lation events. In yeast and vertebrates, phosphorylation of a threonine, Thr167 in yeast and Thr 161 in
Xenopus, is necessary for activation of MPF [72]. In budding yeast a monomeric kinase, CAK1p, has been
identified as responsible for this activating phosphorylation [35]. In vertebrates, the activating phospho-
rylation of this threonine has been shown to be achieved by a kinase originally called CAK (cyclin-de-
pendent kinase activating kinase), recognized now as Cdk7, which together with its cyclin regulating sub-
unit, cyclin H, phosphorylates conserved threonines in other Cdks [32,35]. Cdks 2, 3, 4, and 6 and Cdk7
itself have this conserved threonine, whereas Cdk5 has a serine (Cdk7 is a serine/threonine kinase) residue
and Cdk8 is missing the motif completely (Table 1).
In yeast, negative regulation of MPF is achieved by phosphorylation of Tyr15 on cdc2 by MIK1and
WEE1gene products [73], whereas in animals phosphorylation of both Tyr15 and Thr14 is required to
maintain MPF in an inactive state [72]. In 1993 a human WEE1 kinase was isolated that phosphorylates
Cdk1 on Tyr15 but not Thr14 [74,75], and in 1997 a Wee1-type kinase, MYT1, was shown to phospho-
rylate Thr14 [76]. Cyclin association with Cdk1 is necessary for phosphorylation of these residues
[73,77]. These residues are in the ATP-binding subdomain element GEGTYGV of Cdk1 and it is assumed
that phosphorylation of these sites interferes with ATP binding [21]. As can be seen in Table 1, Cdks 2,
3, 5, and 8 have both sites conserved, Cdks 4 and 6 have only the Tyr15 site, and neither site is present in
Cdk7. Dephosphorylation of Tyr15 in yeast and Tyr15 and Thr14 in animals is necessary to activate in-
active MPF [14]. A tyrosine phosphatase coded by cdc25 in yeast dephosphorylates Tyr15 [78], while its
homologue in animals dephosphorylates both Tyr15 and Thr14 [15,79,80]. Activation of CDC25 is de-
pendent on phosphorylation by cyclin B/Cdk1 causing a positive feedback loop [81]. Another type of ki-
nase, polo-like kinase, has also been shown to be involved in activation of CDC25. Inactivation of CDC25
and thus maintenance of Tyr phosphorylation can cause G 2 delay in response to DNA damage [59]. In-
activation occurs by phosphorylation of CDC25 on a conserved serine residue by the protein kinases
CHK1 and CHK2 in vertebrates [82,83]. Cyclin A also associates with Cdk1 to promote entry into mito-
sis. It is destroyed earlier in mitosis than cyclin B [84]. Inhibitors of cylin/Cdk complexes have been found
(see later). The inhibitor p21 inhibits cyclin A/cdc2 in early G 2 [85].
Besides regulation by cyclin synthesis and destruction and specific phosphorylation-dephosphoryla-
tion of Cdks, studies have shown that controlling the subcellular localization of Cdk-cyclins is also es-
sential for proper cell cycle coordination [86]. Cyclin A is constitutively nuclear but cyclins B1 and B2
accumulate in the cytoplasm and as cells enter prophase B1 is transported to the nucleus [87]. Cyclin D1,
on the other hand, increases in the nucleus during G 1 but is transported as a Cdk/cyclin D1 complex to the
cytoplasm in S phase [88].
- Exit from M Phase
The activation of MPF induces the cell to divide and also sets the stage for its inactivation by activating
the cyclin degradation system [84]. Destruction of cyclins in M phase inactivates p34 protein kinase and
is required for transition from mitosis to interphase [12–14]. Sudden destruction of cyclins just prior to
anaphase is mediated by the ubiquitin pathway of protein degradation [47,89,90]. In addition to inactiva-
tion of p34 protein kinase, reentry into the interphase requires dephosphorylation of proteins involving
protein phosphatase action. Protein phosphatases that are required in late mitosis have been identified in
yeast (“defective in sister chromatid disjoining”—dis; “bypass of wee suppression”—bws1) and As-
pergillus(“blocked in mitosis”—bimG) [91]. Inactivation of MPF is necessary for the cell to complete
cytokinesis and return to a new interphase but not sufficient to inactivate cyclin degradation [92]. Stud-
ies suggest that G 1 cyclin/Cdk activity is required to inactivate mitotic-cyclin destruction [92,93]. Al-
though it was thought that cyclin degradation was necessary for movement from metaphase to anaphase,
experiments showed that cyclin degradation (MPF inactivation) was not required for sister chromatid sep-
aration but separation was linked to ubiquitin-mediated proteasome degradation [94,95]. A chromosome-
tether protein was proposed as a candidate for the necessary degradation.
B. M Phase Regulatory Proteins in Plants
Cell cycle research in plants at the biochemical and molecular level started relatively recently and is
greatly benefiting from the tools and information obtained with fungal and animal systems. The obvious
first step was to find out which of the known cell cycle regulatory components are conserved in plants.
Research during the last several years has yielded some information indicating that at least some of the
key cell cycle regulatory proteins (e.g., p34 protein kinase and cyclins, mitogen-activated protein kinase)
234 REDDY AND DAY