By Monica Heger
Adding to the debate on the importance of non-coding sequences to disease, two separate groups recently used sequencing to identify a repeat expansion in a non-coding region of chromosome 9 as the most common genetic cause of both familial amyotrophic lateral sclerosis and frontal temporal dementia.
The two groups, one led by the National Institutes of Health and the other by the Mayo Clinic Florida, published their results in a recent issue of Neuron.
The finding has diagnostic potential, helps shed light on the relationship between the two diseases, and highlights the importance of non-coding regions to disease. It also illustrates the limitations of next-generation sequencing as the NIH team was not able to identify the mutation until it sequenced an isolated chromosome at 300-fold coverage.
"This is the most common genetic mutation that has been identified for ALS or FTD," said Rosa Rademarkers, a neuroscientist at the Mayo Clinic Florida, who led her group's study. Additionally, it is the "first genetic mutation that can lead to both diseases," she added.
The two diseases have been known to be related for years, but until now, the genetic factors that have been identified were associated with either one disease or the other, not both.
In the short term, the results can be used to develop a PCR-based diagnostic. While there is no cure for either disease, a genetic diagnosis could still be useful for identifying ALS patients who may be at risk for developing cognitive impairments and FTD patients who may be at risk for developing symptoms of ALS, said Bryan Traynor, who led the NIH study.
Long term, the finding could help shed light into disease mechanisms and help with the development of new therapeutic options. Cell lines and animal models can now be created with the genetic mutation to further study the disease and to start testing and developing drugs.
Since the mid-2000s the 9p21 region of chromosome 9 has been known to be linked to familial cases of both ALS and FTD.
In 2006, linkage analysis of ALS identified a 7.8-megabase region on the chromosome. Then, in 2008, Traynor's group at the NIH in collaboration with the University of Helsinki did a genome-wide association study of ALS in Finland, which has one of the highest incidences of ALS and is a genetically homogenous population. The GWAS found a "huge spike" in chromosome 9, said Traynor, narrowing the region down to just 232,000 base pairs.
After the results of the GWAS, Traynor said he thought the team would identify the causative mutation "within two weeks.” First, the team sequenced the coding areas of the region but found nothing. Next, they sequenced the entire 232-kilobase region with Sanger sequencing, but still couldn't identify the mutation.
The researchers then teamed up with labs across the world that had collected samples of ALS families, including from the University of Manchester, Cardiff University, and the VU University Medical Center in the Netherlands.
Using capture techniques to isolate the region on chromosome 9 before sequencing it with next-gen sequencing still did not find the mutation.
Finally, pairing up with a lab in Germany that had perfected a flow-sorting technique, Traynor's team isolated chromosome 9 from one case and one control and then "sequenced the heck out of it," to around 300-fold coverage on the Illumina HiSeq 2000. The team also used oligonucleotide capture baits to isolate and sequence the region from three additional cases and two controls.
The sequencing identified eight potential variants, six of which were within 30 base pairs of each other. Upon closer inspection, Traynor could see that the variants were actually misaligned sequences. Manually aligning the sequences revealed a repeat expansion in the intronic region of the C90RF72 transcript.
The researchers next designed nested PCR primers to test for the repeat expansion in other cases and controls. In a Finnish cohort of 402 cases and 472 controls, the repeat expansion was present in 28 percent of the cases and 0.4 percent of the controls. They also examined additional cohorts in Finland and broader populations of individuals of European descent and smaller cohorts of Asian and African individuals, further linking the expansion to disease.
[ pagebreak ]
Overall, the team found that the repeat explained around 46 percent of familial ALS and 21 percent of sporadic ALS in Finland, and around one-third of familial ALS in the broader European population, making it the single largest genetic cause of the disease to date.
Separately, the Mayo Clinic Florida team identified the repeat expansion in a family with both ALS and FTD, whose disease had been linked to the 9p21 region. Their study, which identified the repeat using Sanger sequencing techniques and Southern blot analysis, found the repeat expansion in 11.7 percent of familial FTD and 22.5 percent of familial ALS.
Traynor said that had his team not sequenced the chromosome to such deep coverage, it never would have found the repeat expansion. Even with 300-fold coverage, coverage across the repetitive region dropped to just two-fold, he said.
Next-generation sequencers are notoriously bad at sequencing through repetitive regions and the hexanucleotide repeat in the ALS cases can be thousands of bases long, he said. In healthy humans, there are fewer than 20 repeats. The nested PCR technique that the team used to screen the cases and controls can only detect up to about 60 repeats, so the "repeat length in a sample carrying the expansion could be far greater than the estimation provided by this technique," the authors wrote.
Indeed, the Mayo Clinic Florida group estimated that the repeat in diseased individuals was between 700 and 1,600 units long, said Rademakers.
Her group, which included researchers at the University of British Columbia, went through a similarly tortuous process as Traynor's group. Rademakers said that the team first focused on the coding regions on chromosome 9.
Eventually, she said, they identified the presence of the repetitive region in healthy individuals. When they tried to amplify that allele in diseased individuals, however, the allele would not amplify. This clued Rademakers in to the fact that the allele was likely a large expansion repeat in the diseased individuals, because she knew that the PCR techniques the team was using for amplification would not work on large expansion repeats.
Even after forming the hypothesis, it still took the team several months to visualize and confirm the expansion repeat, she said. Eventually, the team used a combination of primer PCR, Sanger sequencing, and Southern blot analysis to confirm the repeat expansion in both ALS and FTD patients.
Two Diseases, One Mutation
The fact that the expansion repeat can cause either ALS or FTD, and sometimes both, is not too surprising, said Traynor, because there has been growing evidence suggesting that the two diseases are related, and may actually represent a spectrum of the same disease.
For example, the classic pathological indication of both diseases is bulk ubiquitin inclusions in either the brain or the spinal cord. In ALS they are found predominantly in the spinal cord, while in FTD they are found mostly in the frontal cortex.
The identification of the expansion repeat that is common to both diseases helps to explain a lot of the overlap, said Traynor.
However, what actually causes one person to develop ALS versus FTD remains a mystery, he added. There could be differences in the expansion repeat between the two diseases, other genetic factors, an environmental cause, or something random.
"That is going to be a very important thing to try and work out," he said.
One way to do this is to study families where the repeat expansion has caused either ALS, FTD, or a mixture of the diseases in different family members.
"Why certain individuals develop one disease over the other is completely unknown," Rademakers said.
Additionally, she said she would want to look for the repeat expansion in other cognitive diseases such as Alzheimer's and Parkinson's.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.