Genetic analysis of the FUS/TLS gene in essential tremor

Abstract

Background and purpose

Although essential tremor (ET) has a genetic basis, specific genes have not been identified. Recently, in a large ET family (FET1) from Quebec, a non-sense mutation (p.Q290X) in the amyotrophic lateral sclerosis (ALS) gene fused in sarcoma/translated in liposarcoma (FUS/TLS) was identified by exome sequencing. No confirmatory studies have been published.

Methods

Two-hundred and fifty-nine ET cases and 262 controls were enrolled in a study at Columbia University. We performed a comprehensive analysis of the FUS/TLS gene by sequencing all exons in a subsample of 116 ET cases with early-onset (≤40 years) ET. We evaluated an association between ET and SNPs in the FUS/TLS gene by genotyping four haplotype tagging SNPs in all 259 ET cases and 262 controls. Additionally, seven variants associated with ALS, two variants of unknown pathogenicity detected in ALS cases, eight mis-sense variants predicted to be damaging, and six rare variants were genotyped in these 259 ET cases and 262 controls.

Results

FUS/TLS mutations previously reported in ALS, the FET1 family, or novel mutations were not found in any of the 116 early-onset ET cases. In the case–control analyses, although the power of the performed associations was limited, no significant association between tagging SNPs in FUS/TLS and ET was observed, and none of the analyzed SNPs showed evidence of association with ET.

Conclusion

Our study suggests that pathogenic mutations in FUS/TLS are rare in a sample of early-onset ET cases in North America. We did not find evidence that the FUS/TLS gene is a risk factor for ET.

Introduction

Essential tremor (ET) is one of the most common movement disorders, with prevalence estimated at 4.6% for individuals aged 65 years and older [1] and 20% or higher amongst persons in their 90s and older [2]. Family studies [3–5] and twin studies [6,7] have provided strong evidence for a genetic contribution to ET. Previously, we and others reported that aggregation of ET occurs within families, with many families containing >1 or multiple members with ET. Family and linkage studies of ET suggest an autosomal dominant mode of inheritance with reduced penetrance [8–11]; however, the number of studies is limited and other modes of inheritance including autosomal recessive and complex inheritance patterns may contribute to genetic risk in ET. To date, linkage studies have identified three susceptibility loci at chromosomes 3q13 (ETM1; OMIM:190300), 2p22-25 (ETM2; OMIM:602134), and 6p23 (ETM3; OMIM:611456) [9,10,12]; however, the genes and causal mutations have yet to be identified.

Recently, a genome-wide SNP association study of ET in an Icelandic population identified an associationwith a marker in the LINGO1 gene [13]. Since the initial report, several studies, including our own, have replicated the association in independent ET case–control samples in North America, Singapore, and Europe [14–18]. Collectively, these data suggest that the LINGO1 SNP rs9652490 confers modest increased risk for ET, with odds ratios (ORs) in the range 1.2–1.7 across different studies and populations; some have raised the possibility that it may represent a modifier of age at onset, rather than a causative locus or susceptibility gene [19].

In a recent study of a large family in which multiple family members were diagnosed with ‘definite’ ET, Merner et al. [20] identified variants in the FUS/TLS gene through whole exome sequencing. Mutations in FUS/TLS cause amyotrophic lateral sclerosis (ALS) [21,22] and frontotemporal lobar degeneration (FTLD) [23]. In ALS, FUS/TLS mutations account for approximately 4% of familial cases and <1% of sporadic cases. FUS, a ubiquitously expressed 526-amino-acid protein, encoded by 15 exons, belongs to the FET/TET family (FUS, EWS, and TAF15 proto-oncoproteins) of multifunctional DNA-/RNA-binding proteins [22].

In this study, at Columbia University, we evaluated possible links between the FUS/TLS gene and ET. The study had three aims. First, to identify mutations, we sequenced all FUS/TLS exons in 116 ET cases with early-onset disease (age of onset ≤40 years). Second, we performed an association analysis between ET and SNPs in the FUS/TLS gene; we genotyped four haplotype tagging SNPs in 259 ET cases and 262 controls. Third, to identify previously reported variants associated with ALS, or rare variants that might play a role in ET, we genotyped a total of 23 additional variants in these 259 ET cases and 262 controls (Table S1). Of these 23 variants, seven have been reported as pathogenic in ALS, two have been detected in ALS cases, but have not been confirmed as pathogenic, eight were predicted to be damaging in our ‘sorting tolerant from intolerant’ (SIFT) analysis, and six were rare variants in the ‘Utah residents with Northern and Western European ancestry from the CEPH collection’ (CEU) population with reported polymorphisms in other populations. These variants comprise the majority of all polymorphic variants reported in this gene.

Methods

Study cohort

As described [16], ET cases (N = 259) were recruited in a clinical-epidemiological study [24] at the Neurological Institute of New York, Columbia University, New York (2000–2007). Controls were ascertained from the same set of zip codes as cases and were recruited using random-digit telephone dialing, and frequency-matched on age (5-year strata), gender, and race categories. Each control was initially screened for tremor using a screening questionnaire and later underwent the same detailed videotaped neurological examination as the cases to ensure they did not have ET. All participants underwent a demographic and medical history questionnaire, a family history questionnaire (any first- or second-degree relative with non-specific tremor, ET or PD), and a videotaped neurological examination. Self-reported information on race and ethnic group was obtained. Beginning in 2002, self-reported information on Jewish ancestry was also collected. Data on age of onset of tremor, which we have shown to be reliable [25], were by self-report. On the basis of previous data on the distribution of age of onset in ET, early age of onset was designated as ≤40 years of age [26]. The Institutional review board at Columbia University Medical Center approved the protocol and consent procedures. Written informed consent was obtained from all participants in the study.

After review of the history and videotaped examinations, the diagnosis of ET was then reconfirmed by a senior neurologist specializing in movement disorders (E.D.L.) using published research criteria for possible, probable, or definite ET, which all required moderate amplitude or greater kinetic tremor on several tasks. Definite ET required both a moderate or greater amplitude postural tremor and moderate or greater amplitude kinetic tremor on four or more tasks, in the setting of no other tremor etiology [27]. The presence of bradykinesia or any other sign of parkinsonism (except isolated rest tremor) was an exclusionary criterion for ET. No cases or controls had a history of ALS or evidence of ALS on neurological examination.

There were initially 699 participants, of whom 617 (88.3%, including 328 ET cases and 289 controls) were non-Hispanic white. For the current analyses, we included 262 non-Hispanic white cases and 259 non-Hispanic white controls who had an available blood sample (total N = 521). Power calculations were performed using CaTS (http://www.sph.umich.edu/csg/abecasis/CaTS/index.html) [28,29]. Power calculations show that we have reasonable power to detect susceptibility variants with a genotype relative risk of 1.5–1.8 and higher; weaker variants would be unlikely to be detected by our approach.

Sanger sequencing

Genomic DNA was amplified by PCR using primer sets as described in the study by Merner et al. [20]. Details of PCR primers are available upon request. PCR was carried out using standard reaction and thermocycling conditions. Sanger sequencing of FUS/TLS exons was performed by GeneWiz (www.genewiz.com, South Plainfield, NJ, USA). Sequence was analyzed using Sequencher (Gene Codes Corporation, Ann Arbor, MI, USA) to compare sample sequence to human genome reference hg19 build GRCh37.

Analysis of predicted damaging amino acid substitutions

Predictions of the effects of coding non-synonymous variants on the function of the FUS/TLS protein (GI: 48145611) were performed using the SIFT BLink algorithm provided through J. Craig Venter Institute (http://sift.jcvi.org/) [30].

SNP selection and genotyping

Variants were selected for association analysis using Haploview (Broad Institute, Cambridge, MA, USA) to identify haplotype tagging SNPs in the region of the FUS/TLS gene with an r² threshold of 0.8 (Fig. 1). Additional variants with a minor allele frequency >5% in any recorded population and rare variants previously identified in ALS or ET cases were genotyped. Genotyping was performed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (Sequenom, San Diego, CA, USA) with Sequenom iPlex Gold custom assays designed using MassARRAY assay design software version 4.0 (Sequenom). Amplification was carried out on Applied Biosystems GeneAmp 9700 thermocyclers (Life Technologies, Carlsbad, CA, USA) using standard recommended cycling conditions for iPlex Gold assays. The mass of extension products was measured and recorded using a mass array compact mass spectrometer (Bruker Daltonik, Billerica, MA, USA). Genotypes were clustered and called using SpectroTYPER software version 2.0 (Sequenom). DNA samples were analyzed in duplicate, and genotype calls were assigned without prior knowledge of the diagnostic status of the samples analyzed. Genotypes that were discordant in replicates (N = 2) were manually called as no calls.

Figure 1.

Linkage disequilibrium patterns in FUS/TLS: Haplotype tagging SNPs were identified using Haploview version 4.2, pairwise tagging, with an r² threshold of 0.8 in the CEU HapMap population. Haplotype tagging SNPs are indicated in boldface. Nucleotides for HapMap genotyped SNP in the region are displayed in the top panel. Strong pairwise LD is indicated in red.

Statistical analysis

For our association analysis of variants in FUS/TLS, we genotyped 27 variants in 259 cases and 262 controls (Table S1). Of those 27 variants, eight were polymorphic and the remainder were monomorphic in our cohort. The polymorphic variants were tested for deviation from Hardy-Weinberg equilibrium (HWE) in PLINK (http://pngu.mgh.harvard.edu/~purcell/plink/) [31]. No deviations from HWE were observed in controls, in cases, or in the cohort as a whole. Association analysis was carried out in PLINK using chi-squared analysis to assess genotypic and allelic associations between ET and each of the markers analyzed. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated in PLINK. Haplotype analysis was also performed in PLINK to estimate haplotype blocks and test for association of a haplotype with ET.

Results

Clinical characteristics and demographics of sequenced and genotyped cases and controls

For the FUS/TLS analyses, 259 cases and 262 controls were similar in years of education (Table 1). An association has previously been shown between ET and years ofeducation [32]. A larger proportion of ET cases were males and Ashkenazi Jewish (AJ), and a higher proportion had a family history of ET (Table 1). Exons in FUS/TLR, were sequenced in 116 early-onset cases, defined as age 40 or younger. Of these cases, 35 were diagnosed as having definite ET, 47 had probable ET, and 25 had possible ET; 9 cases were not characterized as definite, possible, or probable. The mean age at the time of enrollment in the early-onset subgroup was 59 years (range: 21–89 years of age, SD: 17.95). The mean age of onset was 20.27 years of age (range: 3–40 years of age, SD: 10.99); 46.6% (N = 54) were males, 27.6% (N = 32) were of AJ ancestry, and 37.9% (N = 44) reported at least one family member with tremor. Of these early-onset cases, 107 were of non-Hispanic white ethnicity and were included in the larger cohort for association analysis. Nine early-onset cases were of Hispanic, African American, or other ancestry and were excluded from association analysis to prevent type-1 error due to population stratification.

Table 1.

Demographic and clinical characteristics of sequenced and genotyped subjects

	ET cases (N = 259)	Controls (N = 262)	Statistical test	P-value
ET diagnosis: definite/probable/possible	75/120/64	0	Not applicable	Not applicable
% male (N)	52.5 (136)	43.1 (113)	χ2 = 4.6	0.03
Mean age at tremor onset (years) (SD)	43.0 (23.1)	Not applicable	Not applicable	Not applicable
Mean years of education (SD)	15.3 (3.8)	15.6 (3.3)	Unpaired t-test	0.34
% with family history of ET (N)**	35.1 (91)	1.1 (3)	χ2 = 101.9	<0.0001
% Ashkenazi Jewish ancestry (N)*	36.7 (95)	22.9 (60)	χ2 = 11.8	0.008

ET, essential tremor.

^**P < 0.0001;

^*P < 0.05.

Sequence analysis of the FUS/TLS gene in early-onset ET cases

All exons in FUS/TLS were sequenced in 116 early-onset ET cases. Variants associated with ALS, rare (MAF < 1%) or novel variants were not detected in the coding regions of FUS/TLS.

FUS/TLS single point association

No significant associations were observed between ET and any of the four tag SNPs tested in the FUS/TLS gene (Fig. 1; Table 2). We also restricted the analysis to clinical subtypes including early-onset (≤40 years of age; Table S2), AJ ancestry (Table S3), or ET cases with a ‘definite’ diagnosis (Table S4); none of the SNPs showed evidence of association with ET.

Table 2.

Association between essential tremor and four haplotype tagging SNPs in the FUS/TLS gene in cases (N = 259) and controls (N = 262)

Chr	SNP	Positiona	Frequency in cases	Frequency in controls	χ 2	P-value	Odds ratio (95% CI)
16	rs741810	31193942	0.25	0.27	0.45	0.50	0.91 (0.69, 1.20)
16	rs1052352	31195279	0.37	0.41	1.44	0.23	0.86 (0.67, 1.10)
16	rs2735393	31196105	0.24	0.28	1.61	0.21	0.84 (0.63, 1.10)
16	rs4889537	31203529	0.24	0.27	1.62	0.20	0.83 (0.63, 1.10)

^aHuman genome reference hg19 build GRCh37.

ALS-associated, predicted damaging, and rare variants in FUS/TLS

We genotyped a total of seven ALS-associated variants in FUS/TLS, two variants of unknown pathogenicity previously detected in patients with ALS, eight variants predicted to be damaging by SIFT, and six rare variants in 259 ET cases and 262 controls (Table S1). None of the ALS-associated or predicted damaging variants were identified in cases or controls. Four of the rare variants were polymorphic in our cohort, but none were associated with ET.

Discussion

We performed a comprehensive genetic analysis of the FUS/TLS gene in a case–control study of ET conducted at Columbia University. Our results suggest that pathogenic mutations are rare in an early-onset ET case sample from North America. Although the power of the performed associations was limited, we found no evidence for the association of ET with vari-ants in the FUS/TLS gene. Notably, no previously unknown exonic variants were detected in 116 cases sequenced. All heterozygotes detected in these cases were common polymorphisms.

The strengths of our study are as follows. (i) We studied a well-characterized set of ET cases and controls that were enrolled in a clinical-epidemiological study at Columbia University; (ii) all cases and controls underwent a demographic and medical history questionnaire, a family history, and a videotaped neurological examination; (iii) we performed allelic association based on clinical subtypes including diagnostic criteria, early-onset disease, and AJ ancestry; (iv) genetic analysis was comprehensive and included sequencing of all coding regions in a subset of early-onset ET cases in addition to association analysis and genotyping of ALS-associated variants in a case–control study of ET.

The majority of clinically associated variants in FUS/TLS have been reported in patients with ALS (reviewed in Valdmanis et al. [33]), although there is a single report of a patient with FTLD with a novel mis-sense mutation [23]. The identification of a nonsense mutation (p.Q290X) in a large ET family (FET1) from Quebec suggests that the FUS/TLS gene may also cause ET [20]. Our current study is the first study to evaluate the role of FUS/TLS in other ET populations and suggests that pathogenic mutations are rare (or absent) and that the FUS/TLS gene is not a risk factor for ET in U.S. populations. Further studies in additional ET populations worldwide will be needed to confirm the relevance and role of FUS/TLS in the pathogenesis of ET.