76676 Med Genet 1992; 29: 766-769 ORIGINAL ARTICLES Comparison of the myotonic dystrophy associated CTG repeat in European and Japanese populations June Davies, Hidehisa Yamagata, Peggy Shelbourne, Jessica Buxton, Toshio Ogihara, Pekka Nokelainen, Masanori Nakagawa, Robert Williamson, Keith Johnson, Tetsuro Miki Abstract repeat at the 3' end of a gene encoding a Gene amplification using polymerase member of the protein kinase family." chain reaction (PCR) was carried out on The repeat copy number has been found to DNA samples from a total of 92 normal be highly variable in the normal population, subjects and 52 subjects with myotonic with a reported size range between 5 and 30 dystrophy (DM) from European and copies."8 DM patients are found to have one Japanese populations, to determine the allele in the normal size range and a larger DM copy number of the CTG repeat associ- specific allele; mildly affected or presympto- Department of ated with DM for each group. In the two matic subjects have at least 50 copies of the Anatomy, Charing populations, the number of repeats on repeat and severely affected patients have Cross and normal chromosomes only were com- alleles containing up to several thousand Westminster Medical School, St Dunstan's pared, as CTG copy number on DM chro- copies of CTG. The extent of the expansion of Road, London W6 mosomes was difficult to determine by the CTG repeat correlates well with age of 8RP. PCR alone. In this study, normal chro- onset and increased severity of symptoms J Davies P Shelbourne mosomes were found which had as many within families: the larger the number of J Buxton* as 35 copies of the repeat, which is larger CTGs, the more severe the effects of the K Johnson than the normal range reported pre- disease, but at least 50 copies are necessary to Department of viously but still does not overlap with the produce detectable clinical symptoms.68 The Geriatric Medicine, repeat number associated with DM path- number of repeated sequences can change in Osaka University ology, which is at least 50 copies. Using succeeding generations, and an increase in Medical School, Japan. data from normal chromosomes from repeat number indicates why children born to H Yamagata unrelated subjects, the frequencies of minimally affected parents can show more T Ogihara five, 11, and 13 copies of the CTG repeat severe symptoms. This explains the phenome- T Miki were found to be significantly different non of genetic anticipation seen in myotonic 10 Laboratory of Medical between the two populations, with five dystrophy families.9 Genetics, National and 11 copies more commonly seen in the Polymerase chain reaction (PCR) using syn- Public Health European population and 13 copies in the thetic oligonucleotides flanking the CTG re- Institute, Helsinki, Finland. Japanese population. This difference may peat allows the specific amplification of the P Nokelainen be the result of natural divergence of the region containing the repeat. The copy normal chromosomes between the popu- number of the repeat is very variable, and has Third Department of Internal Medicine, lation groups. been shown to decrease as well as increase." It Faculty of Medicine, (J Med Genet 1992;29:766-9) is important for greater accuracy in diagnosis Kagoshima to establish the normal range of alleles. This University, Japan. M Nakagawa study directly compares the number of copies Myotonic dystrophy (DM) is the most com- of CTG on non-DM chromosomes in Euro- Department of mon form of adult muscular dystrophy, with pean and Japanese populations. Biochemistry and Molecular Genetics, an average incidence of 1 in 8000 in the Euro- St Mary's Hospital pean population and a lower incidence of 1 in Medical School, 20 000 in Japan.'2 DM is characterised by Methods Imperial College, London. myotonia with muscle wasting and a wide DNA was prepared from peripheral blood or R Williamson range of accompanying symptoms including immortalised lymphoblastoid cell lines.'2 A cataracts, intellectual impairment, and con- total of 144 unrelated subjects was studied, 37 *Present address: duction defects in the heart.' The age of onset normal and 32 DM subjects from Japan and 55 Mothercare Unit of and severity of the disease show extreme varia- normal and 20 DM subjects from Europe Paediatrics, Institute of Child Health, London tion both between and within families. including British, German, Belgian, Swedish, WC1N 1EH. The genetic defect underlying DM was and Finnish families. The largest subgroup in Correspondence to unknown until recently, when it was reported the European population originated from Fin- Dr Johnson. that a fragment of DNA is vastly expanded in land with 16 normal and 11 DM subjects. Received 17 June 1992. Revised version accepted DM patients." This expansion is the result of Statistical values were calculated using the X2 13 July 1992. an increased number of a trinucleotide CTG test in 2 x 2 tables with Yates's correction. The Comparison of the myotonic dystrophy associated CTG repeat in European and Japanese populations 767 50 formamide dye (USB Sequenase version 2-0 sequencing kit, Cambridge Bioscience) and * European run on a 6% denaturing polyacrylamide gel. The 40 Japanese polyacrylamide gel was dried and exposed to Hyperfilm-MP (Amersham) for one to two days. 0j) 01) Results 0 0) Heterozygote frequencies in the normal popu- .0 lation studied were 73% (European) and 81% E (Japanese). The CTG repeat is very informa- z tive and has a PIC value of 0-83 based on the frequency of 19 alleles seen in this study. The CTG repeat alleles showed mendelian inherit- ance through families, and in 146 meioses studied no new mutations were seen on normal 5 8 10 11 12 13 14 15 16 17 20 21 23 24 25 27 32 33 35 chromosomes but the mutation rate on the DM chromosomes was 100%. Copies of CTG The frequencies of the CTG repeat of nor- Figure 1 Frequency distribution of the DM associated CTG repeat in European and mal alleles from the European and Japanese Japanese populations. The European population contains families from Britain, Belgium, Germany, Sweden, and Finland, with the Finnish being the largest subgroup. populations are shown in fig 1. The distribu- No statistical difference in allele frequency was found between the Finnish subgroup tion of alleles is indicated, with a peak repeat and the rest of the European population. number of five, and then a second peak between 11 and 14. Very few repeats were found between five and 11; only a single ex- polymorphic content of the CTG repeat was ample of an eight copy allele (European) and calculated using PIC = 1 -ni=1 Pi2, where Pi is four of 10 copies (three European and one the frequency of allele i.13 Japanese) were found. PCR was carried out using 20 ng of DNA, The modal number of copies is five in the 50 pmol of each primer 101 and 102,6 European population but 13 in the Japanese. 200 gmol/l each of dATP, dTTP, dCTP, and The results of comparing the most common dGTP (Pharmacia), 1 unit of Tth polymerase alleles from unrelated subjects from the Euro- enzyme (Hybaid) in a final volume of 25 gl pean and Japanese populations, using the x2 overlaid with two drops of mineral oil (Sigma). test in 2 x 2 tables with Yates's correction, are Radiolabel was added to the PCR samples in shown in the table. Linkage disequilibrium two ways: (1) end labelling of primer 101 using between the CTG repeat size and flanking [_y-32P]ATP (Amersham); 50 pmol of primer markers on normal chromosomes was looked 101 was incubated at room temperature for 30 for: the sample size for the European popula- minutes with 2 units of T4 polynucleotide tion was too small to determine if disequili- kinase in a 20 gl reaction containing 2 gl of brium was present, but preliminary data from 10 x kinase buffer (final concentration 50 mmol/ the Japanese population indicate that disequi- 1 Tris-HCl pH 7-5, 10 mmol/l magnesium librium exists between 12 and 13 copies of chloride, 5 mmol/I dithiothreitol, 01 mmol/l CTG and the probe p37. 1. No disequilibrium spermidine), and 1 jl ['y-32P]ATP. 0 5 pmol of was found with D19S63. radiolabelled primer 101 was added to each In looking at the normal range of CTG PCR sample. (2) Direct incorporation of copies (including the non-DM chromosome [a- 32P]dCTP (Amersham) by addition of 20 to from affected subjects) four different persons 30 nCi [a-32P] dCTP to each PCR sample. were seen with more than 30 copies of CTG, PCR running conditions were as follows: 32 and 35 copies on two Japanese chromo- initial denaturation at 94°C for five minutes, somes and 33 and 35 on two European chro- then 35 cycles of 94°C for 1-5 minutes, 62°C mosomes (fig 2). In pedigree c in fig 2A, a 35 for one minute, and 72°C for two minutes. copy allele is inherited in a mendelian fashion Five microlitres of the PCR product was first by an affected child (lane 7) on the non-DM checked on a 2% agarose gel (containing equal chromosome as a normal allele from the unaf- amounts of Nusieve and regular agarose), run fected parent (lane 9). In pedigree b in fig 2A, for one hour in 1 x TBE buffer. Then 3 p1 of the reverse occurs where 33 copies of CTG are the PCR product was denatured with 2 gl inherited as a normal allele by an unaffected child from the affected parent. These results give a normal range of five to 35 copies for the x values for the most common CTG alleles found in the European and Japanese CTG repeat in these populations.