New alleles at microsatellite loci in CEPH families mainly arise from somatic mutations in the lymphoblastoid cell lines

Abstract

In the analysis of 40 CEPH families, under the EUROGEM project, with a total of 29 microsatellites (26 CA‐repeats, a TCTA‐repeat within the vWFII‐3 gene, a TTA‐repeat within the PLA‐2 gene, and an AAAT‐repeat intragenic to the NF1 gene) from human chromosomes 12, 17, and 21, we have detected 21 cases of abnormal segregation of alleles in 16 pedigrees for a total of 14 markers (48%). In 11 cases, the abnormal transmissions were of somatic origin, 10 of which (91%) occurred in the lymphoblastoid cell lines. In 9 other cases, it was not possible to determine if the origin of the new alleles was somatic or germline, and in one case hemizygosity in several family members was observed, so its origin was germline. The 20 new mutations detected in the 22,852 meioses analysed represent a mutation frequency of 8.7 × 10⁻⁴ per locus per allele. The germline mutation rate could be as high as 3.9 × 10⁻⁴ per locus per gamete (from 0 to 3.9 × 10⁻⁴), but the rate of somatic mutations detected in the study was much higher (4.8 × 10⁻⁴ to 8.7 × 10⁻⁴ per locus per allele). Individual mutation rates ranged from 0 to 3.8 × 10⁻³. Among the markers analysed, all three that were tri‐ or tetranucleotide repeats showed one or two new alleles, compared to only 10 of the 26 (38%) CA‐repeats showing mutations. Three CEPH families (102, 45 and 1333) each had several mutational events, and one individual (10210) had somatic mutations for two microsatellites from different chromosomes. The mutation rate at microsatellite loci within families, using DNA directly obtained from cells from the individual, is less than 1 × 10⁻⁴ (true germline mutation rate), which should not affect the use of these markers in diagnosis and linkage. However, these results and previous data suggest that for DNA obtained from cell lines, mutations are much more frequent (1 × 10⁻²−1 × 10⁻³).