Protein-protein interaction networks: To party or date?

signaling update home

updates

news

research library

Exclusive to Signaling Gateway

The yeast interactome exhibits organized modularity where a small proportion of proteins – the 'hubs' – interact with many partners. These hubs fall into one of two categories: 'party' hubs, which interact with most of their partners simultaneously, and 'date' hubs, which bind different partners at different locations and times.

Protein-protein interaction networks, or 'interactomes', provide a systems level view of cellular processes like signal transduction, although they have so far been largely considered as static networks. Within these networks, most proteins interact with few partners. However, a small number of proteins – called 'hubs' – interact with many different partners. Scale-free interactome networks are particularly vulnerable to the removal of hubs – knocking out yeast hub genes results in a threefold increase in lethality, compared with the removal of peripheral proteins. Han et al. have analyzed the properties of these hubs taking temporal and spatial dynamics into account. They conclude that hubs fall into one of two categories: 'party hubs', which interact with most of their partners simultaneously, and 'date hubs', which bind different partners at different locations and times.

The biological role of topological hubs may vary depending upon the timing and location of the interactions they mediate. The authors used a filtered yeast interactome (FYI), compiled from different sets of yeast mRNA expression profiling data. The FYI contained 1,379 proteins with an average of 3.6 interactions per protein. The average Pearson's correlation coefficient (PCC) was used to measure the strength of the linear relationship between each hub and its partners. A bimodal pattern of distribution was most clearly apparent for two conditions analyzed, the 'stress response' and 'cell cycle', indicating two distinct hub populations: 'party hubs' with a high average PCC and 'date hubs' with a low PCC.

'Date hub' partners are significantly more diverse in subcellular location than 'party hub' partners. Upon the removal of 'party hubs', general network connectivity was unaffected. However, 'date hub' removal resembled the effects seen by deletion of all hubs. The network remaining after the removal of 'party hubs' is significantly greater than that left behind after the removal of date hubs. In fact, the FYI subnetworks removed with 'date hubs' correspond to small modules of specific biological processes. These sub-networks represent not only stable molecular machinery but also loosely connected regulatory pathways such as osmosensing.

Thus, a model emerges for organized modularity where 'date hubs' represent global connectors between modules, and 'party hubs' function inside modules. This suggests that 'date hubs' have a central role in the modular organization of the yeast proteome. For example, the 'date hub' calmodulin connects four different biological modules, 'cation homeostasis', 'protein folding and stabilization', 'budding, cell polarity and filament formation' and 'endoplasmic reticulum', whereas the party hubs Sec17, Sec22, and Vti1 all function within the 'endoplasmic reticulum' module. The proteome is more sensitive to perturbation of date hubs than that of party hubs.

This spatio-temporal dynamic analysis may be extended to other systems such as the World Wide Web, social and epidemiological networks.

Brenda Riley, Assistant Editor
Signaling Gateway

References

Han, Jing-Dong J, Bertin, Nicolas, Hao, Tong, Goldberg, Debra S, Berriz, Gabriel F, Zhang, Lan V, Dupuy, Denis, Walhout, Albertha J M, Cusick, Michael E, Roth, Frederocl P, & Vidal, Marc Evidence for dynamically organized modularity in the yeast protein-protein interaction network. Nature, 430, 88–93 (01 July 2004); 10.1038/nature02555 | Article | PubMed |