Extent and Timing of Axonal Injury in MS Lesions

The extent of axonal loss in established MS lesions is highly variable between different plaques within a single brain and even more variable between MS plaques from different patients (Fig.1). Some axonal loss is present in all MS lesions; the extent of axonal density reduction in plaques, however, may range from 20% to nearly 100%. Analysis of axonal density in relation to lesion development shows in

Figure I Axonal loss in multiple sclerosis lesions. (A) Subcortical white matter in a patient with primary progressive multiple sclerosis, stained for axons with Bieschowsky's silver impregnation. A small subcortical plaque (P) shows marked reduction in axonal staining, while an adjacent remyelinated shadow plaque (SP) reveals much higher axonal density; in addition there is diffuse axonal loss within the whole white matter (DWMI = diffuse white matter injury; C = cortex) (x 2). (B) Edge of the plaque shown in (A); note the profound reduction in axonal profiles in the plaque (P) compared to the adjacent white matter (WM) (x 100). (C) Small subcortical plaque in a patient with acute multiple sclerosis; extensive reduction in axonal density within the plaque (P). Bielschowsky's silver impregnation (x 5). (D) Edge of the plaque shown in (C); massive reduction of axonal profiles in the plaque and numerous axonal spheroids (acute axonal injury) at the plaque edge (x 300).

Figure I Axonal loss in multiple sclerosis lesions. (A) Subcortical white matter in a patient with primary progressive multiple sclerosis, stained for axons with Bieschowsky's silver impregnation. A small subcortical plaque (P) shows marked reduction in axonal staining, while an adjacent remyelinated shadow plaque (SP) reveals much higher axonal density; in addition there is diffuse axonal loss within the whole white matter (DWMI = diffuse white matter injury; C = cortex) (x 2). (B) Edge of the plaque shown in (A); note the profound reduction in axonal profiles in the plaque (P) compared to the adjacent white matter (WM) (x 100). (C) Small subcortical plaque in a patient with acute multiple sclerosis; extensive reduction in axonal density within the plaque (P). Bielschowsky's silver impregnation (x 5). (D) Edge of the plaque shown in (C); massive reduction of axonal profiles in the plaque and numerous axonal spheroids (acute axonal injury) at the plaque edge (x 300).

initial actively demyelinating lesions an average reduction of 30% to 40%. This reduction is only due in part to axonal loss, but also mediated by local edema and inflammatory infiltration of the tissue. Thus, in later stages of active lesions, as well as in remyelinated shadow plaques, axonal density in general is higher compared to that in acute lesions. In contrast, in permanently demyelinated lesions the reduction of axonal density is on average much more pronounced compared to fresh lesions, reaching average levels of 50% to 70% (Mews et al., 1998). The reason for this further reduction of axonal density in old demyelinated plaques may be twofold. First, the same areas in the CNS may become the target for repeated demyelinating attacks (Prineas et al., 1993), each bout leading not only to new demyelination, but also to further axonal injury and loss. In addition there is a slow burning axonal injury within old inactive demyelinated lesions, which in the long term may contribute to a major extent to the final axonal loss in established plaques (Kornek et al., 2000).

All these studies are based on simple determination of axonal density within a given tissue area; however, this only partially reflects true axonal loss and may lead to false estimates of axonal loss in acute lesions due to tissue edema or in old established plaques due to tissue atrophy. This problem was recently overcome by counting axons within defined tract systems (Bjartmar et al., 2000; Evangelou et al., 2000a). This method allows the determination of true axonal loss, independent from changes due to edema or atrophy. Overall, the values for axonal loss determined by this method were similar to those established by simple determination of axonal density, also showing an average axonal loss in established lesions in the range of 50% to 70%. This indicates that the variation in the extent of axonal loss between plaques of different stages or from different patients is larger than the effect of tissue edema or atrophy.

Another way to determine the extent of axonal damage in MS lesions is to study acute axonal injury at different stages of plaque formation (Ferguson et al., 1997). Earliest signs of axonal injury are reflected by a disturbance of fast axonal transport. Thus proteins or other molecules trafficking along the axons by fast axonal transport accumulate at the sites of axonal injury. A very reliable marker for the disturbance of axonal transport is the focal accumulation of amyloid precursor protein (P-APP). Disturbed axonal transport also leads to the formation of focal axonal swelling, which may occur either at sites of axonal transsection or uninterrupted axons (Fig. 2).

Using these tools to study acute axonal injury in MS plaques, it was found that massive axonal damage occurs during the phase of active demyelination in fresh lesions (Ferguson et al., 1997; Trapp et al., 1998; Kornek et al,. 2000; Bitsch et al., 2000). Thus, axonal loss and damage take place in every newly formed lesion, regardless of whether it develops during the earliest or at late stages of the disease. Analysis of brain biopsies taken in patients at the first bout of the disease even suggested that axonal injury at this stage may be more severe compared to that in lesions formed in the chronic stage (Kuhlmann et al., 2002). This, however, seems to be due to a sampling bias, as biopsies in patients with MS are rare and generally restricted to patients with very severe and atypical disease.

Besides the profound axonal injury, which occurs in actively demyelinating lesions, there is an additional slow burning axonal injury and loss, which is present in inactive demyelinated plaques (Kornek et al., 2000). Although the extent of axonal injury at a given time in these chronic lesions is more than hundred times lower compared to that in active plaques, it may be of major significance. Acute lesions develop within a few days to weeks, whereas chronic demyelinated plaques may persist in the CNS for many years. Thus even a smoldering ongoing axonal injury within chronic lesions may contribute a large share to the total axonal loss in the MS brain. This slow burning chronic axonal injury is absent when plaques are remyelinated.

Axonal injury and loss in MS brains are not restricted to demyelinated plaques (Evangelou et al., 2000b; Bjartmar et al., 2001). In particular in patients with primary or secondary progressive MS, a diffuse inflammatory process is found throughout the whole brain and spinal cord, thus also affecting the so-called normal white matter. This diffuse inflammatory process is associated with diffuse axonal injury and loss, which leads to fiber degeneration and secondary myelin destruction. This process is reflected by diffuse myelin pallor throughout the whole white matter. The extensive activation of local microglia and their expression of inducible nitric oxide synthase suggests that the diffuse white matter damage in chronic MS is mainly mediated through reactive oxygen and nitrogen intermediates.

Within the plaques not all axons are destroyed to the same extent. Thick axons are much better preserved than thin fibers (Evangelou et al., 2001). This could in part be explained by changes in axonal caliber in the course of demyelination. Indeed, it has been shown by light microscopy more than 100 years ago (Marburg, 1906), and later been confirmed by electron microscopy (Shintaku et al., 1988), that demyelinated internodes in MS lesions have thicker calibers than their respective myelinated axonal portions. In addition to a dominant reduction of thin fibers that has been found within the plaques, however, there is also a preferential loss of small neurons at the sites of axonal origin (Evangelou et al., 2001). This can most clearly be seen in the retina in patients with demyelinated plaques in the optic nerve. It thus seems that overall thin axons, originating from small neurons, are more vulnerable in MS lesions compared to thick axons. As discussed later, one major factor responsible for axonal disintegration within plaques seems to be mitochondrial dysfunction and energy failure (Smith and Lassmann, 2002). In such a situation, it is likely that small axons, which contain fewer mitochondria compared to thick ones, are affected more severely.

0 0

Post a comment