Jared Toettcher

Education

Ph.D. Candidate, Biological Engineering, Massachusetts Institute of Technology

B.S., Bioengineering, University of California, Berkeley, 2004

CDP research

Understanding how the damage response and cell cycle networks interact is crucial to our ability to enhance the efficacy of such existing treatments, and perhaps drive development of more specifically targeted therapies. To address this problem, we are taking a combined theoretical and experimental approach to elucidating the systems level properties of IR-induced cell cycle arrest.

We built and validated models of the cell cycle and DNA damage response. We modeled the DNA damage response as consisting of the ATM-Chk2 kinase cascade, resulting in p53 activation. This model was parameterized with single-cell microscopy data from MCF-7 cells. We connected these modules by implementing a variety of G1 and G2 arrest mechanisms dependent on Chk2 and p53 activation.

The model predicts qualitative differences between G2 arrests induced by cyclin inactivation and cyclin transcriptional repression. We measured cyclin E, A and B profiles after arrest in U2OS cells, and were able to fit the computational model to this data only if arrest is mediated by G2 cyclin inactivation. The fit model is able to correctly recapitulate the cyclin levels during arrest, as well as the percentage of time spent in each cell cycle phase during normal cycling.

Why might cells choose to arrest by cyclin inactivation, rather than transcriptional repression? Our model predicts that cells arrested primarily by transcriptional repression of either cyclin B or Cdk1 can properly arrest in G2, but upon cell cycle re-entry may enter a G1-like state, albeit with 4N DNA content. Such cells would be predicted to undergo a second round of DNA replication, a phenomenon known as endoreduplication. We are testing this prediction experimentally in H1299 cells (which normally arrest by cyclin inactivation), expressing an inducible shRNA against Cdk1. Preliminary evidence suggests that these cells do endoreduplicate their DNA upon recovery from IR-induced G2 arrest.

In addition to addressing biological questions, this project presents a variety of computational challenges. In parallel with this work, we are developing tools to integrate large models developed independently of one another, and to compute parameter sensitivities to efficiently optimize the ratio of arrested cyclin levels to their maximal concentration during normal cell cycle progression.

Publications

Weinberger LS, Burnett JC, Toettcher JE, Arkin AP, Schaffer DV. Stochastic gene expression in a lentiviral positive-feedback loop: HIV-1 Tat fluctuations drive phenotypic diversity. Cell. 2005 Jul 29;122(2):169-82.