The Endoreduplication Cell Cycle: Regulation and Function 85
sperm, CDKA-DN out-competed the endogenous CDKA for cyclin binding,
thereby effectively reducing S-phase kinase activity and inhibiting the DNA
replication process. As a result, there was a 50 % reduction in endoredu-
plication during transgenic endosperm development, with virtually no cells
exceeding three cycles of endoreduplication and a maximum ploidy of 24C at
18 DAP. Nuclear size in CDKA-DN endosperm was clearly reduced compared
to control, nontransgenic nuclei, especially within the larger cells in the cen-
tral area of the starchy endosperm, but cell size was virtually unaffected. Only
a minor decrease was measured in the transcription of genes encoding starch
biosynthetic enzymes and zein storage proteins in CDKA-DN endosperms.
An alternative approach involving overexpressing CDKA was inconsequen-
tial, suggesting that this kinase, thought important for endoreduplication, is
not rate-limiting for the process. Investigation of the role of CDKA in maize
endosperm development suggested that endoreduplication and nuclear size
can be uncoupled from cell size and gene expression, thus challenging the
long-standing hypothesis that endoreduplication drives cell growth and sup-
ports higher levels of gene expression.
Several A-, B-, and D-type cyclins have been characterized and found to
be expressed in maize endosperm (Sun et al. 1999b; Dante 2005). CycA1;3-
associated kinase was most active during the mitotic stage and then declined
as the endosperm progressed into the endoreduplication stage, suggesting
that CycA1;3 is more likely involved in the mitotic than the endoreduplication
cell cycle. However, CycB1;3, D5;1, and D2;1 showed a peak in kinase activ-
ity at 11 DAP, a stage in which the endosperm is active in both the mitotic
and endoreduplication cell cycles. The sustained (although reduced) kinase
activity associated with these three cyclins throughout endosperm develop-
ment suggested that they may be involved in the regulation of both types of
cell cycle. This is not surprising for CycD5;1 and D2;1, as CycDs are known
to be involved in S-phase entry. However, the finding that CycB1;3-associated
kinase was active after mitosis ceased was unexpected and suggested a novel
role for CycB1;3 in endoreduplication. Although an important role for CDKA
in regulating endoreduplication in the endosperm has been established, the
cyclin partner has not been identified and remains a challenge for future ex-
periments.
Among the factors known to inhibit CDK activity, a maize Wee1 (Sun
et al. 1999a) and two CKIs, KRP;1 and KRP;2 (Coelho et al. 2005) horthologs,
have been investigated in developing maize endosperm. Ectopic expression of
maize Wee1 in fission yeast inhibited cell division and resulted in elongated
cells, consistent with an inhibitory role in CDK activity. Although specific
data are lacking, the peak in Wee1 expression during endoreduplication sug-
gests that it might contribute to inhibit cell division in endoreduplicating
cells. BothKRP;1andKRP;2are expressed in the endosperm and inhibit
in vitro CycA1;3- and D5;1-associated kinase activities, both of which are
considered to be involved in S-phase. However, KRP;1 protein appeared con-