Advances in synthetic biology: on the path from prototypes to applications link
Purpose: This article discusses various challenges and tools in developing networks in the cell.
First they discuss how cascades influence network design, with various responses to an input signal and sensitivities to noise.
Cascades with higher numbers of members tend to reduce noise in the input signal
They can be made to be extremely digital, with a sharp response to an input crossing a threshold.
Feed-forward and regulatory feedback can be used to create non-linear responses to changes in input, permitting complex systems of reaction to a given input (e.g. repressilators)
Next, systems with spatial patterning are discussed, using cell-cell communication to express genes based on cell density and position.
V fischeri's quorum sensing is useful in creating optimal cell densities as well as forming a basis for many tasks in cell-cell signaling.
By creating artificial distributions of sender and receiver cells, spatial patterns can be made, such as the bullseye created by a central sender colony with concentric rings of receivers. Because the receivers are made to fluoresce when a threshold concentration of signal is present, they form this ring with boundary where concentration of signal drops below the threshold.
Several practical systems have also been developed and are in development; production of the anti-malarial precursor artemisinin and decomposition of the insecticide parathion have already been accomplished using these techniques.
Also, due to the expense and difficulty in actually building every system one needs to test, computational simulation is invaluable to the synthetic biologist; these tools will only become more necessary as synthetic biology continues to ascend as a quantitative engineering discipline.
Purpose: This article discusses Caltech's experiences with nucleic acids as sensors. They developed a system that had sensitivity to caffeine concentration by employing RNA sequences with specific secondary structure. Smolke also mentions applications to more general biosensors.
One significant advantage of nucleic acids as sensors over proteins is their reduced size. While it takes tRNA synthetase 300 amino acids to bind to phenylalanine, an RNA sequence can bind to theophylline, a similar size molecule, with only 30 bp.
The nucleic acid sensors developed by the team consisted of:
Aptamers that will bind to the target molecule (developed by starting with a pool of staggeringly large numbers of unique RNA sequences and checking for those that have affinity for the target)
A switching domain that will respond to the aptamer being bound
An output domain that will unfold a desired sequence upon triggering or untriggering of the switching domain
Generally, the output domain will be an antisense sequence that will suppress the transcription of an specific protein.
These systems can be given extremely precise threshold concentrations through directed mutation or manual adjustment of the sensor's affinity for the target or itself.
In Caltech's system, they created two sensors, one that would trigger suppression of GFP if a high threshold of caffeine was surpassed and one that would trigger production of YFP if a medium threshold of caffeine was surpassed. These were field tested in coffee with various concentrations of caffeine, demonstrating the robustness of the development.
In the long run, sensors like these will form the basis for rapid assays of a patient's blood, whereby dots of RNA in a row can respond only to increasing concentrations of a target molecule. One simply looks at how many spots fluoresce and immediately knows the concentration of the molecule. Dozens of these can be arranged in sequence to analyze as many targets as desired.