Fig 2: Histological changes in TIA1 mutation carriers. Cross section of spinal cord showing severe loss of myelin stain in the corticospinal tracts (CST) (a). Lower motor neurons containing Bunina bodies (arrow) (b) and large, round cored Lewy body-like inclusions (LBLI) (c, d) or round eosinophilic inclusions without distinct cores (e) were present in the medulla and spinal cord. Extra-motor pathology included chronic degeneration with superficial, laminar microvacuolation of the prefrontal cortex (f). a and e, HE/LFB stain; b - d and f, HE stain. Scale bar: a, 1200 µm; b, 7 µm, c, 23 µm; d and e, 15 µm; f, 205 µm

Fig 3: TDP-43 immunoreactive pathology in TIA1 mutation carriers. Numerous predominantly granular TDP-43 immunoreactive (TDP-ir) neuronal cytoplasmic inclusions (NCI, arrows) were present in the prefrontal cortex (a) and primary motor cortex (b). Hippocampal dentate granule cells (c) and dopaminergic neurons in the substantia nigra (d) were consistently affected in all cases. Lower motor neurons (LMN) of the medulla and spinal cord (e – i) contained NCI that were granular (e), filamentous (f) or round and compact (g and h). Single LMN containing combinations of NCI types were not uncommon (i, arrow points to filamentous inclusions in close proximity to a compact, round NCI). Phosphorylation-independent TDP-43 immunohistochemistry. Scale bar: a and i, 25 µm; b, c and e, 18 µm; d and g, 36 µm; f and h, 9 µm

Fig 4: Pedigree of family UBCU2. Family of European ancestry showing an autosomal dominant pattern of inheritance of ALS ± dementia. Black symbols represent clinically affected individuals and diagonal lines indicate those who are deceased. Genetic analysis was performed on the proband (1), her affected niece (14) and her early affected sister (2); all of whom carried the P362L mutation in TIA1

Fig 5: Quantitative comparison of spinal cord lower motor neurons (LMN) and different types of neuronal cytoplasmic inclusions (NCI) in LMN in ALS cases with TIA1 mutations, the C9orf72 mutation (C9) and sporadic ALS (sALS). Quantitation in a-c was performed on HE stained sections; whereas d–f was based on TDP-43 immunohistochemistry. There was no difference in the number of spinal cord LMN among groups (a). Round NCI (b) and Lewy body like inclusions (c) were significantly more frequent in TIA1 mutation carriers than in either C9 or sALS cases. In c, the red circle denotes the data point corresponding to the case with only a single section of upper cervical cord available for evaluation. No differences were seen among groups in the number of granular (d) or filamentous (e) TDP-43 NCI; however, round compact NCI were more frequent in TIA1 mutation carriers (f). (** = p < 0.005, * = p < 0.05 compared to C9 and sALS groups, respectively)