Yeast: angiogenesis model? Yup

Yeast may not have blood vessels, but it could be a powerful model organism for studying angiogenesis, according to linkurl:a study;http://www.pnas.org/content/early/2010/03/11/0910200107.full.pdf+html published online in Proceedings of the National Academy of Sciences yesterday (March 22, 2010) that describes a new, systems-biology approach for identifying surprising model organisms for human diseases. Image: Wikipedia"It's a Eureka moment of, gosh, I can't believe anybody didn't think of this before," said linkurl:Robb Krumlauf,;http://www.stowers-institute.org/labs/KrumlaufLab.asp a developmental biologist and scientific director of the Stowers Institute for Medical Research in Kansas City, Mo., who did not participate in the study. "Most people work in a very directed way" when studying human diseases, he added, but the current study brings to the fore how "looking at a plant meristem or a marine invertebrate would give you tremendous insight" into seemingly unrelated human diseases. Organisms as diverse as plants, worms and vertebrates share related groups of genes that encode identical signaling pathways; though the inputs and outputs to these pathways may differ, the pathways themselves are conserved. "Coming from an evolutionary viewpoint, we've known for a while that you have this remarkable property where organisms can co-opt [genetic networks] for entirely different things," said linkurl:Nipam Patel,;http://patelweb.berkeley.edu/ a developmental biologist at the University of California, Berkeley, who was not involved in the research. One classic example, Patel notes, is a set of genes which regulates dorsal-ventral patterning in Drosophila, but in vertebrates, plays a role in immune system activation. Since the same gene networks are wired up to different processes, said systems biologist linkurl:Edward Marcotte;http://polaris.icmb.utexas.edu/index.php/Main_Page of the University of Texas at Austin, the study's main author, he and his colleagues decided to try to search for these different phenotypes which were linked up to the same genes. Marcotte and his colleagues mined databases and the published literature for gene-phenotype associations to identify genes linked to approximately 300 human diseases and more than 6,000 phenotypes in humans, mice, C elegans, and yeast. Then they looked for the points of overlap in the sequences of these genes between pairs of organisms. Phenotypes generated by mutations in these conserved genes, regardless of whether or not the two organisms shared the same phenotype, were termed phenologs. Because the underlying genetics is the same, Marcotte, "it doesn't matter that the phenotypes don't match." Many of the phenologs the researchers identified pointed to well-known disease models, serving to confirm the validity of the researchers' approach, Marcotte said. For instance, the method identified a knockout mouse whose ciliary defects were phenologs for the human genetic disorder called Bardet-Biedl syndrome -- for which it is already used as a model organism. However, many identified phenologs were completely new. A set of orthologous genes were involved in development of the vasculature in mice, and in yeast regulated the organism's growth rate in the presence of the cholesterol-lowering drug lovastatin, suggesting that yeast could be a model for studying angiogenesis, even though it has no blood vessels. Examining 59 of these yeast genes -- many of which had never before linked with blood vessel development -- in developing frog embryos, the researchers found that five were strongly expressed in the vasculature, suggesting their involvement in angiogenesis-related processes. Indeed, one of the genes caused disruptions in the vasculature when knocked down. "What this paper does is really just drive home the kind of deep homology of all these model systems." said linkurl:John Wallingford,;http://www.bio.utexas.edu/faculty/wallingford/ a developmental biologist also of the University of Texas at Austin, and a coauthor on the study. "In the end, this is a really good example of a really good inroad tool" to pull out some interesting hypotheses, which then must be validated, Patel said. The technique allowed them to significantly speed up the rate at which they were able to discover relevant genes, said Wallingford -- some 21 times faster than by traditional methods. Krumlauf said researchers could use this method to find phenologs of genes identified in genome-wide association studies, and use model organisms to tease out their molecular functions and detect other genes with which they function. "What this tells us is, we don't know where the next major discovery [about human disease] will come from, but as long as we collect the information in a good way and put it in the public domain," unexpected discoveries can occur, he said. The group's computational approach isn't just useful for modeling disease, said Marcotte. By systematically identifying these ancient conserved gene networks, "we can go back and start defining how genes were organized into these systems in the last common ancestor." Patel agreed, noting that it allows researchers to address what is "in and of itself an interesting question -- are certain genetic pathways particularly good at accomplishing specific [molecular] tasks?" **__Related stories:__***linkurl:Supermodels?;http://www.the-scientist.com/blog/display/57225/ [19th March 2010]*linkurl:Modeling with model organisms;http://www.the-scientist.com/article/display/54067/ [January 2008]*linkurl:The trouble with animal models;http://www.the-scientist.com/article/display/53306/ [July 2007]

Yeast: angiogenesis model? Yup

Interested in reading more?

Become a Member of