Relationship of acute axonal damage, Wallerian degeneration, and clinical disability in multiple sclerosis

Abstract

Background: Axonal damage and loss substantially contribute to the incremental accumulation of clinical disability in progressive multiple sclerosis. Here, we assessed the amount of Wallerian degeneration in brain tissue of multiple sclerosis patients in relation to demyelinating lesion activity and asked whether a transient blockade of Wallerian degeneration decreases axonal loss and clinical disability in a mouse model of inflammatory demyelination.

Methods: Wallerian degeneration and acute axonal damage were determined immunohistochemically in the periplaque white matter of multiple sclerosis patients with early actively demyelinating lesions, chronic active lesions, and inactive lesions. Furthermore, we studied the effects of Wallerian degeneration blockage on clinical severity, inflammatory pathology, acute axonal damage, and long-term axonal loss in experimental autoimmune encephalomyelitis using Wallerian degeneration slow (Wld ^S ) mutant mice.

Results: The highest numbers of axons undergoing Wallerian degeneration were found in the perilesional white matter of multiple sclerosis patients early in the disease course and with actively demyelinating lesions. Furthermore, Wallerian degeneration was more abundant in patients harboring chronic active as compared to chronic inactive lesions. No co-localization of neuropeptide Y-Y1 receptor, a bona fide immunohistochemical marker of Wallerian degeneration, with amyloid precursor protein, frequently used as an indicator of acute axonal transport disturbance, was observed in human and mouse tissue, indicating distinct axon-degenerative processes. Experimentally, a delay of Wallerian degeneration, as observed in Wld ^S mice, did not result in a reduction of clinical disability or acute axonal damage in experimental autoimmune encephalomyelitis, further supporting that acute axonal damage as reflected by axonal transport disturbances does not share common molecular mechanisms with Wallerian degeneration. Furthermore, delaying Wallerian degeneration did not result in a net rescue of axons in late lesion stages of experimental autoimmune encephalomyelitis.

Conclusions: Our data indicate that in multiple sclerosis, ongoing demyelination in focal lesions is associated with axonal degeneration in the perilesional white matter, supporting a role for focal pathology in diffuse white matter damage. Also, our results suggest that interfering with Wallerian degeneration in inflammatory demyelination does not suffice to prevent acute axonal damage and finally axonal loss.

Keywords: Axonal damage; Experimental autoimmune encephalomyelitis; Multiple sclerosis; Wallerian degeneration; Wld S.

Fig. 1

Wallerian degeneration is most frequent in the PPWM of lesions with ongoing demyelinating activity. NPY-Y1R IHC was performed on multiple sclerosis biopsy and autopsy tissue samples containing active, early inactive, chronic active, and chronic inactive lesions (a–e). Double IHC of NPY-Y1R (red, a) with the pan-macrophage marker KiM1P (blue, a) showed very little evidence for Wallerian degeneration (arrows) in the actively demyelinating lesion edge of a multiple sclerosis biopsy (a, inset). The density of NPY-Y1R⁺ axonal profiles indicating Wallerian degeneration was significantly higher in PPWM areas of active and chronic active lesions compared to early inactive and chronic inactive lesions in both early and chronic multiple sclerosis patients (f). Significantly higher numbers of NPY-Y1R⁺ axonal profiles (arrows) were detected in the PPWM of early multiple sclerosis patients (biopsies; a–b, f) compared to chronic multiple sclerosis patients (autopsies; c–f). Error bars=SEM, *p < 0.05, **p < 0.01, ***p < 0.001. Scale bars; a 100 μm; b 25 μm, c–e 50 μm; (inset a, e) 10 μm

Fig. 2

The Wld ^S mutation does not influence the clinical severity of EAE. EAE severity in Wld ^S (n = 18) and WT (n = 19) mice was similar in both the acute and chronic disease stage. Mean disease score ± SD during 40 days after disease onset are shown. The graph contains scores from three independent experiments. The days after development of the first clinical symptoms are plotted on the x-axis

Fig. 3

Similar extent of inflammation and demyelination in Wld ^S and WT mice in acute and chronic MOG_35–55-EAE. The mean number of inflammatory infiltrates per spinal cord (SC) cross section (H&E; a–c; g–i), and the percentage of demyelinated white matter (LFB/PAS; d–f; j–l) did not differ between Wld ^S and WT mice in the acute as well as the chronic disease stage (d20 and d40). Error bars=SEM. Scale bars=(a, b, d, and e) 100 μm; (g, h, j, and k) 200 μm

Fig. 4

Similar extent of acute axonal damage but markedly reduced Wallerian degeneration in Wld ^S mice with acute EAE. Comparable densities of APP⁺ axonal profiles were found in spinal EAE lesions of Wld ^S and WT mice in both the acute and chronic disease stage (a–e). In contrast, significantly fewer NPY-Y1R⁺ axonal profiles were observed in Wld ^S as compared to WT mice in acute EAE (f–g and j). In the chronic stage, the densities of NPY-Y1R-immunoreactive axonal profiles were comparable between Wld ^S and WT mice (h–j). Error bars=SEM, *p < 0.05, n = 15 animals per group. Scale bars=(a–b) 100 μm; (inset b) 50 μm; (c–d) 200 μm; (f–g) 150 μm; (h–i) 200 μm

Fig. 5

NPY-Y1R does not co-localize with markers of axonal transport disturbance, neurofilament dephosphorylation or myelin proteins in EAE lesions. Immunofluorescence double labeling performed on EAE lesions in WT mice revealed only rare co-localization of NPY-Y1R (red) and early axonal transport deficits indicated by APP accumulation (green) (a–c, arrow). NPY-Y1R immunoreactivity also did not co-localize with phosphorylated neurofilaments (NF) of healthy axons (SMI31, green) (d–f) or de-, resp. non-phosphorylated NF predominantly found in damaged axons (SMI32, green) (g–i), but partly with the NF high molecular weight (NF-200, green) subunit (j–l, arrows). Double IHC of NPY-Y1R (n, red) with MBP (m, green) did not reveal any co-localization (o, inset). Inset in (o) shows NPY-Y1R⁺ axons, in part enwrapped by myelin, at high magnification (arrows). Scale bars=(a–o) 200 μm; (inset o) 10 μm

Fig. 6

Less axonal loss in the acute disease phase in Wld ^S mutant mice. SC cross sections of Wld ^S mutant and WT mice with EAE were stained with anti-SMI31 antibody (a) and Bielschowsky’s silver impregnation (b) showing axonal loss in white matter lesions 20 days (acute) and 40 days (chronic) after disease onset. Quantification of SMI31⁺ phosphorylated axons revealed that the densities of intact healthy axons were significantly higher in Wld ^S mice as compared to WT mice in acute stage disease (a i–ii, c). However, similarly dense SMI31⁺ profiles were observed in the chronic stage (a iii–iv; lateral funiculus, c). Densities of SMI31⁺ profiles significantly decreased in both Wld ^S and WT mice from the acute to the chronic disease stage (a, c). Similarly, Bielschowsky’s silver impregnation displayed significantly higher relative axonal densities in Wld ^S mice only in the acute stage (b i–ii, d); again, axonal loss in Wld ^S mice was significantly more pronounced in the chronic than in the acute disease stage (b, d). Yellow and red boxes in (a i–ii) are higher magnifications of lesion areas with SMI31 immunostaining in acute disease. Error bars=SEM, n = 15 animals per group, *p < 0.05. Scale bars=(a i–ii) 100 μm; (a iii–iv, b i–ii) 150 μm; (b iii–iv) 200 μm; (insets a i–ii) 25 μm