103only by their extended duration but also by their asynchrony resulting in stages with
uneven number of cells as compared to the exponential increase in cell number in
Xenopus and Zebrafish (Kimmel et al. 1995 ; MacQueen and Johnson 1983 ; Masui
and Wang 1998 ; Newport and Kirschner 1982 ). Furthermore, experiments involving
UV-light and X-Ray irradiation have shown that preimplantation mouse embryos are
sensitive to DNA-damaging agents, suggesting that the DNA damage checkpoint in
addition to the SAC is already functional at this developmental stage (Wei et al.
2011 ; Shimura et al. 2002 ; Artus and Cohen-Tannoudji 2008 ). In contrast, Xenopus
and Zebrafish embryos turn on the aforementioned surveillance mechanisms only
during MBT, raising the question of how early embryonic divisions are controlled in
these organisms (Clute and Masui 1997 ; Ikegami et al. 1997 ).
In contrast to the first prolonged cell cycle in mammals, inhibitory phosphoryla-
tions on Cdk1 are barely detectable during the fast cycles of Xenopus embryos (Ferrell
et al. 1991 ; Hartley et al. 1996 ). This could occur because of decreased activity of the
inhibitory kinases Wee1/Myt1 or increased activity of the activating phosphatase
Cdc25. It has been shown that in Xenopus, an embryonic isoform of Cdc25, Cdc25A
is expressed in addition to the Cdc25C isoform during the fast cycles (Kim et al.
1999 ), which might favour the existence of T14/Y15 unphosphorylated, active Cdk1.
Recent work by Ferrell and co-workers (Tsai et al. 2014 ) confirmed the finding that
Cdc25C is present at constant levels throughout the fast cycles and, in addition,
showed that the amount of Cdc25A increases to half maximal by 75 min after fertilisa-
tion. Blocking this increase in Cdc25A increases the length of the second cell cycle by
5 min and leads to a slight increase in Y15 phosphorylation of Cdk1, indicating that
indeed a shift in the Wee1/Cdc25 ratio could account for altered cell cycle lengths.
In the absence of inhibitory Cdk1 phosphorylations, mechanisms controlling
cyclin-B synthesis and destruction are of key importance to ensure oscillating Cdk1/
cyclin-B activity (Tsai et al. 2014 ). Since, in Xenopus, transcription from the zygotic
genome does not commence until MBT, the expression of cyclin-B during the rapid
early embryonic divisions must be regulated at the posttranscriptional level. For
Xenopus, it has been shown that distinct sequences present in the 3′-untranslated
region (UTR) of maternal mRNAs control their translation in a temporal manner
(Mendez and Richter 2001 ). Cyclin-B mRNAs contain one such element called the
cytoplasmic polyadenylation element (CPE) responsible for recruiting the cytoplas-
mic polyadenylation element-binding protein (CPEB), which mediates polyadenyl-
ation and ultimately translation (Groisman et al. 2000 ; Stebbins-Boaz et al. 1996 ;
Stebbins-Boaz et al. 1999 ). CPEB is activated in a cell-cycle-dependent manner by
the kinase Aurora-A (Mendez et al. 2000 ), leading to active CPEB and
polyadenylation- induced translation of cyclin-B before and during M-phase
(Groisman et al. 2002 ). Aurora-A, CPEB and cyclin-B mRNA have been shown to
co-localise with the mitotic spindle and centrosomes in Xenopus embryos, and
expression of a mutant of CPEB defective in spindle localisation impairs embryo
cleavage (Groisman et al. 2000 ). Similarly, injection of CPEB inhibitory antibodies
reduces total levels of cyclin-B and impairs division (Groisman et al. 2000 ), high-
lighting the importance of regulating cyclin-B synthesis to maintain the integrity
of the fast cycles. Exit from M-phase seems to require translational silencing of
3 Regulation of Cell Division