Neurogenetics

, Volume 4, Issue 4, pp 185–189

Haplotype analysis of the ETM2 locus in familial essential tremor

Authors

    • Center for Human Genetics and Child NeurologyMid-Hudson Family Health Institute
  • Joseph Jankovic
    • Parkinson's Disease Center and Movement Disorders Clinic, Department of NeurologyBaylor College of Medicine
  • Roni Q. Lombardi
    • Center for Human Genetics and Child NeurologyMid-Hudson Family Health Institute
  • Joanna Pucilowska
    • Center for Human Genetics and Child NeurologyMid-Hudson Family Health Institute
  • Eng King Tan
    • Department of Neurology, Parkinson's Disease and Movement Disorders ProgramSingapore General Hospital
  • Tetsuo Ashizawa
    • Department of NeurologyUniversity of Texas Medical Branch
  • Melanie U. Ruszczyk
    • Center for Human Genetics and Child NeurologyMid-Hudson Family Health Institute
Original Article

DOI: 10.1007/s10048-003-0151-2

Cite this article as:
Higgins, J.J., Jankovic, J., Lombardi, R.Q. et al. Neurogenetics (2003) 4: 185. doi:10.1007/s10048-003-0151-2

Abstract

The objective of this study was to analyze a sample of unrelated individuals with autosomal dominant essential tremor (ET) for a genetic association with loci in a candidate region (ETM2) on chromosome 2p24.1 that harbors a disease gene for ET. ET is a common movement disorder that is genetically linked to ETM2 in four large families. It is unknown whether this candidate locus is associated with dominantly inherited ET in other individuals. Based on information from previous genetic linkage studies, a linkage disequilibrium study was designed to compare individuals with a family history of ET (n=45) with normal controls (n=70). Three unreported dinucleotide polymorphic loci (etm1240, etm1231, and etm1234) were identified on a physical map of the ETM2 interval in a region of no recombination. The study sample was tested for allele frequency differences by the CLUMP program and haplotypes were analyzed by the FASTEHPLUS program. The allele frequencies were significantly different between ET cases and the control samples for the loci etm1231 (P≤0.0419) and etm1234 (P<0.0001). A haplotype formed by the loci etm1231 and etm1234 occurred with a frequency of 29% in cases (n=45) and 9% in a white newborn sample (P<0.0001, n=35). The haplotype was not found in normal individuals older than 60 years without tremor (P=0.0063, n=35). This study provides evidence that an ancestral haplotype on chromosome 2p24.1 segregates with the ET disease phenotype in individuals with a family history of the disorder and will facilitate the search for a causative gene.

Keywords

Essential tremorHuman chromosomes, pair 2HaplotypesGenetic markersLinkage disequilibrium

Introduction

Essential tremor (ET) is one of the most-common neurological disorders in humans [1] and its effects are often debilitating [2]. The cardinal feature of ET is a bilateral, 4- to 10-Hz, postural tremor of the arms and hands, with variable involvement of the head, voice, and legs. The age distribution for the onset of tremor is bimodal, usually beginning before the 3rd decade or after the 6th decade [3, 4, 5]. The onset of tremor in the familial form of ET is uncommon after the age of 60 years [3]. ET can be inherited as an autosomal dominant trait with high penetrance [2, 6], but other movement disorders, including Parkinson's disease and dystonia, often coexist in families with ET [7, 8]. This phenotypic variability has fostered the hypothesis that several genes cause an ET "plus" phenotype. In the classic or "pure" form of ET, the identification of two susceptibility loci on chromosomes 2p24.1 (OMIM 602134) and 3q13 (OMIM 190300) [9, 10, 11] suggests genetic heterogeneity. Besides phenotypic variability and genetic heterogeneity, research efforts to find genes for ET are complicated by the presence of phenocopies and the lack of large family pedigrees with a "pure" ET phenotype.

A powerful alternative to family studies involves searching for linkage disequilibrium (LD) in populations by using a case-control strategy. LD studies are based upon the hypothesis that all current copies of the disease gene in a population are descended from a single ancestral mutation. The objective of these studies is to demonstrate that a disease-causing gene and a marker allele have survived in their original combination through long periods of evolution. This is accomplished by analyzing a series of polymorphic loci for allele frequency differences between cases and controls and to determine if certain allele combinations or haplotypes are associated with a disease locus [12]. The preservation of a haplotype that contains the founder mutation is dependent upon the distance from the mutation and the number of copies of the mutant gene in the study population. Testing for a genetic association by this approach can give significant results using markers as far as 400 kb from the putative mutation [12]. The present study uses three novel loci (designated as etm1240, etm1231, and etm1234) in a 133-kb region of no recombination within the ETM2 (http://www.gdb.org) contig to establish the persistence of significant allele and haplotype associations with an ET disease gene.

Materials and methods

Diagnostic criteria

The Mid-Hudson Family Health Institute and the Baylor College of Medicine Institutional Review Boards approved this research study. Study participants in the case group and a control group of individuals older than age 60 years without tremor were recruited from a movement disorders clinic in Houston, Texas. Informed consent was obtained from each individual before genetic and clinical tests. Criteria for study participation included a family history of ET consistent with a dominant mode of inheritance and a diagnosis of "definite" ET as defined by the National Institutes of Health Diagnostic Criteria for Essential Tremor [13] and the Tremor Investigation Group [14]. The diagnosis of "definite" ET was based on the presence of bilateral postural tremor of the hands or forearms that was visible and persistent with 2- to 4-cm excursions in at least one arm. Tremor assessments were made when the subjects' arms were outstretched in front of the body or in a wing-beating position (e.g., the elbows partially flexed and the shoulders abducted in the horizontal plane). Subjects with other movement disorders such as parkinsonism, dystonia, myoclonus, peripheral neuropathy, or "restless" legs syndrome were excluded from the study. Normal allele frequencies of one control sample were determined from blood spot samples of randomly selected, white infants born throughout New York State. To confirm that the data were not dependent upon the selection of the control group, cases were compared with a population of individuals older than age 60 years without tremor, because of their low risk of having the familial form of ET [3].

Genotyping

High molecular weight genomic DNA was isolated from whole-blood lysate by standard phenol/chloroform extraction with isopropanol precipitation or by using the Puregene DNA extraction kit (Gentra Systems, Minneapolis, Minn., USA). We identified 19 polymorphic loci in the course of constructing a physical map of the candidate region. These loci were used in fine mapping studies to delimit the candidate region in four families with ET [10]. Three polymorphic loci designated as etm1240, etm1231, and etm1234 (http://www3.ncbi.nlm.nih.gov; GenBank accession numbers: BV012542, BV012543, BV012544) are contained in a region of no recombination on the bacterial artificial chromosome clone, 572-N-24, (Research Genetics, Huntsville, Ala., USA) at physical distances of etm1240-58 kb-etm1231-75 kb-etm1234. The oligonucleotide primers that flanked the polymorphic regions of these loci included etm1240 (sense 5′-TGCTTGCCTGTAGTCCTAGC-3′, antisense 5′-TGCACCCACCCTAGGTACAATG-3′), etm1231 (sense 5′-TGTCAGTCTATCAGAGACGGAC-3′, antisense 5′-TGCTCATTAGTTTGTGGGAGTTAG-3′), and etm1234 (sense 5′-ACTTGGAGAATCCACCTGTATC-3′, antisense 5′-TCTTCAGGGATAATAAGTGTGGACC-3′). The primers were labeled with IR-DYE 700 (Li-Cor, Lincoln, Neb., USA). Each polymerase chain reaction (PCR) contained 50 ng of genomic DNA in a total volume of 9 μl with 1.0 pmol/μl of each primer, 200 μM of each dNTP, 100 mM TRIS HCl (pH=9.0), 500 mM KCl, 15 mM MgCl2, and 0.5 units of Taq polymerase (Fisher Scientific, Suwanee, Ga., USA). Reactions were performed in a 96-well microtiter plate on a Genius thermocycler (Techne, Princeton, N.J., USA). Amplification was carried out with an initial denaturation step of 5 min at 95°C followed by 36 cycles of denaturation for 20 s at 95°C, 20 s of annealing at either 60°C (etm1231), 58°C (etm1234), or 62°C (etm1240), and 30 s of extension at 72°C. The last extension step was 3 min at 72°C. At the completion of PCR, 5 μl of Stop Loading Buffer (Li-Cor) was added to each sample, denatured for 5 min at 95°C, and immediately placed on ice. The reaction (1 μl) was electrophoresed on a KB plus 6.5% gel matrix (Li-Cor) and visualized on the Li-Cor Global IR2 DNA Analyzer (Li-Cor). The allele sizes were estimated using the SAGAGT version 2.1 software (Li-Cor). Genotypes were assigned without prior knowledge of the study participant's affection status.

Statistics

Allele frequencies were calculated for loci etm1240, etm1231, and etm1234, and analyzed for statistically significant differences between cases and controls by the CLUMP program [15]. The CLUMP program uses a Monte Carlo approach to assess the significance of the departure of allele frequencies between cases and controls when loci with multiple alleles are studied. Linkage disequilibrium was evaluated between the pairs of loci using the FASTEHPLUS program [16]. The FASTEHPLUS is based on the EH program [17] and performs model-free analysis and permutation tests of allelic association.

Results

Demographics

All controls (n=70) and cases with ET (n=45) were white. The ET case group and the control group older than 60 years were of European ancestry. Both sexes were well represented in the control (43% male, 57% female) and case groups (41% male, 59% female). Two control groups, a newborn panel and an older control group [mean age (years)±SD, range; 72±5, 62–81] without tremor, were compared with the ET case group (60±18, 14–84) (Tables 1 and 2). The onset (35±21,12–70) and duration (38±18, 1–57) of tremor varied among cases. No inter-family relationships could be established between the 45 pedigrees constructed for the ET case group. The families had two (n=25), three (n=13), four (n=6), or five consecutive generations (n=1) of members affected by ET.
Table 1.

Analysis of allele frequency differences between essential tremor and control samples

Loci

etm1240

etm1231

etm1234

 

Cases % (n)

Controls ≥age 60 years % (n)

Controls newborns % (n)

Cases % (n)

Controls ≥age 60 years % (n)

Controls newborns % (n)

Cases % (n)

Controls ≥age 60 years % (n)

Controls newborns % (n)

Alleles

1

4.9 (4)

1.4 (1)

1.5 (1)

0.0 (0)

1.7 (1)

1.4 (1)

2.3 (2)

0.0 (0)

0.0 (0)

2

11.0 (9)

2.9 (2)

1.5 (1)

10.0 (9)

5.2 (3)

12.9 (9)

15.9 (14)

0.0 (0)

9.3 (5)

3

3.7 (3)

1.4 (1)

13.2 (9)

14.4 (13)

1.7 (1)

14.3 (10)

20.5 (18)

0.0 (0)

5.6 (3)

4

2.4 (2)

0.0 (0)

4.4 (3)

56.7 (51)

72.4 (42)

64.3 (45)

1.1 (1)

1.9 (1)

18.5 (10)

5

29.3 (24)

28.6 (20)

26.5 (18)

17.8 (16)

19.0 (11)

2.9 (2)

3.4 (3)

7.4 (4)

13.0 (7)

6

6.1 (5)

8.6 (6)

7.4 (5)

1.1 (1)

0.0 (0)

4.3 (3)

12.5 (11)

33.3 (18)

3.7 (2)

7

0.0 (0)

0.0 (0)

4.4 (3)

   

1.1 (1)

9.3 (5)

29.6 (16)

8

0.0 (0)

0.0 (0)

1.5 (1)

   

43.2 (38)

48.1 (26)

20.4 (11)

9

42.7 (35)

57.1 (40)

39.7 (27)

      

P valuea

 

0.0769

0.0829

 

0.0419

0.0349

 

0.0001

0.0001

aThe data were analyzed by the CLUMP program version 2.1. The P value was based on the T4 statistic [15]. The haplotype formed by allele 4 (etm1231) and allele 2 (etm1234) occurs with a frequency of 29% in cases, 9% in the newborn control group (P<0.0001), and none in normal individuals older than 60 (P=0.0063)

Table 2.

Test of allelic combinations for an association with a gene within the ETM2 candidate interval

Cases versus controls (≥age 60 years)

Empirical

Loci combinations

χ2

Degrees of freedom

P value

P valuea

etm1240+etm1231

56.86

19

<0.0001

0.1018

etm1231+etm1234

47.06

19

0.0002

0.0063

etm1240+etm1234

68.39

24

<0.0001

0.5839

etm1240+etm1231+etm1234

75.93

79

0.5771

0.0064

Cases versus controls (newborns)

Loci combinations

etm1240+etm1231

34.18

16

0.0051

0.0348

etm1231+etm1234

79.83

34

<0.0001

<0.0001

etm1240+etm1234

45.45

41

0.2349

0.8744

etm1240+etm1231+etm1234

59.43

90

0.9947

0.0005

aThe data were analyzed by the FASTEHPLUS program [16]

Genotyping

One hundred and fifteen individuals were genotyped at the polymorphic loci, etm1240, etm1231, and etm1234. Figure 1 illustrates the number and sizes of the alleles at these loci. Locus etm1240 has nine alleles with sizes ranging from 264 base pairs (bp) to 290 bp. Locus etm1231 has six alleles with sizes ranging from 177 bp to 195 bp. Locus etm1234 has eight alleles with sizes ranging from 137 bp to 159 bp. The degree of heterozygosity,
Fig. 1.

Polyacrylamide gel electrophoresis of three novel loci within the ETM2 locus on chromosome 2p. Lane 1 is a ladder in base pairs that shows the relative allele sizes of the three loci etm1240 (top), etm1231 (middle), and etm1234 (bottom). The actual allele sizes were assigned using the SAGAGT version 2.1 software (Li-Cor, Lincoln, Neb., USA). Lanes 2 through 11 contain the PCR products of representative samples demonstrating the number of alleles for the loci. Nine alleles were identified at locus etm1240 with sizes of 290, 286, 284, 280, 278, 274, 272, 268, and 264 base pairs. Six alleles were identified at locus etm1231 with sizes of 195, 193, 191, 185, 181, and 177 base pairs. Eight alleles were identified at locus etm1234 with sizes of 159, 155, 153, 151, 146, 144, 142, and 137 base pairs. The horizontal lines drawn at the left correspond to the position and number of alleles for the locus

$$ {\rm{H=1 - }}\sum {{\rm{p}}^{\rm{2}} _{\rm{i}} } $$
for the polymorphic loci is equal to the following values: etm1240=0.68, etm1231=0.55, and etm1234=0.81.

Allele frequency studies

Table 1 shows the results of the allele frequency differences between cases and controls using the CLUMP program [15]. The locus etm1240 does not show significant allele frequency differences between cases and the control groups older than age 60 years (P=0.0769) or newborns (P=0.0829). The loci etm1231 and etm1234 show significant differences between cases and both control groups (P=0.0001). These results suggest an association (i.e., linkage disequilibrium) between these loci and a putative gene for ET in this sample.

Pair-wise analysis of marker associations

The analysis was extended to examine pairs of allelic combinations at two or three sites near an ET gene by using the FASTEHPLUS program [16]. The results (Table 2) show an improvement to the significance (empirical P values) of the loci etm1231 and etm1234 compared with when the loci were analyzed independently (Table 1). For example, combining the allele patterns gives a P value for heterogeneity of 0.0063 in the control group older than age 60 years and less that 0.0001 in newborn controls. The haplotype that distinguished best between cases and controls was the haplotype formed by the loci etm1231 and etm1234. This haplotype consisted of the "4" (etm1231) and the "2" (etm1234) alleles (Fig. 1) with a frequency of 29% in cases and 9% in the newborn control group (P<0.0001) and none in normal individuals older than 60 years (P=0.0063).

Discussion

In the present study we demonstrate LD between a disease gene and the loci etm1231 and etm1234 in unrelated individuals with a family history of ET. These findings are significant for at least two reasons. First, the data show that there is no significant recombination between the loci studied and a putative gene for ET. Second, the results suggest that a common ancestral haplotype segregates with ET in our affected sample. LD studies provide a powerful alternative to traditional linkage analysis, especially when the disorder is multifactorial and the underlying etiology may involve genetic and environmental factors [17, 18]. To date, family studies on ET have been inadequate to narrow the candidate region, ETM2, which contains an ET gene because of uninformative recombinants and the presence of phenocopies. These problems are commonplace in studying prevalent traits such as tremor. The data derived from this study will help refine the ETM2 candidate region by focusing sequencing and mutational analyses on candidate genes at or near the locus etm1234.

This study also illustrates the capability of the CLUMP [15]and FASTEHPLUS [16] programs to detect LD in a situation where traditional linkage analysis is unable to delimit a minimal critical region that contains a disease gene. Other investigators have successfully tested for allele associations between polymorphic markers and known genetic mutations using these methods. In one study, a set of polymorphic loci with known physical distances from a mutation in the aldehyde dehydrogenase 2 (ALDH2) gene were analyzed in controls and in a population of Japanese alcoholics [12, 19, 20, 21]. Significant LD persisted at physical distances up to 400 kb from this ALDH2 gene mutation. The combination of the CLUMP [15] and FASTEHPLUS [16] programs permitted us to analyze 432 possible haplotypes and 272,160 possible genotypes by a permutation strategy that dealt with rare haplotypes. Other investigators have also found this strategy useful in identifying genetic mutations and mapping their region of interest [22].

The approach used in this study is distinctly different from other studies that search for a genetic association between an attractive candidate gene and a disease phenotype. In contrast to these studies, we first mapped the location of an ET gene by traditional genetic linkage analysis [9, 10] and then used a physical map of this region to identify an interval of no recombination. Within this interval, polymorphic loci were used to study a sample of unrelated individuals with a dominant form of ET. In general, genetic association studies do not apply the information from linkage analysis to populations. This feature is worth emphasizing because our results suggest that a single gene may cause ET not only in the four affected families that we previously reported with genetic linkage to chromosome 2p24.1 [9, 10], but also in other individuals with a dominant form of the disorder. These results will be confirmed in larger populations and efforts will focus on delimiting the ETM2 candidate interval by extending the haplotype formed by the loci etm1231 and etm1234 to other contiguous loci.

Acknowledgements.

The experiments in this study comply with Title 45 of the Code of Federal Regulations Part 46, Protections of Human Subjects, of the United States Department of Health and Human Services, National Institutes of Health, Office for Protection from Research Risks. The National Institutes of Neurological Disorders and Stroke provided funding for this research (grant R01 NS39353 to J.J.H.).

Copyright information

© Springer-Verlag 2003