An important biomedical research goal is to reconstitute human tissues from induced pluripotent cells for in vitro interrogation of human disease states. However, our ability to coax stem cells to form complex multi- cellular structures is currently limited by deep conceptual and technical challenges. While high-throughput gene expression and biochemical measurements allow us to reconstruct the molecular circuits that control stem cell fate, these circuits are enormously complex, and the maps produced by genomics are static. Therefore, we do not understand how individual stem cells dynamically integrate signals from their environment while communicating with other cells to coordinate and construct multi-cellular tissues. The major goal of my research is to combine high-throughput single cell imaging and mathematical modeling to derive reduced, predictive models of cell fate circuits and to exploit these models with new optogenetic tools to manipulate stem cell fate with spatial and temporal control. In this application, I use mouse Embryonic Stem (ES) cell differentiation as a powerful model system for quantitative single cell analysis of fate selection n a stem cell population. In response to Wnt and Fgf signals, ES cells leave the pluripotent state and select between two alternate germ layer cell fates. A complex network of transcription factors controls the ES cell, but, in previous research, I showed that a circuit of just two transcription factors, Oct4 and Sox2, controls germ layer fate selection. Now, I use Oct4 and Sox2 as the essential nodes in a quantitative and predictive model of germ layer fate selection that incorporates both single cell information processing and inter-cellular communication. First, I will perform high-throughput single cell time lapse imaging of Oct4 and Sox2 protein levels to quantify the ES cell's response to a large array of Wnt and Fgf inputs. With statistical analysis, will reduce these measurements to a predictive dynamical systems model of signal integration. Second, to determine the impact of inter-cellular communication on ES cell fate selection, I will quantify the spatial and temporal propagation of differentiation signals through the Wnt, Fgf, and Notch pathways in ES cell populations and construct a population level model of cell fate selection. Third, I will combine the model with new optogenetic tools to modulate the single cell response to Wnt and Fgf in order to direct germ layer differentiation with spatial control. Since germ layer differentiation is a foundational process of both mammalian development and in vitro differentiation, optogenetic control of this process would provide a platform for in vitro construction of complex multi-cellular structures from germ layer derivatives. Together, these aims will provide conceptual insight into how stem cells communicate to execute multi-cellular processes like tissue development, homeostasis, and repair. Further, my application will provide a proof of principle for optogenetic light-gated control of in vitro embryonic stem cell differentiation to synthetically generate tissues for studies of human disease.
The goal of this application is to develop mathematical models of stem cell differentiation and to use these models to guide stem cells to form complex multi-cellular structures like muscle tissue and neural circuits in the laboratory. By controlling stem cell differentiation, we might establish in vitro models of human physiology and disease and learn how millions of individual cells in the body work together to form and repair tissues.
|Sivak, David A; Thomson, Matt (2014) Environmental statistics and optimal regulation. PLoS Comput Biol 10:e1003826|