Overview
Precise and accurate measurements ensure the quality of our work to provide reliable results. Using universal or standard measurements allows for our experiments to generate an abundance of data that can be used by other groups for reference or comparison. As a result, an integral part of our project involved measuring levels of fluorescence and their correlation to concentrations of molecules for a variety of molecule-promoter pairs. There were two main experiments in which we measured data: the screens and the titrations.
Screen
The purpose of our measurement was to determine the fluorescence and cell density of each sample to find the pair that produces the largest fold increase. The use of fluorescence offers a clear and detectable signal, especially when the direct measurement of native proteins is difficult or impractical. In our experiment, we used the fluorescence protein sfGFP and measured cell density using the OD600 (optical density at 600 nm) to measure how much light was absorbed or scattered by the bacterial culture at a wavelength of 600 nanometers. To accurately interpret the sfGFP fluorescence data, we divided the sfGFP value by the OD600 (sfGFP/OD600). This ensured that the differences in fluorescence levels reflected changes in the GFP expression, rather than simply variations in cell density, and allowed us to compare GFP production in each culture, while accounting for differences in cell growth. We then found the average reading for each promoter for all the other molecules with the same promoter being tested in the plates read in the same stack, calling this our “background” reading for each promoter. Dividing the sfGFP/OD600 value by this background reading then gave us the fold increase compared to this background fluorescence.
Screening Procedure
1. Start up 96 deep well plates with 250 µL of LB and kanamycin broth and add E. coli with promoters, use The Gilson PLATEMASTER® to pipette all 96 wells at once
2. Cover each plate with sealing film and place all 96 deep well plates into the incubator
3. Back dilute to make 7 plates for each molecule plate, and add cells with promoters to these plates
4. Use PLATEMASTER® to transfer 150 µL to a corresponding clear bottom 96 well plate for each molecule
5. Place stack of clear bottom well plates into the Plate Reader at 37°C with shaking for 10 seconds to measure fluorescence in each well using a 600 nm wavelength
Figure 1. sfGFP and OD600 reading from plate reader for TAR molecule with Promoter Plate 1
6. Create a script that compiles on one sheet the sfGFP and OD600 values for each of the 35 plates and calculate the sfGFP/OD600 values for each plate to normalize the response and take into account cell density
Figure 2. sfGFP/OD600 for TAR with Promoter Plate 1
7. Calculate the average promoter reading for each promoter plate
Figure 3. Promoter Plate 1 average promoter reading (averages of values for the same promoter with different molecules, so the average of corresponding wells in plates 1,8,15,22,29 in this case)
Titration
The second phase of our measurements was done through titrations. Titrations are used to find the optimal molecule-promoter pair based on the changing levels of fluorescence with increasing concentrations of the molecule. A strong positive correlation would allow a user to determine the concentration of a molecule in an unknown substance based on the fluorescence produced. Titrations help determine the concentration range where the fluorescence signal is strong and detectable while minimizing background interference. In measurements for all the titrations, the strains tested should already have a strong fold increase when paired with their corresponding molecules compared to other promoters according to the screen, so finding the background fluorescence was unnecessary and sfGFP/OD600 was used to compare fluorescence.
Figure 4. Fluorescence level of 12 strains of PBA with increasing concentration of the molecule after the titration
With dark green representing larger values and lighter green representing smaller values, it was expected that as the concentration increased, the sfGFP/OD600 values would too, creating a smooth trend with the boxes getting darker when moving across each row. However, the heatmap’s scattered coloration and the lack of dark green boxes in the 1mM and 10mM column shows there was a problem. Furthermore, the consistently lower titration values compared to the screen values in the following bar graph also indicated an issue.
Figure 5. Comparison of fluorescence level between the initial screen and the 1mM well during the titration for 12 strains of PBA
In order to solve this problem, we used accurate measurements to test the fluorescence of the inducer molecules and E. coli promoters under different conditions for troubleshooting. The first of these conditions was redoing the strains that had minimal cell growth during the first set of titrations, therefore affecting the reliability of their measurements. However, these results didn’t produce a very noticeable change, as seen with the corresponding titration data.
Figure 6. Fluorescence level of 12 strains of PRO incorporating the data of the redone PRO strain titrations
To fix any possible errors, we reran the experiment under sterile conditions to eliminate any possible contaminants. Specifically, this included creating fresh media for the experiments after noticing the existing media wasn’t sterile, and autoclaving the media after diluting it with deionized H2O. After troubleshooting, the graph still did not fully reflect what we envisioned, though they did improve. While we expected an upward trend to reflect the increase in sfGRP/OD600 as the concentration increased, the linear trend of the new media data suggested little correlation between the concentration and the sfGFP/OD600 values, which was unexpected. Furthermore, the new values better matched the original screen, but were still low.
Figure 7. Comparison of titration fluorescence levels in new and old media for certain strains grown in both conditions with CND
Figure 8. Comparison of titration fluorescence levels in initial screening, 1mM titration with new media, and 1mM titration with old media for several different strains
The bar graph shows a slight increase between the titrations, meaning there was an improvement, however, the titration value still being lower than the screen value meant there was still an issue. In order to solve this, we grew completely fresh strains and used them for titration, rather than old strains which had been grown several days before. This further helped the trend of the data.
Figure 9. Comparison of titration fluorescence levels with fresh and old strains for several different strains tested with CAR
We also realized that some of the promoters we identified from the original screen were not correctly matched to their names, meaning we had been using the wrong promoters for these cases. After matching the significant wells from the initial screen to their correct promoters, we noticed an increased specificity of one molecule as an inducer for that promoter. This is shown above, where PBA has a high sfGFP/OD600 value for cyoA, a promoter it was found to produce significant fluorescence with, while other molecules do not.
Figure 10. Comparison of fluorescence level of cyoA (a promoter found to produce significant fluorescence with PBA, in red) when tested against each of the ten molecules
Based on our final measurements, we selected the molecule-promoter combinations that produced the best trend in fluorescence. These pairs have the potential to be further developed into biosensors for the molecules and greatly contribute to academic research. In particular, the following molecules and strains were identified to have an increasing trend with increasing concentration of the molecule: BHL with the ydel promoter produced a 1.7x fold increase, CND with the ybcK promoter produced a 2.2x fold increase, CND with the aegA promoter produced a 2.8x fold increase, and DEP with the yfiF promoter produced a 2.0x fold increase. Our development of theoretical parts for the detection of these molecules helps future teams with the measurements needed for their own experiments.
Figure 11-14. Titration graphs for selected strains that have the best potential for future biosensor development based on data analysis
Figure 15. pIGEM1, a theoretical plasmid created with the gadB promoter fused to sfGFP, serving as a potential biosensor for Butanoyl-Homoserine Lactone when transformed into E. coli
Reliability:
In order to ensure the reliability of our measurements, three trials were conducted for each of the titrations. The schematic below indicates the set-up of a 96-well plate, where the top 4 strains of a molecule were each tested in different concentrations of the molecules. Averaging these trails to find and plot the standard deviation indicated that our results are statistically significant and can therefore be trusted. This also helps neutralizes the effect of any contamination or environmental factors that might greatly impact the measurement if it was only taken once.
Figure 16. Example schematic for the titration across varying concentrations of the molecule, where each column represents a trial and the same color indicates the same strain.
Furthermore, we also replicated the initial screening for some of the plates several months later to ensure that it was reproducible after many failed titrations of getting the expected results. We found that the majority of the strains had similar fluorescence to the initial results, indicating that our results are accurate and reasonable.
Figure 17. Comparison of fluorescence levels between the initial screen and the new screen for Promoter Plate 2 tested with PRO
Conclusions:
In general, measurements played a crucial role in our project through a few different aspects. We used measurements in our project to consistently track the accuracy and effectiveness of each molecule-promoter pair under different environmental conditions based on their potential to become a biosensor, allowing us to improve and validate the reliability of our data. We also improve the measurement capability of other scientists by providing a methodology for them to develop their own biosensors and determining which promoters can be used to identify any of the ten molecules tested in this research. For example, our theoretical part can easily be further developed or adapted with other promoters to be used to detect the presence of certain molecules. All our research was performed with multiple trials for reliability, and our cycle of constant testing testifies to our dedication to gather accurate, dependable data.
