UC San Diego

Embryonic stem cells have the capacity to replicate and differentiate into specialized cell types. Current methodologies for driving ES differentiation in vitro are highly inefficient, rely on using relatively large 3D aggregates of cells, and do not permit an adequate diversity of signaling environments. While the differentiation effects of soluble signals has been extensively investigated, extracellular matrix signals, mediated by integrin ligation, have not been systematically examined. Furthermore, the well-established crosstalk between integrin and growth factor signaling has not been systematically explored. In this dissertation, we focus initially on developing a platform for studying combinatorial extracellular matrix signaling environments in parallel. Using robotic spotting techniques, we develop microarray methods for controlled protein immobilization and micropatterned cell culture assays on these confined domains. We demonstrate the utility of the platform by studying the effects of combinatorial ECM protein mixtures on cell differentiation in two contexts: 1) maintaining primary liver cell function; and 2) mouse embryonic stem cell differentiation towards an early hepatic fate. To assess early liver fate specification, we utilize the I114 reporter cell line. Factorial analysis methods identify evidence of potentially synergistic and antagonistic ECM interactions in both hepatocyte and ES differentiation studies. To study the interaction of ECM and GF environments, we next developed a multiwell ECM platform (i.e. 100 ECM spots per well of a 96-well plate) and novel methods for quantitative data acquisition. In particular, we demonstrate the ability to use lineage-specific GFP reporters, conversion of this signal to a fluorescent Cy5- equivalent that can be detected by laser excitation, and validate the use of confocal DNA microarray scanners and software for data acquisition and quantitation. Using controlled population mixtures, we characterize and validate in situ semi-quantitative assays for GFP and DNA. We demonstrate this new technology by studying the effects of 240 unique signaling environments on mES differentiation towards the cardiac lineage using a GFP reporter of MHC-a. This versatile technique is compatible with virtually any set of insoluble and soluble cues, leverages existing software and hardware commonly available, and represents an important step towards the development of the 'systems biology' of stem cells