Target the Bridges in Signaling Networks

Abstract

The ability to identify essential components in signaling networks may allow for targeted disruption of the processes controlled by these networks. Yu et al . wanted to determine whether it was more important that a protein be highly connected (a hub) or that it have a high degree of "betweenness" (a bottleneck), a protein through which many nonredundant paths through the network proceed. They defined four types of nodes in a pathway--hub-bottleneck node, nonhub-bottleneck node, hub-nonbottleneck node, or nonhub-nonbottleneck node--and then analyzed signaling networks, metabolic networks, and protein interaction networks to determine what types of nodes correlated best with essential genes in each type of network. In signaling networks, hub-bottlenecks and nonhub-bottlenecks were likely to be encoded by essential genes, whereas hub-nonbottlenecks were not. Thus, for signaling networks, "betweenness" may be a better indicator of essentialness than is connectedness alone, perhaps because signaling pathways have directed information flow through the network. A further refinement of the nonhub-bottleneck nodes was based on dividing the nodes into those that participate in "permanent" interactions, such as those that hold protein complexes together, versus those that are "transient," mediating temporary protein interactions. Not surprisingly, those nonhub-bottleneck nodes defined as permanent were more likely to be encoded by an essential gene than were the transient nonhub-bottlenecks. Based on their analysis, the authors suggest that nonhub-bottlenecks may represent the points of pathway crosstalk. Cak1p, which is encoded by an essential gene in yeast and is a cyclin-dependent protein kinase-activating kinase, is presented as an example of a nonhub-bottleneck that connects two signaling pathways: regulation of the cell cycle and the sporulation pathway. H. Yu, P. M. Kim, E. Sprecher, V. Trifonov, M. Gerstein, The importance of bottlenecks in protein networks: Correlation with gene essentiality and expression dynamics. PLoS Comput. Biol. 3 , e59 (2007). [PubMed]