Aberrant deposition of stress granule-resident proteins linked to C9orf72-associated TDP-43 proteinopathy

Jeannie Chew,Casey Cook,Lillian M Daughrity,Monica Castanedes-Casey,Jeffrey D Rothstein,Tania F Gendron,John D Fryer,Karen Jansen-West,Giulia Del Rosso,Aishe Kurti,Matthew D Disney,Leonard Petrucelli,Jeannette N Stankowski,Yong-Jie Zhang,Dennis W Dickson

doi:10.1186/s13024-019-0310-z

Abstract

BackgroundA G4C2 hexanucleotide repeat expansion in the noncoding region of C9orf72 is the major genetic cause of frontotemporal dementia and amyotrophic lateral sclerosis (c9FTD/ALS). Putative disease mechanisms underlying c9FTD/ALS include toxicity from sense G4C2 and antisense G2C4 repeat-containing RNA, and from dipeptide repeat (DPR) proteins unconventionally translated from these RNA products. MethodsIntracerebroventricular injections with adeno-associated virus (AAV) encoding 2 or 149 G4C2 repeats were performed on postnatal day 0, followed by assessment of behavioral and neuropathological phenotypes.ResultsRelative to control mice, gliosis and neurodegeneration accompanied by cognitive and motor deficits were observed in (G4C2)149 mice by 6 months of age. Recapitulating key pathological hallmarks, we also demonstrate that sense and antisense RNA foci, inclusions of poly(GA), poly(GP), poly(GR), poly(PR), and poly(PA) DPR proteins, and inclusions of endogenous phosphorylated TDP-43 (pTDP-43) developed in (G4C2)149 mice but not control (G4C2)2 mice. Notably, proteins that play a role in the regulation of stress granules – RNA-protein assemblies that form in response to translational inhibition and that have been implicated in c9FTD/ALS pathogenesis – were mislocalized in (G4C2)149 mice as early as 3 months of age. Specifically, we observed the abnormal deposition of stress granule components within inclusions immunopositive for poly(GR) and pTDP-43, as well as evidence of nucleocytoplasmic transport defects.ConclusionsOur in vivo model of c9FTD/ALS is the first to robustly recapitulate hallmark features derived from both sense and antisense C9orf72 repeat-associated transcripts complete with neurodegeneration and behavioral impairments. More importantly, the early appearance of persistent pathological stress granules prior to significant pTDP-43 deposition implicates an aberrant stress granule response as a key disease mechanism driving TDP-43 proteinopathy in c9FTD/ALS.

Highlights

A G4C2 hexanucleotide repeat expansion in the noncoding region of chromosome 9 open reading frame 72 (C9orf72) is the major genetic cause of frontotemporal dementia and amyotrophic lateral sclerosis (c9FTD/ALS)
To determine whether behavioral abnormalities are associated with markers of neurodegeneration in (G4C2)149 mice, we evaluated astrogliosis and neuronal loss
Given that foci of antisense G2C4 transcripts are detected in postmortem brain tissues of c9FTD/ALS patients [12, 13, 18], and that inverted terminal repeats present within the associated virus (AAV) vector both 5′ and 3′ of the G4C2 repeat have been shown to exert promoter-like activity [16], we evaluated whether antisense G2C4 repeat RNA was present in (G4C2)149 mice by RNA fluorescence in situ hybridization (FISH)

Summary

Introduction

A G4C2 hexanucleotide repeat expansion in the noncoding region of C9orf is the major genetic cause of frontotemporal dementia and amyotrophic lateral sclerosis (c9FTD/ALS). Putative disease mechanisms underlying c9FTD/ALS include toxicity from sense G4C2 and antisense G2C4 repeat-containing RNA, and from dipeptide repeat (DPR) proteins unconventionally translated from these RNA products. Foci containing sense G4C2 or antisense G2C4 repeat RNA bidirectionally-transcribed from the C9orf expansion are observed in c9FTD/ALS postmortem brain tissues, cultured cells, and neurons, and may disrupt RNA metabolism and nucleocytoplasmic transport through sequestration of various RNA-binding proteins [11, 14, 18, 22, 23, 31, 42, 43]. How the many pathognomonic features of c9FTD/ALS cause disease remains to be determined, an effort that is hindered by the lack of a comprehensive model that recapitulates the hallmark features derived from both sense and antisense C9orf repeat expansion transcripts, including TDP-43 pathology

Methods

Results

Discussion

Conclusion