Mitotic kinase oscillation governs the latching of cell cycle switches

In 1996, Kim Nasmyth(1) proposed that the eukaryotic cell cycle is an alternating sequence of transitions from G(1) to S-G(2)-M and back again. These two phases correlate to high activity of cyclin-dependent kinases (CDKs) that trigger S-G(2)-M events and CDK antagonists that stabilize G(1) phase. We associated these "alternative phases'' with the coexistence of two stable steady states of the biochemical reactions among CDKs and their antagonists.(2,3) Transitions between these steady states (G(1)-to-S and M-to-G(1)) are driven by "helper'' proteins. The fact that the transitions are irreversible is guaranteed by a "latching'' property of the molecular switches, as we have argued in previous publications.(4,5) Here, we show that if the latch is broken, then the biochemical reactions can swing back-and-forth across the transitions; either G(1)-S-G(1)-S. (periodic DNA replication without mitosis or cell division) or M-(G(1))-M-(G(1)) ... (periodic Cdc14 release, without fully exiting mitosis). Using mathematical modeling of the molecular control circuit in budding yeast, we provide a fresh account of aberrant cell cycles in mutant strains: endoreplication in the clb1-5 Delta strain(6) and periodic release and resequestration of Cdc14 (an "exit'' phosphatase) in the CLB2kd Delta strain.(7,8) In our opinion, these "endocycles'' are not autonomous oscillatory modules that must be entrained by the CDK oscillator(6,7) but rather inadvertent and deleterious oscillations that are normally suppressed by the CDK latching-gate mechanism.(8)