Transcriptional Control of the Plant Cell Cycle 21
When interpreting experimental data from cell cycle time courses, it is
worthwhile considering potential pitfalls inherent with existing synchroniza-
tion methods: Inhibitor-based synchronization methods are susceptible to
non-specific or unexpected side effects. For example, the commonly used fun-
gal toxin Aphidicolin, which inhibits replicative DNA polymerases, will elicit
some of the cellular responses controlled by the mechanisms that monitor
the completion of DNA replication.A good example is the observation of
S-phase induction of cyclin B1;1 (but not of other cyclins of the cyclin B1
clade) expression in Aphidicolin-treated Arabidopsis cells, but not in cells
synchronized by withdrawal and provision of sucrose (Menges et al. 2005).
Cyclin B1;1 expression has been shown to be specifically induced by DNA
damage and incomplete replication check points (Culligan et al. 2006). Other
inhibitors, including the mitotic spindle inhibitors (propyzamide, colchicine,
amiprophos methyl, oryzalin), are likely to elicit similar artifacts, but have
not yet been analyzed in detail in this respect. Other methods for synchro-
nization, involving the withdrawal and re-provision of nutrients (Sano et al.
1999; Menges et al. 2003) or hormones (Planchais et al. 1997) might cause
fewer such artifacts. Therefore, data genuinely reflecting cell cycle regulated
gene expression is likely to emerge only from the consensus of experiments
performed with different synchronization protocols.
2.1
Analysis of Cell Cycle Regulated Gene Expression in Tobacco BY2 Cells
The tobacco BY2 cell system was the first to offer high levels of synchrony
(Nagata et al. 1992). Synchronization in these cells can be achieved by sev-
eral treatments: Aphidicolin treatments, which result in an early S-phase block;
treatments with microtubule antagonists, leading to an arrest at the metaphase
to anaphase transition; and withdrawal and re-provision of nutrients (Sano
et al. 1999; Menges et al. 2003) or hormones (Planchais et al. 1997).
In a landmark experiment, Breyne and colleagues arrested BY2 cells with
Aphidicolin and followed changes in cell cycle gene expression at hourly in-
tervals (Breyne et al. 2002). The analysis was performed by cDNA AFLP and
the abundance of∼1300 transcripts was found to significantly vary over time
in the synchronized cells. To unequivocally identify transcripts observed to
fluctuate during the cell cycle, the transcript tags were re-amplified and se-
quenced. Although only about half of the identified genes could be placed into
functional categories, numerous genes whose function is associated with the
cell cycle (e.g. functional categories such as “cell cycle regulation”, “cell wall”,
“cytoskeleton”, “protein synthesis”, “replication”), and gene expression (e.g.
“transcription factors”) were identified. Their abundance distribution across
the canonical cell cycle phases was analyzed and striking phase-specific pat-
terns emerged (Breyne et al. 2002). Significant oscillation of transcript abun-
dance in the course of the cell cycle was observed for core cell cycle regu-