Genetic testing in spinocerebellar ataxia in Taiwan: expansions of trinucleotide repeats in SCA8 and SCA17 are associated with typical Parkinson's disease
Abstract
DNA tests in normal subjects and patients with ataxia and Parkinson's disease (PD) were carried out to assess the frequency of spinocerebellar ataxia (SCA) and to document the distribution of SCA mutations underlying ethnic Chinese in Taiwan. MJD/SCA3 (46%) was the most common autosomal dominant SCA in the Taiwanese cohort, followed by SCA6 (18%) and SCA1 (3%). No expansions of SCA types 2, 10, 12, or dentatorubropallidoluysian atrophy (DRPLA) were detected. The clinical phenotypes of these affected SCA patients were very heterogeneous. All of them showed clinical symptoms of cerebellar ataxia, with or without other associated features. The frequencies of large normal alleles are closely associated with the prevalence of SCA1, SCA2, MJD/SCA3, SCA6, and DRPLA among Taiwanese, Japanese, and Caucasians. Interestingly, abnormal expansions of SCA8 and SCA17 genes were detected in patients with PD. The clinical presentation for these patients is typical of idiopathic PD with the following characteristics: late onset of disease, resting tremor in the limbs, rigidity, bradykinesia, and a good response to levodopa. This study appears to be the first report describing the PD phenotype in association with an expanded allele in the TATA-binding protein gene and suggests that SCA8 may also be a cause of typical PD.
The spinocerebellar ataxias (SCAs) are a group of neurodegenerative disorders characterized by cerebellar dysfunction alone or in combination with other neurological abnormalities. The clinical classification of familial SCA has remained difficult because of the variability of the clinical features both among and within families. Nonetheless, the isolation and characterization of the implicated genes affords a better understanding of the molecular basis of SCA. The expansions of coded CAG trinucleotide repeats have been shown to cause dominantly inherited SCA1, SCA2, MJD/SCA3, SCA6, SCA7, SCA17, and dentatorubropallidoluysian atrophy (DRPLA) (1–7). The elongated polyglutamine tract in the respective proteins leads to a gain of function that is toxic to neurons (8). However, SCA8, SCA10, and SCA12 are associated with expansions of 3′ UTR CTG, intronic ATTCT, and promoter CAG, respectively (9–11).
SCA8 was first described by Koob and co-workers (9) as due to a CTG-repeat expansion in the 3′ UTR region of a Kelch-like 1 (KLHL1) antisense RNA gene (12). Since then, the sizing of SCA8 alleles has been clarified in various populations. In general, more than 99% of the normal SCA8 alleles had 16–37 repeats, while unrelated expanded alleles ranging from 68 to 800 repeats were found in familial and sporadic ataxia patients (13–23). Both expanded and normal SCA8 repeats are highly unstable (24). In some affected families, expanded alleles do not always cosegregate with the disease phenotype (9, 13, 15–17, 20, 25, 26). In addition, SCA8 alleles with expansion can be found in rare instances in the general population (9, 13, 15, 16, 21, 23) as well as in patients with psychiatric disorders, Freidreich's ataxia, Parkinson's disease (PD), and Alzheimer's disease (16, 19, 23, 27, 28). SCA17 is caused by a CAG expansion in the coding region of the transcription factor TBP gene (6). In the general population, the polymorphic CAG repeats range in size from 25 to 44 repeats (29–32). Expanded alleles from 43 to 63 repeats have been characterized in familial and sporadic ataxia patients (31–35). In addition to ataxia and cognitive decline, dystonia and/or parkinsonism, as well as reduced penetrance were also observed (31–33, 35).
The prevalence of SCAs differs among populations. Such differences were closely associated with the distributions of large normal alleles in Japanese and Caucasian populations (36). The frequency of SCA has been assessed in Chinese patients, with 5% for SCA1, 6% for SCA2, 48% for MJD/SCA3, and 0% for SCA6, SCA7, and DRPLA (37). A study of ethnic Chinese in Taiwan also revealed MJD/SCA3 as the most common type (47%), followed by SCA6 (11%), SCA2 (11%), SCA1 (5%), SCA7 (3%), DRPLA (1%), and SCA8 (0%) (38). However, there has been no report on SCA10, SCA12, and SCA17 among Chinese populations. In addition, and with the exception of the MJD/SCA3 and SCA6 genes (38, 39), the distribution of SCA repeats in the Chinese population has not been well documented.
To improve our understanding of the trinucleotide- and pentanucleotide-repeat expansions leading to SCA, we have assessed the repeat size at the SCA1, SCA2, MJD/SCA3, SCA6, SCA8, SCA10, SCA12, SCA17, and DRPLA loci in 198 normal controls and 334 Taiwanese patients with ataxia and PD. In addition, we also compared the frequencies of large normal alleles with the relative frequencies of SCA1, SCA2, MJD/SCA3, SCA6, and DRPLA among Taiwanese, Japanese, and Caucasians.
Materials and methods
Subjects
Seventy patients with ataxia, including 39 patients from 28 families with autosomal dominant cerebellar ataxia and 31 patients with sporadic ataxia and 264 patients with idiopathic PD, were enrolled in this study. Clinical diagnoses of ataxia and PD were made according to the published criteria (40, 41). These 334 patients were recruited from the neurology clinics of Chang Gung Memorial Hospital. In addition, 198 unrelated subjects without neurodegenerative disorders were recruited as normal controls. All examinations were performed after obtaining informed consent from patients and control individuals.
DNA analysis
DNA was extracted from peripheral blood leukocytes using the DNA Extraction Kit (Stratagene La Jolla, CA, USA). Molecular analyses of the CAG, CTG, and ATTCT-repeat loci in SCA1, SCA2, MJD/SCA3, SCA6, SCA8, SCA10, SCA12, SCA17, and DRPLA were performed by polymerase chain reaction (PCR) amplification using published primer sequences (1–4, 6, 7, 9–11). The PCR was carried out with 100 ng of genomic DNA, 0.4 µm of each primer, 200 µm dNTPs, 0.8–1.5 mm MgCl2, 10 mm of Tris, pH 8.3, 50 mm KCl, 0.5 U Taq polymerase, and 10% dimethylsulfoxide, in a final volume of 25 µl. The forward primer was fluorescence labeled in each reaction. PCR products were analyzed in a linear polyacrylamide gel on an automated MegaBACE Analyzer. Allele sizes were determined by comparing migration relative to molecular weight standards. DNA sequencing was performed to accurately assess repeat size and the presence of interruptions.
Statistical analyses
Possible differences between the normal and patient groups in normal repeat frequency distributions were assessed using a non-parametric Mann–Whitney U-test. Allele frequencies at each locus were estimated by the gene count method. Statistical analyses of differences in the frequency of large normal alleles (those corresponding to 5 to approximately 10% of the upper tails) (36) for each locus were performed with the Fisher's exact test.
Results
Trinucleotide- or pentanucleotide-repeat distributions
Figure 1 shows the frequency distributions of (CAG/CTG/ATTCT)-repeat lengths at the nine loci in ataxia and PD patients and controls. In the case of the SCA8 locus, the small normal alleles were closely distributed around an allele with 18 CTGs in each group; the second class of normal alleles comprised CTG-repeat sizes of 22–39 units. A third class of extremely large normal SCA8 alleles was found in one SCA3 patient (65 units) and four PD patients (75–92 units). At the SCA17 locus, alleles varied from 30 to 43 in the control group and from 28 to 46 in the patient group; alleles with 36 repeats represented more than 50% of all normal alleles in each group. Expansions of SCA1, MJD/SCA3, and SCA6 were detected in patients with dominant SCA. When these known SCA1, MJD/SCA3, and SCA6 were excluded, the frequency distributions of normal alleles in the ataxia and PD patients were not significantly different from those in the controls at the nine loci studied (data not shown). When alleles corresponding to 5–10% of the upper tails (large normal alleles) at the nine loci were compared, close associations between the prevalence and frequency of large normal alleles of SCA1, SCA2, MJD/SCA3, SCA6, and DRPLA in Taiwanese, Japanese, and Caucasian families (36) were observed (Table 1).
Distribution of (CAG/CTG/ATTCT)-repeat lengths in patients with ataxia and Parkinson's disease (▪) and controls (□) at the nine loci under study. The expanded alleles in SCA1, MJD/SCA3, SCA6, SCA8, and SCA17 are shown in the enlarged closed bars with number indicated below in parenthesis.
| Prevalence (%) | Frequency (repeat number) | |
|---|---|---|
| Locus | Taiwanese/Japanese/Caucasians | Taiwanese/Japanese/Caucasians |
| SCA1 | 3/3/15 | 0.09/0.09/0.26 (>30) |
| 0.05/0.04/0.16 (>31) | ||
| SCA2 | 0/5/14 | 0.03/0.01/0.12 (>22) |
| 0.01/0.01/0.03 (>23) | ||
| MJD/SCA3 | 46/43/30 | 0.07/0.11/0.04 (>28) |
| 0.06/0.07/0.02 (>29) | ||
| SCA6 | 18/11/5 | 0.17/0.20/0.04 (>13) |
| 0.04/0.08/0.00 (>14) | ||
| DRPLA | 0/20/0 | 0.05/0.10/0.03 (>19) |
| 0.01/0.08/0.01 (>20) |
- Prevalence of SCA and frequency of large normal alleles in Japanese and Caucasians according to Takano et al. (36).
Frequency of SCAs and the genetic and clinical features of the patients
Analysis of the loci involved in the SCAs showed that 26 patients from 16 families (67%) had ataxia due to CAG expansion. MJD/SCA3 was the most common type of dominant SCA found in Taiwanese patients in this report, accounting for 18 cases from 10 families (46%), followed by SCA6 (seven cases from five families, 18%) and SCA1 (one case, 3%). The genes responsible for 13 cases from 12 families (33%) of dominant SCA remain to be determined. The clinical phenotypes of these affected individuals were very heterogeneous. All of them showed clinical symptoms of cerebellar ataxia, with or without other associated features. Dystonia as a presenting sign was noted in the patient with SCA1. Bulging eye sign, which is a characteristic feature of MJD/SCA3, was noted in two patients with MJD/SCA3. For MJD/SCA3, the mean age at onset was 36 years (SD, 12.74) and mean expanded allele size was 71 repeats (SD, 4.99) in 18 patients from 10 families. For SCA6, the mean age at onset and mean expanded allele size in seven patients from five families were 48 years (SD, 10.02) and 23 repeats (SD, 0.79), respectively. For the SCA1 patient, the age at onset was 39 years and the expanded allele was 49 repeats.
SCA8 alleles with 75 to 92 repeats in PD patients
Although no expansion of SCA type 8 was detected in patients with dominant SCA, abnormal expansions were detected in four patients (cases 1–4) with PD (1.5%) (Table 2). Initially, the four patients developed resting tremor in the limbs. No evidence of secondary parkinsonism caused by another neurologic disease, known drugs, or toxins was noticed. Cogwheel rigidity, bradykinesia, and/or postural instability were evident at the time of the study. All four patients showed improved symptoms with levodopa therapy. DNA-sequencing analysis revealed that these potentially pathogenic alleles had 75, 82, 88, and 92 combined repeats with a pure, uninterrupted CTG-repeat tract or interruptions (Table 2).
| Subject | Sex/onset age/present age | Clinical features | Family history | SCA: repeat number | Repeat sequence |
|---|---|---|---|---|---|
| Case 1 | Female/60/73 | Dopamine-responsive PD, motor fluctuations, levodopa-induced dyskinesia | None | SCA8: 88 | CTA8CCACTACTGC TACTGCTACTG74 |
| Case 2 | Female/71/81 | Dopamine-responsive PD, motor fluctuations | None | SCA8: 75 | CTA20CTG2CTCCTG52 |
| Case 3 | Female/67/71 | Dopamine-responsive PD | None | SCA8: 82 | CTA12CTG70 |
| Case 4 | Female/57/58 | Dopamine replacement-responsive PD | None | SCA8: 92 | CTA7CTG2CTACTGCTACTG80 |
| Case 5 | Female/75/77 | Dopamine-responsive PD | None | SCA17: 46 | CAG3CAA3CAG6CAACAG CAACAG29CAACAG |
SCA17 allele with 46 repeats in a PD patient
An abnormal CAG expansion in the SCA17 TBP gene was detected in a PD patient (case five, Table 2). This subject developed resting tremor in the right hand at 75 years of age. Stiffness in her right hand and leg, slowness of movement, and diminished facial expression were noted half a year later. She is currently being treated with benserazide/levodopa (25/100) with a significant benefit. DNA-sequencing analysis revealed that the allele in this patient had 46 repeats.
Discussion
The analysis of trinucleotide-repeat sizes in the SCA patients allowed the identification of an expansion in more than half of them. Our study showed that 33% of the autosomal dominant ataxia did not contain expansion in any of the tested gene, which is similar to the previous reports in which causative genes were unknown for approximately 20–40% of the dominant SCAs (36, 38, 42). The genetic loci for the SCA yet to be elucidated may include those unidentified SCA types, protein kinase C gamma gene, and fibroblast growth factor 14 gene (43). The data from a large normal allele study (36) and the present study suggest that the higher prevalence of SCA1 and SCA2 in Caucasians, MJD/SCA3 and SCA6 in Taiwanese and Japanese, and DRPLA in Japanese are related to the higher frequencies of large normal alleles in these gene loci (Table 1).
Abnormal expansions ranging from 75 to 92 repeats of SCA8 alleles were detected in four patients with PD (Table 2). Large expanded repeats in SCA8 were not confined to patients with cerebellar ataxia. More interestingly, expanded alleles of 103, 113, and 90 repeats in the DNA from patients with PD have been reported (16, 23). Because of the large overlap between repeat sizes found in the controls and the affected individuals in their study, the authors suggested that the CTG expansion may be a non-pathogenic polymorphism tightly linked to an ataxic locus. However, in contrast to their studies, no overlap between allele sizes in the controls and the affected individuals was found in our study, which clearly defines the normal and pathological ranges of CTG repeats. We propose that the expanded CTG repeats may have contributed to the pathogenesis of our PD patients. However, in these patients, there is no significant family history, and therefore we cannot demonstrate cosegregation of expansions with the disease. It may therefore be possible that the expanded SCA8 alleles in our PD patients and in others are incidental to the clinical picture.
The pathogenic role of SCA8 expansion remains uncertain. The SCA8 transcript may act as an antisense regulator of KLHL1 expression in various brain tissues (9, 12). If the CTG expansion leads to an accumulation of the SCA8 transcript, it could prevent the expression of the KLHL1 gene. Alternatively, if it leads to a non-functional SCA8 transcript, it could cause an overexpression of the KLHL1 protein and toxicity to the cerebellar tissue. In addition to this possible antisense effect, the expanded CUG tract in the SCA8 transcript may impair nuclear cytoplasmic transport through the formation of extended hairpin loops, resulting in nuclear retention and affecting the transport of other CAG repeats containing mRNAs (44). Also, the extended hairpin loops may sequester CUG-binding protein, which regulates the alternative splicing of specific pre-mRNAs (45). A deleterious, dominant negative effect on the transcription of another gene close to this locus and/or an abnormal interaction at the RNA level is another possible mechanism for SCA8. We speculate that the repeat expansion of the KLHL1-antisense transcript may have played a role in the pathogenesis of our PD patients and may have functioned at the RNA level, altering the expression of a coding gene that is essential for the functional integration of the substantia nigra-striatal circuit.
An abnormal CAG expansion (46 repeats) in the SCA17 TBP gene was detected in a PD patient (Table 2). Previously, the clinical features of cerebellar ataxia, dementia, and behavioral disturbances were reported in a Belgium patient with an expanded allele of 46 repeats, although unaffected at-risk individuals carried the same expanded alleles within the family (33). In a German kindred, four siblings were affected by cerebellar ataxia, chorea, and dementia, while the unaffected mother and two of the siblings also carried the same expanded 48-repeat alleles (35). In addition to the reduced penetrance and/or variable age of onset, there were patients with 44 and 46 repeats in the TBP gene displaying a Huntington's disease-like phenotype (46). Also, non-cerebellar symptoms, such as dopamine-responsive parkinsonism and early-onset tremor, may develop in a specific SCA type (47, 48). As a phenotypical variability of expanded alleles in the SCA gene exists and TBP is a critical factor in transcriptional initiation and is ubiquitously expressed in all cells, the 46-repeat TBP allele in the PD patient may be linked to the neurological manifestations observed.
In conclusion, we screened for SCA mutations in a large cohort of patients with various inherited and non-inherited neurodegenerative disorders. Abnormal expansions of SCA8 and SCA17 genes were detected in patients with PD. Further evidence from studies using cellular and animal models is needed to understand the association of SCA8 and SCA17 expansions with PD.
Acknowledgements
We thank the members of the SCA families and their physicians for their cooperation. A portion of this work was supported by grant NSC90-2311-B-003-013 from the National Science Council, Executive Yuan, ROC.
