We have studied 24 families with multiple affected members with bipolar dis
order to test the hypothesis that in those families clinically showing gene
tic anticipation [Macedo et al., 1999] we would find large repeat expansion
s. The families meeting inclusion criteria had a minimum of two affected me
mbers over two generations and showed marked anticipation both in terms of
age of onset and disease severity, me used the repeat expansion detection (
RED) method to test patients (n = 24) and controls from these families and
unrelated controls (n = 53). me also genotyped patients and family members
from two families with large expansions at the known expansion loci on chro
mosomes 13, 17, and 18. The RED method revealed a higher number of large ex
pansions in patients compared with controls (t-test; P < 0.0055: Mann-Whitn
ey U; P = 0.02). The patients with the largest expansions were typed at the
specific loci on chromosomes 13, 17, and 18 and the chromosome 18 expansio
n locus segregated with disease in one family, and a second family showed s
egregation with the expansion located at the SCA8 locus on chromosome 13. G
enetic anticipation had been analyzed in this cohort of families, with corr
ection for potential ascertainment bias, possible proband effects, cohort e
ffects, regression to the mean, gender effects, and maternal vs. paternal t
ransmission. None of these potential confounds appeared to account for the
observed anticipation. We also identified that the presence of large expans
ions in affected family members derives primarily from two families from th
e genetically isolated Azores population. One family shows segregation with
the chromosome 18 locus, whereas the other family segregates with expansio
ns at the SGA8 locus. (C) 2000 Wiley-Liss, Inc.