Applying RNA fluorescence in situ hybridization to parthenogenetic embryos
with two maternally derived X (X-M) chromosomes and embryos with X chromoso
me aneuploidy such as X(P)0 (X-P, paternally derived X chromosome), (XXXP)-
X-M-X-M and (XXY)-X-M-Y-M. We studied the control of Xist/Tsix expression f
or silencing the entire X chromosome in mice. The data show that the patern
ally derived Xist allele is highly expressed in every cell of the embryo fr
om the 4-cell stage onward, irrespective of the number of X chromosomes in
a diploid cell. The high level of Xist transcription is maintained in non-e
piblast cells culminating in X-P-inactivation, whereas in X(P)0 embryos it
is terminated by the blastocyst stage, probably as a result of counting the
number of X chromosomes in a cell occurring at the morula/ blastocyst stag
e. Xist is also down-regulated in epiblast cells of (XXP)-X-M and (XXXP)-X-
M-X-M embryos to make X-inactivation random. In epiblast cells, Xist seems
to be up-regulated after counting and random choice of the future inactive
X chromosome(s). Although the maternal Xist allele is never activated in fe
rtilized embryos before implantation, some parthenogenetic embryos show Xis
t up-regulation in a proportion of cells. These and other data reported ear
lier suggest that imprinted X-inactivation in non-epiblast tissues of roden
ts had been derived from the random X-inactivation system.