87larger carboxy-terminal lobe (De Bondt et al. 1993 ). However, Cdk2 distinguishes
itself from other kinases by two modifications which keep it inactive when cyclin is
not bound. First, the T-loop (activation loop) blocks access of substrates to the active-
site cleft, and second, residues critical for the correct positioning of ATP are dis-
placed. Binding of cyclin-A relieves Cdk2 from its autoinhibitory state yielding
partially active kinase. Full activation of Cdk2/cyclin-A requires phosphorylation of
a threonine residue (threonine-160 in human Cdk2) within the T-loop of Cdk2 by the
Cdk-activating kinase (CAK), also known as Cdk7 (Russo et al. 1996 ). CAK is con-
stitutively active, and therefore T-loop phosphorylation of Cdk2, as well as of Cdk1,
Cdk4 and Cdk6, occurs in an unregulated manner (Tassan et al. 1994 ). In contrast,
inhibitory phosphorylations occurring on threonine-14 (T14) and tyrosine-15 (Y15)
of Cdks are strictly regulated. Phosphorylation of these residues interferes with the
correct positioning of ATP at the active site. In vertebrates, T14 phosphorylation is
mediated by the kinase Myt1 (membrane-associated tyrosine and threonine kinase),
whereas the phosphorylation of Y15 can be catalysed by either Wee1 (Wee1-like
protein kinase, in the Scottish dialect ‘wee’ means small referring to the fact that
Wee1 mutants divide at half the wild-type size) or Myt1 (Mueller et al. 1995 ; Nurse
and Thuriaux 1980 ). Dephosphorylation of both residues is mediated by the dual-
specific phosphatase Cdc25 (Russell and Nurse 1986 ).
3.2.1 The Cdk1 Autoamplification Loop
High Cdk1/cyclin-B activity triggers entry into mitosis. Precise regulation of this
dimeric complex at the levels of protein synthesis, posttranslational modifications
and protein destruction is therefore essential to ensure timely progression through
the cell cycle (Fig. 3.3). Continuous synthesis of cyclin-B results in the accumula-
tion of cyclin-B as cells progress towards M-phase. However, activation of Cdk1/
cyclin-B does not occur in a linear but in a switch-like manner. In interphase, MPF
activity is suppressed by inhibitory T14/Y15 phosphorylation of Cdk1. Critical for
the activation of MPF in an all-or-none fashion at the transition to M-phase is the
autoamplification loop composed of Cdk1/cyclin-B, the inhibitory kinases Wee1/
Myt1 and the activating phosphatase Cdc25 (Fig. 3.4). Xenopus Myt1 seems to be
ubiquitously expressed, whereas Wee1 is present in mature oocytes and during
embryonic divisions and absent in immature oocytes, which instead express Wee2
(Wee1B) (Han et al. 2005 ; Nakajo et al. 2000 ). In mammalian cells, the Cdc25 fam-
ily of dual-specificity phosphatases contains the three members Cdc25A, Cdc25B
and Cdc25C. In mice, Cdc25A is the only one that is essential for early embryonic
development (Lee et al. 2009 ), whereas Cdc25B and Cdc25C are dispensable for
embryonic development (Ferguson et al. 2005 ). Following cyclin-B accumulation,
a small pool of Cdk1/cyclin-B becomes active and phosphorylates Wee1/Myt1
resulting in the inactivation of these inhibitory kinases. At the same time, Cdk1/
cyclin-B phosphorylates and activates its activator Cdc25 (Hoffmann et al. 1993 ).
The positive feedback loop of Cdc25 activation and the double-negative loop of
Wee1/Myt1 inactivation create a bistable system where Cdk1/cyclin-B cannot exist
3 Regulation of Cell Division