A microbiologist hopes to disrupt bacterial 'decisions'.

Cyclic-di-GMP is small but important. It is an intracellular signalling molecule that controls lifestyle choices in bacteria. When should a bacterium become virulent? When should it differentiate into a new cell type? When might it do better to stop moving around and stay still with many others? Bacteria that gather together tend to encase themselves and their neighbours in a carbohydrate slime, forming what is known as a biofilm. I, like many microbiologists, am keen to find ways to disrupt biofilms, and a better understanding of how cyclic-di-GMP works may provide a way to do this.

Recently, answers have started to emerge. First it was shown that cyclic-di-GMP can bind to certain proteins that modulate the activity of flagellar motors — which propel free-swimming bacteria — and to enzymes that make the biofilm-cementing slime. Then researchers found a protein that 'turns on' some of the slime genes when it attaches to cyclic-di-GMP. But one paper shows a completely new way in which cyclic-di-GMP can control bacterial lifestyle choices: by binding to a regulatory region, called a riboswitch, on a messenger RNA molecule (N. Sudarsan et al. Science 321, 411–413; 2008).

Ronald Breaker and his team at Yale University in New Haven, Connecticut, report how they used various molecular-biology techniques to demonstrate that part of the RNA hitches itself to cyclic-di-GMP. They also proved that cyclic-di-GMP-binding riboswitches from several bacterial strains can function as genetic 'off' as well as 'on' switches.

These findings are noteworthy because humans do not make cyclic-di-GMP, so the molecule could be a target for new antibiotics. Medicines that attack cyclic-di-GMP should be able to treat biofilm-related disorders such as periodontal disease and ear infections, which are often resistant to existing drugs.

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