Experiment and Modeling
What we expect from Experiments and Modeling:
For a proof of concept, we need to know some details about MagicBlock.
1. Velidation of Quorum Sensing Receivers
Used as test standards for other parts. The fluorescence response of QS receivers have to be examined first.
2. Choose Quorum Sensing systems with high orthogonality
The crosstalk of different Quorum Sensing systems problem in working conditions may severely interfere with the signal processing function of Bioblocks, especially when with more than one QS systems are used.
3. Inter-Bioblock communication
Could bioblocks successfully communicate with each other?
4. Working parameters and kinetic model for MagicBlocks
How much time is needed for a bioblock to finish its task and when should the system proceed to the next layer? How much signal containing supernatant should be transfer to the next MagicBlock? To answer these questions, we have to build up a model, determine parameters of the model from experiment data and then guide further experiment in return.
Velidating Fluorescent QS receptors
We tested Fluorescent response for LasR-GFP, RhlR-GFP and RpaR-GFP BioBlocks using their cognate AHLs.
Fig. 1 Fluorescent response of LasR-GFP to 3OC12-HSL
Fig. 2 Fluorescent response of RhlR-GFP to 3OC6-HSL
Fig. 3 Fluorescent response of RpaR-GFP to Coumaroyl-HSL
Orthogonality prediction and test
Alone with literature search, We've used Molecular Docking to predict HSLs' affinity to different QS receiver proteins. We've choosed Las system from P. aeruginosa and RpaR system from R. palustris —of which predicted to be relatively orthogonal, to construct Advanced Bioblocks in our system.
Fig. 4 Structure of LasR and RpaR
We've also tested this prediction after plasmid construction:
Fig. 5 Orthogonality test of LasR-pLas-GFP (i):Fluorescent response to cognate and non-cognate AHLs (ii)Dose-Response curves for cognate and non-cognate AHLs (iii-vi)Fluorescent response to non-cognate AHLs in compared with 3OC12-HSL
Fig. 6 Orthogonality test of RpaR-pRpa-GFP (i):Fluorescent response to cognate and non-cognate AHLs (ii)Dose-Response curves for cognate and non-cognate AHLs (iii-vi)Fluorescent response to non-cognate AHLs in compared with Coumaroyl-HSL
Demostration of Inter-Bioblock communication
To answer this question, we measured AHL production of Generator bioblocks using LC-MS and Fluorescent QS receptor array. Both method confirmed the generation of AHLs, and the quantitive result gives a conclusion that AHL production from a MagicBlock is enough to stimulate another block and actives it's gene expression under our system's working condition.
Fig. 7 LC-MS identification of 3OC12 production from LasI culture
Because LC-MS can only indicate the production of our signal converter approxmately, but this data is too rough to instruct our following work. We get the quantitive relation between input signal concentration and output signal concentration by following experiments and deduction from model which can also indicate an unknown output signal concentration.
Kinetic model of MagicBlock
After expression experiments, we have collected enough data to describe the kinetics properties of a bioblock. We've modeled the response and re-generation of AHLs in typical intermediate bioblocks and determine parameters for liquid handling robot we used to transfer liquid supernatant between bacteria.
We use Matlab to obtain the curve of protein triggered by input signal molecule. X-axis refers to time, Y-axis refers to concentration of protein.
The system goes on when production of protein approaches maximum. After we dilute the input signal concentration, the protein will stop to generate and degrade soon. Following diagrams show the whole reaction and finally give the concentration curve of output signal molecule.
The concentration curve of protein complexity related to time
The concentration curve of output signal molecule related to time
More about this model
Modeling bacteria population
We've observed population drop when AHLs is added to culture during our experiment.
Our mathematical model successfully explained this phenomenon. For details, click