Just days after the first report of a highly eccentric translation mechanism in the major genetic forms of ALS and FTLD comes a second paper showing the same phenomenon. In the February 12 Neuron, researchers led by Leonard Petrucelli at the Mayo Clinic in Jacksonville, Florida, report that their independent study, too, found that the expanded hexanucleotide repeat in the first intron of the C9ORF72 gene is translated. The resulting dipeptide repeat (DPR) proteins aggregate into inclusions found only in people who carry the mutation, and only in neurons, the authors found. “[This work] identifies a novel neuropathology that is very specific for C9ORF72 mutation carriers, and provides insight into disease mechanisms,” Petrucelli told Alzforum. His findings closely mesh with those presented in Science last week by researchers led by Dieter Edbauer and Christian Haass at Ludwig-Maximilians University (see ARF related news story). For the field at large, the papers increase confidence that the finding is true.

Research into the C9ORF72 mutation has mushroomed since 2011, when researchers identified the expanded repeats as the most important genetic risk factor for ALS and FTLD (see ARF related news story; ARF news story). People with the mutation can have 700 or more repeats, in comparison to 25 or fewer in healthy people. Scientists have pursued several hypotheses for how the expansion might promote disease, including RNA toxicity and loss of function. The Science and Neuron papers suggest a new mechanism. “The two papers are nicely complementary and together solidly establish this new disease mechanism in C9ORF72 patients,” Edbauer wrote to Alzforum.

Meanwhile, in the February 5 Acta Neuropathology, Haass and colleagues pinpointed an RNA trafficking protein as another component of these inclusions. The finding may provide additional clues as to how neurons translate this strange intron and form deposits, which could influence the way the disease manifests itself. In the January 21 JAMA Neurology, researchers led by Christine Van Broeckhoven at the University of Antwerp, Belgium, revealed that C9ORF72 mutation carriers show distinctive clinical features, including early onset of disease. The mechanism by which the dipeptide repeat deposits might contribute to such unique characteristics will be a subject of intense scrutiny, researchers told Alzforum.

How might cells translate an intron without a recognizable start codon? In two other neurological diseases, expanded repeats in mRNAs form hairpin structures where ribosomes bind and begin translation (see Zu et al., 2011). Petrucelli and colleagues wondered if this repeat-associated, non-ATG-initiated (RAN) translation occurs in people with the C9ORF72 hexanucleotide expansion. To test this idea, first author Peter Ash used RNA structure prediction software to find that the C9 hexanucleotide repeat can form hairpins that are even more stable than some of those previously shown to promote RAN translation. Notably, the longer the repeats, the more stable the structures become. This jibes with data from Edbauer and colleagues, who reported that constructs had to be at least 38 repeats long to produce detectable DPR proteins in cell culture, with longer repeats cranking out more protein.

To directly detect RAN translation products, Ash and colleagues made polyclonal antibodies against the three possible DPR proteins that could be made from translating the hexanucleotide repeats in each of the three reading frames. They obtained an antibody that primarily reacted to one of the proteins, poly-glycine-proline (GP), and immunohistochemically stained brain tissue from 30 people with the C9ORF72 mutation. They detected the most abundant GP deposits in the cerebellum, hippocampus, and neocortex, but also found them in many other structures, such as the amygdala and the medial and lateral geniculate nuclei. The deposits formed exclusively in neurons, as shown by co-labeling with neuronal markers. Inclusions were absent from nine different peripheral tissues, including peripheral nerves, with the sole exception of a few aggregates in testes. The authors also screened brain tissue from more than 100 people with other neurodegenerative disorders or non-C9ORF72 FTLD/ALS, and found no sign of poly-GP deposits, confirming that they are specific to C9ORF72 mutation carriers.

“The extensive characterization of the RAN-translated peptide pathology here is exquisite and most convincing,” John Trojanowski at the University of Pennsylvania, Philadelphia, wrote to Alzforum. “I agree with the authors’ conclusion that these findings have significant implications for treatment strategies directed at RAN-translated peptides, their aggregation, and the RNA structures necessary for their production.”

Petrucelli noted that his next priority is to understand the toxicity of the DPR protein deposits, as well as how they relate to the RNA foci sometimes seen in mutation carriers. He believes these proteins could potentially serve as a biomarker for diagnosis and for tracking disease progression. In preliminary work, he sees evidence that poly-GP is present in the cerebrospinal fluid of people with the hexanucleotide expansion.

Questions remain about the neuronal inclusions in C9ORF72 mutation carriers. What else lurks there? Previous work showed that they contain the signaling protein p62 and ubiquitin, which tags proteins for degradation. In the Acta Neuropathology paper, Haass and colleagues identify yet another component: heterogenous ribonucleoprotein A3 (hnRNP A3). This protein plays a role in RNA trafficking, and is normally found in the nucleus. It strongly binds to RNA containing the C9ORF72 expanded hexanucleotide repeat, the authors report. They speculate that this binding could lead to export of the mutant C9ORF72 RNA to the cytoplasm, where RAN translation might occur. Depletion of hnRNP A3 from the nucleus might cause problems for neurons, as well, the authors suggest.

The hnRNP A3 finding adds to the evidence that these protein deposits could play a significant role in pathology. John Hardy at University College London, U.K., wrote to Alzforum: “It is worth noting that two other ALS and FTLD genes are VCP and p62, and their protein products have a role in removing abnormal protein deposits.”

Does any of this affect the clinical presentation of the disease? In the JAMA Neurology paper, Van Broeckhoven and colleagues describe features that distinguish FTLD patients with the C9ORF72 mutation from those without. Among almost 300 FTLD patients, those with the mutation tended to develop the disease earlier, typically had the behavioral variant with prominent disinhibition, and about a quarter had ALS as well. C9ORF72 mutation carriers were less likely to have nonfluent aphasia and loss of limb control (apraxia) compared to other FTLD patients. It remains unclear how these clinical features relate to the underlying pathology.––Madolyn Bowman Rogers


  1. In less than a week, two publications reveal the existence of a relatively obscure process of translation—repeat-associated non-ATG (RAN) translation in C9ORF72-associated ALS and FTD. RAN translation has already been shown in two other repeat-associated diseases, myotonic dystrophy type 1 and spinocerebellar ataxia 8, where the extended three-nucleotide repeats in the RNA have been shown to form a hairpin loop, which initiates translation and synthesis of homopolymeric repeats in all three reading frames. Both groups hypothesized that the C9ORF72 GGGGCC repeat forms a similar hairpin loop, which then initiates RAN translation and generates dipeptide repeats (DPRs) in all three frames (GA)n, (GP)n, and (GR)n. They generated antibodies to the three DPRs and, following validation, both groups showed that DPRs colocalize with p62-positive, TDP-43-negative inclusions, which have been proposed as a hallmark of C9ORF72-associated pathology (Al-Sarraj et al., 2011). As such, the strongest signals have been found in the cerebellum and hippocampus, although DPR aggregates are reported in disease-relevant regions as well, i.e., frontal and temporal cortex. The lack of colocalization with TDP-43 is a concern, as well as very strong pathology in brain areas not associated with the disease. Nevertheless, the DPR antibodies will provide a powerful research and diagnostic tool for ALS/FTLD field.

    These two important papers now add RAN translation to two other major hypotheses for the pathological effect of the C9ORF72 mutation, namely, RNA toxicity of the extended repeat and haploinsufficiency of C9ORF72 protein. Interestingly, according to two recent papers (Levine et al., 2013, and Zhang et al.,2013), the C9ORF72 protein may be related to the DENN family of GDP-GTP exchange factors for Rab GTPases involved in membrane trafficking.

    These studies raise new questions:

    1. Are DPRs enough to cause the disease? With such strong cerebellar and hippocampal expression of DPRs, can we consider them as more than a really good marker of C9ORF72 pathology?

    2. Although there is no colocalization between DPRs and TDP-43, C9ORF72 disease is still a TDP-43 proteinopathy. How does the GGGGCC repeat cause TDP-43 aggregation?

    3. The prevailing paradigm in myotonic dystrophy is that the disease is caused by RNA toxicity through aggregation of RNA and sequestration of RNA binding proteins. What are those proteins for the GGGGCC repeat doing in vivo?


    . p62 positive, TDP-43 negative, neuronal cytoplasmic and intranuclear inclusions in the cerebellum and hippocampus define the pathology of C9orf72-linked FTLD and MND/ALS. Acta Neuropathol. 2011 Dec;122(6):691-702. PubMed.

    . The product of C9orf72, a gene strongly implicated in neurodegeneration, is structurally related to DENN Rab-GEFs. Bioinformatics. 2013 Feb 15;29(4):499-503. Epub 2013 Jan 16 PubMed.

    . Discovery of Novel DENN Proteins: Implications for the Evolution of Eukaryotic Intracellular Membrane Structures and Human Disease. Front Genet. 2012;3:283. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. RNA Twist: C9ORF72 Intron Expansion Makes Aggregating Protein
  2. Corrupt Code: DNA Repeats Are Common Cause for ALS and FTD
  3. C9ORF72 Steals the Show at Frontotemporal Dementia Meeting

Paper Citations

  1. . Non-ATG-initiated translation directed by microsatellite expansions. Proc Natl Acad Sci U S A. 2011 Jan 4;108(1):260-5. Epub 2010 Dec 20 PubMed.

External Citations

  1. RNA structure prediction software
  2. VCP

Further Reading

Primary Papers

  1. . hnRNP A3 binds to GGGGCC repeats and is a constituent of p62-positive/TDP43-negative inclusions in the hippocampus of patients with C9orf72 mutations. Acta Neuropathol. 2013 Mar;125(3):413-23. PubMed.
  2. . Distinct clinical characteristics of C9orf72 expansion carriers compared with GRN, MAPT, and nonmutation carriers in a Flanders-Belgian FTLD cohort. JAMA Neurol. 2013 Mar 1;70(3):365-73. PubMed.
  3. . Unconventional translation of C9ORF72 GGGGCC expansion generates insoluble polypeptides specific to c9FTD/ALS. Neuron. 2013 Feb 20;77(4):639-46. PubMed.