Dissecting Protein-Protein Interactions that Regulate the Spindle Checkpoint in Budding Yeast

• Main Author: Lau, Tsz Cham Derek
• Other Authors: Murray, Andrew W.
• Language: en_US
• Published: Harvard University 2013
• Subjects: Biology; Cellular biology; Budding Yeast; Spindle Checkpoint
• Online Access: http://dissertations.umi.com/gsas.harvard:10558; http://nrs.harvard.edu/urn-3:HUL.InstRepos:10364580

Errors in segregation of genetic materials are detrimental to all organisms. The budding yeast ensures accurate chromosome segregation by employing a system called the spindle checkpoint. The spindle checkpoint, which consists of proteins such as Mad1, Mad2, Mad3, Bub1, and Bub3, monitors the attachment of microtubules to the chromosomes and prevents cell cycle progression until all chromosomes are properly attached. To understand how the spindle checkpoint arrests cells in response to attachment errors at the chromosomes, we recruited different checkpoint proteins to an ectopic site on the chromosome by taking advantage of the binding of the lactose repressor (LacI) to the lactose operator (LacO). We found that cells expressing Bub1-LacI arrest in metaphase. The phenotype is in fact caused by dimerization of Bub1 when it is fused to LacI rather than the recruitment of Bub1 to chromosome. The cell cycle arrest by the Bub1 dimer depends on the presence of other checkpoint proteins, suggesting that the dimerization of Bub1 represents an upstream event in the spindle checkpoint pathway. The results with the Bub1 dimer inspired us to fuse checkpoint proteins to each other to mimic protein interactions that may contribute to checkpoint activation. We showed that fusing Mad2 and Mad3 arrests cells in mitosis and that this arrest is independent of other checkpoint proteins. We believe that combining Mad2 and Mad3 arrests cells because both proteins can bind weakly to Cdc20, the main target of the spindle checkpoint, and the sum of these two weak bindings creates a hybrid protein that binds tightly to Cdc20. We reasoned that if Mad3's role is to make Mad2 bind tightly, artificially tethering Mad2 directly to Cdc20 should also arrest cells and this arrest should not depend on any other checkpoint components. Our experiments confirmed these predictions, suggesting that Mad3 is required for the stable binding of Mad2 to Cdc20 in vivo, that this binding is sufficient to inhibit APC activity, and that this reaction is the most downstream event in spindle checkpoint activation. The interactions among spindle checkpoint proteins thus play an important role in cell cycle arrest and must be carefully regulated.