Credit: Reproduced from Morgan, D.O. & Roberts, J. M. Nature 418, 495-496 (2002)

Cyclin destruction and the associated drop in cyclin-dependent kinase (CDK) activity are necessary for mitotic exit. In yeast, degradation of the S-phase cyclin Clb5 was thought to be essential, but a study reported in Nature now raises doubts about the importance of Clb5 in regulating mitotic exit.

In frog embryos, a single oscillator can manage the cell cycle. This negative-feedback oscillator alternates between S and M phases, and involves the regulation of a ubiquitin–protein ligase called the anaphase-promoting complex (APC). This is activated by a molecule called Cdc20, which then recruits cyclins. CDKs phosphorylate and thereby activate Cdc20–APC, which triggers cyclin destruction and mitotic exit. As a result, the CDK activity decreases, which causes APC inactivation and re-accumulation of cyclin.

The cell cycle in somatic cells and yeast, however, is more complex because it includes a G1 phase that is important for cell growth and differentiation. G1 regulation might be achieved by a second oscillator that differs from the first one in using the APC-regulatory subunit Cdh1 instead of Cdc20. CDKs inhibit Cdh1–APC, which means that, in late mitosis, CDK proteolysis causes activation of Cdh1–APC, which then maintains cyclin destruction throughout G1 (see diagram).

To address the question of whether the Clb5 cyclin must be destroyed, Wäsch and Cross constructed a yeast strain (CLB5 Δdb) in which the CLB5 gene lacked the destruction-box sequence that targets Clb5 to Cdc20–APC. However, these cells did not seem to have any defect in mitotic exit, implying that proteolysis of Clb5 is not required. Exit also occurred in CLB5 Δdb cells that lacked Sic1, a CDK inhibitor responsible for further reducing CDK activity in late mitosis. This suggests that mitotic exit can progress without either Clb5 degradation or Sic1.

Next, Wäsch and Cross analysed the mitotic cyclin Clb2 as a potential Cdc20 target, and showed that CLB2 Δdb yeast cells were unable to exit mitosis. Clb2 might therefore be the cyclin that needs to be destroyed for mitotic exit to occur.

Previous studies had shown that cells that lacked Cdh1 and Sic1 are not viable, suggesting that both proteins are essential for mitotic exit, and that the Cdh1–Sic1 oscillator might be the more important regulator. By contrast, Wäsch and Cross found that cells lacking these proteins do not have a marked defect in mitotic exit, and indeed survive, although poorly. However, the cells showed some abnormalities associated with an unstable G1 phase. This finding also highlights the important role of the Cdc20 oscillator in destroying Clb2.

So, the destruction of mitotic cyclins such as Clb2 is crucial to mitotic exit and the Cdc20 and Cdh1–Sic1 oscillators seem to work in a complementary manner, with the latter being important for a stable G1 phase.