Michael Springer - US grants

Intellectual Merit:
Cells are constantly bombarded with environmental signals. Often these signals are in opposition; for example, one signal on its own might lead to growth, another to death. When both signals are simultaneously present, the cell needs to integrate two pieces of information to make a single decision. Currently, there is little understanding of how the integration of multiple inputs is achieved, but it is key to predicting the behavior of cells in complex environments. For yeast, a well-known example of multiple signaling is that of the presence of mixtures of the two sugars glucose and galactose. The textbook description of the integration event is simple: glucose signaling dominates over galactose signaling. This project will test a hypothesis challenging that description - for a broad range of concentrations, the response is far more complex, and involves a novel "ratio-sensor" mechanism. The mechanism would make it possible for the cell to opt for "the best of both" rather than being limited to "all or none". Establishing the existence of such flexibility in yeast responsiveness to sugar mixtures has far-reaching implications, in that it may extend to any number of other trade-off situations between two conflicting physiological or evolutionary objectives where the optimal decision would depend on the relative levels of a plurality of signals.

Broader Impacts:
Many institutions have highlighted the need to introduce quantitative, theoretical and computational approaches into the life sciences mainstream. In response to that need, an initiative has been undertaken to develop an integrated quantitative curriculum for life scientists, beginning with a quantitative "bootcamp". The bootcamp has been extremely well received and materials have been used at other institutions around the world; within the next year a series of modules derived from the bootcamp will be made available via Harvard's online course offerings, HarvardX. The present project will generate data and conceptual approaches that will be used as examples in the bootcamp, and also in other curricula to be developed that will teach, hand-in-hand, experimental design, visualization, and quantitative and analysis methods.

Project Summary Most human traits are complex/quantitative. Similarly, many common human diseases are complex; they typically are not caused by a small number of genes, but instead are influenced by hundreds if not thousands of genes. Little is known about quantitative traits due to conceptual, experimental, and analytical limitations. This proposal aims to address several key questions: 1) what are the genes that can drive a quantitative trait and how are they interrelated, 2) what are the genes that drive variation in a quantitative trait in natural populations, and 3) how do the phenotypes of each individual quantitative gene combine to determine the overall phenotype of the trait, i.e. are gene-gene interactions important. The induction of galactose and phosphate metabolic genes in the budding yeast Saccharomyces cerevisiae are classical Eukaryotic model systems for probing signaling. Preliminary results described in this proposal show that these responses are also complex traits. Our laboratory has developed high- throughput flow cytometry methods that are essential for accurately determining the effects of genes on quantitative traits both among natural variants and mutant strains. Building on our experimental strengths, we will combine fluorescence reporter strains with a series of deletion or dosage perturbation libraries. We will generate the most comprehensive list of quantitative genes yet in each of these traits, and assess the interplay of these quantitative genes within and between traits. Using allele swaps combined with bulk segregant analysis and classical linkage we will determine the extent to which alleles of quantitative genes vary in nature. By combining between zero to four alleles or deletion of quantitative genes, we will be able to directly test the importance of gene-gene interactions. This combination of approaches should greatly enhance our understanding of complex traits and have direct relevance for human disease.