The SCA 1 ( Spinocerebellar ataxia type 1 ) and MJD ( Machado-Joseph disease ) CAG repeats in normal individuals : segregation analysis and allele frequencies

Spinocerebellar ataxia type 1 (SCA1) and Machado-Joseph disease (MJD/SCA3) are autosomal dominant neurodegenerative diseases caused by expansions of a CAG trinucleotide repeat in the SCA1 and MJD genes. These expanded sequences are unstable upon transmission, leading to an intergeneration increase in the number of repeats (dynamic mutation). The transmission of the CAG repeat was studied in normal mother-father-child trios, referred for paternity testing (SCA1, n = 367; MJD, n = 879). No segregation distortion was detected. The CAG allele frequencies were determined in 330 unrelated individuals (fathers from couples tested for paternity). The allele frequency distributions did not differ from those previously reported for European populations. The estimated values for the statistic parameters indicating diversity at the SCA1 locus did not differ much from those reported previously for other STRs in the Brazilian population, while those for the MJD locus were close to or higher than the maximum values of previous reports. This shows that SCA1 and MJD are highly informative loci for applications in genetic and population studies and for forensic analysis.


Introduction
Spinocerebellar ataxia type 1 (SCA1) and Machado-Joseph disease (MJD/SCA3) are autosomal dominant neurodegenerative diseases caused by expansions of a CAG repeat in the SCA1 and MJD genes, respectively.Segregation distortion favoring the transmission of mutated or normal alleles during meiosis has been reported for both genes.(Ikeuchi et al., 1996;Riess et al., 1997;Takiyama et al., 1997;Iughetti et al., 1998).Rubinsztein and Leggo (1997) added to these observations, reporting the preferential transmission of alleles with smaller CAG repeats by normal females.More recently, however, Mac Millan et al. (1999) did not find any evidence of segregation distortion upon transmission of CAG repeat alleles by normal individuals.
In spite of being highly polymorphic, triplet repeats in the normal range are consistently transmitted unchanged from parents to children (Limprasert et al., 1994;Richards and Sutherland, 1994), and the repeat size variation in the DRPLA (dentatorubral-pallidoluysian atrophy) gene has recently been shown to be reliable for use in forensic tests, especially paternity tests (Pelotti et al., 1998).
There are no frequency estimates of the SCA1 and MJD alleles in the Brazilian population, and segregation analysis has produced conflicting results (Rubinsztein and Leggo, 1997;MacMillan et al., 1999).We investigated allele segregation in mother-child and father-child pairs, and estimated the frequencies of SCA1 and MJD alleles in a sample of clinically normal individuals from the northeastern region of the State of São Paulo, Brazil.We also evaluated their applications in paternity investigations, by comparisons with other markers.The absence of segrega-tion distortion herein reported corroborates the data presented by MacMillan et al. (1999).

Material and Methods
The study sample comprised individuals seeking paternity investigation at the Clínica Civil of the Ribeirão Preto University Hospital, University of São Paulo, during the years 1996 and 1997.All individuals were white (skin color was determined visually), originating from Ribeirão Preto and nearby cities.They formed mother-child-alleged father trios, routinely submitted to the investigation of a set of six or more STRs.
For calculating the SCA1 and MJD allele frequencies, we considered mothers and alleged fathers of 165 trios (Sample I) consecutively submitted to the same set of tests for paternity determination, totaling 330 genetically unrelated individuals, or 660 chromosomes.For estimating segregation distortion, 118 trios with confirmed paternity were selected from this sample.In addition, other 249 trios were examined for the SCA1 locus, and 761 trios for the MJD locus.The last two groups (Sample II) were neither consecutively collected, nor submitted to the same set of paternity tests, but the individuals were ethnically similar to those in Sample I, and paternity was confirmed with a probability of 99,99%.The difference in sample size between the two groups reflects nothing but the longer time of use of the MJD locus in the laboratorial routine.
Genomic DNA was extracted from 300 µL of whole blood, as described by Higuchi (1989).The fragments containing the CAG repeats of the SCA1 and MJD genes were amplified by PCR, using the primers and conditions described by Orr et al. (1993) and Kawaguchi et al. (1994), respectively.The amplified products were electrophoresed on a 12% denaturing polyacrylamide gel, followed by silver nitrate staining.The PCR products of about 150 sam-ples were initially genotyped after electrophoresis; the procedure was repeated with samples of apparently equal mobility placed side by side, thus allowing the visualization of an allele ladder and the selection of samples for sequencing, in order to determine the exact number of CAG repeats.These sequenced samples were used as reference.
Exact tests were performed using the GENEPOP program (Raymond and Rousset, 1995).For estimating segregation distortion, we prepared a program that allows to determine the inherited allele by direct genotype analysis of the trios, and records which one of the alleles, the larger or the smaller, was transmitted.The proportions were compared using a χ 2 test, considering the expected ratio of 50%.
This study was approved by the Research Ethics Committee of the Ribeirão Preto University Hospital, Ribeirão Preto School of Medicine, University of São Paulo (HCRP nr.5158/98).

Results
The alleles were designated according to the number of CAG repeats.Nineteen different SCA1 alleles were identified, with 19 to 39 repeats.At the MJD locus, 21 alleles were found, with 14 to 40 repeats.SCA1 alleles 20, 24, 38 and 39, as well as the MJD allele 40, were observed only in Sample II, and were not considered in the allele frequency analysis.In the segregation distortion analysis, 367 trios were analyzed for the SCA1 locus, including 734 meioses (Table 1), and 879 trios for the MJD locus, totaling 1758 meioses (Table 2).No segregation distortion for the SCA1 alleles was revealed by the analysis of informative meioses (p = 0.8516 and p = 0.3604, for 257 maternal and 269 paternal meioses, respectively; Table 1) no distortion was observed for the MJD locus either (Table 2), analyzed in 745 maternal and 695 paternal informative meioses (p = 0.6339 and p = 0.2713, respectively).116 Wiezel et al.The allele frequencies for the SCA1 and MJD loci were determined in 330 unrelated individuals (660 chromosomes) of Sample I: 44 genotypes for SCA1 and 75 for MJD were detected; 76 individuals were homozygous at the SCA1 locus and 50, at the MJD locus.
The allele frequencies found were similar to those reported in the literature for European populations (Figures 1  and 2).At both loci, the most frequent alleles (29 and 30 for SCA1 and 14 and 23 for MJD) were less frequent in African populations.
The parameters of forensic interest were calculated for both loci: Power of Discrimination ( PD

Discussion
This study comprises the largest sample of individuals analyzed for SCA1 and MJD loci reported in the literature, corresponding, respectively, to 24 and 28% of the total.It revealed the largest number of different alleles in a single sample hereto described, though no new allele was outside the range already reported in the literature.
We did not observe preferential transmission of normal SCA1 and MJD alleles: 526 informative meioses (257 maternal and 269 paternal) for the SCA1 locus and 1,440 informative meioses (745 maternal and 695 paternal) for the MJD locus.These results are in accordance with those of Mac Millan et al. (1999), who did not find evidence of segregation distortion in MJD, SCA1 and DRPLA loci in their study of 377 pairs of twins and their normal parents.However, the preferential transmission of the smaller alleles of the MJD locus has been reported by Rubinsztein and Leggo (1997) in normal women whose smaller allele was transmitted in 166 out of the 290 meioses analyzed, while the men transmitted the smaller allele in 126 out of the 269 meioses.
In the present study, the distributions of the allele frequencies at both loci showed a great similarity with the distributions reported in the literature for European populations (Figures 1 and 2).The most prevalent alleles (29 and 30 in the SCA1 locus, and 14 and 23 in the MJD locus, respectively) are less frequent in African populations, which gives them a greater heterozygozity.For the MJD locus, we observed the three modes present in European populations, which correspond to alleles 14, 23, and 27, while, in African populations, up to five modes occur, corresponding to alleles 14, 22, 28, 30, and 33.For the SCA1 locus, our sample showed a mode around allele 30, similar to European populations, while two modes occur in the African populations, corresponding to alleles 26 and 30.The greater heterozygozity in African populations and the profile of frequencies with multiple modes may be either ca-sual or the result of peculiarities of the places of origin of the few samples hereto analyzed.
Since our study is restricted to a sample from a small region of the State of São Paulo, investigations of other series of individuals from different geographical areas of the country are desirable.
For genes with trinucleotide repeats that expand as dynamic mutations, the alleles in the normal range are easily amplified by PCR, but there is a possibility that larger alleles are not amplified, as is the case in myotonic dystrophy (Gennarelli et al., 1998), resulting in an apparently homozygous genotype.Nevertheless, the CAG expanded repeat has been easily diagnosed by PCR (Brice, 1998).The amplification products corresponding to the normal and expanded alleles can be visualized, and the number of repeats accurately determined after denaturing polyacrylamide gel electrophoresis and silver staining (Vuillaume et al., 1998;Maruyama et al., 1996).We did not detect expanded or intermediate alleles in this study, but, considering the sensitivity of the method used to measure the alleles, the frequency of homozygotes ( 23  tion) for the SCA1 locus did not differ much from the values reported previously for other loci in the Brazilian population (Pagotto et al., 1999;Leboute, 2000), while those for the MJD locus were close to or higher than the maximum values reported.This demonstrates that these loci are highly informative markers for general applications in genetics and population studies, as well as for paternity tests in forensic investigations, much in the same way as the DRLPA locus (Pelotti et al., 1998).

Table 1 -
Transmission of larger and smaller SCA1 alleles by normal individuals.

Table 2 -
Transmission of larger and smaller MJD alleles by normal individuals.