How do expanded hexanucleotide repeats in the C9ORF72 gene cause amyotrophic lateral sclerosis and frontotemporal dementia? They give rise to aberrant dipeptide repeat proteins, but even though these DPRs form deposits, they correlate poorly with degeneration. Instead, the pathologic hallmark of C9 carriers are deposits of TDP-43. In the October 1 Brain, researchers led by Frank Hirth at King’s College London now tie DPRs to TDP-43, an RNA-binding protein. At least in fruit flies.

  • In flies, DPRs cause TDP-43 to accumulate in cytoplasm.
  • This snags nuclear import protein karyopherin-α and leads to neurodegeneration.
  • The same sequence may occur in human brain.

Hirth and colleagues report that the DPRs cause TDP-43 to build up in the cytoplasm of neurons. In turn, this stray TDP-43 waylays the nuclear import protein karyopherin-α and worsens DPR deposition, kicking off a cycle of pathology. “DPR accumulation is the first hit, and TDP-43 the second,” Hirth told Alzforum. “Together they initiate a deadly cascade that plugs up the nucleocytoplasmic transport machinery, causing a feedback loop.” The researchers saw more evidence for this in postmortem human brains containing TDP-43 pathology, where karyopherin-α was all but absent from neuronal nuclei.

Other researchers said the findings help shed light on disease mechanisms in ALS/FTD. “This model nicely addresses an apparent paradox in the field, in which fully penetrant genetic mutations present at birth take many years to manifest in a clinical phenotype, but then produce a rapidly progressive disease; the kinetic shift involved in this change is difficult to explain without some kind of feedback loop,” Johnathan Cooper-Knock at the University of Sheffield, U.K., wrote to Alzforum (full comment below).

Researchers agreed the data could point toward new therapeutic strategies. “[This] opens the exciting possibility that one could therapeutically intervene between the appearance of DPRs (an early event that does not correlate with neurodegeneration) and TDP-43 aggregation (a late event that does correlate with neurodegeneration),” Robert Baloh at Cedars Sinai Medical Center in Los Angeles wrote to Alzforum (full comment below). Brian Freibaum at St. Jude Children’s Research Hospital in Memphis, Tennessee, wondered if restoring karyopherin-α location or function would reverse DPR and TDP-43 accumulation (comment below).

Blame Peptides, not RNA. The import protein karyopherin-α4 (green) remains in the nucleus in control flies (left) and in flies accumulating only C9ORF72 RNA (right). It vacates the nucleus in flies making DPR deposits (middle). [Courtesy of Solomon et al., Brain.]

The link between TDP-43 and nuclear import has emerged in the last two years. Some researchers saw TDP-43 aggregates ensnare components of the nucleocytoplasmic transport machinery (Kim and Taylor, 2017; Jan 2018 news), while others caught karyopherin-α and -β breaking apart aggregates of TDP-43, FUS, and other RNA-binding proteins in vitro, suggesting they act as disaggregases (May 2017 conference news). Yet another group found that karyopherin-β prevented FUS aggregates and dissolved existing ones (Apr 2018 news).

Hirth and colleagues wondered how DPRs, TDP-43 pathology, and nuclear import proteins related to each other. Joint first authors Daniel Solomon, Alan Stepto, and Wing Hei Au generated flies that expressed different types of DPR. Poly-GR-expressing flies were worst off. Some died as larvae; survivors were unable to climb a test tube by the time they were adult at 20 days old. By contrast, poly-GA- or poly-GP-expressers developed movement problems late in the life of a fly, around day 40.

In all cases, the fly homologue of TDP-43, TBPH, abandoned the nucleus for the cytosol around the time symptoms appeared. To dissect out why, Solomon and colleagues generated new fly models that made poly-GR or poly-GA without expressing expanded RNA, and compared them with an existing line that expressed C9ORF72 expanded repeat RNA but made no DPRs (Mizielinska et al., 2014). TBPH remained nuclear in flies with RNA foci, but massed in the cytoplasm in flies with DPR deposits. Hirth noted that poly-GR in particular is highly charged and can trap other proteins in its deposits, perhaps explaining why those flies had the most rapidly progressing disease.

Next, the researchers dissected the consequences of cytosolic TBPH. They made a new fly that expressed TBPH without a nuclear localization signal, confining it to the cytoplasm. These insects developed neurodegeneration. Intriguingly, however, neuronal death did not correlate with TBPH aggregates, suggesting that the cytosolic location of TBPH, rather than its aggregation, was the toxic feature. When the authors crossed these flies with C9ORF72 lines, the offspring had worse motor problems and accelerated DPR deposition, hinting at a feedback loop.

Turning to nuclear import, the authors found that DPR deposits correlated with cytoplasmic accumulation of karyopherin-α 2 and 4 (KPNA2/4). Again, poly-GR had the strongest effect, depleting these proteins from the nucleus, while RNA foci alone had no effect on karyopherins (see image above).The authors found no problems with any other nuclear transport proteins during early disease stages in these models. “This indicates that karyopherin pathology is one of the earliest events, before any of the other nucleopore defects,” Hirth told Alzforum.

KPNA2/4 is known to directly bind TDP-43 as well as poly-GR (Freibaum et al., 2010; Lee et al., 2016; Chou et al., 2018). In keeping with this, KPNA2/4 also strayed into the cytoplasm in flies expressing cytosolic TBPH but no DPRs. This places KPNA mis-localization downstream of TDP-43 pathology, suggesting this would be present in sporadic ALS cases, the authors suggested.

Adrian Isaacs at University College London called the link between cytosolic TDP-43 and karyopherin-α intriguing. “This is consistent with the idea that DPRs are the initiators, and TDP-43 the executioner, akin to Aβ and tau in Alzheimer’s disease,” he wrote to Alzforum (full comment below). However, Wilfried Rossoll at the Mayo Clinic in Jacksonville, Florida, doubts the findings will end the controversy over whether C9ORF72 RNA foci or DPRs are more toxic. “It would be interesting to see how the combinations of different DPRs that are present together in human patients affect the phenotype of disease models,” he wrote to Alzforum.

As a first step to relating their data to ALS/FTD, the authors searched for a link between TDP-43 and karyopherin-α pathology in human brain. They compared postmortem samples of frontal cortex from eight healthy controls with samples from eight C9ORF72 expansion carriers who had had FTD and eight people who had had sporadic TDP-FTD. Overall, cases had less nuclear KPNA4, even in neurons without detectable TDP-43 or DPR deposits. The findings hint that soluble cytosolic TDP-43, rather than aggregates, may be the toxic species, matching the fly data, Hirth said. “The phosphorylated, aggregated form of TDP-43 may be an attempt by the cell to protect itself,” he speculated.

Hirth believes the totality of evidence points to a new therapeutic target in TDP-43 diseases. “It’s now clear we need to focus on nucleocytoplasmic transport,” he said. He has found that manipulating karyopherin-α can partially rescue TDP-43 pathology in flies and is also examining whether karyopherin-α pathology marks other neurodegenerative diseases.

Some companies are already targeting nuclear transport in ALS and FTD. Karyopharm Therapeutics in Newton, Massachusetts, collaborates with Biogen to develop its nuclear export inhibitor, KPT-350, for ALS. The related compound KPT-335 has been tested in Phase 1.—Madolyn Bowman Rogers


  1. This paper from Soloman et al. addresses an important problem in C9ORF72 disease research to date. Dipeptide-repeat (DPR) proteins are observed to be extremely toxic in cell and animal models, and TDP-43 pathology correlates with neuronal loss in postmortem material, however, a link between the two has not been clearly established. The authors suggest a feedback loop in which DPR proteins initiate karyopherin-α2/4 pathology, which is both a cause and a consequence of developing TDP-43 pathology. This model nicely addresses another apparent paradox in the field in which fully penetrant genetic mutations present at birth take many years to manifest in a clinical phenotype, but then produce a rapidly progressive disease; the kinetic shift involved in this change is difficult to explain without some kind of feedback loop.

    The models described in this paper lay out a time course in which DPR pathology precedes TDP-43 pathology. This has been proposed previously, but a direct causal link between the formation of DPRs and TDP-43 pathology has been lacking—indeed, it appears that extensive DPR pathology is present for many years prior to TDP-43 pathology (Vatsavayai et al., 2016) and studies of C9ORF72-ALS have failed to demonstrate significant levels of the most toxic DPRs within dying motor neurons (Davidson et al., 2016). This paper addresses this problem in a Drosophila model in which the development of DPRs is temporally associated with TDP-43 mis-localization. It is difficult to be certain that the observed TDP-43 mis-localization is representative of the changes seen in human disease because TDP-43 mis-localization can be a nonspecific response to cell stress, but the findings are persuasive.

    A great strength of this work is the demonstration of karyopherin-α2/4 pathology in postmortem tissue from both sporadic and C9ORF72+ patients. However, the findings were not always consistent with the proposed model, for example, KPNA4 pathology frequently occurred in the absence of DPR/TDP-43 pathology, suggesting that it can occur independently. The authors suggest that soluble oligomers of DPR/TDP-43 could initiate the KPNA4 pathology without producing observable pathological inclusions and so explain this apparent mismatch.

    The key next step is to determine whether reversing karyopherin-α2/4 pathology can modify the progression of neurodegeneration, perhaps utilizing the mouse model from Laura Ranum’s group which recapitulates many of the key features of the human disease, including the various molecular pathologies (Liu et al., 2016). Success in this experiment would herald a promising therapeutic target, not just for C9ORF72-disease but for all TDP-43 proteinopathy.


    . Timing and significance of pathological features in C9orf72 expansion-associated frontotemporal dementia. Brain. 2016 Dec;139(Pt 12):3202-3216. Epub 2016 Oct 22 PubMed.

    . Neurodegeneration in frontotemporal lobar degeneration and motor neurone disease associated with expansions in C9orf72 is linked to TDP-43 pathology and not associated with aggregated forms of dipeptide repeat proteins. Neuropathol Appl Neurobiol. 2016 Apr;42(3):242-54. Epub 2015 Dec 7 PubMed.

    . C9orf72 BAC Mouse Model with Motor Deficits and Neurodegenerative Features of ALS/FTD. Neuron. 2016 May 4;90(3):521-34. Epub 2016 Apr 21 PubMed.

  2. This is a very interesting study that puts together a potential mechanism linking C9ORF72 dipeptide protein aggregates with TDP-43 pathology. By delineating a sequence of events in the fly model involving karyopherin dysfunction and initial TDP-43 mis-localization, they set the stage nicely for testing whether this is a universal pathway seen in other models and human tissues. Likewise, they open the exciting possibility that one can therapeutically intervene between the appearance of DPRs (an early event that does not correlate with neurodegeneration) and TDP-43 aggregation (a late event that does correlate with neurodegeneration). A remaining question will be how to correlate this sequence of molecular events which occurs relatively rapidly in flies with the process in humans, where DPRs occur years or decades before TDP-43 pathology and neurodegeneration.

  3. This new paper by Frank Hirth and colleagues provides clear evidence that poly-GR itself can directly lead to altered TDP-43 localization in flies. It was interesting that poly-GA led to a distinct change in TDP-43 localization, which appeared due to the sequestration of TDP-43 with cytoplasmic poly-GA aggregates. Poly-GR, in contrast, led to a more diffuse mis-localisation of TDP-43 that was not associated with aggregates of poly-GR. This implies the different DPRs induce TDP-43 alterations via different mechanisms. Consistent with previous results, factors involved in nucleocytoplasmic transport were also mis-localized.

    Based on the timing of these events, the authors suggest that TDP-43 mis-localization is the trigger for altered nucleocytoplasmic transport. This is consistent with the idea that DPRs are the initiators and TDP-43 the executioner, akin to Aβ and tau in Alzheimer’s disease (Edbauer and Haass, 2016). This sequence of events is well-supported in the literature, but there is still work to do to better understand the link between DPRs and TDP-43 and why we see very little overlap between TDP-43 and DPR inclusions in patient brains.


    . An amyloid-like cascade hypothesis for C9orf72 ALS/FTD. Curr Opin Neurobiol. 2016 Feb;36:99-106. Epub 2015 Nov 8 PubMed.

  4. I enjoyed reading this publication. I would be interested to know if there would be a way to restore KPNA2/4 function and/or localization and provide a phenotypic rescue downstream of DPR and TDP-43 accumulation in the cytoplasm. With regard to the absence of nuclear pore complex and nucleocytoplasmic transport phenotypes in their publication, I would be curious if they would see emergence of these if they had examined their G4C2-38 flies, which express both G4C2 RNA and GR DPR. It may be that these phenotypes are dependent on both RNA and DPR production simultaneously.

  5. Several years after the discovery of the C9ORF72 repeat expansion as a major genetic cause of ALS/FTD, there is still a lot of controversy over its disease mechanisms, and the role that the formation of RNA foci and the accumulation of DPR proteins play in the disease process. A related question of importance in the field is how C9ORF72 repeat expansions can trigger TDP-43 pathology, the near universal neuropathological hallmark of ALS. There are now several conflicting studies that attribute TDP-43 pathology to either RNA repeat or DPR toxicity. While this new work is interesting, and provides support for an important role of DPRs in this process, it is unlikely to end the controversy. This paper also confirms previous studies on relative DPR toxicity in Drosophila, with arginine-rich peptides being more toxic than other DPR species. It would be interesting to see how the combination of different DPRs that are present together in human patients affect the phenotype of disease models.

Make a Comment

To make a comment you must login or register.


News Citations

  1. TDP-43 Snarls Nuclear Traffic
  2. Protein Liquid-Liquid Phase Transitions: The Science Is About to Gel
  3. Liquid Phase Transition: A Deluge of Data Points to Multiple Regulators

Paper Citations

  1. . Lost in Transportation: Nucleocytoplasmic Transport Defects in ALS and Other Neurodegenerative Diseases. Neuron. 2017 Oct 11;96(2):285-297. PubMed.
  2. . C9orf72 repeat expansions cause neurodegeneration in Drosophila through arginine-rich proteins. Science. 2014 Sep 5;345(6201):1192-1194. Epub 2014 Aug 7 PubMed.
  3. . Global analysis of TDP-43 interacting proteins reveals strong association with RNA splicing and translation machinery. J Proteome Res. 2010 Feb 5;9(2):1104-20. PubMed.
  4. . C9orf72 Dipeptide Repeats Impair the Assembly, Dynamics, and Function of Membrane-Less Organelles. Cell. 2016 Oct 20;167(3):774-788.e17. PubMed.
  5. . TDP-43 pathology disrupts nuclear pore complexes and nucleocytoplasmic transport in ALS/FTD. Nat Neurosci. 2018 Feb;21(2):228-239. Epub 2018 Jan 8 PubMed.

External Citations

  1. Phase 1

Further Reading

Primary Papers

  1. . A feedback loop between dipeptide-repeat protein, TDP-43 and karyopherin-α mediates C9orf72-related neurodegeneration. Brain. 2018 Oct 1;141(10):2908-2924. PubMed.