Team:NCTU Formosa/Engineering
Research
From the research, we know Gloeobacter rhodopsin produces proton motive force using light illumination and it would synthesize ATP by ATP synthase.[1] Besides, we also find data show that the growth rate of E. coli with proton pumping rhodopsin would increase.[2] Simply speaking, the E. Hybrid might grow faster than wild type E. coli.
Image
Short cell cycle and high growth rate are essential for bacteria that are modified for bioengineering usages. For the Wet Lab part, we expected that E. coli with our new part, Gloeobacter rhodopsin could grow faster with the
additional ATP produced by the light-driven proton pump. And for the Dry Lab part, we assumed that the higher growth rate and the shorter cell cycle of E. coli with GR could be precisely determined, analyzed and predicted
by several parameters and functions. Therefore, we kicked off conducting experiments and modelling.
Design
For the Wet Lab, in addition to building a new part for additional ATP production, we designed our experiments by dividing them into two parts: protein expression and functional test. We tried to find the best culture environment for
GR with the simulation of our Dry Lab. And for the Dry Lab, by deriving differential equations based on our assumptions and substituting parameters from published review articles, we could simulate the growth conditions and estimate differences that are possibly produced from the different engineered bacteria and various cultural environments. In this step, we transformed the biological characteristics into mathematical parameters. If the experimental data
meets the predicted results, we could tell how fast the E. Hybrid grows by introducing the parameters in our culture condition to our model.
Build
We built a new part, Gloeobacter rhodopsin which is optimized by harmonization and utilized in our experiments.
Figure1: Colony PCR result
of Harmonized GR after cloning into E. coli
of Harmonized GR after cloning into E. coli
And we simulate that the result would be like the figure shows below. However, we found out that the glucose uptake rate is still under assumption. Therefore, we set the wild type E. coli as control and compare the E. Hybrid to
different glucose uptake rates.
Figure 2: The figure shows the growth curve simulation of wild type E. coli and E. Hybrid for 8 hours.
The growth rate of E. Hybrid is from 1.225 to 0.918 (1/hour) when the glucose uptake rate is assumed from 1.2 to 0.9 (mmol/gram dry weight/hour). The mitosis cycle (cell cycle) is 33.95 and 45.30 (min).
Test
By inputting the data, we got the parameters of E. Hybrid and wild type E. coli.
Figure 3: The experimental data for fitting and model training for wild type E. coli and E.Hybrid for 11 hours.
After calculating growth rate from the data, we know the growth rate of E. Hybrid and wild type E. coli are 0.44 and 0.36 (1/hour) respectively. The ratio of growth rate is about 1.22 and increases about 22.2%. The mitosis cycle
(cell cycle) is 94.52 and 115.52 (min) Clearly, the E. Hybrid growth faster than E. coli. If you want to know more details, please click “Model”.
Figure 4:The prediction result of Lemo21 pET32a and Lemo21 pET32a (GR harmonized) for 14 hours.
Table1:The table is the error rate of growth curve prediction under 22mM Glucose.
The E. Hybrid not only grows fast but the growth curve of it also declines fast from the simulation results.
Learn
We can find that the predicted growth curve of E. Hybrid declines after 11 hours. However, the detection of O.D.600 would include the dead cell. Therefore, the experiment couldn’t precisely describe the growth curve of E. coli when
the E. coli start to decline owing to the glucose. This is a benefit of our model-it can describe the decline. For the validation results, we predict the glucose consumption and use experimental data to validate. At last, we assumed
that if we test the glucose concentration and find the glucose uptake rate of E. Hybrid, we can conduct reasonable and precise predictions which results in our improvement part in our engineering progress.
We find E. Hybrid not only grows fast but the growth curve of it also declines fast from the simulation results. And we speculated that it results from different glucose uptake rates to achieve rapid growth. Then, we discovered that
glucose was used up earlier, and the cell population of E. Hybrid declined. At last, we assumed that if we test the glucose concentration and find the glucose uptake rate of E. Hybrid, we can conduct reasonable and precise predictions
which results in our improvement part in our engineering progress.
Improve
We measured the glucose concentration of wild type and E. Hybrid during our functional test and found out that E. Hybrid accelerates glucose consumption.
Figure 5: The prediction of extracellular glucose concentration and validation with experimental data for Lemo21 pET32a and Lemo21 pET32a (GR harmonized). The R-square is 0.9174 and 0.9671 for E. Hybrid and wild type E. coli.
The glucose uptake rate is 11.718 and 9.667 (mM/OD600/hour) for E. Hybrid and wild type E. coli. By Calculating, we can get the Energy Efficiency (Y) is 0.0382 and 0.039, respectively. Last, we take the glucose uptake rate from
experimental data into the prediction’s parameter to improve the prediction.
We have attained our goals, constructing an E. coli which could grow faster for bioengineering uses and a system that can determine, analyze and predict the growth of E. coli have been successfully achieved. We are able to determine
and analyze the growth of E. Hybrid, and it also grows much faster as our expectations, both of the engineering successes pave promising avenues for uses in the real world.
Reference
- Lee, K. A., et al. (2011). "ATP regeneration system using E. coli ATP synthase and Gloeobacter rhodopsin and its stability." 11(5): 4261-4264.
- Claassens, N. J., et al. (2013). "Potential of proton-pumping rhodopsins: engineering photosystems into microorganisms." 31(11): 633-642.