Skin α-synuclein deposits differ in clinical variants of synucleinopathy: an in vivo study

Abstract

We aimed to characterize in vivo α-synuclein (α-syn) aggregates in skin nerves to ascertain: 1) the optimal marker to identify them; 2) possible differences between synucleinopathies that may justify the clinical variability. We studied multiple skin nerve α-syn deposits in 44 patients with synucleinopathy: 15 idiopathic Parkinson's disease (IPD), 12 dementia with Lewy Bodies (DLB), 5 pure autonomic failure (PAF) and 12 multiple system atrophy (MSA). Ten healthy subjects were used as controls. Antibodies against native α-syn, C-terminal α-syn epitopes such as phosphorylation at serine 129 (p-syn) and to conformation-specific for α-syn mature amyloid fibrils (syn-F1) were used. We found that p-syn showed the highest sensitivity and specificity in disclosing skin α-syn deposits. In MSA abnormal deposits were only found in somatic fibers mainly at distal sites differently from PAF, IPD and DLB displaying α-syn deposits in autonomic fibers mainly at proximal sites. PAF and DLB showed the highest p-syn load with a widespread involvement of autonomic skin nerve fibers.

In conclusion: 1) p-syn in skin nerves was the optimal marker for the in vivo diagnosis of synucleinopathies; 2) the localization and load differences of aggregates may help to identify specific diagnostic traits and support a different pathogenesis among synucleinopathies.

Figure 1

Abnormal intraneural p-syn aggregates in non-synaptic and synaptic fibers. Examples of phosphorylated α-synuclein aggregates in a non-synaptic fiber of a patient with MSA (A) and synaptic fibers of a DLB patient (B) disclosed by confocal microscope (x400). (A)Subepidermal plexus close to the epidermis as confirmed by an isolated epidermal free-ending PGP-ir fibers (arrow) was identified by a PGP staining (A). The plexus showed a positive phosphorylated α-syn (A^I) as neuritic inclusion demonstrated by the merged image (A^II). (B)Nerve fibers innervating a sweat gland tubule were depicted by PGP (B). Some of these fibers showed aggregates of p-syn (B^I) as intraneural inclusions in the merged image (B^II).

Figure 2

NAC and syn-F1 co-localization in a healthy control. Confocal microscope images (X 200) showing a weak syn-F1 staining in sweat gland of healthy control. The syn-F1 staining co-localized with NAC (arrows in A^II) demonstrating that this signal is likely a non-specific signal arising from native α-syn. The asterisks represents sweat tubules autofluorescence.

Figure 3

Abnormal α-syn aggregates characterized by a combination of different primary antibodies. Confocal microscope study (x400) of α-syn aggregates in patients with different variants of synucleinopathy based on a co-localization between p-syn and a neuronal marker (i.e. PGP) or antibodies against abnormal α-syn epitopes expression of C-terminal post-translational modifications or amyloid fibrils (syn-F1). (A) P-syn demonstrated an excellent co-localization with PGP (A^I) in a nerve plexus supporting the intraneural deposition of abnormal α-syn aggregates (A^II). (B) The co-localization in a nerve plexus between p-syn and syn-F1 (B^I) found in the majority of analysed deposits supporting the fibrillar nature of these aggregates (B^II). (C) Sudomotor fibers around a sweat tubule marked by NAC (C^I) were co-localized with p-syn (arrow in C^II) although other sudomotor fibers stained by p-syn were devoid of NAC staining (asterisk in C^II). (D) NY-syn staining was occasionally seen in non-synaptic fibers (D^I) and this staining co-localized with p-syn (D^II). The four different coexistent fibrillar and non-fibrillar α-syn deposits found in skin nerves were similarly distributed among different clinical phenotypes. Nevertheless, these deposits showed important differences in specific variants of synucleinopathy such as their localisation (i.e. only in somatosensory skin fibers in MSA - see Table 2) or the widespread involvement of autonomic annexes (i.e. in PAF and DLB - see Fig. 4). These differences may support a different pathogenesis among synucleinopathies helping to identify specific diagnostic traits.

Figure 4

Distribution of intraneural abnormal α-syn deposits in autonomic annexes. The pattern of p-syn distribution among autonomic annexes disclosed a non-significant difference of deposits around skin vessels (SV) in IPD, PAF and DLB. By contrast abnormal α-syn deposits were significantly higher in sweat glands (SG) and muscle arrector pilorum (MAP) of PAF than in IPD whereas DLB showed an intermediate degree of involvement. These results underlined a widespread extension of deposits in autonomic annexes of patients with autonomic symptoms such as PAF and DLB. ***p < 0.001; °p = 0.07.

Figure 5

P-syn occurrence in consecutive skin sections of a PAF patient. Immunoreactivity of α-synuclein phosphorylated at serine 129 (p-syn) in a single nerve fiber in a broad skin area was analysed considering 6 consecutive free-floating thick skin sections of 50 μm of the same skin sample. A patient with PAF showed a persistent occurrence of p-syn staining in all consecutive skin sections supporting a regular distribution along skin nerves.

Figure 6

Proximal/distal p-syn gradient. The figure illustrates the different pattern of p-syn distribution throughout proximal and distal skin sites in clinical variants of synucleinopathy. (A) PAF showed a homogenous p-syn positivity in proximal and distal sites. (B) IPD (corrected p < 0.01) and DLB (p = 0.05; corrected p = 0.1) displayed a p-syn proximal-distal gradient with higher positivity in proximal sites, mainly the cervical area. (C) MSA showed an opposite pattern of skin nerve p-syn with higher positivity in distal skin sites although the difference was not significant (p > 0.4).