In cells that carry an expanded hexanucleotide repeat in the C9ORF72 gene, toxic dipeptide repeat (DPR) proteins made from the expansion accumulate in the cytoplasm. How these proteins arise has been something of a mystery, however. Because the expanded repeats occur in an intron, normally they should be snipped out of the RNA message in the nucleus and never reach the cytoplasm. In the July 5 Nature Communications, researchers led by Guillaume Hautbergue and Pamela Shaw at the University of Sheffield, and Alexander Whitworth at the University of Cambridge, all in the U.K., describe how these aberrant transcripts escape the nucleus. In cultured cells, the authors found that the nuclear protein SRSF1 glommed onto the expanded repeats and then bound to a nuclear export factor to help smuggle out unspliced C9ORF72 mRNA. Depleting SRSF1, or interfering with its binding to the export factor, blocked expanded RNA from exiting. In both neuronal cells and a Drosophila model of C9ORF72, keeping expanded RNAs in the nucleus abolished DPRs and preserved neuron health and motor function. The approach did not interfere with the export of normal, spliced C9ORF72.

“This is the first time the nuclear export pathway has been targeted as a promising therapeutic strategy in any neurodegenerative disease,” Hautbergue wrote to Alzforum. C9ORF72 expansions cause many cases of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia. Hautbergue believes the same general strategy might also work for other expanded repeat diseases, such as Huntington’s and myotonic dystrophy type 1.

Others shared his enthusiasm, and suggested the findings help answer the nagging question of whether expanded RNAs or proteins are more toxic. “This paper supports our previous prediction that dipeptide repeat proteins are the downstream mediators of C9ORF72 toxicity, but that the structure of the hexanucleotide G4C2 repeat requires specific export machinery,” Brian Freibaum at St. Jude Children’s Research Hospital, Memphis, Tennessee, wrote to Alzforum. The data also hint that nuclear RNA foci might protect cells by sequestering expanded repeat RNA and preventing its export, he added. Data from flies bolsters this idea (see Sep 2015 news). 

Partners in Crime. In the proposed model, nuclear protein SRSF1 (green) binds to expanded repeats (pink) in unspliced C9ORF72 mRNA, allowing the abnormal message to exit the nucleus and be translated as DPRs. [Courtesy of Hautbergue et al., Nature Communications.]

Several recent studies revealed that nuclear-cytoplasmic transport goes haywire in cells carrying expanded C9ORF72, and tied this disruption to neurotoxicity. In one of these studies, Freibaum and colleagues found that deletion of nuclear export factor 1 (NXF1) or the export factor ALYREF protected neurons in flies (Aug 2015 news). In other studies, including one by Shaw, Hautbergue, and colleagues, researchers demonstrated that expanded C9ORF72 directly bound to serine/arginine-rich splicing factor 1 (SRSF1) and to ALYREF (Lee et al., 2013; Cooper-Knock et al., 2014). Another paper suggested that nuclear traffic snarls up in many protein aggregation diseases, not just C9ORF72 (Dec 2015 news). But the mechanisms behind this disrupted traffic, as well as how it contributed to toxicity, remained unclear.

Joint first authors Hautbergue, Lydia Castelli, and Laura Ferraiuolo decided to drill down on the interactions of expanded C9ORF72 with ALYREF and SRSF1. They knocked down each of these nuclear proteins in a Drosophila model of C9ORF72 ALS. Knockdown of ALYREF led to only limited improvement, but suppressing SRSF1 production by about three-fourths squelched DPR production, prevented neurodegeneration in the fly eye, and rescued motor function.

To explore SRSF1’s role further, the authors moved to an N2A mouse neuronal cell line that they transfected with C9ORF72 carrying 38 hexanucleotide repeats. Co-immunoprecipitations confirmed that SRSF1 bound to expanded repeat mRNA in these cells. Notably, the longer the repeats, the more message precipitated with SRSF1. Longer repeats are known to be more toxic. As in flies, knocking down SRSF1 largely prevented DPR production and ameliorated toxicity. The authors achieved similar results in neurons freshly isolated from rat cortex, where knockdown of SRSF1 again lowered DPR levels.

How might SRSF1 be affecting DPRs? Likely by preventing export, since SRSF1 knockdown slashed levels of cytoplasmic-expanded C9ORF72 in both N2A cells and fly brain. SRSF1 is known to interact with NXF1 to stimulate nuclear export (Huang et al., 2003; Huang et al., 2004). This interaction requires four arginine residues in SRSF1. The authors mutated these residues, which abolished SRSF1’s ability to bind NXF1, but left its affinity for C9ORF72 expansions intact. The authors transfected the resulting SRSF1-m4 into the N2A cells. Export of the expanded C9ORF72 RNA crashed by about three-fourths, as did DPR levels. NXF1 plays an essential role in escorting expanded C9ORF72 RNA from the nucleus, the authors concluded.

Finally, the authors generated motor neurons and astrocytes from people carrying the C9ORF72 expansion to see if a similar strategy would work in human cells. When cultured with the astrocytes, the induced motor neurons die off rapidly, losing half their number in four days. Knockdown of SRSF1 in the motor neurons boosted their survival, though not to control levels. The expanded C9ORF72 message stayed in the nucleus, and less DPR protein accumulated in cytoplasm. Importantly, the authors saw no effect on the splicing of the C9ORF72 transcript, and no drop in the export of spliced C9ORF72 mRNA. This suggests that the spliced C9ORF72 message uses a different export mechanism, noted co-author Johnathan Cooper-Knock at Sheffield. The data point to SRSF1 as a viable therapeutic target, he said.

SRSF1 is better known as a splicing factor than a nuclear export factor, and controls many alternative splicing events, said Adrian Krainer at Cold Spring Harbor Laboratory, New York, who is one of the two researchers who first identified this protein. He cautioned against inhibiting SRSF1 as a therapeutic strategy, noting that the knockout mouse dies in utero. “Up or down changes in SRSF1 levels result in extensive changes in alternative splicing of downstream targets. Thus, one would have to worry about likely toxic splicing changes in many transcripts,” Krainer wrote to Alzforum.

The authors believe a better strategy might be to interfere with the interaction between SRSF1 and NXF1. They plan to screen for small molecules that do this, and are speaking with pharma companies about collaborating to design drug candidates. Cooper-Knock noted that because of extensive redundancy among nuclear export factors, a complete lack of SRSF1 only changes the export of a handful of transcripts (Müller-McNicoll et al., 2016). That suggests to the authors that curtailing export by SRSF1 might have few side effects. “We’ve seen no toxicity in cell and animal models,” Cooper-Knock said.—Madolyn Bowman Rogers


  1. This is an important paper. It highlights the importance of SRSF1 in regulating the transport of mutant C9ORF72 transcripts from the nucleus to the cytoplasm. SR proteins are best known for their effects in regulating splicing, with more recent data indicating that SR proteins are also critical for nucleocytoplasmic transport of select RNAs, thereby linking splicing and trafficking. By blocking SRSF1-dependent nucleocytoplasmic transport, the authors were able to prevent C9ORF72 neurotoxicity or pathology in a variety of models (Drosophila, mouse cultures, human cultures). By blocking transport, the translation of mutant transcripts into dipeptide repeat proteins is abrogated.

    The lifecycle of mutant C9ORF72 transcripts is just beginning to be understood, from transcription, splicing, trafficking, translation, and degradation. Understanding each of these processes on a basic level will likely lead to several potentially therapeutic targets. 

    Whether SRSF1 itself (or the interaction between SRSF1 and the nuclear export factor, NSF1) represents a viable therapeutic target will require additional studies to determine whether SRSF1 inhibition is specific enough to be therapeutic in the absence of excess toxicity.

    There are approximately 100,000 to 200,000 SRSF1 binding sites in the transcriptome. Knockdown of SRSF1 leads to differential nuclear export of 225 transcripts, results in differential expression of ~500 transcripts (mostly coding transcripts), and apparently causes altered splicing of >2,500 transcripts (Müller-McNicoll et al., 2016). Therefore, finding an appropriate therapeutic range may be difficult. However, given the absence of any therapy for C9ORF72-related ALS and FTD, all potential therapeutic targets should be rigorously tested before any final judgements are made.


    . SR proteins are NXF1 adaptors that link alternative RNA processing to mRNA export. Genes Dev. 2016 Mar 1;30(5):553-66. PubMed.

  2. We agree with Edward B. Lee that care must be taken when manipulating the expression levels of SRSF1, and we cite the important study by Müller-McNicoll et al., 2016, in our manuscript. However, neuroprotection in our neuronal cell and Drosophila models of C9ORF72-related disease was achieved with moderate depletion of SRSF1 (60-70 percent) compared to a more than 90 percent depletion in the Müller-McNicoll et al., 2016, study.

    This study, and others such as Hautbergue et al., 2009, clearly indicated that single depletion of nuclear export adaptors has moderate effects on the nuclear export of bulk mRNAs and cellular homeostasis in human due to functional compensation.

    On the other hand, we  also showed that expression of an engineered mutant of SRSF1 that is sequestered on pathological C9ORF72-repeat transcripts, but is unable to interact with NXF1 and the nuclear export machinery, also confers neuroprotection similarly to the partial depletion of SRSF1. This constitutes an alternative therapeutic strategy, which will allow avoiding manipulating the endogenous expression levels of SRSF1.

    We are currently developing gene therapy programs in a preclinical mouse model of C9ORF72-related disease to test the efficacy and safety of both the partial depletion of SRSF1 and the exogenous expression of the engineered SRSF1 mutant.


    . UIF, a New mRNA export adaptor that works together with REF/ALY, requires FACT for recruitment to mRNA. Curr Biol. 2009 Dec 1;19(22):1918-24. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. C9ORF72 RNA Foci Acquitted of Toxic Charge—in Fruit Flies
  2. ALS Gene Repeats Obstruct Traffic Between Nucleus and Cytoplasm
  3. Too Much of a Bad Thing: Protein Aggregates Snarl Nuclear Traffic

Paper Citations

  1. . Hexanucleotide repeats in ALS/FTD form length-dependent RNA foci, sequester RNA binding proteins, and are neurotoxic. Cell Rep. 2013 Dec 12;5(5):1178-86. Epub 2013 Nov 27 PubMed.
  2. . Sequestration of multiple RNA recognition motif-containing proteins by C9orf72 repeat expansions. Brain. 2014 Jul;137(Pt 7):2040-51. Epub 2014 May 27 PubMed.
  3. . SR splicing factors serve as adapter proteins for TAP-dependent mRNA export. Mol Cell. 2003 Mar;11(3):837-43. PubMed.
  4. . A molecular link between SR protein dephosphorylation and mRNA export. Proc Natl Acad Sci U S A. 2004 Jun 29;101(26):9666-70. Epub 2004 Jun 21 PubMed.
  5. . SR proteins are NXF1 adaptors that link alternative RNA processing to mRNA export. Genes Dev. 2016 Mar 1;30(5):553-66. PubMed.

Further Reading

Primary Papers

  1. . SRSF1-dependent nuclear export inhibition of C9ORF72 repeat transcripts prevents neurodegeneration and associated motor deficits. Nat Commun. 2017 Jul 5;8:16063. PubMed.