In amyotrophic lateral sclerosis and frontotemporal dementia, the nuclear RNA-binding proteins TDP-43 and FUS mislocalize to cytoplasm, where they aggregate and contribute to pathology. Because both proteins contain putative nuclear export signals (NES) for the receptor exportin-1, some current therapeutic strategies focus on inhibiting this receptor in order to retain the proteins in the nucleus. However, new evidence casts doubt upon this strategy. In back-to-back papers published in the May 4 Scientific Reports, researchers led by Dorothee Dormann at Ludwig-Maximilians University, Munich, and Philip Thomas at the University of Texas Southwestern Medical Center, Dallas, independently reported that the NES in TDP-43 does not work. Instead of active export, TDP-43 appears to exit the nucleus via passive diffusion. Dormann’s group obtained similar data for FUS. These papers follow one in the March 15 Scientific Reports from Sami Barmada and colleagues at the University of Michigan, Ann Arbor, reporting that exportin-1 blockers had no effect on TDP-43 localization.

  • Three papers dispute that TDP-43 and FUS are actively transported from the nucleus.
  • Nuclear export proteins did not affect TDP-43/FUS transport.
  • Instead, both proteins seem to diffuse out of the nucleus.

“This series of manuscripts … shows convincingly that nuclear export of TDP-43 and FUS are not as simple as they seem,” Barmada wrote to Alzforum (full comment below). He thinks the data complicate the rational design of therapies aimed at preventing nuclear export of TDP-43. Au contraire, Wilfried Rossoll, Mayo Clinic, Jacksonville, believes targeting exportin-1 is still a viable strategy. “It appears likely that these drugs address a general defect in nuclear protein import by inhibiting nuclear export, thus restoring a balance between these processes. I don’t think that correcting FUS and TDP-43 localization has been considered the most likely mechanism for the observed therapeutic effect in ALS disease models,” he wrote (see full comment below).

Some groups, such as Karyopharm Therapeutics in Newton, Massachusetts, are developing exportin-1 inhibitors to treat ALS and FTD. In preclinical models, KPT-276, KPT-335, and KPT-350 modestly protected against degeneration and cell death caused by TDP-43 overexpression and C9ORF72 repeats, which lead to TDP-43 mislocalization (Zhang et al., 2015; Chou et al., 2018). Similarly, deletion of the putative NES in FUS reportedly improved survival in flies (Lanson et al., 2011). These findings seemed to support active nuclear export.

To examine how TDP-43 and FUS leave the nucleus, Dormann set out to test the proteins in classic nuclear export assays, which had not been done before. First author Helena Ederle used an interspecies heterokaryon assay, which fuses cells from two different species to measure how much of a nuclear protein from one shows up in the nucleus of the other. To her surprise, TDP-43 and FUS from human cells accumulated in mouse nucleus equally well whether or not their putative NES motifs were disrupted by mutation. This suggested the sequences were not functional. Supporting this, inhibiting or silencing exportin-1 did not slow export of either protein.

How else were the proteins getting out? Ederle and colleagues knocked down the related receptor exportin-5, and silenced part of the mRNA export machinery, which can also shuttle proteins out of the nucleus. No effect, so that wasn’t it.

The authors then considered that the proteins might escape the nucleus by simple passive diffusion. Small molecules of less than 60 kD are known to shuttle through the nuclear membrane within minutes (Popken et al., 2015). TDP-43 is 43 kD, and FUS 53; lo and behold, increasing their size by adding large domains trapped them in the nucleus.

If nuclear egress occurs through diffusion, then binding TDP-43 and FUS to long, unspliced mRNAs in the nucleus should also hold them back. To test this, Ederle and colleagues treated cells with a transcriptional inhibitor for three hours. This resulted in a gradual loss of TDP-43 from the nucleus, suggesting that RNA binding acts to anchor the proteins there. The data also imply that physiological dimerization or oligomerization of TDP-43 and FUS in the nucleus could help retain the proteins, Dormann noted (Jan 2016 news; Afroz et al., 2017). She plans to investigate whether this oligomerization is disrupted in disease models.

Thomas and colleagues also found that the putative NES in TDP-43 was not required for export. They further showed that because this motif was nestled inside the protein, it was not exposed to solvent and would not be accessible for binding to exportin-1. On the other hand, enlarging TDP-43 by adding a large domain prevented egress, again suggesting the protein gets out through diffusion.

Then why has previous research found a beneficial effect from exportin-1 inhibitors? Dormann suggested these inhibitors might act indirectly, by inhibiting export of other nuclear proteins. Ke Zhang at Johns Hopkins University, Baltimore, thinks the inhibitors may prevent formation of stress granules. These structures are believed to induce TDP-43 and FUS to undergo liquid-liquid phase separation and then aggregate. Zhang recently reported that that several stress granule factors are shuttled between the nucleus and cytoplasm (Zhang et al., 2018). If exportin-1 inhibitors interrupt nuclear export of these factors, that could prevent stress granule formation and secondarily suppress TDP-43 and FUS aggregation as well (see comment below).—Madolyn Bowman Rogers.

Comments

  1. Overall, I was very impressed by this study of Ederle et al. This is one among a series of manuscripts, including the Pinarbarsi study and our own (all published in Scientific Reports), that show convincingly that nuclear export of TDP-43 and FUS are not as simple as they seem.

    The annotated nuclear export signal (NES) within TDP-43’s RRM2 was predicted to be a CRM1/XPO1 substrate in 2008, but subsequent data never really supported this hypothesis. There have been several proteomic studies attempting to identify CRM1 substrates, but TDP-43 was never among those listed. Our data indicated that leptomycin B and Karyopharm's selective inhibitor of nuclear export (SINE) compounds—potent inhibitors of CRM1—also failed to affect TDP-43 localization. These conclusions are supported by the Ederle and Pinarbarsi studies, both of which demonstrated that the annotated TDP-43 “NES” failed to function as such. 

    What about the implications for therapeutic design? Although the Karyopharm compounds demonstrated modest neuroprotective effects in ALS models, they did so at low concentrations that had no effect on CRM1-dependent nuclear export. When we used these compounds at higher doses, we noted pronounced inhibition of nuclear export, but also considerable toxicity in primary neuron preparations. Together with the data supporting CRM1-independent transport of TDP-43, these results indicate that the modest neuroprotection afforded by SINE compounds have little to do with TDP-43 localization.

    All three studies showed that CRM1 is not necessary for the nuclear egress of TDP-43. Even so, while no single exporter was necessary for TDP-43 nuclear export, we found several, including CRM1, XPO7, and NXF1, that were sufficient to drive nuclear export of TDP-43. This observation suggests that nucleocytoplasmic transport mechanisms for TDP-43 are partially redundant, as can be seen for some essential RNA binding proteins (i.e., hnRNPs). 

    Whether TDP-43 nuclear egress is passive, or actively mediated by several exporters, the therapeutic implications are similar—both possibilities significantly complicate the rational design of therapies aimed at preventing TDP-43 nuclear export. 

  2. Ever since 2006, when Virginia Lee and colleagues found cytoplasmic mislocalization and aggregation of phosphorylated nuclear proteins TDP-43 (and later FUS and hnRNPs) as a pathological hallmark of ALS and FTD (Neumann et al., 2006), researchers have striven to understand its molecular mechanism. In 2015, three contemporary studies (Zhang et al., 2015; Freibaum et al., 2015; Jovičić et al., 2015) brought nucleocytoplasmic transport to the focus of our attention, suggesting a tempting model that an imbalance of nuclear import and export of TDP-43, FUS, etc., causes the cytoplasmic mislocalization and subsequent phosphorylation and aggregation of these proteins. In accordance with this model, KPTs, a series of chemical compounds inhibiting nuclear export receptor Exportin-1, suppress neurodegeneration in animal and cell models of ALS (Zhang et al., 2015; Chou et al., 2018), likely through correcting this imbalance and thus the TDP-43 pathology.

    However, more recent studies suggested several caveats to this oversimplified model. Firstly, Mark Hipp, Ulrich Hartl, Wilfried Rossoll, and we reported that cytoplasmic protein aggregates, including TDP-43, disrupt nucleocytoplasmic transport through recruiting essential transport factors to these aggregates and/or stress granules induced by these aggregates (Woerner et al., 2016; Chou et al., 2018; Zhang et al., 2018). Furthermore, four recent studies by Jim Shorter, Dorothee Dormann, Yuh Min Chook, Peter St George-Hyslop, and colleagues showed that Transportin-1, the import receptor for FUS, functions as a chaperone preventing cytoplasmic FUS phase separation and aggregation (Guo et al., 2018; Hofweber et al., 2018; Yoshizawa et al., 2018; Qamar et al., 2018). Taken together, these studies suggest that nucleocytoplasmic transport and TDP-43/FUS cytoplasmic aggregation mutually regulate each other, with stress granule assembly/liquid-liquid phase separation as a key mediator. Importantly, this current paper by Dormann and colleagues shows convincing evidence that the nuclear export of TDP-43 does not need Exportin-1, arguing against our earlier explanation of how KPTs suppress neurodegeneration. Consistent with these findings, a prior study led by Sami Barmada reported that despite its protective effect against TDP-43, KPT-350 does not suppress TDP-43 cytoplasmic mislocalization! 

    These interesting findings have not only led us to better understand the ALS/FTD pathophysiology, but also raise many exciting questions. Firstly, what is the function of the nuclear export signals of TDP-43 and FUS if the proteins do not require exportins for their export? Secondly, how about phospho-TDP-43? Do KPTs affect its localization and aggregation? Importantly, and particularly interesting to people exploring the therapeutic potential of KPTs, what mediates the compounds’ protective effects in ALS/FTD models? Although other downstream targets of KPTs can be the answer, recent findings have suggested a possible pathway by which KPTs mitigate TDP-43 and FUS toxicity. Cellular stress disrupts RNA metabolism as well as nucleocytoplasmic transport (Zhang et al., 2018), causing TDP-43 and FUS to localize to cytoplasmic stress granules. Interestingly, as suggested by the current paper, RNA defects may also enhance the export of TDP-43 and FUS. Therefore, some cellular stress response pathways may be the answer. Indeed, several essential stress granule factors (e.g., TIA1 and G3BPs) can undergo nucleocytoplasmic shuttling. As inhibiting stress granule assembly suppresses TDP-43 toxicity and neurodegeneration in multiple ALS/FTD models (Elden et al., 2010; Kim et al., 2014; Becker et al., 2017; Zhang et al., 2018), KPTs may execute their protective roles via inhibiting stress granule assembly, which in turn prevents TDP-43/FUS phase separation and aggregation.

    References:

    . Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science. 2006 Oct 6;314(5796):130-3. PubMed.

    . The C9orf72 repeat expansion disrupts nucleocytoplasmic transport. Nature. 2015 Sep 3;525(7567):56-61. Epub 2015 Aug 26 PubMed.

    . GGGGCC repeat expansion in C9orf72 compromises nucleocytoplasmic transport. Nature. 2015 Sep 3;525(7567):129-33. Epub 2015 Aug 26 PubMed.

    . 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.

    . Cytoplasmic protein aggregates interfere with nucleocytoplasmic transport of protein and RNA. Science. 2016 Jan 8;351(6269):173-6. Epub 2015 Dec 3 PubMed.

    . Nuclear-Import Receptors Reverse Aberrant Phase Transitions of RNA-Binding Proteins with Prion-like Domains. Cell. 2018 Apr 19;173(3):677-692.e20. PubMed.

    . Phase Separation of FUS Is Suppressed by Its Nuclear Import Receptor and Arginine Methylation. Cell. 2018 Apr 19;173(3):706-719.e13. PubMed.

    . Nuclear Import Receptor Inhibits Phase Separation of FUS through Binding to Multiple Sites. Cell. 2018 Apr 19;173(3):693-705.e22. PubMed.

    . FUS Phase Separation Is Modulated by a Molecular Chaperone and Methylation of Arginine Cation-π Interactions. Cell. 2018 Apr 19;173(3):720-734.e15. PubMed.

    . Stress Granule Assembly Disrupts Nucleocytoplasmic Transport. Cell. 2018 May 3;173(4):958-971.e17. Epub 2018 Apr 5 PubMed.

    . Ataxin-2 intermediate-length polyglutamine expansions are associated with increased risk for ALS. Nature. 2010 Aug 26;466(7310):1069-75. PubMed.

    . Therapeutic modulation of eIF2α phosphorylation rescues TDP-43 toxicity in amyotrophic lateral sclerosis disease models. Nat Genet. 2014 Feb;46(2):152-60. Epub 2013 Dec 15 PubMed.

    . Therapeutic reduction of ataxin-2 extends lifespan and reduces pathology in TDP-43 mice. Nature. 2017 Apr 20;544(7650):367-371. Epub 2017 Apr 12 PubMed.

    . Stress Granule Assembly Disrupts Nucleocytoplasmic Transport. Cell. 2018 May 3;173(4):958-971.e17. Epub 2018 Apr 5 PubMed.

  3. This interesting and thought-provoking paper from the Dormann lab reports that TDP-43 and FUS appear to leave the nucleus from passive diffusion rather than active nucleocytoplasmic transport via the Exportin-1/CRM1 export receptor, at least in HeLa cells. This is a surprising finding, since it was widely believed that these RNA-binding proteins shuttle in and out of the nucleus via transport receptors binding to their nuclear localization and export sequences. Recent publications from several labs have demonstrated nucleocytoplasmic transport defects in C9-ALS models, and partial rescue of disease phenotypes via CRM1 inhibitors developed by Karyopharm (e.g., KPT-276 and KPT-350). We have shown similar defects and rescue in TDP-43 proteinopathy models of ALS.

    Does this new finding that TDP-43 and FUS are not actively exported from the nucleus call into question the use of KPT compounds as potential therapeutic interventions for ALS/FTD? I don’t think so.  In my opinion, it appears likely that these drugs address a general defect in nuclear protein import by inhibiting nuclear export, thus restoring a balance between these processes. I don’t think that correcting FUS and TDP-43 localization has been considered the most likely mechanism for the observed therapeutic effect in ALS disease models.

  4. I am surprised to see data that TDP-43 export is occurring through passive diffusion rather than active export. I am not sure this will have huge implications on the effectiveness of KPT-350, as it seems to promote neuronal health independent of direct interactions with TDP-43 or FUS. It does appear that C9ORF72 dipeptides contribute to the mislocalization of TDP-43 but it is unclear whether this is due directly to impairment of import/export through the nuclear pore or indirectly through the accumulation of TDP-43 in stress granules, via the sequestration of nuclear import factors. The authors of this publication did not observe TDP-43 in stress granules, however, this could be due to an issue with the sensitivity of their antibodies.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. How TDP-43’s Disheveled Tail Spells Trouble

Paper Citations

  1. . The C9orf72 repeat expansion disrupts nucleocytoplasmic transport. Nature. 2015 Sep 3;525(7567):56-61. Epub 2015 Aug 26 PubMed.
  2. . 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.
  3. . A Drosophila model of FUS-related neurodegeneration reveals genetic interaction between FUS and TDP-43. Hum Mol Genet. 2011 Jul 1;20(13):2510-23. PubMed.
  4. . Size-dependent leak of soluble and membrane proteins through the yeast nuclear pore complex. Mol Biol Cell. 2015 Apr 1;26(7):1386-94. Epub 2015 Jan 28 PubMed.
  5. . Functional and dynamic polymerization of the ALS-linked protein TDP-43 antagonizes its pathologic aggregation. Nat Commun. 2017 Jun 29;8(1):45. PubMed.
  6. . Stress Granule Assembly Disrupts Nucleocytoplasmic Transport. Cell. 2018 May 3;173(4):958-971.e17. Epub 2018 Apr 5 PubMed.

Further Reading

Primary Papers

  1. . Nuclear egress of TDP-43 and FUS occurs independently of Exportin-1/CRM1. Sci Rep. 2018 May 4;8(1):7084. PubMed.
  2. . Active nuclear import and passive nuclear export are the primary determinants of TDP-43 localization. Sci Rep. 2018 May 4;8(1):7083. PubMed.
  3. . TDP43 nuclear export and neurodegeneration in models of amyotrophic lateral sclerosis and frontotemporal dementia. Sci Rep. 2018 Mar 15;8(1):4606. PubMed.