Abstract
Background
Two recent studies investigated the association of the MAPT H1 haplotype, a known risk factor for neurodegenerative disease including progressive supranuclear palsy and Parkinson’s disease (PD), with essential tremor (ET).
Methods
To confirm this association in a different population we analyzed the distribution of allele and genotype frequencies for the MAPT H1/H2 tagging SNP rs1052553 in ET cases and controls enrolled in a clinical-epidemiological study of ET at Columbia University.
Results
Overall, no association was observed between ET and the MAPT H1 haplotype. We also restricted the analysis to clinical subtypes including early-onset (≤40 years of age), Ashkenazi Jewish ancestry, white non-Ashkenazi, or ET cases with a ‘definite’ or ‘probable/possible’ diagnosis; none of these stratified analyses showed evidence of association with ET. We also performed a meta-analysis of the H1/H2 tagging SNP rs1052553 in published datasets and the H1 haplotype with risk for ET in the current study and did not find evidence for association.
Conclusions
The inconsistent reports of association of MAPT H1 in three emerging studies (our own and two published studies) may reflect sampling issues and/or clinical heterogeneity in these populations.
Keywords: essential tremor, MAPT, SNPs, case-control study, candidate genes
Introduction
Essential tremor (ET) is one of the most common movement disorders, with prevalence estimated at 4.6% for individuals age 65 years and older and as high as 20% or more among persons in their 90s and older [1]. Family studies and twin studies have provided strong evidence for a genetic contribution to ET [2,3]. A number of variants in the MAPT gene have been linked with FTD and an increased risk of PSP, PD, and in some studies, with AD, as well as ET more recently [4]. While one study suggests that the MAPT H1 haplotype is associated with increased risk for ET in a North American population [5], a study in Spain did not find evidence for an association between a familial ET and MAPT H1 haplotype [6]. In this study, we evaluated an association between the MAPT H1 haplotype and ET. First, we analyzed the distribution of allele and genotype frequencies for the MAPT H1/H2 tagging SNP ‘rs1052553’ in ET cases and controls. Second, we performed a meta-analysis of one H1/H2 tagging SNP, rs1052553, with available data in published datasets and the current study with risk for ET.
Methods
Study Cohort
A description of the study cohort is provided in Clark et al (2010) [7]. ET cases (n=249) and controls (n=237) were enrolled in a clinical-epidemiological study at the Neurological Institute, Columbia University, New York (2000–2007). All participants underwent a demographic and medical history questionnaire, family history questionnaire (any first- or second-degree relative with nonspecific tremor, ET or PD), and a videotaped neurological examination. Using published research criteria a diagnosis of ET for possible, probable, or definite ET was confirmed (E.D.L.). The clinical characteristics and demographics of genotyped cases and controls is summarized in Table S1
For the current analyses, genotypes for rs1052553 were available for 249 non-Hispanic white cases and 237 non-Hispanic white controls (total N=486).
The study was approved by the CUMC Institutional Review Board. All participants provided signed informed consent.
SNP Genotyping
All genotyping was performed in duplicate and blinded to case control status.
Genotyping for SNP rs1052553 was performed by Taqman allelic discrimination assay (Life Technologies, Carlsbad, California, USA) (‘C_7563736_10’, catalog number 4351376). All samples genotyped successfully by Taqman allelic discrimination and were 100% concordant with H1/H2 haplotype calls recorded by Sequenom genotyping for SNP rs62063857.
Statistical Analysis
The SNP was tested for deviation from Hardy-Weinberg equilibrium (HWE) in PLINK [8]. No deviations from HWE were observed. Association analysis was carried out in PLINK using chi-squared analysis to assess genotypic and allelic associations. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated in PLINK. Meta-analysis was performed for rs1052553 using METAL [9]. In a secondary analysis we also performed a meta analysis using a random and fixed effects model using the program Comprehensive Meta-analysis (CMA) [10]. Power calculations were performed using G* power version 3.1.7 [11].
Results
MAPT H1 Association and ET
Overall, we did not observe an association of the MAPT H1 haplotype and ET by genotyping the MAPT H1 /H2 tagging SNP rs1052553 (Table 1). We also restricted the analysis to clinical subtypes; although the analysis in white non-AJ cases and controls showed a trend towards association (p=0.07) none of the stratified analyses showed evidence of association of the H1 haplotype with risk of ET.
Table 1.
Association between ET and the MAPT H1/H2 tagging SNP rs1052553
| SNP | Controls | White non-Hispanic cases | Definite diagnosis cases | Probable/Possible diagnosis cases | Early onset <40 years cases | White non-AJ cases and controls | AJ cases and controls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| rs1052553 | Controls (N=237) (Count, freq.) |
Cases (N=249) (Count, freq.) |
P (OR, 95% CI) | Cases (N=75) (Count, freq.) |
P (OR, 95% CI) | Cases (N=174) (Count, freq.) |
P (OR, 95% CI) | Cases (N=99) (Count, freq.) |
P (OR, 95% CI) | Cases (N=157) (Count, freq.) |
Controls (N=180) (Count, freq.) |
P (OR, 95% CI) | Cases (N=92) (Count, freq.) |
Controls (N=57) (Count, freq.) |
P (OR, 95% CI) |
| Genotype | |||||||||||||||
| AA (H1/H1) | 143 (0.60) | 140 (0.56) | 0.16 | 40 (0.53) | 0.18 | 100 (0.57) | 0.27 | 59 (0.60) | 0.55 | 88 (0.56) | 114 (0.63) | 0.12 | 52 (0.57) | 29 (0.51) | 0.64 |
| AG (H1/H2) | 86 (0.36) | 91 (0.37) | 29 (0.39) | 62 (0.36) | 34 (0.34) | 58 (0.37) | 61 (0.34) | 33 (0.36) | 25 (0.44) | ||||||
| GG (H2/H2) | 8 (0.03) | 18 (0.07) | 6 (0.08) | 12 (0.07) | 6 (0.06) | 11 (0.07) | 5 (0.03) | 7 (0.08) | 3 (0.05) | ||||||
| Allele | |||||||||||||||
| A | 372 (0.78) | 371 (0.74) | 0.14 (0.80, 0.59–1.08) | 109 (0.73) | 0.14 (0.73, 0.48- | 262 (0.75) | 0.28 (0.84, 0.60–1.16) | 152 (0.77) | 0.63 (0.91, 0.61–1.35) | 234 (0.75) | 289 (0.80) | 0.07 (0.72, 0.50–1.03) | 137 (0.74) | 83 (0.73) | 0.75 (1.09, 0.64–1.85) |
| G | 102 (0.22) | 127 (0.26) | 41 (0.27) | 86 (0.25) | 46 (0.23) | 80 (0.25) | 71 (0.20) | 47 (0.26) | 31 (0.27) | ||||||
Meta-analysis of MAPT H1/H2 Tagging SNP and Risk of ET
We performed a combined analysis including data for the MAPT H1/H2 tagging SNP rs1052553 from the current study and two published studies; when examined together we found no evidence for association of the ‘A’ (H1 haplotype) allele with ET (p=0.75; OR=1.03, 95% CI: 0.88–1.20) nor in secondary analyses when we removed samples with AJ ancestry (p=0.07; OR=0.72, 95% CI: 0.50–1.03). We also performed meta analysis in METAL; we again failed to find evidence for association of the MAPT H1 haplotype in our study and published studies (p=0.849 in METAL) (Table 2). In a secondary analysis, using the program CMA we performed meta analysis for rs1052553 using a fixed and random effect model for two datasets and the current study removing controls with a family history of ET or PD and samples with AJ ancestry and again found no evidence for association of the MAPT H1 haplotype (Table S2).
Table 2. Meta-analysis of the H1/H2 tagging SNP rs1052553 data from the present study and rs1052553 data from Vilariño-Güell et al.
and García-Martín et al. Analysis was performed in METAL. Effective sample size is corrected for unequal numbers of cases and controls in METAL.
| Study | (White non-Hispanic) Clark et al. (rs1052553) | (White non-AJ) Clark et al. (rs1052553) | Vilariño-Güell et al. (rs1052553) | García-Martín et al. (rs1052553) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cases (N=249) (Count, freq.) |
Controls (N=237) (Count, freq.) |
P (OR, 95% CI) | Cases (N=157) (Count, freq.) |
Controls (N=196) (Count, freq.) |
P (OR, 95% CI) | Cases (N=339) (Count, freq.) |
Controls (N=406) (Count, freq.) |
P (OR, 95% CI) | Cases (N=200) (Count, freq.) |
Controls (N=291) (Count, freq.) |
P (OR, 95% CI) | |
| Genotype | ||||||||||||
| AA (H1/H1) | 140 (0.56) | 143 (0.60) | 0.16 | 88 (0.56) | 125 (0.64) | 0.13 | 221 (0.65) | 231 (0.57) | 0.07 | 104 (0.52) | 158 (0.54) | 0.52 |
| AG (H1/H2) | 91 (0.37) | 86 (0.36) | 58 (0.37) | 65 (0.33) | 100 (0.30) | 148 (0.36) | 75 (0.38) | 111 (0.38) | ||||
| GG (H2/H2) | 18 (0.07) | 8 (0.03) | 11 (0.07) | 6 (0.03) | 18 (0.05) | 27 (0.07) | 21 (0.11) | 22 (0.08) | ||||
| Allele | ||||||||||||
| A | 371 (0.74) | 372 (0.78) | 0.13 (0.80, 0.60–1.07) | 234 (0.75) | 315 (0.80) | 0.06 (0.71, 0.50–1.02) | 542 (0.80) | 610 (0.75) | 0.03 (1.32, 1.03–1.69) | 283 (0.71) | 427 (0.73) | 0.37 (0.88, 0.66–1.17) |
| G | 127 (0.26) | 102 (0.22) | 0.13 (1.25, 0.93–1.67) | 80 (0.25) | 77 (0.20) | 0.06 (1.41, 0.98–2.00) | 136 (0.20) | 202 (0.25) | 0.03 (0.76, 0.59–0.97) | 117 (0.29) | 155 (0.27) | 0.37 (1.13, 0.86–1.51) |
Discussion
We performed an analysis of the MAPT H1 haplotype in a case-control study of ET at Columbia University. Our results suggest that the MAPT H1 haplotype is not a risk factor for ET in a Caucasian sample from North America. We also restricted the analysis to clinical subtypes including early-onset (≤40 years of age), AJ ancestry, white non Ashkenazi, or ET cases with a ‘definite’ or ‘probable/possible’ diagnosis; although the analysis in white non-AJ cases and controls showed a trend towards association (p=0.07) the MAPT H1 haplotype was not significantly associated with ET in any of these stratified analyses. A meta-analysis and secondary meta-analysis including a fixed and random effect model, which included our data and two published datasets, also did not provide evidence for association.
We calculated the statistical power for sample sizes in this study, published data [5,6]) and using pooled data from all three studies based on the allele frequencies and the odds ratio reported in the study from Vilarino-Guell et al. [5]. Our study had similar power to the published study from Garcia Martin et al. [6]. Pooling of data from all three studies shows that the analysis was adequately powered and the respective values with p=0.05 for one-tailed and two-tailed associations were 98% and 96%.
The inconsistent reports of association of MAPT H1 in three emerging studies (our own and two published studies [5,6]) may reflect sampling issues or clinical heterogeneity in these populations. We note the small sample size in the current study; however meta-analysis with published data [5,6] with a combined sample size of 788 ET cases and 934 controls also does not support association. Further association studies of MAPT in ET populations are warranted.
Supplementary Material
Supplementary Table 1. Demographic and Clinical Characteristics of Genotyped Subjects
Supplementary Table 2. Meta analysis for rs1052553 using a Fixed and Random Effect Model
Table 3.
| Meta association | SNP | Study a | Allele1 | Allele2 | Freq Allele 1 | Min Freq | Max Freq | Weight b | Z score c | P value |
|---|---|---|---|---|---|---|---|---|---|---|
| White non-Hispanic | rs1052553 | 1,2,3 | A | G | 0.74 | 0.71 | 0.8 | 1722 | 0.19 | 0.849 |
| White non-AJ | rs1052553 | 1,2,3 | A | G | 0.75 | 0.71 | 0.8 | 1573 | 0.18 | 0.857 |
| Merged association | SNP | Study | Allele1 | Allele2 | Freq Allele1 in cases | Freq Allele1 in controls | OR | L95 | U95 | P |
| White non-Hispanic | rs1052553 | 1,2,3 | A | G | 0.76 | 0.75 | 1.03 | 0.88 | 1.20 | 0.754 |
| White non-AJ | rs1052553 | 1,2,3 | A | G | 0.75 | 0.80 | 0.72 | 0.50 | 1.03 | 0.074 |
Only rs1052553 was genotyped in all studies.
1, in this study, 2, Vilariño-Güell et al., 3, García-Martín et al.
Weight represents the total number of subjects genotyped
The direction of the Z score represents allelic association with allele 1
Acknowledgments
Dr. Louis has received support from the National Institutes of Health: NINDS #R01 NS042859 (PI), NINDS #R01 NS39422 (PI), NINDS #T32 NS07153-24 (PI), NINDS #R01 NS073872 (PI), NINDS #R21 NS077094 (CoI), and NINDS #R01 NS36630 (coI), as well as the Parkinson’s Disease Foundation (PI), the Arlene Bronstein Essential Tremor Research Fund (Columbia University), and the Claire O’Neil Essential Tremor Research Fund (Columbia University).
Dr. Clark is funded by NIH grants R21NS050487 (PI), R01NS060113 (PI), R01NS0738072 (CoPI), P50AG008702 (CoI), P50 NS038370 (CoI), the Parkinson’s Disease foundation (PI) and the Michael J Fox foundation (CoI).
Footnotes
Disclosure of Potential Conflicts of Interest: None
References
- 1.Louis ED, Ferreira JJ. How common is the most common adult movement disorder? Update on the worldwide prevalence of essential tremor. Mov Disord. 2010;25:534–1. doi: 10.1002/mds.22838. [DOI] [PubMed] [Google Scholar]
- 2.Louis ED, Ford B, Frucht S, Ottman R. Mild tremor in relatives of patients with essential tremor: what does this tell us about the penetrance of the disease? Arch Neurol. 2001;58:1584–9. doi: 10.1001/archneur.58.10.1584. [DOI] [PubMed] [Google Scholar]
- 3.Lorenz D, Frederiksen H, Moises H, Kopper F, Deuschl G, Christensen K. High concordance for essential tremor in monozygotic twins of old age. Neurology. 2004;62:208–11. doi: 10.1212/01.wnl.0000103236.26934.41. [DOI] [PubMed] [Google Scholar]
- 4.Kalinderi K, Fidani L, Bostantjopoulou S. From 1997 to 2007: a decade journey through the H1 haplotype on 17q21 chromosome. Parkinsonism Relat Disord. 2009;15:2–5. doi: 10.1016/j.parkreldis.2008.03.001. [DOI] [PubMed] [Google Scholar]
- 5.Vilarino-Guell C, Soto-Ortolaza AI, Rajput A, Mash DC, Papapetropoulos S, Pahwa R, et al. MAPT H1 haplotype is a risk factor for essential tremor and multiple system atrophy. Neurology. 2011;76:670–2. doi: 10.1212/WNL.0b013e31820c30c1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Garcia-Martin E, Martinez C, Alonso-Navarro H, Benito-Leon J, Lorenzo-Betancor O, Pastor P, et al. H1-MAPT and the risk for familial essential tremor. PLoS One. 2012;7:e41581. doi: 10.1371/journal.pone.0041581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Clark LN, Park N, Kisselev S, Rios E, Lee JH, Louis ED. Replication of the LINGO1 gene association with essential tremor in a North American population. Eur J Hum Genet. 2010;18:838–43. doi: 10.1038/ejhg.2010.27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81:559–75. doi: 10.1086/519795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26:2190–1. doi: 10.1093/bioinformatics/btq340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Borenstein M, Hedges L, Higgins J, Rothstein H. Comprehensive Meta-analysis Version 2. Biostat; Englewood NJ: 2005. [Google Scholar]
- 11.Faul F, Erdfelder E, Buchner A, Lang A-G. Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods. 2009;41:1149–1160. doi: 10.3758/BRM.41.4.1149. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1. Demographic and Clinical Characteristics of Genotyped Subjects
Supplementary Table 2. Meta analysis for rs1052553 using a Fixed and Random Effect Model