TDP-43 Cytoplasmic Translocation in the Skin Fibroblasts of ALS Patients

Abstract

Diagnosis of ALS is based on clinical symptoms when motoneuron degeneration is significant. Therefore, new approaches for early diagnosis are needed. We aimed to assess if alterations in appearance and cellular localization of cutaneous TDP-43 may represent a biomarker for ALS. Skin biopsies from 64 subjects were analyzed: 44 ALS patients, 10 healthy controls (HC) and 10 neurological controls (NC) (Parkinson's disease and multiple sclerosis). TDP-43 immunoreactivity in epidermis and dermis was analyzed, as well as the percentage of cells with TDP-43 cytoplasmic localization. We detected a higher amount of TDP-43 in epidermis (p < 0.001) and in both layers of dermis (p < 0.001), as well as a higher percentage of TDP-43 cytoplasmic positive cells (p < 0.001) in the ALS group compared to HC and NC groups. Dermal cells containing TDP-43 were fibroblasts as identified by co-labeling against vimentin. ROC analyses (AUC 0.867, p < 0.001; CI 95% 0.800-0.935) showed that detection of 24.1% cells with cytoplasmic TDP-43 positivity in the dermis had 85% sensitivity and 80% specificity for detecting ALS. We have identified significantly increased TDP-43 levels in epidermis and in the cytoplasm of dermal cells of ALS patients. Our findings provide support for the use of TDP-43 in skin biopsies as a potential biomarker.

Keywords: TDP-43; amyotrophic lateral sclerosis; biomarker; dermis; skin biopsy.

Figure 1

Cytoplasmic TDP-43 immunoreactivity was analyzed in at least two areas of epidermis, papillary and reticular dermis (dotted rectangular area), per case (left panel). In the right panel, representative confocal microscopy images of epidermis (A), papillary dermis (B) and reticular dermis (C) from ALS, HC and NC subjects are shown. TDP-43 was found profusely in epidermis in all cases, although with higher density in the ALS group (p < 0.001) compared to both HC and NC groups. In the dermis of HC and NC subjects, TDP-43 was mostly seen in the nucleus, whereas cytoplasmic aggregations were more present in ALS cases in papillary and reticular dermis (p < 0.001). ALS: amyotrophic lateral sclerosis; HC: healthy controls; NC: neurological controls.

Figure 2

Distribution of the amount of cytoplasmic TDP-43 content and cells with cytoplasmic TDP-43 of the three cohorts in the different skin layers. IR: immunoreactivity; ALS: amyotrophic lateral sclerosis; HC: healthy controls; NC: neurological controls.

Figure 3

Cytoplasmic TDP-43-positive cells (*) were identified as fibroblasts based on vimentin immunoreactivity. Representative confocal microscopy images of ALS dermal fibroblasts co-immunolabeled for TDP-43 (in green), vimentin (in red) and DAPI nuclear staining (in blue).

Figure 4

ROC curves of the percentage of TDP-43 immunoreactivity in both layers of the dermis (AUC 0.911, CI 95% 0.862–0.960, p < 0.001) and of the percentage of dermal cells with cytoplasmic TDP-43 labeling (AUC 0.867, CI 95% 0.800–0.935, p < 0.001) in ALS patients. ROC: receiver operation characteristic; AUC: area under the ROC curve.

Figure 5

Representative images of ALS skin sections immunohistochemically labeled for PGP 9.5 (in red) to identify the innervation and for TDP-43 (in green). In the images, subepidermal nerve plexus and intraepidermal nerve fibers are shown with no colocalization of TDP-43 immunoreactivity.