Motor neuron TDP-43 proteinopathy in progressive supranuclear palsy and corticobasal degeneration

Transactive response DNA-binding protein 43 kDa (TDP-43) is mislocalized from the nucleus and aggregates within the cytoplasm of affected neurons in amyotrophic lateral sclerosis (ALS) cases. TDP-43 pathology has also been found in brain tissues under non-ALS conditions, suggesting mechanistic links between TDP-43-related ALS (ALS-TDP) and various neurological disorders. This study aimed to assess TDP-43 pathology in the spinal cord motor neurons of tauopathies. We examined 106 spinal cords from consecutively autopsied cases with progressive supranuclear palsy (PSP, n = 26), corticobasal degeneration (CBD, n = 12), globular glial tauopathy (GGT, n = 5), Alzheimer’s disease (AD, n = 21), or Pick disease (PiD, n = 6) and neurologically healthy controls (n = 36). Ten of the PSP cases (38%) and seven of the CBD cases (58%) showed mislocalization and cytoplasmic aggregation of TDP-43 in spinal cord motor neurons, which was prominent in the cervical cord. TDP-43-aggregates were found to be skein-like, round-shaped, granular, or dot-like and contained insoluble C-terminal fragments showing blotting pattern of ALS or frontotemporal lobar degeneration (FTLD). The lower motor neurons also showed cystatin-C aggregates, although Bunina bodies were absent in hematoxylin-eosin staining. The spinal cord TDP-43 pathology was often associated with TDP-43 pathology of the primary motor cortex. Positive correlations were shown between the severities of TDP-43 and 4-repeat (4R)-tau aggregates in the cervical cord. TDP-43 and 4R-tau aggregates burdens positively correlated with microglial burden in anterior horn. TDP-43 pathology of spinal cord motor neuron did not develop in an age-dependent manner and was not found in the AD, PiD, GGT, and control groups. Next, we assessed splicing factor proline/glutamine rich (SFPQ) expression in spinal cord motor neurons; SFPQ is a recently-identified regulator of ALS/FTLD pathogenesis, and it is also reported that interaction between SFPQ and fused-in-sarcoma (FUS) regulates splicing of microtubule-associated protein tau exon 10. Immunofluorescent and proximity-ligation assays revealed altered SFPQ/FUS-interactions in the neuronal nuclei of PSP, CBD, and ALS-TDP cases but not in AD, PiD, and GGT cases. Moreover, SFPQ expression was depleted in neurons containing TDP-43 or 4R-tau aggregates of PSP and CBD cases. Our results indicate that PSP and CBD may have properties of systematic motor neuron TDP-43 proteinopathy, suggesting mechanistic links with ALS-TDP. SFPQ dysfunction, arising from altered interaction with FUS, may be a candidate of the common pathway.