The long-term goal of this grant is to exploit budding yeast, Saccharomyces cerevisiae, as a model system for studying eukaryotic biology and its evolution at the system level. One of our specific aims is to deploy methods we have developed that make it possible to perturb steady- state growth in chemostats by inducing exactly one gene or degrading exactly one protein. By following changes in gene expression thereafter, we can observe the direct and indirect consequences of the perturbation genome-wide. The dynamic resolution we can achieve makes it possible to infer causation from temporal order, where previously only correlation was possible. We plan (i) a genome-scale survey (already begun) of the consequences of induction of the known transcription factors in yeast, (ii) the development of synthetic transcription factor based on zinc-finger proteins, (iii) the adaptation of nuclear hormone receptors (in addition to the estrogen receptor), all of which will allow (iv) more sophisticated perturbations, including especially pulses of induction followed by degradation. We will explore the possibility that these methods will produce useful information about regulators other than transcription factors, such as protein kinases, protein phosphatases, acetylating and de-acetylating enzymes, etc. All of these data will be used to construct quantitative models of regulation, as we recently did for the regulation of the combinatorially controlled sulfur assimilation pathway.
A second aim i s to survey the genes in which mutations contribute to fitness in evolving populations of yeast, exploiting our recent study of the dynamic trajectories of beneficial sterile mutations in 600 cultures grown for 1,000 generations. By deep sequencing of the populations over time we will find and follow the diverse mutations in the standing variation that allow subsequent sterile mutations to sweep. In this way we can test further the hypothesis that the constraints of the cellular system biology impose limitations on the number ways increased fitness can be achieved. We will develop fast methods to distinguish causal mutations from the passenger mutations, and reconstruct fitter variants de novo as a test of our inferences. We will study (i) the dynamics of passenger versus causal mutations in populations where the sterile mutation failed to sweep;(ii) comparisons of relative fitness of diverse combinations of beneficial mutations from independent cultures;(iii) characterization of the genome wide expression phenotypes of fitter variants. We hope these methods will shed light on the development of human tumors which, like the yeast in a chemostat, evolve ever-more-fit variants that grow faster in an essentially constant environment.
We study growth of yeast because much of what we find in yeast applies to humans, both at the level of individual proteins, and at the level of cellular system integration. This gives us reason to believe that our results will shed light on the development of human tumors which, like the yeast, constantly evolve ever-more-fit variants that grow faster in an essentially constant environment.
Showing the most recent 10 out of 88 publications
