Hexanucleotide repeat expansions in the C9ORF72 gene are the leading genetic cause of frontotemporal dementia and amyotrophic lateral sclerosis, but alas, researchers do not know whether these repeats wreak havoc by way of wonky RNA molecules or the aggregating dipeptide repeat proteins translated from them. Now, a study published August 7 in Science Express points the finger squarely at the dipeptides. Researchers led by Adrian Isaacs at University College London tweaked C9ORF72 sequences in flies to distinguish between the effects of misfolded RNA and proteins. They report that the translated dipeptides are necessary to cause neurodegeneration and death in the insects. The results may steer researchers toward translation mechanisms that produce the dipeptides.
The hexanucleotide repeat expansions occur within an intron of the C9ORF72 gene (see Sept 2011 news story). Healthy people have fewer than 33 repeats, whereas people with ALS or FTD can carry hundreds to thousands. Transcription of these sequences leads to RNA species that form allegedly toxic nuclear aggregates in neurons (see Jan 2013 news story). However, through an unusual process called repeat-associated non-ATG (RAN) translation, the intronic repeats beget aggregate-prone dipeptide repeat proteins (DPRs) (see Nov 2013 news story and Cleary and Ranum, 2014). The relative contributions of these pathogenic protein and RNA species to disease, and the complex relationships that likely exist between the two, are an area of intense research.
“We have been interested for a long time in dissecting RNA and protein toxicity,” Isaacs told Alzforum. “If we want to develop appropriate therapeutics, we have to know what underlies the neurodegeneration.”
To tease out the contributions of RNA and dipeptides, co-first authors Sarah Mizielinska, Sebastian Grönke, and Teresa Niccoli and colleagues first created “RNA-only” repeats. They cloned seamless, stable strings of three to 288 hexanucleotide repeats, and inserted stop codons every 12 repeats to prevent translation. Using circular dichroism spectroscopy to look at the structure of the resulting RNA molecules, the researchers observed a stable complex called a G-quadruplex. This has been associated with FTD/ALS, forming in RNAs with or without the extra stop codons (see Mar 2014 news story and Fratta et al., 2012). Both types of RNA also formed RNA foci when transfected into neuroblastoma cells, suggesting that the extra codons did not prevent the formation of RNA structures associated with disease.
To discern whether these RNA structures alone could cause neurodegeneration, or if translated peptides were required, the researchers generated several lines of fruit flies expressing various constructs. Strings of 36 or 103 translatable repeats triggered degeneration in the eye—a hallmark of photoreceptor neuron degeneration that produces an obvious phenotype—and the flies died. In contrast, up to 288 “RNA-only” repeats did the flies no harm, even when the researchers ramped up expression of those constructs. Switching on expression of the translatable repeats only in adult fly neurons also decreased survival, whereas adult flies expressing the untranslatable constructs lived a normal lifespan.
Because the RNA-only results suggested that proteins played a crucial role in mediating degeneration, the researchers next created “protein-only” constructs to home in on the role of dipeptide repeats in disease. To make these constructs, the researchers used alternate codons that resulted in repeats in the translated proteins but not in the RNA. They built protein-only constructs for the four types of dipeptide repeat found in C9ORF72 expansions: glycine-alanine, proline-alanine, glycine-arginine, and proline-arginine. Flies expressing 36 or 100 repeats of both arginine-containing dipeptide repeat proteins had grotesque deformities in their eyes (see image above) and they died earlier than usual. The researchers concluded that something about the highly basic nature of arginine residues could be responsible for causing problems with protein homeostasis in cells.
“This paper really puts the RAN translation mechanism in the center of the field,” commented Clotilde Lagier-Tourenne of the University of California, San Diego. “It shows clearly that even short repeats produced through this mechanism are highly toxic.”
Do the results rule out a role for RNA in C9ORF72 toxicity? Probably not, according to Isaacs. The longest string of RNA-only repeats the researchers made contained 288 repeats, and it is possible that longer sequences could cause pathogenicity at the RNA level. “Our data suggests that if you work with short repeats, you’re probably going to model protein toxicity,” he said. “Once we develop longer repeats, then we might start seeing RNA toxicity.”
Lagier-Tourenne agrees. “In a patient with very long repeats, we may have a contribution from RNA toxicity,” she told Alzforum. “We need to accept that you could have dual mechanisms that are concomitant and not focus on just one.”
The dipeptides may also affect the production of RNA, as was reported in a paper in the same print issue of Science (see Aug 2014 news story). In it, the researchers reported that arginine-containing C9ORF72 dipeptides crossed into the nucleolus and bungled the production of ribosomal and messenger RNA in cultured cells. As Isaacs’s paper reported that such peptides were toxic in vivo, the two studies complement each other nicely, Isaacs said.
What do these findings in cells and flies mean for human disease? Isaacs said that the toxicity of arginine-containing dipeptides needs to be confirmed in human cells and tissues, and that levels of the peptides should be measured in cells from patients. However, the strength of the toxicity in flies suggests a likely effect in humans, he said. The findings point toward therapeutic approaches such as small molecules that prevent RAN translation. They also suggest that C9ORF72 expansion carriers who have higher levels of such translation products may be more likely to develop disease symptoms earlier, Isaacs said.—Jessica Shugart
- Corrupt Code: DNA Repeats Are Common Cause for ALS and FTD
- Chicago—RNA Inclusions Offer Therapeutic Target in ALS
- Sense, Antisense: C9ORF72 Makes Both Forms of RNA, Peptides
- RNA-DNA Pairs: At the Root of C9ORF72 Repeat Damage?
- C9ORF72 Killer Dipeptides Clog the Nucleolus
- Cleary JD, Ranum LP. Repeat associated non-ATG (RAN) translation: new starts in microsatellite expansion disorders. Curr Opin Genet Dev. 2014 May 19;26C:6-15. PubMed.
- Fratta P, Mizielinska S, Nicoll AJ, Zloh M, Fisher EM, Parkinson G, Isaacs AM. C9orf72 hexanucleotide repeat associated with amyotrophic lateral sclerosis and frontotemporal dementia forms RNA G-quadruplexes. Sci Rep. 2012;2:1016. PubMed.
- Mohan A, Goodwin M, Swanson MS. RNA-protein interactions in unstable microsatellite diseases. Brain Res. 2014 Apr 4; PubMed.
- Mizielinska S, Grönke S, Niccoli T, Ridler CE, Clayton EL, Devoy A, Moens T, Norona FE, Woollacott IO, Pietrzyk J, Cleverley K, Nicoll AJ, Pickering-Brown S, Dols J, Cabecinha M, Hendrich O, Fratta P, Fisher EM, Partridge L, Isaacs AM. C9orf72 repeat expansions cause neurodegeneration in Drosophila through arginine-rich proteins. Science. 2014 Sep 5;345(6201):1192-1194. Epub 2014 Aug 7 PubMed.