Significant contralaterality of temporal-predominant neuroastroglial tauopathy and FTLD-TDP type C presenting with the right temporal variant FTD

To the Editor: Temporal-predominant neuroastroglial tauopathy is a recently described 4-repeat tauopathy with a unique morphology of tau-positive astrocytes (1, 2). It is associated with TDP-43 proteinopathy and a strong correlation with TMEM106B rs1990622 A/A genotype has been reported (1). Here, we report an autopsy case with a clinically rare right temporal variant frontotemporal dementia (rtvFTD), and the concomitant presence of tau and TDP-43 pathology showing distinct severity in the contralateral hemisphere. A previously healthy right-handed Caucasian man without a history of familial neurodegenerative diseases noticed the insidious onset of progressive word-finding in his native French language and English at the age of 72. He made several semantic paraphasic errors in spontaneous discourse. At age 75, he was assessed, and impairments in phonemic fluency and naming were found. Neurological examination at this time was normal. At age 76, he developed phonagnosia, decline in personal hygiene, perseverative behavior, and a further decline in language characterized by worsening anomia, mild impairment in semantic knowledge, and impaired semantic fluency relative to phonemic fluency, and eventually prosopagnosia. At age 78, he suffered a right internal capsule infarct causing a residual mild left-arm upper motor neuron pattern of weakness. He could not recognize his family members and language skills continued to decline. He died at age of 81. Serial magnetic resonance imaging (MRI) of his brain (ages 66, 75, and 78) revealed progressive right more than left anterior-inferior temporal lobe atrophy (Fig. 1A) and, to a lesser extent, left more than right parietal cortical atrophy. Cerebrospinal fluid examination for β-amyloid and tau was not consistent with Alzheimer disease (AD). The clinical diagnosis was rtvFTD. Distribution of neurodegeneration and laterality. (A) Brain MRI (FLAIR, at age of 78) shows severe atrophy in the anterior temporal lobes, especially on the right side (arrowheads: the anterior part of temporal lobe, asterisk: the inferior horn of lateral ventricle in the level of amygdala and hippocampus). (B) In the temporal lobe, the middle temporal gyrus reveals more severe degeneration including myelin pallor (arrow), in comparison with the superior temporal gyrus (asterisk). (C–H) Neuronal loss with gliosis (C: left, D: right, in the middle temporal gyrus) and TDP-43 pathology (E: left; F: right, in the fusiform gyrus) are predominant on the right side, but tau pathology (G: left, H: right, in the fusiform gyrus) is predominant on the left side. (I) Heat map of neuronal loss (upper), TDP-43 pathology (middle), and tau pathology (lower). The severity of each pathology ranges from white (none: 0) to light (mild: 1, moderate: 2) and dark colors (severe: 3, extremely severe: 4). Gray colored regions indicate that the region was not evaluated. Asterisks indicate complications of AD-associated neurofibrillary pathology. Rt, right; Lt, left. Luxol fast blue/hematoxylin and eosin (B–D), p-TDP-43 immunostaining (E, F), and p-tau (AT8) immunostaining (G, H). Scale bars: (B) = 5 mm; (C–H) = 100 μm. Autopsy revealed frontotemporal brain atrophy with right predominance, especially in the temporal pole, fusiform gyrus, and the inferior and middle temporal gyrus (Fig. 1B). The fixed brain weight was 1044 g. In the above regions, there was severe neuronal loss and gliosis that was more prominent on the right side (Fig. 1C, D, I [upper]). The hippocampus appeared atrophic on both sides, and severe neuronal loss and gliosis were observed in the CA1 and subiculum, representing hippocampal sclerosis. The subcortical nuclei, brainstem, and spinal cord were relatively preserved (Fig. 1I [upper]). We performed systematic immunostainings of several anatomical regions (Table). In the above lesions, many long and thick phosphorylated (p)TDP-43-positive dystrophic neurites and occasional neuronal cytoplasmic inclusions (NCIs) were evident, which corresponds to FTLD-TDP type C (3). Neuronal intranuclear inclusions and astrocytic inclusions were not present. This TDP-43 pathology predominated on the right side (Fig. 1E, F, I [middle] and Fig. 2F). The clinicopathological interpretation was rtvFTD with FTLD-TDP type C (4, 5). As in the present case, rtvFTD patients often exhibit behavioral changes, prosopagnosia (5), and sometimes phonagnosia (6). In particular, it has been reported that a deficit of facial and voice recognition is associated with atrophy of the right anterior temporal lobe and fusiform gyrus (5–7). Histopathological features of cytoplasmic inclusions. (A) Tau pathology in the left temporal cortex. Many tau-positive astrocytes (arrows) with granular back grounds are present. Tau-positive astrocytes are composed of diffuse granular tau accumulations in both perikarya and processes (B). Tau-positive neurons also reveal granular morphology (C). Many tau-positive thick neurites with granular back grounds (D) and patchy accumulation of threads/neurites (E) are features. (F) There are also many p-TDP-43-positive thick and long neurites, which corresponds to FTLD-TDP type C. Some thick neurites are similar to the morphology of tau-positive thick neurites. (G) In the temporal white matter, some focal tau lesions are seen (G, left, asterisk); which consists of granular tau-deposits in astrocytes (G, right) and threads. (H–K) Immunoreactivity for various tau antibodies in the temporal cortex. 4-repeat tau specific antibody (RD4) detects tau-positive astrocytes (H) and other inclusions. These tau-positive inclusions are also AT100 positive (I), but Alz-50/MC-1 negative (J). A small number of NCIs and threads are detected for Tau C3 antibody (K). Some inclusions also show p62-immunoreactivity (L). (M–O) Tau pathology of the Ammon’s horn. Several tau-positive neurons and astrocytes are evident in the dentate gyrus (M). In the CA1 region, p-tau (AT8) antibody detects many tau accumulations (N), but Alzheimer pathology which is shown by GT-38 antibody, is partial (O). (P) Summary of staining profiles of the inclusions. They are evaluated in the inferior temporal gyrus using the following criteria. -, not stained; ±, slightly/inconsistently stained; +, consistently stained. (Q–S) Double-label immunofluorescence of p-tau: p-Thr217 (red) and p-TDP-43 (green) in the left temporal cortex reveals solely tau-positive in the astrocytes (Q), and both positive in the neurons (R) as well as many neurites (S, arrowheads). In addition, some neurites show solely p-TDP-43 positive (arrow), and some neurites/threads show solely p-tau positive (white circle). The sections were treated with Sudan black B and sodium borohydride to suppress autofluorescence. Immunostaining for AT8 (A-E, G, M, N), p-TDP-43 (F), RD4 (H), AT100 (I), Alz-50 (J), Tau C3 (K), p62 (L), and GT-38 (O). Scale bars: A, G (left), N, O = 100 μm; B, C, E, G (right), H, I, J, K, L, S = 20 μm; D, F, M = 50 μm; Q, R = 10 μm. Summary of the antibodies used in this study Immunostaining was performed using the Dako Autostainer Link 48 and EnVision FLEX+ Visualization System, according to manufacturer’s instructions. All sections were counterstained with hematoxylin. p-, phosphorylated; FA, formic acid. *Heat-mediated antigen retrieval was performed incubating the slides with low pH solution (citrate buffer pH6) during 20 minutes at 97°C (Dako PT Link with low pH solution). Summary of the antibodies used in this study Immunostaining was performed using the Dako Autostainer Link 48 and EnVision FLEX+ Visualization System, according to manufacturer’s instructions. All sections were counterstained with hematoxylin. p-, phosphorylated; FA, formic acid. *Heat-mediated antigen retrieval was performed incubating the slides with low pH solution (citrate buffer pH6) during 20 minutes at 97°C (Dako PT Link with low pH solution). In contrast in this case, tau pathology was predominant on the left side (Fig. 1G, H, I [lower]), and accentuated in the temporal lobe, namely in the temporal pole, fusiform gyrus, inferior and middle temporal gyri, hippocampus, and amygdala. There were many tau deposits in the left insular cortex and claustrum, and moderate amounts in the anterior cingulate cortex. They involved other cerebral cortices, subcortical nuclei, and the brainstem. Tau pathology was composed of many astrocytic inclusions, many neurites, and fewer NCIs (Fig. 2A). Morphologically, p-tau deposits featured diffuse granular immunopositivity in astrocytic processes and perikarya (Fig. 2B), which resembled granular or fuzzy astrocytes. As well, granular neuronal cytoplasmic inclusions (Fig. 2C), thick neurites (Fig. 2D), and patchy accumulation of threads/neurites (Fig. 2E) were distinctive. Interestingly, some tau-positive thick neurites were morphologically similar to TDP-43-positive thick neurites (Fig. 2F). Although tau pathology was prominent in the cerebral cortex, some threads and a few coiled bodies were present in the cerebral white matter. In addition, there were several tau-positive astrocytes emphasized at the subcortical white matter in the left temporal lobe (Fig. 2G) reminiscent of white matter aging-related tau astrogliopathy (ARTAG) (8). Tau-positive inclusions showed positivity for 4-repeat (R) tau-, but not for 3-repeat tau-specific antibodies (Fig. 2H, P). Interestingly, they revealed positivity for phosphorylated tau antibodies (AT8, AT100, and p-tau Thr 217), but not for Alz-50/MC-1 antibodies against conformational changes of tau. Positivity of Tau C3, detecting truncated tau, was rare in the neurons and neurites (Fig. 2I–K, P). Altogether, these tau staining profiles suggest distinct processing of pathological tau. Furthermore, Gallyas-Braak silver staining revealed only focal positivity in some thick neurites. Some inclusions also showed p62-immunoreactivity (Fig. 2L, P). Regarding mixed pathologies, Lewy body pathology was not seen. Alzheimer disease (AD) related neuropathological change was mild (Braak stage: II/C, AD neuropathologic change: low [A3B1C1]) (9). In the hippocampus, tau pathology involved the granule cells of the dentate gyrus (Fig. 2M), which is usually seen over Braak NFT stage IV (10) and a few 3R and 4R-tau-positive NFTs were seen in the entorhinal cortex and hippocampus. The AD NFT-specific GT-38 antibody (11) labeled significantly fewer NCIs in comparison with p-tau-positive NCIs (Fig. 2N, O). Since TDP-43 and tau pathology overlapped anatomically (Fig. 1I), we performed double-label immunofluorescence. This revealed that the astrocytic inclusions were solely tau-positive, but the NCIs and neurites often showed colocalization of p-TDP-43 and p-tau (Fig. 2Q–S). As a limitation, and due also to the presence of combined proteinopathies (TDP-43 and tau), our study cannot address whether the astroglial tau pathology alone is associated with evidence for glial cell toxicity. Genetic analyses revealed a TMEM106B rs1990622 A/G genotype, no mutations in the MAPT gene (H1/H1 haplotype), and APOE ε2/ε3 genotype. The present study demonstrates marked contralaterality of the 4R temporal-predominant neuroastroglial tauopathy (1, 2) and FTLD-TDP type C. This case shows the concomitant presence of 2 rare neurodegenerative proteinopathies. Both pathological protein accumulations overlap within some lesions (i.e. left temporal lobe), and we often found colocalization of p-tau and p-TDP-43 in the NCIs and neurites. This suggests that both pathological proteins may be accumulating in response to each other. In fact, interactions between tau and TDP-43 have been observed in several neurodegenerative diseases, for example in corticobasal degenerations, where they sometimes overlap and affect clinical symptoms (12). Although the cause of the highly frequent co-occurrence of 2 different pathological proteins is unknown in this atypical tauopathy, in a previous study Llibre-Guerra et al (1) reported that polymorphisms in TMEM106B strongly correlate with this atypical tauopathy. Interestingly, TMEM106B is involved in the lysosomal system and polymorphisms also correlate with several neurodegenerative diseases, including both tauopathy and TDP-43-proteinopathy (i.e. FTLD-TDP, limbic-predominant age-related TDP-43 encephalopathy, hippocampal sclerosis, AD, and chronic traumatic encephalopathy) (13). The previous study showed that over 90% of neuroastroglial tauopathy-positive cases carry the TMEM106B rs1990622 A/A genotype; however, it is noteworthy that a small number of A/G genotype cases are evident in the present and previous studies (1). There may be other common risk factors associated with several proteinopathies besides the TMEM106B rs1990622 A/A genotype. Although the pathomechanism is not clear, asymmetric neurodegeneration in proteinopathies is sometimes observed in both tau and TDP-43 pathology (14). The cause of this may be related to the propagation of pathological proteins via the neuroanatomic connectome (15). In the present case, 2 pathological proteins might have occurred by common pathogenesis and/or interactions, and then, developed independently. Further study of cases with concomitant neurodegenerative proteinopathies, such as in the present case, may be important in understanding both specific and common mechanisms that cause neurodegeneration. This study was supported by the Edmond J. Safra Philanthropic Foundation, the Rossy Family Foundation, and Canada Foundation for Innovation grant number 40480. The authors would like to thank the patient and his family who made this research possible. The authors have no duality or conflicts of interest to declare.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.