Basic Science and Pathogenesis.

Background: Inclusions of TAR DNA binding protein of 43kDa (TDP-43) constitute the main characteristic pathology in the majority (∼97%) of amyotrophic lateral sclerosis (ALS) cases and approximately 50% of patients with frontotemporal lobar degeneration (FTLD). TDP-43 is a nuclear RNA binding protein; however, in disease, it becomes hyperphosphorylated and/or insoluble, hindering its nuclear function in maintaining RNA homeostasis. Importantly, the incidence of TDP-43 proteinopathy extends to aging brains (LATE) and may be concomitant with Alzheimer's disease (AD) neuropathological changes (LATE/AD) in up to 70% of AD patients. Interestingly, LATE/AD cases are characterized by worse memory and greater hippocampal atrophy than AD alone. Therefore, TDP-43 dysfunction could have diverse, disease-specific effects on pathogenesis, making it imperative to elucidate the mechanism of TDP-43 deposition in the brain and its role in AD-related dementias (ADRDs).

Method: We used standardized molecular biology techniques and single-nuclei RNA sequencing to evaluate TDP-43 dysfunction across brain regions in both LATE/AD and FTLD-TDP, identify common and distinct molecular mechanisms involved in each ADRD, as well as assess the relevance of TMEM106B rs3173615, a genetic variation linked to risk of FTLD haplotype, in TDP-43 pathology.

Result: We detected misspliced, TDP-43-regulated cryptic RNAs in the amygdala and hippocampus of LATE/AD and FTLD-TDP. FTLD-TDP, but not LATE/AD, frontal cortex also showed significant accumulation of these aberrant transcripts. The topographic distribution of cryptic RNAs mimicked that of insoluble, phosphorylated TDP-43, regardless of TDP-43 subtype classification. Despite common misregulated transcripts in these ADRDs, we also identified unique transcriptome changes, even within each TDP-43 subtype. Finally, we observed that TMEM106B core deposition, which is linked to the presence of the risk TMEM106B rs3173615 haplotype, was associated with enhanced TDP-43 dysfunction and pathology, and interactome data suggested a role for TMEM106B core filaments in impaired RNA transport, local translation, and endolysosomal function in FTLD-TDP.

Conclusion: Our results emphasize the use of cryptic RNAs to identify cases with TDP-43 pathology, and raises the possibility that unique transcriptome signatures may further distinguish FTLD-TDP from LATE/AD and within different TDP-43 subtypes. We showed a relationship between increased TMEM106B core deposition and greater susceptibility to TDP-43 dysfunction, potentially through dysregulation of endolysosomal function.