Researchers know the strongest genetic factors that cause frontotemporal dementia and amyotrophic lateral sclerosis: the genes for tau, progranulin, and the C9ORF72 repeat expansion. However, just as in the Alzheimer’s field, a majority of the population genetic risk remains unexplained and may be due to rare genes or ones with small effect sizes. To root them out, geneticists are analyzing affected families and using new statistical techniques. At the 10th International Conference on Frontotemporal Dementias, held August 31-September 2 in Munich, researchers chronicled their incremental progress toward unraveling the remaining genetic signals. There were no big “finds,” but speakers did identify a few new rare genes, including one on chromosome 16 that generated some buzz. Many presentations focused on modifying factors rather than causal ones, discussing genes that affect the onset age or clinical symptoms of the disease. In panel discussions, researchers agreed that the next step is to figure out how genetic factors contribute to disease, so that better therapies targeting these mechanisms can be developed. 

Latest FTD Gene Highlights Autophagy.

Geneticists linked a variant in CYLD, a deubiquitinase that slows autophagy, to familial frontotemporal dementia. [Courtesy of Molecular Cell, Komander et al., 2008]

New Genetic Signals
Linkage analysis of affected families helps researchers pinpoint genes. Carol Dobson-Stone of Neuroscience Research Australia, Sydney, described one such study in an Australian family with a history of dementia and motor neuron disease that spanned four generations. Affected family members had both TDP-43 deposits and tangles of 4R tau in their brains. Linkage analysis revealed a region on chromosome 16 that segregated with the disease. To narrow in on the gene involved, researchers performed whole-exome sequencing on five family members. In the region of interest they found a single variant in the CYLD gene that always segregated with the disease and never appeared in unaffected family members. Variants in CYLD appear to be quite rare, as analysis of an additional 442 people with FTD or motor neuron disease turned up only one additional person with a functional CYLD variant.

CYLD encodes a protein that cleaves ubiquitin chains at their lysine 63 residue. This site promotes a given substrate protein’s degradation by autophagy, so by removing the chains, CYLD puts the brakes on disposal (see Komander et al., 2008; Ferreira et al., 2015). A loss-of-function mutation in CYLD, D681G, leads to more protein degradation and causes the skin disorder cylindromatosis (see Almeida et al., 2008). However, the new FTD variant, M719V, boosts CYLD’s deubiquitinase activity fourfold, Dobson-Stone said. Heightened activity of CYLD in bladder cancer cells has been associated with weakened autophagy (see Yin et al., 2016). That suggests the M719V variant might lead to a buildup of aggregated protein, she speculated, adding, “This reinforces the importance of the autophagy pathway in FTD.”

In addition to its role in autophagy, CYLD also inhibits the activation and nuclear translocation of the transcription factor NFkB. The variant M719V shuts down NFkB more effectively than does its wild-type counterpart, Dobson-Stone noted. Other researchers found this a particularly intriguing potential mechanism for FTD. Finally, Dobson-Stone found that CYLD physically interacts with the protein products of several other FTD risk genes, including OPTN, TBK1, and SQSTM1.

A different linkage site came out of the work of Rita Cacace of the VIB at University of Antwerp, Belgium. A previous study by the group, led by Christine Van Broeckhoven, pointed to a region on chromosome 7q36 in a Dutch family with Alzheimer’s disease (see Rademakers et al., 2005). To pin the gene there, Cacace sequenced the whole genome of four affected family members. She found a four megabase inversion that segregated with disease and disrupted the gene dipeptidyl peptidase 6 (DPP6). To validate the association, the researchers screened 453 FTD, 124 ALS, 335 AD patients, and 755 controls. They found significantly more DPP6 rare and loss-of-function variants among the FTD group than in controls. Alzheimer’s patients tended to have more DPP6 variants as well, but the finding missed significance.

DPP6 increases expression and regulates the function of the potassium channel Kv4.2, which controls the excitability of hippocampal neurons. FTD patients carrying the variants made less DPP6 and less Kv4.2. “Loss of function variants in DPP6 may lead to synaptic hyperexcitability,” Cacace suggested. DPP6 was previously linked to sporadic ALS, but failed to replicate in a large GWAS (see Dec 2007 news; Feb 2009 news; Sep 2009 news). 

Researchers are also using new genetic techniques to turn up fresh associations. Aniket Mishra of VU University, Amsterdam, re-analyzed GWAS data from 3,526 FTD patients and 9,402 controls (see Ferrari et al., 2014), but instead of looking for linkage with single SNPs, he tested the set of all markers within a given gene for association with FTD. This method pumps up the statistical power to find associations (see de Leeuw et al., 2015). 

In this way, Mishra and colleagues identified variants in two genes—ARHGAP35 and SERPINA1—that associated with progressive non-fluent aphasia (PNFA). SERPINA1 is a secreted protease inhibitor found only in peripheral blood and liver, not the central nervous system. However, it enables circulating glucocorticoids to enter the brain, Mishra noted. The risk allele boosts SERPINA1 protein levels, perhaps increasing stress signaling in the brain, he suggested. Meanwhile, ARHGAP35, which is expressed in the brain, binds to glucocorticoid receptor genes and represses their transcription, thus dampening glucocorticoid signaling. Together, these variants suggest that stress signaling may play a role in PNFA, Mishra said. 

Tracking Down Factors That Influence Disease
Many genes with weak effects do not trigger disease by themselves, but instead may hasten or delay its onset, or affect how fast it moves. Cumulatively, such genes may wield a large effect. For example, carriers of GRN mutations and C9ORF72 expansions display great variability in when and whether disease develops. Several researchers at ICFTD shed light on this by reporting numerous modifying factors. Protective factors may be particularly helpful in highlighting potential therapeutic targets, researchers noted.

Cyril Pottier of the Mayo Clinic in Jacksonville, Florida, focused on GRN mutation carriers. Pottier and colleagues analyzed DNA from 491 carriers of 113 different variants, as well as 1,173 matched controls. All were Caucasian, unrelated to each other, and had FTD symptoms whose onset ranged from 39 to 87 years. Two SNPs appeared connected with onset age. One was near TNIK, aka TRAF2 and NCK-interacting kinase, a gene that regulates dendrite growth and synaptic transmission. The other was near RRBP, ribosome-binding protein 1, which regulates the transport and secretion of intracellular proteins.

The researchers also identified an SNP in TMEM106B that affected how likely someone with a GRN mutation was to get the disease. Other SNPs around this transmembrane protein have been previously associated with FTLD-TDP due to GRN and C9ORF72 mutations, although the functional variant is unknown (see Feb 2010 newsAug 2012 newsFeb 2015 news). The new variant, rs6966915, seems to protect people against the disease. About 14 percent of controls carried two copies of the protective T allele, but only 1 percent of GRN carriers did, Pottier reported. In them, the allele delayed onset age by a decade, with TT carriers typically developing symptoms in their 70s instead of their 60s. Pottier is attempting to replicate the association in a larger cohort, including sporadic FTLD patients with TDP-43 pathology and familial patients without known mutations.

New information on a different protective allele of TMEM106B came from Nina Rostgaard of the Danish Dementia Research Centre, Copenhagen. She analyzed DNA from 80 members of a large Danish family with inherited FTD due to a CHMP2B mutation. Of this group, 19 had symptoms, 14 were asymptomatic carriers, and 47 were unaffected family members. CHMP2B carriers who had the minor C allele at SNP rs3173615 of TMEM106B developed symptoms about five years later on average than those without the SNP. This variant was previously reported to protect GRN and C9ORF72 carriers from disease. The new findings imply that TMEM106B modifies FTD across its multiple causative genotypes, Rostgaard said.

In a similar vein, Elisa Rubino of the University of Turin, Italy, dug into the effect of repeat expansions in the ataxin-2 gene. Healthy people have 23 or fewer repeats, while 34 or more repeats cause spinocerebellar ataxia type 2. Between these extremes, intermediate expansions are known to be a risk factor for ALS and FTD/ALS, but have not been associated with pure FTD (see Aug 2010 newsSep 2014 news). Rubino analyzed 243 sporadic FTD patients with a mean age of 65, as well as 176 controls. She found no difference between patients and controls in how common ataxin-2 expansions were. However, FTD patients who had intermediate expansions of 27 to 33 repeats developed disease earlier than repeat non-carriers, and were four times more likely to display symptoms of parkinsonism. Ataxin-2 repeats may affect the clinical features of FTD, Rubino concluded.

Researchers also highlighted new genes that influence when disease strikes. Raffaele Ferrari of University College London and Mario Grassi of the University of Pavia, Italy, described an association analysis of 411 FTD patients from the Italian Network on FTD. The analysis found six SNPs that, when they were summed in a single genetic score, lowered the average age of onset in patients by about four years. The SNPs were found near the genes NRXN, HPCAL1, PTPRD, and GPR137B, which are involved in processes such as neuronal development, myelination, and synaptic transmission. The effect of the genes was independent of which ApoE allele they carried. Furthermore, the researchers’ analysis suggested that about 14 percent of the variability in age of onset of FTD might be accounted for by genetic factors. That figure is much higher than what has been calculated for Alzheimer’s disease, where ApoE alone accounts for almost four percent of the variability, and other genes together about 2 percent, and Parkinson’s disease, where genes are estimated to account for only 0.6 percent of variability in onset age, he noted (see Nov 2014 newsLill et al., 2015). 

With this steady stream of tidbits of information on an ever-growing list of genes, researchers hope to make inroads in understanding the processes that affect disease. “Genetics provides the first entry to new biology,” noted Stephan Züchner of the University of Miami. However, scientists at ICFTD agreed that finding causative and modifying variants is but a first step. To emphasize the importance of subsequent functional studies, Christian Haass of Ludwig-Maximilians University, Munich, referenced a line from Sydney Brenner’s 2002 Nobel Prize lecture, where Brenner noted that genomics discoveries can generate a collective state of “drowning in a sea of data and starving for knowledge.” Haass suggested that geneticists go after the big fish in this sea by prioritizing genes with the strongest effects for follow-up. Those are the ones most likely to have cellular effects, he said.—Madolyn Bowman Rogers


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Sporadic ALS Linked to Potassium Channel
  2. Genomewide Screen for SNPs Linked to Sporadic ALS Finds…Nothing Yet
  3. Research Brief: Latest ALS GWAS Points to Loci on Chromosomes 9, 19
  4. Genetics of FTD: New Gene, PGRN Variety, and a Bit of FUS
  5. FTD Risk Factor Confirmed, Alters Progranulin Pathways
  6. FTLD Gene Bad Actor in Many TDP-43 Proteinopathies
  7. ALS—A Polyglutamine Disease? Mid-length Repeats Boost Risk
  8. Ataxin 2 Variants Take Aim at Motor Neurons, Not Frontal Cortex
  9. Do Alzheimer’s GWAS Hits Speed Up Disease?

Paper Citations

  1. . The structure of the CYLD USP domain explains its specificity for Lys63-linked polyubiquitin and reveals a B box module. Mol Cell. 2008 Feb 29;29(4):451-64. PubMed.
  2. . K63 linked ubiquitin chain formation is a signal for HIF1A degradation by Chaperone-Mediated Autophagy. Sci Rep. 2015 May 11;5:10210. PubMed.
  3. . Five new CYLD mutations in skin appendage tumors and evidence that aspartic acid 681 in CYLD is essential for deubiquitinase activity. J Invest Dermatol. 2008 Mar;128(3):587-93. Epub 2007 Sep 13 PubMed.
  4. . CYLD downregulates Livin and synergistically improves gemcitabine chemosensitivity and decreases migratory/invasive potential in bladder cancer: the effect is autophagy-associated. Tumour Biol. 2016 Jul 22; PubMed.
  5. . Linkage and association studies identify a novel locus for Alzheimer disease at 7q36 in a Dutch population-based sample. Am J Hum Genet. 2005 Oct;77(4):643-52. Epub 2005 Aug 30 PubMed.
  6. . Frontotemporal dementia and its subtypes: a genome-wide association study. Lancet Neurol. 2014 Jul;13(7):686-99. PubMed.
  7. . MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput Biol. 2015 Apr;11(4):e1004219. Epub 2015 Apr 17 PubMed.
  8. . Impact of Parkinson's disease risk loci on age at onset. Mov Disord. 2015 May;30(6):847-50. Epub 2015 Apr 25 PubMed.

Further Reading