Background: Mutations in the gene encoding TDP-43 have been identified in both familial and sporadic amyotrophic lateral sclerosis (ALS).
Methods: A mutation screen and copy number analysis in a motor neuron disease clinic cohort was conducted to characterise the genetic contribution of TARDBP.
Results: A novel missense mutation in a highly conserved region of TDP-43 was identified in a patient with sporadic ALS. The mutation is in close vicinity to previously identified changes. Copy number variation abnormalities were not detected.
Conclusions: The findings stress the importance of TDP-43 in the pathogenesis of sporadic ALS.
Statistics from Altmetric.com
The major disease protein found in ubiquitinated intraneuronal and glial inclusions in frontotemporal lobar degeneration and amyotrophic lateral sclerosis (ALS) is the TAR-DNA binding protein TDP-43.1 In addition to ubiquitinated inclusions, affected cells display nuclear depletion of TDP-43 and redistribution to the cytoplasm. Biochemical analysis of affected tissues is characterised by the presence of phosphorylated C terminally cleaved TDP-43 fragments. Mutations in the TARDBP gene have recently been identified in patients with both familial and sporadic ALS2 3 4 5 6 7 8 but no genetic variants have been found in patients with frontotemporal lobar degeneration. Mutations cluster in exon 6 of TARDBP affecting the C terminal glycine rich domain of the 414 amino acid nuclear protein TDP-43, possibly altering its known functions in exon skipping and splicing inhibition. Overall, TARDBP mutations are rare and do not explain the majority of patients with pathological evidence of TDP-43 accumulation at autopsy. Duplications or triplications of genes cause a number of neurodegenerative diseases,9 suggesting that copy number variation (CNV) of the TARDBP locus is a possible cause of ALS.
In this study, we wanted to investigate the spectrum of TARDBP variants in a motor neuron disease clinic cohort. In addition to mutation analysis by direct DNA sequencing, we performed TARDBP copy number analysis. There was no evidence of CNV in our cohort but we identified a novel protein changing variant in exon 6 in a patient with sporadic ALS.
A total of 128 patients with a clinical diagnosis of ALS according to the revised El Escorial criteria were recruited in the Oxford Motor Neuron Disease Clinic and screened for mutations in and CNV of TARDBP. Fifteen patients had non-SOD1 related familial ALS. Seventy-six patients were male with a mean age of 61 years, and 52 were female with a mean age of 70 years. All patients gave informed consent, and ethics approval for the study was obtained from the Oxfordshire Research Ethics Committee, Oxford, UK.
Patient genomic DNA extracted from blood was amplified using standard PCR techniques with primers for all coding regions of TARDBP (exons 2–6) and an average of 120 nucleotides of flanking intronic regions. All amplicons were sequenced using Big Dye 3.1 dideoxy terminator methods (Applied Biosystems, Foster City, California, USA), using the amplification primers. For sequence variants, the original DNA stock was re-amplified and re-sequenced. Sequence analysis was performed using Chromas software (Technelysium Pty Ltd, Shannon, Co Clare, Ireland). Copy number analysis was performed by real-time PCR and the comparative CT method.
Primers were designed for TARDBP exon 3 using ABI PRISM Primer Express software (forward primer CCG AAC AGG ACC TGA AAG AGT ATT; reverse primer TGT TTG ATG CAA TCG AAG TTT ACC) with β-actin as the internal control (forward primer CAG GGC TTC TTG TCC TTT CCT T; reverse primer GCC CAC ATA GGA ATC CTT CTG A). Real-time PCR reactions were performed in triplicate using a StepOnePlus Real-Time PCR system and FAST SYBR Green Master Mix (Applied Biosystems). Analysis was performed using the StepOne software (Applied Biosystems). TDP-43 protein sequence alignment was performed with the ClustalW method using sequences identified on the European Bioinformatics Institute database (http://www.ebi.ac.uk). Protein prediction analysis using the Polyphen (http://genetics.bwh.harvard.edu/pph/), SIFT (Sorting Intolerant From Tolerant) (http://blocks.fhcrc.org/sift/SIFT.html) and SNAP (http://cubic.bioc.columbia.edu/services/SNAP/) programmes was performed using default settings.
Real-time PCR revealed no evidence of TARDBP CNV in this cohort. Sequence variation was rare, with three single nucleotide polymorphisms (SNPs) observed in 256 chromosomes (fig 1A). Two synonymous SNPs in exon 2 (198 T>C [A66A]) and exon 6 (945 G>A [A315A]) were previously reported in cases of familial and sporadic ALS as well as in healthy controls.10 In one patient, we identified the novel 962 G>C (A321G) variant in exon 6, which is predicted to change alanine for glycine in an evolutionary highly conserved region of the C terminal glycine rich domain (fig 1B). The patient was a 53-year-old Caucasian man who presented with a 6 month history of truncal and asymmetrical arm weakness. At the time of presentation there was wasting of the shoulder girdle, asymmetrical arm weakness and wasting with preserved deep tendon reflexes, respiratory involvement with reduced vital capacity and EMG evidence of normal motor conduction with widespread active denervation. There was no cognitive deficit. The disease progressed rapidly and resulted in death from respiratory failure 20 months after the onset of symptoms. An autopsy was not performed. He had no family history of ALS, and no family members were available for DNA testing. The A321G change has not been observed in previously published control cohorts of over 2500 healthy control subjects in whom TARDBP was sequenced.4 8 10 No immediate detrimental effect on protein function was obvious in the protein prediction applications.
The absence of TARDBP CNV in our cohort is in keeping with previously published reports looking at CNV using different genetic techniques, including multiplex amplicon quantification,11 SNP chip genotyping10 and TaqMan gene expression assays.12 This repeatedly negative finding in over 850 patients suggests that TARDBP CNV is not a common cause of ALS. The silent SNPs identified in our study might be overrepresented in ALS patients and it is conceivable that they confer an increased risk of the disease by altering expression levels or splicing of TDP-43.
Although the novel A321G variant is characterised by a conservative amino acid substitution that is not predicted to significantly change protein function in commonly used prediction algorithms, the absence of this change in a large number of healthy individuals as well as the location in an evolutionary highly conserved residue provide strong indirect evidence of its role in disease pathogenesis. In addition, it is in close vicinity to the majority of previously published mutations (fig 1C) which cluster in the C terminal glycine rich domain of TDP-43. While this is unlikely to be coincidental, our understanding of the biological significance of this finding is limited. The importance of the C terminal domain in splicing inhibition, probably mediated by binding of proteins of the hnRNP class, has been well characterised13 and aberrant splicing resulting from alteration of the C terminal domain is one of several potential mechanisms by which TARDBP mutations might lead to ALS. Others include disruption of nuclear bodies,14 increased aggregation properties or a toxic gain of function through novel protein–protein interactions.8
While the familial A315T,3 M337V4 G290A and G298S5 mutations have been shown to clearly segregate with disease, the majority of TARDBP mutations occur in patients with seemingly sporadic disease, suggesting they are de novo changes or, more likely, low penetrance variants that do not act like classical single gene mutations but require additional genetic or environmental modifiers to exert their effect. Thus TARDBP variants might enhance, rather than directly cause, a similar sequence of events leading to TDP-43 pathology in patients without genetic variance in the TARDBP locus. This is in keeping with the observation that while there is no qualitative difference between the neuropathology of ALS in patients with or without TARDBP mutations, the former might have more severe TDP-43 pathology such as more TDP-43 positive pre-inclusions.5 Functional studies of TARDBP mutations are needed but will have to take into account that the mutations alone might not be sufficient to cause a phenotype in the chosen model system. Of note, our patient did not have any evidence of cognitive involvement, a feature shared by previously published patients with TDP-43 mutations. However, given the small number of reported cases it is probably premature to conclude that sparing of higher cognitive function is characteristic of TARDBP related ALS.
Funding Work in our laboratory is supported by grants from the Motor Neuron Disease Association, the Spinal Muscular Atrophy Trust and the Muscular Dystrophy Campaign.
Competing interests None.
Ethics approval Ethics approval for the study was obtained from the Oxfordshire Research Ethics Committee, Oxford, UK.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.