The synthesis of a bacterial clock

Our bodies have their own internal clocks, in part set up by the cycle of light and dark we live in. The day-night cycle helps create circadian rhythms which control different physiological traits like body temperature and cortisol levels in a roughly 24 hour cycles. At a smaller scale, coordination of broad cellular activity is vital for several processes in the human body. For example, near-synchronous activity of neurons in the brain are required for attention, sleep cycles, and motor movements. Dysregulation of these processes have been implicated in neurological diseases like autism and epilepsy. How the body sets up rhythmic activity is a very complicated problem to solve

So what about an even smaller scale? What controls the temporal dynamics of a population of bacteria? Bacteria may need to overcome its intrinsically noisy signaling to organize a large process, like aggregation or entering into the cell cycle. To do this, some bacteria communicate using a process called quorum sensing. Bacteria that communicate with quorum sensing secrete signaling molecules that are detected by other bacteria. When the concentration of the signaling molecules hits a certain point, this will induce transcription of genes that induce a cooperative action of the bacteria. Researchers from UC San Diego took advantage of quorum sensing to see if they could engineer a rhythmic circuit in bacteria, initiating synchronous and oscillating activity. If they could, it would help us learn more about how synchronous activity is set up in the natural world.

Quorum sensing relies on specific singaling molecules. One such signaling molecule is acyl-homoserine lactone (AHL). This little molecule can cross the cell membrane and diffuse away and into other bacteria to initiate intercellular cooperation. AHL binds to a molecule that's being consistently produced by the cell, named LuxR. When coupled together, the LuxR-AHL complex activates transcription at the luxI promoter. The scientists engineered a strain of E. coli. bacteria to put three different genes, luxI, aiiA, and yemGFP under three copies of the same luxI promoter. In this way, each gene would be activated by increased levels of AHL. The gene luxI codes for the protein LuxI, an enzyme which creates AHL. This represents a positive feedback loop, as increases in AHL leads to increased LuxI creation, which in turn synthesizes more AHL. The gene aiiA codes for the protein AiiA, a protease that breaks down AHL. This will initiate a negative feedback loop as increases in AHL lead to increased AiiA levels, which will in turn break down the AHL, leading to inactivation of the promoter, and a reduction of AiiA levels. The gene yemGFP codes for the protein YemGFP which is a fluorescent and will start glowing when translated. This whole system is outlined in the picture above.

&&

To monitor the bacteria and see if they created rhythmic activity, they used microfluidic devices that allowed them to control the size of the bacterial colony and visualize the GFP activity. If their activity was coordinated between bacteria, you would expect to see oscilations of global GFP expression. They let their engineered bacteria start growing and monitored its GFP activity. At first, there was no fluorescence. There is some naturally produced LuxI that is creating AHL, but when the population of cells is small, there isn't enough AHL being produced by the entire population to initiate large amounts of luxI promoter activity. Some of the AHL is being washed away, and some is being quickly degraded. As the population increases, it will eventually reach a size where the collective amount of AHL produced will be enough to activate the luxI promoter, starting the transcription of luxI, aiiA, and yemGFP. This will induce high levels of fluorescence, increase LuxI levels, and produce even more AHL. It will also upregulate AiiA protein. After a period of time, there will be enough AiiA accumulated to start degrading the AHL and inactivate the promoter. What this circuitry looks like from an outside perspective is shown in the figure above. At the beginning of the cycle, the critical point is reached and the entire colony of bacteria will start to glow. However, as AiiA levels slowly increase, it will degrade AHL, stop transcription of yemGFP, and the colony will stop glowing. this activity will oscillate as the levels of AHL and the effected proteins increase and decreace. In this scenario, AHL has two roles: it mediates cell synchrony at appropriate cell densities based on AHL's concentration, and it directly activates the genes necessary for cellular coordination. It allows for the synchronization of different cells in a population where there once was disorder.

The success of engineering bacteria to have oscillating, coordinated gene expression represents a big advancement in the field of genetic clocks. Most other attempts at engineering cyclic activity in bacteria has either lacked strong coupling, or the synchronous activity would end quickly. In this set-up, the bacteria go through many GFP fluorescent cycles, as is shown by the graph to the right. Also, unlike previous synthetic oscillators, this manages to coordinate the activity throughout an entire population of bacteria, not just a small colony. Using the inherent signaling properties of the bacteria, researchers made an entire colony of bacteria with their own individual cycles robustly react to a signal in unison with increased gene expression - an incredible feat. Designing our own synchronous activity can help us understand how cell activity can become synchronous, whether it's colonies of bacteria, or neurons in the human brain. This type of information will be vital to figuring out how these processes work, and how they can become dysregulated. Even beyond that, engineering cyclic rhythms from scratch is incredibly cool!

References:

Danino, T., Mondragón-Palomino, O., Tsimring, L., Hasty, J. A synchronized quorum of genetic clocks. Nature 463, 326-330 (2010)

Fussenegger, M. Synchronized bacterial clocks. Nature 463, 301-302 (2010).

Image Credits:

The first image is augmented from the Fussenegger paper cited above.

All images are adapted from theh Danino & Mondragón-Palomino et al. paper cited above.