By Deborah Grainger
In the Autumn of 2011, one of the biggest discoveries in the overlapping fields of amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease) and frontotemporal dementia (FTD) research was published in Neuron in two separate papers (DeJesus‑Hernandez et al. 2011, Renton et al. 2011). The findings, reported by independent groups, describe the mutation of a previously obscure gene named C9ORF72 in a significant portion of ALS and FTD cases. The aforesaid mutation is a variable expansion of a non-coding GGGGCC hexanucleotide repeat (G4C2) either in the first intron of the C9ORF72 gene or its promoter region. A number of reports show a decrease in C9ORF72 transcripts in expansion cases, suggesting that loss-of-function of the C9ORF72 protein is a contributory disease mechanism – but more on that later…
The Renton et al. paper looked at the G4C2 insertion in the context of the Finnish population (Finland has one of the highest incidence of ALS in the world) in both inherited (fALS) and spontaneous ALS (sALS). The National Institute of Health team led by Brian Traynor found that this mutation is accountable for around 46% of fALS and 21.1% of sALS in Finland – meaning that 87% of fALS in Finland can now be explained by a single genetic cause (with the D90A SOD1 mutation contributing the additional 41%).
The second paper, coming from a team headed by Rosa Rademakers at the Mayo Clinic, confirmed the mutation in 22.1% of fALS and 11.7% of familial FTD: diseases that were already being increasingly linked to one another (Neumann 2006). Several cases of these diseases were genetically mapped to chromosome 9p21: a genomic region where the C9ORF72 gene is located – making it (along with a few other genes at the time) a prime suspect in their etiology. The DeJesus-Hernandez paper further indicated the role of C9ORF72 in these diseases and provided additional confirmation that similar mechanisms underlie these two neurodegenerative conditions.
These initial findings have enabled several more discoveries since their release in 2011, which have made major headway in this area of research…
Not just FTD and ALS?
C9ORF72 G4C2 intron repeats are seen in both sporadic and familial forms of ALS and FTLD, but they have also been identified in corticobasal and ataxia syndromes. A team based at Copenhagen University Hospital looked for the mutation in 280 patients previously screened for genetic mutations involved in early onset, inherited dementia disorders including FTD and ALS. Eleven of the 14 positive hits found in the study were from patients who presented with either FTD or ALS symptoms. Interestingly, the remaining three cases presented with atypical clinical features and were previously diagnosed with clinical olivopontocerebellar degeneration (OPCD), atypical Parkinsonian syndrome (APS) and a corticobasal syndrome (CBS). This study has widened the potential clinical spectrum of C9ORF72-related disease as well as confirming the hexanucleotide expansion as a prevalent cause of FTD-ALS disorders.
It is worth noting, however, that although the hexanucleotide repeat expansion has also been described in Alzheimer’s patients – at a 1% incidence rate – the authors of that same study proposed that this correlation may be due to misdiagnosis. This hypothesis is supported by the autopsy data available for patients in this largely retrospective study, which showed several cases of ALS or FTD that had actually been mistaken for Alzheimer’s. Currently, definitive diagnosis of Alzheimer’s, ALS or FTD can only be made upon posthumous examination, so it is possible that the 1% incidence of the hexanucleotide repeat in Alzheimer’s may be an artefact of mislabeled ALS or FTD. This is something that could also have played a role in the observations of the previous paper from the Copenhagen group, but, like most things of this nature: it remains to be seen.
Repeat length vs. disease onset
Although commonly used polymerase chain reaction (PCR)-based methods can detect C9ORF72 repeat expansions, they are unable to ‘size’ any inserts beyond about 30 repeats, hampering further genotype-phenotype association studies. Southern blotting remains the only method able to measure the size of G4C2 intron repeats, however, the process is labor-intensive meaning only a limited number of Southern blot studies have been published to date. Southern blot studies that have managed to analyze sufﬁcient numbers of samples have reported a correlation of modal repeat size with age at clinical onset of symptoms, i.e. the more repeats present the earlier the onset of disease (Beck et al. 2013). This working theory is not dissimilar from the pattern of disease onset in Huntingdon’s disease. However, further cross-sectional studies are required to strengthen these findings.
More prevalent than expected
Interestingly, in addition to FTLD and ALS cases, mutations over 32 repeats in length have also been reported in healthy individuals. Some studies have suggested a healthy control upper limit of 30 repeats, but a 2013 American Journal of Human Genetics paper found that large expansions (>400 repeats) in C9ORF72 are not infrequent in the UK population (around 1 in 600 individuals). This is considerably more prevalent than would be expected from epidemiological studies. Due to the sizing issues outlined above, it is difﬁcult to pinpoint exactly at what length the G4C2 intron repeat size becomes pathogenic.
Potential disease mechanisms
There are a number of theories as to how the C9ORF72 G4C2 expansion becomes pathogenic and causes the death of neurons. It’s a tough problem to crack as the gene itself has, as yet, no known function. Despite this, a number of reports proffer loss of the gene’s function as one potential disease mechanism as they have shown decreased levels of C9ORF72 transcripts in expansion cases (Ciura et al. 2013; DeJesus-Hernandez et al. 2013).
However, this lack of transcript could be a consequence of present, but faulty expression of the gene. One hypothesis here is that the C9ORF72 intron repeats get transcribed and translated with the rest of the gene, forming a giant, misshapen strand of RNA. This strand could latch onto proteins normally involved in processing RNA, such as TDP-43, TAF15 and FUS, creating protein tangles that continue to aggregate more and more RNA and proteins. This not only depletes the cell of translation machinery, such protein-RNA clumps are also shown to be toxic to the cell. Indeed, the Rademakers’ group found that RNA foci, formed of G4C2 repeat-containing RNA, are present in the frontal cortex and spinal cord of repeat expansion carriers. These results have since been replicated in the motor cortex, temporal lobe, cerebellum and hippocampus.
Completely new proteins
A separate gain-of-function theory also exists that has garnered a lot of interest and credibility: what if the repeat expansions code for a completely new protein (or proteins)? At first this didn’t seem likely as the expansions seemed to be lacking the start codon necessary to initiate their translation. However, a German-Belgian group has shown it is possible, plus precedents for this have been seen elsewhere, as in the case of repeat-associated non-ATG (RAN) translation. The products resulting from this RAN translation are known as dipeptide repeat (DPR) proteins or C9RAN proteins, and can exist as poly-glycine-alanine (polyGA), poly-glycine-proline (polyGP) and poly-glycine-arginine (polyGR) repeats. These C9RAN proteins are shown to be very ‘sticky’: they are extremely hydrophobic and readily aggregate. Furthermore, “dot-like” cytoplasmic inclusions positive for the polyGA protein have been detected in the neurons of patients carrying the C9ORF72 mutation, and to a lesser extent polyGP and polyGR have also been detected in these inclusions.
Interestingly, prior to the discovery of C9RAN proteins in ALS and FTD cases, a neuropathological hallmark of C9ORF72 mutation carriers had been identified in the cerebella and hippocampi of deceased patients: star-shaped cytoplasmic and intranuclear inclusions negative for TDP-43, but positive for ubiquitin and the ubiquitin-binding proteins p62 and ubiquilin-2 (Brettschneider et al. 2012, Murray et al. 2011). It is now understood that all C9RAN proteins constitute many of these p62-positive, TDP-43 negative inclusions (Mackenzie et al. 2013, Mann et al. 2013, Mori et al., 2013).
To add an extra layer of complexity to C9RAN protein aggregation, these proteins have also been found to colocalize with TDP-43 positive inclusions, typical of classical sALSs and the dominant inclusion type in the FTD subtype frontotemporal lobar dementia (FTLD). TDP-43 and C9RANs usually aggregate in conditions of abundant TDP-43 pathology, in the center of the neuron surrounded by TDP-43. The inverse arrangement — TDP-43 aggregates surrounded by C9RAN proteins — has never been observed, suggesting that C9 pathology precedes TDP-43 pathology.
Overall, since the identification of the C9ORF72 repeat expansion mutation, much uncharted territory has been explored in the fields of ALS and FTD. It brings new hope that full characterization of these diseases will be realized in full in the not-too-distant future. This knowledge will undoubtedly help efforts towards earlier diagnosis and potential treatments for these debilitating and devastating neurodegenerative diseases.
|Target||Catalog no.||Type||Tested Applications|
|C9ORF72||22637-1-AP||pAb||ELISA, WB, IHC, IF, IP|
|C9ORF72||66140-1-Ig||mAb||ELISA, WB, IHC, IF, IP|
|TDP-43 (N-terminal)||10782-2-AP||pAb||ELISA, WB, IHC, IF, IP|
|TDP-43 (C-terminal)||12892-1-AP||pAb||ELISA, WB, IHC, IF, IP|
|TDP-43 (monoclonal)||60019-2-Ig||mAb||ELISA, WB, IHC, IF, IP|
|pTDP-43 (403/404)||22309-1-AP||pAb||ELISA, WB|
|pTDP-43 (409/410)||66079-1-Ig||mAb||ELISA, WB|
|p62/SQSTM1||18420-1-AP||pAb||ELISA, WB, IHC, IF, IP|
|Ubiquilin 2||23449-1-AP||pAb||ELISA, WB, IHC|