Kornek and colleagues reported a similar correlation between activity of MS lesions and density of APP-positive axons15. The occurrence of axonal ovoids in active lesions at an early stage of the disease supports axonal transection from the onset of MS.

The mechanism of axonal damage remains to be elucidated, and we can only speculate on the possibilities. Some of the possible mechanisms are summarized here. First, since the extent of axonal damage in active MS lesions is proportional to the degree of inflammatory activity within the lesion, axonal injury could be a direct result of inflammation per se (Figure 2). Substances such as free radicals, proteolytic enzymes, oxidative products and cytokines produced by activated immune and glial cells are potential mediators of such damage16. Oxidative damage to mitochondrial DNA and impaired activity of mitochondrial enzyme complexes in MS lesions indicate that inflammation can affect energy metabolism, ATP synthesis, and viability of affected cells17. Recently, data indicating that cytotoxic CD8+ T cells can mediate axonal transection in active MS lesions

18 19 20

were provided in MS tissue , in EAE mice and in vitro . Another observation is that

Canavan Disease

Figure 2. Axonal injury cause by inflammatory demyelination in an active MS lesion. Inflammatory substances secreted by activated immune and glial cells may mediate tissue damage including axonal transection. The distal axonal segment undergoes Wallerian degeneration but CNS myelin can persist for a long time. Thus, white matter distal to the active lesion may appear normal despite considerable axonal dropout (From Bjartmar et al.,32 with permission).

Figure 2. Axonal injury cause by inflammatory demyelination in an active MS lesion. Inflammatory substances secreted by activated immune and glial cells may mediate tissue damage including axonal transection. The distal axonal segment undergoes Wallerian degeneration but CNS myelin can persist for a long time. Thus, white matter distal to the active lesion may appear normal despite considerable axonal dropout (From Bjartmar et al.,32 with permission).

treatment with the AMPA/kainate glutamate receptor antagonist NBQX resulted in increased oligodendrocyte survival and reduced axonal damage in experimental autoimmune encephalomyelitis (EAE), an animal model of MS. This suggests that excitotoxicity mediated by glutamate is involved in tissue damage in acute lesions21. Inflammatory edema is a possible culprit as well. This may cause increased extracellular pressure that results in axonal damage, particularly in anatomical locations of the CNS where space for tissue expansion is limited such as the spinal cord22. In support of this hypothesis, the spinal cord cross-sectional area of relapsing-remitting EAE mice was shown to increase by 9% at first attack, but returned to normal at end-stage disease23. Finally, genes involved in axonal responses to inflammation and demyelination could determine the extent of axonal injury in individual patients.

Canavan Disease

Figure 3. Axonal loss in the spinal cord of a paralyzed patient with MS for 22 years. Neurofilament staining demonstrates axonal density in control (A) and in a demyelinated area (B), in the gracile fasciculus of MS cervical spinal cord. This chronic inactive lesion exhibits obvious axonal loss. (From Bjartmar et al,24 with permission).

Figure 3. Axonal loss in the spinal cord of a paralyzed patient with MS for 22 years. Neurofilament staining demonstrates axonal density in control (A) and in a demyelinated area (B), in the gracile fasciculus of MS cervical spinal cord. This chronic inactive lesion exhibits obvious axonal loss. (From Bjartmar et al,24 with permission).

2.2 Cumulative Axonal Loss is Seen in Chronic MS

Despite evidence of early axonal injury, most patients undergo a relapsing-remitting course. For years or decades after the first attack, they show no evidence of residual neurologic dysfunction in between relapses until a certain point when permanent disability ensues. From that point on, they all take a predictable downhill course. This interesting phenomenon is peculiar to MS and raises questions regarding the magnitude of cumulative axonal loss during chronic MS. To quantify total axonal loss in MS lesions, an ax-onal sampling protocol that accounts for both tissue atrophy and reduced axonal density was developed using spinal cord cross sections24. Total axonal loss was quantified in 10 chronic inactive lesions from 5 MS patients with significant functional impairment (EDSS > 7.5) and long disease duration. In these lesions, axons were reduced by an average of 68% (45-84%) compared to controls, while average axonal density (number of axons per unit area) was decreased by 58%. This demonstrates that axonal loss constitutes a significant part of the pathology that many chronic MS lesions develop (Figure 3).

Given that the patients are permanently disabled, the data also support axonal degeneration as the main cause of irreversible neurological disability in non-ambulatory MS patients. A similar reduction in axonal density, 61%, was reported in spinal cord lesions from patients with SP-MS25.

In SP-MS, extensive axonal loss and progression of disability occur in the absence of overt inflammatory activity. This suggests that mechanisms other than inflammatory de-myelination contribute to axonal degeneration. Recently, it was proposed that abnormal expression of sodium channel subtypes in response to demyelination may render axons vulnerable to degeneration, raising the possibility that MS may involve an acquired chan-nelopathy26. More importantly, a number of genes coding for myelin related proteins such as MAG, PLP, PMP22, Po and connexin 32, are being studied in relation to axonal pathology27. In one report, late onset axonal pathology such as atrophy or swelling, cy-toskeleton alterations, organelle accumulation and degeneration was observed in mice lacking MAG27 and PLP28. In PLP-null mice, the axonal pathology was accompanied with progressive clinical disability including impaired gait, tremor and spasticity. Hence, it is postulated that the lack of trophic support from myelin or myelin forming cells may cause degeneration of chronically demyelinated axons7, 29.

2.3 Axonal Pathology is Present in Normal Appearing White Matter

It is well established that axons once severed will undergo relatively rapid Wallerian degeneration distal to the site of transection. Unlike axons, CNS myelin can persist for a long time after proximal fiber transection. Histologically, such remaining myelin sheaths may appear as empty tubes or as degenerating ovoids. Despite this microscopic pathology, however, the white matter may appear normal grossly and on conventional neuroi-maging studies.

Immunohistochemical evidence suggesting Wallerian degeneration, such as discontinuous staining of axonal neurofilaments and presence of terminal axonal ovoids, has been demonstrated in normal appearing white matter from MS brains4. The extent of axonal loss in this region has been addressed quantitatively. Ganter et al.30 working in areas without plaque, reported reductions in axonal density by 19-42% at the lateral corticospi-nal tract of MS patients with lower limb weakness. Lovas and colleagues compared axonal density in lesions and in normal appearing white matter (NAWM) from the cervical spinal cords of SP-MS patients. The average reduction in axonal density in lesions from lateral and posterior columns was 61%25. In normal appearing white matter, however, the average decrease in axonal density was as much as 57%. They also noted that axons with diameter smaller than approximately 3 ^m were more affected than larger axons. In a study that accounted for both decreased axonal density and changes in tissue volume, total axonal loss in the corpus callosum of MS patients with disease durations between 5 and 34 years and various degree of functional impairment was determined31. An average total axonal loss of 53% in normal appearing corpus callosum was reported. Note however that in the same material, the reduction in axonal density was only 34%, emphasizing the need to consider both tissue volume and axonal density to properly assess the degree of axonal loss. These studies suggest that white matter may appear normal upon immunohistochemistry for myelin, or on MRI scans, but may still exhibit a considerable axonal dropout, especially in chronic patients with long disease duration.

Wallerian degeneration in normal appearing white matter has been observed by im-munohistochemistry in an MS patient with short disease duration32. The patient succumbed to a fatal brain stem lesion just after a 9-month-history of relapsing-remitting MS (RR-MS) with few permanent neurologic signs. Demyelinated lesions were not found in the spinal cord postmortem. However, the ventral spinal cord column, containing tracts projecting from the brainstem lesion, exhibited a 20% axonal loss. Microscopy revealed myelin ovoids (Figure 4) and signs of myelin degradation by activated microglia, characteristic of Wallerian degeneration. Since much of the myelin remains, these are 'invisible lesions' as far as MRI and immunostaining for myelin are concerned.

+1 0

Post a comment