Genome-Encoded Cytoplasmic Double-Stranded RNAs, Found in C9ORF72 ALS-FTD Brain, Propagate Neuronal Loss

Steven Rodriguez; Asli Sahin; Benjamin R Schrank; Hawra Al-Lawati; Isabel Costantino; Eric Benz; Darian Fard; Alefiya D Albers; Luxiang Cao; Alexis C Gomez; Kyle Evans; Elena Ratti; Merit Cudkowicz; Matthew P Frosch; Michael Talkowski; Peter K Sorger; Bradley T Hyman; Mark W Albers

doi:10.1126/scitranslmed.aaz4699

. Author manuscript; available in PMC: 2022 Jan 21.

Published in final edited form as: Sci Transl Med. 2021 Jul 7;13(601):eaaz4699. doi: 10.1126/scitranslmed.aaz4699

Genome-Encoded Cytoplasmic Double-Stranded RNAs, Found in C9ORF72 ALS-FTD Brain, Propagate Neuronal Loss

Steven Rodriguez ^1,⁵, Asli Sahin ¹, Benjamin R Schrank ¹, Hawra Al-Lawati ¹, Isabel Costantino ¹, Eric Benz ¹, Darian Fard ¹, Alefiya D Albers ^1,², Luxiang Cao ¹, Alexis C Gomez ¹, Kyle Evans ^1,⁵, Elena Ratti ¹, Merit Cudkowicz ¹, Matthew P Frosch ³, Michael Talkowski ^1,⁴, Peter K Sorger ⁵, Bradley T Hyman ¹, Mark W Albers ^1,^5,^*

PMCID: PMC8779652 NIHMSID: NIHMS1768798 PMID: 34233951

Abstract

Triggers of innate immune signaling in the CNS of amyotrophic lateral sclerosis and frontotemporal degeneration (ALS/FTD) patients remain elusive. We report the presence of cytoplasmic double-stranded RNA (cdsRNA), an established trigger of innate immunity, in ALS-FTD brains carrying C9ORF72 intronic hexanucleotide expansions that included genomically encoded expansions of the G₄C₂ repeat sequences. Presence of cdsRNA in human brains was coincident with cytoplasmic TAR DNA-binding protein 43 (TDP-43) inclusions, a pathologic hallmark of ALS/FTD. Introducing cdsRNA into cultured human neural cells induced Type I interferon (IFN-I) signaling and death that was rescued by FDA-approved JAK inhibitors. In mice, genomically encoded dsRNAs expressed exclusively in a neuronal class induced IFN-I and death in connected neurons non-cell autonomously. Our findings establish that genomically encoded cdsRNAs trigger sterile, viral-mimetic IFN-I induction, and propagated death within neural circuits and may drive neuroinflammation and neurodegeneration in ALS/FTD patients.

One sentence summary:

Genome-encoded cytoplasmic double stranded RNAs elicit neuroinflammation and propagated neuronal death

Type I interferon signaling (IFN-I), a hallmark of aging brains (1, 2), is a pathological feature of adult-onset neurodegenerative diseases, including Amyotrophic Lateral Sclerosis (ALS), Frontal Temporal Dementia (FTD), and Alzheimer’s disease (AD) (3-6). However, unlike, viral encephalitides, such as herpes simplex (7) and Zika virus (8), which are also associated with activated IFN-I signaling and neuronal loss, the triggers for IFN-I induction in ALS, FTD and AD remain elusive. Foci of sense and of antisense RNA have been observed in neurons from ALS patients carrying an intronic expansion of hexanucleotide repeats in the C9ORF72 gene (ALS/FTD-C9ORF72) (9-11), the most common mutation that causes familial and sporadic ALS/FTD. We hypothesized that both strands of the intronic G₄C₂ repeat expansion might generate cytoplasmic double stranded RNA (cdsRNA), which is an established ligand for pattern recognition receptors (PRRs) that initiate IFN-I signaling (12).

Using an antibody that recognizes dsRNA in a sequence-independent manner (13) we stained formalin fixed paraffin embedded (FFPE) tissue sections and observed elevated amounts of cdsRNA in large cells, likely neurons, of the frontal cortex, motor cortex, and cerebellar dentate nuclei of deceased ALS/FTD-C9ORF72 patients (Fig. 1A; fig. S1A-D). We confirmed the specificity of dsRNA immunostaining by pre-incubating the anti-dsRNA antibody with poly I:C, a dsRNA mimetic (dsRNAmi; fig. S2A) and the specificity of the antibody for dsRNA relative to dsDNA by dot blotting (fig. S2B). Co-staining with markers of astrocytes and microglia revealed the presence of cdsRNA in these cell types as well (fig. S3-S4).

Fig. 1. — (A) Immunofluorescence staining with anti-dsRNA and anti-p-TDP43^S409/S410 antibodies on frontal lobe and motor cortex sections from ALS/FTD *C9ORF72* patients (orange arrowheads= dsRNA and p-TDP43^S409/S410 positive cells, cyan arrowhead= dsRNA and p-TDP43^S409/S410 negative cells). (B) Reversed transcribed-PCR amplification with primers to the region of the *C9ORF72* expansion from cdsRNA immunoprecipitation from frontal cortex and cerebellum from ALS/FTD-*C9ORF72* patients (n=3 biological replicates). (C) Quantitation of Western blots of PKR from ALS/FTD-C9 (n=8 biological replicates) or control frontal cortices (n=6 biological replicates). (D) Anti-phospho-PKR^T451 immunostaining of motor cortex from an ALS/FTD *C9ORF72* patient.

Neurons of ALS/FTD-C9ORF72 patients contain cytoplasmic inclusions of the TAR DNA-binding protein 43 (TDP-43). TDP-43 is a nuclear protein, and cytoplasmic inclusions of TDP-43 are thought to confer loss of function, which is associated with increased amounts of dsRNA (14). Co-staining FFPE tissue with antibodies against dsRNA and phosphorylated p-TDP-43^S409/S410 revealed coincidence of TDP-43 cytoplasmic inclusions and cdsRNA in ALS/FTD brains (Fig. 1A; S1D). We did not detect an enrichment of cdsRNA in the frontal cortex of two ALS patients with a pathogenic Superoxide dismutase 1 (SOD1) mutation, a variant of ALS that does not result in formation of TDP-43 inclusions (fig. S1A), or in age-matched controls.

To determine if pathogenic expansion of C9ORF72 is a source of cdsRNA we performed dsRNA-immunoprecipitation (dsRIP) followed by reverse transcribed PCR analysis. Using primers that recognize sequences adjacent to the G₄C₂ repeats in C9ORF72, we successfully amplified cDNAs from dsRIP material purified from both the cerebellum and frontal cortex cytoplasm (Fig. 1B), but not from control cases. Taken together, these results show that cdsRNAs in the cerebellum and frontal cortex of ALS/FTD-C9ORF72 patients were derived, at least in part, from G₄C₂ repeats.

To determine if cdsRNA in the brains of ALS/FTD C9ORF72 patients activates innate immune signaling, a common property of cdsRNA, we measured activation of IFN-I signaling in the frontal cortex brain regions; by Western blotting we observed increased quantities of PKR, an interferon stimulated gene (ISG), in C9ORF72 brains relative to control brains (Fig. 1C). PKR is also a dsRNA-dependent kinase that autophosphorylates T451 in response to binding dsRNA (15), thereby serving as biosensor of cytoplasmic dsRNA. We observed elevated quantities of phosphorylated-PKR^T451 by immunofluorescence microscopy in tissue sections prepared from regions of the motor cortex of ALS/FTD-C9ORF72 patient brains containing both cdsRNA and TDP-43 pathology (Fig. 1D). We also performed an unbiased analysis of previously published RNA-seq data obtained from ALS/FTD-C9ORF72 cerebelli (16). We identified 110 genes with ≥2-fold change in expression relative to control brains (16) (fig. S5A; table S1), and 4 genes that were down-regulated – C9ORF72 was one. Using Enrichr, a Web-based tool for gene set enrichment analysis (GSEA)(17), we found that the most upregulated GO Molecular Function was “double-stranded RNA binding” (GO:0003725; fig. S5B). Interferon gamma and alpha were the most upregulated ligand signatures in the (Library of Integrated Network-based Cellular Signatures) LINCS L1000 datasets (they also overlap; fig. S5C; tables S2-S3). These data are consistent with prior RNA-seq analyses (16) of frontal lobes of the same patients that found elevated IFNα quantities in ALS relative to control tissue. These data suggest a link between the presence of cdsRNA, elevated IFN signaling in frontal lobes and cerebelli, and pathogenic C9ORF72 expansion.

Cytoplasmic dsRNA evokes inflammation and death in human neural cells in vitro

To determine if G₄C₂ cdsRNA is sufficient to induce IFN-I signaling in human neural cells, we transfected synthetic G₄C₂ 66-mer dsRNA into the cytoplasm of cells differentiated from human ReNcell VM neuroprogenitors (18); we have recently shown that these cells express markers characteristic of neurons, astrocytes and oligodendrocytes (19). Following G₄C₂ dsRNA transfection, STAT1 was phosphorylated on Y701 (p-STAT1^Y701), a modification associated with activation (fig. S6). We confirmed that dsRNA is localized to the cytoplasm by transfecting rhodamine-tagged dsRNAmi into differentiated human neural cells (movie S1); 24 hours after dsRNAmi transfection, we observed induction of several ISG proteins, including MDA5, PKR, and p-STAT1^Y701 (Fig. 2A). By 48h, we observed dose-dependent death of human neural cells (Fig. 2B; S7; movie S1). Transfection of dsRNAmi into glutamatergic cortical-like neurons differentiated from the ReNcell CX human neuroprogenitor line also induced neural death after 48 h (fig. S7D). The presence of microglia was not required for death in either setting. To confirm that transfected dsRNAmi was being sensed by intracellular dsRNA PRRs, we generated a knockout of MDA5 by CRISPR-Cas9 (fig. S8). We observed an ~30% decrease in dsRNA-mediated death (Fig. 2C), consistent with a role for MDA5 and redundancy of PRR function. The FDA-approved JAK inhibitors ruxolitinib, baricitinib, and tofacitinib blocked p-STAT1^Y701 and rescued neuronal death in a dose dependent manner (Fig. 2D). By contrast, inhibition of the double stranded RNA dependent protein kinase PKR by several small molecule inhibitors did not rescue neuronal death (fig. S9). These data demonstrate that cdsRNA is sufficient to induce an IFN-I response and death in cultured human neural cells and that death can be rescued by either of three FDA-approved JAK inhibitors.

Fig. 2. — (A) Western blots showing induction of p-STAT1^Y701, MDA5, and PKR in differentiated human neurons 24 hours post-transfection with a dsRNAmi. (B) Quantitation of neuronal viability using the CellTiter-Glo assay 48 hours post transfection with a dsRNAmi at 0.25 (p=0.4807, n=4 biological replicates), 0.5 (p=0.891, n=4 biological replicates), 1 (p=0.0031, n=4 biological replicates), and 2 μg/mL (p=0.0005, n=4 biological replicates). **(C)** Quantitation of cell survival in WT or MDA5 knockout cell lines treated with 2 μg/mL dsRNAmi. **(D)** Quantitation of cell survival in differentiated ReNcell VM cells pre-treated with the FDA-approved JAK inhibitors, ruxolitinib, baracitinib, and tofacitinib, then treated with 2 μg/mL dsRNAmi. (E) Schematic showing mice expressing GFP in mature OSNs were instilled with AAV9 viral constructs expressing either sense or antisense GFP. **(F)** Coronal sections of OE stained by immunofluorescence using an anti-dsRNA in 4 month old mice instilled AAV9 expressing sense or antisense GFP, and coronal sections of mouse OE stained by RNA *in situ* hybridization with a digoxigenin labeled probe for OMP in control or mice instilled with AAV9 expressing sense or antisense GFP. Scale bar= 50 μm (G) Quantitation of height of OE positive for OMP staining (p<0.0001, n=4 pairs of animals, 4 months old) and counts of OMP-positive cells (n=3, p=0.028) each normalized to controls.

Cytoplasmic dsRNA evokes inflammation non cell autonomously in neural circuits in vivo

To determine whether the presence of cdsRNA in neurons is sufficient to induce IFN-I signaling and neuronal death in vivo, we introduced antisense GFP RNA using an AAV9 vector (20) in mice expressing GFP exclusively in mature olfactory sensory neurons ((OSNs); Olfactory Marker Protein (OMP)-Internal Ribosome Entry Sequence (IRES)-GFP mice (21); fig. 2E). As a negative control, we used an AAV9 vector of same serotype that expressed sense GFP and instilled it at the same viral titer into OMP-ires-GFP mice. As a second negative control we instilled AA9-anti-sense GFP into mice expressing tetracycline activator protein (TTA) - not GFP - from the OMP locus. Three to four weeks after viral infection, we observed cdsRNA via immunofluorescence in AAV9 antisense-GFP-infected animals, but not in the negative controls (fig. 2F). The presence of cdsRNA in a subset of neurons was associated with the induction of OASL2 (an ISG), as visualized by RNA in situ hybridization in both infected and adjacent mature neurons, as well as in immature neurons that had not been infected by the virus. These data show that dsRNA-meditated IFN signaling is transmitted to other cell types (fig. S10). We also observed an ~30% reduction in mature OSNs in mice expressing cdsRNA relative to controls (fig. 2F-H). We conclude that cdsRNA encoding a nonpathogenic gene (GFP), but neither sense nor antisense RNA on its own, can induce an IFN-I response and promote cell death of neurons.

Are cdsRNAs encoded by the genome sufficient to elicit neuroinflammation and neuronal death in vivo? To address this we made use of two different transgenic mouse lines that fortuitously express cdsRNA in neurons due to structural rearrangement of an introduced transgene and flanking genomic DNA. The Neurodegeneration 1 (Nd1) (Fig. 3A; fig. S11A,C) (fig. S11B) murine line expresses an isoform of human APP (hAPP) that increases production of amyloid Aβ peptide and the Nd2 line (fig. S11B) expresses an isoform that reduces Aβ peptide production (relative to wild type). Transgene expression was induced specifically in mature OSNs by crossing the Nd1 and Nd2 lines to an OMP-IRES-TTA background (22), (fig S11A,B). Deep-sequencing on whole genome jumping libraries with ~2kb size-selected DNA inserts (23) showed that the transgene in Nd1 animals had undergone a complex genetic rearrangement with more than two breakpoints, two deletions, and at least one inversion (Fig. 3A). We did not detect structural alterations in the transgene of a related mouse line that overexpresses an hAPP isoform using the same genetic strategy (24), but does not induce neurodegeneration (fig. S11E). Sequencing showed that the Nd2 transgene locus had also undergone complex rearrangements with possible inversions. In contrast, a related mouse line, which carried an uncomplicated simple insertion of the same transgene as Nd2, experienced no neurodegeneration (24).

Fig. 3. — (A) Color coded schematic of the transgenic construct, sequencing reads from RNA-seq after dsRIP mapped below, and a cartoon schematic of the structure of the integrated transgenic sequences and their chromosomal locations (flanking sequences) (to scale). Double slash (//) represents regions of the transgene where Sanger sequence did not annotate. (B) Coronal sections of mouse OE stained by tyramide amplified immunofluorescence with an anti-dsRNA antibody. Scale bar= 25μm. (C) PCR amplification of transgenic hAPP from cDNAs generated from RNA isolated by dsRIP, but not from immunoprecipitation with an isotype (IgG2a) and concentration-matched antibody, or dsRIP immunoprecipitates preincubated with a synthetic dsRNA mimetic. (D) Pathway schematic modified from Ingenuity Pathway Analysis software showing genes belonging to IFN-I pathway that are unregulated (red) or unchanged (gray); fold-change in expression is shown for individual genes. (E) Coronal sections of mouse OE stained by RNA *in situ* hybridization with probes for Stat1, IFI44, and OASL2 in control or Nd1 transgenic mice. Scale bar = 50μm (F) Western blots showing the induction of p-STAT1^Y701, PKR, MDA5, and reduction of MAVS proteins in control or Nd1 transgenic mice.

We observed cdsRNA in neurons by immunofluorescence in Nd1 and Nd2, but not the related mouse lines with structurally intact (non-rearranged) transgenes (Fig. 3B; fig. S12A). To confirm that the Nd1 transgene gives rise to dsRNA we generated cDNA libraries using dsRNA purified by dsRIP from cytoplasmic extracts of OE tissue. We then amplified inverted regions (which derive from the hAPP-expressing part of the construct) from these cDNA (13). As controls, we showed that inverted DNA could not be amplified when an isotype control antibody was used for immunoprecipitation or when the antibody against dsRNA was pre-incubated with a dsRNAmi competitor (Fig. 3C). Next generation sequencing of the isolated dsRNAome from Nd1 OE revealed sequences that were enriched in the inverted coding regions, identified by Sanger sequencing. Using RNA in situ hybridization, we also detected cytoplasmic expression of both sense and anti-sense mRNA from the Nd1 transgene (fig. S13).

To determine if genomically encoded cdsRNA expression triggered ISGs, we profiled the transcriptomes of OSNs isolated from Nd1 and littermate controls by RNA-seq (25-27). The IFN-I antiviral signaling pathway was the most upregulated pathway in Nd1 relative to littermate control mice (Fig. 3D-F; Fig. S14; tables S4-5). We confirmed these results using RNA in situ hybridization against three ISGs, Stat1, Oasl2, and Ifi44, and showed that they were induced across the entire olfactory epithelia (OE; Fig. 3E). ISG expression was confirmed by Western blotting for p-STAT1^Y701, PKR, MDA5, and MAVS in Nd1 OE relative to littermate control mice (Fig. 3F; fig. S15A). Stat1, Oasl2, and Ifi44 were also highly expressed in OSNs of Nd2 mice, but not control transgenic mice (fig. S16). While IFNα and IFNγ induced signaling share similar pathways, our RNA-seq data showed that IFNα4 and IFNα12, but not IFNγ or IFNβ, were elevated in Nd1 neurons (table S6), consistent with dsRNA-induced IFN-I induction.

Cytosolic PRRs bind endogenous cdsRNAs (28, 29) and activate a positive feedback loop that increases expression of the receptor genes themselves (30). We found that PKR and MDA5 protein quantities were elevated in the OE of Nd1 mice (relative to littermate controls) and MDA5 and RIG-I mRNA quantities were induced in Nd1 OSNs (Fig. 3F; table S4). Activation of MDA5 and RIG-I lead to the degradation of the mitochondrial anti-viral signaling (MAVS) protein (29, 31, 32). Consistent with these mechanisms, we found that MAVS protein quantities were reduced by 70% in Nd1 animals compared to littermate controls (Fig. 3F; fig. S15A). These data suggest that cdsRNA expressed from the transgene genomic loci activates PRRs in Nd1 mice.

In addition to IFN-I induction in the OE, we observed robust IFN-I signaling in the olfactory bulb (OB) of the brain, where OSN axons terminate (Fig 4A). In the anatomically orderly structure in the OB (Fig 4A), elevated Oasl2 expression is present in at least three structures: (i) periglomerular interneurons synapsed by OSNs axon termini; (ii) mitral cells that are a postsynaptic target of OSNs, and (iii) granule interneurons that do not directly synapse with OSNs in either Nd1 (Fig. 4D) or Nd2 (fig. S17A) mice. We confirmed that the transgene is not expressed in these neuronal types and thus, that it must induce expression in a non-cell-autonomous manner (Fig. 4A-C). IFN-I induction in the brain is associated with increased density of microglia (33), and we observe an increased density of Iba1 staining cells in the OB of Nd1 mice (fig. S18). Thus, genomically encoded cdsRNA expressed exclusively in one class of neurons evoked IFN-I signaling in postsynaptic and other downstream neurons connected in a neural circuit in vivo.

Non-cell autonomous neuronal death evoked by genomically encoded cdsRNA

In addition to IFN-I induction, both Nd1 and Nd2 mice exhibited profound neuronal death. Cleaved-caspase 3 (CCASP3) was increased ~2-fold relative to littermate controls, (Fig. 5A; fig. S12B,C). The steady-state quantities of OSNs in both Nd1 (Fig. 5B) and Nd2 mice (fig. S12B) at 90 days of age were reduced by 40% relative to age-matched controls. Remarkably, <1% of OSNs expressed the transgene (Fig. 3B; 5C; fig. S11C,D). The loss of nearly half of OSNs in mice in which transgene expression is limited to <1% of the population demonstrates a robust non-cell-autonomous neurodegenerative process.

Fig. 5. — (A) Quantitation of the number of cleaved-caspase3 positive cells in coronal sections of the OE in Nd1 mice normalized to littermate controls (green bar) or for mice treated with doxycycline for three months (light green bar); Nd1 p=0.011, n=5 biological replicates; Nd1 + dox p=0.25, n=3 biological replicates. (B) Quantitation of OSNs (all mature neurons labeled with GFP) relative to all cells in the OE quantified by FACS analysis, in Nd1 (green bar) and littermate controls (blue bar); p<0.0001, n=4 biological replicates. (C) Coronal sections of mouse OE stained by RNA *in situ* hybridization with probes for GFP (transgene), OMP (mature neurons) in control or Nd1 transgenic mice. Scale bar = 50μm. (D) OBs were smaller (outlined in red) in Nd1 relative to controls and OBs were significantly smaller by weight (E); p= 0.001, n=4 biological replicates. (F) Coronal sections of mouse OBs stained by immunofluorescence with an antibody for γH2AX in control and Nd1. (G) Quantitation of cells with pan-nuclear γH2AX staining per coronal section of OB normalized to the area of the OB in Nd1 relative to littermate control mice. p < 0.0001, n=3 mice, at lease 6 OB sections per mouse.

The inducible nature of the transgenes in the Nd1 and Nd2 lines afforded critical controls that supported cdsRNA as essential for neuronal death. First, by repressing transgene expression via oral administration of doxycycline we could demonstrate that transgene expression was necessary and sufficient for neuronal death (Fig. 5A). Second, we could rule out that greater quantities of transgene protein products accounted for neuronal death in the Nd lines relative to their non-neurogenerative control lines (fig. S19). Third, we could demonstrate that genomic changes caused by transgene integration, so-called position effects, were not responsible for neuronal loss because mRNA expression was necessary (Fig. 5A). Fourth, we could show that transgene expression did not interfere with neurogenesis, as it was elevated in the Nd1 mouse relative to controls as the expected homeostatic response to neuronal death (fig. S20). Fifth, we established that the half-life of OSNs was reduced in Nd1 mice relative to controls (fig. S20).

Since ruxolitinib reversed dsRNA-mediated death of the FDA-approved JAK inhibitor in human neural cells, we tested if ruxolitinib reverse could reverse the non-cell autonomous IFN-I induction and neuronal death in Nd1 mice. After treating adult Nd1 mice with ruxolitinib by oral gavage for 5 days, we found significant reductions in CCASP3 staining in OSNs (p < 0.0002; fig. S21A) relative to vehicle control treated mice. To confirm that ruxolitinib acted to reduce non-cell autonomous IFN-I signaling, we found marked reductions in phospho-STAT1 immunostaining (fig. S21B). Together, reversal of the neuronal death and IFN-I signaling in greater than 1% of OSNs indicates that ruxolitinib blocks the non-cell autonomous propagation of neuronal death mediated by sparse expression of a genomically encoded cdsRNA, and demonstrates the translational potential of ruxolitinib in neurodegenerative disease.

Propagation of neuronal death and cdsRNA to connected neurons in vivo

Non-cell autonomous death evoked by genomically encoded cdsRNA extended beyond the nose and into the brain. The OBs of Nd1 and Nd2 mice were approximately 20% smaller by weight as compared to littermate controls (Fig. 5D,E; fig. S19B,C). We immunostained fixed tissues for CCASP3 and γH2AX - a marker of damaged DNA that exhibits a pan-nuclear staining pattern (rather than the more common punctate staining) in cells undergoing apoptosis (fig. S22) (34). We observed an 80% increase in pan-nuclear γH2AX staining and in CCASP3 in neurons in Nd1 bulbs (Fig. 5F,G; fig. S22). Taken together, these data confirm that both IFN-I signaling and neuronal death taking place in one neuronal subtype within a neural circuit can propagated to connected and neighboring neurons.

Time-course studies of mRNA and dsRNA in the Nd1 mouse revealed the surprising stability of cdsRNA in vivo. Using oral doxycycline, we suppressed the Nd1 transgene in 3 month old mice and found that mRNA and encoded proteins were undetectable within 2-3 days (Fig. 6A). In contrast, cdsRNA was present at about half of its original quantity by 30 days after transgene suppression (Fig. 6B). We hypothesized that cdsRNA can be transported down axons to connected neurons. Consistent with this idea, we observed detectable expression of MDA5 protein in axon bundles of OSNs of Nd1 mice but not littermate controls (Fig. 6C). In the OB, we found cdsRNA in the second order olfactory neurons of Nd1 mice, and we were able to immunoprecipitate dsRNA encoding the Nd1 transgene from the OB (Fig. 6D, E), even though the transgene is not expressed in these cells (Fig. 4A-C). MAVS protein quantities were also dramatically reduced in Nd1 mouse OBs (Fig. 6F), suggestive of dsRNA PRR activation in the OB, although TLR3 activation in response to extracellular dsRNA cannot be excluded. We conclude that genomically encoded dsRNA can also propagate inflammatory signals to connected neurons.

Fig. 6. — (A) Whole mount images of mice feed Doxycycline (DOX) for 30 blocks transgene expression (green puncta). **(B)** PCR of transgenic sequence from immunoprecipitates using an anti-dsRNA antibody in mice fed DOX for 30 days. **(C)** Immunofluroescence with an anti-MDA5 antibody on sections of olfactory epithelia from Nd1 and littermate control mice, and quantification of integrated intensity, n= 4 biological replicates. (D) PCR of hAPP (transgenic sequence) from DsRIP assay on olfactory bulbs from Nd1 transgenic mice. (E) Immunohistochemistry with an anti-dsRNA antibody in olfactory bulbs from Nd1 transgenic mice. Scale bar = 50μm. (F) Immunofluorescence staining for MAVS in OB of Nd1 and control mice, and quantification of integrated intensity, n= 4 biological replicates.

Discussion

In this paper we show that genomically-encoded cdsRNA in neurons can elicit sterile inflammation resulting in both cell-autonomous and non-cell autonomous neural cell death. This mechanism, which involves dsRNA in the cytoplasm, is distinct from the recently reported accumulation of dsRNA in the nucleus in response to overexpression of the PR dipeptide derived from the antisense strand of the C9ORF72 G₄C₂ expansion (35). It therefore represents an additional mechanism of G₄C₂ pathogenesis complementary to repeat-associated non-ATG (RAN) translation of dipeptide repeats, RNA foci, and loss of function of the C9ORF72 protein itself (36). These four mechanisms may act in concert to promote neuronal death. For instance, increased DNA damage and splicing defects may increase the production of dsRNA molecules (37, 38), the quantity of cdsRNA might be augmented by dysregulated nuclear cytoplasmic transport or transient nuclear envelope breakdown (39), and activation of PKR by cdsRNA can increase RAN translation of dipeptide repeat proteins (40). Derepression of transposons due to loss of nuclear function of TDP-43 (14, 41) is another source of genomically encoded cdsRNAs (in addition to G₄C₂ expansions) in familial and sporadic forms of ALS/FTD with cytoplasmic TDP-43 inclusions (42).

Our data nominate cdsRNA as a biomolecule able to propagate neuronal death within a neural circuit, from cells expressing cdsRNA to those that do not. The surprising stability of cdsRNA (as demonstrated in mice), relative to single stranded mRNA, coupled with its ability to activate the innate immune system, is like to contribute to its potency as an activator of pathologic responses in neighboring and connected cells. It is also likely that the cdsRNA found in glial and microglial cells in ALS/FTD brains can be generated both in situ and in neighboring cells. Such a mode of action is consistent with recent studies implicating viral dsRNA as a species responsible for propagated interferon signaling (43) in Herpes Simplex Virus-1 infections. Secreted cytokines, such as interferon, and prion-like proteins in the IFN-I signaling cascade with low complexity domain proteins (44, 45) may also function in concert with cdsRNA to propagate innate immunity and neuronal death characteristic of neural circuit dysfunction.

Our demonstration that genomically-encoded cdsRNA can mediate neurodegeneration implicates neurons as an activator of the neuroinflammatory cascades observed in ALS/FTD and other neurodegenerative diseases. We hypothesize that cumulative, age-dependent processes, such as genomic instability (46), latent viral reactivation (47, 48), and derepression of endogenous repeat elements, such as retrotransposons and inverted Alu repeats (41, 49), may work in concert to raise the quantities of genomically encoded cdsRNA within neurons and other types of CNS cells. At some point, amounts of cdsRNA cross a threshold where activated innate immune signaling triggers neuronal death; this is reflected in dose-dependent induction of STAT1 phosphorylation and right shift of dose-dependent death triggered by cdsRNA in cultured human neural cells. Crossing this threshold via gradual accumulation of cdsRNA over years to decades may require subsequent pathologic impact on astroctyes, microglia, and neurons and may account for the adult onset of these neurodegenerative diseases. In this view, individuals carrying C9ORF72 G₄C₂ expansions experience accelerated production of cdsRNA relative to non-carriers, and reduced amounts of C9ORF72 in these carriers might lower the threshold where innate immune activation triggers neuronal death (50, 51).

Our study has number of limitations. Our analysis of human tissue was limited to a few brain regions and did not include spinal cord tissue. While we amplified sequences adjacent to the C9 G4C2 hexanucleotide repeats from isolated cdsRNA from C9orf72 human brains, our analysis did not include a full characterization of the isolated cdsRNA. Our mouse studies demonstrated that dsRNA encoded by the transgene in olfactory sensory neurons in the nose was present in neurons in the olfactory bulb in the brain, but the mechanisms of propagation between primary and secondary olfactory neurons require further characterization. The absence of microglia in our cell culture model demonstrate that they are not necessary for the inflammation and neuronal death evoked by cdsRNA, their absence precludes understanding how microglia might alter these inflammatory and death phenotypes in vivo.

From a translational perspective, arrest of the neurotoxic arm of the innate immune response triggered by cdsRNA by pharmacologic intervention, with JAK inhibitors or other drugs, may lead to novel therapeutic strategies that slow neuronal death. We speculate that cdsRNA-dependent mechanisms will be relevant to a majority of sporadic and familial ALS patients, and to the substantial subset of AD patients, who have cytoplasmic TDP-43 inclusions (52, 53). Further clarification of the sources of cdsRNA may also lead to RNA-based therapeutic strategies. The development of imaging and CSF biomarkers that detect and predict IFN-I activation has the potential to identify C9ORF72 carrier patients with disease prior to overt motor or cognitive deficiencies, which would enable a clinical trial to determine if asymptomatic patients are more likely to benefit from targeting neuroinflammatory pathways triggered by cdsRNA in the brain relative to patients with advanced disease.

Supplementary Material

Movie S1

Movie S1. Cytoplasmic dsRNAmi induces death of differentiated human neural cells.

Download video file^{(511.2MB, mp4)}

Table S1

Table S1. Counts table for human C9orf72 and control patient RNA-seq experiment.

NIHMS1768798-supplement-Table_S1.xlsx^{(29.5KB, xlsx)}

Table S2

Table S2. Pathway analysis of human C9orf72 patients- GO Molecular Function term enrichment.

NIHMS1768798-supplement-Table_S2.xlsx^{(7.7KB, xlsx)}

Table S3

Table S3. Pathway analysis of human C9orf72 patients- Lincs L1000 Ligand Perturbation enrichment.

NIHMS1768798-supplement-Table_S3.xlsx^{(6.3KB, xlsx)}

Table S4

Table S4. Sanger sequence of transgene integration sites of the Nd1 and CORMAP lines.

NIHMS1768798-supplement-Table_S4.xlsx^{(9.8KB, xlsx)}

Table S5

Table S5. RNA-seq analysis of RNA from FACS isolated OSNs in Nd1 compared to littermate controls.

NIHMS1768798-supplement-Table_S5.xlsx^{(2.8MB, xlsx)}

Table S6

Table S6. GSEA analysis of differentially expressed genes in Nd1 show enrichment for signatures of interferon, virus stimulated cells, and cancer/genomic instability regulated gene.

NIHMS1768798-supplement-Table_S6.xlsx^{(140KB, xlsx)}

Table S7

Table S7. Interferon-α is the only interferon expressed by neurons in Nd1 transgenic mice.

NIHMS1768798-supplement-Table_S7.xlsx^{(10.2KB, xlsx)}

Fig. S1. CdsRNA, which co-localizes with phosphorylated TDP43, is present in ALS-FTD C9orf72 cases, but not ALS/FTD SOD1.

Fig. S2. Immunohistochemistry with J2, an antibody targeting dsRNA, is specific.

Fig. S3. CdsRNA present in ALS-FTD C9orf72 cases can co-localize with astrocytes.

Fig. S4. CdsRNA present in ALS-FTD C9orf72 cases can co-localize with microglia.

Fig. S5. Gene expression in the cerebellum indicates dsRNA mediated IFN-I signaling in ALS/FTD C9ORF72 patients.

Fig. S6. Phospho-STAT1 quantities are increased by (GGGGCC)66 repeat dsRNA. A

Fig. S7. DsRNA is sufficient to induce dose-dependent cell death in differentiated human neural cells.

Fig. S8. MDA5 partially rescues dsRNAmi mediated toxicity in differentiated human neural cells.

Fig. S9. PKR inhibition does not rescue dsRNAmi mediated toxicity in differentiated human neural cells.

Fig. S10. Instillation of AAV-flex-GFP virus carrying antisense GFP into mice expressing sense GFP exclusively in mature OSNs is sufficient to IFN-I.

Fig. S11. Generation of Nd1 and Nd2 mouse lines.

Fig. S12. Loss of OSNs in Nd2.

Fig. S13. Nd1 transgene produces sense and antisense RNAs.

Fig. S14. FACS for isolation of GFP-positive cells in Nd1 and littermate controls.

Fig. S15. Quantification of RNA-seq, qPCR, and Western blots of extracts from olfactory epithelia from Nd1 normalized to control mice.

Fig. S16. IFN-I signaling is not induced in CORMAP, but is highly induced in Nd2.

Fig. S17. Propagation of IFN-I signaling and neurodegeneration in Nd2.

Fig. S18. Increased activated microglia in Nd1 mice.

Fig. S19. Human-APP expression is equivalent per neuron in CORMAP and Nd1, and CORMAC and Nd2.

Fig. S20. Increased Neurogenesis and decreased neuronal life-span in Nd1 mice.

Fig. S21. Ruxolitinib blocks apoptosis in Nd1 mice.

Fig. S22. Cleaved-caspase 3 and gamma-H2A.X overlap in apoptotic cells.

Fig. S23. An antibody against dsRNA in the olfactory bulb of Nd1 mice is specific to dsRNA.

NIHMS1768798-supplement-9.pdf^{(3.7MB, pdf)}

Acknowledgements:

The authors acknowledge Hannah Brown, Monica Mendelsohn, Jennifer Kirkland for generating the Nd1 mouse line, Sarah Edwards for generating the construct for the Nd2 mouse line. Kristina Holton of the Harvard Medical School Research Computing for assistance with human RNA-seq analysis and use of the Orchestra computing cluster, Clotilde Lagier-Tourenne and Ricardos Tabet for advice about primers for C9, Doo Y. Kim and Rudolph E. Tanzi for the stable ReNcell VM line expressing GFP, and Richard Axel for critical discussions.

Funding:

This work was supported by the NIH (DP2 OD006662, R21 NS094861, R56 AG058063 to M.W.A.), P50 AG005134-34 (to B.T.H.), P50-GM107618 (to P.K.S.), Edward R. and Anne G. Lefler Postdoctoral Fellowship, MGH ECOR Postdoctoral Award, and Diversity supplement 5P50AG005134-34 (to S.R.).

Footnotes

Competing Interests: S.R. is a co-inventor on a patent application based on this work (Methods for Treating Neurodegenerative Disease; US20190352643A1); E.R. is currently an employee of Biogen; B.T.H. is a member of the SAB and owns shares in Dewpoint. He also serves on an advisory panel for Biogen, and his laboratory has current research funding from AbbVie. His wife is an employee and shareholder of Novartis. M.W.A. is a co-inventor on a patent application based on this work (Methods for Treating Neurodegenerative Disease; US20190352643A1), he received grant money from Merck and has consulted for Merck, Genentech, IFF, Takeda, and Transposon Therapeutics.

Data and Material Availability: All data associated with the study are in the paper or supplementary materials. Accession number for Nd1 RNAseq (GSE166307). Materials and mouse lines are available upon request.

References and Notes:

1.Lu T, Pan Y, Kao SY, Li C, Kohane I, Chan J, Yankner BA, Gene regulation and DNA damage in the ageing human brain. Nature 429, 883–891 (2004). [DOI] [PubMed] [Google Scholar]
2.Baruch K, Deczkowska A, David E, Castellano JM, Miller O, Kertser A, Berkutzki T, Barnett-Itzhaki Z, Bezalel D, Wyss-Coray T, Amit I, Schwartz M, Aging. Aging-induced type I interferon response at the choroid plexus negatively affects brain function. Science 346, 89–93 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Hu JH, Zhang H, Wagey R, Krieger C, Pelech SL, Protein kinase and protein phosphatase expression in amyotrophic lateral sclerosis spinal cord. Journal of Neurochemistry 85, 432–442 (2003). [DOI] [PubMed] [Google Scholar]
4.O'Rourke JG, Bogdanik L, Yanez A, Lall D, Wolf AJ, Muhammad AK, Ho R, Carmona S, Vit JP, Zarrow J, Kim KJ, Bell S, Harms MB, Miller TM, Dangler CA, Underhill DM, Goodridge HS, Lutz CM, Baloh RH, C9orf72 is required for proper macrophage and microglial function in mice. Science 351, 1324–1329 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Taylor JM, Minter MR, Newman AG, Zhang M, Adlard PA, Crack PJ, Type-1 interferon signaling mediates neuro-inflammatory events in models of Alzheimer's disease. Neurobiology of aging 35, 1012–1023 (2014). [DOI] [PubMed] [Google Scholar]
6.Wang R, Yang B, Zhang D, Activation of interferon signaling pathways in spinal cord astrocytes from an ALS mouse model. Glia 59, 946–958 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Mancuso R, Baglio F, Agostini S, Cabinio M, Lagana MM, Hernis A, Margaritella N, Guerini FR, Zanzottera M, Nemni R, Clerici M, Relationship between herpes simplex virus-1-specific antibody titers and cortical brain damage in Alzheimer's disease and amnestic mild cognitive impairment. Frontiers in aging neuroscience 6, 285 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Beckham JD, Pastula DM, Massey A, Tyler KL, Zika Virus as an Emerging Global Pathogen: Neurological Complications of Zika Virus. JAMA neurology 73, 875–879 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Mahoney CJ, Beck J, Rohrer JD, Lashley T, Mok K, Shakespeare T, Yeatman T, Warrington EK, Schott JM, Fox NC, Rossor MN, Hardy J, Collinge J, Revesz T, Mead S, Warren JD, Frontotemporal dementia with the C9ORF72 hexanucleotide repeat expansion: clinical, neuroanatomical and neuropathological features. Brain : a journal of neurology 135, 736–750 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Gendron TF, Bieniek KF, Zhang YJ, Jansen-West K, Ash PE, Caulfield T, Daughrity L, Dunmore JH, Castanedes-Casey M, Chew J, Cosio DM, van Blitterswijk M, Lee WC, Rademakers R, Boylan KB, Dickson DW, Petrucelli L, Antisense transcripts of the expanded C9ORF72 hexanucleotide repeat form nuclear RNA foci and undergo repeat-associated non-ATG translation in c9FTD/ALS. Acta Neuropathol 126, 829–844 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Lagier-Tourenne C, Baughn M, Rigo F, Sun S, Liu P, Li HR, Jiang J, Watt AT, Chun S, Katz M, Qiu J, Sun Y, Ling SC, Zhu Q, Polymenidou M, Drenner K, Artates JW, McAlonis-Downes M, Markmiller S, Hutt KR, Pizzo DP, Cady J, Harms MB, Baloh RH, Vandenberg SR, Yeo GW, Fu XD, Bennett CF, Cleveland DW, Ravits J, Targeted degradation of sense and antisense C9orf72 RNA foci as therapy for ALS and frontotemporal degeneration. Proceedings of the National Academy of Sciences of the United States of America 110, E4530–4539 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Leung DW, Basler CF, Amarasinghe GK, Molecular mechanisms of viral inhibitors of RIG-I-like receptors. Trends Microbiol 20, 139–146 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Bonin M, Oberstrass J, Lukacs N, Ewert K, Oesterschulze E, Kassing R, Nellen W, Determination of preferential binding sites for anti-dsRNA antibodies on double-stranded RNA by scanning force microscopy. RNA 6, 563–570 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Saldi TK, Ash PE, Wilson G, Gonzales P, Garrido–Lecca A, Roberts CM, Dostal V. Gendron TF, Stein LD, Blumenthal T, Petrucelli L, Link CD, TDP–1, the Caenorhabditis elegans ortholog of TDP–43, limits the accumulation of double–stranded RNA. 33, 2947–2966 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Mayo CB, Erlandsen H, Mouser DJ, Feinstein AG, Robinson VL, May ER, Cole JL, Structural Basis of Protein Kinase R Autophosphorylation. Biochemistry 58, 2967–2977 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Prudencio M, Belzil VV, Batra R, Ross CA, Gendron TF, Pregent LJ, Murray ME, Overstreet KK, Piazza-Johnston AE, Desaro P, Bieniek KF, DeTure M, Lee WC, Biendarra SM, Davis MD, Baker MC, Perkerson RB, van Blitterswijk M, Stetler CT, Rademakers R, Link CD, Dickson DW, Boylan KB, Li H, Petrucelli L, Distinct brain transcriptome profiles in C9orf72-associated and sporadic ALS. Nat Neurosci 18, 1175–1182 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, Clark NR, Ma'ayan A, Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14, 128 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Kim YH, Choi SH, D'Avanzo C, Hebisch M, Sliwinski C, Bylykbashi E, Washicosky KJ, Klee JB, Brustle O, Tanzi RE, Kim DY, A 3D human neural cell culture system for modeling Alzheimer's disease. Nature protocols 10, 985–1006 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Song Y, Subramanian K, Berberich MJ, Rodriguez S, Latorre IJ, Luria CM, Everley R, Albers MW, Mitchison TJ, Sorger PK, A dynamic view of the proteomic landscape during differentiation of ReNcell VM cells, an immortalized human neural progenitor line. Sci Data 6, 190016 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Sohal VS, Zhang F, Yizhar O, Deisseroth K, Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature 459, 698–702 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Shykind BM, Rohani SC, O'Donnell S, Nemes A, Mendelsohn M, Sun Y, Axel R, Barnea G, Gene switching and the stability of odorant receptor gene choice. Cell 117, 801–815 (2004). [DOI] [PubMed] [Google Scholar]
22.Gogos JA, Osborne J, Nemes A, Mendelsohn M, Axel R, Genetic ablation and restoration of the olfactory topographic map. Cell 103, 609–620 (2000). [DOI] [PubMed] [Google Scholar]
23.Hanscom C, Talkowski M, Design of large-insert jumping libraries for structural variant detection using illumina sequencing. Current protocols in human genetics / editorial board, Jonathan L. Haines … [et al.] 80, 7 22 21–29 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Cao L, Schrank BR, Rodriguez S, Benz EG, Moulia TW, Rickenbacher GT, Gomez AC, Levites Y, Edwards SR, Golde TE, Hyman BT, Barnea G, Albers MW, Abeta alters the connectivity of olfactory neurons in the absence of amyloid plaques in vivo. Nat Commun 3, 1009 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Deleidi M, Hallett PJ, Koprich JB, Chung CY, Isacson O, The Toll-like receptor-3 agonist polyinosinic:polycytidylic acid triggers nigrostriatal dopaminergic degeneration. The Journal of neuroscience : the official journal of the Society for Neuroscience 30, 16091–16101 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Gantier MP, Williams BR, The response of mammalian cells to double-stranded RNA. Cytokine Growth Factor Rev 18, 363–371 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Melton LM, Keith AB, Davis S, Oakley AE, Edwardson JA, Morris CM, Chronic glial activation, neurodegeneration, and APP immunoreactive deposits following acute administration of double-stranded RNA. Glia 44, 1–12 (2003). [DOI] [PubMed] [Google Scholar]
28.Nallagatla SR, Toroney R, Bevilacqua PC, Regulation of innate immunity through RNA structure and the protein kinase PKR. Current opinion in structural biology 21, 119–127 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Wu B, Peisley A, Richards C, Yao H, Zeng X, Lin C, Chu F, Walz T, Hur S, Structural basis for dsRNA recognition, filament formation, and antiviral signal activation by MDA5. Cell 152, 276–289 (2013). [DOI] [PubMed] [Google Scholar]
30.Takeuchi O, Akira S, Pattern recognition receptors and inflammation. Cell 140, 805–820 (2010). [DOI] [PubMed] [Google Scholar]
31.Brubaker SW, Gauthier AE, Mills EW, Ingolia NT, Kagan JC, A Bicistronic MAVS Transcript Highlights a Class of Truncated Variants in Antiviral Immunity. Cell 156, 800–811 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Castanier C, Zemirli N, Portier A, Garcin D, Bidere N, Vazquez A, Arnoult D, MAVS ubiquitination by the E3 ligase TRIM25 and degradation by the proteasome is involved in type I interferon production after activation of the antiviral RIG-I-like receptors. BMC biology 10, 44 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Bialas AR, Presumey J, Das A, Poel C. E. v. d., Lapchak PH, Mesin L, Victora G, Tsokos GC, Mawrin C, Herbst R, Carroll MC, Microglia-dependent synapse loss in type I interferon-mediated lupus. Nature, (2017). [DOI] [PubMed] [Google Scholar]
34.Lu C, Zhu F, Cho YY, Tang F, Zykova T, Ma WY, Bode AM, Dong Z, Cell apoptosis: requirement of H2AX in DNA ladder formation, but not for the activation of caspase-3. Mol Cell 23, 121–132 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Zhang YJ, Guo L, Gonzales PK, Gendron TF, Wu Y, Jansen-West K, O'Raw AD, Pickles SR, Prudencio M, Carlomagno Y, Gachechiladze MA, Ludwig C, Tian R, Chew J, DeTure M, Lin WL, Tong J, Daughrity LM, Yue M, Song Y, Andersen JW, Castanedes-Casey M, Kurti A, Datta A, Antognetti G, McCampbell A, Rademakers R, Oskarsson B, Dickson DW, Kampmann M, Ward ME, Fryer JD, Link CD, Shorter J, Petrucelli L, Heterochromatin anomalies and double-stranded RNA accumulation underlie C9orf72 poly(PR) toxicity. Science 363, (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Cook C, Petrucelli L, Genetic Convergence Brings Clarity to the Enigmatic Red Line in ALS. Neuron 101, 1057–1069 (2019). [DOI] [PubMed] [Google Scholar]
37.Hautbergue GM, Castelli LM, Ferraiuolo L, Sanchez-Martinez A, Cooper-Knock J, Higginbottom A, Lin YH, Bauer CS, Dodd JE, Myszczynska MA, Alam SM, Garneret P, Chandran JS, Karyka E, Stopford MJ, Smith EF, Kirby J, Meyer K, Kaspar BK, Isaacs AM, El-Khamisy SF, De Vos KJ, Ning K, Azzouz M, Whitworth AJ, Shaw PJ, SRSF1-dependent nuclear export inhibition of C9ORF72 repeat transcripts prevents neurodegeneration and associated motor deficits. Nature communications 8, 16063 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Zhang K, Donnelly CJ, Haeusler AR, Grima JC, Machamer JB, Steinwald P, Daley EL, Miller SJ, Cunningham KM, Vidensky S, Gupta S, Thomas MA, Hong I, Chiu SL, Huganir RL, Ostrow LW, Matunis MJ, Wang J, Sattler R, Lloyd TE, Rothstein JD, The C9orf72 repeat expansion disrupts nucleocytoplasmic transport. Nature 525, 56–61 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Kim HJ, Taylor JP, Lost in Transportation: Nucleocytoplasmic Transport Defects in ALS and Other Neurodegenerative Diseases. Neuron 96, 285–297 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Green KM, Glineburg MR, Kearse MG, Flores BN, Linsalata AE, Fedak SJ, Goldstrohm AC, Barmada SJ, Todd PK, RAN translation at C9orf72-associated repeat expansions is selectively enhanced by the integrated stress response. Nat Commun 8, 2005 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Saldi TK, Gonzales PK, LaRocca TJ, Link CD, Neurodegeneration, Heterochromatin, and Double-Stranded RNA. Journal of Experimental Neuroscience 13, 1179069519830697 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Tam OH, Rozhkov NV, Shaw R, Kim D, Hubbard I, Fennessey S, Propp N, Consortium NA, Fagegaltier D, Harris BT, Ostrow LW, Phatnani H, Ravits J, Dubnau J, Gale Hammell M, Postmortem Cortex Samples Identify Distinct Molecular Subtypes of ALS: Retrotransposon Activation, Oxidative Stress, and Activated Glia. Cell Rep 29, 1164–1177 e1165 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Nguyen TA, Smith BRC, Tate MD, Belz GT, Barrios MH, Elgass KD, Weisman AS, Baker PJ, Preston SP, Whitehead L, Garnham A, Lundie RJ, Smyth GK, Pellegrini M, O'Keeffe M, Wicks IP, Masters SL, Hunter CP, Pang KC, SIDT2 Transports Extracellular dsRNA into the Cytoplasm for Innate Immune Recognition. Immunity 47, 498–509 e496 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Cai X, Chen J, Xu H, Liu S, Jiang QX, Halfmann R, Chen ZJ, Prion-like polymerization underlies signal transduction in antiviral immune defense and inflammasome activation. Cell 156, 1207–1222 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Hou F, Sun L, Zheng H, Skaug B, Jiang QX, Chen ZJ, MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response. Cell 146, 448–461 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Lodato MA, Rodin RE, Bohrson CL, Coulter ME, Barton AR, Kwon M, Sherman MA, Vitzthum CM, Luquette LJ, Yandava C, Yang P, Chittenden TW, Hatem NE, Ryu SC, Woodworth MB, Park PJ, Walsh CA, Aging and neurodegeneration are associated with increased mutations in single human neurons. Science, (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Eimer WA, Vijaya Kumar DK, Navalpur Shanmugam NK, Rodriguez AS, Mitchell T, Washicosky KJ, Gyorgy B, Breakefield XO, Tanzi RE, Moir RD, Alzheimer's Disease-Associated beta-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron 99, 56–63 e53 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Readhead B, Haure-Mirande J-V, Funk CC, Richards MA, Shannon P, Haroutunian V, Sano M, Liang WS, Beckmann ND, Price ND, Reiman EM, Schadt EE, Ehrlich ME, Gandy S, Dudley JT, Multiscale Analysis of Independent Alzheimer’s Cohorts Finds Disruption of Molecular, Genetic, and Clinical Networks by Human Herpesvirus. Neuron, (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Chung H, Calis JJA, Wu X, Sun T, Yu Y, Sarbanes SL, Dao Thi VL, Shilvock AR, Hoffmann HH, Rosenberg BR, Rice CM, Human ADAR1 Prevents Endogenous RNA from Triggering Translational Shutdown. Cell, (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Burberry A, Suzuki N, Wang JY, Moccia R, Mordes DA, Stewart MH, Suzuki-Uematsu S, Ghosh S, Singh A, Merkle FT, Koszka K, Li QZ, Zon L, Rossi DJ, Trowbridge JJ, Notarangelo LD, Eggan K, Loss-of-function mutations in the C9ORF72 mouse ortholog cause fatal autoimmune disease. Sci Transl Med 8, 347ra393 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Zhu Q, Jiang J, Gendron TF, McAlonis-Downes M, Jiang L, Taylor A, Diaz Garcia S, Ghosh Dastidar S, Rodriguez MJ, King P, Zhang Y, La Spada AR, Xu H, Petrucelli L, Ravits J, Da Cruz S, Lagier-Tourenne C, Cleveland DW, Reduced C9ORF72 function exacerbates gain of toxicity from ALS/FTD-causing repeat expansion in C9orf72. Nat Neurosci 23, 615–624 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Josephs KA, Murray ME, Whitwell JL, Parisi JE, Petrucelli L, Jack CR, Petersen RC, Dickson DW, Staging TDP-43 pathology in Alzheimer's disease. Acta neuropathologica 127, 441–450 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Josephs KA, Whitwell JL, Weigand SD, Murray ME, Tosakulwong N, Liesinger AM, Petrucelli L, Senjem ML, Knopman DS, Boeve BF, Ivnik RJ, Smith GE, Jack CR Jr., Parisi JE, Petersen RC, Dickson DW, TDP-43 is a key player in the clinical features associated with Alzheimer's disease. Acta neuropathologica 127, 811–824 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Rodriguez S, Sickles HM, Deleonardis C, Alcaraz A, Gridley T, Lin DM, Notch2 is required for maintaining sustentacular cell function in the adult mouse main olfactory epithelium. Dev Biol 314, 40–58 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L, Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28, 511–515 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, Koplev S, Jenkins SL, Jagodnik KM, Lachmann A, McDermott MG, Monteiro CD, Gundersen GW, Ma'ayan A, Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 44, W90–97 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Tsai L, Barnea G, A critical period defined by axon-targeting mechanisms in the murine olfactory bulb. Science 344, 197–200 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Choi SH, Kim YH, Hebisch M, Sliwinski C, Lee S, D'Avanzo C, Chen H, Hooli B, Asselin C, Muffat J, Klee JB, Zhang C, Wainger BJ, Peitz M, Kovacs DM, Woolf CJ, Wagner SL, Tanzi RE, Kim DY, A three-dimensional human neural cell culture model of Alzheimer's disease. Nature, (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials