Modeling the Budding Yeast Cell Cycle

Target model

If the only targets of Cdc14 were Cdc6, Sic1 and Cdh1, as shown in the model, the two mutants cdc14-ts and cdc6∆2-49 cdh1∆ sic1∆ (denoted as triple-antagonist for short) would have the same phenotype. This is not the case since cdc14-ts cells arrest in telophase in vivo, while the triple-antagonist has an unexpected phenotype in which cells undergo DNA synthesis and nuclear division, but not cell division, and then arrest as binucleate cells with 4C DNA content.

To account for this case, we introduce a "target" protein. It has been shown previously that when Cdc14 is released from the nucleolus, it dephosphorylates and activates Cdh1, Swi5, Sic1 (Visintin et al., 1998) and Cdc6. We propose here that Cdc14 not only dephosphorylates and activates these CDK antagonists, but also activates a target protein that allows cells to exit from mitosis in the absence of CDK antagonists.

Flowchart of the "target" model:

Changes from the "no-target" model:



The same parameter set as the "no_target" model is used here except for one value:
kas,rent = 160, down from 200, to make Cdc14 less inhibited by Net1. Hence, more Cdc14 is available for the triple-antagonist mutant to exit mitosis.


(1) The wild type:

(2) Two mutants helped by the "target" model:

  1. sic1∆ cdh1∆ is inviable, but able to exit mitosis for the first cycle.

  2. The mutant experiences licensing problems. ORI doesn't relicense (grey) after the first cycle.

  3. triple-antagonist (sic1∆ cdc6∆2-49 cdh1∆) is inviable, but able to exit mitosis for the first cycle.

  4. The cell can exit mitosis even though the level of Clb2 is high, but the mutant experiences licensing problems.

Reasons for not using this "Target" model:

Despite its success, we do not, at present, endorse the "target" model for two reasons. First, the target switch tends to chatter -- rapid fluctuations of target phosphorylation around the threshold value sometimes occur and must be suppressed by ad hoc modifications of the equations in the current model. Second, the predicted phenotype of the triple-antagonist mutant is very sensitive to changes in the parameters relating to the target and the phenotype of sic1∆ cdh1∆ is highly dependent on the ability of Cdc6 to inhibit Clb2. Introduction of proper checkpoint controls, however, may solve these problems with the target hypothesis.

Download the "target" model:

To download, right click and choose "Save Target As" for Internet Explorer, or "Save Link Target As" for Netscape.