Model may offer better understanding of embryonic development
Rendered image, showing cross sectional slice of a Drosophila embryo (top) and artistic rendition of the development of an image-based finite element model of embryonic pattern formation (bottom). The heat map on the left of the bottom panel represents the population average BMP signaling distribution and the right shows the results of the model with the "winning" mechanism. 9 alternative regulatory schemes and over 20 hypotheses were tested computationally and experimentally by researchers at Purdue University and the University of Minnesota. Understanding the spatial and temporal aspects of cell signaling by a combined experimental/modeling approach provides a wealth of information that may be used to control cells in other settings.
WEST LAFAYETTE, Ind. - A mathematical model developed at Purdue University can predict complex signaling patterns that could help scientists determine how stem cells in an embryo later become specific tissues, knowledge that could be used to understand and treat developmental disorders and some diseases.
During embryonic development, proteins attach to cell receptors and start a cascade of reactions. Understanding those reactions is difficult, however, because feedback signals go back out to the proteins or other molecules along the cascade, constantly changing the reaction pattern. The outcomes of those reactions and the feedback mechanisms - or inputs - are known because they can be observed, but how the inputs lead to the outputs isn't understood.
"We want to understand how stem cells become tissue-specific so that we can manipulate that process to create cells that could be used to treat injuries and diseases," said David Umulis, a Purdue assistant professor of agricultural and biological engineering. "Using a model approach, we can simulate these complex signaling patterns to get a better handle on the process."
Umulis created a model that predicted accurate outcomes when different feedback mechanisms were inserted. His results were published in the current issue of the journal Developmental Cell.
"Fruit fly embryos are a fantastic system to peer into early development since input/output relationships are easy to observe. You have a mutation and an output, but we don't typically know what happens in the middle," he said. "Realistic model embryos proved an additional tool that can be used to aid in that understanding. Models can link that cause and effect."
The study looked at fruit fly, or drosophila, embryos during very early development to decipher what controls the differentiation of these stem cells at their proper locations. During the process, cells take on identities that later specify tissue types in the adult organism. Before directional cues dictate development, the stem cells are capable of becoming many different tissues. Using models to analyze the dynamic signals the cells are receiving may help to better understand how to control similar cells in a laboratory setting.
Umulis said his model is a sort of template to allow researchers to test a number of hypotheses before conducting actual experiments. The information garnered from realistic 3-D models can guide the process and facilitate rapid discovery.
Umulis' next step is to count the number of molecules needed to initiate specific cell responses during embryonic development. The National Institutes of Health and Purdue University funded his work.
Writer: Brian Wallheimer, 765-496-2050, bwallhei@purdue.edu
Source: David Umulis, 765-494-1223, dumulis@purdue.edu
ABSTRACT
Organism-Scale Modeling of Early Drosophila Patterning Via Bone Morphogenetic Proteins
David M. Umulis, Osamu Shimmi, Michael B. O'Connor and Hans G. Othmer
Advances in image acquisition and informatics technology have led to organism-scale spatiotemporal atlases of gene expression and protein distributions. To maximize the utility of this information for the study of developmental processes, a new generation of mathematical models is needed for discovery and hypothesis testing. Here, we develop a data-driven, geometrically accurate model of early Drosophila embryonic bone morphogenetic protein (BMP)-mediated patterning. We tested nine different mechanisms for signal transduction with feedback, eight combinations of geometry and gene expression prepatterns, and two scale-invariance mechanisms for their ability to reproduce proper BMP signaling output in wild-type and mutant embryos. We found that a model based on positive feedback of a secreted BMP-binding protein, coupled with the experimentally measured embryo geometry, provides the best agreement with population mean image data. Our results demonstrate that using bioimages to build and optimize a three-dimensional model provides significant insights into mechanisms that guide tissue patterning.