21 September 2012. One year ago today, researchers uncovered the identity of a mysterious gene on chromosome 9 that had long been linked to frontotemporal dementia and amyotrophic lateral sclerosis. To no one’s surprise, C9ORF72 occupied center stage in dozens of presentations and posters at the 8th International Conference on Frontotemporal Dementias, held 5-7 September in Manchester, U.K. Attendees also discussed the other key genes and proteins that contribute to the disease, including tau, progranulin, TDP-43, and FUS, as featured in Alzforum coverage to come. In addition, scientists debated how to subdivide different types of frontotemporal dementia (FTD) and how amyotrophic lateral sclerosis (ALS) fits into the dementia spectrum.
Repeat expansions in C9ORF72, an open reading frame of unknown function, turned out to explain many cases of both diseases (see ARF related news story on Renton et al., 2011 and Dejesus-Hernandez et al., 2011). An open reading frame is a stretch of DNA between a start codon and a stop codon. As discussed at the meeting, scientists immediately tackled the gene from all angles. Clinicians began developing a consensus on how C9ORF72 disease presents itself. Cell biologists made early forays into understanding how a series of GGGGCC repeats contribute to pathology. And geneticists are working out the expansion’s origins and the rate at which it causes disease. Here is a summary of the state of C9ORF72 science at the conference.
“C9ORF72 is a very frequent gene variant,” said Christopher Shaw, of King’s College London, during his presentation. Researchers reported that it explains from 20 to 67 percent of familial amyotrophic lateral sclerosis (ALS), depending on the population studied, making it the most prevalent genetic mutation in the disease. The expansion also appears in about 25 percent of familial frontotemporal dementia cases (FTD), as well as 7 percent of sporadic ALS and 5 percent of sporadic FTD (Majounie et al., 2012).
One challenge in understanding C9ORF72, Shaw noted, is that the expansion also shows up in 0.3 percent of the healthy population. “Incomplete penetrance is the rule,” he said. Based on families he has seen, Shaw estimated that by age 50, 9 percent of carriers will exhibit symptoms; by age 85, 74 percent would. Geneticists are particularly intrigued by the concept of genetic anticipation, whereby children become sick at a younger age than their parent did. This occurs in perhaps one-quarter of kindreds carrying the C9ORF72 variant; the age of onset appears to be about seven years earlier in children of carriers.
It may be that the repeat region grows in each successive generation, said Bradley Boeve of the Mayo Clinic in Rochester, Minnesota, in an interview with Alzforum. That will be challenging to prove. Brain tissue samples are hard to come by, and the only way to properly measure repeat size—which can range up into the thousands—is via Southern blotting. “That requires a lot of work,” commented Peter Heutink of the VU University Medical Centre in Amsterdam, The Netherlands, in his presentation.
The big step of the anticipation from one generation to the next puzzled some researchers at the meeting. “I do worry a little bit about that seven years,” said Bryan Traynor of the National Institute on Aging in Bethesda, Maryland. “Why isn’t this disease occurring in utero almost?” That is a good question, Boeve agreed. Boeve suggested to Alzforum that perhaps the age of onset does indeed reach a point in some families at which affected embryos do not survive, or that the disease manifests differently in young people.
Attendees had some fun speculating on the origin of the hexanucleotide expansion. Because people with the variant share a haplotype containing dozens of single-nucleotide polymorphisms in the region surrounding the C9 gene, many scientists suspect a single founder. Assuming 15 years per generation, that person might have lived around the time the Roman Empire fell, in the fifth century A.D., Traynor estimated. Since the mutation is most common in Scandinavian populations, he suggested—tongue in cheek—that perhaps Vikings were responsible for spreading it across the globe soon after (see ARF related news story on Ishiura et al., 2012 and Tsai et al., 2012). Shaw, who thought 15 years for each generation might be a bit short, calculated that C9ORF72 began expanding 6,300 years ago, but said it could have been any time between 2,700 and 16,500 years ago (Smith et al., 2012). Heutink suggested that the alternative to a single founder is that the shared haplotype itself is somehow predisposed to repeat expansions. It is possible that multiple massive expansions occurred independently, Shaw agreed in an e-mail to Alzforum.
Physicians could use answers to questions about penetrance and anticipation to better advise people who test positive for a C9 expansion and their families. At the same time, clinicians are learning to recognize signs of the C9 expansion before they even run a DNA sample.
“A lot of clinical descriptions have come forth…. It is clear that [the expansion] is highly heterogeneous in presentation, and it is difficult to distinguish FTD due to C9ORF72 mutations from FTD with other pathology,” wrote Robin Hsiung of the University of British Columbia in Vancouver, Canada, in an e-mail to Alzforum. “However, there are some features emerging that may alert the clinician to a C9ORF72 mutation,” added Hsiung. One such feature is a mixture of ALS and FTD in the family tree. Carriers can have either disease, or signs of both. Rigidity typical of parkinsonism provides another indicator. Approximately one-third of patients with the genetic expansion exhibit this rigidity, Boeve said.
People who present with ALS are more likely to have bulbar rather than limb onset when their ALS is due to C9ORF72 expansion. FTD is mostly of the “behavioral variant” subtype—people have executive dysfunction, apathy, and loss of empathy. People with FTD due to C9ORF72 are more likely than other FTD patients to exhibit “bizarre” beliefs and hallucinations, Boeve added (Snowden et al., 2012). For example, Boeve cares for patients who believe they are being spied upon or in danger from relatives. One believed he was infested by mites with a particular preference for his earlobe; another was convinced plastic bits were extruding from his scalp (Boeve and Graff-Radford, 2012).
Clinicians suspicious of C9 expansions will also find clues in brain images, conference speakers reported. Unlike FTD due to progranulin mutations, which tends to cause atrophy in a focused area on one side of the cortex, C9 pathology is widespread and symmetric between the two halves of the cortex and cerebellum. Spotting the hallmarks of the expansion in a clinical exam could save some money, in that physicians would order genetic tests only for the likeliest suspects, Boeve said
The Pathology Puzzler
One big question is how the C9ORF72 expansion damages the cell, and answers are beginning to come in. In a repeat of the perennial loss-of-function versus gain-of-function debate surrounding many neurodegenerative disease mutations, some scientists theorize that the extended repeats might prevent translation, depriving the cell of the protein’s as-yet unknown physiologic function. Others suggest that the extended RNA itself is toxic, perhaps due to aggregation. Boris Rogelj of the Josef Stefan Institute in Ljubljana, Slovenia, has begun to investigate the latter hypothesis. He presented his findings on the structure of the repeated nucleic acid in a poster.
Rogelj engineered up to four GGGGCC repeats in a DNA strand and examined the secondary structure by nuclear magnetic resonance and circular dichroism. He started with DNA because he worried an RNA might be unstable, but RNA studies are ongoing and thus far look similar to the DNA results, he told Alzforum. With the DNA, Rogelj found that the repeats tended to form a structure called a G-quadruplex. Four repeats line up side by side, as if they formed the four corners of a tall building. These kinds of structures are known to participate in regulating oxidative stress, Rogelj noted.
Researchers still need to discover the natural function of C9ORF72. Here, animal models could help. David Satelle of the University of Manchester told attendees about a C. elegans mutant he found in a standard worm collection. Called ok3062, the strain has a deletion in the nematode analogue of this open reading frame. Wild-type worms normally assume an “S” shape, but these mutants are “kinked” and swim at half the normal speed, Satelle said. These animals could be useful for screening drugs or genetic modifiers, he suggested.
A lighter question meeting attendees tossed around is how to shorten the clunky mouthful that is “C9ORF hexanucleotide intronic repeat expansion mutations.” Shaw abbreviated the phrase to “C9HIREM.” Members of Pam Shaw’s lab, at the Sheffield Teaching Hospital in the U.K., have started calling it the “Dallas mutation” because the telephone area code for the Texas city is 972. Others simply pronounce the gene name “corf” or “snorf.” Votes, anyone?—Amber Dance.