doi:10.1088/1478-3975/5/2/026001

A network scientist highlights active sites of enzymes, cells, brains and society.

For proteins, chemical binding is a tricky business. Special signals must be sent across a sea of water molecules to the desired partner, and complex mutual structural adjustments (a fluctuation fit) must be completed before each successful binding event.

I have long taught that a protein at its lowest-energy conformation still has regions of higher energy. But I've always been intrigued: how is the extra energy of the active sites preserved? And why do we need such big enzymes when their active sites occupy only a tiny region?

Piazza and Sanejouand found part of the answer by identifying special energy-preserving segments of proteins (F. Piazza and Y.-H. Sanejouand Phys. Biol. 5, 026001; 2008). Taking into account the effect of the surrounding water, they modelled proteins with a computer program that arranges oscillating elements in the same pattern as amino acids in real proteins. In most of these proteins, they identified a few easily excitable segments that collected and harboured long-lived, localized vibrations. An analysis of 833 enzymes showed that these segments co-occur with the catalytic active sites; are located on the stiffest parts of the proteins; and have many connections but are surrounded by a less well-connected environment.

The generality of many network properties prompts me to ask: can we find 'active sites' of cells, brains, ecosystems and societies? Piazza and Sanejouand's segments correspond to Ronald Burt's “structural holes” in social networks — whereby areas of greatest economic potential are areas of low connectedness, where brokers can make new connections. Indeed, not only amino acids, but people may also act as brokers, mediators and catalysts. It may be worthwhile to think about creative, broker proteins as drug targets. One could even imagine creative sets of neurons.

Discuss this paper at http://blogs.nature.com/nature/journalclub