C. Pangalos et al., UNDERSTANDING THE MECHANISM(S) OF MOSAIC TRISOMY-21 BY USING DNA POLYMORPHISM ANALYSIS, American journal of human genetics, 54(3), 1994, pp. 473-481
In order to investigate the mechanism(s) underlying mosaicism for tris
omy 21, we genotyped 17 families with mosaic trisomy 21 probands, usin
g 28 PCR-detectable DNA polymorphic markers that map in the pericentro
meric region and Long arm of chromosome 21. The percentage of cells wi
th trisomy 21 in the probands' blood lymphocytes was 6%-34%. There wer
e two classes of autoradiographic results: In class I, a ''third allel
e'' of lower intensity was detected in the proband's DNA for at least
two chromosome 21 markers. The interpretation of this result was that
the proband had inherited three chromosomes 21 after meiotic nondisjun
ction (NDA) (trisomy 21 zygote) and subsequently lost one because of m
itotic (somatic) error, the lost chromosome 21 being that with the low
est-intensity polymorphic allele. The parental origin and the meiotic
stage of NDJ could also be determined. In class II, a ''third allele''
was never detected. In these cases, the mosaicism probably occurred e
ither by a postzygotic, mitotic error in a normal zygote that followed
a normal meiosis (class IIA mechanism); by premeiotic, mitotic NDJ yi
elding an aneusomic zygote after meiosis, and subsequent mitotic loss
(class IIB mechanism); or by a meiosis II error with lack of crossover
in the preceding meiosis I, followed by mitotic loss after fertilizat
ion (class IIC mechanism). Among class II mechanisms, the most likely
is mechanism IIA, while IIC is the least likely. There were 10 cases o
f class I and 7 cases of class II results. Within class I, there were
nine cases with maternal meiotic errors (six meiosis I and three meios
is II errors, on the basis of pericentromeric markers) and one with pa
ternal meiosis I error. The postzygotic loss of chromosome 21 was dete
rmined in eight maternal class I cases, and it was maternally derived
in five cases and paternally derived in three; this suggests that the
postzygotic loss of chromosome 21 is probably random. The mean materna
l age in meiotic class I errors was 31.4 years and in mitotic class II
errors was 27.4 years, as expected.