Here you will observe the protocol we followed to fit the data for the mathematical model and its statistical analysis. To improve our model, we have developed a protocol to quantify the expressed proteins. This, by measuring the fluorescence emitted by a protein expressed in the same cassettes we plan to express our antimicrobial peptides.
One of our goals this year was to generate a mathematical model that, in addition to helping us understand how our expression system works, would help us predict the concentration of peptides we will produce. These peptides are one of the key ingredients to combat wilt. However, to improve our model and make its predictions more efficient, it was necessary to feed our model with experimental data. That is why we performed an experimental design to adjust parameters of our model and improve it. The diagram below shows a summarized methodology for the analysis of the data in this protocol.
Calibration
Texas Red kit tube (iGEM provided kit)
1X PBS
3 96-well plates
200 µL micropipette
Experiment
Transformed bacteria with mCherry constructs grown in LB agar with kanamycin
Calibration
Centrifuge Texas Red kit tube to ensure the pellet is at the bottom of the tube.
Prepare a reference stock solution by resuspending protein powder in 1 mL of 1X PBS at a final concentration of 20 µM (10X).
In a microtube, aliquot 100 µL of the Texas Red reference stock solution and add 900 µL of 1X PBS, obtaining a final concentration of 2 µM (1X).
In a 96 well plate make serial dilutions in columns 1 through 11. Column 12 should only contain PBS.
On well A1 add 200 µL of reference stock solution.
On wells A2 through A12 add 100 µL of 1X PBS.
Transfer 100 µL from well A1 to well A2.
Mix well A2 with 3 resuspensions.
Transfer 100 µL of well A2 to well A3.
Mix well A3 with 3 resuspensions.
Keep transfering and mixing all the way to well A11.
When mixing well A11, take 100 µL and discard it. This step ensures that all wells contain 100 µL.
Repeat this dilutions on rows B and C.
Figure 1. Serial dilution order representation.
Measure fluorescence on rows A, B and C with an excitation wavelength of 561 nm, an emission wavelength of 610 nm, and a bandpass of 20 nm. In case this parameters can't be selected, select the closest ones and write them down.
Keep the plates on ice before measuring.
Calibration
Transform E. coli BL21(DE3) with the mCherry constructs. Plate these transformed cells on LB Agar with kanamycin at a concentration of 34 µg/mL. Incubate overnight at 37ºC.
Pick 5 transformed colonies of each of the constructs (5 biolgical replicates). Grow each colony in 5 mL of LB broth with kanamycin. Additionally, grow a negative control (Non transformed BL21 in LB broth with kanamycin). Incubate overnight at 37ºC and 225 rpm.
Add 3 mL of cell culture to 30 mL of LB broth with kanamycin at 34 µg/mL. Incubate at 37ºC and 225 rpm (as well as the negative control).
Wait until OD600 reaches 0.4 to 0.6 to commence the protein induction.
Make 5 5 mL aliquots of each tube.
Add IPTG at different concentrations: 0 mM, 0.25 mM, 0.50 mM, 0.75 mM and 1.0 mM to each of the tubes. This is explained on figure 2.
Figure 2. Total tubes used with construct denomination and colony order.
Incubate 4 hours at 37ºC and 225 rpm.
In a new, sterile microtube, make a dilution 1:10 with LB broth with the selected antibiotic.
Measure OD600 and register the obtained values.
Dilute each tube to an OD600 of between 0.05 and 0.1 in a microtube and place it on ice.
Add 200 µL in each well as shown in figure 3. Each sample should be measured double. Place the plate on ice.
Meausre fluorescence in a plate reader at an excitation wavelength of 561 nm and an emission wavelength of 610 nm,
Figure 3. Plate 1 with calibration curve and colonies 1, 2 and 3 transformed with construct 1 (with a lacI regulated promoter).
Figure 4. Plate 2 with colonies 3, 4 and 5 transformed with construct 1 (with a LacI regulated promoter) and colonies 1 and 2 transformed with construct 2 (with a lacI regulated / lambda pL hybrid promoter).
Figure 5. Plate 3 with colonies 2, 3, 4 and 5 transformed with construct 2 (with a lacI regulated / lambda pL hybrid promoter).
Three plates were read on a Varioskan LUX microplate reader at 561 nm exitation and 610 nm emission. An excitation bandwidth of 12 nm per 100 ms was also used.
Corrections were made to the flourescence readings by subtracting the blank (LB medium with antibiotic). Subsequently, the data were evaluated by a General Full Factorial Design of two parameters (Promoter and IPTG concentration), where a residual analysis was performed. The results of the residual analysis are shown in Figure 6.
Figure 6. Residual plots of data obtained in this experiment.
According to the Versus Fit plot we can conclude that the flourescence measurements are dependent (Figure 2), so we decided to test with a Durbin-Watson test statistic. The value of this statistic for our experiment was 0.918, being less than 2, this value confirms that within our data there is autocorrelation. To say that the data from the fitted experiment are reliable, there should be no dependence or autocorrelation. In addition, the variance values are high, which indicates an error at the time of carrying out the protocol.
Due to lab constrains, we were unable to finish and repeat our experiment. So, we begin to analyze our protocol. We develop hypotheses of what may have been wrong with our experiment. These are shown below:
The plates were prepared at our university. However, the microplate reader is located at a distance of approximately 20 minutes. Therefore, contamination or mixing of the wells could have occurred during transport. This could cause small differences in volume, which can cause large differences in the signal (ThermoFisher Scientific, 2022).
No equal volumes were loaded during the experiment. This may be due to the micropipette or the operator.
The plate reader did not have the option to choose the exact brand of plates we used for the experiment, so it was read with another configuration.
The experiment was carried out on transparent plates. As they are transparent, fluorescence measurements can affect the signals from neighboring wells (Auld et al., 2020).
In the future, we plan to carry out this protocol again following these considerations:
Use 96-well Black/Clear bottom plate compatible with the plate reader.
Load the plates close to the place where the plate reader is located. This in order to avoid volume displacements between wells, and therefore the variation of signals.
Use freshly calibrated pipettes.
Auld, D. S., Coassin, P. A., Coussens, N. P., Hensley, P., Klumpp-Thomas, C., Michael, S., ... & Dahlin, J. L. (2020). Microplate selection and recommended practices in high-throughput screening and quantitative biology. Assay Guidance Manual [Internet].
ThermoFisher Scientific (2022) Ensayos de microplacas de fluorescencia. Retrieved from: https://www.thermofisher.com/mx/es/home/life-science/cell-analysis/fluorescence-microplate-assays.html
