image: This figure illustrates the organization of gene regulation in yeast: each yellow dot represents a set of transcription factors that regulates a distinct set of genes. In cases where the modules on top are contained in those below, a blue line connects them, revealing how the cell assembles higher regulatory specificity for some genes, and less for other groups of genes. The encircled regions contain groups of modules that are enriched for the indicated transcription factors, with the indicated regions in this network serving specific cellular functions, such as cell-cycle regulation for the region under the transcription factors MBF and SBF. view more
Credit: UCSD
Bioengineering researchers at UCSD and two research institutes in Germany report in the June 16 issue of PLoS Computational Biology that transcription factors act not only in isolation, but also in pairs, trios, and combinations of up to 13 to regulate distinct sets of genes. The researchers, led by UCSD bioengineering professor Trey Ideker, reported a list with 363 combinations of 91 central transcription factors that regulate a large proportion of genes in the yeast genome. The team used rigorous statistical tests to discover active combinations of transcription factors, as if the cells were mixing and matching parts of its regulatory-protein wardrobe to respond to different environmental conditions. The researchers expect that human cells use a similar system of transcription-factor combinations, but on a larger scale.
"A cell's surprising ability to mix and match so many different combinations of these factors to achieve a high degree of complexity and specificity in the expression of its genes is impossible for even the most experienced cell biologists to conceptualize," said Andreas Beyer, a post-doctoral fellow at the UCSD Jacobs School of Engineering's Department of Bioengineering. "That's why we have computers."
The researchers combined the results of their laboratory with other large-scale measurements of transcription factor-gene binding, such as those reported earlier by MIT biology professor Richard A. Young and his collaborators.
Ideker's team was able to identify new transcription factor binding patterns by borrowing a concept from computer science. The team considered the binding of one transcription factor to one gene as analogous to one "hop" of a data packet from one Internet router to another.
In the case of gene regulation, Ideker's team identified "2hop" relationships by first focusing on single transcription factor-gene associations, plus other experimental evidence that indicates that that gene regulates a second gene.
To enlarge the scope of the model further, Ideker's group also incorporated other previously discovered transcription-factor interactions and related genetic results. They relied on a total of eight types of direct and indirect evidence to create a model. That model predicts 980 as-yet-undiscovered transcription factor-gene binding interactions.
"This 'systems biology' approach, using so many different lines of evidence, has given us a much more revealing and detailed picture of how cells orchestrate gene regulation to cope with different environments," said Ideker. "We're far from understanding the full picture of gene regulation in a cell, but this new information should give scientists who are interested in blocking transcription factors a powerful new tool to narrow their search to the most promising candidates."
Other researchers involved in the project include: Christopher Workman, UCSD Jacobs School of Engineering's Department of Bioengineering (now at The Technical University of Denmark); Jens Hollunder and Thomas Wilhelm, Leibniz Institute for Age Research, Fritz Lipmann Institute, Jena, Germany; Dörte Radke and Ulrich Möller, Leibniz Institute for Natural Products Research and Infection Biology, Hans Knöll Institute, Jena, Germany.
Funding was provided by the National Institute of Environmental Health Sciences (NIEHS), the David and Lucille Packard Foundation, the German Academic Exchange Service, and the German Federal Ministry for Education and Research.
Journal
PLoS Computational Biology