Molecular Analysis of CTG/CTA Repeats at SCA8 Locus in South Indian Population

Spinocerebellar ataxias are a group of phenotypically and genetically heterogenous disorders characterized by progressive degeneration of the cerebellum with overlapping symptoms. A novel form of SCA has been described with triplet repeat expansions in the 3 UTR of the SCA8 gene and is caused by expansion of a CTG/CTA repeat in the ataxin-8 opposite strand gene (ATXN8OS) located on chromosome 13q21. Analysis of CTA/CTG repeats in SCA8 gene was performed in 188 ataxia patients and 100 healthy volunteers without any neurological signs or family history. The repeat length was found to be highly polymorphic. We were unable to find any individual with pathogenic repeat length in SCA8 gene, when we used the already established pathogenic repeat criteria. However, repeats >35 (4.7%) were exclusively found in patients only and none of the controls suggesting that these repeat sizes could be pathogenic for our population. The frequency of LN alleles was also found to be higher than reported for other populations. The percentage of LN alleles at SCA 8 locus was 65% and 47% in patients and controls. In the present study three patients also exhibited repeats lower than the normal range and the pathological implications of these needs to be explored.


Introduction
Spinocerebellar ataxias (SCAs) are a heterogenous group of disorders characterized by progressive cerebellar ataxia of gait and limbs, dysarthria, dysphagia, and other neurological signs [1]. The genetic classification of the autosomal dominant types of SCAs is associated with 20 loci. Several of these SCAs (SCA1, 2, 3, 6, 7, 17) are due to CAG repeat expansions in the coding regions of the corresponding genes, translated into abnormally long polyglutamine stretches [2,3]. Apart from these a novel type of mutation in SCAs was described which consisted of trinucleotide repeats expansions in the non-coding regions of certain genes, these include the CAG expansion in SCA12 (OMIM 604326) and CTA/CTG in SCA8 (OMIM 603680) [4][5][6].
The combined CTA/CTG repeat expansion in 3 UTR of the SCA8 gene present on 13q21 (a gene of unknown function) is controversial with regard to its pathologic significance. The defined normal range for SCA8 gene is 15-50 CTG/CTA repeats and pathogenic range is 80-250 CTG/CTA repeats [7][8][9]. Evidence from several populations has suggested that the disease prevalence in a population may be associated with the presence of large normal alleles at the respective loci [8]. The CAG/GAA/CTG/CAA/ATTCT repeats at different loci are highly polymorphic in normal individuals. Once the repeat number crosses a particular threshold, specific to an individual locus, instability is observed leading to disease manifestation. Normal and disease ranges for triplet repeat disorders are reported to vary considerably between populations. In different populations, the disease prevalence of these ataxias has been shown to be associated with the presence of large normal alleles (LNAs) at the respective loci [10,11]. Screening of populations for establishing normal and expanded ranges for that particular geographical region will help in proper molecular diagnosis So, we elucidated the normal, pathological and LN allele frequencies of triplet repeats in SCA8 gene, in both ataxia patients and healthy controls from the south Indian population.

Sample collection and size
The present study was carried on 288 cases which included 186 males and 102 females in the age group of 5-76 years with the mean age of 35.6 15.9 years. A three generation pedigree was recorded with relevant medical history of the family members. Individuals include one hundred eighty eight (n=188) SCA patients (117 males and 71 females) and one hundred (n=100) controls with no symptoms or family history of any neurological disease or ataxia. These patients were clinically diagnosed by a neurologist and detailed information was provided by the referring physician along with radiological and ophthalmological results. Autosomal dominant family history was observed in 28.1% (53/188) and sporadic in 71.8% (135/188) cases. All the cases were ataxic, 29% of patients had slurred speech and deteriorated writing. 22% had optical atrophy with tremor while as speech disturbance with slow eye movements was found in 18% of patients. The remaining cases had imbalance walk with dysarthria or/and dysphagia.

DNA extraction
DNA was obtained from peripheral blood by salting out method using ammonium acetate and ethanol to precipitate the DNA [12].

Polymerase chain reaction
The molecular identification of the genes associated with these neurodegenerative diseases is based on the observation that mutational event consists of the expansion of trinucleotide repeats, hence rapid and specific tests based on the polymerase chain reaction was developed for diagnosis. The application of this technique allows the differentiation of normal and mutated alleles by determining the number of repeats which is adequate for management, as well as, genetic counseling. PCR reactions were performed with specific primers flanking CTA/CTG regions of SCA8 genes, respectively. Primer sequences for these genes were taken from the literature [13]. A three step PCR was carried out by the method published by our group [14]. Briefly 35 cycles with an initial denaturation at 95°C for 5 min followed by cycling at 95°C for 30s, annealing for 40 s at the appropriate temperature, extension at 72°C for 1 min and a final extension at 72°C for 5 min. The amplified products were initially electrophoresed on a 2% agarose gel and to establish the repeat numbers, PCR products were subsequently separated in denaturing 10% polyacrylamide gels . Molecular weight analysis and sizing of alleles was performed by using UVI Tech software (UK).

Results
Two factors make clinical diagnosis difficult for spinocerebellar ataxia. (i) Within a genetic subtype, the clinical signs are highly variable; (ii) there is a striking phenotypic overlap between each type of SCA. As a result, the diagnosis cannot rely on sole clinical evaluation and a molecular diagnosis is required. The SCA8 PCR products were assessed ( Figure 1) and none of the cases had SCA 8 expansion in based on the defined pathological range which is 81-250 CTG/CTA repeats.   The pathogenic (CTG)n repeat in SCA8 gene is adjacent to a nonpathogenic but highly polymorphic, stably transmitted (CTA)n repeat, which makes it technically difficult to determine the precise number of pathogenic (CTG)n repeats. Therefore, the current reference ranges are based on the total number of both the (CTA)n and (CTG)n repeats. Hence sequencing was carried out in random cases and was found that CTA repeats were usually more in patients than controls studied ( Table 1). The pathogenic criteria of SCA 8 is > 80 CTG repeat and in the present study none of the patients falls in this criteria , however, 9 patients with >35 CTG repeats exhibited symptoms of SCA8 type (Table 2) indicating that this may be the pathogenic repeat number for the South Indian patients.
The prevalence of SCAs in various populations has been correlated with the percentage of large normal alleles. To study the frequency of triplet repeat expansion in our two groups relative to the frequency of normal alleles of larger size at the SCA 8 locus, we used the criteria of Takano et al. [8]. In the present study LNA allele distribution were assessed from the healthy individuals (n=100) and from non-disease alleles (n=188). The frequency of large normal alleles >28 repeats at SCA 8 gene in the present study was 0.58. The percentage of large normal at SCA 8 locus was 65% and 47% in patient group and control group respectively (Figure 3). There is no published literature on the frequency of LN alleles at SCA 8 locus from India. The frequency of large normal alleles in the present study is significantly higher than the frequencies reported in Portuguese and Brazilian populations (Table  3). In the present study lower repeats at SCA 8 locus i.e. 16-18 repeats at SCA 8 gene were identified only in the patients ( Table 4). The pathological implications of lower repeats than the normal range needs to be explored.

Discussion
Earlier our group had established the methods of molecular diagnosis, prevalence and role of LN alleles for a panel of ataxias, which includes SCA 1, 2, 3, 6 and 7 in 124 patients with clinical diagnosis of SCA. However only 8% of the cases could be diagnosed as per the established pathogenic range at 5 SCA loci [14]. Hence, establishing non pathological ranges/Pathological range of the trinucleotide repeats in SCA 8 gene characteristics of the south Indian ethnical background, allowed us to introduce in our patients a new test with SCA.
The combined CTA/CTG repeat in the 3 UTR of ATXN8OS gene, present on chromosome locus 13q21, is characterized by a slowly progressive ataxia with disease onset typically occurring in adulthood. The progression is typically over decades regardless of the age of onset. Common initial symptoms are scanning dysarthria with a characteristic drawn-out slowness of speech and gait instability. Some individuals present with nystagmus, dysmetric saccades and, rarely, ophthalmoplegia. Tendon reflex hyperreflexity and extensor plantar responses are present in some severely affected individuals [15]. In the present study >35 repeats were found only in patients only with clinical symptoms of SCA 8 type (Table 2). Hence there is a need to reevaluate the threshold of abnormal repeat expansion and it can be lowered to 35 repeats.
Its pathogenesis has been distinguished among triplet disease disorders both in terms of the transcribed repeat motif (CTG) and for its location in the noncoding region of the gene, resulting in an untranslated expansion at the 3 [5,9,16]. However, in the present study CTG repeats were found less in patients as compared to controls ( Table 1). The existence of an antisense transcript, encoding a novel actin-binding protein (KLHL1), has recently been reported [17,18] suggesting that the pathogenic effect of SCA8 expansion may result in an alteration of KLHL1 messenger RNA stability or processing. However, the role of CTG expansion in the pathogenesis of SCA8 and the molecular mechanism responsible for the disease remain to be clarified [19].
SCA8 form of ataxia accounts for 2-5% of ADAs. SCA8 expansion was found in 4.7% (8/167) ataxia patients in Italy [13]. It is infrequent in Japan and its prevalence is 2.7% in Taiwanese [20,21]. None of the cases in the present study showed expansion at the SCA 8 locus as per the defined pathogenic range of 80-250 CTG/CTA repeats. However, 9 patients with >35 CTG repeats exhibited symptoms of SCA 8 gene (Table 2). This implicates a more sophisticated interpretation of SCA8 gene and raise the question about the diagnostic thresh hold between normal and expanded repeats in our population. The values which are indicated in parentheses are P-values which were obtained from chi-square analysis when comparison was between the present study and the other studies using MedCalc 7.6.0.0.  At SCA8 locus repeats varied from 18-33 with 90% of the alleles having less than 27 repeats in northern Indian population [22] and in the present study repeats varied from 16-37 ( Figure 2) with 44% of the alleles having >27 repeats which indicates that greater repeat size at SCA 8 locus is more in southern population of India. The frequencies of large normal alleles at SCA 8 locus were quite high when compared with other populations (Table 3). LN alleles are relatively unstable and undergo expansion to reach an intermediate range from which further expansion to the disease range take place. Therefore a study on the percentage of large normal alleles in any population would be an indirect reflection of the prevalence of the disease in that population. This was found to be true in African and Israeli populations [23,24]. A similar study involving White and Japanese ADCA patients and normal individuals observed that the prevalence of SCAs is highly correlated with the frequency of large normal alleles in the normal population [8].
The role of triplet repeat expansion beyond thresh hold size in disease pathogenesis has been defined, however the exact role of lower repeats in disease pathogenesis is still unknown in these type of disorders. Notably, the lower repeats may also result in abnormal phenotype, as has been noted in diseases such as SBMA. In the present study lower repeats were not found in the controls which indicate that lower repeats at these loci may play a role in diseases pathogenesis.

Conclusion
SCA 8 gene constitutes a high degree of genetic heterogeneity in our population. It is proposed that pathogenic repeats can be redefined (>35 repeats) based on case/control data of this study at SCA 8 locus, using this criteria 4.7% (n=9) patients can be diagnosed for SCA 8 type. The redefined pathogenic repeats were seen in patients only with clinical characteristics of SCA 8 type. Hence, it is proposed that repeats pathogenic for the South Indian population should be >35 CTG repeats. The higher frequency of LNAs at SCA 8 gene also indicates that it could be more prevalent in our population. Protein misfolding due to expansion of polyglutamine tracts may not be the only mechanism for neurodegenerative diseases, repeats lower than normal range may also result in abnormal phenotype. The repeat range and LN allele frequency of the present study can be of significant help in proper molecular diagnosis at SCA 8 locus.