Team:Waterloo/Results
Results
Cloning
Linuron construct
In this construct, the insert is was cloned into the BglII site between the Paph promoter and bjGFP sequence on the pRJPaph-bjGFP to give a transcriptional fusion with the linuron degradation genes. The insert itself was made by inserting the BBa_B0034 ribosome binding site ahead of each the libA and nat1 genes through PCR. These were then inserted downstream of the Paph promoter in the order of libA then nat1 by using Gibson Assembly/NEBuild. The coding sequences for the libA and nat1 genes were codon optimised for B. diazoefficiens USDA 110.5.
We successfully assembled this construct! It was validated by colony PCR (amplifying inserted genes) and sequencing. We also confirmed operon expression by measuring GFP fluorescence.
Glyphosate construct
Similar to the construction of our plasmid for resistance to linuron, we planned to insert the BBa_B0034 ribosome binding site and coding sequence of the CP4 EPSP synthase gene into the BglII site present between the Paph promoter and bjGFP coding sequence on the pRJPaph-bjGFP plasmid to give a transcriptional fusion of our EPSP gene with the bjGFP. The coding sequence of the CP4 EPSP synthase gene used was derived from the amino acid sequence claimed in the Monsanto patent, reverse translated to be codon optimised for B. diazoefficiens USDA 110.5 Attempts to assemble this construct through NEBuilder were not successful as the no conditions could be found to amplify the coding sequence with required overhangs and ribosome binding site by PCR. This likely was due to the high GC content of the coding sequence (71.2%), however the lengths of the primers could have also contributed to the difficulty. Unfortunately, we decided to drop the glyphosate part of the project and only focus on Linuron due to time constraints.
MIC assay results
Growth curve experiment with wild-type B. diazoefficiens
Although we were unable to evaluate the tolerance of engineered B. diazoefficiens strains due to time constraints, we nonetheless investigated the tolerance of wild-type B. diazoefficiens to linuron and 3,4-Dichloroaniline (3,4-DCA). This was evaluated using growth curve Minimum Inhibitory Concentration (MIC) experiments performed with hourly absorbance readings from 20 to 25 hours of incubation at 30 °C. The absorbances at the five timepoints were plotted versus time and fit with exponential curves of the form y = ae^(kx) + b, where k is growth rate. Sextuplicates were used for each toxin and concentration.
The results of these experiments are presented in figures 5, 6, and 7.
Figure 5. Growth rate versus linuron concentration from 20 to 25 hours. Error bars represent standard deviations of sextuplicates. The growth rate with no toxin was 0.08639 +/- 0.01232 1/h. The sigmoidal fit parameters A, B, C, and D are 3.23E-03, 3.46E-02, 2.96, and 2.30E+01 respectively.
Figure 6. Growth rate versus linuron concentration from 20 to 25 hours. Error bars represent standard deviations of sextuplicates. The growth rate with no toxin was 0.08639 +/- 0.01232 1/h. The sigmoidal fit parameters A, B, C, and D are 7.43E-03, 3.62E-02, 8.32E-01, and 2.08E+01 respectively.
Figure 7. Summary of key inhibitory concentrations for linuron and 3,4-DCA with B. diazoefficiens based on growth curves from 20 to 25 hours.
These differences in toxicity profiles are demonstrated by the key inhibitory concentrations in figure 7. Linuron’s IC50 value is higher than that of 3,4-DCA, demonstrating that linuron is generally less toxic than 3,4-DCA. This agrees with reported toxicity values (Tixier, 2002) (Santos, 2014). Comparing IC10 values shows that 3,4-DCA starts showing toxic effects at a much lower concentration than linuron, due to steeper slope of the plot for linuron. The steeper slope of linuron’s plot also means that linuron’s IC90 is much lower than that of 3,4-DCA. This means that full growth inhibition actually requires a lower concentration of linuron than of 3,4-DCA, despite linuron’s lower IC50. These differences highlight the importance of examining the entire toxicity profile and various inhibitory concentrations, since these convey much more information that the IC50 value alone.
We have therefore experimentally deteremined the toxicity profiles of both linuron and 3,4-DCA toward wild type B. diazoefficiens, validating our experimental approach. This information provides also a reference point against which to compare engineered strains, other toxins, or bacterial species.
Single-timepoint experiment with wild-type B. diazoefficiens
In addition to the growth curve experiment peformed with B. diazoefficiens both in linuron and 3,4-DCA, we also completed this experiment in single-time point format. This was done using one Abs600 after 22 hours incubation at 30 C. Figures 8 and 9 show plots of Abs600 versus toxin concentration. These results are shown in figures 8, 9, and 10. Fitting was performed to the four parameter in equation 1.
Figure 8. Abs600 versus linuron concentration at 22 hours. Error bars represent standard deviations of sextuplicates. The sigmoidal fit parameters A, B, C, and D are 1.28E-02, 5.52E-02, 1.62, and 1.28E+01 respectively.
Figure 9. Abs600 versus linuron concentration at 22 hours. Error bars represent standard deviations of sextuplicates. The sigmoidal fit parameters A, B, C, and D are 1.50E-02, 5.47E-02, 8.97E-01, and 3.08E+01 respectively.
Figure 10. Summary of key inhibitory concentrations for linuron and 3,4-DCA with B. diazoefficiens after 22 hours.
Figures 8 and 9 both show the characteristic sigmoidal shapes expected for MIC experiments. All datapoints are very close to the sigmoidal fit, indicating that this model is a good representation of the data.
Figure 10 summarizes some key inhibitory concentrations as calculated based on these growth curves. Although the two curves have the same general shape as those determined using the growth curve MICs, the calculated inbitory concentrations are somewhat different. The single timepoint data’s IC50 values suggest that linuron is more toxic that 3,4-DCA, which disagrees with the results obtained using growth curves and literature toxicity data (Tixier, 2002; Santos, 2014). This highlights the shortfalls of determining toxicity from Abs600 at a single timepoint and the importance of using growth curves to evaluate a toxin’s overall effect on an organism’s growth rate.
Single-timepoint experiment with E. coli empty vector and constructs
We also chose to evaluate the linuron and 3,4-DCA tolerance of our engineered DH5α strains. We performed single-timepoint MICs after 7 hours incubation at 30 C using empty vector pRJPaph-GFP and two constructs, assigned the numerical values of 14 and 21. These constructs had been vertified by colony PCR, sequencing, and GFP fluorescence reading. Each single-timepoint MIC was completed in triplicate, using LB supplemented with tetracycline to promote vector retention. Figures 3 and 4 show plots of Abs600 versus log toxin concentration for each.
Figure 3. Abs600 versus linuron concentration for 3 engineered DH5α strains at 7 hours. “EV” is pRJPaph-GFP and “14” and “21” each contain the construct with LibA and NAT1. Datapoints are averages of triplicates.
Figure 4. Abs600 versus 3,4-DCA concentration for 3 engineered DH5α strains at 7 hours. “EV” is pRJPaph-GFP and “14” and “21” each contain the construct with LibA and NAT1. Datapoints are averages of triplicates.
Linuron (figure 3) does not show any significant inhibition, even at the highest linuron concentrations. This means that it is not possible to compute parameters such as IC50, so sigmoidal fitting was not applied. Since higher linuron concentrations would not be possible due to solubility constraints, this type of assay is a poor method for evaluating LibA and NAT1 activity in E. coli. Additionally, there are no significant differences between the Abs600 values of the different strains, suggesting that the construct and empty vector do not cause measurable differences in growth rate.
3,4-DCA (figure 4) shows nearly equal inhibition of all three strains beginning at 226 µM 3,4-DCA. However, since the lower “minimal growth rate” region is absent, sigmoidal fitting was not applied, so it is not possible to calculate paramters such as IC50. As with linuron, obtained data at higher toxin concentrations would be challenging due to solubility constraints. Additionally, there are no significant differences between the Abs600 values of the different strains at any toxin concentration, suggesting that the construct and empty vector do not cause differences in growth.
These experiments therefore demonstrate that the construct does not lend linuron or 3,4-DCA tolerance to E. coli DH5α. However, the lack of difference in growth rate provides some evidence that LibA and NAT1 do not impose a significant metabolic burden on DH5α when used in this expression system.
References
Santos, S., Romeu, V., Maria, F., Sharon, C., António, M., Joaqium, V., & Amália, J. (2014). Toxicity of the herbicide linuron as assessed by bacterial and mitochondrial model systems. Toxicology in Vitro, 28 doi:10.1016/j.tiv.2014.04.004
Tixier, C., Sancelme, M., Aït-Aïssa, W., P., Bonneymoy, F., Cuer, A., Fruffaut, N., & Veschambre, H. (2002). Biotransformation of phenylurea herbicides by a soil bacterial strain, arthrobacter sp. N2: Structure, ecotoxicity and fate of diuron metabolite with soil fungi. Chemosphere, 46, 519-526. doi:10.1016/S0045-6535(01)00193-X