Defining the pathogenic relationship of TDP-43 inclusions and cytoplasmic double stranded RNA in AD and FTD

Abstract
Cytoplasmic inclusions of TDP-43 (TAR DNA-binding protein 43) are found in many patients with Alzheimer's
disease (AD). The presence of TDP-43 inclusions predicts a steeper cognitive decline after controlling for
other pathologies. TDP-43 cytoplasmic inclusions are also present in > 45% of cases of frontotemporal
dementia (FTD), including all cases associated with C9ORF72 repeat expansion, which is the most common
genetic mutation causing of FTD. Cytoplasmic aggregates of TDP-43 are hypothesized to sequester the
protein from physiological targets, thereby mimicking loss of function mutations in TARDBP that also lead to
FTD. Loss of function perturbations of TDP-43 in human cells, mice, Drosophila and C. elegans lead to
derepression of endogenous inverted repeat sequences capable of forming double stranded RNA (dsRNA).
In preliminary studies, we have found that both AD and FTD human brains with TDP-43 cytoplasmic
inclusions have more cytoplasmic dsRNA (cdsRNA), a known activator of type I interferon (IFN-I) signaling.
We now propose to investigate the relationship between TDP-43 cytoplasmic inclusions and cdsRNA by
extending our analysis to additional AD and FTD brains. We will accomplish this using a method for highly
multiplexed microscopy (tissue-based cyclic immunofluorescence; t-CyCIF) developed by our team and
recently extended to the analysis of autopsied brains. In Aim 1 we will use t-CyCIF to determine whether
TDP-43 inclusions and cdsRNA are coincident in the same cells, and whether they associated with other
disease features. We will characterize the inflammatory signatures of astrocytes, microglia, and neurons in
the vicinity of cells with TDP-43 inclusions and cdsRNAs. In Aim 2 cdsRNA will be isolated from human AD
and FTD brain regions for unbiased next generation sequencing to define each region's dsRNAome. dsRNA
sequences will be mapped to the genome using novel methods to determine which dsRNAs arise from
endogenous repeat sequences, genomic lesions, and / or viral sequences, particularly members of the
Herpes family. In Aim 3 we will study the pathway linking TDP-43 cytoplasmic inclusions to cdsRNA using
cultured human iPS- derived neurons from patients with pathogenic TDP-43 mutations and from AD and FTD
patients analyzed in Aim 1. We will also elucidate mechanisms of TDP-43 / cdsRNA- mediated
neurodegeneration in two mouse models that express TDP-43 mutations and genomically-encoded cdsRNA,
and probe the role of IFN-I activation in neuronal death using an FDA-approved JAK inhibitor. By defining the
relationship between cytoplasmic TDP-43, dsRNA inclusions, associated inflammatory phenotypes, and
neuronal death, our studies will provide new insight into triggers of neuroinflammation of TDP-43 inclusions in
AD and FTD. We also aim to nominate inflammatory signatures to identify living patients with cdsRNA-
associated AD and FTD pathology who might benefit from a JAK inhibitor in a drug repurposing clinical trial.

Grant Number: 4R01AG078377-02
NIH Institute/Center: NIH
Principal Investigator: MARK ALBERS