Computer Controlled Yeast and an E. coli LCD Screen

Promoters of synthetic biology often fall back on analogies with engineering and computer science. Engineering genetic circuits is just like building electronic ones, and using standard genetic parts is like using standard-thread screws. Recently two papers were published that push this analogy closer to reality. In one, the researchers literally integrated synthetic biology with computer systems, and in the other they programmed E. coli to behave like electronics.

In the first paper, the authors used a computer to control protein production within yeast cells^1,2. Yes, protein synthesis regulated by computer. The computer monitored yellow fluorescence from YFP (yellow fluorescent protein, one of many GFP spinoffs) and kept the amount of YFP at a user defined level using flashing lights. If the amount of YFP dropped, the computer would flash a red light on the yeast; if it became too high the computer flashed a far-red light. Exposure to these wavelengths of light caused the yeast to respond by ramping up or cutting YFP production appropriately.

A second paper programmed E. coli cells to behave like a LCD display³. Unless you have a thing for vintage computers, you're probably reading this on a LCD screen. To mimic a LCD screen, the authors placed E. coli cells into a microfluidic device, where clusters (pixel) of cells are held in particular locations within an array and the flow of media between the clusters (pixels) can be easily controlled. This turns out to be necessary because they wanted the E. coli screen to flash on and off in synchrony, no small feat. A collection of videos showing their E. coli LCD screens at work are available for free here.

To synchronize the E. coli they needed a mechanism for the bacteria to communicate with one another. The go-to approach is typically quorum sensing, an adaptation widespread among bacteria that allows them to coordinate behaviors like producing toxins, growing biofilms, etc. Since this method relies only on diffusion of a small molecule, it works well over short distances, and the team used it to synchronize the cells within a pixel. But over long distances there is a delay in the signal owing to the rate of diffusion. So for long-range synchronization they used hydrogen peroxide (H₂O₂) gas. E. coli already has redox (H₂O₂ is an oxidizing agent) sensing machinery that allowed this to work.

The genetic circuit is shown in the image at the top (Prindle et al.'s Figure 1), for those interested. All four promoters are copies of the well known lux promoter used in the quorum sensing system activated by the signaling molecule 3OC6, and acyl-homoserine lactone (AHL). It is also activated by H₂O₂, which removes an E. coli repressor from the lux promoter.

Each blue rectangle in the image is a pixel containing about 5000 E. coli cells. The heat map and graph show the resolution of the pixels' synchronization for each oscillation. Each cycle of oscillation begins with the enzyme LuxI producing AHL. When enough AHL has accumulated, the cell produces GFP, NDH-2, and AiiA. GFP fluoresces green (making the pixel visible), NDH-2 produces the H₂O₂ signal to help synchronize distant pixels, and AiiA destroys AHL. When AiiA destroys a sufficient amount of AHL, the promoters are inactive, the proteins degrade, and the cycle begins again. Since transcriptions occurs in random bursts, the cycles between cells would quickly lose synchrony without the two mechanisms built into the circuit.

Now that they had a functional setup, the authors adapted their system for useful applications. First you have to appreciate the authors' rationale for adapting their system in this way. When synthetic biologists think about reporters, they usually fall back on color intensity, often GFP fluorescence. This is a big pain because intensity is highly variable because of randomness in gene expression and context specificities. Instead, the information could be encoded in the frequency of oscillations, much like FM radios. By incorporating an arsenic sensitive promoter into their construct, the team built an E. coli LCD that oscillates on and off at a frequency that depends on arsenic concentration, as arsenite. They were able to reliably detect arsenite at concentrations as low as 0.2 μM. This is below the World Health Organization recommended 0.5-μM limit for the developing world.

Overall, this is an exciting paper because the method they used to synchronize their genetic oscillators works over large distances, and their LCD screen is a new approach to biosensors and reporters. Plus, the degree of control microfluidics gives over experiments opens up a lot of doors.

Glossary:

1. GFP, YFP: Proteins that glow green and yellow (respectively) when bombarded with ultraviolet light.

2. Microfluidics: Uses precision control of the flow of small volumes of liquid through channels.

3. Quorum Sensing: A behavior evolved by bacteria allowing them to coordinate activities of mutual benefit based on population size using inter-cell signaling molecules.

4. Promoter: A genetic switch that turns on and off gene expression at the level of transcription of DNA to RNA

Image Credit: Figure 1 from Prindle et al.³

References:

2. Agapagis, C. I Heard You Like Feedback Loops. Oscillator [Blog], Scientific American (2011).

3. Prindle, A. et al. A Sensing Array of Radically Coupled Genetic ‘Biopixels'. Nature. Published online 18 December 2011.